Citation
Inner city high school students' attitudes, beliefs, and intentions toward using the internet for information retrieval

Material Information

Title:
Inner city high school students' attitudes, beliefs, and intentions toward using the internet for information retrieval
Creator:
Schnittgrund, Rita Therese
Publication Date:
Language:
English
Physical Description:
ix, 140 leaves : illustrations, forms ; 29 cm

Subjects

Subjects / Keywords:
High school students -- Attitudes ( lcsh )
Computer-assisted instruction ( lcsh )
Computer-assisted instruction ( fast )
High school students -- Attitudes ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 134-140).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Doctor of Philosophy, Administration, Supervision, and Curriculum Development.
General Note:
School of Education and Human Development
Statement of Responsibility:
by Rita Therese Schnittgrund.

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:
34055331 ( OCLC )
ocm34055331
Classification:
LD1190.E3 1995d .S36 ( lcc )

Downloads

This item has the following downloads:


Full Text
INNER CITY HIGH SCHOOL STUDENTS' ATTITUDES,
BELIEFS, AND INTENTIONS TOWARD USING THE INTERNET
FOR INFORMATION RETRIEVAL
by
Rita Therese Schnittgrund
B.A., St. Mary's College, Notre Dame, 1968
M.B.A., University of Denver, 1970
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Administration, Supervision, and Curriculum Development
1995


1995 by RitaTherese Schnittgrund
All rights reserved.


This thesis for the Doctor of Philosophy
degree by
Rita Therese Schnittgrund
has been approved for the
Graduate School
by
Kathy Escamilla
Brent Wilson
Duane Troxel


Schnittgrund, RitaTherese (Ph.D., Administration, Supervision, and
Curriculum Development)
Inner City High School Students' Attitudes, Beliefs, and Intentions Toward
Using the Internet for Information Retrieval
Thesis directed by Assistant Professor Dian Walster
ABSTRACT
We have entered the Information Age where mass production of
information and technology are primary concerns of society. The ability to
retrieve and use the information that our society is amassing in our personal
and business lives is becoming synonymous with the term "power."
Individuals who can retrieve, evaluate, analyze and use information will be
the leaders of the future. The purpose of this study is to determine the
attitudes, beliefs, and intentions of inner city high school students toward the
use of the Internet for information retrieval as a beginning step in creating a
resource-based curriculum that will affect the future behavior of inner city
students. As future community gatekeepers, it is essential that they be
educated in the use of this important source of power, information. The
study's subjects came from a metropolitan high school with a minority
population of 86%, predominantly Hispanic (79%).
IV


The Fishbein and Ajzen Model of Attitude-Behavior Consistency,
based on the Theory of Reasoned Action, was used to determine the
students behavioral beliefs and outcome evaluations, and normative beliefs
and motivation to comply. Fifty-five subjects who had one week exposure to
Internet navigation were involved. A set of salient behavioral and normative
beliefs, and intentions was determined from interviews with eight subjects.
These beliefs were then incorporated into a questionnaire following specific
Model guidelines that was completed by all subjects.
A correlation coefficient of .59997 or .60 (p < .001) between estimated
behavioral and normative beliefs, and estimated intentions was more than
sufficient to allow one to look at individual beliefs that created the structure.
Results of this study indicate that exposure alone is not sufficient to create a
highly positive belief structure that will insure future use of the Internet for
information retrieval. The values of the aggregate intentions to use the
Internet and normative beliefs are relatively low indicating that future usage
by a number of the subjects is not very promising unless some intervention
to change beliefs occurs.
This abstract accurately represents the content of the candidate's thesis. I
recommend its publication.
Signed tAJjb&r________________
/ Dian Walster
v


I
CONTENTS
CHAPTER
1. INTRODUCTION.................................................1
Information Age..............................................1
Information Literacy......................................4
Educating for Information Literacy........................5
Information Retrieval.......................................11
Electronic Information Systems (EIS)........................13
Electronic Mail..........................................16
USENET News..............................................16
TELNET...................................................17
File Transfer Protocol (FTP).............................17
Browsers and Search Engine Applications..................19
Internet Study..............................................23
Research Question........................................23
Literature Review........................................24
Methodology..............................................25
Study Organization.......................................26
2. LITERATURE REVIEW...........................................27
The Emerging Theory of Information Retrieval................27
Cognitive Science........................................27
VI


Online Searching.............................................32
Electronic Information Systems (EIS).........................37
Research Studies in Information Retrieval........................39
Overview.....................................................39
Information Retrieval Evaluation.............................41
Intermediary Research........................................44
End-User Research............................................47
Summary......................................................56
Telecommunications Research..................................57
Contribution of Current Study................................59
3. METHODOLOGY......................................................62
Introduction.....................................................62
Subjects.........................................................62
Fishbein and Ajzen Model.........................................65
Theory of Reasoned Action....................................65
Model........................................................68
Training of Students.............................................68
Interviews.......................................................69
Sample Selection.............................................69
Interview Questions..........................................70
Interview Procedure..........................................71
Analysis of Interview Responses..................................71
Questionnaire....................................................72
Navigation Logs..................................................73


I
4. ANALYSIS OF DATA..................................................74
Introduction.....................................................74
Interview Analysis...............................................75
Behavioral Beliefs...........................................76
Normative Beliefs............................................78
Intentions...................................................80
Questionnaire Analysis...........................................81
Behavioral Beliefs...........................................81
Evaluators of Behavioral Beliefs.............................83
Behavioral Beliefs (b) Times Evaluators (e)..................84
Normative Beliefs............................................86
Motivation to Comply with Normative Referents................87
Normative Referents (n) Times Motivation (m).................88
Intentions...................................................89
Regression Analysis..........................................90
Gender Comparison............................................93
Summary......................................................95
5. CONCLUSIONS AND IMPLICATIONS......................................96
Introduction.....................................................96
Conclusions......................................................98
Regression Analysis..........................................98
Behavioral Beliefs and Evaluators...........................100
Normative Beliefs and Motivation to Comply..................103
Intentions..................................................106
viii


Gender Comparison..............................107
Navigation Logs................................107
Summary........................................109
Future Research....................................111
APPENDIX
A. INTERVIEW QUESTIONS................................115
B. QUESTIONNAIRE ON INTERNET INFORMATION
RETRIEVAL..........................................116
C. RAW DATA FROM QUESTIONNAIRE........................121
D. HUMAN RESEARCH REVIEW APPROVAL.....................129
LIST OF REFERENCES........................................134
IX


CHAPTER 1
INTRODUCTION
In our lifetime, in the lifetimes of our students, and their
students, people will learn what they need to know by
accessing global electronic databases, and the local
proprietary databases, that will contain the totality of
available information, in forms that will organize that
information, or allow us to reorganize it, into whatever
forms may be most useful for our immediate purposes.
The successor to print literacy will be the set of skills
needed to locate and usefully organize information, for
ourselves and for others in cyberspace (Lemke, 1993,
P- 6).
Should Lemkes prediction come true, the impact on the way we must
educate our young for the future is profound.
Information Age
"In just one short human generation, the primary work of the world has
moved from the use of muscle to the use of machines, machines that move
information rather than goods" (Todd, Lamb, & McNicholas, 1992). We have
undoubtedly entered the Information Age where the mass production of
information and technology are the primary concerns of the society. The
speed and intensity with which this Information Age technology and
information databases are growing is unprecedented in human history.
1


I
I
It took 229 years from the invention of the Newcomen engine in 1708
to the evolution of the jet plane in 1937. However, it only took 36 years to
progress from the first generation of computers to the development of the fifth
generation of computers in the 1990s (Todd et al., 1992). In 1945, Vannevar
Bush, Director of the Office of Scientific Research and Development,
lamented that: "The summation of human experience is being expanded at
a prodigious rate, and the means we use for treading through the
consequent maze to the momentarily important item is the same as was
used in the days of square-rigged ships" (Bush, 1945, p. 102). As scientists
who banded together for research during World War II began to look towards
the beginning of a new age of research at the end of the war, the
"bewildering store of knowledge" (Bush, 1945) appeared to be
overwhelming. Yet, Masuda (1981) states that 90% of the 1981 scientific
knowledge base had been generated in the last 30 years and that this
knowledge base was expected to grow at a rate of 100% in the next ten to
fifteen years. The knowledge base of which Masuda speaks is non-inclusive
of the "overwhelming" database that Bush discussed in 1945.
The ability to retrieve and use this information that our society is
amassing in our personal and business lives is becoming synonymous with
the term "power." Individuals who can retrieve, evaluate, analyze, and use
information will be the leaders of the future. As students, citizens, and
workers, information literacy enhances the individual's power.
"Unfortunately, many of those most at risk, such as the urban
disadvantaged and members of the third world, are frequently the ones least
!
i
2


likely to have a background and the facilities to be empowered through
information" (Kazlauskas & lehl, 1993, p. 3). The reasons are not solely due
to the lack of teaching and availability in this area but often because of the
lack of prerequisite basic skills and predisposition. Basic skills and a
predisposition are important and necessary to the development of
information empowerment.
The purpose of this study is to determine the attitudes, beliefs, and
intentions of inner city high school students toward the use of the Internet for
information retrieval. An individual's beliefs determine intentions, which are
the direct determinants of behavior. Armed with a knowledge of their
predisposition to behave in certain ways, educators will be able to affect
behavioral change that encourages information literacy and the resultant
power that is necessary for survival in the 21st century.
These students are our leaders of the future. Within their own
communities some will act as gatekeepers. Gatekeepers are community
leaders that help others solve problems by accessing information from a
variety of resources (Metoyer-Duran, 1991). Their future behaviors can be
predicted by their intentions which in turn depends upon their attitudes and
beliefs. It is the educators' responsibility to ensure that these students are
positively exposed to the importance of information and electronic retrieval
methods in an attempt to shape their future behavior. Understanding
attitudes and beliefs can guide in curriculum development and presentation
methodology; provide valuable information for countering non-receptive
3


attitudes; and aid in developing an environment in which learning occurs as
a result of retrieval and use of information.
Information Literacy
Information retrieval is a major component in the drive for information
literacy for the 21st Century The term "information literacy" first appeared in
literature by Zurkowski in the mid-1970's (as cited in Kazlauskas & lehl,
1993; Todd et al., 1992). Zurkowski was, at that time, concerned about the
fact that only an estimated one-sixth of the population was information
literate. He called for a national program to achieve information literacy by
1984 (Kazlauskas & lehl, 1993). In today's changing environment, this cry
for information literacy is increasingly echoed.
As defined by a Delphi study group composed of 56 experts
recommended by representatives of the National Forum On Information
Literacy, information literacy is "the ability to access, evaluate, and use
information from a variety of sources" (Doyle, 1992, p. 2). The study
participants expanded on this statement by listing discrete attributes of an
information literate individual:
Recognizes the need for information
Recognizes that accurate and complete information is the
basis for intelligent decision-making
Formulates questions based on information needs
Identifies potential sources of information
Develops successful search strategies
4


Accesses sources of information including computer-based
and other technologies
Evaluates information
Organizes information for practical application
Integrates new information into an existing body of
knowledge
Uses information in critical thinking and problem solving
This new intellectual skill of information literacy will enable the individual to
conquer the new communications and information technologies. Those who
do not obtain these skills will likely be left behind (Kazlauskas & lehl, 1993).
"There is an increasingly urgent need to rethink and restructure
education processes within an information framework in order to provide
existing and future students with the attitudes, knowledge and skills they will
use and apply in their public and private roles as members of an information
society" (Todd et al., 1992, p. 2). In order to prepare our students to live and
work in an information-centered society, we must alter the manner in which
we are educating our children (Hancock, 1992).
Educating for Information Literacy
If information is to be the equivalent of power in the next century, then
educators must themselves be aware of the power of information, electronic
information availability and retrieval methods, and the importance of student
attitude in the development of information literacy behaviors. Extensive
5


teacher indoctrination and training is a requirement for implementation of an
information literacy curriculum.
In order to create an information literate environment, changes are
required in our traditional educational environment. A shift in the roles of
teachers and learners is essential in order to prepare students to live in an
information-centered society. Information literacy needs a resource-based
learning environment in which students are actively engaged and teachers
facilitate this active engagement (Hancock, 1992). Teachers can no longer
be the purveyors of a predetermined set of facts and the learners the
absorbers of this information. An information literate environment means
that teachers must give up the view that teaching is telling, that learning is
absorbing, and that knowledge is static.
The classroom walls must be extended into the local communities
and electronic information sources for information provided by businesses,
social service agencies, citizens' groups, public and university libraries,
cable and network television, radio broadcasts, national and international
print and electronic services, and subject matter experts willing to become
involved in the education of our youth. This type of environment frees the
teacher from being the omniscient expert; enables students to take
responsibility for their own learning; accommodates varied interest and
ability levels; and enables the students to become more effective consumers
of information resources (Sweller & Chandler, 1991). Attitudes and beliefs
toward information retrieval and usage must be positive for this to occur. An
understanding of students' current belief structure is essential for the
6


development of strategies that will enable students to accomplish these
goals.
Paradigms of the traditional view of education and the educational
environment required for the Information Age have been explicitly defined by
both Todd et al. (1992) and Lemke (1994). Todd defines the traditional view
of education as one in which:
teaching is talking and learning is listening
learning is a product, a destination
users of information are passive recipients, the destination of
information
each person's response to information is uniform and conforms to the
expected group response
the power structure is hierarchical and authoritarian
learners are viewed as robot-like information processors into which
information is poured
theoretical abstract book knowledge is primary source of learning
In contrast, the type of education needed for the 21st Century is seen as
emphasizing learning how to learn rather than learning an abstract body of
information selected for all by a hierarchical power. Todd et al. (1992)
envisions this environment as one in which:
learning becomes a never ending journey
autonomy and independent thought are encouraged
the learning environment is shared between student and teacher
what a student learns is not limited by age groupings
7


students assume the responsibility for learning and in the process
develop confidence and self-reliance
community input and even control is encouraged
The learner must be actively involved in information retrieval but, in
addition, must do something with the information to satisfy their learning
needs. Information creates meaning and understanding of a learning
situation that may vary from person to person. Learning becomes an active,
kinetic process enjoyed by both student and teacher alike.
Lemke (1994) predicts a paradigm war which he feels is essential for
the development of a 21st Century model of learning. The two conflicting
paradigms are the Curricular Model of Education, the traditional model, and
the Information Access Model of Learning, based on the norms of free
access to information. In the traditional Curriculum Model:
a uniform, prescribed course of study is designed for all students by
some power.
"A" correct and necessary curriculum is the basis of this model,
teachers are the medium through which information is passed on to
students.
students are evaluated as to the amount of "learning1' that occurred
by tests that measure how much of the information they can
reproduce.
students are guided down a well defined path to particular outcomes,
relevance of information is externally imposed.
uniformity is essential in the evaluation process.
8


"The Curricular Model does not value diversity and creativity, only conformity
and control" (Lemke, 1994, p. 4). Our current public schools epitomize the
Curricular Model of Education.
In opposition, the Information Access Model of Learning requires that
relevant information be determined by the students browsing global
databases in search of information matching their perceived needs and
interests. They must determine their own needs, retrieve information to meet
these needs, evaluate the information retrieved, and then synthesize and
transform this information into a relevant product. Evaluation is done by the
individual as well as others according to functionally based criteria and is
highly individualized. Students direct their own investigation and teachers
assist individuals in obtaining relevant information. Due to individual
student directed learning, the results of the learning process are
unpredictable and definitely not uniform. Therefore, the evaluation is also
individualized, while being based on criteria relevant to specific concerns.
The Information Access Model "embodies the values of free inquiry and
unfettered exploration, transformation, and synthesis of information into
beliefs and valuations" (Lemke, 1994, p. 4). This model is most closely
exemplified by the free public library.
Due to the basic philosophical differences and environmental issues
in these two models, the Information Access Model cannot succeed along
side the Curricular Model (Lemke, 1993). The Curricular Model must be
replaced. Will this ever happen to such a degree that students will begin
learning how to learn rather than learning how to absorb information?
9


Several educational institutions around the world have begun the slow
process of altering their beliefs and processes as to how the young should
be educated.
The Denver Public Schools in Denver, Colorado, have taken perhaps
the first step by redefining its goal of literacy ("A New Literacy for the 21st
Century," 1994). According to the District, the need for this new definition of
literacy is justified by the fact that reading and writing are not adequate for
survival in the highly literate and complex 21st Century. The New Literacy
reaches well beyond the ability to read and write. It includes the ability to:
create, acquire and transform knowledge
be adaptable and flexible
conceptualize and reconceptualize
think critically
capitalize on change
capitalize on complexity
integrate the general with the specific
learn how to learn
build learning organizations
act ethically and understand values
This definition of literacy has been distributed throughout the district, and
students and teachers have been encouraged to move beyond the
classroom and explore the world's full resources. However, at this point,
rhetoric is the primary move toward this philosophy.
10


Several educational systems in Australia have moved beyond the
rhetorical stage and are in the beginning stages of implementing an
information literate environment. This move has been backed by the
Australian government's landmark 1991 document, "Australia as an
Information Society," which asserts that access to information is
"fundamental to our democratic freedom, to the transformation of our
economy, and to the delivery of social justice to all citizens in other words,
to our very survival" (Todd et al., 1992, p. 6). The Marist Sisters' College in
Sydney, Australia, has been involved in restructuring their learning
environment to conform to information literacy principles for the last four
years. Findings from research on their efforts indicated an enormous impact
on the learners' perception of themselves as individuals and as learners, on
the process of learning and learning outcomes, and environmental aspects
such as respect, interest, and collaboration. The process of teaching was
also impacted (Todd et al., 1992).
Information Retrieval
If one is to teach retrieval skills to prepare students for electronic
environments of the Information Age, or design search systems that adapt to
users as their capacity changes, understanding the process and how users
interact with the information retrieval system is essential (Solomon, 1992).
Information retrieval is a cognitive problem solving activity that is dependent
on a wide variety of factors: field or discipline, physical environment, cost,
time constraints, the information problem, search system (content,
11


organization, interface, etc.), user characteristics, personal experiences and
knowledge, and search outcomes (Marchionini, 1989a).
Marchionini (1992) delineates eight sub processes that are involved
in an information seeking activity:
problem recognition and acceptance
problem definition
search system selection
query formulation
query execution
examination of results
information extraction
reflection/iteration/termination.
Problem recognition is the acknowledgment of an information gap or
need and the conscious commitment to fill the gap. Once the commitment
has been made, the most critical step in the whole information-seeking
process is problem definition. One must identify key concepts and
relationships and then choose vocabulary that will transfer these into an
information-seeking task. The choice of search system is linked to a
person's experiences with different systems, their domain knowledge, their
expertise in the information searching process, and the nature of the
information need. Query formulation and reformulation depends on many of
these same factors. Domain and systems knowledge guide the selection of
terms and format of the query. Query execution involves the physical task of
the process: the picking up of a print volume, selecting the correct card in a
12


catalog, or the pressing of the keys on a keyboard. The examination of the
results of the search is another critical step in the process. This is when
relevance decisions are made by the seeker; it is a major decision point.
With the proliferation of information and time restraints, one must make the
decision of what material to examine for the needed information. This is of
particular concern when using a global information database, such as the
Internet.
In electronic searches involving the Internet, one would never have
the time to examine all the available information. The determination of
relevance is the basis for extraction. Depending on the success or
satisfaction with the results of the search, reformulations of the search
process occur. These reformulations require an assessment of the
information seeking process and the results of the process in relation to the
information seeking task. Whether or not to terminate the search depends
on factors such as the determination of success, expectations, time
constraints, standards, and motivation. In the electronic searching
environment, terminating a search appears to be more difficult because of
the proliferation of information and the physical ease of accessibility.
Electronic Information Systems (E1S1
In 1945, Vannevar Bush, the Director of the Office of Scientific
Research and Development, wrote an article that verbalized a dream that
people have striven for ever since: the dream of making available to all
people, through the use of machines, the collective knowledge of mankind
13


(Bush, 1945). The Internet, including its developing applications such as
Gopher, Archie, WAIS, and most recently, the World Wide Web (WWW or W3
or the Web), Netscape, Mosaic, and MacWeb, with their hypermedia
capabilities, is the backbone that is making this dream a reality. It is often
referred to as an electronic highway to information. Simply put, it is an
international network of networks. Each network is administered,
maintained, and paid for by individual educational, research, and other
organizations (Eddings, 1994). Common terminology to describe
information retrieval on the Internet include terms such as "cruising,"
"surfing," and "navigating" the net.
Evolution of the Internet can be traced back to the time of the Russian
Sputnik program in the 1950's and is linked to the Advanced Research
Projects Agency (ARPA) (Engst, 1993). More recently, in the 1980s, the
connection of supercomputer sites across the United States by the National
Science Foundation (NSF) spurred the current NSFNET. The high-speed
networks that connect the NSF supercomputers now form the backbone of
the Internet. However, the user profile has changed from scientist,
researcher, and engineer to include well over forty million users a day
encompassing all age groups (K-12 and up) and all different walks of life
including scientists, government officials, businessmen, researchers,
students, and homemakers.
The current resources on the Internet are unimaginable and growing
daily. One person could never systematically explore everything that is out
there (Eddings, 1994). In 1993 it was estimated that there were 700,000
14


hosts world wide (Brett, 1993). In November, 1994 there were 3,864,000
hosts on the Internet (Treese, 1994) with a total of 81 countries participating
(Treese, 1994). This makes the Internet accessible to over 35 million users
(Ruter, 1995). Currently in the United States, approximately sixty
newspapers and over 70 peer-review scholarly journals publish
electronically on the Internet (Treese, 1994). The K-12 Hotlist numbers 137
elementary and high schools as connected to the Internet (Treese, 1994).
This list does not include at least a dozen schools known to be connected in
a Colorado metropolitan school district. Commercial interest has
dramatically increased in the past two years. The number of business
listings in the commercial Sites Index is 861 with 117 listings added during
the first week of November, 1994. Boston radio station WZLX aired the first
known commercial with a URL (Internet hypertext link) on October 4, 1994
(Treese, 1994). Increases in interest are also demonstrated by attendance
figures at the Internet World Conference. In January 1992, 272 people
attended the Conference; in December 1994, over 10,000 attendees were
recorded (Treese, 1994). Due to unprecedented growth of Internet, all these
figures are outdated by the time they are listed for review.
Information flows through four main rivers on the Internet: electronic
mail (E-mail), USENET news, TELNET, and File Transfer Protocol (FTP)
(Engst, 1993). Each river taps a different source of information. A brief
description of the different information sources and current tools available to
search these sources follows. Students involved in this study were exposed
to and given the opportunity to experience these sources and tools. Since
15


new interfaces and search engines are developed and/or discovered on a
daily basis, this discussion is not inclusive of all tools available. To date, for
example, a list of approximately forty search engines has been compiled for
students' use and a new interface called MacWeb has been added to the list
with Mosaic and Netscape. The following section demonstrates the range of
new information the study's subjects experienced during the exposure
period and aid the reader in understanding future discussions.
Electronic Mail
E-mail is the most prevalent river. It carries most of the directed
information, information that is directed toward a specific recipient. In
addition to recipient specific information, E-mail carries undirected
information to groups of individuals through mailing lists or LISTSERVs.
Through undirected channels all readers can post questions, answer
questions and comment on the posting. The mail is available to all
subscribers to the list. Auto-reply systems that send out information
automatically when requested through E-mail are also available.
Approximately 159 countries are reachable by E-mail (Treese, 1994).
USENET News
USENET is undirected information. USENET is a forum of news
groups of varied interest from neuroscience and education to comic books
and games. The electronic community uses these news groups as a virtual
forum (Eddings, 1994). They are organized into hierarchies by general topic
16


and within each topic are subtopics. USENET News best fits the river
analogy because of the speed with which the information flows. There are
an estimated 180,000 USENET sites with an average of 172 megabytes of
news flowing each day. Estimated world wide disk space used by USENET
news is 61 terabytes (Treese, 1994). Due to the quantity of information, most
of the messages are removed within a week or so. Frequently Asked
Questions (FAQ) are, however, often stored for retrieval for newcomers.
TELNET
TELNET is a protocol that allows users to log on to another computer
on the Internet in an interactive, command-line mode (Fraase, 1993). This
enables users to have access to the programs and information at that
machine and displays the information on the terminal station. But, the
information remains on the host system. Software applications, which make
accessibility to TELNET services via other methods possible, have reduced
the use of TELNET. However, some Internet services still require a TELNET
connection. The major drawback to the TELNET protocol is that each host
often has a unique organization and interface used for searching (Goodrich,
1994). This makes navigation difficult.
File Transfer Protocol (FTP)
FTP allows users to transfer files between their computer and a
remote machine quickly and easily. It is a data-transfer protocol that is
oblivious to the different kinds of computers and their operating system as
17


well as how they are connected (Fraase, 1993). It does not matter if the
machine is in the next room or halfway across the world. FTP is not set up to
let users view a document at the host site. The true relevance of a document
is not determinable until it is downloaded to the terminal site. This can be a
drawback if the files being transferred contain graphics, video, or graphic
images that take a relatively lengthy time to transmit. FTP does allow for the
accomplishment of work at the terminal site that could otherwise not be
done. For example, if a user were scripting a program in C, they would not
need a compiler on their local station. They would simply script at the
terminal site and then FTP the script to an accessible machine with a C-
compiler. Data files would then be stored and compiled at the host machine
and only results would need to be FTP'd back to the terminal station.
Information sources available through anonymous FTP are in all formats:
graphic images, sound, video, text, graphics, spreadsheets, and databases.
Along with the four main rivers of information there are a number of
special or user-level services on the Internet that encompass different
aspects of the standard services. These user-level services have been
developed in an effort to make sense out of the maze of information that is
available but very difficult to navigate through. Finding a single file on 4
million machines is a monumental task. Archie, Gopher, Wide-Area
Information Servers (WAIS), World Wide Web (WWW or W3), Mosaic,
Netscape, and MacWeb are all applications that make the information
seeking task a little more manageable.
18


Browsers and Search Engine Applications
Archie. This robot-like program was developed by the "Archie group"
at McGill University in Montreal (Deutsch, 1992). To date there are over 30
Archie servers around the world, with at least 12 of them located in the
United States. An Archie server is a computer connected to the Internet that
runs Archie software. About once a month the server logs on to every file
server it can access and automatically asks for information on all files at that
site. In this way it updates its own database, drops files that are no longer at
that FTP site and adds new files. Files are indexed based on keywords or
strings. All Archie servers do not have access to the same databases.
Some are regional; some are confined to specific countries; and some are
global. Archie will locate files but will not FTP them back to you. You must
exit Archie and initiate an FTP session.
After the development of Archie, it was not long before the comic book
friends Jughead and Veronica joined Archie to further improve the retrieval
process. Inspired by Archie, Veronica is protocol that queries large numbers
of gopher sites for the name of all their categories and files. The server then
provides a means of searching the list for the items that match the searcher's
criteria. Veronica relies on lists generated once a week. Jughead, on the
other hand, provides up-to-the-minute searching capabilities throughout a
single site's gopher hierarchy (Goodrich, 1994).
Gopher. This protocol was develop at the University of Minnesota
and is very similar to FTP. Gopher, however, has a much more user-friendly
menu based interface, provides access to far more types of information
19


resources, and provides access to online phone books, library catalogs, text
of files and databases of information stored in WAIS, USENET news, and
Archie (Engst, 1993). If you do not know exactly what you are looking for
then Gopher will enable you to sort through a maze of different kinds of
information resources such as pictures, sound, video and computer files as
well as text (Fraase, 1993). It is best used as a browsing tool. Its ease of
use, creation, and maintenance along with its understandable presentation
of information have contributed to the enormous growth of the Internet in the
early 1990s. The 1993 Gopher traffic growth rate of 1076% slowed in 1994
to 197%. In April of 1994 the number of known gopher servers was 6,958
(Treese, 1994).
WAIS. Wide Area Information Servers was develop by four large
companies, Apple Computer, Dow Jones & Co., Thinking Machines
Corporation, and KPMG Peat Marwick, to address the problem of access to
information in the business world (Engst, 1993). WAIS is a distributed
system which means that it is spread out among many computers. It indexes
and provides full-text of collections as well as sophisticated querying
techniques. Each WAIS database is a specialized library. The objective
behind WAIS is to help solve the problem of how to communicate in complex
database query languages. It uses a sophisticated natural language input
system. Unlike some of the other end-user services, it actually looks into the
documents for the terms you are searching, not just in the file title and
description (Eddings, 1994). It applies weights to the selection of terms used
in the searching activity, makes relevance judgments, and then ranks the
20


files. Relevance feedback is also part of the program. If you find several
items that suit your needs you can tell the computer to go out and search for
more items that are similar in nature to those selected.
World Wide Web (WWW or W3). The World Wide Web (Web) is an
online hypertext system that encompasses the entire Internet. Its goal is to
combine all the diverse resources available on the Internet. It is the most
recent, most unstructured, and possibly the most ambitious of what some
have classified as special services or applications available on the Internet.
Its foundation is in hypermedia. The Web began as an academic endeavor
in CERN, a high-energy physics research center in Switzerland. It is able to
negotiate formats without the problems of incompatibility and therefore can
accept any type of data, even multimedia. Its server maintains pointers to
data that is spread out over the entire Internet. Anything that is available via
Archie, Gopher, WAIS, FTP and TELNET is also available via the Web. The
unprecedented growth of the Internet owes a portion of that growth to the
Web. Growth in Web traffic slowed to 1713% in 1994 from 443,931% in
1993. The Web has become so huge and spans so many services and
pieces of information, that it is mind-boggling to use. Currently, applications
and search engines appear monthly that help to navigate this immense sea
of information.
Mosaic. Mosaic is an Internet information browser for use in the Web.
It is currently the most popular navigational tool for use on the Internet. This
application, developed by the National Center for Supercomputing
Application (NCSA), gives access to all the other resources on the Internet
21


through a browsing tool that creates a hypermedia-based menu on your
computer screen. It is estimated that an average of 1600 copies of this
software are downloaded every day (Treese, 1994). The basis for
connections to documents relies on the Uniform Resource Locator (URL)
mechanism, a standard designed to envelop the existing universe of online
information within a single nomenclature (Andreessen & Bina, 1994). URLs
point to documents residing almost anywhere on the Internet through some
networked based access mechanism: Gophers, white pages, newsgroups,
search engines, text documents, graphics, sound, WAIS queries, video, etc.
Netscape and MacWeb are the newest multimedia interfaces. Individual
preferences appear to dictate interface selection. However, action research
indicates that there may be a speed difference between interfaces with the
newest (MacWeb) being the fastest (M. Plitnik, personal communication,
January 27, 1995).
Search Engines. Finding information on the Web can mean following
endless links without success. The information is available but finding it can
be excessively time consuming and frustrating. WebCrawler, World Wide
Web Worm, and Galaxy are three of many search engines that help an
individual navigate the Web. A list of approximately forty engines has been
developed by students in the school where this study was conducted. These
engines search through all the pages of their known links to unexplored
Web pages. The links created by these engines are often electronically
generated. Most are based on a Boolean structure and weighted. The
algorithms used by programmers are rarely discernible and are not uniform.
22


Even the number of hits, number of retrieval documents, varies. Some are
limited and some allow the searcher to choose.
Information retrieval resources on the Internet are in a state of
continual change and appear limitless. It is important to understand
information retrieval on the Internet when discussing the current research in
the field. This network of networks is different in many important aspects
from traditional limited electronic database searches. These differences
could affect the application of underlying cognitive theoretical basis, retrieval
process, and evaluative processes of past studies to future studies of global
database information retrieval. By understanding the information sources
and searching tools available on the Internet, one can better discern if the
current information retrieval research is appropriate and applicable to
Internet information retrieval.
Internet Study
Research Question
In order to educate today's students in the skills of information literacy
required to survive in the 21st Century, research into the resources and
methods of electronic information retrieval is needed. Given the current
power of information and its increasing importance, developing appropriate
attitudes and beliefs in future leaders is essential. Determining the attitudes,
subjective norms and intentions of high school students toward global
electronic information retrieval, specifically the Internet, is a first step in the
23


determination of the best methods to use in the design of an information
literacy curriculum and in building resource-based learning environments.
Attitudes and subjective norms are a function of behavioral belief and
normative beliefs and the direct determinants of intentions. Behavior beliefs
concern what a person thinks about the performance of a behavior.
Normative beliefs deal with what a person thinks specific individuals or
groups think about his performance of a behavior. Since attitudes and
beliefs influence intentions and intentions drive behavior, knowledge of
beliefs and attitudes will help educators understand the culture with which
they are dealing. The questions investigated in this study lead to new
research directions that are necessary to educate our children to succeed in
the Information Age.
Literature Review
No research into the area of information retrieval from a global
database such as the Internet was found. The review of literature for this
study, therefore, consisted of an analysis of the current status of the
emerging theory of information retrieval. Only studies involving electronic
information retrieval were reviewed. This included several that compared
electronic and print information retrieval.
The approach taken was to review the literature in terms of the
theoretical frameworks, affective aspects of the search process, and
evaluation techniques. The emphasis in traditional research has focused on
system performance characteristics and effective search practices. The shift
24


to use of electronic searching by end users dictates a shift in research
emphasis to the end user and the affective aspects of the information
retrieval process research. Several excellent reviews that stress the
traditional view of information retrieval processes are already available .
Methodology
Students at an inner-city school in a large metropolitan district were
the subjects of this study. The school population is predominantly Hispanic
(79%). The minority population is 86%. The subjects ranged in grade level
from sophomores to seniors and in age from 15 to 20 years. The majority of
the students were self-selected in so far as they chose courses in which
information retrieval via the Internet was introduced. Some of the students
were placed in these classes by counselors due to schedule conflicts with
their original choices. All students were given the option of switching
classes. Students in the following classes were involved: Business and
Office Management, Telecommunications, Adventures in SuperComputing,
Advanced Keyboarding, and Business Management.
Students who were enrolled were give one week exposure to the
Internet before completing a questionnaire designed according to the
Fishbein and Ajzen Model of Attitude-Behavior Consistency. The model was
chosen because of its high mean correlation between the relationships of
attitudes, subjective norms and intentions (Walster, 1994). This model has
become a standard against which new models are compared. Further
25


discussion of this model appears in Chapter 4. Analysis of the data
collected follows the model design.
Study Organization
In the Chapter 1 the following topics were discussed:
current trends in our society that dictate a change in the manner in
which we educate our children now and in the future
information retrieval as a process
electronic advances and interfaces that are affecting a change in the
reason for and manner of information retrieval.
Chapter 2 contains a review of the current literature in the field of information
retrieval with the intent of finding a basis for research that will aid in the
development of a resource-based learning environment. Most all of the
research that has been done concerns traditional approaches and statistical
usage information retrieval methods. Initial research into end-user
information retrieval processes has begun and emphasis is placed on these
studies in the review. Chapter 3 details the methodology used in this study
and Chapter 4 the data and findings. A discussion of the results,
implications for implementation of an information literate learning
environment, and suggested future research occurs in Chapter 5.
26


CHAPTER 2
LITERATURE REVIEW
The Emerging Theory of Information Retrieval
Due to technological advances and changing societal needs, the
environment for information retrieval is in a state of change and the theory
that supports this process is in it infancy (Marchionini, 1989b). Research has
been taking place for decades on specific aspects of this volatile and
complex process. Yet, Saracevic, Kantor, Chamis, and Trivison (1988) state
that it is nothing short of amazing how relatively little knowledge and
understanding of this process is available. The development and
understanding of the information retrieval process takes on added
importance and urgency in today's changing environment. Three current
bodies of literature support and guide a theory of information retrieval
(Marchionini, 1989b): cognitive science, online searching, and electronic
information systems (EIS).
Cognitive Science
Cognitive processes initiate the information retrieval process when
the knowledge base for a task is activated and found to need instantiation or
modification (Marchionini, 1989b). Three major cognitive concepts that are
found in information retrieval literature are: mental models, the
27


novice/expert paradigm, and cognitive load. Schema theory is the
predominant theory that underlies all three cognitive concepts. It is an
outcome of constructivist theories concerning how we assimilate information.
We do not remember in the same manner as a tape recorder or video
camera (Sweller & Chandler, 1991); we transform and then construct
representations of the information.
Mental Models. A prevalent theory of cognitive science holds that
complex experiences are understood by an individual through the mental
models they construct. A mental model is a cognitive mechanism used to
form dynamic conceptual representations of their environment. Perkins
(1992) defines a mental image (model) as a holistic highly integrated kind of
knowledge. "It is any unified overarching mental representation that helps
us work with a topic or a subject" (Perkins, 1992, p. 80). The mental model
may be described as the internal side of understanding. One perceives the
world through mental models and when one needs to make a decision, an
individual's mental model allows prediction and evaluation of the
consequences of a particular actions (Johnson-Laird, 1980). Each model is
in a constant state of change and construction as new information and
experiences occur that affect the model.
If the mental model is correct, it will help in understanding
performance. But, if the model is incorrect, inaccurate, incomplete, or
inappropriate, misconceptions and errors may occur (Ajzen & Fishbein,
1980). Mental models of online retrieval systems are often ill-formed,
incomplete and/or incorrect (Liebscher & Marchionini, 1988). One common
28


misconception, for example, is that an online retrieval system compares
every word in the database. This is not true if the database is indexed. Only
a small subset of terms will be compared with the search query.
Normans (1983) concept of conceptual models is also related to
mental models and important to the information seeking process. The
distinction is made between the internal mental model constructed by the
individual and the external conceptual model provided to the individual. In
the information retrieval process this conceptual model may be given to the
user by a designer, trainer, or experimenter as an explanation of the
electronic system (Borgman, 1986). The importance of a conceptual model
in the development process of the mental model has an effect on the type of
training one would design. The information seekers' mental model drives
the information-seeking process. It affects which search system is chosen,
determines the vocabulary used in the search and determines the
expectations for results (Marchionini, 1989a).
Expert/Novice. Schema theory is the predominant theory used to
explain the different problem solving approaches taken by novices and
experts (Sweller & Chandler, 1991). One creates a schema by learning to
classify problems and problem states. Experts in a domain begin the
problem-solving process by building representations or schema of the
problem from which they can draw inferences. They represent the problem
according to fundamental principles which structure the domain knowledge.
Novices tend to approach a problem from a more superficial level. Most of
the investigation surrounding the expertise/novice debate thus far is
29


centered on a particular knowledge domain. Information-seeking is a
fundamental cognitive activity and affects, to different degrees of expertise,
all humans. It involves the general cognitive abilities of problem solving,
learning, planning, and decision making (Marchionini, Dwiggins, Katz, & Lin,
1993). The development of expertise in information retrieval has taken on a
new dimension in the information-intense world of today. Yet, the transition
from novice to expert in the area of information seeking is open to
speculation because there is little hard evidence to support the process
(Solomon, 1992).
Cognitive Load. Cognitive load is concerned with the manner in
which human cognitive resources are distributed. The assumption is that the
human cognitive structure can be characterized by a relatively poor working
memory and a relatively limitless long-term memory capable of storing large
amounts of schemata (Bobis, Sweller, & Cooper, 1993). A schema being a
"generalized cognitive construct that allows us to classify categories of tasks
according to the mental actions appropriate to the task (Purnell et al, 1991,
p. 443). Working memory is affected by cognitive load. Norman (1983)
discovered that subjects chose to perform unnecessary actions in order to
lighten the cognitive load. We are willing to use mental models that are
deficient rather than invest the time and cognitive effort into altering the
model as long as the predictive results are viewed as satisfactory. "Just as
water and electricity seek paths of least resistance, so humans seek the path
of least cognitive load" (Marchionini, 1989b, p. 56).
30


The information retrieval literature discusses cognitive load from two
different viewpoints. One is referred to as information overload, or the
amount of information available to the user on a specific topic. Herbert
Simon's satisficing principle (as cited in Rudd & Rudd, 1986) states that
individuals will only acquire enough information to satisfy their needs; they
will not maximize the amount of information acquired.
The second viewpoint deals with the type of searching process
involved. In discussing the difference between hierarchical and analytical
searching techniques, Marchionini emphasizes that there is a higher
cognitive load on the analytical searching strategies due to the necessity to
learn the Boolean techniques. Whereas, when using the hierarchical
technique of browsing, the user is depending on recognition rather than
recall. Marchionini considers this a lower cognitive load strategy. Conklin
(1987), on the other hand, considers cognitive overload as a distinct
disadvantage with a hypertext, hypermedia environment. Conklin's
reasoning is based on the additional effort and concentration necessary to
keep track of several links or paths at the same time; and the number of
choices that a user must make when deciding which links to follow.
A third dimension not mentioned is the evaluation process required
when browsing in a global database such as the Internet. Hundreds or
thousands of documents can be retrieved from the use of one search
engine. Other search engines would have produced documents that may or
may not be included. The process of determining which documents to
download for a relevancy evaluation and which search engine or tool to use
31


is cause for cognitive overload that does not relate to the learning of
Boolean techniques. Is it really more of a cognitive load to develop a
specific Boolean search that will yield a few exact or partially matched
documents than to determine paths and evaluate the relevance of the
massive information bank that could develop by using a browsing
technique?
Online Searching
Fenichel's (1980) review of the literature of online searching
behaviors demonstrates the extent of early investigations of this process.
Since the early 1980's much additional work has been done (Large,
Beheshti, Breuleux, & Renaud, 1994; Marchionini, et al., 1993; Marchionini &
Teague, 1987; Millsap & Ferl, 1993; Saracevic, Kantor, Chamis, &Trivison,
1988; Smithson, 1994; Wildemuth, Jacob, & Fullington, 1991). These and
other studies can be divided by the subjects studied: traditionally,
intermediaries (professional searchers) and recently, end users. The
variation in strategies, tactics, and usage of the systems, even by
intermediaries, demonstrates the complexity of the problem of developing a
comprehensive online searching methodology. Research encountered to
date has centered on CD-ROM and electronic databases, both
bibliographical and full-text in nature. It has not been related to the use of
global information sources. Only recently has the research concentrated on
the end-user's role in the search process. The complexity of the search
process is currently increasing due to the increase of end-user access and
32


the new global hypermedia electronic information sources with all their
nodes and components.
The two main models of information retrieval are based on analytical
and hierarchical classification systems. Other models currently under
investigation are the Vector, Galois Lattice, rough approximation,
multidimensional, and fuzzy models. A brief description of these models and
their principle information retrieval techniques will reveal the basis of today's
searching methodology.
Analytical Model. This model has traditionally been used for
bibliographic and commercial databases and accessed by professional
intermediaries for factual information needs (Srinivasan, 1991). It requires
the use of well developed query formulations and search strategies
(Marchionini & Liebscher, 1991). Its basis lies in the representation of
information by index terms and, therefore, is most useful when you know
exactly what you are looking for and understand the indexing language
(Godin, Missaoui, & April, 1993). Indexing being the process by which
information is stored for retrieval based on a vocabulary of terms determined
by the indexer. Agreement between indexers on terms and even the same
indexer on different tasks is not constant. Searchers have extensive leeway
in choosing search terms and the terms chosen may not retrieve the
information desired by the end user. Indexing uncertainty and search
uncertainty are the main causes of information retrieval problems (Chen &
Dhar, 1990).
33


Analytical searches require the problem to be broken into key
component parts and are most successfully completed by expert
intermediaries who know Boolean logic, the indexing language, and the
contents of the database. Proper use of the AND, OR, and NOT connectors
is a key to Boolean querying which is most successful with short records
(Morton, 1993).
Hierarchical Model. This model is the principle method for organizing
and finding files in a computer system (Godin et al., 1993). Before
discussing the hierarchical model, two concepts must be discussed:
hypertext and navigating. Hypertext refers to text in electronic form that
provides the user access to text in a nonlinear, nonsequential form.
Hypertext allows browsing through passages of text, expanding items of
interest. As items are expanded, links to the new information are created.
The needs of the user determines the access. Hypertext depends on the
establishment of links between chunks of information before the user
accesses the database, therefore, the user is constrained to the paths
defined by others in navigating the data (Watters & Shepherd, 1994).
Hypermedia functions in the same manner except that it expands the
information retrieval process into other media besides text. Graphics, video,
and sound media are all part of the databases that can be linked.
Navigation according to Newby is human behavior to make sense of
information space. And, "information space is defined as the set of concepts
and relations among them possessed by an information system" (Newby,
1992, p. 20). Navigation is the movement through this information space. In
34


navigating a hierarchical information system the user moves from one class
to another using menus or some direct representation of the hierarchy
(Godin et al., 1993).
The hierarchical model contains information presented in a network of
nodes and connections and has a rich representation of the knowledge in
the subject domain. The hierarchical arrangement of information into folders
and documents that can be browsed is a tree-like structure. This tree-like
structure is the major drawback since there is only one route from root to
item. Branching incorrectly at any point could cause a failed search (Godin,
et al., 1993). With the current advances in hypertext and hypermedia
technologies, where the information is connected by a series of links that
connect related nodes or pieces of information, this problem is being
eliminated.
The major strategy in the hierarchical model is browsing, which is an
alternative form of data retrieval especially suited to casual users and
exploration of new domains (Godin et al., 1993). When browsing, the search
aim is not specific in respect to exact topic, perspective, etc.., and specific
outcomes are not predetermined. The user's query may be vague and ill
formed (Millsap & Ferl, 1993). This technique is well suited for the
exploration of a knowledge domain rather than searching for a specific fact
or piece of data. The immediate feedback provided by the users' browsing
decisions determines the path of the search.
Other Models. Current model development research stresses the
integration of both Boolean and browsing querying techniques. The Galois
35


Lattice model combines the browsing and direct retrieval strategies into one
model by defining information in terms of class and is based on a network
classification structure (Godin et al., 1993). Information retrieval occurs by
direct term specification in which you start with the most general class and
then either refine or enlarge that class. In a recent study (Godin et al., 1993)
based on search time, recall, and precision, no significant difference was
found between Boolean querying and Galois lattice retrieval for subject
searching. There was a significant difference as to recall between the
hierarchical classification retrieval which was lower than the Boolean or the
Galois lattice structures.
The rough approximations model was introduced into domains such
as medical diagnosis by Pawlak in 1982 (Srinivasan, 1991). The important
contribution of this model is the organizing of vocabulary into approximate
classes. Classes are based on semantically equivalent terms. It allows
exact matches based on Boolean logic, allows partial matches, and
approximate matches with regards to the search vocabulary. This model is
very similar to the vector model which provides a framework for retrieving
documents that partially match the query. Boolean logic in the vector model
is limited to the AND connector (Srinivasan, 1991). The fuzzy model also
allows for partial matches between documents and queries by assigning
fuzzy weights to indicate the strength of the associations between queries
and index terms. All of these models show a trend towards a more flexible,
user-centered retrieval system.
36


Electronic Information Systems (E1S1
With new technologies and electronic information sources, most
importantly the Internet, a new information retrieval paradigm has
developed. This paradigm is a user-centered (navigational) rather than
data-centered (bibliographic). In the traditional bibliographic model, the
emphasis is on data collection, classification, and devising strategies for the
retrieval of the information (Kuhlthau, 1991). The system is based on
certainty and order. Newby (1992) describes this traditional model as the
relevance model. In a relevance based system, the goal is the
determination of all relevant documents within a certain database and the
withholding of all non-relevant documents. The relevance issue is
independent on user context, binary (yes/no), non-changing, and based on
the assumption that the language of the user is representative of the
information need. This model is not indicative of human interaction. Human
communication is facilitated by a mental model that each individual has of
the other. These models change continually as interaction occurs. The
more interaction the better the communication becomes because the
ongoing relationship allows people to know each other better (Dervin &
Nilan, 1986). When a human interacts with an information space, a model of
that space is also essential. Presently, it is the job of the human to adjust his
cognitive space to fit the model that a system has of the human.
With this new user-centered, navigational based model, the emphasis
is toward an interaction that models more closely human communication. In
Newby's terms (1992), the paradigm shift is toward a goal of navigation and
37


away from relevance. It is not based on certainty and order because the
end-user's information search process (ISP) usually begins with uncertainty,
anxiety and a lack of precision (Kuhlthau, 1991). The information search
process is a constructive activity in which the user interacts with the
information to find meaning in an attempt to enhance the user's current state
of knowledge on the topic or problem. The person is actively involved in
"sense-making" (Dervin as cited in Kuhlthau, 1991). Sense making is
personal and is constructed based on the individuals personal mental model
of the topic. Relevance of information retrieved is based on the users'
evaluation of the its usefulness in the process of solving the problem.
Information access is an important aspect in the developing
information revolution (Brody, 1993). The shift toward end-user based
systems is necessitated by the proliferation of information technology and
access to this technology. End users are people who have determined that
they must have additional information in order to fulfill a need; who have the
questions; who have to make sense of and use the information; and who are
the ones who are performing the search operations. This is in contrast to
traditional searchers who are professionals in the corporate, educational,
and public sector who find information for another person. These are most
often specific limited searches that yield a database of relevant documents
as interpreted by the intermediary and the informational system used. With
technological advances and information sources available, end users are
the predominant information retrieval personnel today. Students learning to
learn from information, corporate personnel accessing the home pages of
38


various corporations on the Internet for market information and product
sources, citizens accessing governmental databases for information
concerning social services or available government services, and
government purchasing personnel access minority business databases,
these are all end users navigating the vast global information databases
available.
The three bodies of literature that have supported past information
retrieval research are important to current and future research. Cognitive
science provides a theoretical basis in schema, mental models,
expert/novice, and cognitive overload theories. Secondly, online searching
literature provides model development for the search process. Whether
model development by specialists can keep up with the dynamic changes
on WWW is questionable. It appears that search procedures are being
developed by non-information retrieval specialists in order to aid end users
who are getting lost in cyberspace. Finally, electronic information systems
literature supports the rationale for the paradigm shift currently being
experienced in the field.
Research Studies in Information Retrieval
Overview
Current literature reviews on the topic of information retrieval are
discussed but the emphasis is on research of the late 1980's and early
1990's with particular emphasis on end-user research. This emphasis is
39


dictated by the dynamic changes that are currently occurring in the field at
this time. Therefore, presentation of the research is divided by the paradigm
to which it subscribes. Print medium studies were not included because it is
no longer a question of whether electronic information retrieval is a valid
medium. We are currently using electronic information sources extensively
and, will continue to do so at an increasing rate. Studies that considered the
comparison of print and electronic sources of information were included for
their contribution to electronic information retrieval knowledge base.
The presentation of literature begins with a brief discussion of seminal
works in the field covering the 1970's and early 1980's. The majority of this
research deals with the strategies, problems, and system needs of
intermediaries. Professional intermediaries are specialists experienced in
retrieving information from a myriad of sources including electronic search
tools. Characteristics include knowledge of appropriate search strategies
and different information sources, skill in defining the search problem, and
utilization of resources (Marchionini et al., 1993). In addition to being useful
to designers of informational systems, study of expertise in information
retrieval is felt to aid in preparing students for the retrieval process. Selected
studies that contributed to the body of intermediary searching process from
the late 1980's and early 1990's are included. Studies that involve end-user
research constitute the bulk of the review. They fall into two classes: those
that research the process used and those that research the end-user
characteristics. Both approaches seek to understand the search process
and are by no means mutually exclusive.
40


There is no information retrieval research dealing with a dynamic and
unlimited database such as the Internet. There is, however, a beginning; a
realization of this unlimited database in telecommunications research. The
review concludes with a brief summary of this research and the contribution
of the current research to the beginning of an understanding of information
retrieval in such a database and its integral part in establishing a resource-
based learning environment. A brief description of the assessment tools
used to evaluate the success of a search in traditional terms precedes the
discussion in an attempt to prevent any misunderstanding of terms.
Information Retrieval Evaluation
The nature of information retrieval makes its evaluation both difficult
and important. Its importance traditionally lies in three major areas
(Smithson, 1994). Searching in textual databases on the basis of content is
non-conclusive. Some searchers are successful and users are satisfied with
the number of documents and some are deemed successful with relatively
few documents. A benchmark of success is necessary. Secondly,
evaluation is essential in the determination of implementation of changes to
a current system in the development of a new system. Finally, the cost factor
is important not only to the developers of a system but to the user of that
system. Justification of the expense of setting up and operating an online
database as well as justification of users expense in using such a system is
necessary.
41


Relevance. There is widespread consensus that this is the most
important concept in information retrieval theory and practice (Smithson,
1994). Obtaining consensus as to the meaning of this concept is much more
difficult. Traditionally, relevance was determined by the researchers.
Typically a small database was used in the research or subset of larger
databases and all documents "relevant" to the assigned search task were
predetermined by the researcher. In this manner, the researcher could
easily determine the number of "relevant" documents retrieved by the
searcher. This traditional approach has the semblance of scientific
respectability. Yet, it is misleading because the fictional information needs
and artificial relevance judgments are likely to be different from real
situations (Smithson, 1994).
With the increasing use of the user-center paradigm to research, the
matter of relevance is no longer a static, preconcluded factor. The individual
user determines the relevance of each document retrieved at each particular
stage of the information search. Documents that were deemed "relevant"
during the initial stages of the search (or learning process) could be deemed
totally irrelevant by the concluding stages of the process. Relevance in this
paradigm is not a static, didactic, or hierarchical predetermined concept but
an evolving and subjective concept developed by the end user. Saracevic
et al. (1988) breaks down relevance into three categories: relevant, partially
relevant, and non-relevant, eac.' nearly differentiated. In these studies
searchers were intermediaries and relevance evaluators were end users of
the documents. The variety of meanings clearly demonstrates that
42


determination of relevance must be clearly defined and explicitly exposed in
the research.
Effectiveness Ratios. Traditionally, retrieval evaluations have been
based on effectiveness ratios, predominantly precision and recall. These
are measures based on the relevance judgment, either predetermined by
the researcher or determined by the users (Fenichel, 1981).
Precision = No. of relevant items retrieved bv the search
Total No. of references retrieved
Recal = No. of relevant items retrieved bv the search
Total no. of relevant items in the database
In other words, precision can be considered the measure of how well a
system retrieves only the documents deemed relevant or the probability that
a retrieved document will be relevant. Recall is the probability that a
relevant document will be retrieved. Another ratio, contribution, is
considered by Smithson (1994) as an important effectiveness ratio. This is
the proportion of retrieved documents judged as relevant and of which the
user has no previous knowledge.
These quantitative measures are the primary evaluators used during
the 1970s. However, even in current research, measures of search success
are often quantitative rather than qualitative. The goals of information
retrieval have been to locate specific facts or documents not the use of
information retrieval for metacognition.
43


Intermediary Research
Online searching of the 1970's commercial and governmental
bibliographic databases was primarily the domain of trained intermediaries.
Study of expertise in information retrieval is useful to designers of
information retrieval systems as well as the development of information
retrieval theory. The role of the end user is to supply information on needs
and interact with the intermediary. Fenichel (1980, 1981) presents a
comprehensive review of studies from the 1970's. These studies are mainly
exploratory in nature. Variable factors considered are: descriptors
searched, commands used, connect time, retrieved references, recall,
precision, and unit cost. Statistics on the average search based on these
variables is presented in tabular form. Some of the conclusions from
Fenichel's analysis of the research of interest in end-user research are:
Considerable variance is observed among experienced searchers'
use of strategies and vocabulary.
Substantial number of experienced searchers performed simple
searches.
Major problems of experienced searchers and inexperienced
searchers were with search strategies not the mechanics of the
system or language.
Online searching is sensitive to environmental factors such as nature
of the user groups, management policies, and charging procedures.
In Fenichel's (1981) own study, an enormous individual variability in search
behavior of persons of the same experience was encountered. It appears
44


that affective and environmental factors influence search strategies of
professional searchers or intermediaries.
Saracevic and Kantor (1988) studied intermediaries under as real-life
situation as possible. This in depth, three part study considered five
variables: searchers, users, search, questions, and items retrieved.
Cognitive aspects, and human decisions and interactions involved were of
particular interest. Several practical implications resulted from the study.
The importance of the "question" was reaffirmed. Background information
and the intended use of the information are, however, as important as an
understanding of the question. A specific written question has much more to
it than is expressed in written words. Relevant characteristics associated
with successful searchers were found to be skills in word association and a
preference for abstract thinking. Demonstrated cycles in searching indicate
that it is important to view and review intermediate results in order that
adjustments to the search process can be made. High utility marks indicate
that extraneous material may not contribute to higher satisfaction by the
users. Results of different searchers showed a large disparity in the retrieval
process and documents retrieved by the experts. This disparity has
previously been demonstrated by research and indicates that it may be
helpful to complete multiple searchings of the same question using different
searchers.
User perception that online searches performed by intermediaries
were unsuccessful was a primary conclusion of a study of 22 cases
completed by Smithson (1994). A reasonable number of relevant
45


documents appeared to be retrieved and the searches deemed successful
by traditional evaluation methods: recall, precision, relevance, and
contribution However, few of these documents were found useful by the
user. The research schema did not allow for adjustments during the retrieval
process by the users. As the user learns, information needs change. Use of
the case study proved to be beneficial in supplying additional information
with respect to the dynamic nature of the users information needs and the
retrieval process in general.
Blair and Maron (1985) researched retrieval effectiveness in a full-text
retrieval system. Effectiveness was defined by the traditional measures of
precision, recall, and relevance. The information sought was specific
documents that related to a specific litigation by the users, lawyers. To
predict the exact words, word combinations, and phrases used by all
relevant documents and only those documents that are relevant proved
impossible. The study showed that full-text retrieval did not operate at
satisfactory levels. These results counter those obtained by Swanson and
Salton (as cited in Blair & Maron, 1985). The difference was the size of the
database. Swanson and Salton used small experimental databases of 750
items or less. The database used by Blair and Maron contained almost
40,000 documents. Retrieval strategies that are effective in small databases
are not necessarily effective in large databases because of output overload
(Blair & Maron, 1985).
Marchionini, Dwiggins, Katz, and Lin (1993) conducted a series of
four full-text searches to determine the effects of domain expertise and
46


search expertise on the information retrieval process. Three of the
databases used were CD-ROM and one was a hypertext. Experts in search
strategies, intermediaries, and experts in domain knowledge end-users,
were involved in the studies. Results of the studies indicated that domain
knowledge and information retrieval knowledge are not independent
characteristics and that expertise in a domain by itself is not enough to
complete a successful search. Domain experts were content-driven
whereas retrieval experts were problem-driven. Yet retrieval experts
browsed more than was expected while expressing concern over the cost of
the additional browsing time.
Results of intermediary studies emphasized the importance of
individual characteristics and knowledge of retrieval strategies as well as the
cyclical nature of the information retrieval process The opportunity to reflect
and alter the course of the search has an impact on the search outcomes.
End-User Research
The end user is the person or persons who actually uses the
information that is retrieved. Dervin and Nilan (1986) review post-1978
literature, essay and empirical research on information needs and uses. It is
in this time period that the literature began calling for a paradigm shift away
from the traditional perspectives to an alternate paradigm that defines
information as something constructed by people. In an attempt to answer
critics of the state of information retrieval research, this literature review is
from the perspective of conceptualizations that drive research. The literature
47


during the time between 1978 and 1985 calls for development of theories
and conceptual frameworks that have predictive value, and research that is
applicable to practice. Literature critical to past research now calls for an
understanding of information that includes what leads up to and follows the
intersection with electronic systems. Three alternative information needs
assessment approaches are presented from the literature as a result of the
call for a paradigm shift. End-user studies have concentrated on the
medium of online catalogs, CD-ROM technology, and searching of remote
databases. Mischo and Lee's (1987) review and analysis of end-user
research results showed that end-users' searches were not particularly
effective in the traditional sense. Intermediaries retrieved more relevant
documents and spent less time online. End users, although enthusiastic,
seem to have difficulty with search strategies and Boolean logic. They
retrieve less relevant citations and perform less efficiently.
Process Research. There appears to be two major, non-mutually
exclusive forces in end-user research: process research and user
characteristic research. Process research involves studies that deal with the
process, strategies, and techniques of electronic information retrieval in an
attempt to improve the design of information retrieval systems, training, and
instructional methods as well as the searching capabilities of end users.
"We have been studying methods of training search intermediaries on
commercial systems for years ... without learning much that can be applied
to training users of online catalogs" (Kuhlthau, 1988, p. 50).
48


Marchionini and associates have completed numerous studies
involving subjects in K-12 and post-secondary environments. Liebscher and
Marchionini (1988) used a full-text CD-ROM encyclopedia in research with
high school students trained in two different search strategies: analytical
and browse. Traditional forms of evaluation were used: recall, precision,
and analysis of information. Both training groups did equally well offering no
support for the hypothesis that an analytical mental model for search queries
will yield better results than the simpler browse model. In another study
Marchionini and Teague (1987), investigated whether elementary students
of different ages could effectively use selected electronic information
services. Students in the primary levels (second and third grades) were
equally proficient as those students in the upper levels (fourth, fifth, and sixth
graders). All students would use electronic bulletin boards and the online
encyclopedia. The upper level children did use more books to augment
their search and seemed more focused on the topic whereas the primary
levels focused more on the computer system than the topic.
Marchionini (1989a) demonstrated several constructs important to the
mental model development theory of information retrieval in a study of high
school students. Users who had poor information processing skills and
mental models of the information seeking process had a more difficult time
searching an electronic encyclopedia than students with well structured
mental models. Students with strong mental models had no problems
adjusting their models to adapt to the new system or performed successful
searches without making model adaptations. In a similar study with
49


elementary school children, Marchionini (1989b) examines information-
seeking strategies from a mental model perspective. Results demonstrated
that young novice users could successfully use a full-text encyclopedia with
minimal training. Search strategies used were highly interactive rather than
planned and subjects had difficulty formulating effective queries. This
difficulty may have been related to ill-defined mental models of the search
system. The browse strategy served these students well as they scanned
and selected to find the necessary information.
In contrast to Marchionini, Large et al. (1994) used a non-Boolean
search engine to study how children's use of print and CD-ROM versions of
an encyclopedia differed. Neither version proved faster and young users
were able to cope with the complexity of the electronic version. They were
able to distinguish between and efficiently use different retrieval strategies
and the CD-ROM technology with minimal training.
In an attempt to determine the behaviors associated with successful
online bibliographic searching, Fenichel (1981) studied both experienced
and non-experienced searchers as they completed research in a small
subset of the ERIC database. Within groups, however, large differences
were found, sometimes varying by factors of ten or more. Beginner
searchers performed well in comparison with experienced searchers and
the differences were not large. Evidence to support a claim that experience
makes a difference in this database was slight. The simplicity of the search
strategies was similar for experienced and non-experienced alike. In order
to further understand how researchers interactively develop and revise their
50


searches, Wildemuth, Jacob and Fullington (1991) designed a descriptive
study. Five searching strategies were devised by the study of 244 moves in
database searches. This study did indicate an effect for experience since
the number of moves decreased as experience of the searcher increased.
In an attempt to aid in the design of systems that adapt to changes in
users capabilites as their expertise changes over time, Solomon (1992)
observed a group of first graders from their introduction to online searching.
How can a system adapt and help learners who begin and develop different
levels of expertise. Despite the limiting features, in terms of generalization,
of the naturalistic methodology used in this study, it is a beginning in
understanding how electronic information searchers progress from novice to
various stages of expertise. Its importance lies in helping to understand the
transition and how an electronic system can help in promoting the transition
from novice to expert.
Chen and Dhar (1990) appear to differ with Marchioni in their view of
the cognitive load required to perform Boolean and Browse searches.
Expressing the sentiment that subject based searching (browsing) is more
difficult than key-word searching (Boolean), they researched
misconceptions of end users in subject based searching. From the
research they developed a classification schema and taxonomy of
misconceptions (errors or suboptimal procedures) and a semantic network
structure for representing knowledge. Rules for browsing and determining
search options that enable users to make the most out of whatever system
they are accessing also developed from the research. These rules appear
51


to be applicable to most all electronic information systems, perhaps even the
Internet. Further research on the applicability of these rules to electronic
global searches needs to be completed.
From an analysis of transaction logs of remote users of an Online
Public Access Catalog (OPAC) system and demographic questionnaires,
Millsap and Ferl (1993) determined that a large percent (40%) of the
searchers were query searches, searches that were brief and for known
items. The remaining searches were considered browsing searches;
searches in which you must define that which you do not know in order to
find it (Hildreth cited in Milsap & Ferl, 1993). Results indicated that users
definitely need more help in design elements that would lead the user
toward the discovery of sought after information.
Differentiating between model-based and procedural-based training,
Kuhlthau (1988) studied the outcomes of the two training approaches. An
interaction effect between task complexity and training method indicates that
the model-based training is not always superior to procedural-based
training. Both groups demonstrated that users had constructed models even
if not trained with them. No relationship between model articulation and
performance was discovered.
Watters, Shepherd, and Qiu (1994) studied users on a multiversion
information retrieval system. Frants, Shapiro, and Voiskunskii (1993)
defined such a system as one that has more than one algorithm for each
process and the ability to describe a mechanism for choosing the most
appropriate. This type of a system allows the user to choose the methods of
52


retrieval based on information needs. Results indicated that the users do
select the retrieval tool that best fits the situation even if they are not familiar
with them. The subjects' background was not a factor indicating the
adaptability of the user. Contrary to Borgman's (1986, 1989) results,
educational background was not a factor.
Individual Differences Research. Research into searcher
characteristics and the affective aspects of information retrieval process is
becoming prevalent in the literature. Individual differences are an important
aspect in the information retrieval process because they are responsible for
large differences in performance; we can not select those to do the job as
one does in the workplace; and training and design of systems can help
accommodate differences (Egan as cited in Borgman, 1989). Kuhlthau
(1989) through a series of five studies looks at the feelings, thoughts, and
actions of high school students, college students, and public library users as
they experience the search process. The analysis of the studies led to the
development of the Kuhlthau Model of the Search Process which describes
the feelings, thoughts, and actions of the subjects as they progress through
seven stages of an information retrieval process. The theoretical basis of
this model is Kelly's description of the process of construction, Taylor's
levels of information need, and the anomalous state of knowledge
developed by Belkin (Kuhlthau, 1988). Feelings such as uncertainty,
optimism, confusion, frustration, doubt, clarity, confidence, and relief all
appear in the model along with thoughts of ambiguity and specificity.
53


Fidel and Soergel (1983) review the literature from the perspective of
establishing a framework for the organization of independent variables
affecting online bibliographic retrieval. Eight variables were selected for
their significance: setting, user, request, database, search system, searcher,
search process, and search outcome. An extensive taxonomy of factors
affecting these variables is presented. Factors such as users information
seeking style, personality traits of intelligence and cognitive style,
personality characteristics, education, training, experience, and attitudes are
only a few that are included.
Kuhlthau (1988) found that individual differences were important in
the use of a particular online technology. Given equal course training,
engineering and science majors out-performed social science and
humanities majors. This is particularly striking in that it could mean that we
are developing online systems that are not equally accessible to all people.
If new technologies discriminate against particular groups of people,
research into ways to achieve equity through training, design, or additional
assistance must conducted. In a study by Borgman (1989) the relationship
between technical aptitudes, personality characteristics, and academic
orientation and their effect on the information retrieval process was
examined. Technical aptitude and personality characteristics tests along
with completed coursework in related fields in high school and college, and
the success of the subjects in this coursework was considered. Academic
orientation was used as a pointer to underlying characteristics that could
affect the information retrieval process. The results of this study indicate that
54


technical aptitude could be a predictor of performance on different types of
information retrieval systems. The contribution of personality characteristics
to information retrieval performance is likely to be weaker than that of
technical aptitude. No strong relationship existed between technical
aptitudes and personality.
A study conducted by Tenopir, Nahl-Jakobovits, and Howard (1991)
demonstrates the interaction among affective, cognitive, and sensorimotor
search behaviors. The cyclical nature of search behavior from affective
through cognitive to sensorimotor and back through cognitive to the affective
and needs of the searcher that develop are not met by current online
systems. An additional outcome of this qualitative study is that novice
searchers need assistance in modifying strategies as the search continued.
In an attempt to determine if creativity level, intelligence level,
personality traits in regard to masculinity, femininity, and self-esteem are
factors in the information retrieval process, a study of graduate students
enrolled in library and information science programs at six different school
was performed by Bellardo (1985). GRE scores do relate to some extent to
success in some searching tasks. However, the correlation coefficients are
not high. The lack of self-confidence is a factor only in the initial stages and
disappears once the training stage is completed. Recall scores were low
which indicates a need to emphasize techniques for enhancing recall such
as the use of synonyms, near-synonyms, and antonyms; increased use of
the Boolean operator OR; truncation methods, free text searching; and
proximity operators. It is the conclusion of this study that the differences in
55


performance among searchers of the same experience can be, to a small
degree, attributed to their verbal and quantitative aptitude, and inclination
toward critical and analytical creative thinking. It would seem that high
intelligence is not critically necessary for high performance.
Jacobson and Fusani (1991) study the contribution of subject area
knowledge to the successful completion of a search relative to the
contributions of computer background and system knowledge. Success was
based on the searchers' satisfaction with the search results. Prior computer
experience and system knowledge contributed only moderately to the
relevance of retrieved documents. A general effect for subject knowledge is
clearly indicated, however, the reason cannot be discerned from the data
collected.
Summary
Current research in electronic information retrieval is best viewed
from the construct of the two major paradigms. First, the earlier, data-
centered perspective is interested in improving the small to medium
database systems so that all or most of the relevant indexed documents can
be retrieved by intermediaries. Since time meant money, speed was a
contributing evaluative factor. Interest was in improving the systems so that
they could be used more efficiently by experienced intermediaries.
The catastrophic speed of development of electronic technologies
and accessibility to the public has created the user-centered perspective.
Research involving end users began by simply extending the intermediary
56


studies. However, research from both perspectives demonstrated the
importance of individual differences.
Telecommunications Research
Telecommunications is electronic communication across distances.
Collis (1992) reviewed 45 recent studies that dealt with telecommunication
issues in secondary schools. Only six of these studies involved the use of
online databases. The remaining 39 studies were concerned with
computer-mediated communication involving the use of E-mail for student-
to-student exchanges, teacher-to-teacher exchanges, and student-to-
teacher exchanges. None of the studies dealt with use of the Internet.
Research into the use of telecommunications for K-12 is centered
around profiles of early adopters and current usage. Methodological
emphasis is on questionnaires and surveys. Perhaps the largest national
survey was conducted by Bank Street College of Education Center for
Children and Technology. This national survey was designed to gather
information about the range and type of telecommunications activities used
by teachers and students (Honey & Henriquez, 1993). Scientific databases
and news services were perceived as the most valuable. Access to the
Internet was available to less than half of the participants. Those that had
access used it twice as much for professional activities as for student
learning. When used for student learning, the Internet was most commonly
used for E-mail followed by accessing news and bulletin boards, and
57


gaining access to remote computers. The process of retrieving information
on the Internet was not studied.
Hamilton and Thompson's (1992) research on early technology
adopters is based on adoption diffusion theory of Rogers. Results showed
that personal characteristics early adopters have in common include:
education level, social status, social participation, cosmopolitan outlook,
mass media use, personal communication, degree of innovative information
seeking, attitude toward change, attitude toward risk, aspirations, and
attitude toward fatalism. Honey and Henriquez's (1993) research added
experience in teaching and self-taught knowledge of computer technology to
the list.
Janowiak (1990) surveyed 498 K-12 teachers to determine what
educational technologies are used and valued by teachers, and what
emerging technologies they perceived as beneficial. Most highly used
technologies emerging from this study are: microcomputers, overhead
projectors, videocassette recorders, software teaching aids, and movie/slide
projection systems. Multimedia computer projection systems, computer
networks, videodisks, telecommunications, and software to aid problem
solving are seen as valuable emerging technologies to teachers.
Telecommunications was ranked as fourth and included comments about
distance learning, access to databases, student-to student communication,
and professional-to-professional communication. No specific references to
the Internet were encountered.
58


A recent study by Gallo and Horton (1994) investigated what the effect
was on a group of high school teachers who were given direct and unlimited
access to the Internet. Findings suggested:
teachers require ongoing Internet training, technical support, home
Internet access, and time in which to learn and incorporate the
Internet into their classes
Internet use can increase teachers' self-esteem and education
use of the Internet by teachers encourages them to restructure their
classes and schedules to accommodate Internet resources within
their classrooms
These studies show utilization of telecommunications technology but
not the processes of information retrieval utilizing a global database. In
order to understand the information retrieval process utilizing a global
database, specifically designed studies need to be conducted on all aspects
of information retrieval using an unlimited database. Using information
retrieved for learning as well as specific informational data needs in this
environment needs particular attention.
Contribution of Current Study
Available research in the field of information retrieval processes gives
insight into the search process itself, and the effect of training and individual
difference on the ability of a person to conduct a search. Previous studies
deal with severely limited media or databases content in which traditional
evaluation measures of recall, precision, and relevance could be employed.
59


Within these limited databases there is a size variation from small to large.
The large systems tend to create output overload. Retrieval strategies that
work well on small systems do not necessarily work well on large systems
(Blair & Maron, 1985). Traditional research has focused on behaviors of
professional searchers or intermediaries in locating facts quickly and
efficiently. Only in recent years has the paradigm switched from data-
centered to user-centered. Evaluation of success has not, however,
changed since the research was conducted in limited nonglobal databases.
While still considered to be in its infancy, this field is faced with the new
problems of the Internet, a global database, with no maps attached, and a
dramatic paradigm shift. The problems posed by the need to create an
Information Age, resource-based environment with global information
resources are dramatically different than the traditional focus of determining
whether a search was successful based on relevance and effectiveness
ratios.
Information retrieval emphasis has switched from factfinding to
learning. The trend toward user characteristics and traits appears to be in
the right direction. A medium of the future is the Internet (or a similar
construct), as exemplified by the number of primary and secondary schools
with access. It appears that information retrieval research focus and
methodologies must change to aid educators in the establishment of
resource-based learning environments that fulfill the needs of the students of
the 21st Century.
60


This current study on the attitudes and behaviors of inner city high
school students toward the use of the Internet as an information retrieval tool
is an initial exploration to develop educational activities that will enhance
students' existing beliefs, attitudes, and intentions. The students' current
mental models are influenced by their attitudes and beliefs. With knowledge
of these beliefs and attitudes, educators can begin to develop a strategy for
teaching student survival techniques for the 21st Century.
61


!
CHAPTER 3
METHODOLOGY
Introduction
In order to answer the question of how high school students,
predominantly minority, perceive the use of the Internet as an information
retrieval tool, the Fishbein and Ajzen Model of Attitude-Behavior
Consistency was used. Students in an inner city high school taking specific
classes that lend themselves to the instruction of Internet usage were the
study's population. The focus was on obtaining baseline data. Salient
behavioral and normative belief structures as well as student intentions were
determined by interviewing a small sample (n = 8) of the population. A
questionnaire was then constructed and administered following the Fishbein
and Ajzen model (Ajzen & Fishbein, 1980, pp. 261-263).
Subjects
Students attending an urban high school in a metropolitan school
district were the population for this study. The student body is 86% minority,
predominantly Hispanic (79%). Students enrolled in Telecommunications,
Adventures in SuperComputing, Business Management, and Business and
Office Education during the Spring semester of 1995 were the subjects of
this study. These classes were selected because Internet training was part
l
62


of their curriculum. Additionally, students who were enrolled in Advanced
Keyboarding who wished Internet training before graduation were included.
Students were given the option of spending a week on the Internet and then
having it available to them during class time as they desired. Internet
training was added to the curriculum for these students only, not the entire
class. Subjects were drawn from the researcher's classes and all steps of
the study were performed by the researcher. Only the Advanced
Keyboarding class is taught by other teachers. All other classes are taught
solely by the researcher.
A total of fifty-eight (58) subjects were initially involved in the study.
Three of the participants were excluded from the final results. Their answers
to the questionnaire demonstrated that they had not read the questions.
One participant marked all answers as a three (3); another marked them all
with a two (2); and the third participant marked all responses as zero (0).
Follow up discussions with the students revealed that they felt that they had
no difficulties with the language, questions, procedure, or concepts.
Discussions did reveal an attitudinal problem: "The requirement is that I fill it
out. That don't mean I have to read it." and "Its too long. I'm not going to
read it." Consequently, their responses were not considered valid. None of
these three students have a LAU A or B classification. (See page 64 for an
explanation of LAU.)
The fifty-five (55) students that form the study participants included
eleven sophomores, thirteen juniors and thirty-one seniors. There are
twenty-four (24) females and thirty-one (31) males included in the group.
63


Most are self-selected in that they signed up for the courses. Some,
however, were put into the courses by the counselors due to scheduling
problems. All students are, however, given the opportunity to change
classes at the beginning of each semester.
Both grade point average (GPA) and LAU classification were
obtained to check for variation in background. The GPAs for 45 (82%)
subjects are available from the January 1995 School District records.
Recent transfer of students into the school or necessary recalculation of
senior records could account for the missing 10 cases. The available GPA
scores ranged from 0.808 to 4.086 with a mean of 2.634. The distribution of
scores was fairly even: 26.67 % fell below 2.0; 37.78% fell between 2.0 and
3.0; and 35.56% fell above a 3.0.
For some of these subjects English is their second language. The
LAU classification is a language classification based on five categories
(Ovando & Collier, 1985):
A. Monolingual speaker of a language other than English.
B. Predominantly speaks a language other than English.
C. Bilingual
D. Predominantly speaks English.
E. Monolingual speaker of English.
The LAU classifications for only 25 (approximately 45%) of the subjects are
available from the January 1995 School District records. Several reasons
could account for this lack. When a student transfers between districts or
from one school to another, these statistics are not immediately available
64


due to paper work and the required District testing procedure. Secondly, the
LAU classification is only kept for students that have a D classification or
higher. Students could have an E classification. Or, a student could come
from a non-classifiable background, one in which English was the first
language spoken by the student. There are, however, subjects involved in
the study with no LAU classification who are not monolingual English
speakers. These students are most probably new to the District and,
therefore, have not yet been classified. The available classifications have a
mean of 3.120 and frequencies as shown in Table 1.
Table 1
Distribution of Available LAU
Classifications
l_AU Frequency Percent
A 1 4.0
B 4 16.0
C 11 44.0
0 9 36.0
Fishbein and Aizen Model
Theory of Reasoned Action
The Fishbein and Ajzen Model of Attitude-Behavior Consistency was
used to determine the beliefs, attitudes, and intentions of the students (Ajzen
& Fishbein, 1980). The model is based on the Theory of Reasoned Action
which is designed to explain human actions. The goal is to predict and
65


understand human behavior. The basic assumption of this theory is that
humans are rational and make systematic use of information that is available
to them. Whether this information is correct or complete is a separate issue.
The theory is demonstrated by Figure 1 (Walster, 1994). The
immediate determinant of an action is the individual's intentions. One
determinant of intentions is a person's individual attitude toward a behavior;
the individual's judgment about whether performance of the behavior is
good or bad. The second determinant involves the social pressures the
person feels to perform or not to perform the behavior. These social
pressures are called the subjective norms. People tend to perform
behaviors that they judge as positive and that they perceive people
important to them think that they should perform. Weights are assigned to
the two determinants of intentions in order to explain the relative importance
of attitudes and subjective norms.
Beliefs are a function of both attitudes and subjective norms. A person's
attitudes toward a behavior or action are termed behavioral beliefs (b).
Behavioral beliefs alone do not account for a person's attitude. The importance
that the individual places on the behavior or the evaluation (e) of that belief and
the behavioral belief held, account for the person's attitude toward a particular
behavior. Underlying a person's subjective norms are normative beliefs (n).
These beliefs deal with what persons believes people important to them think
about the behavior. Again a second factor is required to make up a person's
subjective norms. This factor is the person's motivation to comply with the
social pressures (m). Although beliefs are integral in the determination of
66


attitude and subjective norms, they require an evaluative factor and a motivation
to comply factor for determination. Therefore, a behavioral change must involve
a change in beliefs, or the evaluative or motivational factors.
Figure 1. Causal Relationships Among Components of the Rshbein and
Ajzen Model.
ATTITUDES
INTENTIONS
BEHAVIORS
weights
SUBJECTIVE
NORMS
Source: D. Walster, 1994, p. 159.
The Theory of Reasoned Action does not include factors such as
personal characteristics, ethnicity, age, sex, social class, intelligence or a
multitude of other factors that have been invoked to explain human behavior.
The factors are considered to be external variables (Ajzen & Fishbein,
1980). They affect the beliefs, evaluations, and motivation to comply of an
individual and are therefore indirectly accounted for by the Theory of
Reasoned Action.
67


Model
When using the model the first issue is the determination of the beliefs
of the individuals involved. The stress is on the "individuals involved"; these
are not suspected beliefs of the researcher or the researcher's beliefs. They
must come from the population involved in the study. A small sample of the
population is interviewed and a list of behavioral and normative beliefs is
developed from the interview results. Stated intentions are also developed
from the interview questions.
Construction of a questionnaire to be distributed to the entire study
population is the next step. The construction of the questionnaire follows
strict quidelines. An essential aspect for validity is to insure that questions
that evaluate the behavioral beliefs and questions that judge the strength of
the normative beliefs measure the same thing. Only the beliefs and
intentions developed from the interview questions are considered when
creating the questionnaire. Results from the questionnaire are statistically
analyzed using regression analysis.
Training of Students
As part of the normal class curriculum, subjects were exposed to
general telecommunications terminology, applications, and usage. A brief
description of the Internet, including its structure, and general content, was
given to all students. Subjects had access to four computers that have a
thinnet connection to the T1 line in the school. The computer lab in which
these computers are located is open for student usage before school, after
68


school, during students' lunch hour, and during most class periods.
Students wishing to use the machines for telecommunication have priority
access.
During class time, subjects worked in groups of two. Each team of
subjects received at least one week exposure to the Internet information
retrieval processes through such interfaces as Gopher, WAIS, Archie, WWW,
Mosaic, Netscape, and various search engines. Eight subjects at a time
work on the Internet while other subjects in class worked on different
projects. Guidance was given as to how to access the different interfaces for
surfing the Internet and was available throughout the time that the students
were engaged in Internet travel. The primary role of the teacher/researcher
was to prevent pornographic information retrieval by monitoring databases
accessed, and to help guide and direct subjects when they had questions or
appeared lost. How (what interface) subjects traveled the Internet and
where they traveled to was left up to each team. No formal curriculum was
developed for the week's access and students were encouraged to explore
and seek information that interested them.
Interviews
Sample Selection
Subjects were placed in groups dependent upon how quickly they
returned their completed Internet contract (Appendix A). The first group of
eight subjects were interviewed individually by the researcher in order to
elicit beliefs and intentions concerning the using the Internet for information
69


retrieval. Six (6) of the students were male and two were female. All eight
students were seniors. Ethnic breakdown consisted of one (1) Asian, one
(1) Anglo, and six (6) Hispanics. There were two LAU B students, one each
Asian and Hispanic. Specific results of these interviews are presented in
Chapter 4.
Interview Questions
Following the guidelines established in the Fishbein and Ajzen Model
of Attitude-Behavior Consistency eight interview questions were constructed.
The behavior of interest must be defined in terms of its four elements: action,
target, content, and time. In this particular case, the action is "using"; "to
retrieve information" is the context; and the "Internet" is the target. The time
was intentionally left open so as not to eliminate any present or future usage.
The first three questions were developed to elicit the behavioral beliefs of
the participants. Questions four, five, and six measure the normative belief
structure of the participants. This structure is composed of beliefs about
what people important to the participant want. The last two questions were
designed to determine the intentions of the students. All eight students
answered the same eight questions in the same sequence:
1. What do you feel are the advantages of using the Internet to retrieve
information?
2. What do you feel are the disadvantages of using the Internet to retrieve
information?
70


3. What else do you think about or want to tell me about using the Internet to
retrieve information?
4. Who do you think would be interested in your using the Internet?
5. Who do you think would approve of your using the Internet?
6. Who do you think would disapprove of your using the Internet?
7. Do you intend to use the Internet to retrieve information?
8. What kind of information do you intend to retrieve from the Internet?
Interview Procedure
The interviews were held during the class period at the desk in the
front of the room while other members of the class were working on
individual and/or group projects. The fact that their responses would not
affect their grade in any way was emphasized. Subjects were allowed as
much time as needed to answer questions. The approximate length of time
for an interview ranged from approximately 5 minutes to 11 minutes with the
average length being approximately 9 minutes. Responses were generally
extremely short and it was necessary to draw out more complete responses.
Care was taken to make sure that a thorough understanding of the question
was held by all subjects. All interviews were performed by the same person.
Analysis of Interview Responses
The first three questions are designed to establish the behavioral
belief structure of the students in order to determine their attitudes.
Questions four through six are designed to establish their normative belief
71


structure. These two factors are the predictors of behavior in the Fishbein
and Ajzen model. The last two questions, seven and eight, are designed to
determine the intentions of the students.
The beliefs and intentions most often mentioned were considered
salient and used in the construction of the questionnaire. A total of twelve
(12) beliefs and eleven (11) intentions were considered salient.
Determination of the salient beliefs and intentions, those most commonly
held, was accomplished by counting the frequency of responses.
Questionnaire
The questionnaire was constructed based on the beliefs and
intentions determined from the interviews and following the model
quidelines. (See Appendix B for the complete questionnaire.) It was then
distributed to fifty-five (55) subjects participating in Internet exposure. Since
a maximum of eight worked on the Internet at any time, the number of
subjects completing the questionnaire at one time was small. The
questionnaire was distributed to each group on their last day of assigned
Internet exploration. Concerns over language ability were addressed by
careful explanation and question asking before completion of the
questionnaire. In addition, questions were encouraged during the process
of completing the questionnaire. The questionnaire administrator stayed
with the group.
72


Navigation Logs
All students have access to the Internet before and after school and
during the lunch hour as well as during class. The only requirement is a
written log of all Internet activity. Students are required to record travel
locations, interface utilized, and reasons for using the Internet. Navigational
logs record actual behavior that reflects follow through from intentions. No
usage is discouraged except that of pornographic information retrieval or
illegal activities. Based upon the belief that the use of Internet for activities
such as games and Multi-User Dungeons (MUDs) is a form of information
retrieval and learning, subjects are not restricted to "academic" endeavors
on the Internet except during the formal class period excluding the initial
exploration session. During this session, only pornography and illegal
activities are restricted.
In summary, the methodology of this study is develop around the strict
quidelines of the Ajzen and Fishbein Model of Attitude-Behavior
Consistency. The determination of salient beliefs and intentions through
interviews and the design of the questionnaire based on these salient
beliefs and intentions are all dictated by the model. Some additional
demographic information was gathered in an attempt to determine subject
variations.
73


CHAPTER 4
ANALYSIS OF DATA
Introduction
In this chapter a systematic analysis of the data will demonstrate the
relevance of the study. The analysis will follow in the sequence in which it
was performed. First, a qualitative analysis of the interview data will be
presented. Once this data was collected and analyzed, a questionnaire was
constructed following the Ajzen and Fishbein Model requirements. Data
from this questionnaire were then entered into SPSS for analysis. Fifty-
seven (57) variables for fifty-five cases were entered. From these variables
an additional twenty-four (24) variables were created. In addition, data was
collected on four (4) demographic variables. A total of 85 variables were
considered. (Appendix C contains all the raw data for the 85 variables.)
The demographic variables were discussed in Chapter 3.
Presentation of data in this chapter will begin with the interview data and
progress through variables relating to behavioral beliefs, then normative
beliefs, followed by intentions. The concluding section will deal with the
regression and comparison tests.
i
74


Interview Analysis
Information elicited during the interviews was categorized into three
different topics: behavioral beliefs, normative beliefs, and intentions. These
were gathered from eight of the fifty-five subjects and used as the basis for
construction of the questionnaire that was distributed to all subjects. The first
eight subjects to return their Internet contracts were selected for the interview
process. Although one might be inclined to believe that these would be the
higher academic and more motivated students, this appears not be the case.
Two of the students interviewed had GPAs above 3.0, two had GPAs
between 2.0 and 3.0, and four subjects had GPAs of less than 2.0. Previous
experience with the students indicates that five of the students were not
highly motivated achievers but underachievers. Perhaps the thought of
exposure to something new and unknown to them was the motivation for
being the first to return their contracts.
Interviews were held during class time so as not to be perceived as
extra work or "punishment." All but one of the eight students was reticent.
They had to be cajoled into expanding their responses. One factor
influencing this could be the fact that the interviewer was also the subjects'
teacher who had expressed approval of the Internet for information retrieval
and whose general enthusiasm for information retrieval on the Internet was
obvious in the classroom. Great care was taken to assure the subjects that
their opinions would in no way effect their grade and that their opinions
would be grouped by content not by individual. After the first minute or two,
75


the students opened up and apparently spoke freely. Negative feelings
about Internet travel were eventually expressed openly and freely.
Behavioral Beliefs
The first three interview questions (Appendix A) are designed to elicit
the behavioral beliefs of the subjects toward the use of the Internet to
retrieve information. In order to develop the salient behavioral belief
structure, all the responses to questions one through three were listed and
arranged by similarity of ideas. Each group was then collapsed into one
item. For example,
Faster than other ways
Saves time rather than the library
Quicker than the library
Quicker
Fast
Can get current information real fast
were condensed to form the belief "My use of the Internet to retrieve
information is fast." The construction of the question was strictly dictated by
the action, target, context, and time of the behavior of interest (Ajzen &
Fishbein, 1980). In the above statement, "using" is the action; "to retrieve
information" is the context; and the "Internet" is the target. The time was
intentionally left open so as to not eliminate any present or future usage.
This level of specificity must be identical throughout the study.
76


A wide variety of responses was recorded and the act of collapsing
them into a manageable number of beliefs undoubtedly resulted in some
degree of loss of distinction in meaning. For example,
Easier than books, libraries, magazines and junk
Easy
Get college information easily
Dont have to do much research-all on the computer
Don't have to leave area
Don't have to pay or travel to get information from all over the
world
Easy access
were all condensed to mean "Easy to use." In addition, several of the
comments could be used in more than one category of beliefs. The
temptation to collapse comments into groups that require a language
comprehensible to the researcher but not suitable for the subjects was
constant. For example,
Difficult to use because the computer freezes
Takes too much time to get a password
Some databases require membership
could be grouped into a single belief: "Use of the Internet is difficult because
of technical application." The meaning conveyed by this statement was felt
to be too complex for the subjects of this study. Therefore, the original
statements were considered singly and not used.
77


The behavioral beliefs classification produced a set of twelve beliefs
(Table 2) that were used in the construction of the questionnaire (Appendix
B, Questions 4-15). Beliefs with a frequency of two or more were included.
Table 2
Behavioral beliefs used and not used in the construction of the
questionnaire and their frequencies.
Behavioral Belief Frequency
Used
Easy to use 7
Makes learning easy 2
Gives you information fast 11
Gives the most current information 2
Gives more information than the library 8
Gets information from ail over the world 3
Allows you to talk to people ail over the world 6
Its fun 4
Its interesting 3
Difficult because there's so much information 4
Difficult if you don't know your way around 6
Difficult because there's too much to read 2
Not Used
Difficult to use because the computer freezes 1
Takes too much time to get a password 1
Can get weather all the time 1
Some databases require membership 1
Normative Beliefs
Determination of normative referents was not as complex a task as
that of determining behavioral beliefs because of the directness of the terms.
Interview questions four through six (Appendix A) were used to compile a set
of normative referents. Parents included any reference to either sex of
78


parent. The category of family members consists of sister, brothers, and
cousins. Cousins were included because of the inclusive nature of the
family in the Hispanic culture. Due to the adjectives used by the participants
to describe the term "friends," this group was divided into two sections:
friends and close friends. In two cases it appeared that the term "close
friends" referred to non-school friends whereas the term "friends" referred to
school friends. This was not consistent over all cases. Table 3 presents the
normative referents that emerged from the interviews and the frequency with
which they occurred.
Table 3 Normative beliefs used and not used in the construction of the questionnaire
Normative Referents Frequency
Used
Employer 3
People who hire in business 2
Parents 4
Family members 4
Teachers 5
People who review college applications 2
Close friends 4
Friends 2
Not Used
Other students 1
No one 4
Those referents with frequencies of two or more were included in the
questionnaire. The referent, no one, with a frequency of four was not used,
79


however, as it would be impossible to establish an evaluation for this non-
existent referent. The high frequency of responses toward employment and
college related referents was unexpected. The eight salient normative
referents determined from the interviews were used to construct questions
twenty-eight through thirty-five (Appendix B).
Intentions
As with the collapsing of the behavioral beliefs, the collapsing of
intentions was involved and caused some loss of meaning. The amount of
reference to historical and cultural material was a surprising development.
The administrator made no reference to such material during the exposure
period. Responses such as:
Cultural information
Religious information
Government and historical information
Background about countries
Politics
Historical characters
History
were collapsed into one category, "Retrieve historical and cultural
information." Only those uses that had a frequency of 2 or more were
included on the questionnaire. A total of eleven intentions were considered
salient and used to form questions forty-four through fifty-four on the
questionnaire (Appendix B).
80


Intentions used and not used in the construction of the questionnaire and
Table 4
intentions used a
their frequencies
Used
Retrieve information about schools and college 6
Retrieve information for career planning 2
Retrieve information about jobs 2
Retrieve information for school reports 7
Retrieve information from other people 2
Retrieve information not in the library 2
Retrieve entertaining information 3
Retrieve historical and cultural information 7
Retrieve information about sports 2
Retrieve information about hobbies interests 6
Retrieve information for all my information needs 3
Not Used
Retrieve information about gang population 1
Retrieve travel information 1
Retrieve computer information 1
Retrieve weather information 1
Questionnaire Analysis
The raw data from the questionnaire on behavioral beliefs is entered
on the data sheet (Appendix C) as variables b4 through b15. The numbers
correspond with the question numbers on the questionnaire. Table 5
contains descriptive statistics on the raw questionnaire data. This and all
data was analyzed on a bipolar scale from 3 to -3 as recommended by the
Behavioral Beliefs
Model's designers.
81


Table 5 Descriptive Statistics on Raw Questionnaire Data of Behavioral Beliefs
Var Mean Std Range
Behavioral Beliefs (b) Dev
Max Min
Easy to use b4 1.127 1.743 3 -3
Makes learning fun b5 1.818 1.389 3 -3
Gives you information fast b6 1.545 1.653 3 -3
Gives the most current information b7 2.164 1.151 3 -3
Gives more information than the library b8 1.673 1.528 3 -3
Gets information from all over the world b9 2.473 1.120 3 -3
Allows you to talk to people all over the world b10 1.782 1.462 3 -3
It's fun b11 1.945 1.177 3 -3
It's interesting b12 2.073 1.200 3 -3
Difficult because theres so much information b13 0.891 1.449 3 -3
Difficult if you don't know your way around b14 1.273 1.830 3 -3
Difficult because there's too much to read b15 0.200 1.704 3 -3
The four lowest standard deviations fall within range of 1.151 to 1.200
and correspond to the four highest means. "Gets information from all over
the world" appears to be the most significant belief of the information
retrieval on the Internet for many subjects as indicated by its mean of 2.473
and its standard deviation of 1.120. Combine this belief with "Allows you to
talk to people all over the world" with a mean of 1.7 and a clear indication
develops that access to information and people around the world is a major
belief of information retrieval through the Internet. "Retrieving information on
the Internet is fun, interesting, and allows one to retrieve the most current
information" are also predominant beliefs held by the subjects. The subjects
consider the Internet as only "slightly" easy to use for information retrieval. In
82


addition, it is "slightly" difficult to use for information retrieval if you dont
know your way around and because of the amount of information and
reading. "Easy to use," "Difficult because there's too much reading," and
"Difficult if you don't know your way around" have a relatively high standard
deviation. It could be that the motivation to expend effort on the part of the
students varied. In addition, the English reading ability of the subjects could
have contributed to this variability in perception of difficulty.
Evaluators of Behavioral Beliefs
The raw data from the questionnaire on evaluators of behavioral
beliefs is entered on the data sheet (Appendix C) as variables e16 through
e27. The numbers correspond with the question numbers on the
questionnaire. Table 6 contains descriptive statistics on the raw
questionnaire data.
The means of all but three of the evaluators are closely grouped in a
range from 2.127 to 2.527. "Retrieving information from all over the world"
again has the highest mean followed by "Retrieving the most current
information" and "Making learning fun." The evaluation of the belief that
information retrieval using the Internet was easy had the lowest mean of the
positive behavioral beliefs. Once again the importance of "Retrieving
information from all over the world," "Making learning fun," and "Retrieving
the most current information" are shown as comparatively important. "Not
knowing your way around the Internet" is "slightly" whereas "Having a lot to
read" is only "slightly" good. There is a variation of range in this data.
83


"Making learning fun" and "Making information retrieval interesting" vary only
from one to three while "Retrieving the most current information" and
"Making information retrieval fun" vary from 0 to three. Again, the importance
of these behavioral beliefs is demonstrated by the means of their
evaluators.
Table 6
Descriptive Statistics on Raw Questionnaire Data of Evaluators of Behavioral Beliefs
Evaluators (e) Var Mean Std Dev Range Max Min
Making it easy to retrieve information e16 1.964 1.290 3 -3
Making learning fun e17 2.509 .690 3 1
Retrieving information fast e18 2.255 1.126 3 -2
Retrieving the most current information e19 2.509 .742 3 0
Retrieving more information than is in the library e20 2.127 1.090 3 -3
Retrieving information from all over the world e21 2.527 .716 3 0
Retrieving information by talking to people all over the ... e22 2.145 1.239 3 -1
Making information retrieval fun is e23 2.255 .907 3 0
Making information retrieval interesting is e24 2.200 .779 3 1
Having enormous amounts of information ... e25 2.291 1.031 3 -1
Not knowing your way around the Internet e26 -1.109 1.939 3 -3
Having a lot to read e27 .673 1.754 3 -3
Behavioral Beliefs (b) Times Evaluators (e)
In order to arrive at one of the components of an estimation of
intentions, a new variable was created that was the result of the behavioral
belief multiplied by the evaluator of that belief. The raw data from the
questionnaire on behavioral belief evaluator (b*e) is entered on the data
sheet (Appendix D) as variables be4 through be15. The numbers
84


corresponds to the question number of the behavioral belief. Table 7
contains descriptive statistics on the raw questionnaire data.
Table 7
Descriptive Statistics on Raw Questionnaire Data of Behavioral Beliefs Evaluators
Behavioral Belief' Evaluator (b*e)
Easy to use
Makes learning fun
Gives you information fast
Gives the most current information
Gives more information than the library
Gets information from all over the world
Allows you to talk to people all over the world
Its fun
It's interesting
Difficult because there's so much information
Difficult if you don't know your way around
Difficult because there's too much to read
Var Mean Std Dev Range Max Min
be4 3.200 3.937 9 -9
be5 4.855 4.107 9 -9
be6 3.564 4.841 9 -9
be7 5.600 3.536 9 -9
be8 4.545 3.511 9 -6
be9 6.455 3.599 9 -9
be10 4.218 4.458 9 -9
be11 4.855 3.719 9 -9
be12 4.727 3.664 9 -9
be13 2.127 4.005 9 -9
be14 -1.891 5.311 9 -9
be15 -.109 3.446 9 -9
The importance, as seen in the previous discussion, of "Gets
information from all over the world" is dramatically demonstrated by the
mean of b*e for this belief and its evaluator. Its mean of 6.455 is well above
the next highest mean of 5.600 for "Gives the most current information."
Estimates of intention is highly influenced by the behavioral beliefs "Gets
information from all over the world," "Gives the most current information,"
"Makes learning fun," "It's fun," and "It's interesting."
85


!
Normative Beliefs
The raw data from the questionnaire on normative referents is entered
on the data sheet (Appendix C) as variables n28 through n35. The numbers
correspond with the question numbers on the questionnaire.
Table 8 contains descriptive statistics on the raw questionnaire data.
Table 8
Descriptive Statistics on Raw Questionnaire Data of Normative Referents
Normative Referents (n) Var Mean Std Dev Range Max Min
Employer n28 .800 1.682 3 -3
People who hire in business n29 1.436 1.500 3 -3
Parents n30 1.109 1.652 3 -3
Family members n31 .818 1.513 3 -3
Teachers n32 2.255 1.322 3 -3
People who review college applications n33 1.600 1.523 3 -3
Close friends n34 .836 1.607 3 -3
Friends n35 .945 1.545 3 -3
The standard deviations of all referents show no dramatic shifts and
the ranges contain all possibilities. The means, however, do give some
interesting information about the strength of the different referents.
"Teachers," "People who review college applications," and "People who
hire in business" have the three highest means. A conjecture derived from
the data may be that people who have an input into the individual's future in
the world of education and business are believed to be interested in the
subjects' use of the Internet to retrieve information. "Family members,"
"Friends," and "Close friends have an unexpectedly low mean. Subjects
i
86


have a relatively strong belief system toward authority figures who could
play a role in the future course of their lives.
Motivation to Comply with Normative Referents
The raw data from the questionnaire on motivation to comply with
normative referents is entered on the data sheet (Appendix C) as variables
m36 through m43. The numbers correspond with the question numbers on
the questionnaire. Table 9 contains descriptive statistics on the raw
questionnaire data.
Table 9
Descriptive Statistics on Raw Questionnaire Data of Motivation to Comply With Normative
Referents
Motivation to Comply with Normative Referents (m) Var Mean Std Range
Dev
Max Min
Employer m36 1.218 1.750 3 -3
People who hire in business m37 1.600 1.594 3 -3
Parents m38 1.491 1.609 3 -3
Family members m39 .618 1.683 3 -3
Teachers m40 1.109 1.729 3 -3
People who review college applications m41 1.618 1.627 3 -3
Close friends m42 .255 1.818 3 -3
Friends m43 -.109 1.760 3 -3
Again the standard deviations are quite stable and the range of all
referents includes all possibilities. The mean motivation to comply for
"Friends" is -.109 and for "Close friends" a low .255. It appears that the
subjects' friends, both categories, are not a motivation to comply for the
87


subjects' normative beliefs to any great extent. The importance of peer
pressure, as is commonly understood, is not demonstrated by the motivation
to comply factor. "People who hire in business," "People who review college
applications," and "Parents" are the referents with whom the subjects are
most motivated to comply. Teachers are not a major factor in motivation.
Normative Referents (n) Times Motivation (m)
In order to arrive at the second component of an estimation of
intentions, a new variable was created resulting from the multiplication of the
normative belief by the motivation to comply with the normative belief (nm).
The raw data from the questionnaire on the multiplication of the normative
belief by the motivation to comply with the normative belief (nm) is entered
on the data sheet (Appendix C) as variables nm28 through nm35. The
numbers correspond to the question number of the normative beliefs. Table
10 contains descriptive statistics on the raw questionnaire data.
The motivation to comply with "Friends and "Close friends" appears
low when looked at separately, combined with the corresponding normative
belief, "Friends" and "Close Friends" become an important factor in the
determination of subjective norms. As could be predicted from the normative
beliefs and motivation to comply with normative beliefs, "People who review
college applications," "Teachers," and "People who hire in business"
become the primary influences in determining normative beliefs.
88


Table 10
Descriptive Statistics on Raw Questionnaire Data of Normative Beliefs Motivation to Comply
With Normative Beliefs (n*m)
Normative Beliefs Motivation to Comply with Normative Referents (nm) Var Mean Std Dev Range Max Min
Employer nm28 1.527 3.910 9 -9
People who hire in business nm29 2.964 4.242 9 -6
Parents nm30 1.800 4.205 9 -9
Family members nm31 .673 3.186 9 -9
Teachers nm32 3.000 4.765 9 -9
People who review college applications nm33 3.491 4.077 9 -9
Close friends nm34 1.909 3.566 9 -6
Friends nm35 1.109 3.505 9 -9
Intentions
The raw data from the questionnaire on intentions is entered on the
data sheet (Appendix C) as variables i44 through i54. The numbers
correspond with the question numbers on the questionnaire. Table 11
contains descriptive statistics on the raw questionnaire data.
The subjects' intentions to use the Internet to retrieve information
were determined by a series of questions. The use of the Internet for
"Retrieving information about career planning," "Retrieve information about
schools and colleges," and "Retrieve information for all my information
needs" had the highest means. Standard deviations for "Retrieve information
for career planning" and "Retrieve information not in the library" were both
relatively low indicating less variance in response. "Retrieve information not
in the library" had the highest mean of 2.255. The Internet as a secondary
89


source of information cannot be assumed from this, however, because the
mean for "Retrieve information for all my information needs" was 2.055.
Table 11
Descriptive Statistics on Raw Questionnaire Data of Intentions (i)
Intentions Var Mean Std Dev Range Max Min
Retrieve information about schools and colleges i44 2.036 1.217 3 -3
Retrieve information for career planning i45 2.164 .977 3 0
Retrieve information about jobs (46 1.873 1.233 3 -3
Retrieve information for school reports i47 1.818 1.454 3 -3
Retrieve information from other people i48 1.727 1.326 3 -2
Retrieve information not in the library i49 2.255 .947 3 0
Retrieve entertaining information i50 1.618 1.472 3 -2
Retrieve historical and cultural information i51 1.673 1.187 3 -2
Retrieve information about sports i52 1.055 1.779 3 -3
Retrieve information about hobbies interests i53 1.655 1.443 3 -3
Retrieve information for all my information needs i54 2.055 1.393 3 -3
Regression Analysis
The last variables to be created were surnbe, sumnm, and sumi, and
sumql. These variables are needed for the regression analysis.
surnbe =

28
sumnm = £nm
35
44
sumi = £i
54
sumql = Iq1a + Iq1b + Iq1c
90


Table 12 Descriptive Statistics on Estimates and Direct Measure of Beliefs and Intentions
Beliefs and Intentions Variable Mean Std Dev Range Max Min
Estimate of
Behavioral Beliefs sumbe 42.145 25.661 81 -52
Normative Beliefs sumnm 16.473 22.054 72 -48
Intentions sumi 19.927 8.987 33 -2
Direct measure of
Behavioral Beliefs sumql 6.665 3.811 12 -2
Normative Beliefs q2 1.364 1.568 3 -3
Intentions q3 2.073 1.200 3 -2
The values for each of the created variables involved can be found on
the data sheet (Appendix C) as variables sumbe 4 through 15, sumnm 28
through 35, sumi 44 through 54 and sumql. The significance of sumbe,
sumnm, and sumi is that they are estimates of the aggregate behavioral
beliefs, normative beliefs, and intentions of the participants. Direct
measures of these variables can be found on the data sheet (Appendix C)
as variables sumql, q2, and q3. Table 12 contains descriptive statistics of
these variables.
The standard deviation of the estimate of intentions is significantly
lower than the standard deviation of the other estimates. The range for
sumbe is 136 and 120 for sumnm. However the range for sumi is only 35. It
91


Full Text

PAGE 1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 2

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 4

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 5

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 6

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 7

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 8

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 9

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 10

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 11

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 12

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 13

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 14

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 15

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 16

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 17

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 18

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 19

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 20

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 21

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 22

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 23

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 24

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 25

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 26

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 27

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 28

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 29

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 30

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 31

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 32

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 33

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 34

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 35

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 36

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 37

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 38

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 39

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 40

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 41

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 42

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 43

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 44

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 45

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 46

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 47

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 48

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 49

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 50

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 51

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 52

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 53

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 54

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 55

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 56

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 57

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 58

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 59

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 60

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 61

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 62

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 63

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 64

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 65

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 66

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 67

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 68

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 69

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 70

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 71

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 72

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 73

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 74

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 75

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 76

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 77

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 78

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 79

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 80

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 81

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 82

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 83

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 84

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 85

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 86

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 87

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 88

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 89

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 90

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 91

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 92

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 93

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 94

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 95

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 96

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 97

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 98

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 99

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 100

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 101

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 102

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 103

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 104

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 105

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 107

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 108

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 109

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 110

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 111

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 112

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 113

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 114

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 115

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 116

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 117

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 118

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 119

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 120

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 121

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 122

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 124

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 125

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 126

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 127

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 128

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 129

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 130

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 131

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 132

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 133

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 134

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 135

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 136

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 137

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 138

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 139

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 140

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 141

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 142

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 143

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 144

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 145

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 146

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 148

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

PAGE 149

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.