Citation
Realizing values in teaching writing with problem-centered instruction

Material Information

Title:
Realizing values in teaching writing with problem-centered instruction development of a cognitive-composition model and instruction inventory
Creator:
Zigmond, Rosalyn H
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
xiv, 209 leaves : ; 28 cm

Subjects

Subjects / Keywords:
English language -- Rhetoric -- Study and teaching (Higher) ( lcsh )
Problem-based learning ( lcsh )
College teaching ( lcsh )
Cognition ( lcsh )
Cognition ( fast )
College teaching ( fast )
English language -- Rhetoric -- Study and teaching (Higher) ( fast )
Problem-based learning ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 199-209).
Thesis:
Educational leadership and innovation
General Note:
School of Education and Human Development
Statement of Responsibility:
by Rosalyn H. Zigmond.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
63757895 ( OCLC )
ocm63757895
Classification:
LD1193.E3 2004d Z53 ( lcc )

Downloads

This item has the following downloads:


Full Text
REALIZING VALUES IN TEACHING WRITING
WITH PROBLEM-CENTERED INSTRUCTION:
DEVELOPMENT OF A COGNITIVE-COMPOSITION MODEL
AND INSTRUCTION INVENTORY
by
Rosalyn H. Zigmond
B. Mus., University of Michigan, 1978
M. A., University of Colorado at Denver, 1993
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Educational Leadership and Innovation
2004


This thesis for the Doctor of Philosophy
degree by
Rosalyn H. Zigmond
has been approved
by
Rodney Muth
Alan Davis
Joanna Dunlap
7- z-j- py
Date
Patricia Sullivan


Zigmond, Rosalyn H. (Ph.D., Educational Leadership and Innovation)
Realizing Values in Teaching Writing with Problem-Centered Instruction:
Development of a Cognitive-Composition Model and Instruction Inventory
Thesis directed by Professor Rodney Muth
ABSTRACT
The purpose of this study was bifold: (1) to develop an instrument that
identifies teachers and students perceptions of instruction and learning strategies,
and (2) to implement the instrument and analyze the relationships between
perceptions of problem-centered instruction, self-directed learning, and critical
thinking in college composition classrooms. To accomplish this, a Cognitive-
Composition Model was created and an Instruction Inventoiy was designed,
validated, and administered. The research review explains relationships among
critical thinking, a problem-based learning approach to teaching, and composition
studies. These constructs constitute the model and facilitated interpretation of data.
The first stage of this study was forming four continua to concretize theory
in problem-based learning, self-directed learning, critical thinking, and composition
studies. These continua led to development of the Instruction Inventory, which was
used to collect data from 15 writing teachers and 247 students. Results show that
teachers believe that they used more problem-centered instructional strategies than
their students believed were used in class but that students perceived themselves to
IV


be using more self-directed learning and critical thinking skills than their teachers
did. An internal consistency seems to exist in how students view their writing
classes in respect to a problem-centered instructional focus and their perceived use
of self-directed learning and critical thinking skills. However, there is no evident
correlation at all between how teachers and students perceive the same classes in
respect to these constructs. A larger sample size may or may not confirm these
results.
A factor analysis confirmed the studys first hypothesis, that a principal
factor underlies all items in the Instruction Inventory. An oblimin rotation
confirmed the second hypothesis, that three constructs underlie the inventory, with a
set of items corresponding to each of these constructs, or dimensions.
Promising uses for the Instruction Inventory and Cognitive-Composition
Model include faculty-development programs and student-centered instructional
design, development, and delivery based on teachers and students perceptions of
teaching and learning in the classroom. Further study can improve the instruments
constructs and reliability; it also may refine the models representation of cognitive
composition.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
Rodney Muth


DEDICATION
My treasured sons, Andy and Frank, gave me encouraging support and inner
strength that kept me reading, writing, and laughing when I needed to laugh. My
wonderful parents have always been by my side, believing in me throughout my
life, and I will always be thankful for their unconditional love and support. I am
especially grateful to my husband, Irving, whose love and devotion helped me to
accomplish my life goal, this dissertation.


ACKNOWLEDGEMENT
This dissertation is the result of support from many special individuals in my life.
My advisor, Rodney Muth, provided unparalleled insight and appreciated smiles.
Alan Davis helped me with the methodology and analyses phases of this study with
endless patience. Other members of my dissertation committee, Joanna Dunlap and
Patricia Sullivan, lent constructive criticism and guidance whenever I needed it.


CONTENTS
Figures...............................................xii
Tables................................................xiii
CHAPTER
1. TEACHING WRITING WITH PROBLEM-CENTERED
INSTRUCTION....................................... 1
Problem-Centered Approach..........................1
Research Questions...........................2
Areas of Research.................................. 5
Merging Problem-Based Learning with Teaching Writing... 7
Problems are Integral........................9
Critical Thinking and Self-Directed Learning as
Problem-Based Learning Skills.............10
The Need for Critical Thinking and Self-Directed
Learning..................................12
The Challenges Critical Thinking and Self-
Directed Pose for Writing Teachers.......14
Unique Tools................................16
Overview of Methodology...........................17
Applications and Significance of Research....19
2. LINKING STUDIES IN CRITICAL THINKING,
PROBLEM-BASED LEARNING, AND
COMPOSITION STUDIES...............................22
Critical Thinking.................................22
Definitions.................................26
Uniqueness of Time and Environment...........29
Problem-Based Learning.............................31
Educational Outcomes and Advantages..........36
Barriers.....................................38
Self-Directed Learning......................42
viii


Assessing Problem-Based and Self-Directed
Learning................................... 46
Teaching Writing.................................... 51
Everyday Necessities.......................... 53
Goal-Directed Thinking........................ 54
Reflective Learning............................56
Discovery Process............................. 59
Problem Solving as Collaborative Planning..... 60
Rhetorical Theory............................. 62
Summary..............................................69
3. DEVELOPMENT AND IMPLEMENTATION OF THE
INSTRUCTION INVENTORY................................74
Framework for the Cognitive-Composition Model....... 75
Cognition and Behaviorism..................... 77
Development of the Instruction Inventory............ 79
Development of the Continua....................80
Construct Validity: Building a Body of Evidence..... 88
Step One: Identity Theoretical Framework
to Develop Items........................... 92
Step Two: Pilot Study......................... 95
Step Three: Factor Analysis....................98
Step Four: Data Analysis...................... 99
Implementation Analyses: Examining Relationships
Between Scores...................................100
Limitations of the Study............................ 101
4. REALIZING VALUES IN TEACHING WRITING WITH
PROBLEM-CENTERED INSTRUCTION........................103
Construct Validity of Instruction Inventory: Results. 104
Roles and Results from Expert Judges...........104
Pilot Study Results...........................106
Principal Component Analysis and
Oblimin Rotation Analysis...................109
Teaching and Learning Relationships in Composition:
Results....................................114
Correlations of Student Subscale Scores.......114
Correlations of Teacher Subscale Scores.......117
Correlation of Teacher and Student Scores......117
ix


Relationships between Perceptions on All Items
before Factor Analysis................120
Relationships between Scoring Trends....122
Items with Largest and Smallest Discrepancies
between Scores........................123
Discussion of Results...................132
5. CONCLUSIONS AND IMPLICATIONS.................... 133
The Cognitive-Composition Model and Teaching Writing... 135
Implications............................135
Further Research........................136
The Instruction Inventory and Perceptions.....140
Implications............................142
Further Research........................146
Coda..........................................147
APPENDIX
A. CODED STATEMENTS IN TEACHERS FORM OF
INSTRUCTION INVENTORY THAT STEM FROM
PROBLEM-BASED LEARNING CONTINUUM................... 149
B. CODED STATEMENTS IN TEACHERS FORM OF
INSTRUCTION INVENTORY THAT STEM FROM
SELF-DIRECTED LEARNING CONTINUUM................... 153
C. CODED STATEMENTS IN TEACHERS FORM OF
INSTRUCTION INVENTORY THAT STEM FROM
CRITICAL THINKING CONTINUUM.........................157
D. JUDGESDIRECTIONS..................................161
E. EMAIL MESSAGE TO FACULTY REQUEST THEM TO
PARTICIPATE IN STUDY............................... 182
F. INSTRUCTIONS FOR TEACHERS ADMINISTERING
INSTRUCTION INVENTORY...............................183
G. STUDENT FORM OF INSTRUCTION INVENTORY..............185
x


H. TEACHER FORM OF INSTRUCTION INVENTORY.......189
I. RELATIONSHIP BETWEEN TEACHERS' AND STUDENTS
MEANS FOR INSTRUCTION INVENTORY ITEMS.......193
REFERENCES.......................................199
xi


FIGURES
Figure
1-1 Relationships between critical thinking and self-directed
learning skills in a problem-centered composition classroom.....11
1 -2 Links between theories, Cognitive-Composition Model, and
Instruction Inventory construction..............................18
3- 1 Cycle between the theories, model, and survey instrument
used in this study..............................................81
4- 1 Teachers' and students' problem-based learning scores...........119
4-2 Teachers' and students' self-directed learning scores..........119
4-3 Teachers' and students' critical thinking scores............... 120
Xll


TABLES
Table
2-1 Continuum of Learners Critical Thinking Skills...................24
2-2 Continuum of Student Characteristics of Problem-Based Learning...32
2-3 Continuum of Learners Self-Directed Learning Skills..............43
2-4 Continuum of Instructional Strategies in Rhetorical Theories......64
2- 5 Synthesis of Critical Thinking, Problem-Based Learning, and
Teaching Writing in Student Attributes..........................70
3- 1 Cognitive-Composition Model.......................................76
3-2 Continuum of Student Characteristics of Problem-Centered
Instruction.....................................................82
3-3 Continuum of Learners Self-Directed Learning Skills..............84
3-4 Continuum of Learners Critical Thinking Skills...................85
3- 5 Continuum of Instructional Strategies in Rhetorical Theories......87
3 -6 Origin and Coding of Items........................................94
4- 1 Teachers and Students Comments after Pilot Study of
Instruction Inventory..........................................108
4-2 Principal Component Matrix.......................................Ill
4-3 Component Analysis with Oblimin Rotation.........................113
4-4 Item Distribution in Individual Subscales........................115
xiii


4-5 Correlations of Student Subscore Totals............................116
4-6 Correlations of Teacher Subscore Totals............................118
4-7 Highest and Lowest Scored Items by Students and Teachers...........124
4-8 Items with Largest Discrepancies between Students and
Teachers Scores.................................................126
4- 9 Items with Most Similarities between Students and Teachers
Scores...........................................................131
5- 1 Characteristics of Constructs Used in Study........................134
xiv


CHAPTER ONE
TEACHING WRITING WITH PROBLEM-CENTERED INSTRUCTION
This study is an exploration of college writing students use of critical
thinking and self-direction in classes whose teachers use an explicit problem-
centered instructional approach. Social, or collaborative, aspects of writing are
included in this study also, as Crosby (2003) describes writing as an open-ended,
relational process that engages one with othersother writers, speakers, and
readers, other thoughts, other words, other times (p. 629). Understanding students
and teachers perceptions of problem-centered writing classes, collaborative
learning, self-directed learning, and critical thinking inspired development of the
model and survey instrument in this study.
Problem-Centered Approach
Problem-based learning, a structured approach to problem-centered
instruction, originated in medical curricula in the 1960s and promotes critical
thinking and self-directed learning in social learning environments (Aspy, Aspy, &
Quinby, 1993; Barrows, 1994). Accordingly, writing teachers who intentionally
design instruction with a problem-centered approach should also be able to
emphasize collaborative aspects of critical thinking and self-directed learning
(Allen, Sargeant, Mann, Fleming, & Premi, 2003; Copland, 2003). Assuming that
1


writing students use social skills in classes with problem-centered instruction, how
do their perceptions compare to their teachers perceptions of the teaching and
learning in their composition classes? Understanding such educational dynamics is
the heart and inspiration of this study.
Research Questions
The following research questions guide this study:
1. What evidence demonstrates the construct validity of the Instruction
Inventory, an instrument that measures teachers and students perceptions
of instruction and learning?
2. What are the relationships between teachers and students perceptions of the
following constructs in the college composition class?
a. problem-centered instruction
b. self-directed learning
c. critical thinking
The first question emphasizes the first phase of the study, the critical process of
validating the constructs defined in the instrument under consideration. The second
question identifies the college writing class as the target audience for which the
instrument was designed and tested. Although the initial impetus for this study was
to understand how a problem-centered approach to teaching writing relates to
students use of critical thinking and self-directed learning skills, a survey
instrument had to be developed and validated first. Without a valid instrument that
2


measures teachers and students perceptions in teaching and learning,
understanding relationships between problem-centered instruction, self-directed
learning, and critical thinking would be uncertain.
To begin to answer these questions, theory and research were reviewed in
problem-based learning, critical thinking, and composition studies. From this
research, continua were developed, which grew into a model of cognitive writing.
After the model was built, an instrument was created to measure teachers and
students perceptions of problem-centered instructional strategies and students use
of self-directed learning and critical thinking. The underlying intention of this
instrument was to find a way to identify perceptions of such strategies and skill use
(Dolmans, Schmidt, & Gijselaers, 1994b; Do Imams, Gijselaers, Moust, Grave, &
Van Der Vleuten, 2002).
Uncertainty in how students perceive instruction can pose challenges for
teachers. Teachers can, for example, under or overestimate students use of a skill or
strategy, under or overestimate students understanding of a topic, design classroom
activities that do not target students real learning needs, give assignments that bore
or frustrate students, evaluate students work on misunderstood criteria, mislead
discussions, and the list continues. Because of these uncertain perceptions, teachers
may design learning assignments that do not meet their students needs, resulting in
a teacher-centered classroom (Brooks & Brooks, 1999). A teacher-centered, or
content-centered classroom is significantly different from a student-centered
3


classroom. A teacher-centered classroom strives to meet the needs of a teacher
needing to accomplish content-driven goals regardless of students' progress in
learning (Brooks & Brooks). A student-centered classroom strives to present content
in such a way that it complements students learning progress (Yost, 2000). The
important point here is that incorrect perceptions by teachers and students may
strain learning (Cashin & Downey. 1995).
Students misperceptions of instruction can create problems because
miscommunications with their teachers can create negative attitudes about learning
(Tan, Goh, Chia, & Treagust, 2001). For example, students in a recent study
reported that noi only did they find learning qualitative analysis tedious and
challenging in a lecture-format, they thought the practical, recitation sessions were
unrelated to their class lessons even though the teacher designed the practice
sessions to be relevant (Tan, Goh, Chia, & Treagust). Students may also believe that
their teachers are expecting too much from them or not challenging them enough.
They may fed that their teachers are presenting instruction with no regard for their
personal learning needs, or students may think that their teachers are under or
overestimating their acquisition of the content and learning objectives for the
course. On the other hand, students perceptions of humor in instruction, for
example, can result in increased learning (Wanzer & Frymier, 1999). In short,
students perceptions of instruction do seem to affect their learning (Cashin &
Downey, 1995).
4


To understand these types of classroom and instructional idiosyncrasies, a
Cognitive-Composition Model and an Instruction Inventory were designed for this
study. Their purpose is neither to assess teaching or learning nor to determine the
effect of different instructional strategies on different types of learning. Rather, the
Cognitive-Composition Model and Instruction Inventory reveal information about
relationships between constructs. Moreover, the study focuses on the relationships
between teachers and students perceptions of problem-centered instructional
strategies, self-directed learning, and critical thinking in college composition
classes.
Areas of Research
Main areas of research for this study include problem-centered instructional
strategies, self-directed learning, and critical thinking skills. Problem-centered
instruction is a logical approach to help writing students learn how to write well
(Flower, 1989). It is grounded in problem-based learning, a particular approach that
focuses students learning on solving problem scenarios, exploring what they know
and need to know in order to solve the problem, and evaluating their learning as
individuals and team members (Barrows, 1985, 1988; Barrows & Tamblyn, 1980).
Self-directed learning is a desirable practice for mature adults to hone (Guglielmino
& Guglielmino, 2001; Knowles, 1975). It is a strategy that encourages self-concepts
of responsibility and control in the learning process, including identification of
individual learning objectives, activities, and assessments (Candy, 1991;
5


Zimmerman & Lebau, 2000). Critical thinking is a proficiency that enables people
to make difficult life decisions (Barnet & Bedau, 2002; Greenfield, 1987). It is a
skill that emphasizes purposeful, rationalized analysis and evaluation of others
statements, written and spoken; critical thinking also entails problem-solving,
intentional reflection, and creativity (Barnet & Bedau, 2002; Drake, 1976; McPeck,
1990). To establish a common-ground understanding of the constructs in this study,
problem-centered instruction is an informal way of teaching. It is grounded in
problem-based learning, a particular approach that focuses students learning on
solving problem scenarios, exploring what they know and need to know in order to
solve the problem, and evaluating their learning as individuals and team members
(Barrows, 1985, 1988; Barrows & Tamblyn, 1980). Self-directed learning is a
strategy that encourages self-concepts of responsibility and control in the learning
process, including identification of individual learning objectives, activities, and
assessments (Candy, 1991; Zimmerman & Lebau. 2000). Critical thinking is a skill
that emphasizes purposeful, rationalized analysis and evaluation of others
statements, written and spoken. It entails problem-solving, intentional reflection,
and creativity (Barnet & Bedau, 2002; Drake, 1976; McPeck, 1990).
The Instruction Inventory is a short, discipline-independent survey designed
around problem-centered instruction, self-directed learning, and critical thinking.
The Cognitive-Composition Model depicts attributes of two polarized types of
writers on a continuum. It illustrates one type of writer as a self-directed, critically
6


thinking problem-solver and the opposite type of writer as directed by others, quick-
thinking, and avoiding conflict. Somewhere between these outliers are authentic,
logical descriptions of real student writers. The important point of the model is that
characteristics of problem-centered instruction, self-directed learning, and critical
thinking in writing classes manifest themselves in varying degrees, not as
predefined, polarized attributes. The relationships between such variations as they
are perceived by teachers and students are the essence of this study.
Merging Problem-Based Learning with Teaching Writing
This chapter is titled Explicitly Teaching Writing with Problem-Centered
Instruction because teaching writing has rarely been formally studied in a problem-
centered context. Instructional strategies integral to problem-centered learning,
however, are not all that new to composition instruction. For example, teaching
writing in accord with epistemic and expressivist rhetorical theory entails
instructional strategies such as teaching writing as a process, encouraging deep
thinking about issues, and stimulating discovery throughout the learning process
(Berlin, 1987). These types of instructional strategies focus this study.
The strength of exploring the discipline of composition studies in problem-
centered instruction is that both naturally integrate self-directed learning and critical
thinking skills (Flower, 1989; Zimmerman & Lebau, 2000). Problem-centered
learning environments have been found to develop self-directed learning and critical
thinking skills in non-composition classes, although these findings are not
7


quantified (Allen, Duch, & Groh, 1996; Stepien & Gallagher, 1993; Woods, 1994).
Albanese and Mitchell (1993), for example, conducted a meta-analvsis of problem-
based learning environments for medical curricula and suggested that problem-
based learning is a valuable approach to learning, but that it should be studied
continuously and its benefits analyzed to understand and appreciate its strength in
developing cognition. These types of encouraging reports provide all the more
reason, and need, to develop an instrument that identifies perceptions of problem-
centered instruction, self-directed learning, and critical thinking skills.
If problem-centered instruction can promote students critical thinking and
self-directed learning skills in medical, engineering, and business curricula
(Barrows, 1994; Stinson & Milter, 1996; Woods, 1994), it should also achieve
similar outcomes in composition classes. Problem-centered instruction advances
authentic problem solving, another characteristic of critical thinkers and self-
directed learners:
Students, as they are increasingly posed with problems relating to
themselves in the world and with the world, will feel increasingly
challenged and obliged to respond to that challenge. Because they
apprehend the challenge as interrelated to other problems within a
total context, not as a theoretical question, the resulting
comprehension tends to be increasingly critical and thus constantly
less alienated. Their response to the challenge evokes new
challenges, followed by new understandings; and gradually the
students come to regard themselves as committed. (Freire, 1990,
pp. 68-69)
8


According to Freire (1990), if problem-engagement during the learning
process heightens students commitment to learn, then exploration of problems in
problem-centered instruction should also be a viable way to teach writing that
engages learners thinking (Barrows, 1996; Edens, 2000; Gijselaers, 1996).
Accordingly, problem-based learning should help students to develop their critical
thinking as well as their identities as maturing writers, as its foundation is solving
problems, clearly an intellectual activity.
Problems Are Integral
Problem-centered instruction is grounded in problem-based learning,
student-centered learning in that students identify what they know and what they
need to know, in consultation with their teachers, to solve problems (Barrows, 1985,
1986, 1994, 1996; Barrows & Tamblyn, 1980). Students work in small groups and
discuss their collective knowledge to teach and learn from each other (Bridges &
Hallinger, 1997). The instructor, or tutor, assumes the role of a coach and asks
metacognitive questions that challenge students thinking and writing about the
problems (Barrows. 1988; Gijselaers, 1996; Stepien, & Gallagher, 1993).
The relationships between problem-centered instruction, critical thinking
skills, and self-directed learning skills in a composition classroom are illustrated in
Figure 1-1. Norris and Ennis (1989) explain: Since critical thinking takes place
within a problem-solving context, and often in the context of interacting with other
people, the critical thinker needs to be able to function effectively within these
9


contexts (p. 8). Norris and Ennis also emphasize the importance of collaborative
learning for critical thinkers, which is also inherent in the problem-based learning
environment (Bruffee, 1993). In Figure 1-1, the connection between the college
composition classroom and pressures from the real world on composition classes
(arrows) increase the importance of preparing students for real-world challenges.
Critical thinking skills and self-directed learning skills are two of several elements
of problem-based learning environments that are highlighted in this study.
Critical Thinking and Self-Directed Learning as
Problem-Based Learning Skills
In essence, critical thinking involves reflection, active engagement,
suspended conclusions, open-mindedness, evaluation, analysis, formulating
inferences, making decisions, and problem solving (Barnet & Bedau, 2002;
McPeck, 1981; Norris, 1992; Ruggiero, 1988). Attitude and application are integral
to critical thinking also, as people must use their critical thinking skills regularly,
not dust them off to decipher occasional crises (Halpem. 1989; Norris & Ennis,
1989). Thinking critically requires sustained effort and commitment to fulfill its
value and nurture its growth over time. These types of cognitive skills help to solve
problems. Self-directed learning involves accepting responsibility for and
controlling one's own learning (Brockett & Hiemstra, 1991; Brookfield, 1993). One
of the more important premises of self-directed learning is that instructional
activities should be based on the learners perceived needs (Hiemstra, 1998). The
10


Real World
Figure 1-1. Relationships between critical thinking and self-directed learning skills
in a problem-centered composition classroom.
11


distinction between self-directed learning and other-directed learning is that self-
directed learners identify their own needs and objectives, oftentimes in conjunction
with their teachers input. Likewise, students also assess their own learning. Self-
directed learning, by definition, is clearly student-centered and serves individuals
well when attempting to solve their learning problems and needs.
The Need for Critical Thinking and Self-Directed Learning
People need critical thinking skills to make sense of the onslaught of
information that bombards them every hour of every day (Meyers, 1986). These are
the skills that enable people to think through daily floods of information: In a
world of complex facts, events and ideas, there simply is no short cut to analyzing
arguments apart from understanding these complexities (McPeck, 1981, p. 93).
People must think critically to make sensible decisions about the economy, natural
resources, health, education, investments, and nuclear weapons that affect every
aspect of life. Critical thinking skills are important lifelong-learning skills to
develop because most people complete their formal education in their early twenties
but can live into their eighties. They will not learn everything in college that they
will need the rest of their fifty or sixty years of life (Rubinstein & Firstenberg,
1987). Developing students thinking is particularly acute today, when our
cultures output of information far exceeds our ability to think critically about that
information (Meyers, 1986, p. xi).
12


People need self-directed learning skills because they usually know better
than anyone what they know, need to learn, want to learn, and how to learn
(Dolmans & Schmidt, 2000). In an age where people must always be learning at
work to keep their jobs, earn promotions, converse intelligently, and understand
newsworthy issues around the globe, they must learn new things. The workplace
requires constant learning because the quantity and quality of information grows
daily in every field. Self-directed learners know what they want and need in their
lives and manage their learning to meet their own objectives (Hiemstra, 1998).
Critical thinking can help people to write well because thinking and writing
reflect each other (Coe, 1974). High quality writing reflects high quality thinking
(Emig, 1977; Flower & Hayes, 1988). When students rush through the writing
process, they do not give themselves time to think about the message they intend to
communicate; the result is usually poor, ineffective writing. Conversely, when
students engage in the writing process and take time to think critically, draft, elicit
feedback, and revise, their written products tend to be well-structured presentations
of their thoughts (Young, 2003). Writing teachers have the opportunity and
responsibility to boost their students critical thinking and self-directed learning
skills (McPeck, 1981).
13


The Challenges Critical Thinking and Self-Directed Learning
Pose for Writing Teachers
Writing teachers face the problem, however, of enabling students to move
from using ordinary, everyday, social language to individualized, intentional critical
thinking, or from knowledge telling to knowledge transforming (Bereiter &
Scardamalia, 1987). Knowledge telling is reporting information. Knowledge
transforming is creating new ideas. It is imperative that citizens of the 20th and 21st
centuries think critically, yet recent tests have shown that only 25% to 50% of 1st
year college students have the skills needed for logical thought (Halpem, 1989, p.
37). Without challenges to go beyond everyday language use and transform
knowledge, students tend to use shallow language and thinking in their writing
(Flower, 1994a).
Consider the following writing assignment as a way for students to
transform knowledge: Identify a problem that you are currently facing as a
freshman at this school that other freshmen also face. What are your resources for
solving the problem? What realistic solutions might eliminate or alleviate the
problem? Write a feature story for the school newspaper that evaluates and argues
for different solutions depending on different types of problems that freshmen
typically encounter. In contrast, an assignment cue that endorses knowledge telling
might be phrased as follows: What is the most common problem that college
freshmen face? List and explain five ways to deal with it. To write an essay
14


responding to this question, students would engage in knowledge telling because the
requirement is to merely recall information from an external source, not to critically
think about it, synthesize, apply, or evaluate it.
Part of the challenge, responsibility, and opportunity in moving students
critical thinking and self-directed learning skills from telling to transforming lies in
empowering students to collaborate and converse as writers:
College and university teachers (and textbook writers) cannot tell
students how to write. Instead, because writing is itself a displaced
form of conversation, teachers have to find ways for students to
learn to engage in constructive conversation with one another
about writing. (Bruffee, 1993, p. 57)
To improve their writing, teachers and students should write and speak as writers,
not as students feigning to be writers (Bruffee). Authentic workshop activities
engage students as real writers to improve their writing effectiveness for real
audiences. Such workshops transcend the issue of grades and focus on writing well
(Flower, 1994c).
Professional writers usually ask their peers for feedback on their work, and
so should student writers. Basically, the more authentic the learning environment,
the more authentic, and critical, the learning; in a problem-centered learning
environment, students are genuinely and conscientiously engaged (Barrows 1985).
Thus, a problem-centered learning environment redefines the rules and regulations
of classroom learning, compels the instructor to release control of the learning
process, allows students to develop as authentic writers, and frees students to do so
15


(Cottell, 2002). The challenge for writing teachers is letting go of control and
creating the kind of learning environment that encourages engaged critical thinking
and self-directed learning.
Another challenge that critical thinking and self-directed learning pose for
writing teachers is how to evaluate these skills. An effective way to evaluate such
skills is to ask the learners evaluate themselves, even though problems inherently
arise when assessing writing. Huot (2002) notes two assumptions about assessment
and teaching writing: one assumption is that in literate activity, assessment is
everywhere; the other assumption is that being able to assess writing is an
important part of being able to write well (p. 165). Hout continues to explain how
self-assessment is valuable for writers because it helps them to develop as critical
thinkers and self-directed learners. After all, learning to describe what one sees in a
text is an important part of being able to develop the critical consciousness
necessary for a developed evaluative discourse about writing (p. 177). In sum, the
challenges that writing teachers face when promoting self-directed learning and
critical thinking involve assessing learning and progress.
Unique Tools
In an attempt to understand the relationships between thinking critically and
writing in a problem-based learning environment, a Cognitive-Composition Model
and Instruction Inventory were developed for this study. The Instruction Inventory
merges these constructs into a discipline-generic survey that elicits teachers and
16


students perceptions about how instruction is delivered and received. The
Cognitive-Composition Model provides a continuum that combines the Instruction
Inventory with rhetorical theories, whereby opposite ends of the continuum contrast
cognitive and behavioral attributes of writing students. Links between theories,
Cognitive-Composition Model, and Instruction Inventory are detailed in the Review
of Literature in Chapter Two. Methodologies of creating these tools are detailed in
Chapter Three. Chapter Four presents analyses of the results of the instrument
implementation and conclusions and future research comprise Chapter Five of this
dissertation.
Overview of Methodology
Survey research methodology and construct validity generated the data for
this study. Theory and research comprised the construction of four continua, from
which a Cognitive-Composition Model was created. An Instruction Inventory, or
survey, was then written. To keep it discipline-independent, the survey
encompassed only three of the continua, excluding the continuum illustrating
rhetorical theories for teaching writing. Figure 1-2 illustrates which continua
contributed to the model and which contributed to the survey instrument. The
double-headed arrow shows how the instrument and model inform each other.
The Instruction Inventory is composed of two sections: a section on
problem-centered attributes of instruction and a section on self-direction and critical
thinking in learning. Units of measure were fifteen college composition classes with
17


eighteen students and one teacher in each class. Complete anonymity was assured
because teachers created their own codes and shared them only with
Figure 1-2. Links between theories, Cognitive-Composition Model, and Instruction
Inventory construction.
their students. Questions included fifty-four Likert-type scaled statements in one
survey for teachers and a complementary survey for students.
A detailed process was followed to validate the constructs scrutinized in the
study: (a) a draft of the survey was distributed to a panel of judges to evaluate its
contents, (b) the survey was revised, (c) a pilot study was implemented in which the
survey generated qualitative data, (d) the survey was revised again, (e) the survey
was distributed to writing instructors and their students, (f) data were analyzed.
Analysis included a series of reliability tests and factor analyses, including factor
18


and principal component analyses and oblimin and varimax rotations. It is important
to note here that the implementation of the Instruction Inventory in composition
classes was necessary to test the datas reliability and validity. Equally, its
implementation revealed relationships between teachers and students perceptions
of self-directed learning and critical thinking in a problem-centered instructional
environment. To determine the perceived relationships between the constructs,
students total scores were compared to their teachers total scores on the Instruction
Inventory.
Applications and Significance of Research
Awareness of the Cognitive-Composition Model can give writing teachers
insight into understanding their students thinking and writing. If a writing teacher
sees a student listing information in an essay without critically thinking about it,
telling knowledge, he or she can encourage that student to synthesize, or transform,
the information into a purposeful essay on the topic. By contrasting cognitive with
behavioral composition on a continuum, the model can help teachers identify if a
student is more cognitive or behavioral and encourage their development in a
particular direction with metacognitive questions (Barrows, 1988; Stepien &
Gallagher. 1993; Wilkerson. 1996).
The Instruction Inventory can help teachers understand how their
perceptions of instruction and learning compare to their students perceptions. In so
doing, they can make informed decisions about their instructional design and
19


delivery of activities for students. It can reveal discrepancies and similarities
between teachers and their students perceptions of instructional strategies used in
the composition class and the nature of those discrepancies. It is important to note,
however, that this instrument should be implemented near the end of a semester, the
time when students are best in tune with their teachers instructional style and
teachers are most knowledgeable about their students' learning styles.
Consequently, information gained from the Instruction Inventory will inform the
teachers subsequent class designs, not necessarily the current class. In short, the
Instruction Inventory functions better as a summative, not formative, assessment
tool.
When teachers know whether, or to what degree, their students are using
particular skills, they can refine their instructional design and delivery to improve
their teaching effectiveness (Roche & Marsh, 2000). If teachers believe their
students are using self-directed and critical thinking skills when they are actually
not, the instructors efforts may be ineffective and useless. Changes that teachers
make based on students" perceptions can and should be made, however, as student
ratings of college faculty are typically valid assessments (Koon & Murray, 1995).
Why should a teacher blindly assume students are developing as self-directed,
critical thinkers when the Instruction Inventory can validate or negate their
assumptions? Ultimately, the Cognitive-Composition Model and Instruction
Inventory can help writing teachers of all levels, not only in higher education, to
20


understand their students use of critical thinking, problem-solving, and self-
directed learning skills. It can also guide them through the continuous challenge of
how to encourage students to move from superficial to critical thinking, or how to
reflect knowledge creation in their writing.
Learning the skills of clear thinking can help everyone to recognize
propaganda and thus not fall prey to it, to analyze unstated
assumptions in arguments, to realize when there is deliberate
deception, to consider the credibility of an information source, and
think a problem or a decision through in the best way possible.
(Halpem, 1989, p. 7)
In essence, writing teachers have the delightful duty to create learning environments
that inspire their students to be self-directed learners, critical thinkers, and
constructive writers.
21


CHAPTER TWO
LINKING STUDIES IN CRITICAL THINKING, PROBLEM-
BASED LEARNING, AND COMPOSITION STUDIES
This literature review contains three major topics: critical thinking, problem-
based learning, and composition studies. The first section defines critical thinking
and explains its uniqueness. The second section describes problem-based learning as
a foundation of problem-centered instruction, barriers to problem-based learning, its
outcomes and advantages, and self-directed learning, an essential component of
problem-based learning. The final section reviews everyday necessities, goal-
directed thinking, reflective learning, discovery process, and rhetorical theory as
they pertain to teaching writing. The summary of this literature review explains the
connections between a problem-centered orientation in writing classes and students
relevant propensities toward critical thinking, problem-solving, and self-directed
learning skills.
Critical Thinking
Critical thinking empowers adults to make sensible decisions and solve
challenging problems every day of their lives. Thinking organizes reality for us,
and our own place in those realities, and creates alternative realities that we might
hope to achieve (or to avoid) (Smith, 1990, p. 30). A widely accepted definition of


critical thinking emphasizes both its process and product: '''Critical thinking is
reasonable and reflective thinking that is focused upon deciding what to believe or
do (Norris & Ennis, 1989, p. 1). The process explains how to think critically,
reasonably, and reflectively. The product describes the results of thinking critically,
or making decisions about future actions. Table 2-1 illustrates a continuum of
learners critical thinking skills.
Critical thinking should receive proper attention in school: The
development of thinking skills has always been a problematical aspect of teaching;
but it is particularly acute today, when our cultures output of information far
exceeds our ability to think critically about that information (Meyers, 1986, p. xi).
Developing thinking skills does not have to be problematic; however, the following
seems evident in college composition courses:
Written assignments are equally valuable to the instructor because
they reveal, at least in part, what student thinking processes are
like. Though it is a time-consuming process, sitting down with
student to go over papers allows teachers to help the students see
more explicitly their own thought processes and thus become more
aware of their progress in developing new modes of critical
thinking. (Meyers, 1986, p. 86)
Writing instruction can promote effective critical thinking because teaching
thinking is analogous to teaching writing (Ruggiero, 1988, p. 11). Two critical
thinking skills, problem solving and self-directed learning, are traits of mature adult
learners (Knowles, 1975), yet sometimes overlooked in college composition
23


Table 2-1
Continuum of Learners Critical Thinking Skills
Critical Thinking ii if wi Surface Thinking
Focus Student Attributes
1. Problem solving Cl. Analyzes problems. (Norris, 1992) G1. Stays away from problems when possible.
2. Evaluation skills C2. Evaluates claims, assumptions, and evidence. (Smith, 1990) G2. Accepts claims, assumptions, and evidence on value.
3. Learning Attitude C3. Shows imaginative open-mindedness and intellectual curiosity. (Norris & Ennis, 1989) G3. Shows an objective, indifferent approach to learning.
4. Idea generation C4. Considers new ideas. (Barnet & Bedau, 2002) G4. Depends on former ideas.
5. Summarizing skills C5. Summarizes arguments accurately. (Barnet & Bedau, 2002) G5. Summarizes arguments incorrectly.
6. Self-reflection C6. Uses judicious, or reflective, skepticism. (McPeck, 1981) G6. Accepts information without much reflection or contemplation.
7. Construction of solutions Cl. Constructs alternatives and solutions in problem-solving. (Meyers, 1986) G7. Applies first solutions developed in problem- solving.
8. Argument construction C8. Can recognize and construct sound arguments. (Ruggiero, 1988) G8. Does not deliberately recognize or analyze sound arguments.
9. Application of logic C9. Applies principles of formal and informal logic. (Barnet & Bedau, 2002) G9. Vague application of formal and informal logic principles.
10. Understanding fallacies C10. Avoids fallacies in reasoning. (Ruggiero, 1988) G10. Uses logical fallacies in reasoning.
24


Table 2-1 (continued)
Focus Critical Thinking Surface Thinking Student Attributes
11. Making conclusions C11. Suspends conclusions. (McPeck, 1981) G11. Makes quick conclusions and judgments.
12. Accepting truths C12. Sets aside own ideas of the truth and reflects on alternatives. (Meyers, 1986) G12. Depends on othersideas 1 of the truth without reflection.
13. Motivation C13. Internally motivated about learning. (Halpem, 1989) G13. Externally motivated about learning.
14. Logical Inferences C14. Can judge and make logical inferences. (Norris, 1992) G14. Hastily judges and draws logical inferences.
courses. One of the biggest challenges for college teachers is to help their students
develop these thinking skills. Teachers should also expose students to new values
and perceptions beyond their immediate learning experiences, a vital ingredient to
critical thinking and writing (Meyers, 1986). Instead, many writing teachers
traditionally pay attention to writing research papers with painstaking attention to
correct punctuation and grammar (Berlin, 1988). Writing instruction can and should,
however, advance critical thinking skills because Writing is a traditional, obvious,
and easy way to encourage students to become creative and critical themselves
(Smith, 1990, p. 126). In turn, these creative and critical thinkers can become savvy
information consumers and productive, intelligent citizens. Even if teachers believe
they are promoting critical thinking, how do they know if their students are truly
25


using these skills? The snag here could be that the definition of critical thinking
varies from teacher to teacher and student to student, making its assessment
difficult.
Definitions
Almost one hundred years ago, Dewey (1910) wrote the following about
thinking:
The most important factor in the training of good mental habits
consists in acquiring the attitude of suspended conclusion, and in
mastering the various methods of searching for new materials to
corroborate or to refute the first suggestions that occur. To
maintain the state of doubt and to carry on systematic and
protracted inquirythese are the essentials of thinking, (p. 13)
McPeck (1981) concurs with Dewey's concept of suspension and defines
critical thinking as the appropriate use of reflective skepticism within the problem
area under consideration. Knowing how and when to apply this reflective skepticism
effectively requires the thinker to know something about the field in question (p.
7). McPeck continues to explain that skepticism and the suspension of making quick
decisions is a significant attribute of critical thinking. His use of reflective
skepticism mirrors Deweys definition of reflective thought, which Dewey defines
as Active, persistent, and careful consideration of any belief or supposedform of
knowledge in the light of the grounds that support it, and the further conclusions to
which it tends (p. 6). In essence, McPecks and Deweys definitions of critical
26


thinking and reflective thought are in sync with each other and emphasize the notion
of intentional deliberation of the issue at hand.
Skepticism is an intellectual obligation, however, according to Elbow
(1973). The seventeenth century harbored a skeptical ideology; that is, scientists
prided themselves in doubting theories and not being gullible. In turn,
We now have a state of affairs where to almost anyone in the
academic or intellectual world, it seems as though when he plays
the doubting game he is being rigorous, disciplined, rational, and
tough-minded. And if for any reason he refrains from playing the
doubting game, he feels he is being unintellectual, irrational, and
sloppy, (p. 151)
Elbow claims such skepticism and doubt tends to promote a disengaged, or closed
perception toward new ideas. While this type of doubt has an intellectual value, it is
only haif of the game. The believing game is an approach to thinking in which the
only chance of correcting a mistake is to affirm or believe, but not to argue. The
more people believe and experience more things instead of doubt them, the more
open they are to change their minds, open their perspectives, take risks, work in
groups, and listen to others. Elbow cautions that believing takes time because it is
engaging and deliberative; interestingly, Dewey and McPeck describe critical
thinking as engaging and deliberative also (Greenfield, 1987; Halpem, 1989).
Norris (1992) adds a different spin on the role of reflection in critical
thinking and posits that critical thinking is actually a description of how people
reflect on others thinking in writing and speech. Critical thinking, then, is thinking
27


about thinking, what some have referred to as metacognition (Norris, p. 184).
Thinking about others thinking involves analysis and evaluation of others
statements. Barnet and Bedau (2002) see such roles of analysis and evaluation
prevalent in critical thinking:
Thinking critically involves, along with imagination (so that we
can see our own beliefs from another point of view), a twofold
activity: analysis, finding the parts of the problem and the
separating them, trying to see how things fit together; and
evaluation, judging the merit of our claims and assumptions and
the weight of the evidence in their favor, (p. 4)
Halpem (1989) also emphasizes evaluation and analysis in critical thinking, as they
relate to problem-solving and decision-making:
The term critical thinking is used to describe thinking that is
purposeful, reasoned, and goal directed. It is the kind of thinking
involved in solving problems, formulating inferences, calculating
likelihoods, and making decisions. The critical part of critical
thinking denotes an evaluation component, (p. 5)
Another definition that incorporates evaluation in critical thinking is Drakes (1976)
definition: Judgments about problem terms, arguments, laws, principles,
hypotheses, and statements comprise critical thinking, then (p. 46).
In addition to the aforementioned definitions of critical thinking, McPeck
(1981, p. 13) offers a list of the major descriptions of critical thinking: (a) critical
thinking is not a distinct subject because people are always thinking about
something in particular, (b) criteria to evaluate critical thinking vary from field to
field, (c) critical thinking does not necessarily entail disagreement, rejection, or
28


deviation from norms, (d) critical thinking includes not only assessment of
statements, but problem solving and active engagement in certain activities as well,
(e) critical thinkers in one subject may not be critical thinkers in another subject, (f)
the study of logic does not constitute critical thinking, (g) critical thinking is a
dimension of rationality.
In sum, critical thinking involves reflection, active engagement, process,
product, suspended conclusions, open-mindedness, skepticism, careful
consideration, metacognition, evaluation, analysis, formulating inferences, making
decisions, domain specificity, assessment of statements, achievement, problem
solving, and self-directed learning. Attitude and application is integral to critical
thinking, also, as people must use the critical thinking skills they possess to keep
them sharp (Halpem, 1989; Norris & Ennis, 1989). Thinking critically requires
sustained effort and commitment to fulfill its value and nurture its growth over time.
Uniqueness of Time and Environment
Time distinguishes critical thinking from other thinking. Dewey (1910)
wrote, If the suggestion that occurs is at once accepted, we have uncritical
thinking, the minimum of reflection (p. 13). This implies that uncritical thinking is
done impetuously and that reflection requires deliberation. Evaluating thinking
processes, outcomes of the thought process, and conclusions also require significant
time (Halpem, 1989). Greenfield (1987) implies the valuable time requisites of
teaching thinking skills: Any kind of attention to such skills and type of instruction
29


in thinking skills lead to benefits. Ignoring problem solving entirely or assuming
that students acquire thinking skills simply by practice in solving problems, drill, or
osmosis does not (Greenfield, p. 20). The time it takes to strengthen critical
thinking and problem solving in school is more important than drilling on content
for some teachers: I believe that paying more attention to process is worth the
sacrifice of some content in our courses (Stice, 1987, p. 98). Although time and
content are two costs of teaching critical thinking and reasoning skills, the rewards
are worthwhile (Lochhead & Whimbey, 1987).
A unique learning environment also distinguishes critical thinking from
other thinking. The learning atmosphere must value process over product, discovery
over correctness. It must tolerate mistakes and encourage students to learn from
them. The environment that nurtures critical thinking must encourage students to be
adventurous and creative, free from embarrassing criticism and hurt feelings from
teachers and classmates (Ruggiero, 1988).
To promote critical thinking, the learning atmosphere should also provide
frequent challenges in the form of problems and issues appropriate to students level
of intellectual development, problems and issues that afford them practice in
thinking skills (Ruggiero, p. 97). These challenges allow them to struggle with
problems and issues long enough to develop strategies for dealing with confusion
and frustration, but not so long as to defeat them (Ruggiero, 1988, p. 97). Even
though the critical thinking classroom atmosphere is invigorating, tolerant, and
30


creative, many colleges and universities continue to deliver much instruction via
lecture format. Lectures tend to be passive, non-creative, teacher-centered forms of
instruction that do not typically promote critical thinking (Meyers, 1986).
Problem-Based Learning
Problem-based learning can create a learning environment that is
invigorating, tolerant, and creative (see Table 2-2). It is a process in which problems
form the organizing focus and stimulus for learning, teachers are facilitators or
guides, and learning is student-centered generally occurring in small student groups
(Ayres, 2002, Barrows, 1996; Barrows & Tamblyn, 1980). Some educators (Ayers,
2002; Barrows, 1996; Woods, 1994) see problem-based learning as a learning
environment, or situation, in which the problem drives the learning. In such an
environment or process, learners actively construct knowledge with problem-based
learning (Brooks & Brooks, 1999; Flower, 1994a; Gijselaers, 1996).
Although problem-based learning is often conceptualized as a form
of discovery learning, it also provides a framework where
extensive social or cooperative learning is encouraged through
curricular supports, such as faculty- or peer-group-derived course
objectives, resource lists, and evaluative criteria. (Zimmerman &
Lebeau, 2000, p. 311)
Barrows (1986) describes problem-based learning as exhibiting variable
instructional strategies, including lecture, case-based lectures, case methods, and
closed loop or reiterative problem-based methods.
31


Table 2-2
Continuum of Student Characteristics of Problem-Based Learning
Focus Problem-Based Learning Teacher-Based Learning Student Attributes
1. Content P1. Identifies content to be learned. (Barrows, 1996) Tl. Learns information determined by others.
2. Personalized learning P2. Personalizes own learning based on personal interests. (Barrows & Tamblyn, 1980) T2. Generalizes own learning.
3. Knowledge construction P3. Constructs knowledge actively when learning. (Barrows & Tamblyn, 1980) T3. Does not knowingly construct knowledge when learning.
4. Problem analysis P4. Analyzes problems. (Birch, 1986) T4. Avoids problems.
5. Reasoning skills P5. Practices logical reasoning and problem-solving skills. (Edens, 2000) T5. Thinks compulsively and rashly.
6. Participation activity P6. Participates actively and is motivated when learning. (Woods, 1994) T6. Participates on passive, low level of involvement when learning.
7. Group problem solving P7. Encourages group problem solving. (Allen, Duch, & Groh, 1996) T7. Withdraws from group problem solving.
8. Solution implement- ation P8. Implements solutions to problems solved. (Bridges & Hallinger, 1996) T8. Does not complete solution implementation.
9. Prior knowledge P9. Activates own prior knowledge. (Bridges & Hallinger, 1996) T9. Does not acknowledge prior knowledge.
10. Independence P10. Learns independently. (Gijselaers, 1996) T10. Depends on others throughout learning process.
11. Critical thinking P11. Thinks critical Iv. (Magnussen, Ishida, & Itano, 2000) Til. Possesses illogical, irrational reasoning skills.
12. Learning transfer P12. Transfers learning responsibility from teacher to self. (Barrows & Tamblyn, 1980) T12. Does not transfer responsibility for learning from teacher.
32


In problem-based learning, students assume responsibility for their learning,
identify what they need to know, and personalize their learning (Bridges &
Hallinger, 1997). Small groups of three to five students engage in collaborative
learning (Dolmans, Schmidt, & Gijselaers, 1994a). Students also engage with
problem solving through self-study and develop their expertise in specific subject
matter. Teachers assume the role of tutors and ask guiding, metacognitive questions
of students (Stepien & Gallagher, 1993; Wilkerson, 1996).
The anchoring problems in problem-based learning are real-life scenarios
that serve as the stimulus for acquiring new knowledge (Bridges & Hallinger, 1997).
The problems tell engaging stories in settings to which the students can relate. They
are open-ended enough to challenge students, engender controversy, and require
students to demonstrate thinking skills beyond simple knowledge and
comprehension (Allen, Duch, & Groh, 1996). These problems present a dilemma, or
puzzle, that will elicit different responses from different learners (Kaufman, 1998).
Problems should be authentic, engaging, and ill-structured (Gijselaers, 1996;
Stinson & Milter, 1996).
The difference between ill-structured problems and case studies commonly
used in business curricula is that ill-structured problems do not have ready-made
solutions, or answers, as case studies often have (Stinson & Milter, 1996). The
function of the problem in problem-based learning is to give students a focus for
integrating their learning from many disciplines (Barrows, 1996). Basing learning
33


around problem solving also requires students to identify sub-problems and
challenges throughout the process, as Getzels (1985) notes:
And perhaps most important, instruction in finding and
formulating problems is well worth educational attention not only
as incidental to problem solving, which is typically what has been
done in school if it was done at all, but as an important focus of
interest in its own right, (p. 60)
The basic sequence for problem-based learning begins with students
identifying the educational, or learning, objectives for the project to focus their
learning; tutors also define learning objectives but may or may not disclose them to
the students (Barrows, 1985, 1986, 1988). A common theme in problem-centered
instruction is that the teacher identifies the learning objectives, sometimes with
input from students, and creates the problem to facilitate development (Woods,
1996). Next, students interact with the assignment in group discussions and self-
directed study. They do not hear lectures about the topic or read textbooks their
teachers assign for background information. As a result of exploring the nature of
the problem at hand, students note the questions that they cannot solve with their
prior, or current, knowledge bases. They identify the resources that can fill in their
information gaps, much as a doctor requests further testing to make an informed
diagnosis or a writer pursues further research to write an article.
The next step is to apply the information that they acquired in group
discussions and self-study to the original problem to determine whether the solution
is viable. Important steps come late in the problem-based learning process, such as
34


application of the solution to the problem, but they are perhaps the most overlooked
steps: students review, synthesize, and evaluate their learning process and products.
Reflection throughout the problem-based learning process enhances it considerably
(Woods, 1994). Problem solving, then, is a cognitive process to derive the best
answer to an unknown, the problem, which is something the learner has not
encountered earlier (Woods, 1996).
A problem-based learning teacher is known as a tutor (Barrows & Tamblyn,
1980). Instead of directing students to recite the correct answer to a problem, tutors
coach and guide students through repeated practices in reasoning and self-directed
study, improving their self-assessment skills in the process (Edens, 2000). Tutors do
not function as ultimate knowledge sources depositing knowledge into students
heads (Freire, 1990). While tutors may use more direct instruction in a lesson
initially, to help scaffold the reasoning processes, eventually they withdraw as
students learn to assume responsibility for their own learning (Stepien & Gallagher,
1993/ A demanding role for a tutor is to find a balance between allowing students
to explore issues and guiding them to discover critical learning issues (Ayers, 2002;
Gijselaers, 1996). Tutors are generally experts in their area of study; similarly, they
must also be experts in the problem-based learning tutor role.
Although the role of the tutor is clear, researchers do not fully understand
tutors conceptions of learning that results from problem-based learning (Dolmans,
Schmidt, & Gijselaers, 1994b; Dolmans, Gijselaers, Moust, Grave, & Van Der
35


Vleuten, 2002). Lack of this literature engenders a need to understand tutors
impressions of teaching and learning in the classroom. Tutors, and all teachers,
would benefit from knowing how their metacognitive questions affect students
learning (Barrows, 1988).
Educational Outcomes and Advantages
Educational outcomes supported by problem-based learning include
acquisition of an extensive knowledge base, development of an effective reasoning
process, advancement of effective self-directed learning skills, increased motivation
for learning, early immersion into a learning culture, structured knowledge for use
in clinical contexts, and development of team skills (Bridges & Hallinger, 1997).
Educational advantages include logical reasoning, discovery of personal educational
needs relevant to future related tasks, learning information in the context of
understanding a problem, reinforcing information with reuse, integrating
information from many disciplines, learning and evaluating self-evaluation, and
self-directed learning skills (Barrows & Tamblyn, 1980).
One of the most important outcomes of problem-based learning is acquiring
an extensive knowledge base (Dolmans & Schmidt, 2000). Developing a reasoning
process should be conscious so that students can be aware of how they reason
through a problem and transfer it to future learning situations (Bridges & Hallinger,
1997). The development of self-directed learning skills, another educational
objective of problem-based learning, is vital for learners, as self-directed learners
36


assume responsibility for their learning, are able to identify their own learning
needs, and know how to acquire and assess them (Candy, 1991). Improving
communication and interpersonal skills are lifelong skills for adults to foster, as is
kindling an internal motivation for learning throughout life (Barrows, 1996; Bruhn,
1997). Cultivating team skills, another educational objective of problem-based
learning, encourages students to learn in small groups and helps them to generate
and discover ideas that they might not have encountered on their own (Barrows,
1985; Bridges & Hallinger, 1987; Wilkerson, 1996; Woods, 1994).
Educational advantages of problem-based learning include student
responsibility for learning, acquisition of interdisciplinary content, instructor
involvement as mentor and coach, and participation of students in the evaluative
process (Allen & Rooney, 1998; Rangachari, 1996). Problem-based learners have
shown that they retain their knowledge, know how to apply it appropriately,
demonstrate mastery of leadership skills, invest significant effort while learning, and
view their preparation as meaningful and enjoyable (Bridges & Hallinger, 1997).
Students activate their prior knowledge in problem-based learning, create a context
that resembles the situation in which students later will use their knowledge, and
elaborate their newly acquired knowledge at the time it is initially learned (Bridges
& Hallinger). Problem-based learners also appear to have broader social and
interpersonal skills, a greater appreciation of the complexity of problems and the
resources available for solution, and a heightened motivation for continued self-
37


learning (Bruhn, 1997, p. 67). Such learners also tend to become more independent
as a result of the experience they gain in thinking about problems (Seltzer, Hilber,
Maceli, Robinson, & Schwartz, 1996). Further, students learning sharpens when
they experience some type of failure and reflect on how and why that happened, a
common occurrence in problem-based learning (Bridges & Hallinger, 1996, 1997).
Central purposes of problem-based learning are to foster critical thinking,
develop students confidence and skill in dealing successfully with unanticipated
issues under practical constraints, and transfer responsibility for learning from the
teacher to the student (Kaufman, 1998). Problem-based learning provides a method
of inquiry which characterizes research and supports the idea that teaching and
learning are realized in a research atmosphere (Birch, 1986). In this type of learning,
lectures become discussions where students actively engage in learning, class focus
shifts from facts to meaningful information that could help solve problems, and
teachers become resources for processes rather than sources of answers to be
memorized in order to pass tests (Aspy, Aspy, & Quinby, 1993; Bridges &
Hallinger, 1997; Bruffee, 1993). Problem-based learning also promotes transfer of
learning between disciplines (Bridges & Hallinger, 1997).
Barriers
Although the educational outcomes and advantages of problem-based
learning seem evident, research does not definitively confirm evidence to support it.
Albanese and Mitchell (1993) for example, conducted a meta-analysis of problem-
38


based learning environments for medical curricula. They suggested that problem-
based learning is a valuable approach to learning, but that evidence of its strengths
is not conclusive. They strongly suggested that it be continuously studied and its
benefits analyzed to fully understand problem-based learning.
Problem-based learning, however, can challenge and inhibit tutors and
students in many ways. Problem-based learning projects seldom run smoothly
(Bridges & Hallinger, 1996, p. 58). One of the primary problems with gaining an
accurate assessment of the degree of success problem-based learning enjoys is that
no overall format or scheme delineates points at which data should be gathered, the
type of data needed, and how it should be analyzed (Bruhn, 1997, p. 68). Also,
students frequently ask for more structure and guidance than is typically given in a
problem-based learning environment. They often complain of vague assignments
because they are used to direct instruction from their primary and secondary
teachers. Students new to problem-based learning are typically uncomfortable with
their tutors encouragement to discover learning strategies on their own (Edens,
2000).
A common impediment to problem-based learning has been academics
conservatism, whose tendency is to avoid interactive methods of teaching and
learning and student self-assessments. These instructors prefer to do research to
further their professional advancements, not to learn how to engage students in
unstructured, spontaneous learning (Birch, 1986). In successful problem-based
39


learning, curricula offer interdisciplinary content similar to more conservative
approaches. The difference is that problem-based learning students collaborate as
they work on their lessons and their instructors act as mentors and coaches (Allen &
Rooney, 1998).
Collaborative skills are sometimes underdeveloped in problem-based
learners and can hinder learning, another barrier to address. Woods (1994) stresses
the importance of collaboration in learning: We learn more effectively and
efficiently if we work actively and cooperatively to learn knowledge. Yet, working
in groups is hard work. For groups to be effective, we need to be skilled in
interpersonal relations and in group process (pp. 4-3). Problem-based learning
participants develop effective interpersonal and interaction skills, including
consensual decision-making skills, dialogue and discussion skills, team maintenance
skills, conflict management skills, and team leadership skills (Ayers, 2002).
Fenwick (1998) presents yet another critical perspective toward problem-
based learning and writes, We suggest that a problem-based approach to
professional education is ontologically narrow and epistemologically inconsistent
with the lived nature of professional practice (p. 53). That is, the nature of the
problems in problem-based learning are contrived, according to Fenwick, and,
therefore, inauthentic, contrary to a major prerequisite of problem scenarios. He
continues to explain that framing life in terms of problems locks the world into an
undesirable chaos that needs a redeeming order. Fenwicks bias is averse to
40


focusing learning around chaotic, confused situations, assuming, of course, that the
problem scenarios are indeed disorganized. These adversities and potential barriers,
plus fear of the unknown, scarcity of resources, and lack of student empowerment
may explain why problem-based learning is not more popular than it is today
(Ayers, 2002).
Perceived weaknesses of problem-based learning can be erroneous, however.
For example, Getzels (1985) claims that discovering problems is a creative process,
integral to creative thinking and problem solving. Opponents of problem-based
learning, supporters of Fenwick (1998), could claim that the creative, problem-
discovery phase is lost in problem-based learning because the teacher supplies the
problem for the students. However, learning is complex and incorporates many
levels of sub-problems that students can discover while working through the
problem-solving process. Getzels explains that problem finding and problem solving
meld into each other: At the core of an effective solution is a productive problem.
It is this that makes what I have referred to as the problem of the problem such an
important subject for inquiry and instruction in our schools (Getzels, 1985, pp. 60-
61). In short, discovering and solving problems is inherent in the learning process
and demands attention in school. Problem solving requires an integrated use of
thinking skills and an appropriate knowledge or data base (Ruggiero, 1988, p. 23).
Students who are focused on discovering and solving their own learning needs are
self-directed learners and engaged in the problem-solving process.
41


Self-Directed Learning
Self-directed learning is an educational outcome of problem-based learning
(Barrows, 1994) and merits individual attention because self-directed learners tend
to be critical thinkers, characteristics that schools aim to develop in students:
1 also think that every act of teaching should have built into it some
provision for helping the learner become more self-directing. I
recognize that there are situations in which a person is indeed
dependent in some respects and that in these situations it is
appropriate for him or her to be taught or directed. But 1 dont
think that it is healthyor even humanefor a person to be kept
permanently dependent upon a system or upon another person.
(Knowles. 1975, pp. 10-11)
As illustrated in Table 2-3, several things are known about self-directed
learning: (a) individuals can become empowered to assume responsibility for
decisions about their learning, (b) self-direction is a continuum of characteristics
that exists in all people, (c) self-directed learning does not imply learning in
isolation, (d) self-directed learners tend to transfer learning between situations, and
(e) self-directed study can involve many activities and resources, and roles for
instructors in self-directed learning environments are as knowledgeable resources
(Hiemstra, 1998). Self-directed learning, then, is a process in which the learner feels
empowered enough to assume the responsibility to control her or his learning by
making decisions involving the activities and resources that will promote
constructive learning (Mullen, 1992). Individuals are more likely to learn when they
are able to respond to an event with self-directed thinking (Morris & Moore, 2000).
42


Table 2-3
Continuum of Learners Self-Directed Learning Skills
Focus - 1 1-1 i n Self-Directed Learning Other-Directed Learning
Student Attributes
1. Self-reliance SI. Self-reliant in finding information. (Hiemstra, 1998) Ol. Needs help finding information.
2. Personal goals S2. Sets personal learning goals. (Garger, 1999) 02. Does not set personal learning goals.
3. Learning in isolation S3. Enjoys learning in isolation. (Hiemstra, 1998) 03. Does not enjoy learning in isolation.
4. Time for learning S4. Finds time for learning. (Licklider, 1997) 04. Does not find time for learning.
5. Feedback and criticism S5. Accepts feedback and criticism. (Anfield, 1997) 05. Does not accept feedback and criticism.
6. Learning control S6. Takes control of personal learning. (Brookfield, 1993) 06. Does not take control of personal learning.
7. Seeking learning S7. Seeks out learning opportunities. (Gerber, Lankshear, S vensson, 1995) 07. Does not seek out learning opportunities.
8. Problem- solving S8. Enjoys problem solving. (Hiemstra, 1997) 08. Does not enjoy problem solving.
9. Challenging assumptions S9. Challenges assumptions. (Pilling-Cormick, 1997) 09. Does not challenge assumptions.
10. Responsibility for learning S10. Assumes responsibility for own learning. (Knowles, 1975) 010. Does not assume responsibility for own learning.
11. Self-evaluation SI 1. Evaluates and advocates own learning. (Pilling- Cormick. 1997) Oil. Does not evaluate or advocate own learning.
12. Self-reflection S12. Reflects upon and makes choices about own learning. (Candy, 1991) 012. Does not reflect upon or make choices about own learning.
43


Self-directed learners can select an appropriate learning style from a range of
styles according to the demand of the situation and their own learning capability
(Robotham, 1995). These learners fall on a linear scale that represents preferences
toward solving problems independently or solving problems within a group setting
(Hiemstra, 1997; Hiemstra& Brockett, 1991); indicators on the self-directedness
scale include self-reliance in finding information, setting personal learning goals,
enjoying learning in isolation, structuring personal learning, finding time for
learning, accepting feedback and criticism, taking control of persona! learning,
setting personal learning approaches, seeking out learning opportunities, and
enjoying solving learning problems in isolation. It is the desire, attitudes, values,
and abilities that will ultimately determine the degree of self learning that will take
place (Guglielmino & Murdick, 1997, p. 11).
Learner control is a central aspect of self-directedness (Brookfield, 1993;
Lowry, 1989; Tremblay, 1992). Self-directed learners typically challenge
assumptions their teachers make about their learning (Pilling-Cormick, 1997).
Giving responsibility to learners has shown to benefit those learners (Guglielmino &
Guglielmino, 2001; Hatcher, 1997; Hiemstra, 1998). When adults learn how to
assume responsibility and empower themselves for their own sake, their self-
directedness also promotes their self-advocacy, self-determination, and
metacognitive learning (Hiemstra & Brockett, 1991). Self-direction in learning is a
44


combination of forces both within and outside the individual that stress the learner
accepting ever-increasing responsibility for decisions associated with the learning
process (Brockett & Hiemstra, p. 9). The role of the teacher in self-directed
learning environments shifts from instructor, to facilitator, to learning resource as
the locus of responsibility for the learning shifts from the instructor to the learner
(Robotham, 1995). This allows the learner to retain the freedom to explore and
develop personally within an environment where the teacher provides a minimal
amount of structure.
Independent thinking is hardly likely to occur when students are
told precisely what they should learn, and what they should think
about it. And students are also not likely to improve as independent
thinkers when they see teachers themselves being told precisely
what they should teach, and how they should teach it. (Smith,
1990, p. 128)
As self-directed learners reflect on their learning, they determine,
investigate, and evaluate their needs (Pilling-Cormick, 1997). When learners reflect
and make choices about their learning, they demonstrate openness and willingness
to consider learning approaches from different perspectives (Gerber, Lankshear, &
Svensson, 1995). Reflection is an important aspect of the self-directed learning
process because it increases awareness of how cultural factors can influence the
pursuit of exemplary learning projects (Brookfield, 1993).
Barriers to self-directed learning also include challenges that many learners
face because their school teachers may have conditioned them to be other-directed
learners (Hiemstra, 1998). Other-directed learning is teacher-centered learning, in
45


that the teacher decides what, how, and when students will learn. Initial resistance to
self-directed learning is common among learners until they have a successful
experience with it (Hatcher, 1997). Assuming control of learning, however, brings
the learner into direct conflict with entrenched issues; as learners decide their own
learning plans, they may clash with authoritative voices that had been responsible
for deciding what and when they learn (Brookfield, 1993). Critics of self-directed
learning believe that it is a way for instructors to be less accountable for their
teaching because it is a diluted form of education (Hiemstra, 1998). However, these
critics evidently do not recognize self-directed learning as a significant contribution
to the advancement of adult education (Brookfield, 1993).
Assessing Problem-Based and Self-Directed Learning
Tools and techniques for assessing problem-based and self-directed learning
include integrative essays in which students discuss what they have learned during a
project and how they might use the knowledge and skills in the future, protocols or
standards that students may use to evaluate their own performance or products, and
models or examples of products completed by expert practitioners against which to
compare their own products. They also include knowledge review exercises that test
students ability to apply their knowledge to typical situations, forms created by
students to elicit feedback from their peers on aspects of their performance,
structured observations that provide descriptive information about individual and
46


group performance, and probing questions for students to consider in relation to
their final performance products (Bridges & Hallinger, 1996).
Assuming responsibility in learning includes self-assessment of learning
needs (Thomas & Harper, 2001). Adults prefer to assume responsibility for their
own learning rather than be at the mercy of someone else deciding what, how, and
when they should learn (Hiemstra. 1998 ). In their research involving continuing
professional education of certified public accountants (CPAs), Thomas and Harper
claim that self-directed learning activities should stem from learning objectives
based on self-assessments of learning needs. This way, CPAs will be able to self-
assess the impact of these activities based on professional competencies developed
by industry professionals.
Woods (1996) studied his classes of chemical engineering students learning
in a problem-based learning environment and declared its success. Over the past
thirteen years, we have evaluated our program. On virtually all measures, our
students show a marked superiority' in the skills that are in demand in the real
worldstrong support for what started as only a small piece of a larger pie
(Woods, p. 98). Woods also claims that students can self-assess accurately if
teachers empower them regularly.
An example of employees documenting their self-directed learning,
evaluating it, and getting credit for it is Motorolas self-directed learning lab
(Filipczak & Ganzel, 1999). One method Motorola used was to compare learning
47


from using self-directed learning methods with instructor-led methods. Another
strategy to measure self-directed learning is with manager-employee interviews and
documentation analyzing whether the amount of learning is increased by self-
directed learning for the same or less cost to Motorola (Guglielmino & Murdick,
1997).
Fenwick (1998) is as skeptical about evaluating problem-based learning as
he is about the quality of the problems: "Some critical evaluation of problem-based
learning has emerged, but this tends to focus on student outcomes, conceptualized in
terms of the amount of knowledge residue generated by problem-based learning
compared to lecture approaches'' (p. 53). If evaluation is equated only with research
and measurement, Fenwicks biased concerns are well-founded; if not, his concerns
are shallow.
Schmidt and Dolmans (1995) corroborate Fenwicks sentiments and explain
that outcome data does not provide the quality of information that can judge a small-
scale, course-level curriculum:
Thus what is needed, in addition to outcome studies as an
instrument for accountability, is a strategy that deals with the
curriculum in its own right, provides information about its
shortcomings, and enables a school to implement improvements
whenever and wherever necessary, (p. 82)
Schmidt and Dolmans evidently see evaluation as more than objective observation
and measurement; they see it as curriculum evaluation, in which the whole
48


curriculum receives evaluation, not just the learning outcomes of individual
students.
Part of the challenge in evaluating problem-based learning is that formal
assessment procedures are designed to evaluate traditional instructor-led learning.
The process and intent of problem-based learning are different, however, from
instructor-led learning and require different means of evaluation. According to
Barrows (1996), not many formal assessment procedures can address problem-based
learnings idiosyncrasies because they are generally insensitive to the cognitive and
behavioral differences observed in problem-based learning. Assessments of the
effectiveness of PBL are generally favorable, yet its cost in money and time, as well
as its effect on the future behavior of practitioners, has not been established
(Bruhn, 1997, p. 66).
A challenge in evaluating self-directed learning is that it involves the
learners life experiences, including learning plans, self-directed learning teams,
experience-based learning, competency-based assessment and education, and past
expertise (Boyatiz, Cowen, & Kolb, 1995). The value of experiences is difficult, if
not impossible, to measure. Educators should acknowledge such an array of
contributors to the overall learning process, but to devise a dependable, rational
evaluation plan to do so is challenging. These life experiences can also include
dependency of the learner on the instructor to provide learning resources, identify
learning strategies, and evaluate learning progress (Robotham, 1995). The instructor
49


then carries the responsibility to be an expert in the field, but not to overpower
already anxious learners (Robotham).
Evaluation of problem-based and self-directed learning occurs within the
learning process itself, making it a formative assessment for learners (Barrows,
1986). Part of the evaluation process involved in problem-based learning is that
teachers should be personally involved in the problem-based process. Firsthand
involvement allows for authentic assessments to understand the issues involved in
problem and reasoning skills (Edens, 2000). Evaluating the outcomes of problem-
based learning should also include an assessment of the observed interaction of
students and their instructors (Bruhn, 1997).
To assess problem-based learning fully, Barrows (1996) claims that teachers
should observe problem-based learning in action. Problem-based learning students
use an iterative process of assessing what they know, identifying what they need to
know, gathering information, and collaborating on the evaluation of hypotheses in
light of the data they have collected (Stepien & Gallagher, 1993). As students reflect
on their learning processes, se f-directed and problem-based learners move beyond
superficial questioning into the realm of critical learning (Pilling-Cormick, 1997). In
the development of identifying these needs, learners can also devise appropriate,
relevant, intellectual assessment strategies.
50


Teaching Writing
Writing is a complicated intellectual process that involves a variety of
problem-solving strategies (Flower & Hayes, 1988; Hillocks, 1995). Within this
problem-solving process lies a progression of discoveries for students that reflects a
unique mode of learning (Emig, 1977; Perl, 1988). Writing promotes successful
learning because it connects and integrates ideas from many perspectives while
students can be active and personally engaged (Emig, 1977). Accordingly, writers
draw from a repertoire of strategies for building meaning and learning for
themselves (Flower, 1994c).
Teaching writing involves teaching thinking, problem-solving, and self-
directed learning skills simultaneously with composition skills because they are
each interrelated and interdependent (Coe, 1974; Flower, 1994c; Rohman, 1965).
Good thinking can produce good writing, but without it, good writing is impossible;
bad thinking never leads to good writing (Rohman, 1965). However, as writing
students encounter new ideas and ways of thinking in their college writing courses,
their rhetorical training (understanding knowledge communication between writer,
audience, and purpose) proves incapable of dealing with their new intellectual
experience (Berlin, 1987). Writing students need to develop their problem-solving,
self-directed learning, and critical thinking skills to write well about meaningful
topics and to develop lifelong skills of thinking critically to solve whatever
problems they may face. Teachers can and should provide guidance for students, but
51


they should make it clear to students that correctness is not as important as having
the ability to perceive a problem and unravel it (Meyers, 1986, p. 31).
Composition studies, or teaching writing, addresses the role of problem
solving in many ways. A problem is simply any situation in which you are at point
A and need to find some way to get to your goal, point B (Flower, 1989, p. 44). For
writers, this can be moving from a list of potential writing topics to a focused thesis,
a rough draft to an edited draft, or fifty pages to ten pages of content. One of the
most basic reasons for writing, which students, academic writers, journalists, and
business people share, is to discuss and deal with problems (p. 135).
Expert and novice writers often use different problem-solving strategies and
the problem-solving approach assumes that there is often a better way and that
writers can substantially expand their repertory of strategies (Flower, 1989, p. 3).
For example, a novice writer might approach topic development by listing ten ideas
that interest him and choosing one. The expert writer might brainstorm a list,
discuss the items with a friend, revise the list, write for a while, and then choose an
idea that he has already analyzed and discussed. The theoretical framework for
problem solving in teaching writing includes rhetorical theory, the concept that
writing consists of a process that always involves the writer, reality, reader, and
language. The following section reviews composition literature and the
manifestation of problem solving in teaching writing.
52


Everyday Necessities
Problems are inherent in everyday life, making the process of finding their
solutions an everyday necessity: In teaching writing, we are not simply offering
training in a useful technical skill that is meant as a simple complement to the more
important studies of other areas. We are teaching a way of experiencing the world, a
way of ordering and making sense of it (Berlin, 1988, p. 58). Because the writing
process itself is a thinking process (Flower & Hayes, 1988), it has much in common
with common problem-solving processes, such as taking exams, understanding a
patients illness, or completing a tax form correctly (Flower, 1989). Problem-
solving is clearly an everyday necessity.
Yet another reason that problem solving is an everyday necessity is because
it is a response to the uncertainty inherent in most rhetorical situations:
communications with ambiguities in the writers intentions, readers capabilities,
and the basic message can obscure the effect of the communication. Writers
acquire problem-solving strategies from social interaction, but academic writing is
still a set of heuristics, not an algorithm or formula (Flower, 1994a, p. 23; Bruffee,
1993). Problem solving is integral to our everyday activities:
Solving problems is essentially no different from anything else the
brain does. Problem-solving cannot be separated from decision-
making, categorizing, classifying, inferring, and other aspect of
thinking, often listed as distinct skills. Problem-solving can be seen
to involve, and be involved in. all these aspects of thought. They
are not distinctive elements but simply ways of looking at, or
53


talking about, one coherent thing, the brains construction of
realities, the imposition of order on chaos. (Smith, 1990, p. 19)
Goal-Directed Thinking
People turn to problem solving when they have a problem or a goal that they
want to understand in the form of goal-directed thinking (Flower, 1989). For some
writers, problem solving transforms the writing process into a goal-directed
journey intent on individual writing purposes (Flower, p. 4). One such situation
might be a writer composing a letter to the local newspaper editor because of a
strong belief about an issue. Problems exist when people feel a conflict between
where they are and where they want to be or between the present state and the
desired state (Flower); if the goal is attainable with minimal effort, the problem is
also minimal.
Flower and Hayes (1988) conducted a study in which they wanted to know
how writers define problems for themselves. They also wanted to know how writers
analyze their audience, if they do at all, if they develop a full representation of their
problem and if it helps them generate new ideas. Another objective was to
understand if good and poor writers go about this task differently. They found that
good writers respond to all aspects of the rhetorical problem, or writing task. These
writers build a unique representation of their audience, assignment, and goals
involving the audience. They also found that good writers create a rich network of
goals to affect their readers and these goals helped the writers generate new ideas.
54


Another finding was that good writers represent the problem in more breadth and
depth than novice writers, they continue to develop their image of the reader, and
they detail their goals as they write. The main conclusion of our study is this: good
writers are simply solving a different problem than poor writers (p. 101). Flower
and Hayes' conclusions coincide with Getzeis' (1985) claims that discovering
problems is a creative process, integral to creative thinking and problem solving.
They conclude with the following:
This important study of creativity in fine art suggested that
problem-finding is a talent, a cognitive skill which can lead to
creativity. The parallels between these two studies suggest that
problem-finding in both literature and art is related not only to
success, but in some less well defined way to creativity itself, (p.
101)
Other implications of the Flower and Hayes study are that teachers can successfully
teach student writers how to explore a rhetorical problem, that problem
comprehension is not a mysterious act. Teaching students to explore their own
problems freely can help them to create inspiration instead of wait for it (p. 102).
Flowers (1989) research points out that when writers think aloud, they
mention their goals freely. Typically, writers have four main goals: how they want
to affect a reader, the relationship that they want with their reader, their attempt to
build a coherent network of ideas and meaning, and the conventional features of a
written text (Flower & Hayes, 1988, pp. 98-99). Flower has heard student writers
give themselves instructions and monitor their writing progress and effort in think-
55


aloud research. She also explains that writers are absorbed in reaching their goals,
just like all problem solvers. If the goal is developing a topic, for instance, once the
writer has generated an outline of main and subtopics, he or she does not typically
remember the process that generated the outline. Further research with writers' think
alouds should reveal more insight into the process that develops their goal-directed
thinking.
Reflective Learning
The writing process entails many creatively cognitive steps and strategies,
just as creating a painting or composing a song requires. Flower (1989) has
attempted to understand the cognition behind writers strategies so teachers can
develop instructional strategies to help students to write well. Thus, my experience
with this methodology has convinced me that personal learning through self-
reflective epistemic writing is as powerful a learning tool as we may have in our
repertoire as college teachers (Brown. 1998, p. 138). Flower explores this question
of what actually happens when we write and offers several conclusions. She claims
that the writing process, because it is a thinking process, involves deep thinking to
the point that sometimes, after writing for hours, writers have no recollection of
what they did during the time. This is common, even though the two hours were
spent actively planning, setting goads, generating new ideas, drawing inferences
from old ideas, looking for relationships or patterns, creating trial text, evaluating
our prose, detecting errors, and diagnosing problems and planning ways around
56


them (Flower, pp. 31-32). Flower has also identified steps in the composing
process and a variety of strategies for taking these steps:
Planning
Step 1: Explore the rhetorical problem
Step 2: Make a plan
Generating ideas in words
Step 3: Generate new ideas
Step 4: Organize your ideas
Designing for a reader
Step 5: Know the needs of your reader
Step 6: Transform writer-based prose into reader-based prose
Revising for effectiveness
Step 7: Review your paper and your purpose
Step 8: Test and edit your writing
Step 9: Edit for connections and coherence (Flower, 1989, p. 51)
Flower (1989) describes the steps in analyzing a problem, which are
reminiscent of steps in the composing process. First, the writer defines the conflict
or key issues. This is different from merely stating a topic, such as pollution, in
that it is stating the problem within the topic, such as corporations should take an
active role in reducing pollution in this country. The second step in analyzing a
problem is to place the problem in a larger context, such as understanding the
effects of pollution on human well-being. The third step is to operationalize the
problem so that the writer can tackle the problem on a manageable scale: Ninety
percent of manufacturing corporations in this country should decrease their airborne
pollutants by 75 percent in the next 24 months.
57


The next step in the writers process of analyzing a problem is to explore
subparts of the problem to understand it thoroughly, such as the number of
manufacturing companies that are polluting the air and at what rates, the progress
already made in pollution over the last three years, legislation that has gone into
effect, particular health risks of pollution, and the like. The fifth and final step in
problem analysis is to come to an open-minded conclusion. This is when the writer
comes to a solution or a new definition of the problem; the weakness of most
problem solving is leaping too quickly to a solution (Dewey, 1910). A rhetorical
problem can be a vast situation:
[It] is like a rather large, uncharted territory that contains you, your
reader, your ideas, and all the things you possibly could do. As you
make a mental tour of this territory selecting the goals that matter
and the strategies you will use, you are building your own image or
representation of the task. (Flower, 1989, p. 76)
Because of the enormity of the rhetorical problems that writers can face, the five-
step process Flower outlines offers a logical approach, albeit flexible.
Reflection plays a key role in the problem-solving and writing process (Perl,
1988). Writing is a special form of reflection through which new meaning can be
created, new understanding of problems can become circumscribed, and new ways
of organizing experiences can be developed (Bringle & Flatcher, 1999, p. 180).
When people read and write, they are not only constructing meaning, they are
observing and reflecting using a repertoire of strategies for building meaning
(Bruffee, 1993; Flower, 1994b). Writers also reflect on their own goals,
58


assumptions, strategies, and habits of making meaning (Courtney & Abodeeb, 1999;
Flower, 1994a). Collaborative planning is a specific strategy that invites student
writers to reflect on what happened in their planning sessions and to consider what
they are learning about themselves and thinkers, problem-solvers, and writers
(Wallace, 1994; Yost, 2000).
Discovery Process
In their research, Flower and Hayes (1988) found that discovery and the new
insights it creates are the end result of a complicated intellectual process. The act of
creating and discovering ideas is the essence of effective writing (Flower & Hayes).
Finding ideas for writing topics is akin to reporting research findings, such as
reporting the number of factories that are violating pollution standards. Discovering
ideas, on the other hand, is synthesizing findings, such as explaining the number of
polluting factories combined based on their locations and the type of product they
manufacture to create a new insight into the pollution problem. Problem solving is a
form of discovery, similar to writing and thinking processes. For some writers,
writing about problems can change their perceptions about the problem, or their
definition of the problem because it opens new doors of understanding (Flower,
1989).
If you can define the problem, you may have solved it. The hardest
part of solving many problems is trying to discover what the
problem really is and define the conflict that makes it a problem.
Writing is a powerful way to think problems through, because it
59


helps you describe and name the conflicting parts of your own
thinking, (p. 31)
Bruner (1961) also saw discovery as an intellectual process and defines it as
all forms of obtaining knowledge for oneself by the use of ones own mind (p.
237). He explains that discovery is best accomplished by those learners who are
prepared to discover new knowledge and emphasizes that discovery is a process of
transforming evidence (p. 237) that allows the learner to create new insights.
Bruner also hypothesized that discovery enriches learners problem-solving skills
because it allows people to acquire information viably.
Problem Solving as Collaborative Planning
Collaborative planning is an integral part of the writing process because it
involves writing workshops and giving and receiving feedback on writing. It is
socially supported conversation and an opportunity for student writers to talk about
their ideas in a socially supportive environment where peers listen and ask critical
questions (Wallace, 1994, p. 50). Bruffee (1993) concurs:
In collaborative learning students work on focused but open-ended
tasks. They discuss issues in small consensus groups, plan and
carry out long-term projects in research teams, tutor one another,
analyze and work problems together, puzzle out difficult lab
instructions together, read aloud to one another what they have
written, and help one another edit and revise research reports and
term papers, (p. 1)
One of the strategies that expert writers use is planning; trial and error tactics belong
more to novice writers (Flower, 1989). Likewise, problem-solvers rely on planning
60


because it saves them significant time in the writing process. Collaborative planning
is more of a process than a writing strategy whereby students can reflect on their
thinking (Flower, 1994b).
Teachers set the tasks that writing students tackle in collaborative planning.
This requires forethought, practice, and planning on the teachers part (Bruffee,
1993, p. 35). Tasks may be questions to be answered by the consensus of the group.
For example, articles could be given to small groups of writers for them to decide on
the argument and major points of support. Another task for collaborative planning
could be problems to be solved so that all group members are satisfied, such as
writing a summary of an article. Teachers do not tell students if they are right or
wrong because right and wrong answers do not exist as such (Bruffee). The
teachers role is to give students opportunities to work on specific, open-ended
activities collaboratively. This contradicts the traditional role of the teacher as sole
source of knowledge and can threaten the teachers authority. Accordingly, most
writing students tend not to trust collaborative activities and, therefore, do not
benefit from its learning opportunities (Bruffee). This could also be an explanation
of why peer critiquing in a writing class tends to be superficial and strained.
Another challenge with collaborative planning is that many student writers
in college struggle with knowing how to plan; they have been conditioned in school
to focus on story telling, a form of knowledge telling, and gathering information
(Wallace, 1994). Also, student writers need to know how to collaborate, to be active
61


listeners, and ask for elaboration with probing questions (Bruffee, 1993; Wallace,
1994). Inevitably, however, multiple perspectives in a discussion stir conflict.
Consequently, writing students present alternative solutions and strategies to each
other and clarify their criteria for evaluating their solutions to the problem at hand
and elucidate their grounds for supporting their rationale behind the solution
(Bruffee, 1993).
Collaborative learning brings to the surface the relationship
between the authority of knowledge and the authority of teachers.
By challenging the traditional, foundational understanding of the
authority of knowledge, collaborative learning helps college and
university teachers begin thinking in quite a different way about
what it means to teach, (p. 7)
In addition to reviewing the literature about problem solving in composition
studies, teachers of writing can benefit from understanding the evolution of
rhetorical theory and how the new rhetoric aligns with problem-based learning and
self-directed learning (Lauer, 2002). Writing teachers should also accept the
responsibility to understand basic cognitive structures and how they develop in
order to provide productive learning experiences for students (Berlin, 1987; Brooks
& Brooks, 1999; Hillocks, 1995). This literature review of composition studies
continues with a review of the literature of rhetorical theory.
Rhetorical Theory
The writing process involves the writer, reality, reader, and language; these
elements are also known as rhetorics and rhetorical theories are the principles that
62


attempt to explain how these elements interact (Berlin, 1987). See Table 2-4. The
difference between rhetorical theories lies in which of these elements are
emphasized in the classroom based on how the teacher and students conceive them
(Berlin). Tne rhetorical theories that shape composition studies are loosely
categorized as objective or subjective. Objective rhetorics maintain that rhetorical
elements are scientific, observable, and measurable. Subjective rhetorics are
grounded in the epistemic and expressionist movements that build on the concept
that reality in writing is personal, private, and intellectual.
Objective rhetorics are positivistic theories that locate reality in the material
world (Berlin, 1987, p. 139). Based on Skinners (1965) stimulus-response theory,
reinforced behaviors in the writing process will be repeated by writers, making
writing an observed behavior. For example, grammar workbooks are commonly
used in behavioral rhetoric because the teacher can reward the student for writing
correct answers in the workbook and discipline the writer for being wrong. This
behavior is also evident in weekly spelling, punctuation, and grammar quizzes and
the noxious red pen of many writing teachers. Behaviorists focus on product rather
than process because it is relatively easy to measure the correctness of the product
and difficult to assess the writing process. A weakness in behaviorist rhetorics is
that because the writing process inherently involves complex thinking activities, the
behaviorist tendency to observe, mark, and fix errors erroneously leads student
63


Table 2-4
Continuum of Instructional Strategies in Rhetorical Theories
Focus Epistemic C urrent-Traditional
Instructional Strategies
1. Teaching emphasis E1. Emphasizes process of writing. (Berkenkotter, 1988) CT1. Emphasizes written product.
2. Exemplary writing E2. Stresses connections between good writing and good thinking. (Rohman, 1965) CT2. Stresses connections between good writing and observable behaviors.
3. Theoretical sources E3. Endorses reciprocal relationship between subjective thought and language artistry. (Berlin, 1987) CT3. Endorses reciprocal relationship between objective analysis and grammar.
4. Function of writing E4. Uses writing as heuristic to discover ideas. (Flower. 1989) CT4. Uses writing as tool to report information.
5. Viewpoint concerns E5. Encourages reading different perspectives on the writing topic. (Sommers, 1982) CT5. Encourages exposure to limited points of view in writing topic.
6. Philosophic basis E6. Embraces truth and knowledge as dynamic and dialectical. (Berlin, 1987) CT6. Embraces truth and knowledge as concrete and static.
7. Audience concerns E7. Considers audience as a primary role in writing, influencing content and organization. (Wallace, 1994) CT7. Considers audience as a minor role in writing, second to punctuation and syntax.
writers to believe that mechanical elements in writing are more important than
intellectual development (Berlin).
Epitomizing objective rhetoric is the current-traditionalist theory, which
takes an objectiv e, scientific stance toward rhetoric (Berlin, 1987). This theory
64


dominates teaching writing today, evident in the profuse use of workbooks and
textbooks promoting its positivist tenets (Berlin, 1988). It emphasizes product over
process, emphasizing words, sentences, paragraphs, usage, syntax, spelling, and
punctuation. Current-traditionalists have a strong preoccupation with the informal
essay and research paper (Gere, 2000). The world is rational, according to current-
traditionalists, and its system of knowledge is discovered through the scientific,
experimental method.
The weakness in the current-traditionalist rhetoric is that the intellectuality
of writing poses inherent problems that challenge empirical verification; in short,
intellectual problems require intellectual solutions. Observable behaviors do not
always provide answers to intellectual dilemmas. Current-traditional rhetoric posits
that truth is empirically based and can only be achieved through subverting a part of
the human response to experience (Berlin, 1988). For these reasons, several
pedagogies have rejected the current-traditional rhetoric emphasis on product and
correctness as antidemocratic because it excludes the dominant discourse of creative
expressionism (Schick, 2002). For example, writing teachers who edit student
papers on early drafts and emphasize scores on weekly grammar tests may be
depriving their students of the opportunity to think critically about supporting their
claims with rational support and multiple perspectives.
65


To deprive learners of critical thought processes is similar to controlling
their thinking and conditioning them to obey the word of the teacher who is always
correct. Shortcomings of current-traditionalist rhetoric are further elucidated:
Thinking of writing as social, collaborative, and constructive tells
us a good deal about how college and university teachers (and
textbooks) should be teaching writing and expecting students to
learn it. One implication is that, as much as they might like to,
college and university teachers (and textbook writers) cannot tell
students how to write. Instead, because writing is itself a displaced
form of conversation, teachers have to find ways for students to
learn to engage in constructive conversation with one another
about writing. (Bruffee, 1993, p. 57)
Engaging students in constructive conversation about writing involves thinking
critically about their ideas and constructively criticizing others ideas, attributes of
subjective rhetorics. Meyers (1986) explains another shortcoming in the current-
traditional rhetoric approach to teaching writing:
As teacher? drop the defenses of detachment and authority,
students become more likely to challenge their pronouncements.
This can be threatening to teachers who see 'rightness as the
primary' sign of competence. But when teachers come to regard
encouraging reflection and pressing for informed judgments as
important goals, they will welcome questions and challenges as
signs that students are beginning to trunk on their own. (Myers,
1986, p. 94)
Subjective rhetorics were a reaction to the current-traditionalists and stem
from the expressionist movement prominent in the 1960s and 1970s, which
emphasizes the learners experience, interest, and motivation. Expressivists see truth
as a private vision, contacted through writing, discovered by the individual, and
66


demanding original use of language (Berlin, 1988). This rhetoric also posits that no
one can or should evaluate writing because of its emphasis on unique expression
and self-discovery throughout the writing process, and especially in the prewriting
phases (Bruner, Goodnow, & Austin, 1956; Gere. 2000); this approach is clearly the
antithesis of current-traditionalists who believe all rhetorical elements are
measurable and either correct or incorrect. Expressivists, on the other hand, see
reality as arising out of the interaction of the private vision of the individual and
the language used to express this vision' (Berlin, 1987, p. 146). Writing teachers
with an expressionist approach encourage students to ffeewrite to discover what
they are thinking and write reflections that will not be evaluated by anyone. Elbow
(2000) reflects on the power of his freewriting:
Unfocused exploring is probably my main use of freewriting: I
have a thought, perhaps out of the blue or perhaps in the midst of
writing something (even while writing something else), and I give
myself permission to pursue it on paper in an uncontrolled way
wherever it wants to go, even if it digresses (as it usually does).
This kind of freewriting is precious to me because my mind seems
to work bestat the level of ideas as well as syntaxwhen I allow
it to be uncontrolled and disorganized. 1 cannot find as many ideas
or perceptions if 1 try to stay on one track or be organized. And the
not-stopping seems to build mental momentum, helps me get
wound up or get rolling so that more ideas come. (p. 118)
Possible weaknesses in expressionist rhetoric lie with the anarchist
expressionists who want complete and uninhibited freedom in writing, flouting all
written conventions (Gere, 2000). This approach to teaching writing negates the
importance of adhering to written conventions and grammar that make text
67


intelligible, which is neither acceptable nor feasible when writing to an authentic
audience. An underlying assumption of the subjective rhetorics is enabling
individuals to arrive at self-understanding and self-expression will inevitably lead to
a better social order (Berlin, 1987, p. 155), directly refuting the anarchist
expressionists view of breaching written conventions.
John Deweys (1938) questions illustrate concerns of expressivism in
education well:
Can we find any reason that does not ultimately come down to the
belief that democratic social arrangements promote a better quality
of human experience, one which is more wisely accessible and
enjoyed, than do non-democratic and anti-democratic forms of
social life? Does not the principle of regard for individual freedom
and for decency and kindliness of human relations come back in
the end to the conviction that these things are tributary to a higher
quality of experience on the part of a greater number than are
methods of repression and coercion or force? Is it not the reason
for or preference that we believe that mutual consultation and
convictions reached through persuasion, make possible a better
quality of experience than can otherwise be provided on any wide
scale? (p. 34)
Deweys words emphasize the prominence of social experiences and human
relations in rhetoric, attributes evident in epistemic rhetoric.
Epistemic, or new, rhetoric is another subjective rhetoric. The difference
between epistemic rhetoric and expressivist rhetoric is that epistemic rhetoric sees
language as a social, not private phenomenon; language as meaning-maker holds a
central role (Berlin, 1987): Epistemic rhetoric holds that language is the key to
understanding the dialectical process involved in the rhetorical act (p. 166). The
68


new rhetoric emphasizes that instruction corresponds with reality (Gere, 2000) and
because it posits the role of language to create reality and not relay it, writing is the
act of making meaning in this rhetoric. Therefore, epistemic writing is writing that
aids in thinking, learning, and self-knowledge (Brown, 1998, p. 136). In addition,
new rhetoric considers mechanical conventions in writing to be important because it
contributes to the overall meaning-making process (Berlin, 1988). Writing teachers
who ascribe to epistemic rhetoric typically ask students to discuss their writing in
workshops and conduct individual conferences with students.
Truth and knowledge are dynamic and dialectical, not static, in epistemic
rhetoric (Berlin, 1988; Gere, 2000). This perspective emphasizes the tenet in new
rhetoric, that language, written language, creates meaning based on the writers
cognition. This rhetoric is also referred to as cognitive rhetoric which includes
writers thinking and comprehension strategies. Flower (1994a) explains that
cognitive rhetoric emphasizes metacognitive awareness and the conscious control
of ones options (p. 121). She explains that cognitive rhetoric emphasizes the
learners own processes of observation and reflection and notes that even though the
endorsement of cognitive rhetoric is a matter of degree, it can change what teachers
do and students learn.
Summary
This literature review corroborates a logical, positive synthesis of critical thinking,
problem-based learning, and teaching writing (see Table 2-5). Critical thinking
69


Table 2-5
Synthesis of Critical Thinking, Problem-Based Learning, and Teaching Writing in
Student Attributes
Focus Cognitive Writers Behavioral Writers
Student Attributes
1. Content acquisition C1. Identifies content to be learned. B1. Learns information determined by others.
2. Personalized learning C2. Individualizes learning based on personal interests. B2. Generalizes own learning.
3. Knowledge construction C3. Constructs knowledge actively when learning. B3. Does not knowingly construct knowledge when learning.
4. Problem analysis C4. Analyzes problems. B4. Avoids problems.
5. Problem solving C5. Defines what information is pertinent to solve a problem. B5. Accepts readily available information to solve a problem.
6. Reasoning ggskills C6. Possesses logical reasoning and problem- solving skills. B6. Thinks compulsively and rashly.
7. Information use C7. Reuses information reinforced through use. B7. Discards information after use.
8. Participation activity C8. Participates actively and is motivated when learning. B8. Participates on a passive, low' level of involvement when learning.
9. Group problem solving C9. Encourages group problem solving. B9. Retracts from group problem solving.
10. Solution implement- ation C10. implements solutions to problems solved. BIO. Does not complete problem solution implementation.
11. Prior knowledge C11. Activates own prior knowledge. B11. Does not acknowledge prior knowledge.
12. Learning independence C12. Learns independently. B12. Depends on others to identify content to be learned.
13. Critical thinking C13. Thinks critically. B13. Possesses unreasonable, irrational reasoning skills.
70


Table 2-5 (continued)
Focus Cognitive Writers Behavioral Writers
Student Attributes
14. Learning transfer C14. Transfers learning responsibility ffom the teacher to self. B14. Does not transfer responsibility for learning from teacher.
15. Emphasis of teaching C15. Emphasizes process of discovery. B15. Emphasizes product of correctness, not process.
16. Model writing C16. Connects between good writing and good thinking. B16. Takes positivist approach to observing and measuring good writing.
17. Viewpoint concerns Cl7. Reads authors who have different perspectives on writing topic. B17. Reads limited number of authors and perspectives.
18. Function of writing C18. Believes writing aids cognition because writers use language to discover ideas. B18. Believes writing fits proper formats of research papers and discourse classification modes.
19. Philosophic basis Cl9. Believes truth and knowledge is dynamic and dialectical, not static. B19. Believes truth and knowledge are concrete and definite.
directly entails problem solving and analysis, necessary skills to be taught in college
and university (Meyers, 1986). Educated people must be autonomous enough to
make important decisions, respect the rights of others, seek understanding of factual
information, and remain open-minded to others perspectives; these are the basic
characteristics of critical thinkers (Norris & Ennis, 1989). These are also the same
characteristics of an epistemic rhetorical approach to teaching writing in a problem-
based learning environment. McPeck (1990) emphasizes the necessity of self-
directed learning:
71


True critical or autonomous thinking is, by definition, doing ones
own thinking; therefore, students must be cut loose from their
dependency on authority. Yet teachers in a classroom are in a de
facto position of authority, and this fact has a deleterious effect
upon the free and open exchange of ideas, (p. 52)
Meyers (1986) also emphasizes the natural fit between teaching writing and
teaching critical thinking by pointing out that written assignments reveal what and
how students are thinking. This insight allows writing teachers to nurture new
modes of critical thinking for student writers.
Enriching what writing teachers do and students learn is the impetus of this
study. A problem-centered instructional orientation to teaching writing makes sense
because it is aligned with epistemic rhetoric that promotes cognition, personal voice,
reflection, goal-setting, and discovery. These attributes empower the writer as a
creative, critical thinker. Attributes of problem-based learning, such as the
development of problem-solving and self-directed learning skills, teacher role as
tutor and resource instead of sole source of knowledge, and socially collaborative
activities also reflect epistemic and expressive rhetorics. Current-traditional rhetoric
is more aligned with lecture formats of knowledge telling. It conflicts with evidence
that writing is a complex, intellectual activity and seems to disregard students
needs to develop critical thinking skills.
Recognizing and promoting the correlations among critical thinking,
problem solving, and writing, writing teachers can contribute to the cognitive
development of their students into responsible adults. If we teach our students how
72


to use what they know and how to learn what they will need to know, it seems to me
that both the individuals and society at large will be better served" (Stice, 1987, p.
98). This is a responsibility for writing teachers of great magnitude, but one that
should be savored every day:
If we teach our student to care, or at least show them that we care,
our teaching of critical thinking may foster a future technology
more concerned with healing than with destruction. Then students
may learn that the wonders of life on our lfail planet are not only
puzzles that engage our problem-solving abilities but also
mysteries worthy of preserving simply for the beauty and joy they
offer. (Meyers, 1986, p. 119)
Indeed, society will be better served when students are equipped to solve problems,
direct their own learning, think critically, and write effectively about the world
around them.
73


CHAPTER THREE
DEVELOPMENT AND IMPLEMENTATION OF
THE INSTRUCTION INVENTORY
The Instruction Inventory was created and validated to understand the role of
problem-centered instruction in composition. Before the instrument was created,
however, the Cognitive-Composition Model was developed. This chapter explains
the methodology underlying the development of the model and the instrument and
how they provided responses for two research questions. The first question asks the
following:
1. What evidence demonstrates the construct validity of the Instruction
Inventory, an instrument that measures teachers and students perceptions
of instruction and learning?
The second stage of this study involved implementation of the model and
instrument. This chapter also explains methods behind the data collection and
analyses involved in the implementation of the instrument. These methods provided
responses for the second research question in the study:
2. What are the relationships between teachers and students perceptions of the
following in college writing class?
74


a. Problem-centered instruction
b. Self-directed learning use
c. Critical thinking use
Several main sections comprise this chapter: the framework of the Cognitive-
Composition Model, the development of the Instruction Inventory, the construct
validity process, and the implementation of the Instruction Inventory and data
analyses.
Framework for the Cognitive-Composition Model
The Cognitive-Composition Model juxtaposes elements from problem-based
learning, critical thinking, self-directed learning, and rhetorical theories (see Table
3-1). The model illustrates writers cognitive orientations by contrasting them with
writers behavioral orientations; these orientations are defined according to
problem-based learning, self-directed learning, critical thinking, and rhetorical
theory. The integration of problem-based learning with rhetorical theories is
somewhat novel. By combining knowledge of problem-based learning tenets with
composition theories and strategies, teachers may be able to identify ways to
promote their students problem-solving, self-directed learning, and critical thinking
skills.
75


Table 3-1
Cognitive-Composition Model
Focus Cognitive Writers Behavioral Writers Student Attributes
1. Content acquisition C1. Identifies content to be learned. B1. Learns information determined by others.
2. Personalized learning C2. Individualizes learning based on personal interests. B2. Generalizes own learning.
3. Knowledge construction C3. Constructs knowledge actively when learning. B3. Does not knowingly construct knowledge when learning.
4. Problem analysis C4. Analyzes problems. B4. Avoids problems.
5. Problem solving C5. Defines what information is pertinent to solve a problem. B5. Accepts readily available information to solve a problem.
6. Reasoning skills C6. Possesses logical reasoning and problem- solving skills. B6. Thinks compulsively and rashly.
7. Information use Cl. Reuses information Reinforced through use. B7. Discards information after use.
8. Participation activity C8. Participates actively and is motivated when learning. B8. Participates on a passive, low level of involvement when learning.
9. Group problem solving C9. Encourages group problem solving. B9. Retracts from group problem solving.
10. Solution implement- ation CIO. Implements solutions to problems solved. B10. Does not complete problem solution implementation.
11. Prior knowledge C11. Activates own prior knowledge. B11. Does not acknowledge prior knowledge.
12. Learning independence Cl2. Learns independently. B12. Depends on others to identify content to be learned.
13. Critical thinking C13. Thinks critically. B13. Possesses unreasonable, irrational reasoning skills.
76


Table 3-1 (continued)
Focus Cognitive Writers Behavioral Writers
Student Attributes
14. Learning transfer C14. Transfers learning responsibility from the teacher to self. B14. Does not transfer responsibility for learning from teacher.
15. Emphasis of teaching C15. Emphasizes process of discovery. B15. Emphasizes product of correctness, not process.
16. Model writing Cl6. Connects between good writing and good thinking. B16. Takes positivist approach to observing and measuring good writing.
17. Viewpoint concerns C17. Reads authors who take different perspectives on writing topic. B17. Reads limited number of authors and perspectives.
18. Function of writing Cl8. Believes writing aids cognition because writers use language to discover ideas. B18. Believes writing fits Proper formats of research papers and discourse classification modes.
19. Philosophic basis C19. Believes truth and knowledge is dynamic and dialectical, not static. B19. Believes truth and knowledge are concrete and definite.
Cognition and Behaviorism
The Cognitive-Composition Model is a continuum, not a static diagram,
because it ascertains the extent of problem-based leamingness in composition
instruction. Most notable about the Cognitive-Composition Model is the contrast
between the extremes of the continuum, cognitive and behavioral writers. This
contrast reflects contemporary learning theories (Bigge & Shermis, 1999): Their
great difference centers upon the behavioristic assumption that human beings are
passive or reactive and the cognitive-interactionist assumption that they are
interactive in relationship with their environments (p. 44).
77


Cognitive writers tend to address problem scenarios that they could actually
encounter in a real-world environment. They assume responsibility' for solving
problems inevitable in the writing process with active, critical thinking. Cognitive
writers typically use problem-solving and self-direction skills, they actively
construct knowledge, define resources needed to solve a problem, assume
responsibility for learning, and experiment with learning. Thus, and most
importantly, cognitive writers can write effectively for any given audience or
purpose with engaged cognition and critical thinking because they are able to
understand and write appropriately in different rhetorical situations.
In sharp contrast to cognitive writers are behavioral writers, students who
typically learn to write in an environment in which the teacher crafts the writing
assignment and controls students learning. Teachers focus instruction on sentence-
level mechanics and unknowingly do not encourage problem-solving, self-direction,
or critical thinking. Characteristics of behavioral writers include avoidance of
problem-solving and critical thinking, concentration on language and composition
mechanics, emphasis on product over process, and dependency on the teacher to
define all learning objectives and control learning. Behavioral writers tend to
depend solely on their teachers instructions and accuracy because they are more
aware of the written format and accurate mechanics than their audience and purpose
for writing. These students tend to engage superficially with their topics also, which
usually reflects shallow thinking throughout the writing process.
78


Problem-centered instruction can enhance students learning, as it has in
many disciplines. In science courses, students were challenged to make and justify
estimations and assumptions (Allen, Duch, & Groh, 1996). Business students had
the opportunity to solve everyday business problems (Holt, & Willard-Holt, 2000).
In health care fields, students had the opportunity to solve complex,
multidisciplinary, problems (Bruhn, 1997). Engineering students exposed to a
problem-based learning program scored one standard deviation higher than those
who were not enrolled in the program (Woods, 1996). Calculus students became
more independent and involved in mathematics when they were exposed to a top-
down approach of identifying subproblems before major problems (Seltzer, Hilber,
Maceli, Robinson, & Schwartz, 1996). Business students learned content as needed
by managing learning problems (Stinson & Milter, 1996). Similar positive
experiences have been noted in social work (Heycox, & Bolzan, 1991) and law
(Winsor, 1991). These studies illustrated the functionality of problem-centered
instruction in various disciplines. They did not, however, utilize a valid instrument
to measure teachers and students perceptions of teaching and learning, hence the
need for the construct validity of the Instruction Inventory.
Development of the Instruction Inventory
Although one purpose of this study was to develop a discipline-independent
instrument that informs teachers about how problem-centered instruction, self-
directed learning, and critical thinking are perceived by teachers and students, it was
79


not the first construction of this study. Likewise, neither was the model the first
construction. The starting point was the development of four continua to concretize
theory and research. Figure 3-1 illustrates the cycle between the theories and the
development of the Cognitive-Composition Model and Instruction Inventory.
Development of the Cominua
The four continua concern problem-based learning, self-directed learning,
critical thinking, and rhetorical theory7. The continuum of problem-based learning
theory and methods identifies attributes of problem-centered instruction and
contrasts them with its antithesis, teacher-based learning, as depicted in Table 3-2.
On the left side of the continuum is problem-based learning, which is student-
centered learning because the problems essentially promote self-directed learning
and problem-solving skills with the guidance of a tutor and transfer the
responsibility of learning from teacher to student (Barrows, 1988; Seltzer, Hilber,
Maceli, Robinson, & Schwartz, 1996). Problem-based learning helps students to
develop problem-solving and reasoning processes, self-directed learning skills, and
collaborative working skills (Allen & Rooney, 1998; Barrows & Tamblyn, 1980;
Rangachari, 1996). Also, problem-based learning stimulates internal motivation for
learning because it requires student involvement throughout the learning process.
Such a problem-centered approach to instruction also tends to develop self-
evaluation skills and activation of prior knowledge (Barrows, 1988; Bridges &
Hallinger, 1997; Bruhn, 1997).
80


Instruction
Inventory
Problem-based Learning,
Composition Studies, and
Critical Thinking Theories
Cognitive-Composition Model
Figure 3-1. Cycle between the theories, model, and survey instrument used in this
study.
Problem-centered instruction is integrated differently in different classrooms
by different teachers. The continuum illustrated in Table 3-2, however, depicts the
learning approach according to Barrows (1988). Even so, Barrows (1986) explains
variations in its applications:
The term problem-based learning must be considered a genus for
which there are many species and subspecies. Each addresses
different objectives to varying degrees. All descriptions and
evaluations of any PBL method must be analysed in terms of the
type of problem used, the teaching-learning sequences, the
responsibility given to students for learning and the student
assessment methods used. Any teacher who wishes to employ PBL
should decide on desired educational objectives and then select the
method that fits best. (p. 485)
On the right side of the problem-based learning continuum is teacher-based
learning. This type of learning is teacher-centered because it encourages students to
depend on their teachers to tell them what to learn, how to learn, and when to learn.
Freire (1990) referred to this method as banking because the teacher, known as
81


Table 3-2
Continuum of Student Characteristics of Problem-Centered Instruction
Focus Problem-Based Learning Teacher-Based Learning
Student Attributes
1. Content PI. Identifies content to be learned. (Barrows, 1996) Tl. Leams information determined by others.
2. Personalized learning P2. Personalizes own learning based on persona! interests. (Barrows & Tamolyn, 1980) T2. Generalizes own learning.
3. Knowledge construction P3. Constructs knowledge actively when learning. (Barrows & Tamblyn, 1980) T3. Does not knowingly construct knowledge when learning.
4. Problem analysis P4. Analyzes problems. (Birch, 1986) T4. Avoids problems.
5. Reasoning skills P5. Practices logical reasoning and problem-solving skills. (Edens, 2000) T5. Thinks compulsively and rashly.
6. Participation activity P6. Participates actively and is motivated when learning. (Woods, 1994) T6. Participates on passive, low level of involvement when learning.
7. Group problem solving P7. Encourages group problem solving. (Allen, Duch, & Groh, 1996) T7. Withdraws from group problem solving.
8. Solution implement- ation P8. Implements solutions to problems solved. (Bridges & Hallinger. 996) T8. Does not complete solution implementation.
9. Prior knowledge P9. Activates own prior knowledge. (Bridges & Hallinger, 1996) T9. Does not acknowledge prior knowledge.
10. Independence P10. Leams independently. (Gijselaers, 1996) T10. Depends on others throughout learning process.
11. Critical thinking PI 1. Thinks critically. (Magnussen, Ishida, & Itano, 2000) Til. Possesses illogical, irrational reasoning skills.
12. Learning transfer PI2. Transfers learning responsibility from teacher to self. (Barrows & Tamblyn, 1980) T12. Does not transfer responsibility for learning from teacher.
82


the oppressor, deposits knowledge into students heads. Banking minimizes
students creativity. It serves interests of instructors who teach prescribed, contrived
lessons with no intentions to share real-world situations with students. These
teachers react instinctively against any experiment in education which stimulates
the critical faculties (Freire, p. 60), further emphasizing students dependence on
their teachers control of the learning process.
The continuum of self-directed learning theory and methods, Table 3-3,
contrasts self-directed learners with other-directed learners. Self-directed learners
identify their own learning goals, activities, and assessments (Candy, 1991). They
are comfortable learning by themselves and in groups (Knowles, 1975). Self-
directed learners have a strong sense of control and responsibility regarding their
own learning (Brockett & Hiemstra, 1991). In contrast, other-directed learners look
to their teachers to direct their learning. They typically do not assume much control
or responsibility in their learning, nor do they make time for learning or accept
criticism well. The major differences between self-directed and other-directed
learners concern the impetus of the learning and the desire to learn.
In Table 3-4, the continuum of critical thinking theory and methods contrasts
critical thinkers with surface thinkers. Critical thinkers are people who analyze
problems, evaluate claims, suspend conclusions, and judge logical inferences
(Barnet & Bedau, Norris. & Ennis, 1989; Ruggiero, 1988). Surface thinkers are
83


Table 3-3
Continuum of Learners Self-Directed Learning Skills
Focus Self-Directed Learning Other-Directed Learning
1. Self-reliance SI. Student At Self-reliant in finding information. (Hiemstra, 1998) ributes Ol. Needs help finding information.
2. Personal goals S2. Sets personal learning goals. (Garger, 1999) 02. Does not set personal learning goals.
3. Learning in isolation S3. Enjoys learning in isolation. (Hiemstra, 1998) 03. Does not enjoy learning in isolation.
4. Time for learning S4. Finds time for learning. (Licklider, 1997) 04. Does not find time for learning.
5. Feedback and criticism S5. Accepts feedback and criticism. (Anfield. 1997) 05. Does not accept feedback and criticism.
6. Learning control S6. Takes control of personal learning. (Brookfield. 1993) 06. Does not take control of personal learning.
7. Seeking learning S7. Seeks out learning opportunities. (Gerber, Lankshear, Svensson, 1995) 07. Does not seek out learning opportunities.
8. Problem- solving S8. Enjoys problem solving. (Hiemstra, 1997) 08. Does not enjoy problem solving.
9. Challenging assumptions S9. Challenges assumptions. (Pilling- Cormick, 1997) 09. Does not challenge assumptions.
10. Responsibility for learning S10. Assumes responsibility for own learning. (Knowles, 1975) 010. Does not assume responsibility for own learning.
11. Self-evaluation SI 1. Evaluates and advocates own learning. (Pilling- Cormick, 1997) OIL Does not evaluate or advocate own learning.
12. Self-reflection S12. Reflects upon and makes choices about own learning. (Candy, 1991) 012. Does not reflect upon or make choices about own learning.
84


Table 3-4
Continuum of Learners Critical Thinking Skills
Critical Thinking Surface Thinking
Student Attributes
1. Problem solving C1. Analyzes problems. (Norris, 1992) G1. Stays away from problems when possible.
2. Evaluation skills C2. Evaluates claims, assumptions, and evidence. (Smith, 1990) G2. Accepts claims, assumptions, and evidence on value.
3. Learning Attitude C3. Shows imaginative open-mindedness and intellectual curiosity. (Norris & Ennis, 1989) G3. Shows an objective, indifferent approach to learning.
4. Idea generation C4. Considers new ideas. (Barnet & Bedau, 2002) G4. Depends on former ideas.
5. Summarizing skills C5. Summarizes arguments accurately. (Barnet & Bedau, 2002) G5. Summarizes arguments incorrectly.
6. Self-reflection Co. Uses judicious, or reflective, skepticism. (McPeck, 1981) G6. Accepts information without much reflection or contemplation.
7. Construction of solutions C7. Constructs alternatives and solutions in problem-solving. (Meyers, 1^86) G7. Applies first solutions developed in problem- solving.
8. Argument construction C8. Can recognize and construct sound arguments. (Ruggiero, 1988) G8. Does not deliberately recognize or analyze sound arguments.
9. Application of logic C9. Applies principles of formal and informal logic. (Barnet & Bedau, 2002) G9. Vague application of formal and informal logic principles.
10. Understanding fallacies CIO. Avoids fallacies in reasoning. (Ruggiero, 1988) G10. Uses logical fallacies in reasoning.
11. Making conclusions C11. Suspends conclusions. (McPeck, 1981) G11. Makes quick conclusions and judgments.
85


Table 3-4 (continued)
Critical Thinking Surface Thinking


Student Attributes
12. Accepting truths Cl2. Sets aside own ideas of the truth and G12. Depends on others ideas of the truth without
reflects on reflection.
13. Motivation alternatives. (Meyers, 1986) C13. Internally motivated G13. Externally motivated
14. Logical Inferences about learning. (Halpem, 1989) C14. Can judge and make logical inferences. about learning. G14. Hastily judges and draws logical inferences.
(Norris, 1992)
mundane thinkers (Norris & Ennis). The contrast between critical and surface
thinking is that critical thinking is reasonable and reflective; that is, a critical
thinker makes decisions conscientiously and deliberately, whereas a surface thinker
is spontaneous (Norris & Ennis). Critical thinking involves purpose, reason, goals,
problem-solving, inferences, evaluations, reflection, and analysis (Halpem, Norris).
Surface thinking is neither reasonable nor reflective, making it everyday, functional
thinking without deliberation or goal orientation.
The continuum of instructional strategies in rhetorical theories in Table 3-5
contrasts two rhetorical theories, epistemic and current-traditional (Gere, 2000, pp.
10-25). Epistemic rhetoric, on the left side of the continuum, is oriented toward
problem-centered instruction because it stresses problem-solving, the process of
writing, discovery learning, multiple perspectives, reciprocal relationships between
86


Table 3-5. Continuum of Instructional Strategies in Rhetorical Theories
Focus Epistemic Current-Traditional
Instructional Strategies
1. Teaching emphasis E1. Emphasizes process of writing. (Berkenkotter, 1988) CT1. Emphasizes written product.
2. Exemplary writing E2. Stresses connections between good writing and good thinking. (Rohman, 1965) CT2. Stresses connections between good writing and observable behaviors.
3. Theoretical sources E3. Endorses reciprocal relationship between subjective thought and language artistry. (Berlin. 1987) CT3. Endorses reciprocal relationship between objective analysis and grammar.
4. Function of writing E4. Uses writing as heuristic to discover ideas. (Flower, 1989) CT4. Uses writing as tool to report information.
5. Viewpoint concerns E5. Encourages reading different perspectives on the writing topic. Sommers, 1982) CT5. Encourages exposure to limited points of view in writing topic.
6. Philosophic basis E6. Embraces truth and knowledge as dynamic and dialectical. (Berlin, 1987) CT6. Embraces truth and knowledge as concrete and static.
7. Audience concerns E7. Considers audience as a primary role in writing, influencing content and organization. (Wallace, 1994) CT7. Considers audience as a minor role in writing, second to punctuation and syntax.
thought and language, and audience. The right side of the continuum, current-
traditional rhetorical theory, is oriented toward teacher-based learning because it
emphasizes the teacher's control in the classroom and focuses on the quality of
written products and the dissection of words and sentences, usage and style, and
87


Full Text

PAGE 1

REALIZING VALUES lN TEACHING WRITING WITH PROBLEM-CENTERED INSTRUCTION: DEVELOPMENT OF A COG.l'HTIVE-COMPOSITION MODEL .t\ND INSTRUCTION INVENTORY by Rosalyn H. Zigmond B. Mus., University of Michigan, 1978 M A., University of Colorado at Denver. 1993 A thesis submitted to the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Doctor of Philosophy Educational Leadership and Innovation 2004

PAGE 2

This thesis for the Doctor of Philosophy degree b y Rosalyn H. Zigmond has been approved by Rodney Muth Alan Davis Joanna Dunlap Date

PAGE 3

Zigmond, Rosalyn H. (Ph.D., Educational Leadership and Innovation) Realizing Values in Teaching Writing with Problem-Centered Instruction: Development of a Cognitive-Composition Model and Instruction Inventory Thesis directed by Professor Rodney Muth ABSTRACT The purpose of this study was bifold: (1) to develop an instrument that identifies teachers' and students' perceptions of instruction and learning strategies, and (2) to implement the instrument and analyze the relationships between perceptions of problem-centered instruction, self-directed learning, and critical thinking in college composition classrooms. To accomplish this, a CognitiveComposition Model was created and an Instruction Inventory was designed, validated, and administered. The research review explains relationships among critical thinking, a problem-based learning approach to teaching, and composition studies. These constructs constitute the model and facilitated interpretation of data. The first stage of this study was forming four continua to concretize theory in problem-based learning, self-directed learning, critical thinking, and composition studies. These continua led to development of the Instruction Inventory, which was used to collect data from 15 writing teachers and 247 students. Results show that teachers believe that they used more problem-centered instructional strategies than their students believed were used in class but that students perceived themselves to iv

PAGE 4

be using more self-directed learning and critical thinking skills than their teachers did. An internal consistency seems to exist in how students view their writing classes in respect to a problem-centered instructional focus and their perceived use of self-directed learning and critical thinking skills. However, there is no evident correlation at all between how teachers and students perceive the same classes in respect to these constructs. A larger sample size may or may not confirm these results. A factor analysis confirmed the study's fust hypothesis, that a principal factor underlies all items in the Instruction Inventory. An oblimin rotation confumed the second hypothesis, that three constructs underlie the inventory, with a set of items corresponding to each of these constructs, or dimensions. Promising uses for the Instruction Inventory and Cognitive-Composition Model include faculty-development programs and student-centered instructional design, development, and delivery based on teachers' and students' perceptions of teaching and learning in the classroom. Further study can improve the instrument's constructs and reliability; it also may refine the model's representation of cognitive composition. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. RodneyMuth v

PAGE 5

DEDICATION My treasured sons, Andy and Frank, gave me encouraging support and inner strength that kept me reading, writing, and laughing when I needed to laugh. My wonderful parents have always been by my side, believing in me throughout my life, and I will always be thankful for their unconditional love and support. I am especially grateful to my husband, Irving, whose love and devotion helped me to accomplish my life goal, this dissertation.

PAGE 6

ACKNOWLEDGEMENT This dissertation is the result of support from many special individuals in my life. My advisor, Rodney Muth, provided unparalleled insight and appreciated smiles. Alan Davis helped me with the methodology and analyses phases of this study with endless patience. Other members of my dissertation committee, Joanna Dunlap and Patricia Sullivan, lent constructive criticism and guidance whenever I needed it.

PAGE 7

CONTENTS Figures ...................................................................... xii Tables ....................................................................... xiii CHAPTER 1. TEACHING WRITING WITH PROBLEM-CENTERED INSTRUCTION .................................................... 1 Problem-Centered Approach ...................................... 1 Research Questions ... ................... .. .. . . ......... 2 Areas of Research ........................................ ........... 5 Merging Problem-Based Learning with Teaching Writing ... 7 Problems are Integral. ..................................... 9 Critical Thinking and Self-Directed Learning as Problem-Based Learning Skills ..................... 10 The Need for Critical Thinking and Self-Directed Learning .................................................... 12 The Challenges Critical Thinking and SelfDirected Pose for Writing Teachers ...................... 14 Unique Tools ............................................... 16 Overview of Methodology ........................................ 17 Applications and Significance of Research ............ 19 2. LINKING STUDIES IN CRITICAL THINKING, PROBLEM-BASED LEARNING, AND COMPOSITION STUDIES ....................................... 22 Critical Thinking ....................................... ............... 22 Definitions .................................................. 26 Uniqueness of Time and Environment. ................ 29 Problem-Based Learning .......................................... 31 Educational Outcomes and Advantages ............... 36 Barriers ..................................................... 38 Self-Directed Learning ................................... .42 Vlll

PAGE 8

Assessing Problem-Based and Self-Directed Learning ................................................. 46 Teaching Writing ................................................... 51 Everyday Necessities ..................................... 53 Goal-Directed Thinking .................................. 54 Reflective Learning ....................................... 56 Discovery Process ......................................... 59 Problem Solving as Collaborative Planning ........... 60 Rhetorical Theory. . . . . . . . . . . .. . . . . . . . . . 62 Summary ............................................................. 69 3. DEVELOPMENT AND IMPLEMENTATION OF THE INSTRUCTION INVENTORY .................................. 74 Framework for the Cognitive-Composition Model. ........... 75 Cognition and Behaviorism.............................. 77 Development of the Instruction Inventory............ . . . . . 79 Development of the Continua ............................ 80 Construct Validity: Building a Body ofEvidence ............. 88 Step One: Identify Theoretical Framework to Develop Items ....................................... 92 Step Two: Pilot Study .................................... 95 Step Three: Factor Analysis .............................. 98 Step Four: Data Analysis ................................. 99 Implementation Analyses: Examining Relationships Between Scores. . . . . . . . . .. . . .. . . . . . .. . . . .. . 1 00 Limitations of the Study.. .... ........ .. . ..... .. ... .... ....... ... .. ... ... 1 01 4. REALIZING VALUES IN TEACHING WRITING WITH PROBLEM-CENTERED INSTRUCTION ..................... 1 03 Construct Validity of Instruction Inventory: Results ......... 104 Roles and Results from Expert Judges ................. 104 Pilot Study Results ........................................ 106 Principal Component Analysis and Oblimin Rotation Analysis ........................... 109 Teaching and Learning Relationships in Composition: Results ................................................... 114 Correlations of Student Subscale Scores ............... 114 Correlations of Teacher Subscale Scores ............... 117 Correlation of Teacher and Student Scores ............ 117 lX

PAGE 9

Relationships between Perceptions on All Items before Factor Analysis ......................................... 120 Relationships between Scoring Trends ................ 122 Items with Largest and Smallest Discrepancies between Scores ...................................... .. 123 Discussion of Results ............................... ..... 132 5. CONCLUSIONS AND IMPLICATIONS .......................... 133 The Cognitive-Composition Model and Teaching Writing ... 135 Implications ................................................ 135 .Further Research .......................................... 136 The Instruction Inventory and Perceptions ..................... 140 Implications ................................................ 142 Further Research .......................................... 146 Coda ............................................................. .. ... 147 APPENDIX A. CODED STATEMENTS IN TEACHERS' FORM OF INSTRUCTION INVENTORY THAT STEM FROM PROBLEM-BASED LEARNING CONTINUUM ............... ...... 149 B. CODED STATEMENTS IN TEACHERS' FORM OF INSTRUCTION INVENTORY THAT STEM FROM SELF-DIRECTED LEARNING CONTINUUM ....................... 153 C. CODED STA TEMFNTS IN TEACHERS' FORM OF INSTRUCTION INVENTORY THAT STEM FROM CRITICAL THINKING CONTINUUM ................................. 157 D. JUDGES' DIRECTIONS ................................................... l61 E. EMAIL MESSAGE TO FACULTY REQUEST TIIEM TO PARTICIPATE IN STUDY ............. ................................. 182 F. INSTRUCTIONS FOR TEACHERS ADMINISTERING INSTRUCTION INVENTORY ......................................... ..... 183 G. STIJDENT FORM OF INSTRUCTION INVENTORY ............... 185 X

PAGE 10

H. TEACHER FORM OF INSTRUCTION INVENTORY ............... 189 I. RELATIONSHIP BETWEEN TEACHERS' AND STUDENTS' MEANS FOR INSTRUCTION INVENTORY ITEMS ............... 193 REFERENCES ....................................................................... 199 XI

PAGE 11

FIGURES Figure 1-1 Relationships between critical thinking and self-directed learning skills in a problem-centered composition classroom ........... 11 1 -2 Links between theories, Cognitive-Composition Model and Instruction Inventory construction ..... . . .. . ....... ... ............. .... 18 3-l Cycle between the theories, model, and survey instrument used in this study .............................................................. 81 4-1 Teachers' and students' problem-based learning scores .. ............ 119 4-2 Teachers' and students' self-directed learning scores ................... 119 4-3 Teachers' and students' critical thinking scores ......... .. ............ 120 xu

PAGE 12

TABLES Table 2-l Continuum ofLeamers' Critical Thinking Skills ........................ 24 2-2 Continuum of Student Characteristics of Problem-Based Learning ... 32 2-3 Continuum of Learners' Self-Directed Learning Skills ................ .43 2-4 Continuum of Instructional Strategies in Rhetorical Theories ......... 64 2-5 Synthesis of Critical Thinking, Problem-Based Learning, and Teaching Writing in Student Attributes ................................... 70 3-1 Cognitive-Composition Model. ............................................ 76 3-2 Continuum of Student Characteristics of Problem-Centered Instruction ...................................................................... 82 3-3 Continuum of Learners' Self-Directed Learning Skills ................. 84 3-4 Continuum of Learners' Critical Thinking Skills ........................ 85 3-5 Continuum of Instructional Strategies in Rhetorical Theories ......... 87 3-6 Origin and Coding ofltems .................................................. 94 4-1 Teachers' and Students' Comments after Pilot Study of Instruction Inventory ........................................................ 1 08 4-2 Principal Component Matrix ............................................... 111 4-3 Component Analysis with Oblimin Rotation ............................ 113 4-4 Item Distribution in Individual Subscales ................................ 115 Xlll

PAGE 13

4-5 Correlations of Student Subscore Totals ........ .. ..... ..... ........... l 16 4-6 Correlations of Teacher Subscore Totals . .. . .................... .. ... 118 4-7 Highest and Lowest Scored Items by Students and Teachers ......... 124 4-8 Items with Largest Discrepancies between Student s and Teachers' Scores ............................ .. ... ... . . . ................. 1 2 6 4-9 Items with Most Similarities between Students and Teachers' Scores ....... . .............. ............... . ................. ............. 131 5-1 Characteristics of Constructs Used in Study .. ....................... .. 134 XIV

PAGE 14

CHAPTER ONE TEACHING WRITING WITH PROBLEM-CENTERED INSTRUCTION This study is an exploration of college writing students' use of critical thinking and self-direction in classes whose teachers use an explicit problem centered instructional approach. Social or collaborative, aspects of writing are included in this study also, as Crosby (200 3 ) describes writing as an open-ended, relational process that engages one with others--other writers, speakers, and readers, other thoughts, other words, other times" (p. 629). Understanding students' and teachers' perceptions of problem-centered writing classes, collaborative learning self-directed learning, and critical thinking inspired development of the model and survey instrument in this study. Problem-Centered Approach Problem-based learning, a structured approach to problem-centered instruction originated in medical curricula in the 1960s and promotes critical thinking and self-directed learning in social learning environments (Aspy Aspy, & Quinby, 1993 ; Barrows 1994). Accordingly writing teachers who intentionally design instruction with a problem-centered approach should also be able to emphasize collaborative aspects of critical thinking and self-directed learning (Allen Sargeant, Mann, Fleming & Premi 2003; Copland, 2003). Assuming that

PAGE 15

students use social skills in classes with problem-centered instruction, how do their perceptions compare to their teachers' perceptions of the teaching and learning in their composition classes? Understanding such educational dynamics is the heart and inspiration of this study. Research Questions The following research questions guide this study: 1. What evidence demonstrates the construct validity of the Instruction Inventory. an instrument that measures teachers' and students' perceptions of instruction and learning? 2. What are the relationships between teachers' and students' perceptions of the following constructs in the college composition class? a. problem-centered instruction b. self-directed learning c. critical thinking The first question emphasizes the first phase of the study, the critical process of validating the constructs defined in the instrument under consideration. The second question identifies the college writing class as the target audience for which the instrument was designed and tested. Although the initial impetus for this study was to understand how a problem-centered approach to teaching writing relates to students' use of critical thinking and self-directed learning skills, a survey instrument had to be developed and validated first. Without a valid instrument that 2

PAGE 16

measures teachers' and students' perceptions in teaching and learning, understanding relationships between problem-centered instruction. self-directed learning, and critical thinking would be uncertain. To begin to answer these questions, theory and research were reviewed in problem-based learning, critical thinking, and composition studies. From this research, continua were developed. which grew into a model of cognitive writing. After the model was built, an instrument was created to measure teachers' and students' perceptions of problem-centered instructional strategies and students' use of self-directed learning and critical thinking. The underlying intention of this instrument was to find a way to identify perceptions of such strategies and skill use (Dolmans, Schmidt, & Gijselaers, 1994b; Dolmans, Gijselaers, Moust, Grave, & Van Der Vleuten, 2002). Uncertainty in how students perceive instruction can pose challenges for teachers. Teachers can, for example, under or overestimate students' use of a skill or strategy, under or overestimate students' understanding of a topic, design classroom activities that do not target students' real learning needs, give assignments that bore or frustrate students, evaluate students' work on misunderstood criteria, mislead discussions, and the list continues. Because of these uncertain perceptions, teachers may design learning assignments that do not meet their students' needs, resulting in a teacher-centered classroom (Brooks & Brooks, 1999). A teacher-centered, or content-centered classroom is significantly different from a student-centered 3

PAGE 17

classroom. A teacher-centered classroom strives to meet the needs of a teacher needing to accomplish content-driven goals regardless of students' progress in learning (Brooks & Brooks). A student-centered classroom strives to present content in such a way that it complements students' learning progress (Yost, 2000). The important point here is that incorrect perceptions by teachers and students may strain learning (Cashin & Downey, 1995). Students misperceptions of instruction can create problems because miscommunications with their teachers can create negative attitudes about learning (Tan, Goh, Chi a, & Treagust, 2001 ). For example, students in a recent study reported that not only did they find learning qualitative analysis tedious and challenging in a lecture-format, they thought the practical, recitation sessions were unrelated to their class lessons even though the teacher designed the practice sessions to be relevant (Tan, Goh, Chia, & Treagust). Students may also believe that their teachers are expecting too much from them or not challenging them enough. They may feel that their teachers are presenting instruction with no regard for their personal learning needs, or students may think that their teachers are under or overestimating their acquisition of the content and learning objectives for the course. On the other hand, students' perceptions of humor in instruction, for example, can result in increased learning (Wanzer & Frymier, 1999). In short, students' perceptions of instruction do seem to affect their learning (Cashin & Downey, 1995). 4

PAGE 18

To understand these types of classroom and instructional idiosyncrasies, a Cognitive-Composition Model and an Instruction Inventory were designed for this study. Their purpose is neither to assess teaching or learning nor to determine the effect of different instructional strategies on different types of learning. Rather, the Cognitive-Composition Model and Instruction Inventory reveal information about relationships between constructs. Moreover, the study focuses on the relationships between teachers' and students' perceptions of problem-centered instructional strategies, self-directed learning, and critical thinking in college composition classes. Areas of Research Main areas of research for this study include problem-centered instructional strategies. self-directed learning, and critical thinking skills. Problem-centered instruction is a logical approach to help writing student-; learn how to write well (Flower, 1989). It is grounded in problem-based learning, a particular approach that focuses students' learning on solving probiem scenarios, exploring what they know and need to know in order to solve the problem, and evaluating their learning as individuals and team members (Barrows, 1985, 1988; Barrows & Tamblyn, 1980). Self-directed learning is a desirable practice for mature adults to hone (Guglielmino & Guglielmino, 2001; Knowles, 1975). It is a strategy that encourages self-concepts of responsibility and control in the learning process, including identification of individual learning objectives, activities, and assessments (Candy, 1991; 5

PAGE 19

Zimmerman & Lebau, 2000). Critical thinking is a proficiency that enables people to make difficult life decisions (Barnet & Bedau, 2002; Greenfield, I 987). It is a skill that emphasizes purposeful, rationalized analysis and evaluation of others' statements, written and spoken; critical thinking also entails problem-solving, intentional reflection, and creativity (Barnet & Bedau, 2002; Drake, 1976; McPeck, 1990). To establish a common-ground understanding of the constructs in this study, problem-centered instruction is an informal way of teaching. It is grounded in problem-based learning, a particular approach that focuses students' learning on solving problem scenarios, exploring what they know and need to know in order to solve the problem, and evaluating their learning as individuals and team members (Barrows, 19S5, 1988; Barrows & Tamblyn, 1980). Self-directed learning is a strategy that encourages self-concepts of responsibility and control in the learning process, including identification of individual learning objectives, activities, and assessments (Candy, I 991; Zimmerman & Lebau, 2000). Critical thinking is a skill that emphasizes purposeful, rationalized analysis and evaluation of others' statements, written and spoken. It entails problem-solving, intentional reflection, and creativity (Barnet & Bedau, 2002; Drake, 1976; McPeck, 1990). The Instruction Inventory is a short, discipline-independent survey designed around problem-centered instruction, self-directed learning, and critical thinking. The Cognitive-Composition Model depicts attributes of two polarized types of writers on a continuum. It illustrates one type of writer as a self-directed, critically 6

PAGE 20

thinking problem-solver and the opposite type of writer as directed by others, quick thinking, and avoiding cont1ict. Somewhere between these outliers are authentic, logical descriptions of real student writers. The important point of the model is that characteristics of problem-centered instruction, self-directed learning, and critical thinking in writing classes manifest themselves in varying degrees, not as predefined, polarized attributes. The relationships between such variations as they are perceived by teachers and students are the essence of this study. Merging Problem-Based Learning '-Vith Teaching Writing This chapter is titled "Explicitly Teaching Writing with Problem-Centered Instruction" because teaching writing has rarely been formally studied in a problem centered context. Instructional strategies integral to problem-centered learning, however are not all that new to composition instruction For example teaching writing in accord with epistemic and expressivist rhetorical theory entails instructional strategies such as teaching writing as a process, encouraging deep thinking about issues, and stimulating discovery throughout the learning process (Berlin, 1987). These types of instructional strategies focus this study. The strength of exploring the discipline of composition studies in problem centered instruction is that both naturally integrate self-directed learning and critical thinking skills (Flower, 1 989; Zimmerman & Lebau, 2000). Problem-centered learning environments have been found to develop self-directed learning and critical thinking skills in non-composition classes although these findings are not 7

PAGE 21

quantified (Allen, Duch, & Groh, 1996; Stepien & Gallagher, 1993; Woods, 1994). Albanese and Mitchell (1993), for example, conducted a meta-analysis of problembased learning environments for medical curricula and suggested that problembac;ed learning is a valuable approach to learning, but that it should be studied continuously and its benefits analyzed to understand and appreciate its strength in developing cognition. These types of encouraging reports provide all the more reason, and need, to develop an instrument that identifies perceptions of problemcentered instruction, self-directed learning, and critical thinking skills. If problem-centered instruction can promote students' critical thinking and self-directed learning skills in medical, engineering, and business curricula (Barrows, 1994; Stinson & Milter, 1996; Woods, 1994), it should also achieve similar outcomes in composition classes. Problem-centered instruction advances authentic problem solving, another characteristic of critical thinkers and selfdirected learners: Students, as they are increasingly posed with problems relating to themselves in the world and with the world, will feel increasingly challenged and obliged to respond to that challenge. Because they apprehend the challenge as interrelated to other problems v.ithin a total context, not as a theoretical question, the resulting comprehension tends to be increasingly critical and thus constantly less alienated. Their response to the challenge evokes new challenges, fol1owed by new understandings; and gradually the students come to regard themselves as committed. (Freire, 1990, pp. 68-69) 8

PAGE 22

According to Freire ( 1990), if problem-engagement during the learning process heightens students' commitment to learn then exploration of problems in problem-centered instruction should also be a viable way to teach writing that engages learners' thinking (Barrows, 1996; Edens 2000; Gijselaers, 1996). Accordingly, problem-based learning should help students to develop their critical thinking as well as their identities as maturing writers, as its foundation is solving problems, clearly an intellectual activity. Problems Are Integral Problem-centered instruction is grounded in problem-based learning, student-centered learning in that students identify what they know and what they need to know in consultation with their teachers to solve problems (Barrows, 1985 1986, 1994 1996; Barrows & Tambl:yn, 1980) Students work in small groups and discuss their collective knowledge to teach and learn from each other (Bridges & Hallinger, 1997). The instructor, or tutor, assumes the role of a coach and asks metacognitive questions that challenge students' thinking and writing about the problems (Barrows, 1988; Gijselaers, 1996; Stepien, & Gallagher, 1993). The relationships between problem-centered instruction critical thinking skills, and self-directed learning skills in a composition classroom are illustrated in Figure 1-1. Norris and Ennis (1989) explain: "Since critical thinking takes place within a problem-solving context and often in the context of interacting with other people, the critical thinker needs to be able to function effectively within these 9

PAGE 23

contexts" (p. 8). Norris and Ennis also emphasize the importance of collaborative learning for critical thinkers, which is also inherent in the problem-based learning environment (Bruffee, 1993). In Figure 1-1, the connection between the college composition classroom and pressures from the real world on composition classes (arrows) increase the importance of preparing students for real-world challenges. Critical thinking skills and self-directed learning skills are two of several elements of problem-based learning environments that are highlighted in this study. Critical Thinking and Self-Directed Learning as Problem-Based Learning Skiils In essence, critical thinking involves reflection, active engagement, suspended conclusions, open-mindedness, evaluation, analysis, formulating inferences, making decisions, and problem solving (Barnet & Bedau, 2002; McPeck, 1981; Norris, 1992; Ruggiero, 1988). Attitude and application are integral to critical thinking also, as people must use their critical thinking skills regularly, not dust them off to decipher occasional crises (Halpern. 1989; Norris & Ennis. 1989). Thinking critically requires sustained effort and commitment to fulfill its value and nurture its growth over time. These types of cognitive skills help to solve problems. Self-directed learning involves accepting responsibility for and controlling one s own learning (Brockett & Hiemstra, 1991; Brookfield, 1993 ). One of the more important premises of self-directed learning is that instructional activities should be based on the learner's perceived needs (Hiemstra, 1998). The 10

PAGE 24

Real World College composition classroom Problem-solving context / Real World Figure 1-1. Relationships between critical thinking and self-directed learning skills in a problem-centered composition classroom. 11

PAGE 25

distinction between self-directed learning and other-directed learning is that self directed learners identify their own needs and objectives, oftentimes in conjunction with their teachers' input. Likewise, students also assess their own learning. Self directed learning, by definition, is clearly student-centered and serves individuals well when attempting to solve their learning problems and needs. The Need for Critical Thinking and Self-Directed Learning People need critical thinking skills to make sense of the onslaught of information that bombards them every hour of every day (Meyers, 1986). These are the skills that enable people to think through daily floods of information: "In a world of complex facts, events and ideas, there simply is no short cut to analyzing arguments apart from understanding these complexities" (McPeck, 1981, p. 93). People must think critically to make sensible decisions about the economy natural resources, health, education, investments, and nuclear weapons that affect every aspect oflife. Critical thinking skills are important lifelong-learning skills to develop because most people complete their formal education in their early twenties but can live into their eighties. They will not learn everything in college that they will need the rest of their fifty or sixty years of life (Rubinstein & Firstenberg, 1987). Developing students thinking is "particularly acute today when our culture's output of information far exceeds our ability to think critically about that information" (Meyers, 1986, p xi) 12

PAGE 26

People need self-directed learning skills because they usually know better than anyone what they know, need to learn. want to learn, and how to learn (Dolmans & Schmidt, 2000). In an age where people must always be learning at work to keep their jobs earn promotions converse intelligently and understand newsworthy issues around the globe, they must learn new things. The workplace requires constant learning because the quantity and quality of information grows daily in every field. Self-directed learners know what they want and need in their lives and manage their learning to meet their own objectives (Hiemstra, 1998). Critical thinking can help people to write well because thinking and writing reflect each other (Coe, 1974). High quality writing reflects high quality thinking (Emig, 1977; Flower & Hayes, 1988). When students rush through the writing process. they do not give themselves time to think about the message they intend to communicate; the result is usually poor, ineffective writing. Conversely, when students engage in the writing process and take time to think critically, draft, elicit feedback, and revise, their written products tend to be well-structured presentations oftheir thoughts (Young, 2003). Writing teachers have the opportunity and responsibility to boost their students' critical thinking and self-directed learning skills (McPeck 1981 ). 13

PAGE 27

The Challenges Critical Thinking and Self-Directed Learning Pose for Writing Teachers Writing teachers face the problem, however, of enabling students to move from using ordinary, everyday, social language to individualized, intentional critical thinking, or from knowledge telling to knowledge transforming (Bereiter & Scardamalia, 1987). Knowledge telling is reporting information. Knowledge transforming is creating new ideas. "It is imperative that citizens of the 201h and 2151 centuries trunk critically, yet recent tests have shown that only 25% to 50% of 1st year college students have the skills needed for logical thought" (Halpern, 1989, p. 3 7). Without challenges to go beyond everyday language use and transform knowledge, students tend to use shallow language and thinking in their writing (Flower, 1994a). Consider the following writing assignment as a way for students to transform knowledge: "IdentifY a problem that you are currently facing as a freshman at this school that other freshmen also face. What are your resources for solving the problem? What realistic solutions might eliminate or alleviate the problem? Write a feature story for the school newspaper that evaluates and argues for different solutions depending on different types of problems that freshmen typically encounter." In contrast, an assignment cue that endorses knowledge telling might be phrased as follows: "What is the most common problem that college freshmen face? List and explain five ways to deal with it." To write an essay 14

PAGE 28

responding to this question, students would engage in knowledge telling because the requirement is to merely recall information from an external source, not to critically think about it, synthesize, apply, or evaluate it. Part of the challenge, responsibility, and opportunity in moving students' critical thinking and self-directed learning skills from telling to transforming lies in empowering students to collaborate and converse as writers: College and university teachers (and textbook writers) cannot tell students how to write. Instead, because writing is itself a displaced form of conversation, teachers have to find ways tor students to learn to engage in constructive conversation with one another about writing. (Bruffee, 1993, p. 57) To improve their writing, teachers and students should write and speak as writers, not as students feigning to be writers (Bruffee ). Authentic workshop activities engage students as real writers to improve their writing effectiveness for real audiences. Such workshops transcend the issue of grades and focus on writing well (Flower, 1994c ). Professional writers usually ask their peers for feedback on their work, and so should student writers. Basically, the more auihentic the learning environment. the more authentic, and critical, the learning; in a problem-centered learning environment, students are genuinely and conscientiously engaged (Barrows 1985). Thus, a problem-centered learning environment redefines the rules and regulations of classroom learning, compels the instructor to release control of the learning process, allows students to develop as authentic writers, and frees students to do so 15

PAGE 29

(Cottell, 2002). The challenge for writing teachers is letting go of control and creating the kind of learning environment that encourages engaged critical thinking and self-directed learning. Another challenge that critical thinking and self-directed learning pose for writing teachers is how to evaluate these skills. An effective way to evaluate such skills is to ask the learners evaluate themselves, even though problems inherently arise when assessing writing. Huot (2002) notes two assumptions about assessment and teaching writing: one assumption is that "in literate activity assessment is everywhere"; the other assumption is that "being able to assess writing is an important part of being able to write well" (p. 165). Hout continues to explain how self-assessment is valuable for writers because it helps them to develop as critical thinkers and self-directed learners. After all, "learning to describe what one sees in a text is an important part of being able to develop the critical consciousness necessary for a developed evaluative discourse about writing" (p. 177). In sum, the challenges that writing teachers face when promoting self-directed learning and critical thinking involve assessing learning and progress. Unique Tools In an attempt to understand the relationships between thinking critically and writing in a problem-based learning environment, a Cognitive-Composition Model and Instruction Inventory were developed for this study. The Instruction Inventory merges these constructs into a discipline-generic survey that elicits teachers' and 16

PAGE 30

students' perceptions about how instruction is delivered and received. The Cognitive-Composition Model provides a continuum that combines the Instruction Inventory with rhetorical theories, whereby opposite ends of the continuum contrast cognitive and behavioral attributes of writing students. Links between theories, Cognitive-Composition Model, and Instruction Inventory are detailed in the Review of Literature in Chapter Two. Methodologies of creating these tools are detailed in Chapter Three. Chapter Four presents analyses of the results of the instrument implementation and conclusions and future research comprise Chapter Five of this dissertation Overview of Methodology Survey research methodology and construct validity generated the data for this study. Theory and research comprised the construction of four continua, from which a Cognitive-Composition Model was created. An Instruction Inventory, or survey, was then written. To keep it discipline-independent the survey encompassed only three of the continua, excluding the continuum illustrating rhetorical theories for teaching writing. Figure 1-2 illustrates which continua contributed to the model and which contributed to the survey instrument. 1be double-headed arrow shows how the instrument and model inform each other. The Instruction Inventory is composed of two sections: a section on problem-centered attributes of instruction and a section on self-direction and critical thinking in learning. Units of measure were fifteen college composition classes with 17

PAGE 31

-----------------------------------------eighteen students and one teacher in each class. Complete anonymity was assured because teachers created their 0\\11 codes and shared them only with Problem-based Learning/ Teacher-based Learning Continuum Self-directed Learning/ Other-directed Learning Continuum Critical Thinking/ Good Thinking Continuum Epistemic Rhetoric/ Current Traditronal Rhetoric Continuum ---.... Instruction Inventory Figure 1-2. Links between theories Cognitive-Composition Model and Instruction Inventory construction. their students. Questions included fifty-four Likert-type scaled statements in one survey for teachers and a complementary survey for students. A detailed process was followed to validate the constructs scrutinized in the study: (a) a draft of the survey was distributed to a panel of judges to evaluate its contents, (b) the survey was revised, (c) a pilot study was implemented in which the survey generated qualitative data, (d) the survey was revised again (e) the survey was distributed to writing instructors and their students (f) data were analyzed. Analysis included a series of reliability tests and factor analyses, including factor 18

PAGE 32

and principal component analyses and oblimin and varimax rotations. It is important to note here that the implementation of the Instruction Inventory in composition classes was necessary to test the data's reliability and validity. Equally, its implementation revealed relationships between teachers and students perceptions of self-directed learning and critical thinking in a problem-centered instructional environment. To determine the per\:eived relationships between the constructs, students' total scores were compared to their teachers' total scores on the Instruction Inventory Applications and Significance of Research Awareness ofthe Cognitive-Composition Model can give Miting teachers insight into understanding their students' thinking and Miting. If a writing teacher sees a student listing information in an essay without critically thinking about it, telling knowledge, he or she can encourage that student to synthesize, or transform, the information into a purposeful essay on the topic. By contrasting cognitive with behavioral composition on a continuum, the model can help teachers identify if a student is more cognitive or behavioral and encourage their development in a particular direction with metacognitive questions (Barrows, 1988; Stepien & Gallagher. 1993; Wilkerson. I 996). The Instruction Inventory can help teachers understand how their perceptions of instruction and learning compare to their students' perceptions. In so doing, they can make informed decisions about their instructional design and 19

PAGE 33

delivery of activities for students. It can reveal discrepancies and similarities between teachers' and their students' perceptions of instructional strategies used in the composition class and the nature ofthose discrepancies. It is important to note, however, that this instrument should be implemented near the end of a semester, the time when students are best in tune with their teachers' instructional style and teachers are most knowledgeable about their students' learning styles. Consequently, information gained from the Instruction Inventory will inform the teachers' subsequent class designs, not necessarily the current class. In short, the Instruction Imentory functions better as a summative, not formative, assessment tool. When teachers know whether, or to what degree, their students are using particular skills, they can refine their instructional design and delivery to improve their teaching effectiveness (Roche & Marsh, 2000). If teachers believe their students are using self-directed and critical thinking skills when they are actually not, the instructors' efforts may be ineffective and useless. Changes that teachers make based on students perceptions can and should be made, however, as student ratings of college faculty are typically valid assessments (Koon & Murray, 1995). Why should a teacher blindly assume students are developing as self-directed, critical thinkers when the Instruction Inventory can validate or negate their assumptions? Ultimately, the Cognitive-Composition Model and Instruction Inventory can help writing teachers of all levels, not only in higher education, to 20

PAGE 34

understand their students' use of critical thinking, problem-solving, and selfdirected learning skills. It can also guide them through the continuous challenge of how to encourage students to move from superficial to critical thinking, or how to reflect knowledge creation in their Learning the skills of clear thinking can help everyone to recognize propaganda and thus not fall prey to it, to analyze unstated assumptions in argwnents, to realize when there is deliberate deception, to consider the credibility of an information source, and think a problem or a decision through in the best way possible. (Halpern, 1989, p. 7) In essence, writing teachers have the delightful duty to create learning environments that inspire their students to be self-directed learners, critical thinkers, and constructive writers. 21

PAGE 35

CHAPTER TWO LINKING STUDIES IN CRITICAL THINKING, PROBLEM BASED LEARNING, AND COMPOSITION STUDIES This literature review contains three major topics: critical thinking, problembased learning, and composition studies. The first section defines critical thinking and explains its uniqueness. The second section describes problem-based learning as a foundation of problem-centered instruction, barriers to problem-based learning, its outcomes and advantages, and self-directed learning, an essential component of problem-based learning. The final section reviews everyday necessities, goaldirected thinking, reflective learning, discovery process, and rhetorical theory as they pertain to teaching writing. The summary of this literature review explains the connections between a problem-centered orientation in writing classes and students' releYant propensities toward critical thinking, problem-solving, and self-directed learning skills. Critical Thinking Critical thinking empowers adults to make sensible decisions and solve challenging problems every day of their lives. "Thinking organizes reality for us, and our own place in those realities, and creates alternative realities that we might hope to achieve (or to avoid)" (Smith, 1990, p. 30). A widely accepted definition of 22

PAGE 36

critical thinking emphasizes both its process and product: "Critical thinking is reasonable and reflective thinking that is focused upon deciding what to believe or do" (Norris & Ennis, 1989, p. 1 ). The process explains how to think critically, reasonably, and reflectively. The product describes the results of thinking critically, or making decisions about future actions. Table 2-l illustrates a continuum of learners' critical thinking skills. Critical thinking should receive proper attention in school: "The development of thinking skills has always been a problematical aspect of teaching; but it is panicularly acute today, when our culture's output of information far exceeds our ability to think critically about that information" (Meyers, 1986, p. xi). Developing thinking skills does not have to be problematic; however, the following seems evident in college composition courses: Written assignments are equally valuable to the instructor because they reveal, at least in part, what student thinking processes are like. Though it is a time-consuming process, sitting down with student to go over papers allows teachers to help the students see more explicitly their own thought processes and thus become more aware of their progress in developing new modes of critical thinking. (Meyers, 1986, p. 86) Writing instruction can promote effective critical thinking because "teaching thinking is analogous to teaching writing" (Ruggiero 1988, p. 11 ). Two critical thinking skills, problem solving and self-directed learning, are traits of mature adult learners (Knowles, 1975) yet sometimes overlooked in college composition 23

PAGE 37

Table 2-1 Continuum of Learners Critical Thinking Skills Critical Thinking Surface Thinking Focus ... ... Student Attributes 1. Problem solving Cl. Analyzes problems. 01. Stays away from (Norris. 1992) .r_roblems when possible. '1 Evaluation skills C2. Evaluates claims, 02. Accepts claims. assumptions, and assumptions and evidence (Smith, evidence on value. 1990) 3. Learning A nitude C3. Shows imaginative 03. Shows an objective open-mindedness and indifferent approach to intellectual curiosity. learning (Norris & Ennis, 1989i 4. Idea generation C4. Considers new ideas G4. Depends on former ideas. (Barnet & Bedau, 2002) I 5. Summarizing CS. Summarizes G5. Summarizes arguments skills arguments accurately I incorrectly. (Barnet & Bedau, I 2002) 6. Self-reflection C6. Uses judicious, or G6. Accepts information reflective, skepticism without much retlection (McPeck, 1981) i or contemplation. 7. Construction of C7. Constructs alternatives I 07. Applies first solutions solutions and solutions in developed in problemproblem-solving. soiving. (Meyers, 1986) 8. Argument C8. Can recogniz e and 08. Does not deliberately construction construct sound recognize or analyze arguments (Ruggiero, sound arguments 1988) 9. Application of C9. Applies principles of 09. Vague application of logic formal and informal formal and informal logic logic. (Barnet & principles. Bedau, 2002) I 0. Understanding CJO. A voids fallacies in I 010. Uses logical fallacies in fallacies reasoning (Ruggiero, reasoning. 1988) 1 24

PAGE 38

Table 2-1 (continued) Critical Thinking Surface Thinking Focus Student Attributes 11. Making C II. Suspends conclusions conclusions (McPeck, 1981) 12. Accepting truths C 12. Sets aside own ideas G 12. Depends on others ideas 13. Motivation 14 Logical Inferences of the truth and I I of the truth without reflects on reflection. alternatives (Meyers j 1986) I C13. Internally motivated about learning. (Hal rn 1989) Cl4. Can judge and make logical inferences. (Norris, 1992) G l3. Externally motivated about learning. G 14. Hastily judges and draws logical inferences. courses. One of the biggest challenges for college teachers is to help their students develop these thinking skills. Teachers should also expose students to new values and perceptions beyond their immediate learning experiences, a vital ingredient to critical thinking and writing (Meyers. 1986). Instead, many writing teachers traditionally pay attention to writing research papers with painstaking attention to correct punctuation and grammar (Berlin. 1988). Writing instruction can and should however, advance critical thinking skills because "Writing is a traditional, obvious, and easy way to encourage students to become creative and critical themselves (Smith, 1990 p. 126). In turn, these creative and critical thinkers can become savvy information consumers and productive, intelligent citizens. Even if teachers believe they are promoting critical thinking, how do they know if their students are truly 25

PAGE 39

using these skills? The snag here could be that the definition of critical thinking varies from teacher to teacher and student to student, making its assessment difficult. Definitions Almost one hundred years ago, Dewey (191 0) wrote the following about thinking: The most important factor in the training of good mental habits consists in acquiring the attitude of suspended conclusion, and in mastering the various methods of searching for new materials to corroborate or to refute the first suggestions that occur. To maintain the state of doubt and to carry on systematic and protracted inquiry-these are the essentials of thinking. (p. 13) McPeck (1981) concurs with Dewey's concept of"suspension" and defines critical thinking as "the appropriate use of reflective skepticism within the problem area under consideration. Knowing how and when to apply this reflective skepticism effectively requires the thinker to know something about the field in question" (p. 7). McPeck continues to explain that skepticism and the suspension of making quick decisions is a significant attribute of critical thinking. His use of reflective skepticism mirrors Dewey's definition of reflective thought, which Dewey defines as "Active. persistent. and careful consideration of any belief or supposed form of knowledge in the light ofthe grounds that support it. and the further conclusions to which it tends" (p. 6). In essence, McPeck's and Dewey's definitions of critical 26

PAGE 40

thinking and reflective thought are in sync with each other and emphasize the notion of intentional deliberation of the issue at hand. Skepticism is an intellectual obligation, however, according to Elbow ( 1973). The seventeenth century harbored a skeptical ideology; that is, scientists prided themselves in doubting theories and not being gullible. In turn, We now have a state of affairs where to almost anyone in the academic or intellectual world, it seems as though when he plays the doubting game he is being rigorous, disciplined, rational, and tough-minded. And if for any reason he retrains from playing the doubting game, he feels he is being uninteHectual, irrational, and sloppy. (p. 151) Elbow claims such skepticism and doubt tends to promote a disengaged, or closed perception toward new ideas. While this type of doubt has an intellectual value, it is only haif of the game. The "believing game" is an approach to thinking in which the only chance of correcting a mistake is to affirm or believe, but not to argue. The more people believe and experience more things instead of doubt them, the more open they are to change their minds, open their perspectives, take risks, work in groups, and listen to others. Elhow cautions that believing takes time because it is engaging and deliberative; interestingly, Dewey and McPeck describe critical thinking as engaging and deliberative also (Greenfield, 1987; Halpern, 1989). Norris ( 1992) adds a different spin on the role of reflection in critical thinking and posits that critical thinking is actually a description of how people reflect on others' thinking in writing and speech. "Critical thinking, then, is thinking 27

PAGE 41

about thinking, what some have referred to as metacognition" (Norris, p. 184). Thinking about others' thinking involves analysis and evaluation of others' statements Barnet and Bedau (2002) see such roles of analysis and evaluation prevalent in critical thinking: Thinking critically involves, along with imagination (so that we can see our own beliefs from another point of view), a twofold activity: analysis, finding the parts of the problem and the separating them, trying to see how things fit together; and evaluation, judging the merit of our claims and assumptions and the weight of the evidence in their favor. (p. 4) Halpern ( 1989) also emphasizes evaluation and analysis in critical thinking, as they relate to problem-solving and decision-making: The term critical thinking is used to describe thinking that is purposeful, reasoned, and goal directed. It is the kind of thinking involved in solving problems, formulating inferences, calculating likelihoods, and making decisions. The "critical" part of critical thinking denotes an evaluation component. (p. 5) Another definition that incorporates evaluation in critical thinking is Drake's (1976) definition: "Judgments about problem terms, arguments, laws, principles, hypotheses, and statements comprise critical thinking, then" (p. 46). In addition to the aforementioned definitions of critical thinking, McPeck (1981, p. 13) offers a list of the major descriptions of critical thinking: (a) critical thinking is not a distinct subject because people are always thinking about something in particular, (b) criteria to evaluate critical thinking vary from field to field, (c) critical thinking does not necessarily entail disagreement, rejection, or 28

PAGE 42

deviation from norms, (d) critical thinking includes not only assessment of statements, but problem solving and active engagement in certain activities as well, (e) critical thinkers in one subject may not be critical thinkers in another subject, (f) the study of logic does not constitute critical thinking, (g) critical thinking is a dimension of rationality. In sum, critical thinking involves reflection, active engagement, process, product, suspended conclusions, open-mindedness, skepticism, careful consideration, metacognition, evaluation, analysis, formulating inferences, making decisions, domain specificity, assessment of statements, achievement, problem solving, and self-directed learning. Attitude and application is integral to critical thinking, also, as people must use the critical thinking skills they possess to keep them sharp (Halpern, 1989; Norris & Ennis, 1989). Thinking critically requires sustained effort and commitment to fulfill its value and nurture its growth over time. Uniqueness ofTime and Environment Time distinguishes critical thinking from other thinking. Dewey ( 191 0) wrote, "If the suggestion that occurs is at once accepted, we have uncritical thinking, the minimum of reflection'' (p. 13 ). This implies that uncritical thinking is done impetuously and that reflection requires deliberation. Evaluating thinking processes, outcomes of the thought process, and conclusions also require significant time (Halpern, 1989). Greenfield (1987) implies the valuable time requisites of teaching thinking skills: "Any kind of attention to such skills and type of instruction 29

PAGE 43

in thinking skills lead to benefits. Ignoring problem solving entirely or assuming that students acquire thinking skills simply by practice in solving problems, drill, or osmosis does not" (Greenfield, p. 20). The time it takes to strengthen critical thinking and problem solving in school is more important than drilling on content for some teachers: "I believe that paying more attention to process is worth the sacritlce of some content in our courses" (Stice, 1987, p. 98). Although time and content are two costs of teaching critical thinking and reasoning skills, the rewards are worthwhile (Lochhead & \Vhimbey, 1987). A unique learning environment also distinguishes critical thinking from other thinking. The learning atmosphere must value process over product, discovery over correctness. It must tolerate mistakes and encourage students to learn from them. The environment that nurtures critical thinking must encourage students to be adventurous and creative, free from embarrassing criticism and hurt feelings from teachers and classmates (Ruggiero 1988). To promote critical thinking, the learning atmosphere should also "provide frequent challenges in the form of problems and issues appropriate to students' level of intellectual development, problems and issues that afford them practice in thinking skills" (Ruggiero, p. 97). These challenges "allow them to struggle with problems and issues long enough to develop strategies for dealing with confusion and frustration, but not so long as to defeat them" (Ruggiero, 1988, p. 97). Even though the critical thinking classroom atmosphere is invigorating, tolerant, and 30

PAGE 44

creative, many colleges and universities continue to deliver much instruction via lecture format. Lectures tend to be passive, non-creative, teacher-centered forms of instruction that do not typically promote critical thinking (Meyers, 1986). Problem-Based Learning Problem-based learning can create a learning environment that is invigorating, tolerant, and creative (see Table 2-2). It is a process in which problems form the organizing focus and stimulus for learning, teachers are facilitators or guides and learning is student-centered generally occurring in small student groups (Ayres, 2002 ; Barrows, 1996; Barrows & Tamblyn, 1980). Some educators (Ayers, 2002; Barrows, 1996; Woods, 1994) see problem-based learning as a learning environment, or situation, in which the problem drives the learning. In such an environment or process, learners actively construct knowledge with problem-based learning (Brooks & Brooks, 1999; Flower, 1 994a; Gijselaers 1996) Although problem-based learning is often conceptualized as a form of discovery learning, it also provides a framework where extensive social or cooperative learning is encouraged through curricular supports, such as facultyor peer-group-derived course objectives, resource lists, and evaluative criteria. (Zimmerman & Lebeau,2000,p.311) Barrows (1986) describes problem-based learning as exhibiting variable instructional strategies, including lecture, case-based lectures, case methods, and closed loop or reiterative problem-based methods. 31

PAGE 45

Table 2-2 Continuum o.fStudent Characteristics of Problem-Based Learning Problem-Based Learning Teacher-Based Learning Focus IIIII .. Student Attributes 1. Content Pl. Identifies content to be Tl. Learns infonnation learned. (Barrows, 1996) detennined by others. 2. Personalized P2. Personalizes own learning T2. Generalizes own learning. learning based on personal interests. (Barrows & Tamblyn 1980) 3. Knowledge P3. Constructs knowledge T3. Does not knowingly construction actively when learning. construct knowledge (Barrows & Tamblyn, when learning. 1980) 4. Problem P4. Analyzes problems tBirch, I T4. A voids probiems 1986) 5. Reasoning P5. Practices logical reasoning I T5. Thinks compulsively and skills and probiem-soiving skills rashly (Edens, 2000) 6. Participation P6. Participates and is T6 Participates on passive, activity motivated when learning. low level of involvement (Woods. l 994) when learning. 7. Group problem P7. Encourages group problem T7. Withdraws from group solving solving (Allen, Ouch, & I problem solving. Groh 1996) 8. Solution P8. Implements solutions to T8. Does not complete implementproblems solved. (Bridges solution implementation. ation & Hallinger, 1996) 9 Prior P9. Activates own prior T9. Does not acknowledge knowledge knowledge. (Bridges & I prior knowledge Hallinger, 1996) 10. Independence PIO. Learns independently. TIO. Depends on others (Gijselaers, 1996) throughout learning process II Critical PI I. Thinks critically. Tit. Possesses illogical, thinking (Magnussen. Ishida, & irrational reasoning skills. Itano, 2000) 12. Learning Pl2. Transfers learning I T12. Does not transfer transfer responsibility from teacher responsibility for learning to self. (Barrows & I from teacher. Tamblyn, I 980) 32

PAGE 46

In problem-based learning, students assume responsibility for their learning, identify what they need to know, and personalize their learning (Bridges & Hallinger, 1997). Small groups of three to five students engage in collaborative learning (Dolmans, Schmidt, & Gijselaers, 1994a). Students also engage with problem solving through self-study and develop their expertise in specific subject matter. Teachers assume the role of tutors and ask guiding, metacognitive questions of students (Stepien & Gallagher, 1993; Wilkerson, 1996). The anchoring problems in problem-based learning are real-life scenarios that serve as the stimulus for acquiring new knowledge (Bridges & Hallinger, 1997). The problems tell engaging stories in settings to which the students can relate. They are open-ended enough to challenge students, engender controversy, and require students to demonstrate thinking skills beyond simple knowledge and comprehension (Allen, Duch, & Groh, 1 996). These problems present a dilemma, or puzzle, that will elicit different responses from different learners (Kaufman, 1998). Problems should be authentic, engaging, and ill-structured (Gijselaers, 1996; Stinson & Milter, 1996). The difference between ill-structured problems and case studies commonly used in business curricula is that ill-structured problems do not have ready-made solutions, or "answers," as case studies often have (Stinson & Milter, 1996). The function of the problem in problem-based learning is to give students a focus for integrating their learning from many disciplines (Barrows, 1996). Basing learning 33

PAGE 47

around problem solving also requires students to identify sub-problems and challenges throughout the process, as Getzels (1985) notes: And perhaps most important, instruction in finding and formulating problems is well worth educational attention not only as incidental to problem solving, which is typically what has been done in school if it was done at all, but as an important focus of interest in its own right. (p. 60) The basic sequence for problem-based learning begins with students identifying the educational, or learning, objectives for the project to focus their learning; tutors also define learning objectives but may or may not disclose them to the students (Barrows, 1985, 1986, 1988). A common theme in problem-centered instruction is that the teacher identifies the learning objectives, sometimes with input from students, and creates the problem to facilitate development (Woods, 1996}. Next, students interact with the assignment in group discussions and selfdirected study. They do not hear lectures about the topic or read textbooks their teachers assign for background information. As a result of exploring the nature of the problem at hand, students note L;.e questions that they cannot solve with their prior, or current, knowledge bases. They identify the resources that can fill in their information gaps, much as a doctor requests further testing to make an informed diagnosis or a writer pursues further research to write an article. The next step is to apply the information that they acquired in group discussions and self-study to the original problem to determine whether the solution is viable. Important steps come late in the problem-based learning process, such as 34

PAGE 48

application of the solution to the problem, but they are perhaps the most overlooked steps: students review, synthesize, and evaluate their learning process and products. Reflection throughout the problem-based learning process enhances it considerably (Woods, 1994 ). Problem solving, then, is a cognitive process to derive the best answer to an unknown, the problem, which is something the learner has not encountered earlier (Woods, 1996). A problem-based learning teacher is kno\.\n as a tutor (Barrows & Tamblyn, 1980). Instead of directing students to recite the correct answer to a problem, tutors coach and guide students through repeated practices in reasoning and self-directed study, improving their self-assessment ski1ls in the process (Edens, 2000). Tutors do not function as ultimate knowledge sources depositing knowledge into students' heads (Freire, 1 990). While tutors may use more direct instruction in a lesson initially, to help scaffold the reasoning processes. eventually they withdraw as students learn to assume responsibility for their own learning (Stepien & Gallagher, 1993). A demanding role for a tutor is to find a balance between allowing students to explore issues and guiding them to discover critical learning issues (Ayers, 2002; Gijselaers, 1996). Tutors are generally experts in their area of study; similarly, they must also be experts in the problem-based learning tutor role. Although the role of the tutor is clear, researchers do not fully understand tutors' conceptions oflearning that results from problem-based learning (Dolmans, Schmidt, & Gijselaers, 1994b; Dolmans, Gijselaers, Moust, Grave, & VanDer 35

PAGE 49

Vleuten, 2002). Lack of this literature engenders a need to understand tutors' impressions of teaching and learning in the classroom. Tutors, and all teachers, would benefit from knowing how their metacognitive questions affect students' learning (Barrows, 1988). Educational Outcomes and Advantages Educational outcomes supported by problem-based learning include acquisition of an extensive knowledge base, development of an effective reasoning process, advancement of effective self-directed learning skills, increased motivation for learning, early immersion into a learning culture, structured knowledge for use in clinical contexts, and development of team skills (Bridges & Hallinger, 1997). Educational advantages include logical reasoning, discovery of personal educational needs relevant to future related tasks learning information in the context of understanding a problem. reinforcing information with reuse, integrating information from many disciplines, learning and evaluating self-evaluation, and self-directed learning skills (Barrows & Tamblyn, 1980). One of the most important outcomes of problem-based learning is acquiring an extensive knowledge base (Dolmans & Schmidt, 2000). Developing a reasoning process should be conscious so that students can be aware of how they reason through a problem and transfer it to future learning situations (Bridges & Hallinger, 1997). The development of self-directed learning skills, another educational objective of problem-based learning, is vital for learners, as self-directed learners 36

PAGE 50

assume responsibility for their learning, are able to identify their own learning needs, and know how to acquire and assess them (Candy, 1991 ). Improving communication and interpersonal skills are lifelong skills for adults to foster, as is kindling an internal motivation for learning throughout life (Barrows, 1996; Bruhn, 1997). Cultivating team skills, another educational objective of problem-based learning, encourages students to learn in small groups and helps them to generate and discover ideas that they might not have encountered on their own (Barrows, 1985; Bridges & Ballinger, 1987; Wilkerson, 1996; Woods, 1994). Educational advantages of problem-based learning include student responsibility for learning, acquisition of interdisciplinary content, instructor involvement as mentor and coach, and participation of students in the evaluative process (Allen & Rooney, 1998; Rangachari, 1996). Problem-based learners have shown that they retain their knowledge, know how to apply it appropriately, demonstrate mastery of leadership skills invest significant effort while learning, and view their preparation as meaningful and enjoyable (Bridges & Hallinger, 1997). Students activate their prior knowledge in problem-based learning, create a context that resembles the situation in which students later will use their knowledge, and elaborate their newly acquired knowledge at the time it is initially learned (Bridges & Hallinger). Problem-based learners also appear to have "broader social and interpersonal skills, a greater appreciation ofthe complexity of problems and the resources available for solution, and a heightened motivation for continued self37

PAGE 51

learning" (Bruhn, 1997, p. 67). Such learners also tend to become more independent as a result of the experience they gain in thinking about problems (Seltzer Hilber, Maceli Robinson, & Schwartz, 1996). Further, students' learning sharpens when they experience some type of failure and reflect on how and why that happened a common occurrence in problem-based learning (Bridges & Hallinger, 1996, 1997). Central purposes of problem-based learning are to foster critical thinking, develop students' confidence and skill in dealing successfully with unanticipated issues under practical constraints and transfer responsibility for learning from the teacher to the student (Kaufman, 1998). Problem-based learning provides a method of inquiry which characterizes research and supports the idea that teaching and learning are realized in a research atmosphere (Birch, 1986 ). In this type of learning, lectures become discussions where students actively engage in learning class focus shifts from facts to meaningful information that could help solve problems, and teachers become resources for processes rather than sources of answers to be memorized in order to pass tests (Aspy, Aspy, & Quinby, 1993; Bridges & Hallinger, 1997; Bruffee 1993). Problem-based learning also promotes transfer of )earning between disciplines (Bridges & Hallinger, 1997). Barriers Although the educational outcomes and advantages of problem-based learning seem evident. research does not definitively confirm evidence to support it. Albanese and Mitchell ( 1993) for example conducted a meta-analysis of problem-38

PAGE 52

based learning environments for medical curricula. TI1ey suggested that problem based learning is a valuable approach to learning, but that evidence of its strengths is not conclusive. They strongly suggested that it be continuously studied and its benefits analyzed to fully understand problem-based learning. Problem-based learning, however, can challenge and inhibit tutors and students in many ways. "Problem-based learning projects seldom run smoothly" (Bridges & Hallinger, 1996, p 58). "One of the primary problems with gaining an accurate assessment of the degree of success problem-based learning enjoys is that no overall format or scheme delineates points at which data should be gathered, the type of data needed, and how it should be analyzed" (Bruhn, 1997, p. 68). Also, students frequently ask for more structure and guidance than is typically given in a problem-based learning environment. They often complain of vague assignments because they are used to direct instruction from their primary and secondary teachers. Students new to problem-based learning are typically uncomfortable \\-ith their tutors' encouragement to discover learning strategies on their own (Edens, 2000). A common impediment to problem-based learning has been academics' conservatism, whose tendency is to avoid interactive methods of teaching and learning and student self-assessments. These instructors prefer to do research to further their professional advancements, not to learn how to engage students in unstructured, spontaneous learning (Birch, 1986). In successful problem-based 39

PAGE 53

learning, curricula offer interdisciplinary content similar to more conservative approaches The difference is that problem-based learning students collaborate as they work on their lessons and their instructors act as mentors and coaches (Allen & Rooney, 1998 ). Collaborative skills are sometimes underdeveloped in problem-based learners and can hinder learning, another barrier to address. Woods (1994) stresses the importance of collaboration in learning: ''We learn more effectively and efficiently if we work actively and cooperatively to learn knowledge. Yet working in groups is hard work. For groups to be effective, we need to be skilled in interpersonal relations and in group process" (pp. 4-3). Problem-based learning participants develop effective interpersonal and interaction skills, including consensual decision-making skills. dialogue and discussion skills, team maintenance skills, conflict management skills and team leadership skills (Ayers, 2002). Fenwick (1998) presents yet another critical perspective toward problem based learning and writes "We suggest that a problem-based approach to professional education is ontologically narrow and epistemologically inconsistent with the lived nature of professional practice'' (p. 53). That is the nature ofthe problems in problem-based learning are contrived according to Fenwick and therefore, inauthentic, contrary to a major prerequisite of problem scenarios. He continues to explain that framing life in terms of problems locks the world into an undesirable chaos that needs a redeeming order. Fenwick's bias is averse to 40

PAGE 54

focusing learning around chaotic, confused situations, assuming, of course, that the problem scenarios are indeed disorganized. These adversities and potential barriers, plus fear of the unknown, scarcity of resources, and lack of student empowerment may explain why problem-based learning is not more popular than it is today (Ayers, 2002). Perceived weaknesses of problem-based learning can be erroneous, however. For example, Getzels ( 1985) claims that discovering problems is a creative process, integral to creative thinking and problem solving. Opponents of problem-based learning, supporters of Fenwick (1998), could claim that the creative, problem discovery phase is lost in problem-based learning because the teacher supplies the problem for the students. However, learning is complex fuid incorporates many levels of sub-problems that students can discover while working through the problem-solving process. Getzels explains that problem finding and problem solving meld into each other: "At the core of an effective solution is a productive problem. It is this that makes what I have referred to as the 'problem ofthe problem' such an important subject for inquiry and instruction in our schools" (Getzels, 1985, pp. 6061 ). In short, discovering and solving problems is inherent in the learning process and demands attention in school. "Problem solving requires an integrated use of thinking skills and an appropriate knowledge or data base" (Ruggiero, 1988, p. 23). Students who are focused on discovering and solving their own learning needs are self-directed learners and engaged in the problem-solving process. 41

PAGE 55

Self-Directed Learning Self-directed learning is an educational outcome of problem-based learning (Barrows, 1994) and merits individual attention because self-directed learners tend to be critical thinkers, characteristics that schools aim to develop in students: I also think that every act of teaching should have built into it some provision for helping the learner become more self-directing. I recognize that there are situations in which a person is indeed dependent in some respects and that in these situations it is appropriate for him or her to be taught or directed. But I don't think that it is healthy--or even humane-for a person to be kept permanently dependent upon a system or upon another person. (Knowles. 1975, pp. 10-11) As illustrated in Table 2-3, several things are known about self-directed learning: (a) individuals can become empowered to assume responsibility for decisions about their learning, (b) self-direction is a continuum of characteristics that exists in all people, (c) self-directed learning does not imply learning in isolation, (d) self-directed learners tend to transfer learning between situations, and (e) self-directed study can involve many activities and resources, and roles for instructors in self-directed learning environments are as knowledgeable resources (Hiemstra, 1998). Self-directed learning, then, is a process in which the learner feels empowered enough to assume the responsibility to control her or his learning by making decisions involving the activities and resources that will promote constructive learning (Mullen, 1992). Individuals are more likely to learn when they are able to respond to an event with self-directed thinking (Morris & Moore, 2000). 42

PAGE 56

Table 2-3 Continuum of Learners' Se(f-Directed Learning Skills Self-Directed Learning Other-Directed Learning Focus Student Attributes I. Self-reliance S1. Self-reliant in finding 01. Needs help finding information. (Hiemstra. information. 1998) 2. Personal goals S2. Sets personal learning I o2. Does not set personal goals. (Garger, 1999) learning goals. 3. Learning in S3. Enjoys learning in 03. Does not enjoy learning isolation isolation. (Hiemstra. in isolation. 1998) 4. Time for learning S4. Finds time tor learning. 04. Does not find time tor (Licklider, 1997) learning. 5. Feedback and S5. Accepts feedback and 05. Does not accept feedback criticism criticism. ( Anfield, 1997) and criticism. 6. Learning control S6. Takes control of personal 06. Does not take control of learning. (Brookfield, personal learning. 1993) 7. Seeking learning S7. Seeks out learning I 07. Does not seek out opportunities. (Gerber, learning opportunities. Lankshear, Svensson, 1995) 8. Problemsolving S8. Enjoys problem solving. 08. Does not enjoy problem (Hiemstra. 1997) solving. 9. Challenging S9. Challenges assumptions. 09. Does not challenge assumptions (Pilling-Cormick, 1997) assumptions. 10. Responsibility S10. Assumes responsibility 010. Does not assume for learning for own learning. responsibility for own (Knowles, 1975) learning. I I. Self-evaluation Sl1. Evaluates and advocates 011. Does not evaluate or own learning. (Pilling-I advocate own learning. Cormick. 1997) 12. Self-reflection Sl2. Reflects upon and makes I 012. Does not reflect upon or choices about own make choices about own learning. (Candy, 1991) l learning. 43

PAGE 57

Self-directed learners can select an appropriate learning style from a range of styles according to the demand of the situation and their own learning capability (Robotham, 1995) These learners fall on a linear scale that represents preferences toward solving problems independently or solving problems within a group setting (Hiemstra, 1997; Hiemstra & Brockett, 1991 ); indicators on the self-directedness scale include self-reliance in finding information, setting personal learning goals, enjoying learning in isolation, structuring personal learning, finding time for learning, accepting feedback and criticism, taking control of personal learning, setting personal learning approaches, seeking out learning opportunities, and enjoying solving learning problems in isolation. "It is the desire, attitudes, values, and abilities that will ultimately determine the degree of self learning that will take place" (Guglielmino & Murdick, 1997, p. 1 1 ). Leamer control is a central aspect of self-directedness (Brookfield, 1 993; Lowry, 1989: Tremblay, 1992). Self-directed learners typically challenge assumptions their teachers make about their learning (Pilling-Cormick, 1997). Giving responsibility to learners has shown to benefit those learners (Guglielmino & Guglielmino, 2001; Hatcher, 1997; Hiemstra, 1998). When adults learn how to assume responsibility and empower themselves for their own sake, their self directedness also promotes their self-advocacy, self-determination, and metacognitive learning (Hiemstra & Brockett, 1991 ). "Self-direction in learning is a 44

PAGE 58

combination of forces both within and outside the individual that stress the learner accepting ever-increasing responsibility for decisions associated with the learning process" (Brockett & p. 9). The role of the teacher in self-directed learning environments shifts from instructor, to facilitator, to learning resource as the locus of responsibility for the learning shifts from the instructor to the learner (Robotham, 1995). This allows the learner to retain the freedom to explore and develop personally within an environment where the teacher provides a minimal amount of structure. Independent thinking is hardly likely to occur when students are told precisely what they should learn, and what they should think about it. And students are also not likely to improve as independent thinkers when they see teachers themselves being told precisely what they should teach, and how they should teach it. (Smith, 1990,p. 128) As self-directed learners reflect on their learning, they determine, investigate, and evaluate their needs (Pilling-Cormick, 1997). When learners reflect and make choices about their learning, they demonstrate openness and willingness to consider learning approaches from different perspectives (Gerber, Lankshear, & Svensson, 1995). Reflection is an important aspect ofthe self-directed learning process because it increases awareness ofhow cultural factors can influence the pursuit of exemplary learning projects (Brookfield, 1 Q93). Barriers to self-directed learning also include challenges that many learners face because their school teachers may have conditioned them to be other-directed learners (Hiemstra, 1998). Other-directed learning is teacher-centered learning, in 45

PAGE 59

that the teacher decides what, how, and when students will learn. Initial resistance to self-directed learning is common among learners until they have a successful experience with it (Hatcher, 1997). Assuming control of learning, however, brings the learner into direct conflict with entrenched issues; as learners decide their own learning plans, they may clash with authoritative voices that had been responsible for deciding what and when they learn (Brookfield 1993). Critics of self-directed learning believe that it is a way for instructors to be less accountable for their teaching because it is a diluted form of education (Hiemstra, 1998). However these critics evidently do not recognize self-directed learning as a significant contribution to the advancement of adult education (Brookfield, 1993). Assessing Problem-Based and Self-Directed Learning Tools and techniques for assessing problem-based and self-directed learning include integrative essays in which students discuss what they have learned during a project and how they might use the knowledge and skills in the future protocols or standards that students may use to evaluate their own performance or products, and models or examples of products completed by expert practitioners against which to compare their own products. They also include knowledge review exercises that test students' ability to apply their knowledge to typical situations, forms created by students to elicit feedback from their peers on aspects of their performance, structured observations that provide descriptive information about individual and 46

PAGE 60

group performance, and probing questions for students to consider in relation to their final performance products (Bridges & Hallinger, 1996). Assuming responsibility in learning includes self-assessment of learning needs (Thomas & Harper, 2001 ). Adults prefer to assume responsibility for their own learning rather than be at the mercy of someone else deciding what, how, and when they should learn (Hiemstra 1998 ). In their research involving continuing professional education of certified public accountants (CPAs), Thomas and Harper claim that self-directed learning activities should stem from learning objectives based on self-assessments of learning needs. This way, CPAs will be able to self assess the impact of these activities based on professional competencies developed by industry professionals. Woods (1996) studied his classes of chemical engineering students learning in a problem-based learning environment and declared its success. "'Over the past thirteen years, we have evaluated om program. On virtually all measures, our students show a marked superiority in the skills that are in demand in the real world-strong support for what started as only a small piece of a larger pie" (Woods, p. 98). Woods also claims that students can self-assess accurately if teachers empower them regularly. An example of employees documenting their self-directed learning, evaluating it, and getting credit for it is Motorola's self-directed learning lab (Filipczak & Ganzel, 1999). One method Motorola used was to compare learning 47

PAGE 61

from using self-directed learning methods with instructor-led methods. Another strategy to measure self-directed learning is with manager-employee interviews and documentation analyzing whether the amount of learning is increased by selfdirected learning for the same or less cost to Motorola (Guglielmino & Murdick, 1997). Fenwick (1998) is as skepticai about evaluating problem-based learning as he is about the quality of the problems: 'Some critical evaluation of problem-based learning has emerged, but this tends to focus on student outcomes conceptualized in terms of the amount of knowledge residue generated by problem-based learning compared to lecture approaches" (p. 53). If evaluation is equated only with research and measurement, Fenwick's biased concerns are well-founded; if not, his concerns are shallow. Schmidt and Dolmans (1995) corrobordte Fenwick's sentiments and explain that outcome data does not provide the quality of information that can judge a smallscale, course-level curriculum: Thus what is needed, in addition to outcome studies as an instrument for accountability, is a strategy that deals with the curriculum in its own right, provides information about its shortcomings, and enables a school to implement improvements whenever and wherever necessary. (p. 82) Schmidt and Dolmans evidently see evaluation as more than objective observation and measurement; they see it as curriculum evaluation, in which the whole 48

PAGE 62

curriculum receives evaluation, not just the learning outcomes of individual students. Part of the challenge in evaluating problem-based learning is that formal assessment procedures are designed to evaluate traditional instructor-led learning. The process and intent of problem-based learning are different, however, from instructor-led learning and require different means of evaluation. According to Barrows ( 1996), not many formal assessment procedures can address problem-based learning's idiosyncrasies because they are generally insensitive to the cognitive and behavioml differences observed in problem-based learning. '"Assessments of the effectiveness of PBL are generally favorable, yet its cost in money and time, as well as its effect on the future behavior of practitioners, has not been established" (Bruhn, 1997, p. 66). A challenge in evaluating self-directed learning is that it involves the learner's life experiences, including learning plans, self-directed learning teams, experience-based !earning, competency-based assessment and education, and past expertise (Boyatiz, Cowen, & Kolb, 1995). The Yalue of experiences is difficult, if not impossible, to measure. Educators should acknowledge such an array of contributors to the overall learning process, but to devise a dependable, rational evaluation plan to do so is challenging. These life experiences can also include dependency of the learner on the instructor to provide learning resources, identify learning strategies, and evaluate learning progress (Robotham, 1995). The instructor 49

PAGE 63

then carries the responsibility to be an expert in the field, but not to overpower already anxious learners (Robotham). Evaluation of problem-based and self-directed learning occurs within the learning process itself, making it a formative assessment for learners (Barrows, 1986). Part of the evaluation process involved in problem-based learning is that teachers should be personally involved in the problem-based process. Firsthand involvement alJows for authentic assessments to understand the issues involved in problem and reasoning skills (Edens, 2000). Evaluating the outcomes of problem based learning should also include an assessment of the observed interaction of students and their instructors (Bruhn, 1997). To assess problem-based learning fully, Barrows ( 1996) claims that teachers should observe problem-based learning in action. Problem-based learning students use an iterative process of assessing what they know, identifying what they need to know, gathering information, and collaborating on the evaluation of hypotheses in light of the data they have collected (Stepien & Gallagher, 1993). As students reflect on their learning processes, se,f-directed and problem-based learners move beyond superficial questioning into the realm of critical learning (Pilling-Connick, 1997). In the development of identifying these needs, learners can also devise appropriate, relevant, intellectual assessment strategies. 50

PAGE 64

Teaching Writing Writing is a complicated intellectual proc e ss that involves a variety of problem-solving strategies (Flower & Hayes, 1988; Hillocks 1995). Within this problem-solving process lies a progression of discoveries for students that reflects a unique mode of learning (Emig, 1977; Perl, 1988). Writing promotes successful learning because it connects and integrates ideas from many perspectives while students can be active and personally engaged (Emig 1977). Accordingly writers draw from a repertoire of strategies for building meaning and learning for themselves (Flower, 1994c). Teaching writing involves teaching thinking, problem-solving, and self directed learning skills simultaneously with composition skills because they are each interrelated and interdependent (Coe, 1974 ; Flower, 1994c ; Rohman 1965). Good thinking can produce good writing but without it, good writing is impossible; bad thinking never leads to good writing (Rohman, 1965). However, as writing students encounter new ideas and ways of thinking in their college writing courses, their rhetorical training (understanding knowledge communication between writer, audience, and purpose) proves incapable of dealing with their new intellectual experience (Berlin 1987) Writing students need to develop their problem-solving self-directed learning, and critical thinking skills to write well about meaningful topics and to develop lifelong skills of thinking critical1y to solve whatever problems they may face. Teachers can and should provide guidance for students, but 51

PAGE 65

they should make it clear to students that correctness is not as important as having the ability to perceive a problem and unravel it (Meyers, 1986, p 31 ) Composition studies or teaching writing, addresses the role of problem solving in many ways. A problem is simply any situation in which you are at point A and need to find some way to get to your goal, point B" (Flower 1989, p. 44). For writers, this can be moving from a list of potential writing topics to a focused thesis, a rough draft to an edited draft, or fifty pages to ten pages of content. "One of the most basic reasons for writing which students academic writers journalists and business people share, is to discuss and deal with problems" (p. 135) Expert and novice writers often use different problem-solving strategies and "the problem-solving approach assumes that there is often a 'better' way and that writers can substantially expand their repertory of strategies (Flower, 1989 p. 3) For example a novice writer might approach topic development by listing ten ideas that interest him and choosing one. The expert writer might brainstorm a list, discuss the items with a friend, revise the list write for a while, and then choose an idea that he has already analyzed and discussed. The theoretical framework for problem solving in teaching writing includes rhetorical theory, the concept that writing consists of a process that always involves the writer reality reader, and language. The following section reviews composition literature and the manifestation of problem solving in teaching writing. 52

PAGE 66

Everyday Necessities Problems are inherent in everyday life. making the process of finding their solutions an everyday necessity: "'In teaching writing, we are not simply offering training in a useful technical skill that is meant as a simple complement to the more important studies of other areas. We are teaching a way of experiencing the world, a way of ordering and making sense of it" (Berlin, 1988, p. 58). Because the writing process itself is a thinking process (Flower & Hayes, 1988), it has much in common with common problem-solving processes, such as taking exams, understanding a patient's illness, or completing a tax form correctly (Flower, 1989). Problemsolving is clearly an everyday necessity. Yet another reason that problem solving is an everyday necessity is because it is a response to the uncertainty inherent in most rhetorical situations: communications with ambiguities in the writer's intentions, reader's capabilities, and the basic message can obscure the effect of the communication. "Writers acquire problem-solving strategies from social interaction, but academic writing is still a set of heuristics, not an algorithm or formula" (Flower, 1994a, p. 23; Bruffee, 1993 ). Problem solving is integral to our everyday activities: Solving problems is essential1y no different from anything else the brain does. Problem-solving cannot be separated from decision making, categorizing, classifYing, inferring, and other aspect of thinking, often listed as distinct skills. Problem-solving can be seen to involve, and be involved in. all these aspects ofthought. They are not distinctive elements but simply ways of looking at, or 53

PAGE 67

talking about, one coherent thing, the brain's construction of realities, the imposition of order on chaos. (Smith, 1990, p. 19) Goal-Directed Thinking People turn to problem solving when they have a problem or a goal that they want to understand in the form of goal-directed thinking (Flower, 1989). For some writers, problem solving transforms the writing process into a "goal-directed journey intent on individual writing purposes ' (Flower, p. 4). One such situation might be a writer composing a letter to the local newspaper editor because of a strong belief about an tssue Problems exist when people feel a conflict between where they are and where they want to be or between the present state and the desired state (Flower); ifthe goal is attainable with minimal effort, the problem is also minimal. Flower and Hayes ( 1988) conducted a study in which they wanted to know how writers define problems for themselves. They also wanted to know how writers analyze their audience, if they do at all, if they develop a full representation oftheir problem and if it helps them generate new ideas. Another objective was to understand if good and poor writers go about this task differently. They found that good writers respond to all aspects of the rhetorical problem or writing task. These writers build a unique representation of their audience, assignment, and goals involving the audience. They also found that good writers create a rich network of goals to affect their readers and these goals helped the writers generate new ideas. 54

PAGE 68

Another finding was that good writers represent the problem in more breadth and depth than novice writers, they continue to develop their image of the reader, and they detail their goals as they write. "The main conclusion of our study is this: good writers are simply solving a different problem than poor writers'' (p. 1 0 1 ). Flower and Hayes conclusions coincide with Getzels' ( 1985) claims that discovering problems is a creative process, integral to creative thinking and problem solving. They conclude with the following: This important study of creativity in fine art suggested that problem-finding is a talent a cognitive skill which can lead to creativity. The parallels between these two studies suggest that problem-finding in both literature and art is related not only to success, but in some less well defined way to ''creativity" itself. (p. 1 01) Other implications of the Flower and Hayes study are that teachers can successfully teach student writers how to explore a rhetorical problem, that problem comprehension is not a mysterious act. Teaching students to explore their own problems freely can help them "to create inspiration instead of wait for it" (p. 1 02). Flower's ( 1989) research points out that when writers think aloud, they mention their goals freely. Typically, writers have four main goals: how they want to affect a reader, the relationship that they want with their reader, their attempt to build a coherent network of ideas and meaning, and the conventional features of a written text tflower & Hayes, 1988, pp. 98-99). Flower has heard student writers give themselves instructions and monitor their writing progress and effort in think55

PAGE 69

aloud research. She also explains that writers are absorbed in reaching their goals, just like all problem solvers. If the goal is developing a topic, for instance once the writer has generated an outline of main and subtopics, he or she does not typically remember the process that generated the outline. Further research with writers' think alouds should reveal more insight into the process that develops their goal-directed thinking. Reflective Learning The writing process entails many creatively cognitive steps and strategies, just as creating a painting or composing a song requires. Flower (1989) has attempted to understand the cognition behind writers' strategies so teachers can develop instructional strategies to help students to write well. "Thus, my experience with this methodology has convinced me that personal learning through self reflective epistemic writing is as powerful a learning tool as we may have in our repertoire as college teachers" (Brown, 1998, p. 138). Flower explores this question of what actually happens when we write and offers several conclusions. She claims that the writing process, because it is a thinking process, involves deep thinking to the point that sometimes, after writing for hours, writers have no recollection of what they did during the time. This is common, even though the two hours were spent actively "planning, setting goals, generating new ideas, drawing inferences from old ideas, looking for relationships or patterns, creating trial text, evaluating our prose, detecting errors, and diagnosing problems and planning ways around 56

PAGE 70

them" (Flower, pp. 31-32). Flower has also identified steps in the composing process and a variety of strategies for taking these steps: Planning Step 1: Explore the rhetorical problem Step 2: Make a plan Generating ideas in words Step 3: Generate new ideas Step 4: Organize your ideas Designing for a reader Step 5: Know the needs of your reader Step 6: Transform writer-based prose into reader-based prose Revising for effectiveness Step 7: Review your paper and your purpose Step 8: Test and edit your writing Step 9: Edit for connections and coherence (Flower, 1989, p. 51) Flower ( 1989) describes the steps in analyzing a problem, which are reminiscent of steps in the composing process. First, the writer defines the conflict or key issues. This is different from merely stating a topic, such as "pollution,'' in that it is stating the problem within the topic, such as "corporations should take an active role in reducing pollution in this country." The second step in analyzing a problem is to place the problem in a larger context, such as understanding the effects of pollution on human well-being. The third step is to operationalize the problem so that the writer can tackle the problem on a manageable scale: "Ninety percent of manufacturing corporations in this country should decrease their airborne pollutants by 75 percent in the next 24 months." 57

PAGE 71

The next step in the writer s process of analyzing a problem is to explore subparts of the problem to understand it thoroughly, such as the number of manufacturing companies that are po11uting the air and at what rates, the progress already made in pollution over the last three years, legislation that has gone into effect, particular health risks of pollution, and the like. The fifth and final step in problem analysis is to come to an open-minded conclusion. This is when the writer comes to a solution or a new definition of the problem; the weakness of most problem solving is leaping too quickly to a solution (Dewey, 191 0). A rhetorical problem can be a vast situation: [It] is like a rather large, uncharted territory that contains you, your reader, your ideas, and all the things you possibly could do. As you make a mental tour of this territory selecting the goais that matter and the strategies you will use, you are building your own image or representation of the task. (Flower, 1989. p. 76) Because of the enormity of the rhetorical problems that writers can face, the fivestep process Flower outlines offers a logical approach, albeit flexible. Reflection plays a key role in the problem-solving and writing process (Perl, 1988). "Writing is a spe c ial form of reflection through which new meaning can be created. new understanding of problems can become circumscribed, and new ways of organizing experiences can be developed (Bringle & 1--Iatcher, 1999 p. 180). When people read and write, they are not only constructing meaning, they are observing and reflecting using a repertoire of strategies for building meaning (Bruffee, 1993; Flower, 1994b ). Writers also reflect on their own goals, 58 --------

PAGE 72

assumptions, strategies, and habits of making meaning (Courtney & Abodeeb, 1999; Flower, 1994a). Collaborative planning is a specific strategy that invites student writers to reflect on what happened in their planning sessions and to consider what they are learning about themselves and thinkers, problem-solvers, and writers (Wallace, 1994; Yost, 2000). Discovery Process In their research, Flower and Hayes ( 1988) found that discovery and the new insights it creates are the end result of a complicated intellectual process The act of creating and discovering ideas is the essence of effective writing (Flower & Hayes). Finding ideas for writing topics is akin to reponing research findings, such as reponing the number of factories that are violating pollution standards. Discovering ideas, on the other hand, is synthesizing findings, such as explaining the number of polluting factories combined based on their locations and the type of product they manufacture to create a new insight into the polluiion problem. Problem solving is a form of discovery, similar to writing and thinking processes. For some writers, writing about problems can change their perceptions about the problem, or their definition of the problem because it opens new doors of understanding (Flower, 1989). If you can define the problem, you may have solved it. The hardest part of solving many problems is trying to discover what the problem really is and define the conflict that makes it a problem. Writing is a powerful way to think problems through, because it 59

PAGE 73

helps you describe and name the conflicting parts of your own thinking. (p. 31) Bruner ( 1961) also saw discovery as an intellectual process and defines it as "all forms of obtaining knowledge for oneself by the use of one's own mind" (p. 237). He explains that discovery is best accomplished by those learners who are prepared to discover new knowledge and emphasizes that discovery is a process of "transforming evidence" (p. 23 7) that allows the learner to create new insights. Bruner also hypothesized that discovery enriches learners' problem-solving skills because it allows people to acquire information viably. Problem Solving as Collaborative Planning Collaborative planning is an integral part of the writing process because it involves writing workshops and giving and receiving feedback on writing. It is socially supported conversation and an opportunity for student writers to talk about their ideas in a socially supportive environment where peers listen and ask critical questions (Wallace, 1994, p. 50). Bruffee (1993) concurs: In collaborative learning students work on focused but open-ended tasks. They discuss issues in small consensus groups, plan and carry out long-term projects in research teams, tutor one another, analyze and work problems together, puzzle out difficult lab instructions together, read aloud to one another what they have written, and help one another edit and revise research reports and term papers. (p. 1) One of the strategies that expert writers use is planning; trial and error tactics belong more to novice writers (Flower, 1989). Likewise, problem-solvers rely on planning 60

PAGE 74

because it saves them significant time in the writing process. Collaborative planning is more of a process than a writing strategy whereby students can reflect on their thinking (Flower, 1994b ). Teachers set the tasks that writing students tackle in collaborative planning. This requires forethought, practice, and planning on the teacher's part (Bruffee, 1993, p. 3 5). Tasks may be questions to be answered by the consensus of the group. For example, articles could be given to small groups of writers for them to decide on the argument and major points of support. Another task for collaborative planning could be problems to be solved so that all group members are satisfied, such as writing a summary of an article. Teachers do not tell students if they are right or wrong because right and wrong answers do not exist as such (Bruffee ). The teacher's role is to give students opportunities to work on specific, open-ended activities collaboratively. This contradicts the traditional role of the teacher as sole source of knowledge and can threaten the teacher's authority Accordingly, most writing students tend not to trust collaborative activities and therefore, do not benefit from its learning opportunities (Bruffee). This could also be an explanation of why peer critiquing in a writing class tends to be superficial and strained. Another challenge with collaborative planning is that many student writers in college struggle with knowing how to plan; they have been conditioned in school to focus on story telling, a form of knowledge telling, and gathering information (Wallace, 1994). Also, student writers need to know how to collaborate, to be active 61

PAGE 75

listeners, and ask for elaboration with probing questions (Bruffee, 1993; Wallace, 1994). Inevitably however, multiple perspectives in a discussion stir conflict. Consequently, writing students present alternative solutions and strategies to each other and clarify their criteria for evaluating their solutions to the problem at hand and elucidate their grounds for supporting their rationale behind the solution (Bruffee, 1993 ). Collaborative learning brings to the surface the relationship between the authority of knowledge and the authority of teachers. By challenging the traditional. foundational understanding of the authority of knowledge, collaborative learning helps college and university teachers begin thinking in quite a different way about what it means to teach. (p. 7) In addition to reviewing the literature about problem solving in composition studies, teachers of writing can benefit from understanding the evolution of rhetorical theory and how the new rhetoric aligns with problem-based learning and self-directed learning (Lauer, 2002). writing teachers should also accept the responsibility to understand basic cognitive structures and how they develop in order to provide productive karning experiences for students (Berlin, 1987; Brooks & Brooks, 1999; Hillocks, 1995). This literature review of composition studies continues with a review of the literature of rhetorical theory. Rhetorical Theory The writing process involves the writer, reality, reader, and language; these elements are also known as rhetorics and rhetorical theories are the principles that 62

PAGE 76

attempt to explain how these elements interact (Berlin, 1987). See Table 2-4. The difference between rhetorical theories lies in which of these elements are emphasized in the classroom based on how the teacher and students conceive them (Berlin). Tne rhetorical theories that shape composition studies are loosely categorized as objective or subjective. Objective rhetorics maintain that rhetorical elements are scientific, observable, and measurable. Subjective rhetorics are grounded in the epistemic and expressionist movements that build on the concept that reality in writing is personal, private, and intellectual. Objective rhetorics are "positivistic theories that locate reality in the material world" (Berlin, 1987, p. 139). Based on Skinner's (1965) stimulus-response theory, reinforced behaviors in the writing process will be repeated by writers, making writing an observed behavior. For example, grammar workbooks are commonly used in behavioral rhetoric because the teacher can reward the student for writing "correct" answers in the workbook and discipline the writer for being "wrong." This behavior is also evident in weekly speiJing, punctuation, and grammar quizzes and the noxious red pen of many writing teachers. Behaviorists focus on product rather than process because it is relatively easy to measure the correctness of the product and difficult to assess the writing process. A weakness in behaviorist rhetorics is that because the writing process inherently involves complex thinking activities, the behaviorist tendency to observe, mark, and fix errors erroneously leads student 63

PAGE 77

Table 2-4 Continuum of Instructional Strategies in Rhetorical Theories Epistemic CurrentTraditional Focus IIIII Instructional Strategies I. Teaching El. Emphasizes process of CTl. Emphasizes written emphasis writing. (Berkenkotter, product. 1988) 2. Exemplary E2. Stresses connections CT2. Stresses connections writing between good writing and between good writing and good thinking. (Rohman, observable behaviors. 1965) 3. Theoretical ..;. Endorses reciprocal CT3. Endorses reciprocal sources relationship between relationship between subjective thought and objective analysis and language artistry. (Berlin, grammar. 1987) 4. Function of E4. Uses writing as heuristic CT4. Uses writing as tool to \Vliting to discover ideas. report information (Flower. 1989) 5 Viewpoint E5. Encourages reading CT5. Encourages exposure to concerns different perspectives on limited points of view in the writing topic. writing topic. f--(Sommers, 1982) 6 Philosophic E6. Embraces truth and 1 CT6. Embraces truth and basis knowledge as dynamic know ledge as concrete and and dialectical. (Berlin, static. 1987) 7. Audience E7. Considers audience as a I CT7. Considers audience as a concerns primary role in writing, minor role in writing influencing content and second to punctuation and organization. (Wallace, syntax. 1994) writers to believe that mechanical elements in writing are more important than intellectual development (Berlin). Epitomizing objective rhetoric is the current-traditionalist theory, which takes an objective, scientific stance toward rhetoric (Berlin, I 987). This theory 64

PAGE 78

dominates teaching writing today, evident in the profuse use of workbooks and textbooks promoting its positivist tenets (Berlin, 1988). It emphasizes product over process, emphasizing words, sentences, paragraphs, usage, syntax, spelling, and punctuation. Current-traditionalists have a strong preoccupation with the informal essay and research paper (Gere, 2000). The world is rational, according to current traditionalists, and its system of knowledge is discovered through the scientific, experimental method. The weakness in the current-traditionalist rhetoric is that the intellectuality of writing poses inherent problems that challenge empirical verification; in short, intellectual problems require intellectual solutions. Observable behaviors do not always provide answers to intellectual dilemmas. Current-traditional rhetoric posits that truth is empirically based and can only be achieved through subverting a part of the human response to experience (Berlin, 1988). For these reasons, several pedagogies have rejected the current-traditional rhetoric emphasis on product and correctness as antidemocratic because it excludes the dominant discourse of creative expressionism (Schick, 2002). For example, 'hTiting teachers who edit student papers on early drafts and emphasize scores on weekly grammar tests may be depriving their students of the opportunity to think critically about supporting their claims with rational support and multiple perspectives. 65

PAGE 79

To deprive learners of critical thought processes is similar to controlling their thinking and conditioning them to obey the word of the teacher who is "always correct." Shortcomings of current-traditionalist rhetoric are further elucidated: Thinking of writing as social, collaborative, and constructive tells us a good deal about how college and university teachers (and textbooks) should be teaching writing and expecting students to learn it. One implication is that, as much as they might like to, college and university teachers (and textbook writers) cannot tell students how to write. Instead, because writing is itself a displaced form of conversation, teachers have to find ways for students to learn to engage in constructive conversation with one another about writing. (Bruffee, 1993, p. 57) Engaging students in constructive conversation about writing involves thinking critically about their ideas and constructively criticizing others' ideas, attributes of subjective rhetorics. Meyers (1986) explains another shortcoming in the currenttraditional rhetoric approach to teaching writing: As teacheE drop the defenses of detachment and authority, students become more likely to challenge their pronouncements. This can be threatening to teachers who see 'rightness' as the primary sign of competence. But when teachers come to regard encouraging reflection and pressing for informed judgments as important goals, they will welcome questions and challenges as signs that students are beginning to think on their own. (Myers, 1986, p. 94) Subjective rhetorics were a reaction to the current-traditionalists and stem from the expressionist movement prominent in the 1960s and 1970s, which emphasizes the learner's experience, interest, and motivation. Expressivists see truth as a private vision, contacted through writing, discovered by the individual, and 66

PAGE 80

demanding original use of language (Berlin, 1988) This rhetoric also posits that no one can or should evaluate writing because of its emphasis on unique expression and self-discovery throughout the writing process and especially in the prewriting phases (Bruner Goodnow & Austin, 1956; Gere 2000); this approach is clearly the antithesis of current-traditionalists who believe all rhetorical elements are measurable and either correct or incorrect. Expressivists, on the other hand see "reality as arising out ofthe interaction ofthe private vision of the individual and the language used to express this vision'' (Berlin, 1987, p. 146). Writing teachers with an expressionist approach encourage students to freewrite to discover what they are thinking and write reflections that will not be evaluated by anyone Elbow (2000) reflects on the power of his freewriting: Unfocused exploring is probably my main use of freewriting: I have a thought, perhaps out of the blue or perhaps in the midst of writing something (even while writing something else), and I give myseif permission to pursue it on paper in an uncontrolled way wherever it wants to go, even if it digresses (as it usually does). This kind offreewriting is precious to me because my mind seems to work best-at the level of ideas as well as syntax-when I allow it to be uncontrolled and disorganized I cannot find as many ideas or perceptions if 1 try to stay on one track or be organized. And the not-stopping seems to build mental momentum, helps me get wound up or get rolling so that more ideas come. (p. 118) Possible weaknesses in expressionist rhetoric lie with the anarchist expressionists" who want complete and uninhibited freedom in writing, flouting all written conventions (Gere 2000) This approach to teaching writing negates the importance of adhering to written conventions and grammar that make text 67

PAGE 81

intelligible, which is neither acceptable nor feasible when writing to an authentic audience. An underlying assumption of the subjective rhetorics is "enabling individuals to arrive at self-understanding and self-expression will inevitably lead to a better social order" (Berlin, 1987, p. 155), directly refuting the anarchist expressionists' view of breaching written conventions. John Dewey's (1938) questions illustrate concerns of expressivism in education well: Can we find any reason that does not ultimately come down to the belief that democratic social arrangements promote a better quality of human experience, one which is more wisely accessible and enjoyed. than do non-democratic and anti-democratic forms of social life? Does not the principle of regard for individual freedom and for decency and kindliness of human relations come back in the end to the conviction that these things are tributary to a higher quality of experience on the part of a greater number than are methods of repression and coercion or force? Is it not the reason for or preference that we believe that mutual consultation and convictions reached through persuasion, make possible a better quality of experience than can otherwise be provided on any wide scale? (p. 34) Dewey's words emphasize the prominence of social experiences and human relations in rhetoric, attributes evident in epistemic rhetoric. Epistemic, or new, rhetoric is another subjective rhetoric. The difference between epistemic rhetoric and expressivist rhetoric is that epistemic rhetoric sees language as a social, not private phenomenon; language as meaning-maker holds a central role (Berlin, 1987): "Epistemic rhetoric holds that language is the key to understanding the dialectical process involved in the rhetorical act" (p. 166). The 68

PAGE 82

new rhetoric emphasizes that instruction corresponds with reality (Gere, 2000) and because it posits the role of language to create reality and not relay it writing is the act of making meaning in this rhetoric. "Therefore, epistemic writing is writing that aids in thinking, learning and self-knowledge" (Brown, 1998 p 136). In addition, new rhetoric considers mechanical conventions in writing to be important because it contributes to the overall meaning-making process (Berlin, 1988). Writing teachers who ascribe to epistemic rhetoric typically ask students to discuss their writing in workshops and conduct individual conferences with students. Truth and knowledge are dynamic and dialectical, not static, in epistemic rhetoric (Berlin, 1988; Gere, 2000). This perspective emphasizes the tenet in new rhetoric, that language, written language, creates meaning based on the writer s cognition. This rhetoric is also referred to as cognitive rhetoric which includes writers' thinking and comprehension strategies. Flower (1994a) explains that cognitive rhetoric emphasizes metacognitive awareness and the conscious control of one's options" (p. 121). She explains that cognitive rhetoric emphasizes the learner s own processes of observation and reflection and notes that even though the endorsement of cognitive rhetoric is a matter of degree, it can change what teachers do and students learn. Summary This literature review corroborates a logical positive synthesis of critical thinking, problem-based learning, and teaching writing (see Table 2-5). Critical thinking 69

PAGE 83

Table 2-5 Synthesis of Critical Thinking, Problem-Based Learning and Teaching Writing in Student Attributes Cognitive Writers Behavioral Writers Focus .. Student Attributes I. Content Cl. Identifies content to be 81. Learns information acquisition learned. determined by others. ., Personalized C2 Individualizes leammg I 82 Generalizes own learning. learning based on personal interests. I 3. Knowledge C3 Constructs knowledge I 83. Does not knowingly construction actively when learning. construct knowledge when learning 4. Problem C4. Analyzes problems 84. A voids problems analysis 5. Problem C5 Defines what information 85. Accepts readily available solving is pertinent to solve a I intormation to solve a problem. problem. 6. Rea s oning C6. Possesses logical 86. Thinks compulsively and ggskills reasoning and problemrashly. solving skills 7. Information C7. Reuses information 87. Discards information after use reinforced through use. use 8 Participation C8 Participates actively and 88. Participates on a passive activity is motivated when low level of involvement learning. I when learning. 9. Group problem C9. Encourages group 89. Retracts from group solving problem solving. problem solving. 10. Solution CIO. imple ; nents solutions to 810. Does not complete implementproblems solved. problem solution at ion implementation. 11. Prior CII. Activates own prior 811. Does not acknowledge knowledge knowledge. prior knowledge. 12. Learning Cl2. Learns independently. 812. Depends on others to independence I identify content to be learned. 13. Critical Cl3. Thinks critically 813. Possesses unreasonable, thinking irrational reasoning skills. 70

PAGE 84

Table 2-5 (continued) Cognitive Writers Behavioral Writers Focus Student Attributes 14. Learning Cl4. Transfers learning B14. Does not transfer transfer responsibility from the responsibility for learning teacher to self. from teacher. 15. Emphasis of C15. Emphasizes process of I B15. Emphasizes product of teaching discovery. correctness, not process. 16. Model writing Cl6. Connects between good Bl6. Takes positivist approach writing a:1d good to observing and thinking. I measuring good writing. I 17. Viewpoint C17. Reads authors who have I B17. Reads limited number of concerns different perspectives on I authors and perspectives. writing topic. 18. Function of C18. Believes writing aids Bl8. Believes writing fits writing cognition because writers proper formats of research use language to discover papers and discourse ideas. classification modes. 19. Philosophic C19. Believes truth and I Bl9. Believes truth and basis knowledge is dynamic know ledge are concrete and dialectical, not static. and definite. directly entails problem solving and analysis. necessary skills to be taught in college and university (Meyers, 1986). Educated people must be autonomous enough to make important decisions, respect the rights of others, seek understanding of factual information, and remain open-minded to others' perspectives; these are the basic characteristics of critical thinkers (Norris & Ennis, 1989). These are also the same characteristics of an epistemic rhetorical approach to teaching writing in a problembased learning environment. McPeck ( 1990) emphasizes the necessity of selfdirected learning: 71

PAGE 85

True critical or autonomous thinking is, by definition, doing one's own thinking; therefore, students must be cut loose from their dependency on authority. Yet teachers in a classroom are in a de facto position of authority, and this fact has a deleterious effect upon the free and open exchange of ideas. (p. 52) Meyers (1986) also emphasizes the natural fit between teaching writing and teaching critical thinking by pointing out that written assignments reveal what and how students are thinking. This insight allows writing teachers to nurture new modes of critical thinking for student writers Enriching what writing teachers do and students learn is the impetus of this study. A problem-centered instructional orientation to teaching writing makes sense because it is aligned with epistemic rhetoric that promotes cognition, personal voice, reflection, goal-setting, and discovery. These attributes empower the writer as a creative. critical thinker. Attributes of problem-based learning, such as the development of problem-solving and self-directed learning skills, teacher role as tutor and resource instead of sole source of knowledge, and socially collaborative activities also reflect epistemic and expressive rhetorics. Current-traditional rhetoric is more aligned with lecture formats of knowledge telling. It conflicts with evidence that writing is a complex, intellectual activity and seems to disregard students' needs to develop critical thinking skills. Recognizing and promoting the correlations among critical thinking, problem solving, and writing, writing teachers can contribute to the cognitive development of their students into responsible adults. "If we teach our students how 72

PAGE 86

to use what they know and how to learn what they will need to know, it seems to me that both the individuals and society at large will be better served" (Stice, 1987, p. 98). This is a responsibility for writing teachers of great magnitude, but one that should be savored every day: If we teach our student to care, or at least show them that we care, our teaching of critical thinking may foster a future technology more concerned with healing than with destruction. Then students may learn that the wonders of life on our frail planet are not only puzzles that engage our problem-solving abilities but also mysteries worthy of preserving simply for the beauty and joy they offer. (Meyers, 1986, p. 119) Indeed, society will be better served when students are equipped to solve problems, direct their own learning, think critically, and write effectively about the world around them. 73

PAGE 87

CHAPTER THREE DEVELOPMENT AND IMPLEMENTATION OF THE INSTRUCTION INVENTORY The Instruction Inventory was created and validated to understand the role of problem-centered instruction in composition. Before the instrument was created, however, the Cognitive-Composition Model was developed. This chapter explains the methodology underlying the development ofthe model and the instrument and how they provided responses for two research questions. The first question asks the following: 1. \\'nat evidence demonstrates the construct validity of the Instruction Inventory, an instrument that measures teachers' and students' perceptions of instruction and learning? The second stage of this study involved implementation of the model and instrument. This chapter also explains methods behind the data collection and analyses involved in the implementation of the instrument. These methods provided responses for the second research question in the study: 2. What are the relationships between teachers' and students' perceptions of the following in college writing class? 74

PAGE 88

a. Problem-centered instruction b. Self-directed learning use c. Critical thinking use Several main sections comprise this chapter: the framework of the Cognitive Composition Model, the development of the Instruction Inventory, the construct validity process, and the implementation of the Instruction Inventory and data analyses. Framework for the Cognitive-Composition Model The Cognitive-Composition Model juxtaposes elements from problem-based learning, critical thinking, self-directed learning, and rhetorical theories (see Table 3-1). The model illustrates writers' cognitive orientations by contrasting them with writers' behavioral orientations; these orientations are defined according to problem-based learning, self-directed learning, critical thinking, and rhetorical theory. The integration of problem-based learning with rhetorical theories is somewhat novel. By combining knowledge of problem-based learning tenets with composition theories and strategies, teachers may be able to identifY ways to promote their students' problem-solving, self-directed learning, and critical thinking skills. 75

PAGE 89

Table 3-1 Cognitive-Composition Model Cognitive Writers Behavioral W Titers Focus Student Attributes 1. Content Cl. Identifies content to be B 1. Learns information acquisition learned. determined by others. 2. Personalized C2. Individualizes learning 82. Generalizes own learning. learning based on personal interests. 3. Knowledge C3. Constructs knowledge 83. Does not knowingly construction actively when learning. I construct knowledge when learning. 4. Problem C4. Analyzes problems. 84. A voids problems. analysis 5. Probiem C5. Defines what information 85. Accepts readily available solving is pertinent to solve a information to solve a 2roblem. problem. 6. Reasoning C6. Possesses logical I 86. Thinks compulsively and skills reasoning and problemrashly. solving skills. I 7. Information C7. Reuses information I 87. Discards information after use Reinforced through use. use. 8. Participation C8. Participates actively and 88. Participates on a passive, activity is motivated when low level of involvement learning. when learninz. 9. Group problem C9. Encourages group B9. Retracts from group solving problem solving. problem solving. 10. Solution C 10. Implemerl.s solutions to B I 0. Does not complete implementproblems solved. problem solution ation implementation. 11. Prior C11. Activates own prior B 11 Does not acknowledge knowledge knowledge. prior knowledge. 12. Learning C12. Learns independently. B 12. Depends on others to independence I identifY content to be learned. 13. Critical C 13. Thinks critically. B 13. Possesses unreasonable, thinking irrational reasoning skills. 76

PAGE 90

Table 3-1 (continued) Cognitive Writers Behavioral Writers Focus .. Student Attributes 14. Learning CI4. Transfers learning Bl4. Does not transfer transfer responsibility from the responsibility for teacher to self. learning from teacher. 15. Emphasis of CIS. Emphasizes process of 1 B15. Emphasizes product of teaching discoverv. i correctness. not process. 16. Model writing Cl6. Connects between good Bl6. Takes positivist approach writing and good I to observing and thinking. i measuring good writing. 17. Viewpoint Cl7. Reads authors who take Bl7. Reads limited number of concerns different perspectives on I authors and perspectives. writing topic. 18. Function of CI8. Believes writing aids 818. Believes writing fits writing cognition be:;ause writers Proper formats of use language to discover research papers and ideas. I discourse classification I modes. 19. Philosophic Cl9. Believes truth and B19. Believes truth and basis knowledge is dynamic knowledge are concrete and dialectical. not static. and definite. Cognition and Behaviorism The Cognitive-Composition Model is a continuum, not a static diagram, because it ascertains the extent of problem-based learningness in composition instruction. Most notable about the Cognitive-Composition Model is the contrast between the extremes of the continuum, cognitive and behavioral writers. This contrast reflects contemporary learning theories (Bigge & Shermis, 1999): "Their great difference centers upon the behavioristic assumption that human beings are passive or reactive and the cognitive-interactionist assumption that they are interactive in relationship with their environments" (p. 44 ). 77

PAGE 91

Cognitive writers tend to address problem scenarios that they could actually encounter in a real-world environment. They assume responsibility for solving problems inevitable in the writing process with active, critical thinking. Cognitive writers typically use problem-solving and self-direction skills, they actively construct knowledge, define resources needed to solve a problem, assume responsibility for learning, and experiment with learning. Thus, and most importantly, cognitive writers can write effectively for any given audience or purpose with engaged cognition and critical thinking because they are able to understand and write appropriately in different rhetorical situations. In sharp contrast to cognitive \\Titers are behavioral writers, students who typically learn to write in an environment in which the teacher crafts the writing assignment and controls students' learning Teachers focus instruction on sentence level mechanics and unknowingly do not encourage problem-solving, self-direction, or critical thinking. Characteristics of behavioral writers include avoidance of problem-solving and critical thinking, concentration on language and composition mechanics, emphasis on product over process, and dependency on the teacher to define all learning objectives and control learning. Behavioral writers tend to depend solely on their teachers' instructions and accuracy because they are more aware of the written format and accurate mechanics than their audience and purpose for writing. These students tend to engage superficially with their topics also, which usually reflects shallow thinking throughout the writing process. 78

PAGE 92

Problem-centered instruction can enhance students' learning as it has in many disciplines In science courses, students were challenged to make and justify estimations and assumptions (Allen, Duch, & Groh, 1996). Business students had the opportunity to solve everyday business problems (Holt & Willard-Holt, 2000). In health care fields, students had the opportunity to solve complex multidisciplinary problems (Bruhn, 1997) Engineering students exposed to a problem-based learning program scored one standard deviation higher than those who were not enrolled in the program (Woods, 1996). Calculus students became more independent and involved in mathematics when they were exposed to a top down approach of identifying subproblems before major problems (Seltzer, Hilber, Maceli, Robinson, & Schwartz, 1996). Business students learned content as needed by managing learning problems (Stinson & Milter 1996). Similar positive experiences have been noted in social work (Heycox, & Bolzan 1991) and law (Winsor, 1991 ) These studies illustrated the functionality of problem-centered instruction in various disciplines. They did not however, utilize a valid instrument to measure teachers' and students' perceptions of teaching and learning hence the need for the construct validity of the Instruction Inventory. Development of the Instruction Inventory Although one purpose of this study was to develop a discipline-independent instrument that informs teachers about how problem-centered instruction self directed learning, and critical thinking are perceived by teachers and students, it was 79

PAGE 93

not the first construction of this study. Likewise, neither was the model the first construction. The starting point was the development of four continua to concretize theory and research. Figure 3-1 illustrates the cycle between the theories and the development of the Cognitive-Composition Model and Instruction Inventory. Development of the Cominua The four continua concern problem-based learning, self-directed learning, critical thinking, and rhetorical theory. The continuum of problem-based learning theory and methods identifies attributes of problem-centered instruction and contrasts them with its antithesis. teacher-based ]earning, as depicted in Table 3-2. On the left side ofthe continuum is problem-based learning, which is student centered learning because the problems essentially promote self-directed learning and problem-solving skills with the guidance of a tutor and transfer the responsibility of learning from teacher to student (Barrows, 1988; Seltzer, Hilber, Maceli, Robinson & Schwartz, 1996). Problem-based learning helps students to develop problem-solving and reasoning processes, self-directed learning skills, and collaborative working skills (Allen & Rooney 1998; Barrows & Tamblyn, 1980; Rangachari 1996). Also, problem-based learning stimulates internal motivation for learning because it requires student involvement throughout the learning process. Such a problem-centered approach to instruction also tends to develop self evaluation skills and activation of prior knowledge (Barrows, 1988; Bridges & Hallinger, 1997; Bruhn, 1997). 80

PAGE 94

Instruction Inventory Problem-based Learning, Composition Studies, and Critical Thinking Theories Cognitive-Composition Model Figure 3-1 Cycle between the theories, model, and survey instrument used in this study. Problem-centered instruction is integrated differently in different classrooms by different teachers. The continuum illustrated in Table 3-2, however, depicts the learning approach according to Barrows (1988) Even so, Barrows (1986) explains variations in applications: The tenn problem-based learning must be c o nsidered a genus for which there are many species and subspe c ies Each addresses different objectives to varying degrees. All descriptions and evaluations of any PBL method must be analysed in terms of the type of problem used the teaching-learning sequences, the responsibility given to students for learning and the student assessment methods used Any teacher who wishes to employ PBL should decide on desired educational objectives and then select the method that fits best. (p. 485) On the right side of the problem-based learning continuum is teacher-based learning This type of learning is teacher-centered because it encourages students to depend on their teachers to tell them what to learn, how to learn, and when to learn. Freire (1990) referred to this method as "banking" because the teacher, known as 81

PAGE 95

Table 3-2 Continuum of Student Characteristics of Problem-Centered Instruction Problem-Based Learning Teacher-Based Learning Focus ... Student Attributes 1. Content Pl. Identifies content to be Tl. Learns information learned. (Barrows, 1996) determined by others. 2. Personalized P2. Personalizes own learning T2. Generalizes own learning based on pers onal interests. learning. {Barrows & Tamblyn, 1980) 3 Knowledge P3. Constructs ki1owledge T3. Does not knowingly construction actively when learning. construct knowledge (Barrows & Tamblyn, 1980) when learning. 4. Problem P4. Analyzes problems. (Birch, I T4. A voids problems. analvsis 1986) 5. Reasoning PS. Practices logical reasoning T5. Thinks compulsively skills and problem-solving skills and (Edens, 2000) I 6 rashly. 6. Participation P6. Participates actively and is IT. Panicipates on pa'>sive, activity motivated when learning. low level of (Woods, 1994) I involvement I when learning. 7. Group P7 Encourages group problem IT7 Withdraws from group problem solving. (Allen Ouch, & problem solving. solving Groh, 1996) 8. Solution P8. Implements solutions to I T8. Does not complete implementproblems solved. (Bridges & I solution ation Hallinger. 1996) implementation. 9. Prior P9. Activates own prior I T9. Does not acknowledge knowledge knowledge. (Bridges & prior knowledge. Hallinger, 1996i 10. Independence PlO. Learns independently. TIO. Depends on others (Gijselaers, 1996) throughout learning process. 11. Critical PI I. Thinks critically. (Magnussen Tl!. Possesses illogical, thinking Ishida, & Itano 2000) irrational reasoning I skills. 12. Learning Pl2. Transfers learning I n2. Does not transfer transfer responsibility from teacher to responsibility for self. (Barrows & Tamblyn, I learning from teacher. 1980) 82

PAGE 96

the "oppressor," deposits knowledge into students' heads. Banking minimizes students' creativity. It serves interests of instructors who teach prescribed, contrived lessons with no intentions to share real-world situations with students. These teachers "react instinctively against any experiment in education which stimulates the critical faculties" (Freire, p. 60), further emphasizing students' dependence on their teachers' control of the learning process. The continuum of self-directed learning theory and methods, Table 3-3, contrasts self-directed learners with other-directed learners. Self-directed learners identifY their own learning goals, activities, and assessments (Candy, 1991 ). They are comfortable learning by themselves and in groups (Knowles, 1975). Self directed learners have a strong sense of control and responsibility regarding their own learning (Brockett & Hiemstra, 1991 ) In contrast, other-directed learners look to their teachers to direct their learning. They typically do not assume much control or responsibility in their learning, nor do they make time for learning or accept criticism well. The major differences between self-directed and other-directed learners concern the impetus of the learning and the desire to learn. In Table 3-4, the continuum of critical thinking theory and methods contrasts critical thinkers with surface thinkers. Critical thinkers are people who analyze problems, evaluate claims, suspend conclusions, and judge logical inferences (Barnet & Bedau, Norris. & Ennis, 1989; Ruggiero, 1988). Surface thinkers are 83

PAGE 97

Table 3-3 Continuum of Learners' Self-Directed Learning Skills Self-Directed Learning Other-Directed Learning Focus 4 .. Student Attributes 1. Self-reliance Sl. Self-reliant in finding 01. Needs help finding information. (Hiemstra. information. 1998) 2. Personal goals S2. Sets personal learning 02. Does not set personal goals. (Garger, 1999) learning goals. 3 Learning in S3. Enjoys learning in 03. Does not enjoy learning isolation isolation. (Hiemstra. in isolation. 1998) 4. Time for learning S4. Finds time for learning. 04. Does not find time for (Licklider, 1997) learning 5. Feedback and S5. Accepts feedback and 05. Does not accept feedback criticism criticism. (Anfield. and criticism. 1997) 6. Learning control S6. Takes control of 06. Does not take control of personal learning personal learning. ( Brookfield. 1993) 7. Seeking learning S7. Seeks out learning 07. Does not seek out opportunities. (Gerber, learning opportunities. Lankshear, Svensson, 1995) 8. Problemsolving S8. Enjoys problem 08. Does not enjoy problem solving. (Hiemstra, solving. 1997 ) 9. Challenging S9 Challenges 09. Does not challenge assumptions assumptions. (Pillingassumptions. Connick, 1997) 10. Responsibility SIO. Assumes responsibility 010. Does not assume for learning for own learning. responsibility for own (Knowles 1975) learning 11. Self-evaluation SIJ. Evaluates and I 011. Does not evaluate or advocates advocate own learning. own learning. (Pilling-Connick, 1997) 12. Self-reflection 512. Reflects upon and 012. Does not reflect upon or makes make choices about own choices about own learning. learning. (Candy, 1991 ) 84

PAGE 98

Table 3-4 Continuum of Learners' Critical Thinking Skills Critical Thinking Surface Thinking .. Student Attributes 1. Problem solving Cl. Analyzes problems. GI. Stays away from (Norris, 1992) problems when possible. 2. Evaluation skills C2. Evaluates claims, G 2 Accepts claims asst:mptions and assumptions, and evidence. c : .,mith, evidence on value. 199Ql 3. Learning Attitude C3. Shows imaginative G3 Shows an objective, open mindedness and indifferent approach to intellectual curiosity. learning (Norris & Ennis, 1989) ---4 Idea generation C4 Considers new ideas. G4 Depends on tonner ideas. (Barnet & Bedau, 2002) 5. Summarizin:;: CS. Summarizes I G5. Summarizes arguments skills arguments accurately I incorrectly. (Barnet & Bedau I 2002) 6. Self-reflection co. Uses judicious, or I G6. Accepts information reflective, skepticism. without much reflection I (McPeck 1981) l or contemplation. 7. Construction of C7. Constructs alternatives G7. Applies first solutions solutions and solutions in developed in problemproblem-solving. solving. ( Me_vers, I Q86) 8 Argument C8. Can recognize and G8. Does not deliberately construction construct sound recognize or analyze arguments. (Rug,.giero, sound arguments 1988) 9. Application of C9 Applies principles of G9. Vague application of logic formal and informal formal and informal logic logic (Barnet & principles Bedau, 2002) 10. Understanding CIO. A voids fallacies in GIO. Uses logical fallacies in fallacies reasoning. ( Ruggiero, reasoning. 1988) 11. Making Cll. Suspends conclusions. Gil. Makes quick conclusions conclusions (McPeck, 1981) and judgments. 85

PAGE 99

Table 3-4 (continued) Critical Thinking Surface Think in g IIIII .. Student Attributes I 2 Accepting truths CI2. Sets aside own ideas Gl2. Depends on others' ideas of the truth and of the truth without reflects on reflection. alternatives (Meyers, 1986) 13. Motivation C13. Internally motivated G 13. Externally motivated about learning about learning. (Halpern, 1989) 14. Logical Cl4. Can judge and make G14. Hastily judges and draws Inferences logical inferences. logical inferences. (Norris, 1992) mundane thinkers (Norris & Ennis). The contrast between critical and surface thinking is thm critical thinking is reasonable and reflective; that is, a critical thinker makes decision!> conscientiously and deliberately, whereas a surface thinker is spontaneous (Norris & Ennis). Critical thinking involves purpose, reason, goals, problem-solving inferences, evaluations, reflection, and analysis (Halpern, Norris). Surface thinking is neither reasonable nor reflective making it everyday functional thinking without deliberation or goal orientation. The continuum of instructional strategies in rhetorical theories in Table 3-5 contrasts two rhetorical theories, epistemic and current-traditional (Gere, 2000, pp. 1 0-25). Epistemic rhetoric, on the left side of the continuum is oriented toward problem-centered instruction because it stresses problem-solving, the process of writing discovery learning, multiple perspectives reciprocal relationships between 86

PAGE 100

Table 3-5. Continuum of Instructional Strategies in Rhetorical Theories Epistemic CurrentTraditional Focus .. Instructional Strategies I. Teaching El. Emphasizes process CTI. Emphasizes v.Titten emphasis of writing. product. (Berkenkotter, 1988) 2. Exemplary E2. Stresses connections CT2. Stresses connections v.Titing between good writing between good writing and and good thinking. observable behaviors. (Rohman. 965) 3. Theoretical E3. Endorses reciprocal CT3. Endorses reciprocal sources relationship bet\veen relationship between subjective thought objective analysis and and language artistry. grammar. (Berlin. 1987'1 4. Function of E4. Uses writing as CT4. Uses writing as tool to writing heuristic to discover report information ideas. (Flower, 1989) 5. Viewpoint ES. Encourages reading CT5. Encourages exposure to concerns different perspectives I limited points of view in on the writing topic. I writing topic. Sommers, 1982) 6 Philosophic basi s F6. Embraces truth and CT6. Embraces truth and knowledge as knowledge as concrete dynamic and and static. dialectical. (Berlin 1987) 7. Audience E7. Considers audience len Considers audience as a concerns as a primary role in minor role in writing, writing, influencing second to punctuation and content and syntax. organization. (Wall ace 1994) thought and language, and audience. The right side of the continuum, currenttraditional rhetorical theory, is oriented toward teacher-based learning because it emphasizes the teacher's control in the classroom and focuses on the quality of written products and the dissection of words and sentences usage and style, and 87

PAGE 101

mechanics. Interestingly, the contrast between epistemic and current-traditional rhetorics mirrors the contrast between problem-based learning and teacher-based learning theories. Epistemic rhetoric and problem-centered instruction are cognitive and subjective; current-traditional rhetoric and other-directed learning are behavioral and objective. Regardless of their orientations, the validity of constructs in these four continua were then subjected to testing. Construct Validity: Building a Body of Evidence The construct validity process builds a body of evidence for measuring abstract constructs such as problem-centered instruction, self-directed learning, and critical thinking. Krathwohl ( 1998) explains, "Construct validity is the unifying framework for all evidence regarding validity. Basically the evidence sought is that the measure behaves as would be expected if it were a valid measure" (pp. 426-27). The construct validation is both theoretical and empirical because predictions regarding the construct stem from theory and confirmation of these predictions are grounded in evidence. "A construct," states Krathwohl, "is a characteristic which is presumed to exist but which cannot be directly measured" (p. 426). He continues, "Its presence and strength or amount is inferred from the sample of behavior gathered by a measure designed to assess it" (p. 426). The nature of a construct, then, is tenuous, only to be actualized in a valid means to assess it. Construct validation is a critical concern for scientists, mainly because they cannot do without constructs (Nunnally, 1978). Instead of being isolated, observable 88

PAGE 102

dimensions, constructs represent "hypotheses that a variety of behaviors will correlate with one another in studies of individual ditTerences" (Nunnally p. 96). The constructs under consideration in this study should correlate positively with each other. Nunnally points out that all theories concern constructs and not particular, observable variables, emphasizing the importance of validating them reliably. Constructs also vary in how the observable variables describing them can vary in number and definition (Nunnally 1978). Larger domains of variables used to describe the construct can result in difficulties defining which variables belong in the domain. Smaller domains of variables often show that any variable in that domain can adequately measure the construct. In this study, each of the three constructs under scrutiny have ten to fifteen dimensions operationalizing them creating a challenging situation when defining exactly which ones defme the construct. "For many constructs, the domain of related observables has 'fuzzy edges,' and scientists are not sure of the full meaning of their own constructs" (Nunnally, p. 97). The domain of observables is inexact. Therefore it makes sense to combine results from a number of measures of observables to validate the construct, as was the case in this study Stanley and Hopkins (1990) concur with Nunnally (1978) and explain that construct validation is a ''systematic analysis of test scores in terms of psychological constructs" (p 104 ). They emphasize that many criteria are required to build a body 89

PAGE 103

of evidence and determine what the test actually measures and for the instrument to "possess a degree of construct validity" (p. 1 05). The systematic approach to validate constructs is lucid, yet flexible. The process to measure constructs is comprised of three major aspects: ( 1) specifying the domain of observables related to the construct; (2) from empirical research and statistical analyses, determining the extent to which the observables tend to measure the same thing, several different things, or many different things; and (3) subsequently performing studies of individual differences and/or controlled experiments to determine the extent to which supposed measures of the construct produce results which are predictable from highly accepted theoretical hypotheses concerning the construct. (Nunnally, 1978, p. 98) Interestingly, Nunnally explains that the three major aspects of validating constructs rarely occur in order. It is more likely that the researcher will identify a construct and then jwnp to studying relationships, or correlations, between the measure of that construct and other measures of other constructs. When many scientists jump to analyzing correlations of similar constructs, the result is that a single construct may have a number of proposed measures. But these measures may not be measuring the same thing. Then, what usually happens is that a scientist will list the observables related to the construct. Nunnally also explains that scientists usually wo:rk. as individuals resulting in rarely planned measures of constructs and neatly itemized procedures for validating constructs. The standard, realistic procedure for developing a body of evidence in support of the construct validity of a measure begins with establishing the logical 90

PAGE 104

relationship of the items in the Instruction Inventory to their underlying constructs by a panel of experts: "Develop a set of tasks or items based on theory-a rational analysis of the construct (Stanley & Hopkins, 1990, p. 1 05). After revisions have been made according to experts' recommendations, the survey is field tested with a sample of participants: "Deduce testable predictions regarding the relationship between the construct and other empirical measures" (pp. 105-1 06). A factor analysis is then conducted to determine clusters of items that measure the same constructs: "Obtain the data required to investigate these predicted relationships empirically" (p. I 06). Next, the instrument is revised according to the factor analysis results and distributed to another sample set of participants, different from the first set that field tested the survey. After data collection, results and correlations are analyzed: Eliminate items or tasks that operate contrary to theory (or revise the theory) and processed again with steps 2 and 3" (p. 106). Implications for applications of the instrument and data analysis are explored throughout this validation process. This study follows a slightly amended version of the ideal construct validity procedure described above; the purpose of these amendments are due to logistical constraints of time and participant resources, however, and do not diminish the study's integrity. This particular study did not carry out all ofthe recommended steps in construct validity, such as contrasting divergent and convergent evidence, because the Instruction Inventory is a new instrument, and building the case for 91

PAGE 105

construct validity is an ongoing, recursive process. It is not possible to establish construct validity in a single study fully (Stanley & Hopkins, 1990). Step One: IdentifY Theoretical Framework to Develop Items The first step of construct validity is to identify the theoretical framework underlying each of the main dimensions of the instrument (see Literature Review in Chapter Two). Direct links between specific theoretical postulates and operationalized constructs formed the Instruction Inventory and the Cognitive Composition. This initial step of construct validity includes hypothesizing about the relationships between constructs based directly on explicit theories in this first step also (Crocker & Algina, 1986). Forming hypotheses deepens the identification of the theoretical framework. In the case of this study, a positive relationship was expected to exist among all three constructs; that is, stronger problem-centered orientations of instruction should promote stronger self-directed learning and critical thinking use in students. A measurement instrument consisting of "items that are specific, concrete manifestations of the construct" (Crocker & Algina, 1986, p. 231) can then be developed. As is the case of survey construction, inferences were made throughout the process of researching and creating the Instruction Inventory. The Instruction Inventory proposed approximately seventy operationalized statements for the three constructs of problem-based learning, self-directed learning, and critical thinking. 92

PAGE 106

After the Cognitive-Composition Model was created, items for the instrwnent to measure problem-based learning (Appendix A), self-directed learning (Appendix B), and critical thinking were drafted (Appendix C). The process involved coding each of the polarized ends of the continua. Four positive items were written for the left side of each subscale, creating positively-scored items for the Inventory. After several drafts. some items were deleted and most were reworded. Because the right side of each subscale represents a stark contrast of the left side, the three items that were written for them were negatively scored in the Inventory. This process was followed for each of the descriptors in each of the subscales. Table 3-6 explains the origin and coding of items based directly on the continua. The descriptor is the attribute being measured, the placement is the location on the continuum that also shows how the item is scored, and the code identifies the location on the continuum. To determine if the items measuring each dimension logically reflect the theoretical definition of the dimension, a draft of the Instruction Inventory was sent to a panel of experts in problem-based and self-directed learning and critical thinking (see Appendix D). Judges evaluated the logical fit of each item with the underlying construct it was intended to measure with a Likert-scaled survey. Data collected from the judges for their evaluations of the Instruction Inventory statements and internally validating statements helped to revise and increase the validity of the Instruction Inventory. Items were asked differently in 93

PAGE 107

Table 3-6 Origin and Coding of Items PROBLEM-BASED LEARNING CONTINUUM Descriptor: Placement: Code: Identifying content to be learned. Left side-positive score P1 ("P"roblem orientation) a. I ask students to tell me what they want to learn in class. b. I invite discussion from students. about .;ourse content. c. I encourage students to talk to each other about what they want to learn. d. It's okay for students to change their minds about what they want to learn in class. Descriptor: Placement: Code: Learning information determined by others. Right side-negative score Tl ("T"eacher orientation) e. I assume students will depend on my choices for class content. f. I ask students to accept the university's course description. g. Students should trust their instructors' decisions on course content. SELF-DIRECTED LEARNING CONTINUUM Descriptor: Placement: Code: Self-reliant in .finding information. Left side-positive score S I ("S"elf orientation) a. My students don't need much help when it comes to research. b. My students spend time outside class exploring topics that interest them. c. My students are able to find whatever they need in the library. d. My students are able to find whatever they need online. Descriptor: Placement: Code: Needs help finding information Right side-negative score 0 I ("O"ther orientation) e. My students depend on me to help them with their research. f. My students ask me for research suggestions before they begin to research. g. My students need help finding information in libraries they have not yet used. 94

PAGE 108

Table 3-6 (continued) CRITICAL THINKING CONTINUUM Descriptor: Placement: Code: Analyzes problems. Left side-positive score C 1 ("C"ritical thinking) a. My students are eager to analyze any problem I present to them in class. b. My students are familiar with a logical sequence of solving problems. c. My students enjoy solving problems in small groups. d. My students identify their own problems for which they can generate logical solutions. Descriptor: Placement: Code: Stavs away from problems when possible Right side negative score G 1 ("G 'ood thinking) e. My students would rather I offer solutions for their problems instead of them generating their own solutions. f. My students enjoy listening to me lecture more than they like identifying their learning challenges in class. g My students have expressed a clear dislike of discussing learning dilemmas in class. different sections of the survey to strengthen their consistency and internal reliability. Step Two: Pilot Study A pilot study was conducted to amend any administrative or comprehension issues that may exist with the instrument. An email was sent to the entire faculty of seventy instructors (see Appendix E) explaining the study and requesting participation from only four teachers for the pilot study. The same email asked those instructors who did not participate in the pilot study to participate in the tinal 95

PAGE 109

implementation approximately three weeks after the pilot study. The pilot study was explained merely as an inquiry into the perceptions of writing teachers and their students of instructional strategies and lessons learned in the class. To reduce incidence of false responses and increase validity of the results, teachers were told nothing else about the objectives of the study. The email message explained that the teacher and the students were to complete the anonymous survey simultaneously during class in thirty minutes, the extent of their total participation in the study. Teachers implementing the survey had complete directions for them to read verbatim to their students (see Appendix F). Students had the choice to abstain from completing the survey without any ramifications from anyone whatsoever. After completing the survey and placing them in individual envelopes, each of the envelopes were placed in a larger envelope to ensure their anonymity. I then entered the classroom and told the teacher and students that I did not want to know how they scored any item. Instead, I asked the questions, "What difficulties did you encounter completing the survey?" "Were any items confusing or ambiguous?" "Do you have any suggestions to make the survey easier to take?" Copious notes were taken during this discussion and analyzed after discussions with the pilot classrooms, lending qualitative data to triangulate the data analysis. This discussion took five to ten minutes. An advantage to conducting a pilot study is that the remainder of the faculty was available to participate in the field test and resulting factor analysis and means comparisons. A 96

PAGE 110

reliability test was conducted after the pilot study to show that the scales that were intended to yield subscores were reliable. Results of these analyses are reported in Chapter Four. After the pilot study comprising two classrooms was complete the revised instrument was distributed to the instructors who volunteered to participate in its final implementation. All surveys were hard copies and not electronically delivered. Each unit of measurement consisted of one teacher's survey and eighteen students surveys (course enrollment is limited to eighteen students). Although instructors teach more than one section of composition some first-year composition and some upper-division courses they were asked to choose only one section of their choice to participate in the study (see Appendixes G and H for the student and teacher forms of survey, respectively). The rate of response from the students was one hundred percent because the teacher distributed the surveys to the students during class time An option was given to the students, however, to opt out of completing the survey if they so chose when their teacher distributed the surveys, as in the pilot study. Upon completion the teacher placed all of the surveys, including unused ones, in the envelope that originally contained them sealed it and delivered them to me. Logistic restraints dictated the following schedule for data collection in this study : the pilot study of the Instruction Inventory was given during the fifth week of a fifteen-week semester. Distribution of the revised version of the instrument took 97

PAGE 111

place in the eighth week of the semester. The factor analysis and comparison of total scores completed the schedule. Ideally the Instruction Inventory should have been distributed in the final third of a course or any time after the tenth week of a fifteen-week semester. This would have given students ample time to develop their perceptions about the instructional strategies and their learning acquisition in the course This timing probably would have given more accurate assessments than they did in the fifth and eighth weeks of the semester. Future research should test this hypothesis or determine whether students' perceptions are as consistent in the second third as well as the final third of the course. Step Three: Factor Analysis The next step of construct validity V.' as to analyze the internal structure of the measure using factor analysis This step confirmed the identity of subscales or dimensions that made up the larger measure. Krathwohl ( 1998) refers to this step as "'evidence based on internal structure; or factor analysis. The factor analysis '"is a statistical procedure that by examining interrelationships among items or tests help to identify the dimensions underlying a measure and hence what it is measuring" (Krathwohi p. 430). These dimensions, or factors, correspond to what the test intended to measure and establish empirical evidence for the construct validity. In the case of this study the factor analysis measured and established empirical 98

PAGE 112

evidence for measuring problem-centered instruction, self-directed learning, and critical thinking. In essence, the factor analysis is a confirmatory analysis, examining two hypotheses concerning the internal structure of the Instructional Inventory. The first hypothesis is that there is a principal factor underlying all items in the Instruction Inventory: a principal component analysis tested this. The second hypothesis is that three dimensions underlie the inventory: problem-centered instruction, self-directed learning, and critical thinking, with a set of items corresponding to each of these dimensions. This hypothesis was examined by means of oblimin and verimax rotations, constrained to three factors each. Factor analysis generates a matrix of intercorrelations for each item in the Instruction Inventory (Krathwohl, 1988). Because the Instruction Inventory was designed to measure three constructs, the analysis revealed clusters of items that measured the same factors within these constructs. This allowed for reliable inferences about the constructs and shows relationships among items. Within each construct were positive and negative items illustrating each attribute Negatively worded items were reverse-scored so that all items measuring a common attribute are expected to be positively correlated Step Four: Data Analysis After the field test, the instrument was revised again to reduce the number of items in the instrument. The data were then analyzed and results are elucidated in 99

PAGE 113

Chapter Four. Planned data analysis methods consisted of differential comparisons of the students' and teachers' total scores on the survey. These differentials determined discrepancies between them. Positive correlations were expected between students' and teachers' scores on each subscale. That is the more teachers perceived their instruction to be problem-centered, the more the students should perceive it also. Likewise, the more teachers and students perceive the instruction to be problem-centered, the more they should perceive their learning to be self-directed and the higher caliber they should see their thinking as critical. These results are discussed in Chapter Four. Implementation Analyses: Examining Relationships between Scores To provide insight into the second research question inquiring about the relationships between the three constructs in college composition courses means were compared on three subscales not just an item-by-item analysis. Means of students scores for each item of the Instruction Inventory were compared to teachers' scores for each item Such a comparison answered severa1 questions : (a) Are teachers scoring higher than their students in any areas? (b) Are students scoring higher than their teachers? (c) How much higher or lower are teachers' and students mean scores and how much discrepancy exists between scores on each item? (d) On which items, and constructs, do teachers and their students agree? (e) What is the nature of discrepancies between means in each of the three constructs 100

PAGE 114

under scrutiny? (f) Do the trends show significant standard deviations in students' scores in any of the constructs? In effect, the relationships between teachers' and students' scores regarding each construct were analyzed and the actual scores for each construct were also analyzed to determine how strongly teachers and students felt. These approaches to comparisons of means sufficiently answered the second research question. Because the mean scores reflect students' and teachers' perceptions ofleaming and instruction, these comparisons focused on discrepancies and similarities in perceptions. Limitations of the Study Ideally, the Instruction Inventory should have been distributed in the final third of a course, or any time after the tenth week of a fifteen-week semester. This would have given students ample time to develop their perceptions about the instructional strategies and their learning acquisition in the course. This timing probably would have given more accurate assessments than they did in the fifth and eighth weeks of the semester. Future research should test this hypothesis or determine whether students perceptions are as consistent in the second third as well as the final third of the course. Other limitations of this study are that it surveyed only fifteen teachers' classes, when a sample of fifty or more classes might have generated stronger trends in data. The teachers were each members of a single writing faculty in a large, 101

PAGE 115

public university, potentially limiting generalizability to other writing faculties or other types of colleges and universities. Even in light of these possible limitations, producing the quantity and quality of data that the Instruction Inventory does, lends itself to many types of analyses. This is a benefit of the instrument. Macro and micro analyses offer rich perspectives on the fundamental question that originally instigated this study: What is the viability of teaching writing with problem-based learning? The results are quite telling. 102

PAGE 116

CHAPTER FOUR REALIZING VALUES IN TEACHING WRITING WITH PROBLEM-CENTERED INSTRUCTION Various data analyses produced results that met this study's objectives: (a) to develop and validate a survey instrument that reveals how problem-centered instruction self-directed learning and critical thinking are perceived by teachers and students and (b) to examine the relationships among these three constructs as perceived by teachers and their students, in college composition classes. A factor analysis confirmed the study's first hypothesis, that a principal factor underlies all items in the Instruction Inventory. An oblimin rotation confirmed the second hypothesis that three dimensions underlie the inventory: problem-centered instruction self-directed learning and critical thinking with a set of items corresponding to each of these dimensions. This chapter is organized into two major sections : (a) results of the construct validity process of the survey instrument and (b) results of the comparisons of means that elucidate teaching and learning relationships in composition classes. 103

PAGE 117

Construct Validity of Instruction Inventory: Results A specific process was followed to begin the establishment of a body of empirical evidence to validate the constructs in the Instruction Inventory (Nunnally, 1978; Stanley & Hopkins, 1990). Because this is a new instrument and the construct validation process is ongoing, this study initiates the process but does not complete it. Plans to continue the validation process are detailed in Chapter Five. Roles and Results from Expert Judges The first step in the process was to elicit comments from judges, experts in problem-centered instruction, self-directed learning and critical thinking. Their comments on the draft of the Instruction Inventory were wide-ranging, from theoretical philosophizing to meticulous wordsmithing. Of the five judges who were asked to participate in tt !s research, three responded to the request; the other two declined to participate because of a lack of time. The judges generally agreed that the items in the Instruction Inventory represented the dimensions that they were intended to represent. The judges were asked to respond to a five-pomt Likert-type scale about how difficult it was for them to make their decisions about the appropriateness of the item. The most prevalent response was a "4," meaning that the judges agreed that the item was clearly associated with the underlying dimension it presented. One of the judges expressed concern about whether the first section of the instrument reflected enough critical elements of problem-based learning exactly. 104

PAGE 118

Upon close review, the draft of the survey did seem to be more focused on problem-centered instruction than problem-based learning. The difference here is noteworthy. Problem-based learning is a particular method of teaching and learning (Barrows, 1994), characterized by a particular learning process and outcomes (see Literature Review in Chapter Two). Problem-centered instruction, on the other hand, is less structured than problem-based learning in that it is a general instructional approach that integrates prohlems as the center and impetus of learning (Flower, 1989; Copeland, 2003). The comments from this judge were heeded and the construct was redefined as problem-centered instruction in lieu of problem-based learning. Each of the judges was concerned with the length and usability of the survey. The first drafts o ;_. the Instruction Inventory contained a total of 38 dimensions, each containing four positive and three negative items for each dimension, totaling 266 items in the survey (See Appendix D for the 266-item form of the Instruction Inventory sent to the judges). It was painfully long and Wlusable. The reason for this excess de'. elopment was to give the judges a choice of deleting the least relevant items. After revisions were made based on the judges' feedback, the pilot study form of the Instruction Inventory contained seventy items. an 85% reduction in size. Three redundant dimensions were removed and one positive and one negative item were given for each of the remaining 35 dimensions. 105

PAGE 119

Word choice, the judges consented, was also somewhat biased, which could lead to misleading, or false, results in the Instruction Inventory. Specifically, some of the negatively scored items were negatively worded and influenced the item. For example, "I encourage my students not to identify any of their learning goals" was changed to "My students are not always sure of how to define their personal learning goals." Not many teachers would admit to the first statement, but they would probably feel more comfortable admitting to the latter statement. Another example of negatively biased wording wa<> in the original statement, "My students do not fully accept responsibility for their own learning,'' which was changed to "My students take responsibility for their learning," a positively worded item. An astute comment from a judge was that several of the items involved motivation, not action. An original item that was motivational read, "My students believe that learning is their right to pursue." It was revised to "My students always look for new learning opportunities." Because the purpose of the survey is to operationalize constructs, action-oriented items were necessary. Motivation and beliefs do not represent what .1ctually happens in a classroom. The items in the revised Instruction Inventory describe students' and teachers' actions in the classroom. not beliefs and motivation. Pilot Study Results After the judges reviewed the instrument and revisions were made based on their recommendations, a pilot study was conducted in which the instrument was 106

PAGE 120

-------tested in two writing classes. Discussions with participants who completed the pilot study reported that the Instruction Inventory was easy to administer and complete. To identifY participants for the pilot study, an email was sent to all seventy members of the writing faculty of a large, public university requesting their voluntary participation (See Appendix E). Two instructors volunteered their classes to participate in the study. They received a packet of materials containing the surveys and envelopes. Each teacher reported that the entire process, including reading the directions and completing the survey took ten to twelve minutes. The teachers and their students were eager to discuss the survey immediately after they completed it. Their comments on its administration and their understanding of the items are noted in Table 4-1. Neither teacher h a d any questions about the directions that they received in the packet about how to administer the survey Students and teachers reported that the survey was easy to complete in less than fifteen minutes. Most comments about understanding the survey items were positive from the teachers, but the students said that some of the items seemed to be repetiti ve The Instruction Inventory was revised on the informative qualitative data collected in the pilot studies from the teachers and students. Eliminating a few redundant items comprised the major changes made in the piloted survey. Some redundancy was kept however, to strengthen the instrument's internal reliability. 107

PAGE 121

Table 4-1 Teachers' and Students' Comments after Pilot Study of Instruction Inventory Administration of Survey-Teachers' Comments I had no questions about the directions. It was all very clear. Thank you for providing such clear directions for me and the students. Administration of Survey-Students' Comments It took about ten minutes to take. I had no problems taking the survey. None of us had any questions about the directions. Understanding the Survey-Teachers' Comments All of the items were quite clear. I didn't have any questions about what any of the items meant. Understanding the Survey-Students' Comments Could you separate the similar items a little more? Some of the items seemed redundant. Could you use easier words to understand? Can you include a "Not Applicable" column? Some of these items don't apply. I think it's probably too early in the semester to do this. We aren't all that familiar with how our teacher does things. Some of the same questions were repeated. The negative questions are a little confusing. I wasn't sure if you were looking for a higher or lower number score. Some of the words are not in my vocabulary, like logical fallacy. I could have used 5 more minutes. The questions switched between good and bad. Other revisions included changing wording and moving items that pertained to similar issues. These changes reduced the instrument from seventy items in the pilot 108

PAGE 122

survey to 54 in the fmal survey, a reduction of 13%, or 80% of the original 266 items in the survey reviewed by the judges. See Appendixes G and H for the 54item student and teacher forms of the survey used in the final implementation, respectively). Principal Component Analysis and Oblimin Rotation Analysis A principal component factor analysis was conducted using all 54 items and produced a problematic result. Not all items loaded positively on a principal factor, and no three-factor rotation corresponded closely to the three constructs that the instrument was intended to measure. Accordingly, an exploratory analysis was conducted to arrive at a reduced set of items that would clearly reflect the three underlying scales. To test the first hypothesis, that a principal factor underlies all items in the Instruction Inventory, a factor analysis was conducted of all 54 items in the instrument. Initial analysis did not confirm the hypothesis, which is typical of preliminary factor analyses. To begin the process of reducing items that show a factor analysis and reliability analysis that support each other, items that clearly represented problem-centered instruction in the instrument were circled The following two items were used as the core on which to construct a measure of problem-based learning: (a) "My teacher tells us that applying solutions to problems is unimportant; (b) "My teacher tells us that group work can interfere with the pace of our learning. Both of these items dealt directly with working together to solve problems, and the correlation between them was moderate (.311 ). They were 109

PAGE 123

summed together, and the correlation between their sum and other problem-based learning items was computed. The highest correlation was with Item #19, and this item was added and the process repeated until a core of nine problem-based learning items had been identified with displayed internal consistency. Problem-based learning items that did not correlate with this set of items and did not load strongly with groups of self-directed learning items or critical thinking items were dropped. The same process was repeated with self-directed learning and critical thinking items, and finally as factor analysis was conducted with the reduced set of items. Reliability concerns whether an instrument will produce the same results each time it is given to the same person in the same environment. Because this study intended to measure three different constructs, each construct is considered a subscale and tested separately. This reliability within a subscale is also known as the internal consistency of the subscale and ultimately, the entire instrument. The total subscale correlation scores show the correlation of a single variable, or item, with the sum of the other variables in the subscale. The principal component analysis of the revised three subscales reports that a principal component does, indeed, underlie all items retained in the Instruction Inventory (see Table 4-2). The common theme of the principal component is that all items left in the instrument, after the aforementioned deletions, measure a particular philosophy of problem-centered instruction, self-directed learning, and critical thinking. In most general terms. this common element is student-centered learning. 110

PAGE 124

Table 4-2 Principal Component Matrix Component 1 PBLN6 .216 PBLN8 .164 PBLN9 .144 PBLP11 .247 PBLN12 .228 PBLN13 .282 PBLN15 .248 PBLN16 .328 PBLN17 .222 PBLN19 .336 SDLP21 .519 SDLN23 .415 SDLP24 .297 SDLP25 .570 SDLPJO .292 SDLP31 .442 SDLP33 .496 SDLN34 .299 SDLN35 .463 CTP37 .511 CTN39 .286 CTN40 .369 CTN43 .472 CTN45 .511 CTN47 .460 CTN48 .529 CTP49 .561 CTN51 .248 Note. Extraction Method: Principal Component Analysis. 3 components extracted. 111

PAGE 125

To confirm the second hypothesis, that three distinct dimensions comprise the inventory, an oblimin rotation analysis was conducted (see Table 4-3). Results reveal that three items did not load on the expected factor. Upon inspection of the wording of the individual items, the explanation for these exceptions became clear. For example, item #11, "My teacher encourages us to learn according to our interests," measures self-directed learning closer than it does problem-centered instruction. One of the main tenets of self-directed learning is that students have the leeway to design their o\\'n learning activities based on their own interests and needs (Candy, 1991 ). In a writing class, this means that students can choose topics that interest them instead of being asked to write about subjects that their teachers choose for them. Item #33, "I am comfortable challenging my classmates' ideas," more likely measures critical thinking than self-directed learning. Critical thinking promotes analysis and evaluation of others' ideas, so asking students to challenge their classmates' ideas, as is typical in a writing workshop, directly supports critical thinking views (Barnet & Bedau, 2002). lbe final item that seems misplaced in the survey is item #49, "I team because I want to learn," which seems to measure self directed learning more than critical thinking. Self-directed learners characteristically possess the internal motivation necessary to pursue learning, make time for it and fmd learning opportunities around them (Gerber, Lankshear, & Svensson, 1995). Results of the factor analysis, then, identified how each of these three items, #11, 112

PAGE 126

Table 4-3 Component Analysis with Oblimin Rotation Component I '1 3 PBLN6 .573 PBLN8 .453 PBLN9 .504 PBLPII .438 PBLN12 .388 PBLNI3 .497 PBLN15 .391 PBLNI6 .353 PBLNI7 .471 PBLN19 .656 SDLP2l .616 SDLN23 .359 SDLP24 .502 SDLP25 .693 SDLP30 .508 SDLP31 617 SDLP33 -.411 SDLN34 .331 SDLN35 .367 CTP37 .335 -.425 CTN39 -.498 CTN40 -.6!4 CTN43 -.500 CTN45 -.538 CTN47 -.524 CTN48 -.618 CTP49 664 CTN51 -.337 Notes: Extraction Method: Principal Component Analysis. Rotation Method: ObJimin with Kaiser Normalization. Rotation converged in 12 iterations. 113

PAGE 127

#33, and #49, loaded logically into components different from their original subscales. Reliability analyses for problem-centered learning was r = .63; self directed learning was r = .73; and critical thinking was r = .70. In essence. the principal component analysis confirmed the underlying principle that all items in the problem-centered instruction, self-directed learning, and critical thinking subscales measure an aspect of student-centered learning. The oblimin rotation confirmed that these three subscales are reliably distinct. Table 4-4 contains the distribution of items in each subscale. Teaching and Learning Relationships in Composition: Results In addition to the reliability testing and factor analyses to validate constructs in the Instruction Inventory, comparisons of teachers' and students' total scores were conducted to answer the second research question: What are the relationships between teachers' and students' perceptions of the following in the college writing class? a. Problem-centered instruction b. Self-directed learning c. Critical thinking Correlations of Student Subscale Scores A total score for each student was computed for each subscale to determine the relationships between subscales. As shown in Table 4-5, the strongest correlation between student subscore totals is r = .426. between self-directed 114

PAGE 128

Table 4-4 Item Distribution in Individual Sub s cales Item# 6 (N) 8 (N) 9 (N) 12 (N) 13 (N) 15 (N) 16 (N) 17 (N) 19 (N) 11 21 23 (N) 24 25 30 31 34 (N) 35 (N) 49 33 37 39 (N) 40 (N) 43 (N) 45 (N) 47 (N) 48 (N) 51 (N) Problem-Centered Instruction My teacher tells us that group work can interfere with the pace of our learning My teacher tells us that applying solutions to problems is unimportant. My teacher coaches us to solve problems quickly. My teacher suggests that we set aside related information we have learned in other classes. My teacher expects us to depend on her/him to further our learning. My teacher advises us to use only the strategies that we learn in class for our assignments. My teacher designs learning activities solely to meet the prescribed course objectives. My teacher encourages us to accept all experts research. My teacher has implied that the point of critical thinking is getting the right answers Self-Directed Learning My teacher encourages us to learn according to our interests. I always look for new learning opportunities I do not evaluate my own learning unless I'm asked to do so. I decide how, what, wh y, and when I learn. I spend time outside class exploring topics that interest me. I evaluate my own work fairly I am comfortable setting my own goals for learning. I think. writing reflections about my learning progress is generally a waste of time. I am unsure of how to define my own learning goals I learn because I want to learn Critical Thinking I am comfortable challenging my classmates ideas. I can dependably recognize gaps in logic I am uncomfortable experimenting with new ways to learn. 1 often confuse logical inferences with logical arguments. I construct logicai inferences from what I read and hear with difficulty I usually need help to analyze my own problems I frequently confuse paraphrasing with summarizing. I prefer to use current ideas than create new ideas I solve problems with my heart not my brains. Note (N) denotes a negatively scored item; scores have already been reversed, so to understand the scorer's intent, a negative fonn of the item must replace those above All other items are positively scored. 115

PAGE 129

Table 4-5 Correlations of Student Subscore Totals Problem-Based Learning (PBL) Self-Directed Learning (SOL) SOL Critical Thinking (CT) .180* .329* .426* Note. *Correlation is significant at the 0.01 leveL N = 228. learning and critical thinking. In other words, the stronger the students' perceptions that they were using self-directed learning skills in their writing class, the stronger they seem to believe they were also using critical thinking skills, although this cannot he a strong conclusion because of the relatively weak correlation (George & Mallery, 2001 ). These results also seem to report that there is internal consistency in how students view their writing classes in respect to a problem-centered instructional focus and their perceived use of self-directed learning and critical thinking skills. This is a logical finding, though. Self-directed learning and critical thinking are closely related, in that sett:.directed learners are typically critical thinkers, and vice versa ( Guglielmino & Guglielmino, 2001; Ruggiero, 1 988). However, similar positive correlations were expected between student total scores for problem-centered instruction with self-directed learning and critical thinking, but each of these correlations is particularly weak. Most noteworthy is the correlation between problem-centered instruction and self-directed learning, r = 116

PAGE 130

.180. These results actually show no correlations, positive or negative involving problem-based instruction with either of the other two constructs. Correlations ofTeacher Subscale Scores A total score for each teacher was also computed for each subscale. As shown in Table 4-6, the only significant correlation (r = .649) is between teachers' perceptions of self-directed lean and critical thinking. This positive correlation indicates that the more teachers perceive their students are using self-directed learning skills in their writing classes, the more they also perceive them to be using critical thinking skills. This is a logical finding. No significant correlation exists between teachers' perceptions of problem centered instruction and self-directed learning or critical thinking in their classes. This indicates that teachers do not perceive a significant increase, or decrease, of their students' self-directed learning or critical thinking skills with their own use of problem-centered instruction. As mentioned earlier, a positive correlation was expected between teachers' and students' perceptions problem-centered instruction and self-directed learning, so the finding of no correlation in teachers' scores is surprising. Perhaps a larger sample size and the elimination of outliers would indicate significant positive correlations in teachers' perceptions. Correlations of Teacher and Student Scores Just as noteworthy as the correlations within students' and teachers' subscale scores are the correlations between the teacher's score and her or his students' 117

PAGE 131

Table 4-6 Correlations ofTeacher Subscore Totals Self-Directed Learning (SDL) Critical Thinking (CT) .649* /'lote. *Correlation is significant at the 0.01 level. N = 15. scores. With all 15 ciasses included in the correlation, correlation between teachers' and their students scores did not exist. This is possibly because in such a small data set, a few outliers could mask a moderate correlation. A few teachers in other words perceive their teaching very differently from how their students perceive them. Scores in problem-centered instruction self-directed learning, and critical thinking are illustrated in scatter plots in Figure 4-l, Figure 4-2, and Figure 4-3, respectively. If the same classes were outliers in each of the three subscale correlations, another correlation could be analyzed without them to determine if any significant correlations exist. However, as evident in the plots, no consistent class outliers exist. These results seem to report that there is internal consistency in how students view their writing classes in respect to a problem-centered instructional focus and their perceived use of self-directed learning and critical thinking skills. However, there is no evident correlation at all between how teachers and students perceive the 118

PAGE 132

52 50 48 46 c c 44 c 42 c c G) 0 c 0 40 II) ....1 CD a.. 38 II) 36 .r; 0 Ill G) 34 t-34 36 38 40 42 Students' mean PBL scores Figure 4-1. Teachers' and their students' problem-based learning scores. 40 c ::_ c 0 0 0 30 II) = ....1 0 en II) c c "L CD c .c: 0 Ill CD 20 t-30 31 32 33 35 36 Students' mean SOL scores Figure 4-2. Teachers and their students' self-directed learning scores. 119

PAGE 133

40 Cl) ..... c 0 30 0 II) (] (] 1-(] (.) II) ..... -Cl) D .s:::. 0 0 m Cl) 20 129 30 31 32 33 34 35 36 Students' mean CT scores Figure 4-3 Teachers' and their students' critical thinking scores. same classes in respect to these constructs. Expected results were that teachers and students would hold similar perceptions of the degree of problem-centered instruction, self-directed learning, and critical thinking. A larger sample size may or may not bear these results. Relationships between Perceptions on All Items before Factor Analy s es Relationships between teachers' and students' perceptions in all 54 items in the inventory corroborate aforementioned comparisons of mean scores (see Appendix 1). Results from the problem-based learning instructional strategy subscale data show that teachers generally scored themselves at least one point 120

PAGE 134

higher, in many cases two or three points higher, than their students on most items in this section. This noteworthy discrepancy indicates that teachers generally believe that they were using more problem-based learning instructional strategies than their students believed were used in class. Results from the self-directed learning subscale data show some agreement between the teachers' and students' beliefs regarding students' acquisition of self-directed learning skills. A discrepancy did exist, however, in that students believe their self-directed learning skills are higher than what their teachers believe in several classes. Results from the critical thinking subscale data are similar to those of the self-directed learning items. Analysis shows that teachers and their students agree on many items in the level of students' acquisition of critical thinking skills. The major discrepancy between perceptions is that students scored their learning of critical thinking skills generally higher than their teachers scored them. Although a high level of agreement in learning between teachers and students is significant and desirable in any class, the discrepancies showing students' perceptions oftheir self-directed learning and critical thinking as higher than their teachers' perceptions is remarkable. This observation, in conjunction with teachers scoring themselves higher than their students in using problem-based instructional strategies, is also intriguing. 121

PAGE 135

In essence. teachers believe they used more problem-based learning instructional strategies than their students believed were used, but students believed themselves to be learning more self-directed learning and critical thinking skills than their teachers believed they were learning. Theoretically, because self-directed learning and critical thinking are elements of problem-based learning, all three constructs should be positively correlated; as perceptions of problem-based learning instructional strategies increase, so should perceptions of self-directed learning and critical thinking. This was not demonstrated by the Instruction Inventory in this study, however. A positive correlation does seem to exist between self-directed learning and critical thinking, but it is not necessarily negatively correlated with problem-based learning instructional strategies, either. The unknown construct that may explain these relationships is the very nature of perception itself. Further research in perceptions in learning is discussed in Chapter Five. Relationships between Scoring Trends Teachers scored more Ss than students in most items, especially in the problem-centered instruction items. This indicates that teachers perceive themselves to be "usually" or "almost always" using problem-based instructional strategies. than their students perceive them to be used, evident in such a high occurance of 5s. Students perceive teachers to be using problem-centered strategies "sometimes" or "usually." Although students perceive their teachers as delivering more problem-122

PAGE 136

centered instruction than less of the time, their scores were still not as high as their teachers' scores. Teachers and students are also more positive than negative about acquiring self-directed learning and critical thinking skills. Teachers' scores account for a significant majority of 1 s ("almost never") and 5s ("almost always"), whereas gstudents rarely scored these learning skills with 1 s or 5s. These extreme scores accounted for the largest discrepancies between teachers' and students' scores. Using a wider range of scores could indicate that either teachers were able to make clearer differentiations in self-directed learning and critical thinking skills than students were able to make, or students perceive their learning skills occurring in mid ranges. Most likely, a combination of both of these reasons account for the discrepancies in scores between teachers' and students perceptions in learning. Items with Largest and Smallest Discrepancies between Scores Items in which teachers and students scored the highest and lowest show minimal overlap with each other, as depicted in Table 4-7. Students and teachers scored items 6 and 12 high. They also each scored items 28 and 53 low. Item 6 (My teacher tells us that group work can interfere with the pace of our learning) and item 12 (My teacher suggests that we set aside related information we have learned in other classes) are each negatively scored items. That is, high scores on each of these items means that teachers and students agree that group work does not interfere with 123

PAGE 137

Table 4-7 Highest and Lowest Scored Items by Students and Teachers Highest Scored Items Lowest Scored Items Students' My teacher tells us that group #4 My teacher asks us to write about Scores work can interfere with the pace what we Jearn in class every of our learning. (N) day. 8 My teacher tells us that applying #27 I tend to be quiet when my solutions to problems is teacher gives us problems to unimportant. (N) solve. (N) 12 My teacher suggests that we set #28 I often request additional time to aside related information we have complete my homework. learned in other classes. (N) #44 I like to solve problems quickly 142 I often evaluate what I read and (N) hear. #53 Grades motivate me. (N) 49 I learn because I want to learn. Teachers #6 My teacher tells us that group #5 I frequently ask students to Scores work can interfere with the pace evaluate their own learning of our learning. (N) #28 I often request additional time to #12 My teacher suggests that we set complete my homework. aside related information we have learned in other classes. (N) I #29 My students are solely responsible for their learning. #15 I advise students to use only the strategies that they learn in class #50 My students rarely jump to for our assignments. (N) conclusions. #19 I have implied that the point of #53 Grades motivate my students. (N) critical thinking is getting the right answers. (N) #54 My students usually make time just to think about things in their #20 My students ask their classmates lives. for feedback on their work. Note. (N) denotes a negatively scored item; scores have already been reversed, so to understand the scorer's intent, a negative form of the item must replace those above. All other items are positively scored. Bold numbers are duplicates 124

PAGE 138

learning and that the teachers are not advising their students to set aside relevant information learned in other classes. The low scoring items, 28 (I often request additional time to complete my homework) and negatively-scored 53 (Grades motivate my students) reflect teachers' and students' agreement that students rarely request additional time to complete homework and that grades do motivate them. The items in the Instruction Inventory that showed the largest discrepancies between teachers' and students' scores(> .5) are shown in Table 4-8 Interestingly, most of the major discrepancies between studenls' and teachers' overall mean scores (> .5 points) occurred in the problem-based learning items; very few items in the self-directed and critical thinking learning sections showed major discrepancies between scores Most scores fell between 3.0 and 4.0, meaning that the responses ranged from sometimes true" to .. almost always true." Results show that of the 27 final items on the Instruction Inventory, 9 items, or 33%, show notewonhy discrepancies between teachers' and students' perceptions, or> .5. It is important to note that the focus here is perceptions, not evaluations, of teaching and learning. Teachers and students will inevitably disagree on the value of teaching and learning, such as whether or not keeping a journal benefits writing. They should, however, essentially agree on the type of instructional strategies used in the class, such as whether or not students discuss learning goals with each other. Teachers and students should also essentially agree on learning levels in self-directed and critical thinking skills, such as considering multiple 125

PAGE 139

Table 4-8 Items with Largest Discrepancies between Students' and Teachers Scores Students' Teachers Difference Mean Score Mean Score 3. My teacher often gives us problems to 3 .50 4.27 77 analyze and solve. 5. My teacher frequently asks us to evaluate 3.20 2.67 .53 our own learning. 7. My teacher requires us to write reasons for 3.65 4.47 .82 our claims ll. My teacher encourage:; us to learn 3 65 4.33 .68 according to our interests. I 31. I am comfortable setting my own goals for 3.93 3.27 .66 learning. 35. I am unsure of how to define my own 3.70 3.00 .70 learning goals (N) 36. I typically consider many perspectives 3 .71 3.00 .71 when solving problems. I I I 45. I usually need help to analyze my own 3.52 1 2.8o ...,, 'problems (N) I I I I 49. I Jearn because I want to learn. 4.18 3.60 .58 Note (N) denotes a negatively scored item : scores have already been reversed so to understand the scorer's intent, a negative form of the item must replace those above. All other items are positively scored perspectives when solving problems. Lack of agreement on perspectives most likely means some type of miscommunication between teachers and students. 126

PAGE 140

Probably the most perplexing discrepancies in scores concern goal setting in items 31 and 35. The results seem to contradict each other because one states comfort and the other states discomfort in goal setting, yet students scored themselves relatively high on each item and teachers scored their students relatively low on each item. These results are confounded by responses to item 2, "My teacher tells us to talk to each other about what we want to learn in class." In it, teachers believe that they ask students to discuss their ]earning goals more than students think that they do ask them. Why would teachers believe they ask students to talk to each other about what they want to learn in class if they do not believe in students' abilities to define their learning goals? Either these results are inconclusive and the items will require revision in future versions of the Instruction Inventory, or they demonstrate different understandings of goal setting between teachers and students. This could indicate destructive results in the writing class, as much of the writing process integrates goal setting and planning. In item 7, "My teacher requires us to write reasons for our claims," teachers believe that they ask students to rationalize their claims more than students believe they make this request. However, in item 45, "I usually need help to analyze my own problems." students believe they need more help than their teachers believe, yet teachers think they ask students to rationalize their claims more than their students believe. These items do not seem to corroborate each other. If teachers really do not believe their students need help analyzing problems, then why do they ask students 127

PAGE 141

to rationalize their claims? The Cognitive-Composition Model favors students identifYing their own strengths and weaknesses as writers-results that are seemingly supported by this study. In item 5 "My teacher frequently asks us to evaluate our own learning, teachers do not believe they are making this request of students, but students believe they are being asked to evaluate their own learning. Self-evaluation is part of becoming a self-directed critically thinking writer It is also one of the many instructional strategies inherent in problem-based learning. Implications of the results in this study could indicate a misconception of what self-evaluation is or how self-evaluation is done. Why do students perceive their teachers to be asking them to self-evaluate but their teachers do not share the perception? Perhaps students are developing as mature writers more than their teachers realize and are actua1ly evaluating their writing on their own volition. In item 3, "My teacher often gives us problems to analyze and solve teachers believe they are utilizing problem-solving strategies considerably more than their students believe they are being used. Discrepancies could be the result of teachers and students defining "problems" differently. They may also be interpreting "analyzing" and "solving" differently. Above and beyond semantic differences, however, the teachers' high scoring on this item indicates that they believe they incorporate problems in their instructional design. If they see writing assignments as problems and the writing process as a series of problems to solve, as 128

PAGE 142

is projected in the Cognitive-Composition Model, the results make sense. But the students evidently do not see writing assignments and the writing process as problems that require solutions. This is the type of miscommunication that can interfere with learning in the writing classroom. Another seeming miscommunication between students' and teachers' perceptions regards personalization in learning. In item 11, "My teacher encourages us to learn according to our interests," teachers strongly believe they do encourage students to learn according to their interests, but students do not hold such strong beliefs. However, teachers do not think their students Jearn because they want to learn; but claim they learn because they have chosen to do so, as noted in item 50, "I learn because I want to learn." Why would a teacher claim to encourage students to learn according to their personal interests, yet not give students significant credit for learning based on their own desires? In a writing class, this is akin to allowing students to choose their own writing topics, yet underestimate students' intentions to write because they want to write. Designing instruction based on the learners' personal interests typically piques the learner's interest. Writing teachers endorse this premise by giving students topic choice. However, teachers should also believe that students write because they want to write. Item 3 7, "I typically consider many perspectives when solving problems," also reveals noteworthy discrepancies between teachers' and students' perceptions. Students believe that they do consider multiple perspectives, but their teachers are 129

PAGE 143

not as positive. Perhaps the process of considering several perspectives is "hidden," or inherent, in students' learning processes, rendering their teachers unaware that they are considering multiple perspectives. Perhaps students translate "multiple" as two or three perspectives, while their teachers see "multiple" as four or five. Another explanation could be that teachers may want students to overtly acknowledge and qualify counter perspectives when solving a problem or developing a logical argument, but students do not see the value in giving attention to perspectives that could potentially defeat their solution or weaken their argument. Writing teachers often encourage students to wrjte from several perspectives about a single topic to promote their critical thinking. If students do not see the value of this, they will not write about multiple perspectives, even though they may be thinking about them. The items in the Instruction Inventory that showed the smallest discrepancies between teachers' and students' scores(< .1) are shown in Table 4-9. Students and teachers evidently agree with teachers' advice that group work does not interfere with learning, applying solutions to problems is important, and students often evaluate what they read and hear. These first two items also reflect agreement between problem-based learning instructional strategies. Agreement is also evident, although with slightly less conviction, in that teachers do not coach students to solve problems quickly and teachers regularly give students time in class to work in groups, both desirable problem-based learning instructional strategies. Agreement l30

PAGE 144

Table 4-9 Items with Most Similarities bern een Students' and Teachers 'Scores Item Students' Teachers' Mean Mean Scores Scores 6. My teacher tells us that group work can interfere with the 4.73 4.67 pace of our learning. (N) 8. My teacher tells us that applying solutions to problems is 4.49 4.47 unimportant. (N) 9 My teacher coaches us to solve problems quickly. (N) 3.52 3.53 18. My teacher regularly gives us time in class to work in 3 86 3.86 groups. 39. I am uncomfortable experimenting with new ways to learn. 3.42 3.40 (N) 41. I prefer not to judge other peoples' claims. (N) 3.32 3.40 42 I often evaluate what I read and hear. 4.00 4 07 Note. (N) denotes a negatively scored item ; scores have already been reversed, so to understand the scorer's intent a negative fonn of the item must replace those above. All other items are positively scored. between students and teachers also exists in that students are comfortable experimenting with new ways to learn and students prefer not to judge other peoples' claims. In essence, teachers and students virtually agree on 4 of 19 problem-based learning instructional strategies (21% ), 3 of 19 critical thinking skills (16%), and no self-directed learning skills. 131

PAGE 145

Discussion of Results In sum, internal consistency exists in how students perceive instruction and learning in their writing classes, but teachers and their students perceive instruction and skill use much differently. Teachers generally scored themselves higher than students in the problem-based learning instructional strategies and the studentsscored themselves higher than the teachers scored them in acquisition of self-directed learning and critical thinking. The question is, did the teachers overrate their instruction and underrate their students learning? Or did the students overrate their learning and underrate their teachers instruction? Because more overall discrepancies occurred in the problem-based learning instructional strategy items and more similarities occurred between scores in the self-directed learning and critical thinking items, it is more likely that the teachers were overrating their instruction and underrating their students than the students were overrating their learning. Implications of these results for writing teachers and other teachers interested in problem-based learning, self-directed learning, and critical thinking abound. 132

PAGE 146

CHAPTER FfVE CONCLUSIONS AND IMPLICATIONS In addition to further research and implications of the Cognitive Composition Model and Instruction Inventory, the conclusion of this study merits explanation of two fundamental matters: defining constructs and differentiating between classification and evaluation. Defining and operationalizing constructs is challenging. It was especially difficult in this study because the constructs, or subscales, overlap so much. Problem-centered instruction seems to promote self directed learning and critical thinking, while self-directed learning and critical thinking tend to thrive in a problem-centered instructional environment (Barrows, 1985; Candy, 1991; McPeck, 1990). Delineating these constructs was, consequently, necessary to understand the relationships between them. Table 5-1 summarizes the characteristics of each construct as they \\'ere used in this research. The need to clearly define and operationalize the constructs was also necessary to classify the instruction and learning skills. Factor analyses further classified them. Only then could the constructs be compared and their relationships analyzed. A dear difference exists between reliably cJassifying a classroom according to an approach and evaluating the value or worth of the class. Evaluation of the constructs was not a concern of this study, though. No attempt was made to 133

PAGE 147

Table 5-1 Characteristics of Constructs Used in Studv / Problem-Centered Instruction Self-Directed Learning Critical Thinking Personalized Self-reliant Imaginative Active Goal-directed Open-minded Analytical Accepting Curious Collaborative Responsible Skeptical Prob !em-oriented Controlled Logical Reflective Motivated Deliberative judge the worth of problem-centered instruction, self-directed learning, or critical thinking in college composition classrooms. No statement was made to claim one approach as more effective than the other. If it had, a definition of criteria would have been necessary and the study would have had a completely different intent. The purpose of this study was to develop an instrument that identifies teachers' and students' perceptions of instruction and learning, and to analyze the relationships between their perceptions of problem-centered instruction, self-directed learning, and critical thinking in college composition classrooms. This necessitated the matters of defining and classifYing the constructs. 134

PAGE 148

Future research opportunities with the Cognitive-Composition Model and Instruction Inventory can, ideally, involve every department of every college in this country, but especially writing departments. Implications of their use are that they can improve learning and thinking by students as well as teachers. It is important to note, however, that the instrument is developmental and requires continuing work to improve its constructs, face validity, and reliability because it did not fully meet the original hopes and needs of this particular study. This final chapter is divided into two main sections: implications of and future research of(a) the Cognitive-Composition Model and teaching writing and (b) the Instruction Inventory and perceptions in teaching and learning. The Cognitive-Composition Model and Teaching Writing The Cognitive-Composition Model synthesizes problem-centered instruction, self-directed learning, and critical thinking. Therefore, when items in the Instruction Inventory are interpreted in the terms of these constructs, they are simultaneously interpreted in terms of the Cognitive-Composition Model. The results of this study indicate fascinating implications about teaching writing. They also stimulate many ideas for further research. Implications The Cognitive-Composition Model can inform instructors who teach writing in any capacity, including those in non-English and non-Composition disciplines. Writing across the curriculum has received much attention recently because teachers 135

PAGE 149

in all disciplines are realizing that writing helps students to think about the learning activity at hand. Their challenges center around how to design writing assignments and then how to evaluate them. Non-writing teachers, then, need to understand the dynamics of writing as cognition to be able to incorporate it into their instructional design. They should know how to create writing cues that stimulate thinking that can not only help their students critically comprehend their lessons, but they also want to know how they can provide maximum assistance to their students. Writing is a practical strategy to incorporate into any instructional design, but it poses challenges to teachers who do not fully understand the dynamics of cognition in composition. The Cognitive-Composition Model provides a succinct guide for teaching writing because it describes the cognitive attributes of composition. Because the Model also depicts behavioral aspects of writing to show contrast, it offers a comprehensive picture of writers. It can stimulate constructive ideas for non composition teachers to incorporate into their instruction. Further Research The Cognitive-Composition Model provides an initial, solid foundation for understanding basic premises of teaching writing. Even so, relationships between the constructs revealed in this study pose provocative questions about teaching writing: To what degree do writing classes exhibit consistent problem-centered environments? Do students' self-directed learning orientations and critical thinking 136

PAGE 150

improve the more problem-based learning instructional strategies are used? How is critical thinking related to self-directed learning in writing students? In essence, these questions probe the nature of positive or negative correlation among the three constructs. If more constructs are added to the Instruction Inventory, such as intentional and reflective learning orientation, their correlation with the other three constructs would be interesting to explore. Because reflection and goal setting are integral to problem-based learning approaches, their correlation would probably be positively correlated with problem-ba'ied learning instructional strategies, self-directed learning, and critical thinking. How should additional constructs be included in the Cognitive-Composition Model? Thus, fmther research is needed to triangulate data and strengthen teachers' convictions and rationale to promote the traits comprising the Cognitive-Composition Model for students. Further questioning also leads to augmenting this study with qualitative data. Establishing a quantitative data base, analyzing it, and then augmenting it with qualitative data create a full perspective on the issue. Or, the converse approach of collecting qualitative data first analyzing it, and completing the cycle with quantitative data gives a wide-ranging view of the topic. The order of these research strategies is determined primarily by the initial research question. In this study, a quantitative approach was necessary to validate the constructs and then analyze their 137

PAGE 151

relationships. Thus, the next steps for this research could include collecting qualitative data such as text and discourse analyses Because this study found that no correlation existed between teachers and their students perceptions of problem-centered learning. self-directed learning, and critical thinking, interviews of teachers and students who have extreme differences of perception would be in order. A dialogue including the teacher and her or his students could bring out the discrepant points of view. Analysis of this open, spontaneous discourse would detail personal insights into teachers' and students' perceptions of instructional strategies and acquisition of learning skills. Questions posed in the discussion would also focus conversation around how the different instructional strategies and learning skills are related to each other. Analyzing course syllabi could also report informative qualitative data to inform this study. These documents can be analyzed in terms of attributes described in the Cognitive-Composition Model. For example, a syllabus that identifies all of the content to be learned in class without giving students any opportunity to write about or develop their personal learning goals would receive a low mark on the Cognitive-Composition Model. If the syllabus requests that students identify what they want to learn according to their interests, it would receive a favorable mark on the Cognitive-Composition Model. Syllabi could be evaluated in terms of how many attributes of the Model that they do or do not represent, perhaps in a point system or rubric. 138

PAGE 152

Another way to collect data that could inform relationships between problem-based learning instructional strategies and learning skills would be to observe and analyze class discussions and activities. Noting what teachers and students say and do in the classroom and how students perceive them would reveal the culture of the classroom and how the syllabus is operationalized. A discourse analysis of classroom conversatiuns could also be "scored" according to traits described on the Cognitive-Composition Model, complementing the syllabus analysis rubric. Analyses of teachers' comments written on students' papers in a writing class would also lend insight into triangulating future research. For example, comments from teachers that are directly instructive or scribbles that delete sections of the student's writing that alter the student's intentions are more teacherthan student-oriented and would score low on the Cognitive-Composition Model. On the contrary suggestive. reader-based comments in the form of questions are more student-centered and respectful of the writer because they encourage and trust the writer to make effective decisions, scoring high on the Model. A semester-long study of a teacher's comments on several students' papers would reveal valuable information about how that teacher operationalizes her syllabus and philosophy of teaching. Each of these research modes-validating additional concepts in the Instruction Inventory and analyzing syllabi, classroom discourse, teachers' 139

PAGE 153

comments, and focus contribute to teachers' understanding of how their instruction is perceived by them and their students. They can also reveal teachers' and students' perceptions of instruction and learning. Then again, careful analysis will answer the question, "To what extent does the Cognitive-Composition Model reliably express different aspects of problem-based learning instruction, self directed learning, critical thinking, intentional learning, and reflective learning?" This is yet another research question for further research. The Instruction Inventory and Perceptions The purpose of the Instruction Inventory in this study was to develop an instrument that measures perceptions of teaching and learning. To do so, a model was developed. From the instrument and model, conclusions were drawn to understand implications in teaching writing with problem-based learning. The instrument uncovered perceptions about the relationships between the problem centered instruction and students' use of critical thinking and self-directed learning skills. Considerations, for example, were: Do teachers or students determine the majority of what is learned in class? Are whole-classroom discussions led more by the teacher or the students? Do students look forward to group problem-solving activities? Continua illustrating contrasts in instructional approaches were developed to define such attributes of problem-based learning, self-directed learning, critical thinking, and rhetorical theory in composition. Follow-up interviews could also be conducted to examine these discrepancies in detail. 140

PAGE 154

Teachers can use the Instruction Inventory as a gauge midway through a course to determine how students are aligned with instruction. For instance, if a teacher believes that her or his students are assuming control of how they are learning to sharpen clarity in writing and enjoy writing as problem solving in groups, he or she can use the Instruction Inventory. Results reveal several things. Teachers can identify the degree of discrepancy between their responses and their students' responses. If the mean of the students' scores are similar to the teacher's scores, then they agree. If general agreement exists between the scores in the first section (instructional section) but not in the second section, then general agreement exists on instructional strategies but not on how students are acquiring self-directed learning and critical thinking skills. The Instruction Inventory, then, can be used as a formative assessment for teachers to align their instruction with students' needs. The Instruction Inventory can also be used as a summative assessment for teachers to judge the value of their instruction at the end of a course. In a sense, this instrument is a valid final course evaluation, able to replace many of the course questionnaires currently used in higher education. Another benefit of the Instruction Inventory is that it was designed to be discipline-independent, making it applicable in any field of study. Further research will test reliability of the instrument in various disciplines (i.e., physics, engineering, literature, history, psychology). As a survey instrument, the Instruction Inventory is valid, based on a body of empirical evidence collected during the construct-validity process implemented 141

PAGE 155

in this study; that is, it measures what it was intended to measure. Even so, the validation process is not complete. Additional empirical evidence is needed to reinforce the instrument's representation of problem-based learning, self-directed learning, and critical thinking. The nature of the construct validity process is that it is ongoing, continuously building a body of evidence to validate the constructs being measured. The adage that perception is reality carries some truth. For example, the way teachers perceive their instruction affect the way they design it. The way they see students' learning also affects her instructional design. Yes, tests, papers, and discussions yield evidence oflearning. Standard methods of evaluating students' learning, however, do not reveal perceptions about the process involved in the learning. In some cases, written examinations assess more of the students' test taking abilities than their knowledge or application of that knowledge. Grades that teachers assign to students' tests and papers are based on criteria and standards created by the teacher. In such cases, the teacher determines to what degree the students have "learned" a set of information. The missing link in traditional grading systems is that they do not reflect students' and teachers' perceptions of instruction and learning. Implications The Instruction Inventory gives educators a valid, reliable tool that ascertains teachers' and students' perceptions of instruction and learning in a 142

PAGE 156

classroom. While most instruments attempt to measure or assess the effect of a particular instructional strategy on learning, this tool focuses on perceptions of problem-based learning instructional strategies, self-directed learning and critical thinking. The implication here is that perceptions affect the teaching and learning processes just as much or more than instruction affects learning. When teachers understand their students' perceptions, and their own perceptions on what is happening in class, they can design student-centered activities that effectively improve students' learning. Another implication of the Instruction Inventory is that the three constructs under scrutiny problem-based learning instructional strategies. self-directed learning, and critical thinking-are empirically evident and can be measured The construct-validity process examines the survey instrument from severa l perspectives. It begins with thorough research and understanding of theories pertaining to the constructs under scrutiny. The process ends with factor analyses that mathematically test the reliability and validity of the constructs. Throughout this process, theory is transformed into succinct items that operationalize the theory. Academic debates typically explore transitions from theory to practice, but much of these discussions remain interestingly theoretical. The discussion concerns to what degree the value of theory is preserved or sacrificed during the shift. The transformation of theory to practice in the construct validity process however does not sacrifice the integrity of theories in any way. On the contrary, 143

PAGE 157

construct validity creates yet another perspective to understand underlying theories In tum, more theories can be generated, grounded, and ultimately qualified in the construct validity process. And so the cycle continues. Another use for Instruction Inventory is as a faculty development tool. It can serve as a formative assessment for teachers and as the foundation for a comprehensive faculty development program As a formative assessment, it can be administered about halfway through the semester to inform teachers how their students are perceiving instruction and learning. The teacher can compare the students mean scores with her own scores. Large discrepancies indicate disagreement, small discrepancies indicate some agreement, and similarities between scores indicate that teachers and students are in sync. After collecting and analyzing this data, instructors can hone their instructional design delivery and materials to increase students' understanding of lessons and decrease their misperceptions of it. The Instruction Inventory can. in this light. empower a teacher to create a highly student-centered environment, a goal that many teachers strive for but are unable to attain. As a comprehensive faculty development program, data collected from the Instruction Inventory across departments can be used as training materials for constructive professional discussions about perceptions in teaching and learning. Forums can present opportunities for teachers to share insights into their uses of the Instruction Inventory Strengths and weaknesses in instructors' teaching and 144

PAGE 158

students' learning can be identified and become the next workshop topic for faculty. Students can be invited to attend these discussions and contribute to teachers' understanding of how to interpret students' perceptions of teaching and learning. Instead of collecting and harboring data generated with the Instruction Inventory, it should be an opportunity for teachers and students to continue the understanding of the role of perceptions in education. To judge the effectiveness of the faculty development program, teachers can administer the Instruction Inventory in subsequent semesters and compare discrepancies and similarities between their scores and their students' scores. The goals are, of course, smaller discrepancies and more similarities between perceptions. Implications from the results of the Instruction Inventory can also be used to augment faculty course questionnaires in which students judge their teachers' effectiveness. Large discrepancies between a teacher's scores and her or his students' mean scores probably indicate larger disagreements than smaller discrepancies. Instead of a fill-in-the-bubble ''test" about the teacher's effects on their learning, students would have the opponunity to share their perceptions about their teachers' instruction and their own learning. In this capacity, the Instruction Inventory would give teachers valuable information about the teaching and learning that is actually happening, as perceived by their students. 145

PAGE 159

Further Research Continually building a body of evidence for the construct validity of the Instruction Inventory is a necessity for further research for this study; although the results were statistically reliable, they could be much stronger. Revising the instrument to include forty to fifty significantly reliable items is another goal for continued research in this study. Other prospects for future research include adding constructs such as reflective learning (metacognition and self-assessment (Courtney & Abodeeb, 1999; Leach, Neutze, & Zepke, 2000) and intentional learning (self-regulated, goal directed learning (Connelly, 2000; Nolen, 1996)). This would allow for a deeper understanding of teachers' and students' scope of perceptions in learning than is available with the current version of the Instruction Inventory. Reflective and intentional learning are also features of problem-based learning and adult learners, so their attributes would logically fit in the Cognitive-Composition Model. Further research activities could include planning, implementing, and evaluating a multi-year, substantial grant designed to re-validate and improve the Instruction Inventory, re-develop the Cognitive-Composition Model, and examine data across disciplines in a university setting. The Instruction Inventory would be distributed to hundreds of instructors and thousands of students in a university. Results within and between schools (i.e., Arts and Sciences, Business, Architecture, Education, Music, Law, and Journalism) could reveal 146

PAGE 160

provocative discrepancies--or alignments-in perceptions pertaining to teaching and learning in different types of classroom. The results can then be the focus of a comprehensive faculty development program described above. The sheer volume of data collected can reveal trends in perceptions in different schools and colleges unattainable by any other means. The implementation process would be the same basic construct validation process guiding this study. In such a grant project, the Instruction Inventory can also be used as an adjunct to faculty course questionnaires, the results of which would benefit teachers and their administrators. The role of perceptions in education inspired this study. Although much has been learned here, this study only begins to shed light on how teachers' and students' perceptions affect teaching and learning in a writing (or any other) class. The logical assumption imposed in this study is that teachers' perceptions of students' learning and students' perceptions of teachers' instruction do affect learning. But how direct is this influence? One of the strengths of this study is that it focused on relationships among perceptions, not effects of teaching on learning. Investigating the influence of perceptions in learning is, therefore, a serious topic for further research. Coda The quest to understand relationships between various instructional strategies and learning will never, and should never, end. This study provides a preliminary step in the exploration of teaching writing with problem-centered 147

PAGE 161

instruction. T caching writing is a huge responsibility because it can directly affect students' thinking and learning. Not only are composition and written conventions the subject of study in writing class, but so are self-directed learning, critical thinking, making decisions, and solving problems. Students who develop these skills are fortunate. Their teachers are just as privileged to be able to understand their relationships. This development of the Cognitive-Composition Model, Instructional Inventory, and values in teaching writing with problem-based learning is merely the commencement of exhilarating, enlightening, essential research and knowledge. 148

PAGE 162

Appendix A Coded Statements in Teachers' Form of Instruction Inventory That Stem from Problem-Based Learning Continuum Pl. Identifying content to be learned. a. I ask students to tell me what they want to learn in class. b. I invite discussion from students about course content. c. I encourage students to talk to each other about what they want to learn. d. It's okay for students to change their minds about what they want to learn in class. Tl. Learning determined by others. e. I assume students will depend on my choices for class content. f. I ask students to accept the university's course description. g. Students should trust their instructors' decisions on course content. P2. Personalizing own learning based on personal interests. a. I encourage students to learn according to their interests, not for a grade. b. In their introductions at the beginning of the course, I ask students what they want to learn during the semester. c. I encourage students to reflect on how their personal interests and experiences can enhance their learning. d. I give students time in class to think about how the course affects their growth. T2. Generalizing own learning. e. I tell students that their personal interests are extra, added concerns in our class. f. I design learning activities solely to meet the prescribed course objectives. g. I remind students that their personal interests are separate from course objectives. P3. Constructing knowledge actively when learning. a. I encourage student-; to construct knowledge when learning. b. I frequently ask students to write about what they are learning in this class. c. I regularly ask students to write about what they want to learn in class but have not mastered. d. I ask students to talk about relationships among the topics we discuss in class. 149

PAGE 163

T3. Constructing knowledge when learning. e. I want students to listen quietly while I lecture. f. I remind students that class materials are their main source of information g. I expect students to depend on me to build their content base. P4 problems. a. I ask students to explain the challenges that they face in their learning. b. I provide problems for students to solve in assignments. c. I encourage students to include others' perspectives in their thinking. d. I ask students to think about potential outcomes of their proposed solutions to problems. T4. Avoiding problems. e. I ask students to consider problems at face value. f I encourage students to solve problems intuitively. g. I coach students to be able to solve problems quickly. P5. Practicing logical reasoning and problem-solving skills. a. I frequently d1scuss the role of logic in learning with students. b. I require students to write their reasons for the claims they make. c. I always ask students to explain their opinions. d. I frequently challenge students' arguments. T5 Thinking compulsively and rashly e. I advise students to accept expens' research on its face value. f. I keep class debates to a minimum. g. I tell students that quick thinking is an art. P6. Participating actively when learning. a. I ask each student questions in class every day. b. I require students to write in class every day. c. I tell students to ask questions when they arise, not to wait till the end of class. d. I design small group activities every day T6 Participating on a passive low level of involvement when learning. e. I do not like to hear a lot of talking in class. f. I expect students to sit quietly while I teach. g. I am uncomfortable when students laugh unless I have told a joke. P7. Encouraging group problem solving. a. I regularly give students time to discuss assignments with each other. b. I encourage students to work with each other outside of class. c. I coach students how to critique each others' class work constructively. d. I ask students to help each other by sharing their proven strategies 150

PAGE 164

T7. Retractingfrom group problem solving. e. I tell students that group work complicates grading. f. I want students to focus on their own learning difficulties, not their classmates g. I explain to students that group work can sometimes interfere with the pace of their learning. P8. Implementing solutions to problems solved. a. I require students to show evidence of their application of solutions to problems. b. I ask students to write about the consequences of their solutions. c. I help students understand implementation as the final step in problem solving. d. I schedule class time to engage students in problem solving. T8. Not completing problem solution implementation. e. I explain to students that solutions that fail are faulty in some way. f. I tell students that implementation is a minor phase of problem solving. g I give students more help developing solutions than testing them. P9. Activating own prior knowledge. a. I ask students to write about what they've learned in the past about topics we are studying in our class. b. I design assignments that require students to use what they already know to complete them. c. I advise students to use their experiences when solving problems. d. I directly teach students how to use prior knowledge to benefit learning. T9. Not activating own prior knowledge. e I tell students that things they learned in other classes can complicate their learning in our class. f 1 encourage students to focus on learning new information to complete our class assignments. g. I advise students to use only new approaches that we learn in our class for our assignments. P 10. Learning independently. a. I support students who like to work on their own. b. I design class activities for students to learn independently. c. I want students to be comfortable learning alone as needed. d. I tell students that learning independently can be rewarding. TlO. Depending on others throughout learning process. e. I tell students that the best way to learn is to depend on others' input. f. I discourage students from working on their own. g. I explain to students that they should depend on their teachers to learn successfully. 151

PAGE 165

Pll. Thinkin g critically a. I encourage students to think deeply about most issues b I design class activities for students to go beyond the surface of the topic. c. I discuss how critical thinking can benefit my students' learning. d. I evaluate students' critical thinking on a daily basis Til. Posse s sing unreasonable irrational reasoning skills e. I evaluate students products of critical thinking but give them guidance on their processes. f. I tell students that the point of critical thinking is to get the right answers. g. I want students to be able to think critica11y on their feet about any topic. Pl2. Transftrring learning responsibility from the teacher to self a. I require students to accept responsibility for their learning. b. I tell students they are responsible for what they learn. c. I am responsible for encouraging students to own their learning d I show students how to be accountable for their learning. Tl2. Not transferring responsibility for learning from teacher. e. I expect students to depend on their teachers to further their learning. f. I tell students to share their responsibility for learning with their teachers. g. I want students to know that their teachers are completely responsible for their success. 152

PAGE 166

Appendix B Coded Statements in Teachers' Form of Instruction Inventory That Stem from Self-Directed Learning Continuum S I. Self-reliant in finding information a. My students don't need much help when it comes to research. b. My students spend time outside class exploring topics that interest them. c. My students are able to find whatever they need in the library. d. My students are able to find whatever they need online. 01. Needs help finding information. e. My students depend on me to help them with their research. f. My students ask me for research suggestions before they begin to research. g. My students need help fmding information in libraries they have not yet used. S 2. Sets persona/learning goais. a. My students write out their learning goals for our class. b. My students tell me what they want to learn in class. c. My students discuss their learning objectives with me. d. My students are comfortable setting their own goals for learning. 02. Does not set persona/learning goals e. My students may not be able to meet their learning goals if they were to identifY them. f. My students are not always sure of how to define their personal learning goals g. My students ask me to help them determine their learning goals for our class. S 3. Enjoys learning in isolation. a. My students are content to work on their own. b. My students prefer to work on projects individually. c. My students typically do better working in iso\ation than with a group. d. My students complain that we do too much group work in class. 03. Does not enjoy learning in isolation. e My students do not like class activities that require them to work individually. f. My students prefer small group work over individual seat work. g. My students believe that the best way to work is in small groups. 153

PAGE 167

S4. Finds time for learning. a. My students find enough time outside class to complete their homework. b. My students pursue personal learning opportunities outside cla ss. c. My students believe that learning opportunities exist around them every day. d. My students believe that learning takes place outside the classroom as much as inside it. 04. Does not find time for learning e. My students often express lack of time to complete their homework. f. My students prefer to work on assignments in class so they don't have to do them as homework. g. My students cram for tests and to write final papers. S5. Accepts feedback and criticism. a. My students accept criticism openly. b. My students ask their ciassmates to criticize their work c My students ask for my feedback on projects they re working on outside of our class. d. My studgents apply feedback from past assignment to future assignments. 05. Does not accept feedback and criticism. e. My students are defensive when I evaluate their work. f. My students withdraw from discussions after I critique their work. g. My students refer to their classmates' feedback infrequently after they receive it. S6 Takes control o_(personallearning a. My students claim ownership of their learning. b. My students know why they want to learn in our class. c. My students can explain their personal learning needs. d. My students control how, what why, and when they learn in our class. 06. Does not take control of persona/learnin g e. My students depend on me to monitor their learning. f. My students appreciate it when I set strict deadlines for assignments. g. My students always want clear, detailed explanations of assignments. S7 Seeks out/earning opportunities. a. My students frequently look for learning opportunities outside the classroom b. My students find exciting learning opportunities in most of what they do. c. My students are aware of what they learn from everyday experiences. d. My students talk to me about the lessons that they've learned outside our class. 154

PAGE 168

07. Does not seek out learning opportunities. e. My students believe that the best learning takes place in school. f. My students are aware of what they learn in classrooms more than personal experiences. g. My students accomplish most of their learning through our class assignments. S8. Enjoys problem solving. a. My students look forward to solving problems I present to them in class. b. My students are comfortable identifying challenges that arise for them in their learning. c. My students believe that problem solving is an integral part of learning. d My students give me suggestions for my challenges that I share with them. 08. Does not enjoy problem solving. e. My students would rather listen to a lecture than solve a problem. f. My students withdraw from class discussions when I present a problem in need of a solution. g. My students have inconsistent problem solving strategies. S9. Challenges assumptions. a. My students are comtortable challenging their classmates' ideas. b. My students play devil's advocate frequently in class. c. My students constructively criticize their classmates' opinions. d. My students often question me. 09. Does not challenge assumptions. e. My students know that an expert's ideas are accurate. f. My students trust that textbooks are correct. g. My students accept what their classmates say on face value. S 10. Assumes responsibility'for own learning. a. My students believe they are responsible for their learning. b. My students are comfortable assuming responsibility for their learning. c. My students hold a high sense of ow11ership of their learning. d. My students believe that learning is their right to pursue. 0 l 0. Does not assume responsibility for own learning. e. My students hold me accountable for their learning in class. f. My students would agree that their learning is directly dependent on my level of accountability to them. g. My students believe that they cannot be held responsible for their learning or lack of it. 155

PAGE 169

S 11. Evaluates and advocates own learning. a. My students grade their own work fairly. b. My students accurately assess the value of their learning processes. c. My students like to evaluate their work. d. My students want to know about strategies to assess their own learning. 011. Does not evaluate and advocate own learning. e. My students trust my evaluation oftheir work. f. My students blame me if they receive a poor grade. g. My students know that their teachers are their best learning advocates. S12. Reflects upon their own learning. a. My students are comfortable writing reflections of their learning. b. My students are able to reflect on the weaknesses as well as the strengths of their learning. c. My students' reflections of their learning help them to identify new strategies. d. My students are happy to reflect on their learning in whole class discussions. 012. Does not reflect upon own learning e. My students complain when I ask them to write reflections of their learning. f. My students write very little in their reflections. g. My students think that reflecling on their own learning is useless. 156

PAGE 170

Appendix C Coded Statements in Teachers' Form oflnstruction Inventory That Stem from Critical Thinking Continuum C 1. Analyzes problems a. My students are eager to analyze any problem I present to them in class b. My students are familiar with a logical sequence of solving problems. c. My students enjoy solving problems in small groups. d. My students identifY their own problems for which they can generate logical solutions. G 1. Stays away from problems when possible. e. My students would rather I offer solutions for their problems instead of them generating their own solutions f. My students enjoy listening to me lecture more than they like identifying their learning challenges in class. g. My students have expressed a clear dislike of discussing learning dilemmas in ciass. C2. Evaluates claims. assumptions, and evidence. a. My students identify claims in what they read and hear b. My students judge the value of claims in what they read and hear. c. My students identifY assumptions in what they read and hear. d. My students judge the value of assumptions in what they read and hear. G2. Accepts claims, assumptions and evidence on face value. e. My students misconstrue evidence for arguments in what they read and hear. f. My students often judge the role of an assumption in an argument incorrectly. g. My students misinterpret claims in what they read and hear. C3. Shows imaginative open-mindedness and intellectual curiosity. a. My students display their imaginations when developing their thoughts with unconventional ideas. b. My students accept al1 of their classmates' ideas. c. My students typically ask many conceptual questions in class. d. My students enjoy discussing a topic at length in class without getting bored. G3. Shows an objective approach to learning c. My students turn their attention to a topic on or off at will. f. My students accept what is said in a lecture and do not ask any questions. g. My students enjoy a detached approach to learning 157

PAGE 171

C4. Considers new ideas. a. My students consider new ideas when developing their own opinions about a topic. b. My students are open to their peers' opinions. c. My students are comfortable trying new ways to learn. d. My students ask their peers and me about new approaches to old tasks. G4. Depends onformer ideas. e. My students are more comfortable applying tried strategies to a problem than testing new strategies. f. My students depend more on former ideas than new ideas when writing. g. My students prefer to recycle former thoughts than create new thoughts. C5. Summarizes arguments accuratel_-v. a. My students accurately identify key points in an argument. b. My students create a hierarchy of relationships between the main points in a presentation. c. My students get to the point when they summarize what they heard or read. d. My students summarize what they heard or read accurately. G5. Summarizes arguments Tentatively. e. My students hesitate when identifying the highlights of an argument. f. My students misread the relationships between key points of what they heard or read. g. My students include details of a presentation when summarizing it. C6. Uses judicious, or reflective, skepticism. a. My students judge arguments before responding to them. b. My students think before speaking. c. My students take their time to reflect on what they read and hear. d. My students are skeptical when confronted with new concepts. G6. Accepts information without much reflection or contemplation. e. My students accept information without questioning it. f. My students are quick to respond to new theories. g. My students would rather respond than reflect on what they read and hear. C7. Constructs alternatives and solutions in problem-solving. a. My students are able to see both sides of an issue clearly. b. My students identify several choices before making decisions. c. My students spend time developing as many solutions as possible for problems they are solving. d. My students believe that identifying alternatives to problems is a critical phase of the problem-solving process. 158

PAGE 172

G 7. Applies first solutions developed in problem-solving. e. My students apply the first solutions they develop when problem solving. f. My students solve problems quickly. g. My students solve problems with their gut intuition. C8. Can recognize and construct sound arguments. a. My students recognize good arguments in most cases. b. My students are skilled at constructing sound arguments. c. My students identify clear claims in arguments. d. My students create logical support for their arguments. G8. Does not deliberately recognize or analyze sound arguments. e. My students misinterpret claims in arguments. f. My students confuse reports for sound arguments. g. My students do not go out of their way to analyze arguments. C9. Applies principles of formal and informal logic. a. My students are clear thinkers. b. My students think about problem situations before they respond to them emotionally. c. My students analyze problems to understand them before they begin to develop solutions for them. d. My students are objective problem solvers. G9. Va&TUe application of formal and informal logic principles. e. My students are subjective problem solvers. f. My students solve problems with their hearts, not their brains. g. My students are unfamiliar with formal and informal logic. CIO. Avoidsfailacies in reasoning. a. My students know what logical fallacies are. b. My students avoid fallacies in their reasoning at all costs. c. My students recognize fallacies in others' arguments. d. My students look for errors in their arguments when they are developing support. G 10. Makes fallacies in reasoning. e. My students commonly see a fallacy in their reasoning as poor writing. f. My students misidentify fallacies in reasoning. g. My students underestimate fallacies in reasoning. Cll. Suspends conclusions. a. My students consider as many alternatives as possible before making decisions. b. My students take their time when solving problems. c. My students do not rush their problem solving. d. My students do not jump to conclusions. 159

PAGE 173

G 11. Makes quick conclusions and judgments. e. My students solve problems quickly. f. My students make decisions quickly. g. My students do not waste time drawing conclusions. C12. Sets aside own ideas of the truth and reflects on alternatives. a. My students consider others ideas before they act on their own ideas. b. My students do not consider their concept of truth as the final word. c. My students seek out others perspectives when solving problems d. My students realize their ideas are not the final say on an issue. G 12. Depends on own ideas of the truth without e. My students are confident enough of their ideas not to seek other perspectives f. My students don't reflect on the value of others' ideas g. My students are quick to depend on their ideas to solve problems. C13.lnternally motivated about learning. a. My students learn because they want to learn. b. My students decide for themselves why they want to learn c. My students do not respond to rewards for learning d. My students learn for themselves, not others. G 13. Externally motivated about learning. e My students learn for a grade. f. My students invest effort when the reward is worthwhile. g. My students mistake passion for learning with their desire for high grades. C 14. Can judge and make logical inferences. a. My students identify logical inferences from illogical ones. b. My students judge the value of inferences. c. My students construct logical inferences from what they read and hear. d. My students write creative inferences. G 14. Hastil y judges and draws logical inferences e. My students are quick to judge inferences. f. My students do not waste time evaluating the merit of an inference g. My students confuse logical inferences with logical arguments. 160

PAGE 174

Appendix D Judges' Directions I am developing a survey for my doctoral dissertation research. and would greatly appreciate your expertise in validating the survey questions. I respectfully request your input by Thursday, January 15,2004. The survey will reveal relationships between writing teachers' and students' perceptions toward (1) problem-based learning instructional strategies used in their class and (2) the extent to which students acquired self-directed learning and critical thinking skills. Although the focus of my research is teaching writing with problem based learning, the survey can be used for any subject. The following definitions operationalize key terms: Problem-based learning an approach to learning that focuses students' learning on solving problem scenarios, exploring what they know and need to know m order to solve the problem, and evaluating their learning as individuals and team members. Self-directed learning-an approach to learning that encourages students' self concept of responsibility and control of their own learning throughout their learning process. including identification of their individual learning objectives, activities, and assessments. Critical thinking an approach to thinking that emphasizes purposeful, rationalized analysis and evaluation of others' statements, written and spoken, entailing problem-solving, intentional reflection, and creativity. Please note that the attached survey is the teachers' form. The students' form will contain parallel statements written expressly for students. 161

PAGE 175

Directions The survey contains two main sections: (1) problem-based learning attributes of instruction and (2) self-directed learning and critical thinking attributes of students. A numbered instructional strategy is followed by 7 statements that exemplifY the strategy for each section. 1. Circle the number in the Likert scale that best describes how clearly the lettered item is associated with the underlying dimension presented in italics above it. 2. Write comments as necessary. I will compile your feedback and revise the survey before I field test it and conduct the factor analysis in February. Please return your responses by Thursday, January 15,2004. Thank you in advance for your time and expertise developing my research. Rosalyn H. Zigmond 162

PAGE 176

Pl. Identifying content to be learned I a. I ask students to tell me what they want to learn I in class. 1 2 3 4 5 b. I invite discussion from students about course 1 2 3 4 5 content. 1 2 3 4 5 c. I encourage students to talk to each other about what they want to learn. 1 2 3 4 5 d. It's okay for students to change their minds about what they want to learn in class. Tl. Learning information determined by others. e. I assume students will depend on my choices for 1 2 3 4 5 class content. f. I ask students to accept the university s course 1 2 3 4 5 description. g. Students should trust their instructors' decisions 1 2 3 4 5 on course content. P2. Personalizing own learning based on personal interests. a. I encourage students to learn according to their 1 2 3 4 5 interests, not for a grade. b. In their introductions at the beginning of the 1 2 3 4 5 course, I ask students what they want to learn during the semester. c. I encourage students to reflect on how their 1 2 3 4 5 personal interests and experiences can enhance their learning. 1 2 3 4 5 I I d. I give students time in class to think about how I the course affects their growth. T2. Generalizing own learning. e. I tell students that their personal interests are 1 2 3 4 5 extra, added concerns in our class. f. I design learning activities solely to meet the 1 2 3 4 5 prescribed course objectives. g. I remind students that their personal interests are 1 2 3 4 5 separate from course objectives. 163

PAGE 177

P3. Constructing knowledge actively when learning. a. I encourage students to construct knowledge 1 2 3 4 5 when learning. b. I frequently ask students to write about what they 1 2 3 4 5 are learning in this class. c. I regularly ask students to write about what they 1 2 3 4 5 want to learn in class but have not mastered. d. I ask students to talk about relationships among 1 2 3 4 5 the topics we discuss in class. T3. Constructing knowledge unknowingly when learning. 1 2 3 4 5 e. I want students to listen quietly while I lecture. f. I remind students that class materials are their 1 2 3 4 5 main source of information. g. I expect students to depend on me to build their 1 2 3 4 5 I content base. P4. Analyzing problems. I a. I ask students to explain the challenges that they 1 2 3 4 5 face in their learning. b. I provide problems for students to solve in 1 2 3 4 5 assignments. c. I encourage students to include others' 1 2 3 4 5 perspectives in their thinking. d. I ask students to think about potential outcomes 1 2 3 4 5 oftheir proposed solutions to problems. T4. Avoiding problems. e. I ask students to consider problems at face value. 1 2 3 4 5 I f. I encourage students to solve problems 1 2 3 4 5 I I intuitively. 1 2 3 4 5 g. I coach students to be able to solve problems quickly. 164

PAGE 178

I P5. Practicing logical reasoning and problem-solving skills. a. I frequently discuss the role of logic in learning 1 2 3 4 5 with students. b. I require students to write their reasons for the 1 2 3 4 5 claims they make. c. I always ask students to explain their opinions. 1 2 3 4 5 d. I frequently challenge students' arguments. 1 2 3 4 5 T5. Thinking compulsively and rashly. e. I advise students to accept experts' research on its 1 2 3 4 5 face value. f. I keep class debates to a minimum. 1 2 3 4 5 g. I tell students that quick thinking is an art. 1 2 3 4 5 I P6. Participating actively when learning. a. I ask each student questions in class every day. 1 2 3 4 5 b. I require students to write in class every day. 1 2 3 4 5 c. I tell students to ask questions when they arise, 1 2 3 4 5 not to wait till the end of class. d. I design small group activities every day. 1 2 3 4 5 T6. Participating on a passive, low level of involvement I when learning e. I do not like to hear a lot of talking in class. 1 2 3 4 5 f. I expect students to sit quietly while I teach. 1 2 3 4 5 g. I am uncomfortable when students laugh unless I 1 2 3 4 5 have told a joke. 165

PAGE 179

P7. Encouraging group problem solving. a. I regularly give students time to discuss 1 2 3 4 5 assignments with each other. b. I encourage students to work with each other 1 2 3 4 5 outside class. c. I coach students how to critique each others' 1 2 3 4 5 class work constructively. d. I ask students to help each other by sharing their 1 2 3 4 5 proven strategies. 1 T7. Retracting.from group problem solving. e. I tell students that group work complicates 1 2 3 4 5 grading. f. I want students to focus on their own learning 1 2 3 4 5 difficulties, not their classmates'. g. I explain to students that group work can 1 2 3 4 5 I some_times interfere with the pace of their learmng. P8. Implementing solutions to problems solved a. I require students to show evidence of their 1 2 3 4 5 application of solutions to problems. I b. I ask students to write about the consequences of 1 2 3 4 5 their solutions. c. I help students understand implementation as the 1 2 3 4 5 fmal step in problem solving. d. I schedule class time to engage students in 1 2 3 4 5 problem solving. T8. Not completing problem solution implementation. e. I explain to students that solutions that tail are 1 2 3 4 5 faulty in some way. f. I tell students that implementation is a minor 1 2 3 4 5 phase of problem solving. g. I give students more help developing solutions 1 2 3 4 5 than testing them. 166

PAGE 180

P9. Activating own prior knowledge a. I ask students to write about what they've learned in the past about topics we are studying in our 1 2 3 4 5 class. b. I design assignments that require students to use 1 2 3 4 5 what they already know to complete them. c. I advise students to use their experiences when 1 2 3 4 5 solving problems. d. I directly teach students how to use prior 1 2 3 4 5 knowledge to benefit learning. T9. Not activating own prior knowledge. e. I tell students that things they learned in other 1 2 3 4 5 classes can complicate their learning in our class. f. I encourage students to focus on learning new 1 2 3 4 5 information to complete our class assignments. g. I advise students to use only new approaches that 1 2 3 4 5 we learn in our class for our assignments. PIO. Learning independently. a. I support students who like to work on their own. 1 2 3 4 5 b. I design class activities for students to learn 1 2 3 4 5 independently. c. I want students to be comfortable learning alone 1 2 3 4 5 I as needed. d. I tell students that learning independently is 1 2 3 4 5 rewarding. T 1 0. Depending on others throughout learning process. I e. I tell students that the best way to learn is to 1 2 3 4 5 depend on others' input. f. I discourage students from working on their own. 1 2 3 4 5 g. I explain to students that they should depend on 1 2 3 4 5 their teachers to learn successfully. 167

PAGE 181

j Pll. Thinking critically. a. I encourage students to think deeply about most 1 2 3 4 5 Issues. b. I design class activities for students to go beyond 1 2 3 4 5 the surface of the topic. c. I discuss how critical thinking can benefit my 1 2 3 4 5 students' learning. d. I evaluate students' critical thinking on a daily 1 2 3 4 5 basis. T 1 1. Possessing unreasonable irrational reasoning skills. e. I evaluate students' products of critical thinking, 1 2 3 4 5 but give them guidance on their processes. f. I tell students that the point of critical thinking is 1 2 3 4 5 to get the right answers. g. I want students to be able to think critically on 1 2 3 4 5 their feet about any topic. Pl2. Transferring learning responsibility from the teacher to self a. I require students to accept responsibility for their 1 2 3 4 5 learning. b. I tell students they are responsible fur what they 1 2 3 4 5 learn. 1 2 3 4 5 c. I am responsible for encouraging students to own their learning. d. I show students how to be accountable for their 1 2 3 4 5 I learning. l 112. Not transferring responsibility/or learning.from teacher. I e. I expect students to depend on their teachers to 1 2 3 4 5 further their learning. f. I tell students to share their responsibility for 1 2 3 4 5 learning with their teachers. I g. I want students to know that their teachers are 1 2 3 4 5 completely responsible for their success. 168

PAGE 182

S 1. Self-reliant in finding iriformation. a. My students don't need much help when it comes 1 2 3 4 5 to research. b. My students spend time outside class exploring 1 2 3 4 5 topics that interest them. c. My students are able to find whatever they need 1 2 3 4 5 in the library. d. My students are able to find whatever they need 1 2 3 4 5 I online. 01. Needs help finding iriformation. e. My students depend on me to help them with 1 2 3 4 5 their research. f. My students ask me for research suggestions l 1 2 3 4 5 before they begin to research. g. My students need help finding information in I 1 2 3 4 5 libraries thev have not yet used. S 2. Set s personal/earning goals. a. My students write out their learning goals for our 1 2 3 4 5 class. b. My students tell me what they want to learn in 1 2 3 4 5 class. c. My students discuss their learning objectives \\'ith 1 2 3 4 5 me. d. My students are comfortable setting their own 1 2 3 4 5 goals for learning. 02. Does not set persona/learning goals. e. My students may not be able to meet their 1 2 3 4 5 learning goals if they were to identify them. f. My students are not always sure of how to define 1 2 3 4 5 their personal learning goals. I g. My students ask me to help them determine their 1 2 3 4 5 I learning goals for our class. J 169

PAGE 183

S 3. Enjoys learning in isolation. I a. My students are content to work on their own. 1 2 3 4 5 b. My students prefer to work on projects 1 2 3 4 5 individually. c. My students typically do better working in 1 2 3 4 5 isolation than with a group. d. My students complain that we do too much group 1 2 3 4 5 work in class. 03. Does not enjoy learning in isolation. e. My students do not like class activities that 1 2 3 4 5 require them to work individually. f. My students prefer small group work over 1 2 3 4 5 individual seat work. g. My students believe that the best way to work is 1 2 3 4 5 in small groups. S4. time for learning. a. My students find enough time outside class to 1 2 3 4 5 complete their homework. b. My students pursue personal learning 1 2 3 4 5 opportunities outside class. c. My students believe that learning opportunities 1 2 3 4 5 exist around them every day. d. My students believe that learning takes place 1 2 3 4 5 outside the classroom as much as inside it. 04. Does not find time for learning. e. My students often express lack of time to 1 2 3 4 5 complete their homework. f. My students prefer to work on assignments in 1 2 3 4 5 class so they don't have to do them as homework. g. My students cram for tests and to write final 1 2 3 4 5 papers. 170

PAGE 184

SS. Accepts feedback and criticism. a. My students accept criticism openly. 1 2 3 4 5 b. My students ask their classmates to criticize their 1 2 3 4 5 work. I c. My students ask for my feedback on projects 1 2 3 4 5 they're working on outside of our class. d. My students apply feedback from past 1 2 3 4 5 assignment to future assignments. 05. Does not accept feedback and criticism. e. My students are defensive when I evaluate their 1 2 3 4 5 I work. f. My students withdraw from discussions after I 1 2 3 4 5 critique their work. g. My students refer to their classmates feedback 1 2 3 4 5 infreguently after thev receive it. S6. Takes control of personal learning. a. My students claim ownership of their learning. 1 2 3 4 5 b. My students know why they want to learn in our 1 2 3 4 5 class. c. My students can explain their personal learning 1 2 3 4 5 needs. d. My students control how, what, why, and when 1 2 3 4 5 they learn in our class. 06. Does not take control ofpersonal learning. e. My students depend on me to monitor their 1 2 3 4 5 learning. f. My students appreciate it when I set strict 1 2 3 4 5 deadlines for assignments. I g. My students always want clear, detailed 1 2 3 4 5 explanations of assignments. 171

PAGE 185

S7. Seeks out learning opportunities. a. My students frequently look for learning 1 2 3 4 5 opportunities outside the classroom. b. My students find exciting learning opportunities 1 2 3 4 5 in most of what they do. c. My students are aware of what they learn from 1 2 3 4 5 everyday experiences. d. My students talk to me about the lessons that 1 2 3 4 5 they've learned outside our class. 07. Does not seek out learning opportunities. e. My students believe that the best learning takes 1 2 3 4 5 place in school. f. My students are aware of what they learn in 1 2 3 4 5 classrooms more than personal experiences. g. My students accomplish most of their learning 1 2 3 4 5 through our class assignments. S8. Enjoys problem solving. a. My students look forward to solving problems I 1 2 3 4 5 present to them in class. b. My students are comfortable identifying 1 2 3 4 5 challenges that arise for them in their learning. c. My students believe that problem solving is an 1 2 3 4 5 integral part of learning. d. My students give me suggestions for my 1 2 3 4 5 challenges that I share with them. 08. Does not enjoy problem solving. e. My students would rather listen to a lecture than 1 2 3 4 5 solve a problem. f. My students withdraw from class discussions 1 2 3 4 5 when I present a problem in need of a solution. I g. My students have inconsistent problem solving 1 2 3 4 5 strategies. 172

PAGE 186

S9. Challenges assumptions. I a. My students are comfortable challenging their 1 2 3 4 5 classmates' ideas. b. My students play devil s advocate frequently in 1 2 3 4 5 class. c. My students constructively criticize their 1 2 3 4 5 classmates opinions d. My students often question me. 1 2 3 4 5 09. Does not challenge assumptions e. My students know that an expert's ideas are accurate 1 2 3 4 5 f. My students trust that textbooks are correct. 1 2 3 4 5 g. My students accept what their classmates say on 1 2 3 4 5 face value. S 10. Assum e s rejponsibility for own learning. a. My students believe they are responsible for their 1 2 3 4 5 learning. b. My students are comfortable assuming 1 2 3 4 5 responsibility for their learning. c. My students hold a high sense of ownership of 1 2 3 4 5 their learning. d. My students believe that learning is their right to 1 2 3 4 5 pursue. 010. Does not assume responsibility for own learning e. My students hold me accountable for their 1 2 3 4 5 learning in class. f. My students would agree that their learning is 1 2 3 4 5 directly dependent on my level of accountability to them. g. My students believe that they cannot be held 1 2 3 4 5 responsible for their learning or lack of it. 173

PAGE 187

j S 11. Evaluates and advocates own learning. a. My students grade their own work fairly. 1 2 3 4 5 b. My students accurately assess the value of their 1 2 3 4 5 learning processes. c. My students like to evaluate their work. 1 2 3 4 5 d. My students want to know about strategies to 1 2 3 4 5 j assess their own learning. I 011. Does not evaluate and advocate own learning. e. My students trust my evaluation of their work. 1 2 3 4 5 f. My students blame me if they receive a poor 1 2 3 4 5 grade. g. My students know that their teachers are their 1 2 3 4 5 best learning advocates. S 12. Reflects upon their own learning. 1 2 3 4 5 a. My students are comfortable writing reflections of their learning. 1 2 3 4 5 b. My students are able to reflect on the weaknesses as well as the strengths of their learning. 1 2 3 4 5 c. My students' reflections of their learning help them to identify new strategies. 1 2 3 4 5 d. My students are happy to reflect on their learning in whole class discussions. 012. Does not reflect upon own learning e. My students complain when I ask them to write 1 2 3 4 5 I reflections of their learning. f. My students write very little in their reflections. 1 2 3 4 5 g. My students think that reflecting on their own 1 2 3 4 5 learning is useless. 174

PAGE 188

1 Cl. Analyzes problems a. My students are eager to analyze any problem I 1 2 3 4 5 present to them in class. b. My students are familiar with a logical sequence 1 2 3 4 5 of solving problems. c. My students enjoy solving problems in small 1 2 3 4 5 groups. d. My students identifY their own problems for 1 2 3 4 5 which they can generate logical solutions G 1. Stays away from problems when possible. e. My students would rather I offer solutions for 1 2 3 4 5 their problems instead of them generating their own solutions. f. My students enjoy listening to me lecture more 1 2 3 4 5 than they like identifYing their learning challenges in ciass. g. My students have expressed a clear dislike of 1 2 3 4 5 discussin learnin dilemmas in class. C2. Evaluates claims, assumptions, and evidence a. My students identify claims in what they read and 1 2 3 4 5 hear. b. My students judge the value of claims in what 1 2 3 4 5 they read and hear. c. My students identifY assumptions in what they 1 2 3 4 5 read and hear. d. My students judge the value of assumptions in 1 2 3 4 5 what they read and hear. G2. Accepts claims, assumptions, and evidence on face value. e. My students misconstrue evidence for arguments 1 2 3 4 5 in what they read and hear. f. My students often judge the role of an 1 2 3 4 5 assumption in an argument incorrectly. g. My students misinterpret claims in what they read 1 2 3 4 5 and hear. 175

PAGE 189

C3. Shows imaginative open-mindedness and intellecrual curiosity a My students display their imaginations when 1 2 3 4 5 developing their thoughts with unconventional ideas. b. My students accept all of their classmates ideas. 1 2 3 4 5 c My students typically ask many conceptual 1 2 3 4 5 questions in class. d. My students enjoy discussing a topic at length in 1 2 3 4 5 j class without getting bored. G3. Shows an objective indifferent approach to learning 1 2 3 4 5 e. My students turn their attention to a topic on or I off at wilL 1 2 3 4 5 I f. My students accept what is said in a lecture and do not ask any questions 1 2 3 4 5 g. My students enjoy a detached approach to learning. C4. Considers new ideas. a. My students consider new ideas when developing 1 2 3 4 5 their o\\n opinions about a topic. b. My students are open to their peers' opinions. 1 2 3 4 5 c. My students are comfortable trying new ways to 1 2 3 4 5 learn I d. My students ask their peers a.d me about new 1 2 3 4 5 approaches to old tasks. I G4. Depends on former ideas. e. My students are more comfortable applying tried 1 2 3 4 5 strategies to a problem than testing new strategies. 1 2 3 4 5 I f. My students depend more on former ideas than new ideas when writing. 1 2 3 4 5 g. My students prefer to recycle former thoughts than create new thoughts 176

PAGE 190

C5. Summarizes arguments accurately. a. My students accurately identifY key points in an 1 2 3 4 5 argument. b. My students create a hierarchy of relationships 1 2 3 4 5 between the main points in a presentation. c. My students get to the point when they 1 2 3 4 5 summarize what they heard or read. d. My students summarize what they heard or read 1 2 3 4 5 accurately. G5. Summarizes arguments e. My students hesitate when identifYing the 1 2 3 4 5 highlights of an argument. f. My students misread the: relationships between 1 2 3 4 5 key points of what they heard or read. g. My students include details of a presentation 1 2 3 4 5 when summarizing it. C6. Uses judicious, or reflective, skepticism. a. My students judge arguments before responding 1 2 3 4 5 to them. b. My students think before speaking. 1 2 3 4 5 c. My students take their time to reflect on what 1 2 3 4 5 they read and hear. d. My students are skeptical when confronted with 1 2 3 4 5 new concepts. G6. Accepts information without much reflection or contemplation. I e. My students accept information without 1 2 3 4 5 questioning it. f. My students are quick to respond to new theories. 1 2 3 4 5 g. My students would rather respond than reflect on 1 2 3 4 5 what they read and hear. 177

PAGE 191

C7. Constructs alternatives and solutions in problemsolving. a. My students are able to see both sides of an issue 1 2 3 4 5 clearly. b. My students identifY several choices before 1 2 3 4 5 making decisions. c. My students spend time developing as many 1 2 3 4 5 solutions as possible for problems they are solving. 1 2 3 4 5 d. My students believe that identifYing alternatives to problems is a critical phase of the problemsolving process. G 7. Applies first solutions developed in problemsolving. 1 2 3 4 5 e. My students apply the first solutions they develop when problem solving. 1 2 3 4 5 f. My students solve problems quickly. 1 2 3 4 5 g. My students solve problems with their gut intuition. C8. Ca recognize and construct sound arguments. a. My students recognize good arguments in most 1 2 3 4 5 cases. b My students are skilled at constructing sound 1 2 3 4 5 arguments. c. My students identifY clear claims in arguments. 1 2 3 4 5 d. My students create logical support for their 1 2 3 4 5 arguments. G8. Does not deliberately recognize or analyze sound arguments. e. My students misinterpret claims in arguments. 1 2 3 4 5 I f. My students confuse reports for sound arguments. 1 2 3 4 5 g. My students do not go out of their way to analyze 1 2 3 4 5 arguments. 178

PAGE 192

I C9. Applies principles ojformal and iriformal logic. j a. My students are clear thinkers 1 2 3 4 5 I b. My students think about problem situations 1 2 3 4 5 before they respond to them emotionally. c My students analyze problems to understand 1 2 3 4 5 them before they begin to develop solutions for them. 1 2 3 4 5 d. My students are objective problem solvers. G9. Vague application ojformal and iriforma//ogic principles. e. My students are subjective problem solvers. 1 2 3 4 5 f. My students solve problems with their hearts, not 1 2 3 4 5 their brains. I g. My students are unfamiliar with formal and 1 2 3 4 5 informal logic. ClO. Avoids fallacies in reasoning. I a. My students know what logical fallacies are. 1 2 3 4 5 b. My students avoid fallacies in their reasoning at 1 2 3 4 5 all costs. c. My students recognize fallacies in others' I 1 2 3 4 5 arguments. d. My students look for errors in their arguments 1 2 3 4 5 when they are developing support. G 10. Makes fallacies in reasoning. e. My students commonly see a fallacy in their 1 2 3 4 5 reasoning as poor writing. f. My students misidentify fallacies in reasoning. 1 2 3 4 5 g. My students underestimate fallacies in reasoning. 1 2 3 4 5 Cll. Suspends conclusions. a. My students consider as many alternatives as 1 2 3 4 5 possible before making decisions. b My students take their time when solving 1 2 3 4 5 problems. c. My students do not rush their problem solving. 1 2 3 4 5 d. My students do not jump to conclusions. 1 2 3 4 5 G 11. Makes quick conclusions and judgments e. My students solve problems quickly. 1 2 3 4 5 f. My students make decisions quickly. 1 2 3 4 5 g. My students do not waste time drawing 1 2 3 4 5 conclusions. 179

PAGE 193

Cl2. Sets aside own ideas of the truth and reflects on alternatives. a. My students consider others' ideas before they 1 2 3 4 5 act on their own ideas b. My students do not consider their concept of truth 1 2 3 4 5 as the final word. c. My students seek out others perspectives when 1 2 3 4 5 solving problems. d. My students realize their ideas are not the fmal 1 2 3 4 5 say on an issue. Gl2. Depends on own ideas of the truth without reflection. e. My students are contident enough of their ideas 1 2 3 4 5 not to seek other perspectives f. My students don't reflect on the value of others' 1 2 3 4 5 ideas. g. My students are quick to depend on their ideas to 1 2 3 4 5 solve problems. C 1 ggglnternally motivated about learning. a. My students learn because they want to learn. 1 2 3 4 5 b. My students decide for themselves why they want 1 2 3 4 5 to learn. c. M y students do not respond to rewards for 1 2 3 4 5 learning. d. My students learn for themselves, not others. 1 2 3 4 5 G 13. External(.,.: motivated about learning. e. My students learn for a grade. 1 2 3 4 5 f. My students invest effort when the reward is 1 2 3 4 5 worthwhile. g. My students mistake passion for learning with 1 2 3 4 5 their desire for high grades. 180

PAGE 194

judge and make logical inferences. My students identify logical inferences from 1 2 3 4 5 I illogical ones. b. My students judge the value of inferences. 1 2 3 4 5 c My students construct logical inferences from 1 2 3 4 5 what they read and hear. d. My students write creative inferences. 1 2 3 4 5 G 14. Hastily judges and draws logical inferences. e. My students are quick to judge inferences. 1 2 3 4 5 f. My students do not waste time evaluating the 1 2 3 4 5 merit of an inference. g. My students confuse logical inferences with 1 2 3 4 5 lo cal ar ents. 181

PAGE 195

Appendix E Email Message to Faculty Requesting Them to Participate in Study Greetings. I need your help in my doctoral research. I have created a survey that explores writing teachers' and their students' perceptions of instructional strategies and lessons learned in the class. I am asking that you take 30 minutes and distribute it to any one of your classes. I need four classes to panicipate in the pilot study and distribute the survey between Monday, February 9 and Friday, February 13.ldeally, I'd like to visit with your classes immediately after you complete it to discuss the general experience, but not to discuss any responses to the questions-these will be kept strictly anonymous. I will then need the rest of you to distribute the fmal version of the survey between Monday, March 1 and Wednesday, March 10. Please know that each survey will come with an envelope so that each of your students a.1d you can seal it immediately and no one will see the results. You will also make up a code number that you share with your students, so that I can compare results, but that I will not be able to identifY with anyone. This will ensure anonymity. After I hear from you, I will leave a sealed envelope in your mailbox with detailed instructions and the survey forms and envelopes, of course! Your participation means a lot to me personally and the success of my doctoral research. Thank you in advance for your much needed help. Your respectful colleague, Rosalyn Zigmond 182

PAGE 196

Appendix F Instructions for Teachers Administering Instruction Inventory In the attached envelope are the following items: ..,. 1 survey marked "TEACHER FORM" ..,. 18 surveys marked "STUDENT FORM" ..,. 19 letter-size envelopes ..,. 1 large envelope marked "ALL SURVEY FORMS" After you and your students are done, place all surveys, including unused ones, and consent forms in the proper envelopes, seal the envelope, and put them in my mailbox in ENVD by ... 2:00p.m. on Friday, March 5, 2004. Please discard the envelope that I addressed to you containing these items. Thank you once again for participating in this study! Rosalyn PLEASE READ THE DIRECTIONS ON THE A IT ACHED PAGE TO STUDENTS! 183

PAGE 197

Please read the following to students before administering the survey: Class-Rosalyn Zigrnond, an instructor with CU's Program for Writing and Rhetoric, is conducting research for her doctoral project. The focus of the research is to examine writing teachers' and students' perceptions of various instructional strategies and the learning that results from them. Our total participation consists of completing a survey, which should take approximately 10 minutes. Please respond to the questions candidly and quickly, and in pen. I will be completing a survey also. Our identity will be kept strictly anonymous--even Ms. Zigmond will not be able to identify our surveys from other classes. Under no circumstances will I see your survey responses or you see mine. We will not put our name or any identifying information on the surveys. Instead, I have made up a code number for our class that only we will know. Please write this number in the upper right comer of your survey: XXXX:X. This is a number I have made up and will not tell anyone, except you! When you have completed the survey, place it in the envelope I will give each of you and seal it. I will then place all of your envelopes in this larger one and seal it. If you decide not to participate, rest assured that no repercussions will result from your decision. This survey is completely voluntary. Please read or write quietly until everyone has completed the survey. 184

PAGE 198

Appendix G Student Form of Instruction Inventory STUDENT FORM INSTRUCTIONS: This survey is designed to gather data on teaching and learning in your classroom. After reading each item, indicate the degree to which you feel the statement is true of your class today. Read each choice carefully, but try not to spend too much time on any one item Your first reaction is usually the most accurate --------------KEY--------------1 Almost never true I hardly ever do this 2 Not often true I do this less than half the time 3 -Sometimes true I do this about half the time. 4 Usually true I do this more than half the time 5 Almost always true. I usually do this 1. My teacher encourages quiet classrooms 1 2 3 4 5 2. My teacher tells us to talk to each other about what we want 1 2 3 4 5 to learn in class. 3. My teacher often gives us problems to analyze and solve. 1 2 3 4 5 4. My teacher asks us to write about what we learn in class 1 2 3 4 5 every day. 5 My teacher frequently asks us to evaluate our own learning. 1 2 3 4 5 6. My teacher tells us that group work can interfere with the 1 2 3 4 5 pace of our learning. 7 My teacher requires us to write reasons for our claims. 1 2 3 4 5 8. My teacher tells us that applying solutions to problems is 1 2 3 4 5 unimportant. 9 My teacher coaches us to solve problems quickly. 1 2 3 4 5 10. My teacher tells us that we are responsible for what we learn. 1 2 3 4 5 11. My teacher encourages us to learn accord ing to our interests. 1 2 3 4 5 185

PAGE 199

--------------KEY--------------1 Almost never true. I hardly ever do this. 2 Not often true. I do this less than half the time. 3 Sometimes true. I do this about half the time. 4 Usually true. I do this more than half the time. 5 -Almost always true. I usually do this 12 My teacher suggests that we set aside related information 1 2 3 4 5 we have learned in other classes 13. My teacher expects us to depend on her/him to further our 1 2 3 4 5 learning. 14. My teacher encourages us to take our time when thinking. 1 2 3 4 5 15. My teacher advises us to use only the strategies that we learn 1 2 3 4 5 in class for our assignments 16. My teacher designs learning activities solely to meet the 1 2 3 4 5 prescribed course objectives. 17. My teacher encourages us to accept all experts' research. 1 2 3 4 5 18. My teacher regularly gives us time in class to work in groups. 1 2 3 4 5 19. My teacher has implied that the point of critical thinking is 1 2 3 4 5 getting the right answers. 20. I often ask my classmates for feedback on my work. 1 2 3 4 5 21. I always look for new learning opportunities. 1 2 3 4 5 22. I believe that textbooks are always correct. 1 2 3 4 5 23. I do not evaluate my own learning unless I'm asked to do so 1 2 3 4 5 24. I decide how what why, and when I learn. 1 2 3 4 5 25. I spend time outside class exploring topics that interest me. 1 2 3 4 5 26. I depend on my teacher to monitor my learning. 1 2 3 4 5 27. I tend to be quiet when my teacher gives us problems to 1 2 3 4 5 solve 186

PAGE 200

KEY 1 Almost never true. I hardly ever do this. 2 Not often true I do this less than half the time. 3 -Sometimes true. I do this about half the time. 4 Usually true. I do this more than half the time. 5 Almost always true I usually do this 28 I often request additional time to complete my homework. 1 2 3 4 5 29. I am solely responsible for my learning. 1 2 3 4 5 30. I evaluate my own work fairly. 1 2 3 4 5 31. I am comfortable setting my own goals for learning. 1 2 3 4 5 32. I feel bad about myself when my teacher criticizes my work 1 2 3 4 5 33. I am comfortable challenging my classmates ideas 1 2 3 4 5 34. I think writing reflections about my learning progress is 1 2 3 4 5 generally a waste of time. 35. I am unsure of how to define my own learning goals. 1 2 3 4 5 36 I typically consider many perspectives when solving problems 1 2 3 4 5 37. I can dependably recognize gaps in logic. 1 2 3 4 5 38. I usually think before speaking. 1 2 3 4 5 39. I am uncomfortable experimenting with new ways to learn. 1 2 3 4 5 40. I often confuse logical inferences with logical arguments 1 2 3 4 5 41. I prefer not to judge other peoples' claims. 1 2 3 4 5 42. I often evaluate what I read and hear. 1 2 3 4 5 43. I construct logical inferences from what I read and hear with 1 2 3 4 5 difficulty. 44. I like to solve problems quickly. 1 2 3 4 5 45. I usually need help to analyze my own problems 1 2 3 4 5 187

PAGE 201

KEY 1 Almost never true I hardly ever do this. 2 Not often true I do this less than half the time 3 Sometimes true I do this about half the time. 4 Usually true. I do this more than half the time. 5 -Almost always true I usually do this 46. I generally speak before thinking. 1 2 3 4 5 4 7 I frequently confuse paraphrasing with summarizing 1 2 3 4 5 48. I prefer to use current ideas than create new ideas. 1 2 3 4 5 49. I learn because I want to learn. 1 2 3 4 5 50 I rarely jump to conclusions. 1 2 3 4 5 51. I solve problems with my heart, not my brains 1 2 3 4 5 52. I develop my thoughts imaginatively 1 2 3 4 5 53. Grades motivate me 1 2 3 4 5 54. I usually make time just to think about things in my life. 1 2 3 4 5 188

PAGE 202

Appendix H Teacher Form of Instruction Inventory TEACHER FORM INSTRUCTIONS: This survey is designed to gather data on teaching and learning in your classroom After reading each item, indicate the degree to which you feel the statement is true of your class today Read each choice carefully, but try not to spend too much time on any one item. Your first reaction is usually the most accurate. KEY 1 Almost never true. I hardly ever do this. 2 Not often true I do this Jess than half the time 3 Sometimes true. I do this about half the time. 4 Usually true. I do this more than half the time. 5 -Almost always true. I usually do this. 1. I encourage quiet classrooms 1 2 3 4 5 2. I tell students to talk to each other about what they want to 1 2 3 4 5 learn in class 3. I often give students problems to analyze and solve. 1 2 3 4 5 4. I ask students to write about what they learn in class every day 1 2 3 4 5 5 I frequently ask students to evaluate their own learning. 1 2 3 4 5 6. I tell students that group work can interfere with the pace of 1 2 3 4 5 their learning 7 I require students to write reasons for their claims. 1 2 3 4 5 8. I tell students that applying solutions to problems is 1 2 3 4 5 unimportant. 9 I coach students to solve problems quickly 1 2 3 4 5 10. I tell students that they are responsible for what they learn 1 2 3 4 5 11. I encourage students to learn according to their interests. 1 2 3 4 5 189

PAGE 203

KEY 1 A/most never true. I hardly ever do this. 2 Not often true. I do this less than half the time 3 Sometimes true. I do this about half the time 4 Usually true. I do this more than half the time. 5 Almost always true. I usually do this. 12. gl suggest that students set aside related information they have 1 2 3 4 5 learned in other classes. 13. I expect students to depend on me to further their learning. 1 2 3 4 5 14. I encourage students to take their time when thinking 1 2 3 4 5 15. I advise students to use only the strategies that they learn in 1 2 3 4 5 class for our assignments 16. I design learning activities solely to meet the prescribed 1 2 3 4 5 course objectives. 17. I encourage students to accept all experts' research. 1 2 3 4 5 18. I regularly give students time in class to work in groups. 1 2 3 4 5 19. I have implied that the point of critical thinking is getting the 1 2 3 4 5 right answers 20. My students ask their classmates for feedback on their work. 1 2 3 4 5 21. My students always look for new learning opportunities. 1 2 3 4 5 22. My students believe that textbooks are always correct. 1 2 3 4 5 23. My students do not evaluate their own learning unless asked 1 2 3 4 5 to do so. 24. My students decide how, what, why, and when they learn 1 2 3 4 5 25 My students spend time outside class exploring topics that 1 2 3 4 5 interest them. 26. My students depend on me to monitor their learning. 1 2 3 4 5 27. My students tend to be quiet when I give them problems to 1 2 3 4 5 solve. 190

PAGE 204

--------------KEY--------------1 Almost never true. I hardly ever do this. 2 Not often true. I do this less than half the time. 3-Sometimes true I do this about half the time. 4 Usually true. I do this more than half the time. 5 Almost always true. I usually do this. 28. My students often request additional time to complete their homework. 29. My students are solely responsible for their learning. 30. My students evaluate their own work fairly. 31. My students are comfortable setting their own learning goals. 32. My students feel bad about themselves when I criticize their work. 33. My students are comfortable challenging their classmates' ideas 34. My students think writing reflections about their learning progress is generally a waste of time 35. My students are unsure of how to define their own learning goals. 36 My students typically consider many perspectives when solving problems 37. My students can dependably recognize gaps in logic. 38 My students usually think before speaking 39. My students are uncomfortable experimenting with new ways learn. 40. My students often confuse logical inferences with logical arguments. 41. My students prefer not to judge other peoples' claims. 42. My students often evaluate they read and hear. 191 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 to 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

PAGE 205

KEY 1 Almost never true. I hardly ever do this. 2 Not often true I do this less than half the time 3-Sometimes true I do this about half the time. 4 Usually true. I do this more than half the time. 5 Almost always true. I usually do this. 43. My students construct logical inferences from what they read 1 2 3 4 5 and hear with difficulty 44. My students like to solve problems quickly. 1 2 3 4 5 45. My students usually need help to analyze their own problems. 1 2 3 4 5 46. My students generally speak before thinking. 1 2 3 4 5 47. My students frequently confuse paraphrasing with summarizing 1 2 3 4 5 48. My students prefer to use current ideas than create new ideas. 1 2 3 4 5 49. My students learn because they want to learn. 1 2 3 4 5 50. My students rarely jump to condusions. 1 2 3 4 5 51. My students solve problems with their hearts, not their brains. 1 2 3 4 5 52. My students develop their thoughts imaginatively. 1 2 3 4 5 53. Grades motivate my students. 1 2 3 4 5 54. My students usually make time just to think about things in 1 2 3 4 5 their lives. 192

PAGE 206

1. 2. 3. 4. 5. 6. 7. Appendix I Relationships between Teachers' and Students' Means for Instruction Inventory Items Teacher Teacher Students Students Students scoring scoring scoring scoring scoring 2 points 1 point 2 points 1 point the same Item higher higher higher higher as their than than than than teacher students students teacher teacher N !% I N % N % N % N % My teacher 6 40 4 27 1 7 3 20 l 7 encourages quiet classrooms. My teacher tells us to 7 147 l 7 l 7 5 34 1 7 talk to each other about what we want to learn in class. I My teacher often 3 20 7 47 0 0 2 13 3 20 gives us problems to analyze and solve. My teacher asks us to 2 13 3 20 0 0 3 120 7 47 I write about what we learn in class every day. My teacher frequently 4 '27 ,., 13 0 10 5 34 4 27 .... asks us to evaluate I our own learning. My teacher tells us 0 lo 2 13 1 7 2 13 10 67 that group work can interfere with the pace of our learning. My teacher requires 2 13 9 60 0 0 0 0 4 27 us to write reasons for our claims. 193

PAGE 207

Teacher Teacher Students Students Students scoring scoring scoring scoring scoring 2 points 1 point 2 points 1 point the same Item higher higher higher higher as their than than than than teacher students students teacher teacher N % N % N % N % N % 8. My teacher tells us 0 0 6 40 3 20 2 l3 4 27 that applying solutions to problems I is unimportant. : 9. My teacher coaches 2 I l3 2 13 3 20 3 20 5 34 us to solve problems quickly 10. My teacher tells us 7 47 5 34 0 10 0 0 3 20 that we are responsible for what we learn. 11. My teacher 2 13 8 53 1 7 1 7 3 20 encourages us to learn according to our interests. 12. My teacher suggests 0 0 II i3 0 0 1 7 3 20 that we set aside related information we have learned in other classes. 13. My teacher expects us 2 13 6 1 40 0 0 3 20 4 27 to depend on her/him I to further our 1-learning. 14. My teacher 0 O 7 47 0 0 4 27 4 27 encourages us to take our time when thinking. 15. My teacher advises us 3 20 9 60 1 7 1 7 I 7 to use only the strategies that we learn in class for our assignments. 194

PAGE 208

Teacher Teacher Students Students Students scoring scoring scoring s coring scoring 2 points 1 point 2 points 1 point the same Item higher higher higher higher as their than than than than teacher students students teacher teacher N % N % N % N % N % 16. My teacher designs 1 7 6 40 0 0 4 27 4 27 learning activities I solely to meet the prescribed course objectives. 17. My teacher 0 0 1 1 73 0 0 ., 13 2 13 .. encourages us to accept all experts' research. 18. My teacher regularly 1 7 8 53 2 13 3 20 l 7 gives us time in class to work in groups. 19. My teacher has 2 13 10 67 0 0 0 0 3 20 implied that the point of critical thinking is getting the right answers. 20. I often ask my 7 47 7 47 0 0 0 0 l 7 classmates for feedback on my work. 21. I always look for 1 7 l 7 0 0 6 40 7 47 1 new learning opportunities. 22. I believe that 0 0 7 47 1 7 3 20 4 27 textbooks are always correct. 23. I do not evaluate my 1 7 3 20 1 7 7 47 3 20 own learning unless I'm asked to do so. 24. I decide how, what, 1 7 3 20 2 13 5 33 4 27 why and when I learn 195

PAGE 209

Teacher Teacher Students Students Students scoring scoring sconng scoring scoring 2 points 1 point 2 points 1 point the same Item higher higher higher higher as their than than than than teacher students students teacher teacher N % N % N % N % N % 25. I spend time outside 1 7 4 27 1 7 3 20 6 40 class exploring topics that interest me. 26. I depend on my I 7 13 3 20 7 I 47 2 13 .:.. teacher to monitor my learning. 27. I tend to be quiet 1 7 3 20 0 0 1 7 IO 67 when my teacher gives us problems to solve. 28. I often request 3 20 5 33 0 10 0 0 7 47 additional time to complete my homework. I I 29. I am solely 0 0 1 7 4 27 7 47 3 20 responsible for my learning. 30. I evaluate my own 0 0 4 27 I 7 5 33 5 33 work fairly. 31. I am comfortable 0 jO 0 iO 4 27 2 13 9 60 setting my own goals for learning. 32. 1 feel bad about 0 0 2 13 '""' 13 6 40 5 33 .1.. myself when my I teacher criticizes my l work. 33. I am comfortable 0 0 6 40 1 7 2 13 6 40 challenging my classmates' ideas. 34. I think writing 0 0 8 53 0 0 2 13 4 27 reflections about my learning progress is generally a waste of time. 196

PAGE 210

Teacher Teacher Students Students Students scoring scoring scoring scoring scoring 2 points 1 point 2 points 1 point the same Item higher higher higher higher as their than than than than teacher students students teacher teacher N % N % N % N % N % 3 5. I am unsure of how 0 0 1 7 2 13 9 60 3 20 to define my own I learning goals. I I 36. I typically consider 0 lo 2 13 0 0 5 33 8 53 many perspectives I I I when solving I problems. 37. I can dependably 1 7 0 0 3 20 3 20 8 153 recogntze gaps m I logic. 38. I usually think 0 'o l 7 1 17 8 53 5 33 before speaking. 39. I am uncomfortable l 7 3 20 0 0 5 I 33 6 40 experimenting with new ways to learn. 40. I often confuse 1 7 1 7 1 7 5 33 7 47 logical inferences with logical I arguments. I 41. I prefer not to judge 2 113 2 i 13 1 !7 2 13 8 53 other peoples' claims. I 42. I often evaluate what I 7 4 27 0 '0 4 27 6 40 I read and hear. 43. I construct logical 1 '..,. 0 0 1 7 5 33 8 53 I inferences from what I read and hear with difficulty. 44. I like to solve 2 13 4 27 2 13 2 13 5 33 problems quickly. 45. I usually need help 0 0 1 7 3 20 5 33 6 40 to analyze my own problems. 197

PAGE 211

Teacher Teacher Students Students Students scoring scoring scoring scoring scoring 2 points 1 point 2 points 1 point the same Item higher higher higher higher as their than than than than teacher students students teacher teacher N % N % N % N % N % 46. I generally speak 0 0 2 13 2 13 5 33 6 40 before thinking. 4 7. I frequently confuse 0 0 1 7 2 13 5 33 7 47 paraphrasing with summarizing. 48. I prefer to use 1 13 1 7 6 40 5 33 ... current ideas than create new ideas. 49. I learn because I 0 0 1 7 2 13 5 33 7 47 want to learn. 50. I rarely jump to 0 tO 0 0 0 0 7 47 8 53 conclusions. 51. I solve problems 1 7 0 0 0 10 6 i 40 8 153 with my heart, not my brains. 52. I develop my 1 7 5 33 0 0 5 33 4 27 thoughts imaginatively. 53. Grades motivate mt. 1 7 2 13 2 13 3 '20 7 47 54. I usually make time 0 0 0 10 6 40 5 33 4 27 just to think about I I things in my iife. I Note. Items are from student form of survey. N = number of incidents how the teacher scored compared to the students in her or his class. 198

PAGE 212

REFERENCES Albanese, M. A., & Mitchell, S. (1993). Problem-based learning: A review of literature on its outcomes and implementation issues. Academic Medicine, 68(1), 52-81. Allen, D. E., Ouch, B. J., & Groh, S. E. (1996). The power of problem-based learning in teaching introductory science courses. In L. Wilkerson & W. H. Gijselaers (Vol. Eds.), New Directions for Teaching and Learning: Vol. 68. Bringing problem-based learning to higher education: Theory and practice, (pp. 43-52). San Francisco: Jossey-Bass. Allen, M., Sargeat, J., Mann, K., Fleming, M., & Premi, J. (2003). VideoconferGencing for practice-based small-group continuing medical education: Feasibility, acceptability, effectiveness, and cost. Journal of Continuing Education in the Health Professions, 23(1), 38-47. Allen, R., & Rooney, P. (1998, June). Designing a problem-based learning environment for ESL students in business communication. Business Communication Quarterly, (61)2. Retrieved March 20, 2002, from http://web5infotrac.galegroup.cornlitw/infomark/914/743/21060279w5.purl= rcl GBFM --Anfield, J. (1997). Joint development. People Management, 3(3), 34-37. Aspy, D. N., Aspy, C. B.. & Quinby, P.M. (1993). What doctors can teach teachers about problem-based learning. Educational Leadership, 50(7), 22-25. Ayres, F. (2002, February). Problem-based learning: The benefits to students and organizations. Training Journal. Retrieved March 20, 2002 from http://www.trainingjoumal.co.uk/abstracft/30202.htm Barnet, S., & Bedau, H. (2002). Critical thinking. reading, and writing: A brief guide to argument (4th ed.). Boston: Bedford/St. Martin's. Barrows, H. (1985). How to design a problem-based curriculum for the preclinical years. New York: Springer. 199

PAGE 213

Barrows, H. ( 1986). A taxonomy of problem-based learning methods. Medical Education, 20, 481-486. Barrows, H. (1988). The tutorial process. Springfield: Southern Illinois University School of Medicine. Barrows, H. (1994). Practice-based learning: Problem-based learning applied to medical education, Springfield: Southern Illinois University School of Medicine. Barrows, H. (1996). Problem-based learning in medicine and beyond: A brief overview. In L. Wilkerson & W. H. Gijselaers (Vol. Eds.), New Directions for Teaching and Learning: Vol. 68. Bringing problem-based learning to higher education: Theory and practice (pp. 3-12 ). San Francisco: Jossey Bass. Barrows, H., & Tamblyn, R. M. (I 980). Problem-based learning: An approach to medical education. New York: Springer. Bereiter, C., & Scardamalia, M. ( 1987). The psychology of written composition. Hillsdale, NJ: Lawrence Erlbaum. Berkenkotter, C. (1988). Decisions and revisions: The planning strategies of a publishing writer. In G. Tate & E. P. J. Corbett (Eds.), The writing teacher's sourcebook (2nd ed., pp. 128-139). New York: Oxford University Press. Berlin, J. ( 1987). Rhetoric and reality: Writing instruction in American colleges, 1900-1985. Carbondale: Southern Illinois University Press. Berlin, J. (1988). Contemporary composition: The major pedagogical theories. In G. Tate & E. P. J. Corbett (Eds.), The writing teacher's sourcebook (2nd ed., pp. 47-59). New York: Oxford University Press. Bigge, M. L., & Shermis, S. S. (1999). Learning theory for teachers (6th ed.). New York: Longman. Birch, S. (1986). Towards a model for problem-based learning. Studies in Higher Education, 11(1), 73-82. 200

PAGE 214

Boyatiz, R. E., Cowen, S. S., & Kolb, D. A. (1995). A learning perspective on executive education. Selections, 11(3), 47-55. Bridges, E. M., & Hallinger, P. (1996). Problem-based learning in leadership education. In L. Wilkerson & W. H. Gijselaers (Vol. Eds.), New Directions for Teaching and Learning: Vol. 68. Bringing problem-based learning to higher education: Theory and practice (pp. 53-62). San Francisco: Jossey Bass. Bridges, E. M .. & Hallinger, P. (1997). Using problem-based learning to prepare educational leaders. Peabody Journal of Education, 72(2), 131-146. Bringle, R. G., & Hatcher, J. A. (1999). Reflection in service learning: Making meaning of experience. Educationai Horizons, 77( 4 ), 179-185. Brockett, R. G., & Hiemstra, R. (1991 ). Self-direction in adult learning: Perspectives on theory, research, and practice. New York: Routledge. Brookfield, S. (1993). Self-directed learning, political clarity and the critical practice of adult education. Adult Education Quarterly, 43(4). Retrieved November 1, 2001, from http://nlu.nl.edu/ace/Resources/Documents/ Political Clarity .html Brooks, J. G., & Brooks, M. G. (1999). The case for constructivist classrooms. Alexandria, VA: Association for Supervision and Curriculum Development. Brown, W. S. ( 1998). Power of self-reflection through epistemic writing. College Teaching, 46(4), 135-138. Bruffee, K. A. (1993). Collaborative learning: Higher education, interdependence, and the authority of knowledge. Baltimore: Johns Hopkins University Press. Bruhn, J. G. (1997). Outcomes of problem-based learning in health care professional education: A critique. Family and Community Health, 20(1), 66-69. Bruner, J. (1961). The act of discovery. Harvard Educational Review, 31, 21-32. Bruner, J. H., Goodnow, J. L., & Austin, G. A. (1956). A study of thinking. New York: Wiley. 201

PAGE 215

Candy, P. L. (1991). Self-direction in lifelong learning: A comprehensive guide to theory and practice. New York: Routledge. Cashin, W. E., & Downey, R. G. (1995). Disciplinary differences in what is taught and in students' perceptions of what they learn and of how they are taught. In M. Theall & J. Franklin (Vol. Eds.), New Directions for Teaching and Learning (no. 64, pp. 81-92). San Francisco: Jossey-Bass. Coe, R. M. (1974). Rhetoric 2001. Freshman English News, 3, 1-3,9-13. Connelly, R. J. (2000). Intentional learning: The need for explicit informed consent in higher education. Journal of General Education, 49(3), 211-230. Copland, M.A. (2003). Developing prospective principals' problem-framing skills. Journal of School Leadership, 13. 529-548. Cottell, P. G. (2002). Problem-based learning-A bean counter takes the plunge. Journal ofStudent Centered Learning, 1(1), 7-18. Courtney, A. M., & Abodeeb, T. L. (1999). Diagnostic-reflective portfolios. Reading Teacher, 52, 708-714. Crocker, L., & AI gina, J. ( 1986). Introduction to classical and modern test theory. Orlando, FL: Holt, Rinehart and Winston. Crosby, C. (2003). Writer's block, merit, and the market: Working in the university of excellence. College English, 65, 626-645. Dewey, J. (1910). How we think Boston: D. C. Heath & Co. Dewey, J. (1938). Experience and education. New York: Touchstone. Dolmans, D. H. J. M., Gijselaers, W. H., Moust, J. H. C., Grave, I. H. A. P., & VanDer Vleuten, C. P.M. (2002). Trends in research on the tutor in probJem-based learning: Conclusions and implications for educational practice and research. Medical Teacher, 24(2), 173-180. Dolmans, D. H. J. M., & Schmidt, H. G. (2000).What directs self-directed learning in a problem-based curriculum? In D. H. Evensen & C. E. Hmelo (Eds.), Problem-based learning: A research perspective on learning interactions (pp. 251-262). Mahwah, NJ: Lawrence Erlbaum. 202

PAGE 216

Dolmans, D. H. J. M., Sclunidt, H. G., & Gijselaers, W. H. (1994a). The relationship between student-generated learning issues and self-study in problem-based learning. Instructional Science, 22, 251-267. Dolmans, D. H. J. M., Schmidt, H. G., & Gijselaers, W. H. (1994b). Use of student generated learning issues to evaluate problems in a problem-based curriculum. Teaching and Learning in Medicine, 6, 199-202. Drake, J. A. (1976). Teaching critical thinking: Analyzing, learning, and teaching critical thinking skills. Daville, IL: Interstate Printers. Edens, K M. (2000). Preparing problem solvers for the 21st century through problem-based learning. College Teaching, 48(2), 55-66. Elbow, P. (1973). Writing without teachers. New York: Oxford University Press. Elbow, P. (2000). Everyone can write: Essays toward a hopeful theory of writing and teaching writing. New York: Oxford University Press. Emig, 1. ( 1977). Writing as a mode of learning. College Composition and Communicaiion, 28, 122-28. Fenwick, T. J. (1998). Boldly solving the world: A critical analysis of problem based learning as a method of professional education. Studies in the Education of Adults, 30(1), 53-66. Filipczak, B., & Ganzel R. G. (1999). Motorola's self-directed learning lab. Training, 35(6), 18-20. Flower, L. (1989). Problem-solving strategies for writing (3rd ed.). Orlando, FL: Harcourt Brace Jovanovich. Flower, L. ( 1994a). The construction of negotiated meaning Carbondale: Southern Illinois University. Flower, L. (1994b). Rewriting collaborative planning. In L. Flower, D. L. Wallace, L. Norris & R. E. Burnett (Eds.), Making thinking visible: writing, collaborative planning, and classroom inquiry (pp. 132-136). Urbana, IL: National Council of Teachers of English. 203

PAGE 217

Flower, L. (1994c). Writers planning: Snapshots from research. In L. Flower, D. L. Wallace, L. Norris, & R. E. Burnett (Eds.), Making thinking visible: Writing, collaborative planning, and classroom inquiry (pp. 37-47). Urbana, IL: National Council of Teachers of English. Flower, L. & Hayes, J. R. (1988). The cognition of discovery: Defining a rhetorical problem. In G. Tate & E. P. J. Corbett (Eds.), The writing teacher s sourcebook (2"d ed., pp. 92-1 02). New York: Oxford University Press. Freire, P. (1990). Pedagogy of the oppressed (M. B. Ramos, Trans.). New York: Continuum. (Original work published 1970) Garger, E. M. (1999). Goodbye training, hello learning. Workforce, 78(11), 35-42. George, D., & Mallery, P. (2001). SPSSfor windows step by step : A simple guide and reference 10. 0 update (3'd ed). Needham Heights, MA: Allyn & Bacon. Gerber, R., Lankshear, C. L., & Svensson, L. S. (1995). Self-directed learning in a work context. Education & Training, 37(8), 26-32. Gere, A. R. (2000). Teaching writing: The major theories. In J. C. McDonald (Ed.), The Allyn and Bacon sourcebookjor coi/ege writing teachers (pp. 10-25). Needham Heights, MS: Allyn & Bacon. Getzels, J. W. (1985). Problem fmding and the enhancement of creativity. NASSP Bulletin 69(482), 55-61. Gijselaers, W. H. (1996). Connecting problem-based practices with educational theory. In L. Wilkerson & W. H. Gijselaers (Vol. Eds.), New Directions for Teaching and Learning: Vol. 68. Bringing problem-based learning to higher education, Theory and practice, (pp. 13-22). San Francisco: Jossey-Bass. Greenfield, L. B. (1987). Teaching thinking through problem-solving. In J. E. Stice (Vol. Ed.), New Directions for Teaching and Learning: Number 30 Developing critical thinking and problem-solving abilities (pp. 5-22). San Francisco: Jossey-Bass. Guglielrnino, P. J., & Guglielrnino, L. M. (2001 ). Moving toward a distributed learning model based on self-managed learning. S. A. M Advanced Management Journal, 66(3), 36-43. 204

PAGE 218

Guglielmino, P. J., & Murdick, R. G. (1997). Self-directed learning: The quiet revolution in corporate training and development. SAM Advanced Management Journal. 62(3 ) 10-18. Halpern, D. F. (1989). Thought and knowledge: An introduction to critical thinking. Hillsdale, NJ: Lawrence Erlbaum. Harman H. H. (1967) Modern factor analysis (2"d ed.). Chicago: University of Chicago Press. Hatcher, T. G. (1997). The ins and outs of self-directed learning. Training & Development, 51(2), 34-36 Heycox, K., & Bolzan, N. ( 1991 ). Applying problem-based learning in first-year social work. In D. Boud & G. Feletti (Eds.), The challenge of problem-based learning (pp. 186-193 ). New York: St. Martin's Press. Hiemstra, R. ( 1998, March). Self-advocacy and self-directed learning: A potential confluence for enhanced personal empowerment. Paper presented at the SUNY Empire State College Conference, Rochester, NY. Hillocks, G. ( 1995). Teaching writing as reflective practice. New York: Teachers College Press. Holt, D. G., & Willard-Holt, C. (2000). Let's get real: Students solving authentic corporate problems. Phi Delta Kappan, 82(3 ) 243-246. Hout, B. (2002). Toward a new discourse of assessment for the college writing classroom. College English, 65(2), 163-180. Kaufman, D. M. (1998, Winter/Spring). Problem-based learning: Using cases to teach about how to deal with ethical problems. NCEHR Communication. Retrieved March 20, 2002, from http://www.ncehr-cnerh.org/englishl communique2/PBLeaming html Knowles M. (1975). Self-directed learning: A guide for learners and teachers. Parsippany, NJ: Pearson Learning. Koon, J., & Murray, H. G. (1995). Using multiple outcomes to validate student ratings of overall teadher effectiveness. Journal of Higher Education, 66(1 ), 61-81. 205

PAGE 219

Krathwohl, D. R. (1998). Methods of educational and social science research: An integrated approach (2"d ed.). New York: Addison-Wesley. Lauer, J. M. (2002). Rhetoric and composition studies: A multimodal discipline. In C. R. McDonald and R. L. McDonald (Eds.), Teaching writing: Landmarks and horizons (pp. 121-133) Springfield: Southern Illinois University Press. Leach, L., Neutze, G., & Zepke, N. (2000). Learners' perceptions of assessment: gTensions between philosophy and practice. Studies in the Education of Adults, 3 2(1 ), 107-119. Lochhead, J., & Whimbey, A. (1987). Teaching analytical reasoning through thinking aloud pair problem solving. In J. E. Stice (Vol. Ed.), New Directions for Teaching and Learning: Number 30. Developing critical thinking and problem-solving abilities (pp. 73-92). San Francisco: Jossey Bass. McPeck, J. E. (1981). Critical thinking and education. New York: St. Martin's Press. McPeck, 1. E. (1990). Teaching critical thinking. New York: Routledge, Chapman and Hall. Meyers, C. ( 1986). Teaching students to think criticanv: A guide for faculty in all disciplines. London: Jossey-Bass. Morris, M. E., & Moore, P. C. (2000). The lessons we (don't) learn: Counterfactual thinking and organizational accountability after a close call. Administrative Science Quarter{v, 45(4), 737-765. Mullen, T. P. (1992). Integrating self-directed teams into the management development curriculum. Journal of Management Development, 11(5), 4354. Nolen, S. B. (1996). Why study? How reasons for learning influence strategy selection. Educational Psychology Review, 8(4), 335-355. Norris, S. P. (1992). The generalizability of critical thinking: Multiple perspectives on an educational ideal. New York: Teachers College. 206

PAGE 220

Norris, S. P ., & Ennis, R. H. ( 1989). Evaluating critical thinking. Pacific Grove, CA: Midwest Publications Critical Thinking Press. Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill. Perl, S. (1988). Understanding composing. In G. Tate & E. P. J. Corbett (Eds.), The writing teacher's sourcebook (2nd ed., pp. 113-118). New York: Oxford University. Pilling-Cormick, J. (1997). Transfonnative and self-directed learning in practice. New Directions for Adult and Continuing Education, 74 69-77. Rangachari, P. K. (1996). Twenty-up: Problem-based learning with a large group. In L. Wilkerson & W. H. Gijselaers (VoL Eds.), New Directions for Teaching and Learning : Vol 68. Bringing problem-based learning to higher education: Theory and practice (pp. 63-72). San Francisco: Jossey-Bass. Robotham, D. (1995). Self-directed iearning: The ultimate learning style? Journal of European Industrial Training, 19(1), 3-7. Roche, L.A., & Marsh, H. W. (2000). Multiple dimensions of university teacher self-concept: Construct validation and the influence of students' evaluations ofteaching. Instructional Science, 28(5-6), 439-468. Rohman, D. G. ( 1965). Pre-writing: The stage of discovery in the writing process. College Composition and Communication, 16, 106-112. Rubinstein, M. F., & Firstenberg, I. R. ( 1987). Tools for thinking. In J. E. Stice (Vol. Ed.) New Directions for Teaching and Learning: Number 30. Developing critical thinking and problem-solving abilities (pp. 23-36). Ruggiero, V. R. (1988). Teaching thinking across the curriculum. New York: Harper & Row. Schick, K. (2002). Valuating academic writing. In C. R. McDonald & R. L. McDonald (Eds.), Teaching writing: Landmarks and horizons (pp. 228-248). Springfield: Southern Illinois University Press. Schmidt, H .G., Dolmans, D. H. J. M., & Gijselaers. W. H. (1995). Theory-guided design of a rating scale for course evaluation in Problem-Based curricula. Teaching and Learning in Medicine, (7)2, 82-91. 207

PAGE 221

Seltzer, St., Hilber, St., Maceli, J., Robinson, E., & Schwartz, D. (1996). An active approach to calculus. In L. Wilkerson & W. H. Gijselaers (Vol. Ed.s.), New Directions for Teaching and Learning: Vol. 68. Bringing problem-based learning to higher education: Theory and practice (pp. 83-90). San Francisco: Jossey-Bass. Skinner, B. F. (1965). Science and human behavior. New York: Free Press. Smith, F. (1990). To think. New York: Teachers College. Stanley, J. C., & Hopkins, K. D. (1990). Educational and psychological measurement and evaluation (7th ed. ). Needham Heights, MA:Allyn & Bacon Stepien, W., & Gallagher, S. (1993). Problem-based learning: As authentic as it gets. Educational Leadership, 50(7), 25-29. Stice, J. E. ( 1987). Learning how to think: Being earnest is important, but it's not enough. In J. E. Stice (Vol. Ed.), New Directions for Teaching and Learning: Number 30. Developing critical thinking and problem-solving abilities (pp. 93-99). San Francisco: Jossey-Bass. Stinson, J. E., & Milter, R. G. (1996). Problem-based learning in business education: Curriculum design and implementation issues. In L. Wilkerson & \V. H. Gijselaers (Vol. Eds.), New Directions for Teaching and Learning: Vol. 68. Bringing problem-based learning to higher education: Theory and practice (pp. 33-42). San Francisco: Jossey-Bass. Tan, K. D., Goh, N., Chia, L., & Treagust, D. (2001). Secondary students' perceptions about learning qualitative analysis in inorganic chemistry. Research in Science and Technological Education, 19(2), 223-234. Thomas, P. B., & Harper, B.S. (2001). CPE: Changing with the profession. CPA Journal, 72(2), 30-34. Wallace, D. L. (1994). Teaching collaborative planning: Creating a social context for writing. In L. Flower, D. L. Wallace, L. Norris, & R. E. Burnett (Ed.s.), Making thinking visible: Writing, collaborative planning, and classroom inquiry (pp. 48-66). Urbana, IL: National Council of Teachers of English. 208

PAGE 222

Wilkerson, L. (1996). Tutors and small groups in problem-based learning: Lessons from the literature. In L. Wilkerson & W. H. Gijselaers (Vol. Eds.), New Directions for Teaching and Learning: Vol. 68. Bringing problem-based learning to higher education : Theory andpractice (pp. 23-32). San Francisco: Jossey-Bass. Winsor, K. (1991). Applying problem-based learning to teaching the practice of law. In D. Boud & G. Feletti (Eds.), The challenge of problem-based learning (pp. 217-224). New York: St. Martin's Press. Woods, D. R. (1994). Problem-based learning: How to gain the most .from PBL. Waterdown, ON: Donald R. Woods. Woods, D. R. (1996). Problem-based learning for large classes in chemical engineering. In L. Wilkerson & W. H. Gijselaers (Vol. Eds.), New Directions for Teaching and Learning: Vol. 68. Bringing problem-based learning to higher education: Theory and practice (pp. 91-1 00). San Francisco: Jossey-Bass. Yost, D. S. (2000). An examination of the construct of critical reflection: Implications for teacher education programming in the 21st century. Journal ofTeacher Education, 51(1), 39-49. Young, A. (2003 ). Writing across and against the curriculum. College Composition and Communication, 54(3), 472-485 Zimmerman, B J., & Lebau, R. B. (2000). A commentary on self-directed learning. In D. H. Evensen & C. E. Hmelo (Eds.), Problem-based learning: A research perspective on learning interactions (pp. 299-314). Mahwah, NJ: Lawrence Erlbaum. 209