Citation
Comparison of program designs, pull-out and in-class for Chapter I programs

Material Information

Title:
Comparison of program designs, pull-out and in-class for Chapter I programs
Creator:
Baber, Darlene M
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
xvi, 360 leaves : forms ; 29 cm.

Subjects

Subjects / Keywords:
Children with social disabilities -- Education (Elementary) -- Colorado ( lcsh )
Education -- Experimental methods -- Colorado ( lcsh )
Remedial teaching -- Colorado ( lcsh )
Children with social disabilities -- Education (Elementary) ( fast )
Education -- Experimental methods ( fast )
Remedial teaching ( fast )
Colorado ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ed. D.)--University of Colorado at Denver, 1990. Education
Bibliography:
Includes bibliographical references.
General Note:
School of Education and Human Development
Statement of Responsibility:
by Darlene M. Baber.

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:
26039872 ( OCLC )
ocm26039872

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

-... --..... COMPARISON OF PROGRAM DESIGNS, PULL-OUT AND IN-CLASS FOR CHAPTER I PROGRAMS by Darlene M. Baber B.S., University of Minnesota, 1968 M.A., College of St. Thomas, 1971 M.A., University of Northern Colorado, 1981 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Education School of Education 1990

PAGE 2

This thesis for the Doctor of Education degree by Darlene M. Baber has been approved for the School of Education by Myrle E. Hemenway Philip DiStefano [Cj'f() I

PAGE 3

Copyright 1990 Darlene M. Baber All Rights Reserved

PAGE 4

Baber, Darlene M. (Ed.D., Education) Comparison of Program Designs, Pull-Out and In-Class for Chapter I Programs Thesis directed by Professor Bob L. Taylor Two program designs, in-class and pull-out, were compared to determine differences in the academic achievement and selfconcept of Chapter I students, grades 3 to 5. Teachers of this student sample, Chapter I and regular classroom teachers, and Chapter I coordinators/directors from the State of Colorado were surveyed to determine the benefits and detrimental effects of both program designs. Reasons were given for program design preferences. Academic achievement scores of 235 Chapter I students from three Colorado school districts were compared for the 1986/1987 school year using analysis of variance. Normal curve equivalency scores (NCE) from standardized tests compared pretest and posttest scores for students. Self-concept scores for this sample were derived from the administration of the Self-Description Questionnaire (SDQ). A multiple analysis of variance (MANOVA) was applied to Normal Curve Equivalency (NCE) posttest scores and toe self-concept scores of students from the combined three districts. No significant differences were found in the achievement or self-concept of students of the pull-out program as compared to students.of the in-class program.

PAGE 5

iv Samples of teachers and coordinators were surveyed. Teachers and coordinators preferred the program design, whether pull-out or in-class, by which they delivered Chapter I services. Beneficial and detrimental effects of both program designs were described by teachers and coordinators. The conclusions are: 1. There were no differences by program design in the academic achievement and self-concept of the Chapter I students. 2. Both program designs were clearly judged by teachers and coordinators to be sensible and beneficial to pupil progress and adjustment school despite detrimental effects resulting from program design. 3. Both program designs benefitted students similarly, as no differences were found in achievement or self-concept, program design decisions. Therefore, decisions regarding program design selection can be made with regard to solving problems that are impacting the efficiency and effectiveness of Chapter I programs. 4. Teachers and coordinators of both program designs view the benefits of the program as outweighing the detrimental effects of the design, both in-class and pull-out. After more than 20 years of study and results reported in the literature, the detrimental effects of both programs remain apparent.

PAGE 6

v The form and content of this abstract are approved. I recommend its publication. Signed Bob L. Taylor

PAGE 7

DEDICATION To my father, Roy W. Baber, for a lifetime of love and encouragement, and to my children, Jamen Margaret Parkey and Micheal Rey Parkey II, who will carry on.

PAGE 8

ACKNOWLEDGMENTS The contributions of support, assistance and a great amount of time to this thesis by the people who encouraged its development are hereby acknowledged. Among those to whom sincere gratitude and appreciation are extended are the following: The major advisor of the thesis, Dr. Bob Taylor, who accepted the development and completion of this dissertation following the of two previous major advisors, Dr. Hazlett Wubben and Dr. Jim Rose. With continuing guidance, support and encouragement, Dr. Taylor directed the design of the study and promoted its completion. Dr. Myrle Hemenway, who edited thesis language with great care and provided needed constructive criticism. Dr. Philip DiStefano, who contributed to the design of the study and its approval. Ms. Virginia Plunkett, Chapter I Coordinator for the Colorado State Department of Education, who originally requested the study and encouraged coordinators to participate in the research. Without her influence, the study would not have been possible. Dr. Alan Davis, formerly of the Technical Assistance Center of Denver, Colorado, and presently on the staff at

PAGE 9

the University of Colorado-Denver, who assisted with the research design and statistical analysis and evaluation. viii Ann Underwood, who whenever needed typed every letter and number of this document and contributed invaluable advice regarding the requirements and form of this document. Kim Westerlund of the Rose Medical Center, who provided computer assistance with data input and statistical analyses. Toby Olberding, a manager of computer program research and development for U.S. West Communications, who provided assistance with computer program analysis, and friendship when needed. Each teacher and student who participated in the study, Chapter I coordinators, and Lucille Gallegos Jaramillo, James Pahlau, Sylvia Kreider, Colleen Rickert and Jo Ann Deller, who contributed information and a great amount of time for data collection and cooperation with teachers. The dedication and knowledge of these people inspired the work of this thesis.

PAGE 10

CONTENTS CHAPTER I. INTRODUCTION. 1 Rationale of the Study. 6 Problem of the Study. 9 Research Questions. 10 Hypotheses. 10 Definitions 11 Delimitations 13 Limitations 13 Organization of Dissertation. 14 II. REVIEW OF THE LITERATURE .. 16 The Incidence and Context of "Pull Out" 28 The Effects of "Pull Out" 28 Studies of "Pull-Out" and "In-Class" Programs ..... 31 Summary of the Literature 91 III. RESEARCH DESIGN AND PROCEDURES. 101 Overview. 101 Selection of Districts. 101 Methodology 102 Survey Design 104 The Sample 154 Students. 108

PAGE 11

Teachers ... 109 Coordinators. 111 Instrumentation for Academic Assessment 111 Standardized Achievement Tests. 111 The Instruments . 112 Standardized Test by District for Chapter I Student 112 Research Design 113 Data Processing for Academic Achievement; 114 Demographic Data for Achievement Analysis 114 Hypothesis I 115 Instrumentation for Self-Concept Assessment 120 Research Design 120 Variables 230 Selection of the Instrument 122 Background of the SDQ 123 Factor Analysis 124 Reliability and Construct Validity. 126 Pilot Study Descriptive Statistics for Self-Concept Data Processing and Analysis of SelfConcept Data. . . Summary of NCE's and SDQ. Instrumentation for Teachers' and Coordinators' Survey--Subproblem. Pilot Study 127 130 136 137 138 138 Data Collection and Analysis for Subproblem 139 Demographic Results of Coordinator Survey . 140 X

PAGE 12

IV. ANALYSIS AND RESULTS Overview. Problem of the Study. Research Question 1 Research Question 2 Subproblem. . Analysis of Research Question 1 District A ANOVA. . . Major Findings for Academic Achievement Combined Districts MANOVA for Academic Achievement . . . Major Findings for Achievement by Combined Districts .... Analysis of Research Question 2 Results of District A ANOVA for Self-Concept . . Results of Combined Districts MANOVA for Self-Concept. . . . . Major Findings of MANOVA for Self-Concept Summary Statement of Problem Analysis Summary Analysis of the Subproblem. Relevant Questions of Teacher Survey Results . . . . . Relevant Questions of Coordinators/ Directors Survey Analysis . Comparative Analysis of Teachers' and Coordinators' Surveys Open-Ended Questions .. xi 143 143 143 144 144 145 145 146 146 148 149 149 150 151 156 156 157 159 160 206 227 237

PAGE 13

Major Findings .... Research Question 1 Research Question 2 Subproblem. . . V. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS .. Problem of the Study. Research Questions. Hypotheses. Subproblem. Research Design and Procedures. Students. Teachers. Major Findings. Conclusions . Recommendations REFERENCES APPENDICES A. THE SELF-DESCRIPTION QUESTIONNAIRE. B. INSTRUCTIONS FOR HAND SCORING THE SDQ SCORING SHEET . . . . C. LETTER FROM DR. HERBERT W. MARSH IN SUPPORT OF RESEARCH . . . . . . . . D. PILOT STUDY OF THE SELF-DESCRIPTION QUESTIONNAIRE . . . . . E. REVISED SELF-DESCRIPTION QUESTIONNAIRE INSTRUCTIONS TO STUDENT . . . . . F. PILOT STUDY OF THE TEACHER QUESTIONNAIRE, 239 239 240 241 244 244 245 245 245 245 246 247 247 252 253 256 263 264 269 272 274 279 INTRODUCTORY LETTER SURVEY. . . . 283 xii

PAGE 14

xiii G. TEACHER QUESTIONNAIRE TO DISTRICTS B AND C, INTRODUCTORY LETTER, TEACHER SURVEY 288 H. CHAPTER I COORDINATOR/DIRECTOR QUESTIONNAIRE, INTRODUCTORY LETTER, COORDINATOR SURVEY 293 I. LETTER OF SUPPORT FROM DR. BOB L. TAYLOR. 298 J. COMPLETE MATERIALS GIVEN TO TEACHERS ADMINISTERING THE S.D.Q.. . . 300 K. QUESTIONS REGARDING THE STATUS OF THE CHAPTER I PROGRAM . 313

PAGE 15

TABLES Table l. "What Are the Two Major Strengths of the Title I Program?" Elementary School Staff Responses 57 2. Should Pull-Out Be the Major Program Model for Title I 59 3. Studies of Pull-Out Designs 94 4. Studies of In-Class Design 98 5. Teacher Questionnaire: Building Program Design, "Pull-Out," "In-Class" 110 6. NCE Pretest, Ethnicity by Grade Level. 116 7. NCE Pretest, Ethnicity by Sex. 117 8. NCE Posttest, Ethnicity by Grade 117 9. NCE Posttests, Ethnicity by Sex. 118 10. NCE Pretest to Pos.ttest Ethnicity Gains. 119 11. NCE Pretest to Posttest Ethnicity by Sex 119 12. Variables from Data for Combined Districts 122 13. Physical Appearance. 132 14. General School 133 15. Physical Ability 133 16. Reading. 134 17. Parent Relations 136 18. Mathematics. 136 19. General Self 136 20. District A: NCE Pretest Between Group Comparison 147

PAGE 16

20. District A: NCE Pretest Between Group Comparison ..... 147 21. District A: NCE Posttest Between Group Comparison 147 22. Tests of Between-and Within Subjects Effects. 149 23. District A ANOVA Self-concept as Measured by the SDQ. Self-Description Questionnaire. Variance Between In-class and Pull-out Programs . . . . . . . . . . 152 24. District A Self-Concept Variables Showing Significance at the a = .05 Level. . ... 153 25. Combined Districts MANOVA for Effects of Self-Concept by Grade, Sex, Ethnicity and Program Type . . . . . . . . . 153 26. District A MANOVA Comparison of Group Means by Ethnicity and Type of Program . . . 155 27. Persons Who Decided Upon Program Design in Building . . . . . 163 28. Reasons for Selection of Design. 164 29. Question 5: The Most Beneficial Aspects of the Chapter I Program. . . . . . 167 30. Question 6: The Most Detrimental Effects of the Chapter I Program. . . . . . . 177 31. Question 7: Problems Impacting Success with the Program. . . . . . . 184 32. Question 8: Students Miss No Academic Subject 189 33. Students Miss Regular Classroom Instruction to Receive Chapter I Reading . . . . 192 34. Students Miss Regular Classroom Instruction to Receive Chapter I Math . . . . . . 195 35. The Chapter I Program Supplants or Supplements the Regular Education Program. . . . . 197 36. Chapter I Program Design in Building Is Judged by Teachers to be a Sensible and Beneficial Plan . . . . . . . . . . . 199 XV

PAGE 17

37. Question 23: Worriesor Concerns About Program or Chapter I Students in Relation to Program Design in Building . . . . . . 201 38. Person(s) Who Decided Upon Program Design in Your District . . 207 39. Reason Design Was Selected 209 40. Question 9. Most Beneficial Effects of Chapter I Program. . . . 211 41. Most Detrimental Effects of the Chapter I Program. 214 42. Problems Impacting Success of the Program. 218 43. Do Chapter I Students Miss Regular Classroom Instruction to Receive Chapter I Reading or Math . . . . . ... 222 44. Chapter I Program Design Judged to be Sensible and Beneficial 224 45. Worries or Concerns About Program or Chapter I Students in Relation to Program Design . . 225 xvi

PAGE 18

CHAPTER I INTRODUCTION The Chapter I compensatory education program is a federally funded program targeting the need to improve the educational opportunity of economically disadvantaged children. There are arguments for continuing Chapter I services, especially in a time when there is a growing number of needy and homeless children (Plunkett, 1985), and the problem of educating the poor and disadvantaged has not lessened since President Kennedy challenged the nation in 1961 "to abolish all forms of human poverty." As our nation adds to its roles of poor, minority, transient, welfare children, there is a growing number of homeless who need to be helped to get off the streets and to be educated so that they can participate in the mainstream of society. Although the state has the primary responsibility to develop and implement educational programs to provide for a spectrum of needs for these children, the federal government has assumed a share of the financial burden. Targeting the children of poverty who have great academic need, the Federal Government created funding through the Elementary and Secondary

PAGE 19

Education Act (ESEA), April 11, 1965. Chapter I is authorized by ESEA and is a subsection of ESEA. 2 Currently, the Federal Government, through Chapter I, provides an unrestricted $500 per child for cooperative or coordinated service models. This amount clearly cannot provide the entire intervention needed to assist the targeted children; hence there is a need to discover the most effective, creative, and responsible way to provide the services with limited funds by both the government and the educational community. Those who are held directly accountable for the limited funds, such as Chapter I coordinators, teachers, and regular education teachers, often do not have adequate input, knowledge or the empirical data to make decisions regarding program design. School district central office personnel, including Chapter I coordinators, determine program design. According to Virginia Plunkett (October, 1987), Chapter I State Coordinator, study of program design effectiveness is necessary to provide data to assist the personnel who are accountable for Chapter I funds in decisions regarding program design. Program decisions are often made as a result of misconceptions about Chapter I requirements and with an obvious lack of research or empirical data (Poyner and Others [sic], 1981). There has been a misconception by many districts that Chapter I services must remain distinct and separate from regular education services, as in a pull-out program design. However,

PAGE 20

3 Chapter I regulations require that program funds be expended in addition to and not in place of regular education funds (Gaffney & Schember, 1982). Services, therefore, are not required to be provided in a different location or at a different time as in a pull-out program. In-classroom program designs have been developed as workable alternatives to pullout programs, but the reasons for developing alternative program designs in many districts remain unclear (Allington and Others [sic], 1985; Gaffney & Schember, 1982; Halasa, 1983; Harpring, 1985). States and districts have needed a broader approach to interpreting the rules and regulations that govern categorical programs in general and Chapter I in particular (Savage, 1987). Savage promotes radical change in program design. Refocusing centering on evaluation of program effectiveness rather than on fiscal accountability is an alternative presented in this study. The effectiveness of pull-out programs as the predominant design for providing Chapter I services has been demonstrated to be a cost-effective means for improving the academic achievement of the students in need of compensatory education (Yap, 1985). However, the pull-out program has been criticized as causing poor self-concepts among the population experiencing the service (Glass & Smith, 1987). This criticism has been leveled without concern for the self-concept of the student who has experienced difficulty as a learner prior to Chapter I services. In fact, Chapter I requires that the student achieve

PAGE 21

4 well below the national norm as a criterion for remedial services. These students may also have experienced poverty, poor relationships in the family and/or community, and other socioeconomic handicaps prior to entering the school environment (Levine & Stark, 1981). A further criticism of the pull-out program is that the Chapter I educator and the regular educator do not communicate and cooperate in sharing responsibility for the progress of the student (Allington and Others, 1985). They are isolated in time and space and often do not share curricular content, goals, and activities. Ideally, in-class programs are designed to bring students and their teachers together in time and space to share responsibility for the curriculum and progress of the students and to eliminate the segregation that allegedly caused the stereotyping and poor self-concept of Chapter I students in the pull-out programs (Fretchling & Hammond, 1978). Many school districts have tried an alternative to the pull-out design without researching the possible beneficial and detrimental effects of the pull-out model or the in-class model (Bean & Eichelberger, 1985; Gaffney & Schember, 1982; Halasa, 1983; Neumann, 1985; Vasquez-Nuttall, 1982). A final judgment in the report by Glass of the University of Colorado in 1977 condemned pull-out programs as detrimental to student selfconcept and student achievement. Time away from the regular curriculum, and subjects missed, as well as poor preparation of

PAGE 22

5 Chapter I teachers, were further criticisms adding to the skepticism concerning the pull-out design (Glass & Smith, 1977). Studies revealed that the in-class programs which were created to save Chapter I from these mounting criticisms can result in another set of benefits and detrimental effects to students and to teachers. Information concerning studies of in-class design is given in Chapter II of this study. Students were targeted as remedial students while remaining within the classroom, which did not help improve their self-concepts. Chapter I teachers have had to obtain and effectively use space within the regular classroom setting. Problems come from two teachers in the same classroom and the resulting rise in noise level as well as who teaches what to whom and when? (See Chapter II.) It does not seem that the pros and cons of either design have been adequately considered in order to meet the needs of the district, the building site, and, finally, the individual child. David Savage (1987) does not recommend taking time to consider which program design would be more effective in improving academic achievement without negatively affecting student self-concept. Radical changes such as coordinated service models are being promoted by such authors as Savage whose article on the topic appeared in the Kappan, April, 1987 and Felix, Hertlein, McKenna and Rayborn (1987) whose article appeared in the Kappan in June 1987. This model would group

PAGE 23

students receiving services according to need and would then track them separately for compliance purposes. This model implies a needed review of the past research on tracking students, even though separate tracking may guard districts against commingling and supplanting categorical funds. 6 Before this new design is adopted, it is of great importance to evaluate the effectiveness of the present program designs. According to Plunkett, in a rationale for this study when Plunkett had a conversation with the researcher in September 1989, "Teachers and students should have input into the planning of the design in which they are to participate. They should help determine the future use of pull-out and inclass designs." The primary goal of Chapter I services should be focused on the improved academic achievement and positive self-concept of the student. Research must continue to aim at the attainment of these goals. As funds and resources continue to decrease and as the need for categorical services rise, designs that most effectively accomplish these goals are critical. Rationale of the Study Educators have suspected a relationship between student self-concept and cognitive achievement. Even if there was no direct correlation between the two, self-concept would be an important indicator of children's happiness in a place where

PAGE 24

they are required to spend many hours per day and per week (Henerson, Morris & Fitz-Gibbon, 1978, p. 43). Improved academic achievement may or may not directly cause improved self-concept. The reverse also may or may not be true. However, it is important to understand how a compensatory education program design can maximize the expected beneficial outcomes that the services should provide to improve both the academic achievement and self-concept of the students that are serviced. 7 The pull-out design, as the predominant service plan for Chapter I of ESEA, has been demonstrated through research to allow disadvantaged students to make gains on normal curve equivalency (NCE) scores beyond the stated expected gains of the programs, and it remains a cost-effective service design (Yap, 1983, 1985). Yet, the impetus for pull-out instruction is derived largely from two provisions in the Chapter I legal framework. The first requirement is that Chapter I must supplement, not supplant or replace, the regular education curriculum. The second provision requires Chapter I funds be used only to fund the excess costs of Chapter I programs and must not supplant funds from non-federal sources. Since the pull-out designs were implemented basically to solve the possible audit problems of these two provisions, few research studies were completed to predetermine expected positive outcomes for the students and teachers. Research was not available to guide the

PAGE 25

8 choice of program design based upon the impact of design choice on achievement and the self-concept of the learner. A more flexible interpretation of federal regulations has allowed that compensatory education could occur in the same place and at the same time as regular education. Thus, programs developed in response to criticisms of pull-out programs as detrimental to student self-concept.! Student achievement and student self-concept data supporting the success of in-class programs as an alternative to pull-out models are unavailable as a basis for program design decisions and imple-mentation in school districts. There may be some other bases for selecting program designs if problems with audits for federal rules and regulations have been overcome with both designs. Problems with federal audits have burdened school districts since prior to 1980. If program designs, in-class and pull-out can be demonstrated to be equally effective in improving student cognitive achievements without negatively affecting student self-concept, then program design decisions can be made considering the impact on staff, building and district as well as the impact on students. Certainly, teacher attitudes toward the design and their cooperation and coordination with the entire school staff, may be critical factors in program effectiveness. The teachers' !However, adequate measurement of student self-concept, as a result of empirical data from Chapter I students, has not been available.

PAGE 26

9 and Chapter I coordinators' views of how the program design directly affects students can be determined through research. If it is determined that the Chapter I curriculum does not supplant or replace regular education curriculum, then it must be supplementary to, and coordinated with, regular education cur.riculum. It is important to demonstrate that pull-out designs and in-class designs can be predicted to bring about positive student outcomes in achievement and student self-concept. Chapter I teachers, coordinators, and regular classroom teachers can provide data to indicate positive effects and negative effects of either design. Those who will be making decisions regarding program design can begin to think critically toward the expected degree of satisfaction by teachers, students and staff with the section of school called "Chapter I." Problem of the Study The problem of this study was to compare the effectiveness of a Chapter I program pullout design with a Chapter I in-class design by comparing student achievement scores from students in each type of delivery system and by comparing the students' self-concept results for students in each of the two delivery systems in selected schools. Hence, the study provided data to determine the effectiveness of the program designs using two independent variables, student achievement as measured by Normal

PAGE 27

10 Curve Equivalency (NCE) pretest and posttest scores, and student self-concept data as measured by the Self Description Questionnaire (SDQ) for analyzing the sample of students, grades 3 to 5. A subproblem of this study was included to determine the support among school personnel for the two approaches. This could have long-range implications for future decisions on which approach will become established practice. Which program, and for what reasons, was favored by the Chapter I teachers, the Chapter I coordinators, and the regular classroom teachers? Research should aid in designing programs that will most effectively meet the needs of disadvantaged youth in a time of limited and decreasing resources. Research Questions This study investigated the following research questions. 1. Is the pull-out design more effective than the in-class design for improving student academic achievement? 2. Is the pull-out design more effective than the in-class design for improving student self-concept? Hypotheses This study tested the following hypotheses: Hypothesis I -There was no significant difference between academic achievement scores of Chapter I students in the pull-

PAGE 28

11 out program and achievement scores of Chapter I students in the in-class program. Hypothesis II -There was no significant difference in the self-concept of Chapter I students, as measured by the Self Description Questionnaire in the pull-out program and the selfconcepts of Chapter I students in the in-class program. Definitions For the purpose of this study, the following definitions are appropriate. Pull-out: The pulling out of participating Chapter I students from the regular classroom to receive remedial instruction services from the Chapter I teacher and/or aide in a small group setting (Yap, 1983, p. 5). In-class: The classroom setting within a school building providing instruction by a certified non-remedial teacher and a certified remedial teacher and/or aide. Regular classroom: The classroom setting within a school building pro.viding instruction by a certified non-remedial teacher to a heterogenous grouping of students, usually over 20 in number. Combination: The provision of instruction services in a combination of pull-out and regular classroom settings. Disadvantaged. A child is at a disadvantage if, because of social or cultural characteristics (e.g., social class, race,

PAGE 29

12 ethnic origin, poverty, sex, and geographical location) he/she comes into the school system with knowledge, skills and attitudes which impede learning and contribute to a cumulative academic deficit. The disadvantage may persist throughout school life and contribute to restricting later economic and social opportunities (Passow, 1986, p. 16). Self-Concept. The feelings or beliefs of an individual about herself/himself as reported in relation to general school, general self, reading, mathematics, physical ability, physical appearance, peer relationships and parent relationships (Marsh, Barnes, Cairns, & Tidman, 1984). Attitude: A position or manner indicative of feeling, opinion, or intention toward a person or thing (Morehead, Albert, & Loy, 1972). Environment: The aggregate of surrounding things or conditions; the totality of external influences (Morehead et al., 1972). Self-Report: A direct type of attitude assessment of a person's behavior, feelings, or beliefs about himself or herself, determined through responses made by the person about herself/himself. NCE Scores: Normal Curve Equivalents expressed as standard scores where the mean and standard deviations are fixed, with equal interval scales that can be equated across standardized tests (Glass & Hopkins, 1984).

PAGE 30

Delimitations There were several delimitations to the study resulting from difficulties in getting the cooperation of school districts utilizing Chapter I programs. 1. Four of six school districts contacted agreed to cooperate with the study. Three were selected. These school districts had Chapter I programs which could provide data for the two delivery designs, pull-out and in-class programs. 13 2. Only data from Chapter I pull-out and in-class programs from districts utilizing both programs were used. This study did not concern special education pull-out programs. 3. Only Chapter I students in grades 3 through 5 were used from the three school districts. 4. The study was conducted only during the academic school year, 1986-1987. Limitations There were several limitations to the study which should be identified. 1. Both pull-out and in-class programs included in the study were not identical in all three school districts. 2. The reading difficulties of the Chapter I students may have impacted the results of the Self-Description Questionnaire, though a pilot test to check vocabulary was conducted.

PAGE 31

3. The ethnic makeup of the student populations varied for the three school districts. 14 4. The instruments used varied in quality. The ones to test student data had established validity and reliability. An investigator-developed instrument, which had only undergone pilot testing with a graduate class, was used to survey teachers and coordinators. 5. The teachers of the classes used in the study administered tests so that neither a single nor trained administrator administered the tests providing the scores used in the study. 6. Only three districts were used in the study, and these were selected because they met specific criteria. The sample was not randomly drawn from the school districts of the state. 7. The self-concept test was administered once only at the end of the treatment. Organization of Dissertation The dissertation is organized into five chapters. Chapter I presents a problem statement, with a subproblem and hypotheses, and also includes the delimitations, limita-tions, and definitions. Chapter II presents a review of relevant literature, including a summary table of benefits and detrimental effects of pull-out and in-class designs contained in the literature. Chapter III describes the procedures followed in the

PAGE 32

15 investigation, including a description of the sample, instruments used, pilot studies of the instruments, and methods of testing and analyzing data. Chapter IV analyzes statistical findings in relation to the hypotheses involved in the investigation. Survey data are analyzed to address the subproblem of the study. Chapter V contains a summary of the study, conclusions and recommendations for further research.

PAGE 33

CHAPTER II REVIEW OF THE LITERATURE A wealth of research and literature, developing as an upside down pyramid, has amassed to shed light on compensatory education in this country. There seems to be no question remaining regarding the need for compensatory education to target the economically disadvantaged. However, what remains to be learned is the effect of various instructional methods, models and designs on the performance and perception of the self of the academically disadvantaged. How did we come to the sub-question of which program design may be most effective in providing the needed services? To begin at the top of the upside down pyramid, we must look at the Law. and Regulations that led to the Chapter I Programs and Federal support for the nation's poor in the educational setting. The first major federally funded compensatory education effort took a straight shot at poverty in 1965 with the development of Title I. Its target was the education system to enhance the opportunities of the poor in public schooling. Following Title I of the Elementary and Secondary Education Act of 1965 (ESEA), compensatory education developed into a sophisticated and effective program and sharply focused effort

PAGE 34

17 against the sources of educational failure (Douthitt, 1985). In a Phi Delta Kappan article by Virginia Plunkett, State Coordinator of the Colorado Department of Education, Chapter I, Plunkett traced the evolution of compensatory education from Title I to Chapter I (Plunkett, 1985). The problems of educating the poor and disadvantaged children began to emerge in 1957 with the Soviet launching of Sputnik I. Congress responded with a number of measures, but the issue of poverty did not attract national attention until President Kennedy addressed the nation in 1961, stating that "man holds in his mortal hands the power to abolish all forms of human poverty.'' In 1964, President Lyndon Johnson hit the poverty bullseye by proclaiming that one of four major tasks of the nation was to better the education of millions of disadvantaged youth who need it the most. President Johnson signed ESEA into law on April 11, 1965. This act was not a response to public pressure. According to Plunkett, ". the ESEA is an example of Presidential and Congressional leadership acting solely in response to a largely unrecognized national problem" (Plunkett, 1985). The hope of the Congress was that ESEA would encourage school districts to develop imaginative new approaches for meeting the educational needs of poor children. Through the 70s, the nation's lawmakers provided more specific directions on how Title I funds were to be spent.

PAGE 35

The funds were to go only to schools with concentrations of children from low income families. Children with the greatest deficits in basic skills were to be provided services in reading, math, and language arts. Family income was not to be a consideration. Specific educational approaches were to be designed based on a comprehensive needs assessment of each child. The number of children per program was limited so as not to dilute Title I services. These Title I services were to serve as supplements to, not as replacements for, the regular school program. The law also required Title I programs to include other educationally sensible components: o clearly stated objectives, o coordination with the regular classroom program and with other special programs, o parent o dissemination to the Title I staff of research data and information on promising practices, and o dissemination to parents and the community of information about the Title I program. (Plunkett, 1985, p. 535) 18 Congress strengthened the component of parent involvement, significantly increased funds for state administration, and strengthened and clarified the requirements. Title I programs increased in quality and sophistication while program outcomes became increasingly positive. Then in 1978 educators viewed the new Title I requirements related to parent involvement as an encroachment on local prerogative inhibiting local initiative and creativity. Educators made their dissatisfaction known to members of Congress. Reenactment of Title I as Chapter I of the Education Consolidation and Improvement Act of 1981 (ECIA)

PAGE 36

19 reduced or eliminated many of the requirements that had spurred dissatisfaction from educators. Between 1965 and 1981 Congress was again aroused to action to counteract the Reagan Administration's direction to reduce the federal role. Accordfng to administrative plan, the Chapter I budget was to be cut by $2.48 billion. Congress coalesced a movement by 200 of their own members, parents, educators and children to spare the drastic hurt to school children, since only 45% of all eligible were served as it was in 1981-82. Congress was able to reduce the budget cuts from 14% to 11% and then to pass a supplemental appropriation of $148 million over President Reagan's veto. However, inflation won out. The slight rise in appropriations over the next few years could not keep pace. Plunkett voiced a continued concern that Chapter I funding, which fails to keep up with the rate of inflation, is sufficient only to provide services to about half the children that are eligible. Legislative changes that weaken the program are another concern. Despite these concerns, Plunkett concluded that we have made substantial progress toward narrowing the achievement gap for disadvantaged children. She attributed this progress of Title I/Chapter I over the years to two major factors: (1) A clear mission to help educationally deprived children from low income areas and (2) an annual appropriation of funds by Congress. Plunkett maintained that the rules and

PAGE 37

20 regulations that require funded programs to focus on children's educational needs have been generally reasonable. Chapter !/Title I programs have steadily improved the academic achievement of disadvantaged children. Plunkett (1985) attributed this success to: o improved evaluation procedures o a better understanding of how to run the program o matching instruction to children's needs o using a diagnostic/prescriptive approach o providing appropriate staff development o coordinating Chapter I activities with the regular classroom o involving parents However, which program designs, of the traditional designs, could mold all of these components into the most effective program model? This question remained without address in the Plunkett article (1985). In 1986, Dr. Plunkett called for an extensive study of program designs and her questioning resulted in intensive research for this thesis. Francis Archambault reviewed the literature on the traditional models, pull-out and in-class. He found that although the pull-out was the predominant model used in compensatory education, the in-class model was on the rise in studies cited to 1984 (Archambault, 1987). VanEcko, Ames, and Archambault (1980) did not attribute the wide popularity of the

PAGE 38

21 pull-out model to pedagogical reasons, but rather argued that the design was selected to comply with previously stated provisions, and state and local perceptions of federal requirements. Two provisions in the Title !/Chapter I legal framework appear to have provided the impetus for pull-out instruction. The first is the supplement-not-supplant provision as reviewed by Plunkett. These funds also could not be used to supplant or replace funds from non-Federal sources. In addition, theexcess costs requirement provided that Title I funds could be used to fund only the excess costs of Title I programs and projects. Archambault (1987) referred to a summary from Gaffney and Schember (1982) on how these two provisions affected Title I instruction: In earlier years there was some uncertainty about what the supplement-not-supplant/excess costs provisions required of schools to ensure that Title I participants received extra services. When auditors from the U.S. Department of Health, Education and Yelfare cited school districts for in-class programs that also served non-Title I students (a "general aid" violation), some states and districts perceived the problem to be "supplanting" and turned more to pullout models as the solution. Consequently, though Title I never so stated, some program administrators believed pullout designs were legally required, or necessary to avoid audit problems; and some states refused to approve any in-class programs. Thus, "uncertainty or misconceptions about the meaning of certain requirements and the fear of possible audit violations" led some states to promulgate "overly restrictive policies concerning the types of programs Title I may fund. (pp. 6-7) Gaffney and Schember concurred that more frequent use of pull-out model may have increased the number of supplement and

PAGE 39

22 not supplant violations. They speculated that Title I instruction substituted for, instead of supplemented, state and locally funded instruction, especially where students were pulled out for long periods of time. Congress tried to correct this situation with two directives in 1978. First, Congress stated that Title I did not require any particular educational strategy. Secondly, a directive to the office of Education set forth regulations which told administrators how to design in-class as well as pull-out models. In 1981, Title I regulations offered four more program design models in addition to these two. Also, in the summer of 1981, Congress went even further. The excess cost provision was repealed and a new provision was added. Pull-out projects could not be required to show compliance with the new Educational Consolidation and Improvement Act (ECIA). Archambault (1987) concurred with Gaffney and Schember: The following conclusions appear to flow from the above discussion: (1) prior to the passage of ECIA in 1981 pullout programs may have been instituted in large part as a means of complying with what were perceived to be Federal regulations, for the purpose of avoiding sanctions resultingfrom violations of these regulations; (2) that at least from the vantage point of 1981 Congress claimed that they never required pullout for districts to be in compliance with the provisions of the regulations; and, (3) that in the last days of the Title I regulations, but even more clearly under Chapter I regulations, the Department of Education has specifically suggested that instructional models other than pullout should be considered. However, all of these pronouncements appear to have been made without conclusive evidence about the pedagogical effectiveness of these various alternatives. III-67)

PAGE 40

23 Archambault (1987) in his paper under the subsection titled "Instructional Setting: Key Issue or Bogus Concern," reviewed the somewhat sparse and sometimes confusing literature surrounding the instructional setting debate. He outlined the views of the critics of the pull-out programs. -It has been used to demonstrate compliance with certain program regulations and not justifiable on pedagogical grounds. Pull-out diminishes cooperation between regular and compensatory instructors. -It reduces teacher responsibility for pupil progress. -Valuable content is lost when pull-out instruction is received. -It unnecessarily labels and stigmatizes students. Pull-out leads to resegregation. (p. III-60) Proponents seem to have comparatively little evidence to justify its widespread use, but state in return: Pull-out provides opportunity for more concentrated instruction in a small group setting. -It provides better trained teachers. (p. III-60) While proponents of pull-out have little evidence to justify its widespread use, critics similarly cannot justify a widespread shift to in-class instruction.

PAGE 41

24 Proponents have countered that pullout provides the opportunity for more concentrated instruction delivered to smaller groups of students by better-trained teachers. Thus, the lines have been drawn, with both sides contending that they have convincing evidence for the superiority of their favorite strategy. What is disconcerting, however, is that the proponents of pullout have comparatively little evidence to justify its widespread use and the critics of pullout would be similarly hard-pressed to justify a wholesale shift to in-class instruction. (Archambault, 1987, p. III-60) Regarding the types of students served and the nature of instruction, Archambault referred to a set of studies conducted following the mid-to-late 1970s (Breglio, Hickley & Beal, 1978; Carter, 1984). Of the approximately 15% of all elementary grade school children in the U.S. that were served by Title I, 45% of low achieving students received compensatory education. Fifty-four percent did not receive services. More non-poor than poor students were receiving Title I and more regular achieving students than low achieving were receiving Title I instruction. About 2,000,000 low achieving students were not receiving services. Finally, the Sustaining Effects Study researchers found that Hispanic and Black students in large cities and rural areas were proportionally less likely to receive compensatory education services (Archambault, 1987, p. III-61). Of those receiving services prior to a 1977 report of the National Institute of Education (NIE), about 9.5% of Title I districts offered remedial reading or language arts as part of the compensatory education program. Forty-five percent offered

PAGE 42

25 remedial mathematics, and less than 10% of the districts offered any other type of remedial instruction. Students spent about three hours and 45 minutes per week in Title I language arts and about three hours per week in mathematics. Archambault and St. Pierre (1980), KimborQugh and Hill (1981), Lignon and Doss (1982), and Allington and Others [sic] (1985) all reached similar conclusions regarding Title I students. They received more total language arts instruction in school than regular education students in the same district. Although they received more total language instruction, combining regular and compensatory instruction, the compensatory instruction replaced a portion of the regular education instruction. In addition, according to Carter (1984), while Title I students received considerably more instruction in reading and math than the regular education students, it was not clear that they experienced a net gain in instruction, as they missed instruction and activities that regular students experienced (Archambault, 1987, p. III-63). NIE (1977) and all national studies did concur that smaller class size was an advantage for compensatory education students. The average class size for instruction was 9 students in reading and 12 students in both mathematics and language arts. Related to class size and grouping, the advantage of smaller class size allowed greater opportunity for teacher flexibility. It did report, from the literature, that teachers of Title I students

PAGE 43

used different methods and practices than teachers of regular students (Carter, 1984). Moreover, according to NIE (1977), Cooley and Leinhardt (1980) and Fretchling and Hammond (1978), it also appeared that at least some of these techniques (i.e., alternative learning paths and sequencing for individual children, individual and small group pacing, assignment of specific learning objectives or activities to individual children, diagnostic and prescriptive teaching, etc.) may be viewed as components of individualized instructional programs. Finally, Fretchling and Hammond (1978) reported that many different instructional approaches could work (Archambault, 1987, p. III-64). 26 The teacher flexibility was there to provide different approaches to Title I students, but were the instructional specialists delivering the compensatory instruction (about 66%) and paraprofessionals (about 24%) qualified to do so? According to VanEcko, Ames, and Archambault (1980), Coulson et al.l977) and Carter (1984), these education specialists were generally more highly educated, with somewhat more coursework and inservice training than regular teachers. However, they may have had somewhat less teaching experience. Cooley (1981) did report two complications of Title I regulations for the Title I teachers providing additional instructional services. First, teachers were not to serve students who were ineligible for services. Secondly, they could

PAGE 44

not be the only teacher to instruct eligible students. Thus developed the concept that schools had no alternative but to pull students out of regular classrooms to provide "extra" instruction in Title I classes (Archambault, 1987, p. III-64). 27 Prior to 1977, NIE reported that almost 75 percent of the Title I students in reading programs received instruction through a pull-out program. Title I language arts instruction was delivered in a pull-out setting to 41% of the eligible students and mathematics instruction entertained 45% in this small group setting. Coulson et al. (1977) concluded that the ideal amount of time students spent in pull-Qut was on the rise between 1973-74 and 1974-75, the two years for which the data were analyzed. Then in 1977, with pull-out programs on the rise, and literature reviews seemingly positive in nature, one of the most influential early studies criticized pull-out programs, claiming_ that pull-out programs were detrimental to students. A new question became the target of research. Were pull-out programs detrimental to student "Pull Out" in Compensatory Education, a study prepared by Gene Glass and Mary Smith (1977) was a major criticism of pullout programs for Chapter I and Special Education. Special Education involves a population that is diagnosed as handicapped. Therefore, the conclusions of the Glass study pertaining to special education were not considered relevant to this research review. Another criticism of the Glass study addressed the

PAGE 45

28 persons interviewed on the "pull-out" issue. Of 23 persons interviewed, only three were employed by public schools, titles unknown. The concern may have been lack of personal involvement with program design at the school building level. The "Conclusions, Observations, and Recommendations" have influenced proponents of in-classroom designs. Conclusions about the "pull out" technique are as follows: 1. Pulling Title I eligible pupils out of regular classrooms for compensatory instruction is virtually universal. 2. The "pull out" procedure per se has no-clear academic or social benefits and may, in fact, be detrimental to pupils' progress and adjustment to school. 3. The "pull out" procedure is used by schools more to satisfy Title I regulations than because it is judged by teachers to be a sensible and beneficial plan. (Glass & Smith, 1977, p. 46). The chapter of the Glass study on the nature and effects of "pull out" programs contained further conclusions that form the basis of the Coordinators Survey and Teachers Survey of this study (see Appendix B). The Incidence and Context of "Pull Out" The Effects of "Pull Out" The national statistics for the amount of the entire instructional day spent in the "pull out" setting were on the rise in the 1970s, from around 5% in 1973-1974 to around 9% in 1974-1975. Pupils missing regular classroom instruction in the subject from which they were removed was the concern addressed by Federal regulations, that Chapter I instruction does not

PAGE 46

29 supplant or replace regular reading or mathematics instructions. Glass stated, "By some chop logic we do not understand, supplanting is not supplanting at all if one supplants science and social studies" (Glass & Smith, 1977, p. 2). At the elementary level when a pupil is "pulled out" for remedial reading, language, or math, the chances are: 1 in 3 that no academic instruction is missed. 1 in 5 that the same subject for which he is pulled out is missed. 1 in 4 that instruction in social studies is missed. 1 in 7 that instruction in science is missed. (Glass, p. 17) The remedial subject matter specialist, Chapter I teacher, who removes students from the classroom setting, merely rechristened regular classroom teachers, concluded Glass, has the most cynical assessment of the teacher's role and contribution. Glass maintained that these teachers received very little extra training and virtually no extra pay. Pupils have three chances in four of receiving remedial instruction from a "specialist" when "pulled out" of the regular classroom. A study by the National Institute of Education (NIE) (September 1977) showed beneficial effects of "pullout at certain grade levels and in certain subjects and detrimental effects elsewhere." Glass and Smith (1977) found little support for these conclusions. The academic gains made by "pulled out" and mainstreamed compensatory education pupils in the NIE data are virtually

PAGE 47

identical, differing over all grades and subjects by less than one quarter in grade equivalent units. (p. 3) Glass concurred with a finding of consistent negative rela-tionship between the percentage of time pupils spend in the "pull out" setting and their math and reading achievement, 30 consistent across all grades and subjects. These data are found in the files of the Emergency School Aid Act (ESAA) conducted by Systems Development Corporation. In findings from over 40 experiments, Glass found that invidiously labeling pupils as "mentally retarded," "intellectually slow," or "academically weak" reduces the academic achievement of these pupils. Teachers' judgment of the success, motivation, and social competence of labeled pupils is reduced by nearly one standard deviation (below those for comparable pupils not so labeled). Glass stated, In our opinion, the "pulled out11 pupil is placed in moderate jeopardy of being dysfunctionally labeled, of missing opportunities in peer tutoring and role modeling, and of being segregated from pupils of different ethnic groups. (p. 5) The instance of the pull-out problem was one explanation according to Glass. Pull-out is an artifice created by ESEAA Title I (Chapter I) regulations concerning "supplementing, not supplanting" and minimizing "excess costs." The data following the Glass study did not uphold the contention that virtually no support was found for the "pull out" concept among educators in their professional organiza-tions. Two statements by Glass were questioned in the surveys

PAGE 48

31 of this dissertation. First, Glass maintained that teachers worry that pulling pupils out of class creates discontinuities in their schooling and makes coordination of teaching difficult. Secondly, the regular classroom teachers felt less responsible for pupils whose needs are presumably being met somewhere else by a special teacher (Glass and Smith, 1977, p. 6). There were no quantitative data cited to date to support the second of these statements. Pull out programs do not promote racial integration of classrooms. The smaller Title I (Chapter I) classrooms are probably ethnically and racially more homogeneous than the regular classrooms in the schools in which they reside. However, reviewing literature on grouping students according to ability, Glass found the research was immense and inconclusive. Finally, Glass and Smith (1977) called for further research on the effects of pull-out on student self-concept, teacher expectancy, and strong expectancy biases in pupils and teachers alike. The difficulty in drawing an analogy between "pull out" and allied forms of instruction is that too little is known about exactly what happens to pupils who are "pulled out" and what they, their peers, and their teachers think about it. (p. 40) Studies of "Pull-Out" and "In-Class" Programs A major criticism of the Glass study was found in Archambault (1987, p. III-72). Glass and Smith apparently were willing to base their conclusions about pull-out on a single

PAGE 49

32 study, ESEA. A review of the impact of other variables such as ability grouping, labeling, peer tutoring, racial desegregation, and mainstreaming the handicapped was included in other studies. However, Archambault referred to the ESEA study as the "sine qua non" of their argument. From this single study, Glass and Smith could argue that pull-out was not only ineffective in improving achievement, but it produced harmful labeling and poor self-concept in students. From ESEA they further concluded a consistent negative relationship between percentage of time pupils spend in pull-out and their reading and mathematics achievement. Pupils pulled out of their regular classroom must receive remarkably compensatory instruction to offset the risks. Pull-out serves as a means for resegregating students, and so forth. Archambault (1987) disagreed, stating that In our opinion, the pulled out pupil is placed in moderate jeopardy of being dysfunctionally labeled, of missing opportunities for peer tutoring and role modeling, and of being segregated from pupils of different ethnic groups. (p. III-72) Archambault (1987) discussed a reanalysis of the IDS study conducted for NIE and the ESEAA study to support his findings that pull-out students were in "moderate jeopardy." He sum-marized the two studies as follows: One of the first large scale (n = 8319) investigations of the effect of pullout was the third year evaluation of the Emergency School Aid Act (Coulson et al., 1977). In this study, the impact of pullout and 18 other variables was investigated across two types of compensatory education programs (Pilot and Basic

PAGE 50

33 Programs), two content areas (reading and mathematics) and three grade levels (3, 4, and 5). Pullout was defined here as the proportion of reading or mathematics instruction received outside the regular classroom, and along with 18 other variables, it served as a predictor variable in 12 regression equations (2 programs 2 x content areas x 3 grade levels yields 12 equations). For eight of these analyses no significant differences were found for the pullout variable. In the other four instances three of which were concerned with reading, pullout was found to have a significant negative effect on achievement. Leinhardt and Pallay (1982) interpret these findings to mean that a change from receiving one-half instruction in reading to receiving one-half instruction in pullout would be associated with a small reduction in the dependent measure. These researchers argue that no matter what the dependent measure was, the effect was very small. Perhaps even more important than the small size is the strong possibility that pullout was associated with less gain because (1) less able students tended to get pullout instruction and (2) the pretest did not control adequately for such differences. Further, and as suggested by Coulson et al. (1977), schools were more likely to use pullout instruction when they had students with more severe problems. The Instructional Dimensions Study (IDS) conducted for NIE by the Learning Research and Development Center (LRDC) at the University of Pittsburgh also was concerned with the effect of pullout. This research focused on reading and mathematics instruction in first and third grade and used classrooms as the basic unit analysis. Here the variable "setting" was scaled as a composite of the time students were in supplemental instruction outside the regular classroom and the number of children in a classroom that were pulled out for instruction. NIE (1977) reported that the results uncovered in this research differed by content area and grade level. First-grade students receiving instruction in the regular classroom (i.e., the in-class setting) made significantly larger gains in reading and mathematics than those in pullout settings. However, at the third-grade level, setting had no significant effect on reading achievement and pullout was associated with larger achievement gains in mathematics. NIE interpreted these results to mean that neither setting is consistently associated with greater instructional effectiveness. Arguing that the test was too difficult Glass and Smith (1977) criticized the NIE study for using the achievement pretest as a covariate. Consequently, even researchers at LRDC argued that the covariate did "very little to adjust

PAGE 51

34 for initial differences among first-grade students, and the use of gain as a dependent variable (worked) no better (probably worse) than posttest" (Leinhardt and Pallay, 1982, p. 568). (Archambault, 1987, pp. III-68-69) Criticisms followed the Glass report beginning with Fretchling and Hammond (1978) in a re-analysis and refinement of the data subsequent to monitoring the IDS study. Their conclusion remained consistent for in-class instruction at grade one for both reading and mathematics. Instruction was found to have a positive effect on achievement. Previously, grade three findings for reading instruction stated that instructional setting made no difference, while pull-out was found to be more effective for mathematics. Fretchling's re-analysis maintained that rechecking and cleaning the data showed in-class to be more effective for reading, while no significant difference was found for instructional setting for mathematics. Despite these changes in the original data, Fretchling and Hammond (1978) did not support a wholesale change to in-class instruction. Reasons for their position demonstrate the need for further convergent evidence: (1) the IDS findings were based on data from districts that were "special" and not nationally representative; (2) they had a gut feeling that mainstreaming (i.e., in-class instruction) may pose problems for many teachers and simply not be possible in some instances; and (3) they were unwilling to generalize from a single study (p. 9). Similarly, Glass and Smith (1977) were perhaps

PAGE 52

35 premature in this same instance to generalize from a single study. Cooley and Leinhardt (1980) and Leinhardt and Pallay (1982) also reported findings different from those reported by NIE. Setting was related to student achievement gain only in third grade reading and where pull-out was associated with less gain. Changing the population to include only children who had received pull-out instruction and redefining certain variables, Leinhardt and Pallay found further differences across two analyses (Archambault, 1987, pp. III-70-71). The most plausible interpretation of these results is that the non-pullout students had more academic knowledge than the pullout students, that is, the poorer students were chosen for pullout. Hence, pullout was associated with less achievement because it had less able students and not because it was an inferior practice. Pretest did not adequately adjust for all the initial differences. In this case, the underadjustment would be particularly faulty for the first grade since (as Glass & Smith, 1977, pointed out) the pretest was not a good measure of initial abilities. Examining the results with the non-pullout students removed, we see the following: Pullout is still associated with less gain in third grade reading but not in the other three cases. Overall, the more time a student spent in the segregated pullout class, the greater the gain. It seems likely that effective practices were being implemented in these settings. It cannot be determined whether students who were pulled out suffered some initial losses due to having been removed from the regular class. We can say that once pulled out, they increasingly gained from more pullout instruction. (p. 568, in Archambault, 1987, p. III-72). In addition to the achievement findings, Leinhardt and Pallay also reported on the stigma attached to labeling a child as eligible for Chapter I services. Many educators believe that compensatory education students are likely to feel

PAGE 53

36 different and rejected rather than rescued when pulled out of their regular classrooms for instruction. Research findings do not support this belief. Noddings (1978) argued that stigmati-zation occurs in mainstream programs and may well be even greater in the mainstream setting. According to Hayes (1983), in the regular classroom setting, compensatory education students are both targeted and remediated in front of their classmates. Further, it is argued that in a pull-out setting, most or all of the students are in similar need of remedial instruction, and labeling may be meaningless (Madhere, 1981). However, within the larger building setting, pulled-out students may be easily identifiable (Archambault, 1987, pp. III-76, III-77). Archambault concluded that it is not the setting itself that aggravates the feeling of being different, but the sensitivity or insensitivity of teachers within the setting. Noddings (1978) and Kennedy (1978) were in agreement, and Archambault (1987) has found Kennedy's advice to be particularly compelling: Compensatory education students are less likely to be subjected to labeling and its negative effects in environments where teachers actively encourage children's respect for and appreciation of a variety of human differences; in environments where similarities between learning tasks and materials are emphasized; and in environments where Title I and non-Title I children frequently move, in an organized way, to other parts of the building or classroom to receive special instruction. (p. 35; in Archambault, p. III-77) The question of stigmatization and attitudes toward

PAGE 54

37 compensatory education (CE) laboratories was again addressed in a study by Poyner and Others [sic] (1981). A Description of Compensatory Services in High Poverty Schools is a part of an extensive series of studies on CE and its long-term effect. The report described CE programs and compared their activities with those of regular education. Data were collected in 55 high poverty schools through interviews, structured observations, and the generation of narrative protocols based on semi-structured observation. Principals, regular classroom teachers, CE teachers and students were included in the study. Positive opinions from the principals on the effects of CE programs far outweighed the negative effects. Benefits mentioned by principals included more individualized instruction for non-participants by virtue of reduced pupil-teacher ratio while served students received compensatory instruction, non-participants' use of equipment and materials, more appropriate instruction for nonparticipants, more expert or competently provided instruction for non-participants, and improved non-participant attitudes toward school, self and other students (p. 83). The detrimental effects cited by a few principals related to the observation that non-participants suffered from not being able to receive the benefits of the CE programs. Poyner (1981) concluded that the results of principals' positive opinions might be expected, considering that principals are generally

PAGE 55

involved in the process of selecting CE programs and working out the details. they are supporting their own selections and On the other hand, teachers often are not 38 In a sense, decisions. consulted, frequently and programs which they may or may not like are imposed upon them. (Poyner and Others, p. 86) Regular teachers in six different schools had positive feelings toward the CE programs and felt they were helpful. However, 27 of the 325 regular classroom teachers (in 14 schools) voiced problems and complaints that were truly varied. Complaints reported by teachers in two schools were problems with disruptions of the regular classroom associated with pulling students out for CE instructions. Several teachers questioned the professional competence and working habits of the CE staff. In contrast, however, the dedication of pull-out CE teachers and/or aides was described and noted by observers in 10 different schools. Much evidence was cited. such as teachers reported taking extra courses and inservice workshops in order to learn new approaches to compensatory students. Poyner and Others (1981) gave pointed advice to later researchers regarding the complaints and problems of the teachers. Again, the particular complaints are varied. Several reflect the scheduling and location themes discussed in previous sections of this report, and several seem to reflect internal communication problems or faulty understanding of fear of violation of program intents and guidelines. Unlike the complaints registered by regular classroom teachers. however, several of these complaints seem to reflect situations involving school administration (e.g., use of teachers as substitutes, policies pertaining

PAGE 56

39 to parent conferences, meetings, and assemblies, and materials not arriving on time). Perhaps principals could be made more aware of the need to monitor and prevent problems in areas such as these. (p. 85) Poyner's final conclusion was that the issue of stigmatization was of little importance in the opinions of principals, teachers and students. The students seemed to enjoy themselves in the pull-out laboratories. Compared to some 13 expressions of positive attitudes toward pull-out laboratories compiled from 12 different schools in daily topical summaries, only three statements reflecting unfavorably on pull-out laboratory situations were discovered. In one of the few instances that hinted at stigmatization, several students expressed resentment at having to leave the regular classroom while a party was going on, to attend the CE lab. The results of this study seemed to directly contradict the conclusions of Glass and Smith (1977) andd others who mentioned that students are in extreme jeopardy from stigmatization in a pull-out setting. Also, it might be pointed out that principals, teachers and students were consulted for the first time in the Poyner study. Further, the study done by Poyner and Others (1981) indicated that while there was great variation among the different programs studied to 1981, general characteristics were noted. In general it was found that principals were wellsatisfied with the effectiveness of the CE programs, students had positive attitudes toward the programs, teachers had mixed attitudes, and there was little

PAGE 57

evidence of stigmatization of program participating students. (1981, abstract) In 1981, a national movement was developing to eliminate Title I pull-out arrangements. In a summary and conclusions examining the Chicago Learning Reading Program (CMLRP), Levine and Stark (1981) examined information, hoping to improve the general effectiveness of inner-city elementary and intermediate schools. Selected schools in Los Angeles, California, New York, New York, and Chicago, Illinois, 40 identified factors in augmented achievement gains. Five schools in Los Angeles and Chicago implemented school-wide approaches to improve instruction for students in concentrated poverty areas. School-wide approaches were defined as elimination of Title I arrangements that fragment the education of low achievers. Effectiveness, or success of big city elementary schools, was generally defined as academic achievement higher than expected, given the socioeconomic composition of the student body. Efforts were made to identify the characteristics of unusually successful inner-city schools. Efforts aimed to improve academic achievement and create more successful inner-city schools involved the development and implementation of materials to teach reading comprehension skills through the Chicago Mastery Learning Reading Program (CMLRP). Another important effort involved school-wide approaches for improving achievement through systematic school-by-school planning in Title I schools. For example, during the

PAGE 58

41 1980-1981 school year, the superintendent of the New York Community School District (including 21 elementary schools in Brooklyn), Frank Arricale, and staff, decided that a more concentrated and comprehensive approach was needed to raise achievement. Levine and Stark reported that Arricale's decision was encouraged by research that Title I "pull out" programs which take students out of the regular classrooms for special instruction generally should be avoided because they tend to be fragmented in their operation and impact on achievement (Levine & Stark, 1981, p. 4). In addition, the New York City United Federation of teachers was also strongly supportive of mastery learning approaches for improving achievement in a "within-theurban-classroom" approach. The school-wide project in Los Angeles was a particu-larly interesting attempt to eliminate pull-out instruction to aid desegregation. In 1978, the Los Angeles Unified School District (LAUSD) initiated a court-ordered desegregation plan which dealt with racially balanced schools and required that efforts be made to improve achievement in schools which remained Racially Isolated Minority Schools (RIMS). Most RIMS were defined as inner city schools whose students could not be included in the reassignment plan. A particularly compelling account of budget for the Los Angeles Project and a rationale of pull-out programs as creating obstacles for improvement of achievement follows:

PAGE 59

42 For the 1979-80 school year, approximately $40,000,000 was spent to improve education in RIM schools. Approximately $5,000,000 of this amount was expended as part of the Schoolwide Project at 74 schools participating in Title I. For the 1980-81 school year, 73 Title I schools were added to the Schoolwide Project. The Schoolwide approach, authorized in Section 133 of PL 95-561, allows a school.to use Title I funds for all students rather than limiting expenditures to low-achieving students. Until PL 95-561 was passed in 1978, Title I was mostly limited in practice to "pullout" instruction in which eligible students are removed from regular classrooms for special instruction in reading, math, and other subjects. Pullout programs create obstacles in working to improve the achievement of disadvantaged students. For example: (1) students removed for special instruction are labelled as "dummies" by other students; (2) scheduling complications and confusion detract from instruction in the regular classroom; (3) instructional methods and materials differ between Title I and regular classroom instruction, thus sometimes confusing students or even damaging achievement; (4) regular classroom teachers are encouraged to feel they are not responsible for the performance of low-achieving students--this very difficult task can be perceived as the responsibility of Title I teachers; (5) students whose achievement improves significantly are returned to the regular classroom where their achievement may not continue to improve; (6) materials and equipment purchased for Title I students either sit in cabinets unused or teachers are forced to violate regulations to use excellent materials with all their students; (7) "disincentives" are created wherein Title I personnel may lose their positions if too many students improve too much in achievement; (8) confusion is created concerning the principal's authority as compared with central Title I office authority to supervise instruction. Principals are tempted either to relinquish responsibility or to use Title I personnel for inappropriate tasks, or both; and (9) inordinate amounts of staff time are spent keeping records related to the eligibility of students. (Levine & Stark, 1981, pp. 32, 33) Levine and Stark concluded that given the above stated difficulties in the model Title I approach (pull-out), it was not surprising that achievement of students had not risen to a level commensurate with the billions of dollars spent nationally

PAGE 60

43 each year. Further, to avoid these difficulties Title I funds could now be spent for all students in a school providing that participating schools serve a population not less than 75 percent from low income families and that the district adds funds for non-Title I students at the same per-pupil level as. Title I provides for eligible students. However, the extensive cost and financial limitations were cited as reasons why the LAUS would eliminate the Schoolwide Project during the 1981-82 school year. Additional concerns were questions about how well it had worked and how important a force for change it had been, especially in view of the desegregation crisis (Levine & Stark, 1981, p. 33). Levine and Stark concluded that despite the intention to eliminate the expensive Schoolwide Projects, the Project had made a significant difference, or, at least facilitated substantially improved instruction in some schools. However, Levine and Stark referred to "descriptions of research concerning the inherent unsoundness of most pull out approaches" as found in the Glass and Smith study (1977). The final conclusion emphasized that problems inherent in pull-out could be overcome or minimized at inner city schools even in the absence of participation in a Schoolwide Project. Levine and Stark looked to support for this generalization in a description of two Chicago elementary schools which seemed to have worked out effective school-wide approaches within Chicago guidelines

PAGE 61

for implementation of Title I. No specifics were cited in regard to the inherent problems with pull-out or how problems were overcome. The homogeneous grouping for reading instruction (part time) in all schools studied did indicate success in achievement. Homogeneous grouping seemed to be more 44 successful than is frequently true because it seldom involved pulling students out of their regular classroom. Grouping was based on essential skills providing homogeneity of grouping, unlike programs utilizing the basal reader. Levine and Stark maintained that the alternative materials used decreased rather than magnified low self-image among slower students. No matter the factors involved in the success or lack thereof, the arrangements and processes described in the study "functioned in an inter-related manner to help inner city schools get off 'dead center'" (Levine & Stark, p. 62). While the Schoolwide Projects were projected to close in 1982, in large part due to excessive costs, others were studying the excess costs provisions of Title I and their impact on program design. In a study by Michael Gaffney and Daniel M. Schamber (1982), The Effects of Title I Supplement and Not Supplant and Excess Costs on Program Design Decisions, the findings concerning program design practices, including pullout, were detailed in the reports central section.

PAGE 62

45 The purpose of the study was stated as follows: The Title I District Practices Study was conducted by Advanced Technology, Inc. for the U.S. Department of Education's Planning and Evaluation Service. One goal of this study was to describe how local districts operated projects funded by Title I of the Elementary and Secondary Education Act [ESEA] in the 1981-82 school year. A second, related goal was to document local educators' rationales for their program decisions, their perception of the problems and benefits of requirements contained in the 1978 Title I Amendments, and their assessments of the expected effects of Chapter 1 of the Education Consolidation and Improvement Act [EDIA] on school district operations of Title I projects. The study was specifically to draw cross-time comparisons with the findings of the Compensatory Education Study conducted by the National Institute of Education [NIE] and to provide baseline data for subsequent analyses of Chapter 1, ECIA's administration. The results of the Title I District Practices Study are presented in this and eight other special reports, plus the study's Summary Report. These reports synthesize data collected from a mail questionnaire sent to Title I Directors in more than 2,000 randomly selected school districts, structured interviews and document reviews in 100 nationally representative Title I districts, and indepth case studies in 40 specially selected Title I districts. (Gaffney & Schamber, 1982, p. 7) As was stated earlier, Chapter I of the ECIA of 1981 (P.S. 97-35) made two general changes in the Title I framework for supplement not supplant and excess costs provisions. Effective October, 1982, the excess costs provision was eliminated. Chapter I then allowed certain "special state and local program funds" to be excluded from determinations of compliance with the supplement-not-supplant provision. The Gaffney and Schember report was concerned with the effects of the provisions on program design prior to 1982 and to possible program design issues under the supplement-not-supplant provisions.

PAGE 63

46 The amount of data collected for the study was impressive. The data from the Title I District Practices Study (DPS) conducted by Advanced Technology for the U.S. Department of Education (ED) previously summarized, supported general conclusions. o Most districts surveyed (92 percent) use a pullout design for part or all of their Title I program; this over whelming reliance on the pullout approach for delivering part or all of Title I services holds true across districts of various enrollment sizes. o About a third (30 percent) of the districts surveyed use an inclass model for part or all of their program, and use of inclass designs is increasing; very large districts (79 percent) use the inclass approach for part or all of their program much more than do small districts (28 percent). o A belief in the educational superiority of a program design for part or all of a district's program was the most frequent reason given for use of the pullout design (81 percent) and/or the inclass design (75 percent). o Compliance with Title I's funds allocation provision was the second most frequent reason (61 percent) given by districts for use of the pullout design and the third most frequent reason (45 percent) was "state Title I office has advised use of this design." o Past misconceptions about the supplement-not-supplant provision still exist and affect program design practices. Program design practices under Chapter 1 may be influenced by several factors: o Some districts may reexamine the extent of their use of pullouts in light of the Chapter 1 provision saying pullouts cannot be required to prove compliance with the supplement-not-supplant provision. o Use of the inclass design, which was already increasing under Title I, may accelerate in some districts if

PAGE 64

appropriations for Title !/Chapter 1 continue to decline. o Some districts may eventually understand better how to design projects that provide supplemental, rather than substituted, services as a result of the models in 47 Title I guidelines and Chapter l's draft of nonregulatory guidance document. o The cumulative impact of changes wrought by Chapter 1, and the fiscal situation may, over time, produce shifts in program design practices. (Gaffney & Schamber, 1982, pp. 11-12) These selected findings illustrated district program design practices, the reasons given for such practices and the interaction between program design practices and the supplement-not-supplant and excess cost provisions. Both the districts surveyed by mail and the Title I director interviewed on site were reported to be representative samples. The findings from these samples did not vary significantly by district size, nor did the size of a district's Title I budget correlate with the use of pull-out designs. However, the use of in-class designs significantly correlated with district size. According to the findings, when 30% of the mail questionnaire districts that used in-class design were broken down by district size, very large districts (47%) were far more likely to employ the in-class model for part or all of their program than small districts (28%) and medium districts (32%). Of the mail survey districts using in-class, 32% had increased their use of this design over a three-year

PAGE 65

48 period. Only 10% had decreased the in-class designs. However, no comparable net change for pull-out was reported, with 10% increase in use of design, and a 10% decrease also reported. Reasons given for decreasing use of pull-out design did not vary significantly by district size. However, the most frequent reason given for the decrease (74%) was teachers' or principals' recommendations. The next most frequently cited reason (51%) for all district sizes was the "informal assessment of program performance" (Gaffney & Schember, 1982, p. 16). The same reasons were given by respondents who increased the use of pull-out design (72%), citing recommendations by principals and leaders. Results of Title I evaluations were also an important reason for increasing pull-out use (63%), but a less important reason (43%) for decreasing pull-out use. It would seem from the results reported that districts more often changed to the pull-out design to improve program effectiveness as reported by the evaluations. However, the data confirmed by principal interviews in districts using a pull-out approach echoed a considerable concern regarding subject matter missed when Title I services were proviqed. Three-fifths of the principals reported having a policy on kinds of regular instruction students could or could not miss:

PAGE 66

40% could not miss reading 22% could not miss math 12% could not miss P.E. 16% could not miss any basic subject. 49 Of ongoing interest is the variety of reasons behind district choices. In contrast to the Levine and Stark study where districts converted to in-class designs based upo(l98l)n the 1977 Glass study, teachers and local administrators held strong views, not just in favor of their own designs, but against other approaches. A very important or somewhat important reason for design choice given by 92% of those surveyed was that the design was "educationally superior for all or part of the program." Of the two designs, 81% of those districts with pull-out programs made this claim as compared to an equally emphatic 75% of those with in-class. A variety of reasons and criticism for each program design were contrasted in the study, but the site visits using the pull-out design clearly indicated that one-third of the directors chose this approach because it was easier to administer. A smaller number, 10%, mentioned a concern about the supplement-notsupplant provision as a reason for using the pull-out design, while another 11% cited past compliance problems (Gaffney & Schember, p. 126). Therefore, it could be concluded that while data from the DPS supported the continuing difficulty in complying with the supplement-not-supplant provision, the cause

PAGE 67

50 of this problem seemed to be a continuing misconception of the requirement. It can be concluded from the Gaffney and Schember study that misconceptions about the supplement-not-supplant provision still existed in 1982 and affected program design practice.s. An analysis of three years of Title I applications from 18 districts in over a dozen states showed that district applications were inconclusive. There was not sufficient information demonstrating that Title I children would receive their fair share of state and locally funded instructional time. Setting and the relationship of program design to legal provisions remained unclear (Gaffney & Schember, 1982, p. 45). However, the study gave direction for further research: Further inquiry into the interaction of program design and the supplement-not supplant provision under Chapter 1 would be a useful way of continuing to examine implementation of the requirement that federally funded compensatory education provide supplemental rather than substituted services. (p. xii) The Gaffney and Schember study gave insight as to why the pull-out design continued to receive wide support despite documentation both nationally and locally by those highly critical of the typical pull-out method. An example was a case study district in which teachers and administrators who preferred the extended pull-out approach were also highly critical of the method. Many of the regular classroom teachers stated that they wanted full responsibility for their students. They did not know what the students were doing in the Title I

PAGE 68

51 class, and the pull-out model subjected the Title I students to conflicts between the philosophies of the Title I and the regular teacher. However, teacher conflicts in other districts using the in-class design were also cited. These teachers stated that two adults in the same classroom with a common in-class approach was distracting to students and resulted in classroom discipline and management problems. Another study in 1982 by Thomas L. Good criticized pull-out programs on the grounds that they often created difficult teaching/learning situations and likened them to instructional practices such as tracking and ability grouping. The Kimbrough and Hill study (1981) was cited. Findings stated that pull-out programs posed problems for students who received special instruction, as well as for regular teachers. The study again addressed the issue of supplanting or replacing: These investigators found that pull-out programs posed problems for students who received special assistance as well as for regular teachers. In some schools, children were out of classes for categorical programs so frequently that teachers had their total classes only 1 1/2 hours daily and therefore were unable to implement the state mandated curriculum. Fragmented instruction was especially a problem for Hispanic students because they qualified for so many special programs (six or seven daily!). Indeed, even though many Hispanic students had attended school for five years, they had received TIQ formal instruction in science or social studies. Special programs were replacin,, not supplementing, the core curriculum for many students. Because of scheduling problems created by multiple pull-outs, many districts allowed special categorical programs to replace core programs. Many lowachieving, disadvantaged students thus received only special instruction, though they were entitled to regular instruction in math and reading as well as supplementary instruction in those subjects. (Good, 1982,p. 46)

PAGE 69

52 Kimbrough and Hill (1981) also brought attention to confusion by many children due to conflicting approaches taken by special and regular teachers. In several cases they found teaching methods and materials used in special and regular classrooms were incompatible. Good added further criticism, pointing out that children who have the fewest time-management skills had to know when to leave class, when to return, and how to negotiate with their teachers about work missed. To darken the picture, children reentering the classroom, no matter how carefully attempted, often disrupted regular classroom activities. Considering teachers' general negative reactions to interruptions, the pupils could view themselves as causing the teachers' irritation. How, then, might a teacher react to a student who would seek to find out what work had been missed (Good, 1982, p. 49)? Good concluded further that the pulled-out student had less access to the social language and the social identity of hisjher classroom group. The student might well miss the chance to learn more refined knowledge about the implicit norms of behavior that the teacher and fellow students associate with certain classroom contexts. Hence the pull out arrangement works to virtually assure that the student will be deprived of valuable social knowledge about the classroom and lack of such knowledge makes it likely that the student will violate teacher and/or peer expectations. (Good, 1981, p. 47)

PAGE 70

53 Good argued that what a student learns is substantially affected by the school a student attends and the teachers and groups of students that he or she encounters. It would seem then, that conflicts among individuals, whether teacher to teacher, student to peers, or teacher to student are inherent in the pull-out design and are apt to affect the student's view of himself in a negative way, according to Kimbrough and Hill, and Good. They were in agreement with critics of pull-out, as cited in the Gaffney and Schamber study (1982, p. 47). Due largely to these and other critics, in-class programs were on the rise in 1983. Nevertheless, despite the growing incidence of in-class programs, three of four studies published in 1983 favored the pull-out design. The first of these studies, "Effects of Instructional Setting and Approach in Compensatory Education: A Statewide Analysis," evaluated data from Title I projects under the Hawaii State Department of Education. Effects of different instructional settings and approaches on the achievement of Title I students receiving remedial reading and mathematics instruction from 1978 to 1981 were examined. Project profiles were developed, based on descriptive statistics indicating enrollments, hours of instruction, costs, student achievement gains and other characteristics of the different Title I Projects within the state. A premise of the study stated that the adoption or adaptation of instructional settings is often a matter of

PAGE 71

54 logistical convenience or fiscal necessity rather than an outcome of careful deliberations of benefits and impact. Noting that negative perceptions had been expressed by some researchers indicating the lack of efficacy of the pull-out setting, there had been little direct evidence as found in the preceding studies discussed from the early 1980s. After reviewing the research, the authors had concluded that setting was not a clearly overwhelming variable in and of itself. They undertook to address the following research questions: 1. What are the effects of different instructional settings on the achievement of students participating in Chapter I projects? 2. What are the effects of different instructional approaches on achievement of students participating in Chapter I projects (Yap, 1983, p. 5)? Over a three-year period data were collected from evaluation reports for some 300 Chapter I projects. Yap stated that no experimental manipulation was used in the study. Three project settings were used: pull-out, in-class, and a combi-nation setting. The latter involved the provision of instruc-tional services in a combination of pull-out and regular classroom instruction. In general, for the three school years considered, 1979 to 1981, the findings were: (1) students in lower grades made greater Normal Curve Equivalent (NCE) score gains than those in higher grades;

PAGE 72

(2) smaller school and project sizes were associated with greater NCE gains; (3) students with lower pretest score_s made greater gains than those with higher pretest scores; (4) higher per pupil costs were associated with greater NCE gains; (5) absenteeism was inversely related to achievement gains; (6) the pull-out setting had a more favorable impact on achievement than other instructional settings; and (7) combined instructional approaches (e.g., a prescribed system supplemented by teacher-made materials) produced greater gains than an approach that used commercially packaged materials with a prescribed instructional system, or one that used both commerciallypackaged and teacher-made materials without a prescribed instructional system. (Yap, 1983, abstract) A majority of the reading programs for which information 55 was made available utilized the pull-out setting, while a small number provided instruction in the regular classroom. Some projects used a combination of settings. Results of analysis of covariance (with effects of pretest and cost partialed out) suggest_ that pull-out represented an effective setting for providing Chapter I services (Yap, 1983, p. 16). Yap concluded that students who received instruction in this setting performed as well as, if not better than, their counterparts in other settings, although the difference did riot reach significance at the .OS level. The combination approach was shown to be as effective as, if not more so than, other approaches. The overall analysis P!oduced a discernible trend indicating superiority of pull-out setting when grade level was not taken into consideration. The study led to similar conclusions found in previous studies such as the Glass and Smith study (1977) regarding class size and expenditures, finding class size and achievement

PAGE 73

56 to be related. Further, attendance rate, once socioeconomic factors were accounted for, was not found to be associated with achievement. Project hours were not found to be associated with achievement, and previous studies also held that mere length of the school day or class period does not necessarily influence student achievement. Yap concluded that the positive relationship depends on how the available time was used, not just the amount of time available, as did Slavin, 1981. Studies have found that reducing class size has beneficial effects on cognitive and affective outcomes and on the teaching process itself. Therefore, since the pull-out setting often means smaller class size and generally provides greater instructional support to students, it could be expected to produce greater achievement (Tobias, 1982; Yap, 1983, p. 2). The focus of a study by Vasquez-Nutall Associates (1982-1983), "ESEA Title I Program, Interim Report," concerned opinions of administrators and teachers. Data were collected for the first half of the school year through a combination of classroom visits, interviews and questionnaires. Three districts in Boston Public Schools included 37 elementary schools and six middle schools. Twenty-nine of the 54 elementary schools visited were reported. As in the Yap study, the majority of respondents in the elementary schools wanted to retain the pull-out instructional model. A substantial

PAGE 74

minority at secondary level stated a preference that Title I be offered as regularly scheduled instruction. These results were reported from a population of all building administrators, all Title I teachers and professionals. Also filling out questionnaires were twice as many non-Title I teachers as Title I teachers with non-Title I teachers selected by Title I coordinators. Table 1 represents responses to the question, "What are the two major strengths of the Title I program?" Table 1 "What Are the Two Major of the Title I Elementary School Staff Responses 1. Individualized instruction 2. Small group instruction 3. Improvement in student skills 4. Improvement in student self image 5. Materials 6. Assistance of paraprofessionals Title I Teachers N = 50 % 64 36 12 12 10 10 Paraprofs N = 29 % 34 30 14 14 School Admin. N 27 % 37 30 15 11 7 Non-Title I Teachers N 73 % 53 34 8 7 8 57

PAGE 75

Table 1 (continued) 7. Diagnostic and assessment capabilities of program 8. Competency of Title I teachers 9. Supplemental training in reading 10. Well-trained paraprofession-als Title I Teachers N = 50 % 8 Paraprofs N = 29 % 14 7 School Admin. N 27 % 22 7% Non-Title I Teachers N-73 % 5 5 26 58 From: Vasquez-Nutall Associates, Chapter 1 Elementary Reading Programs. Interim Evaluation Report (ERIC Document Reproduction Service No. Ed 238 968), 1982, p. 24. To the question, "Should Pull-out be the Major Program Model for Title I?" the following table illustrates that the elementary school staff wanted to keep the Title I program much as it was then (see Table 2). Administrators were somewhat more willing than teachers or paraprofessionals to respond that minor changes were needed. However, most of the changes that were suggested concerned scheduling, grouping of students, and selection of students rather than a different model (Vasquez-Nutall Associates, 1982,

PAGE 76

59 p. 34). The major complaint about the pull-out model came from classroom teachers claiming that the program was "disruptive" and that "it was too difficult for children to make up work they missed" in the regular classroom (Vasquez-Nutall Associates, 1982, p. 35). Vasquez-Nutall Associates (1982, 1983) also reported on an in-classroom pilot which was not well received. No advantages Table 2 Should Pull-Out Be the Major Program Model for Title I? N No Unsure Yes, Yes, Yes, No with vith vith resp. major mi. nor no changes changes changes ' ' ' Adm1.nistrators 27 7 4 41 52 T.I. teachers 50 22 76 2 NonT.I.teach. 73 4 7 22 67" Paras 29 3 21 76 From VasquezNutal1, 1982, p. 34. were found. The biggest disadvantage reported too much distraction to run two instructional programs at the same time in the same room. Seventy-three percent of the pilot staff preferred separate Chapter I rooms the next year and 95% of the nonpilot staff preferred to retain their separate rooms. Other disadvantages included inadequate space ("turf") struggles between teachers, limitations on materials and

PAGE 77

60 equipment, and the inability to sufficiently individualize Chapter I instruction. A survey by Mary Hayes (1983) revealed that most teachers on 200 attitude scales had negative attitudes toward in-class compensatory reading programs. Hayes attempted to replicate the results of a master's thesis by Marie Davis (1982) at Keen College, New Jersey (1982) in which she found that teachers surveyed preferred an in-class compensatory program to the pull-out program. In the Hayes study an opinionnaire of nine questions was administered to classroom teachers in grades 1 to 5 in the Elizabeth, New Jersey, Public School District. Results of the study were in opposition to the master's thesis. In the first conclusion that pull-out was favored over the in-class design, it was stated that the pull-out program was the lesser of two evils! The in-class program was too distracting. However, a student leaving or entering from the pull-out program, was also a disruption, as in the findings of Vasquez-Nutall Associates (1983). An overwhelming response in the comments section of the opinionnaire clearly demonstrated the importance of a teacher's attitude upon the learning environment of the child. Hayes included. the research on this topic: Another important issue is attitude, both the of the teacher and the student. When a teacher is aware of the potential of her pupils, it influences her attitude and the child's performance as well (Gordon and Wilkerson, 1966). A warm and accepting teacher who expects that children can learn will be effective. Thorsten R.

PAGE 78

61 Carlson (1972) states that reading achievement depends on the child's attitudes, information, and general background. (Hayes, 1983, p. 24) The Hayes study concurred with research showing a direct correlation between teacher attitudes and the attitudes of the student. To demonstrate student attitude toward pull-out design, the Hayes study presented evidence to show that a child responds more positively to a pull-out compensatory reading program than to in-class, which may result in reduced student embarrassment and fewer distractions. In favor of the in-class design, however, a significant number of teachers felt that the in-class compensatory reading program enabled more feedback on student progress and provided a better opportunity for communication between the classroom teacher and the remedial teacher. Further, 54% of those surveyed chose either the in-class mode or an alternative to either in-class or pull-out. When asked which compensatory program they would suggest if they had input into the administrative decision, close to half the sample suggested that the pull-out model should be implemented in their school district. Overall, findings showed a pattern favoring a pull-out program. The belief of teachers surveyed, clearly showed that pull-out programs provided the best learning environment for the remediation of the child (Hayes, 1983, p. 21).

PAGE 79

In one of the studies of classrom and project teachers, published in August, 1983, the Reading Strategy Project (RSP) rated the inclassroom project very highly. The inclass project was designed in Cleveland, Ohio, to assist upper ele mentary school pupils who had deficits in basic reading skills as reflected in performance on citywide reading tests. Participants in the project in the 1982 school year 62 served 3,978 public school students (grades 4) and 318 non public pupils, grades 1. According to Halasa (1983), participants increased their reading comprehension scores on the California Test of Basic Skills significantly more than would have been expected under Chapter I treatment conditions, and observed gains exceeded the Project criterion. Halasa concluded that the size of the gains was influenced both by the model used and by the grade level. While Halasa reported the success of the Reading Strategy Project in terms of improving pupils' basic reading skills through increased achievement gains, the very supportive attitude of the teachers may have been a contributing factor. In contrast, the pilot inclass project in Boston was not well received (VasquezNutall Associates, 1983). The Boston pilot model was implemented in nine schools, one in each district. It is of interest that principals and not teachers were asked to volunteer for this program. In return they were told that they would not lose any Chapter I teachers. Yet half of the

PAGE 80

principals expressed reservations about the program and well over one-third of all participating teachers were extremely dissatisfied. Many of the conflicts mentioned between Chapter I teachers and regular classroom teachers were not reported in the Halasa study. Although distraction was the major complicative factor, scheduling the activities to 63 accommodate another clas.s had been forced more often upon the regular classroom teacher than the Chapter I teacher. Ninetyfour percent of classroom teachers said they had to modify activities, usually avoiding whole class instruction. The Chapter I teachers who were most successful in coping with the shared classroom were those who showed unusually interesting and quiet activities. Yet, distraction continued to be a problem to more than 80% of all teachers. The success or failure of the in-classroom design, then, may be more a factor of teacher attitude, satisfaction or dissatisfaction, and may stem from their involvement in the decision as to program design. There were no advantages in the Boston in-class program, while achievement gains were a clear advantage in the Halasa study. However, the contrast in teacher attitudes in the two studies was most compelling. Therefore, in all four of the studies from 1983, teacher attitude was a factor. All but one study favored the pull-out design. However,

PAGE 81

the Halasa study did not describe achievement from a pull-out control group from which comparison could be derived. 64 Yet again in 1985 the cry came, "Let's Pull Out of the Pullout Program" (Taylor, 1985). Citing A Nation at Risk and John Goodlad in A Place Called School, Taylor again condemned. pull-out programs as disruptive. Interruptions decrease the amount of time teachers can devote to instruction, taking children away from regular class activities, and should be reduced. Nothing new so far, as this premise goes, but Taylor commented regarding effects on classroom teacher morale. She claimed resentments build over the school year due to the greater freedom of the support staff and differences in job responsibilities. In many schools, Taylor claimed support staff have a handful of students in a group, aides, and some have telephones. Classroom teachers accompany their students to field trips and assemblies. While the pull-out programs command a large share of the school budget, classroom teachers are not compensated for the extra work required by the pull-out programs. Taylor claimed that pull-outs create situations that discriminate against certain groups. Also, children know too well when they are labeled. And what are the effects on students? They may have to miss classes they like in the regular classroom while other children who need to practice responsibility escape the accountability of the regular

PAGE 82

65 classroom, declaring that they were not present when an assignment was given. Four solutions to the problem were offered. First, make pull-outs more cost effective by offering them before or after school. Second, greatly reduce the number. Third, employ paraprofessionals in place of professional support staff. Finally, a solution that would seem viable was to let each school be free to work out its own solution to problems involving remedial education. Taylor offered no research data or supporting studies. However, the fact that this article was published in Principal (Taylor, 1985) could have an effect on those administrators who were making decisions regarding program design. So, while negative and positive attitudes about both program designs from all arenas continued to fan the flames of controversy, administrators and researchers continued to look for a solution to the program design dilemma. Although there had been no evidence to support the detrimental effects of pull-out as claimed by Glass and Smith (1977), criticisms such as Taylor's continued throughout the 1980s. David Savage had asked in 1981, "What conclusions can we draw about the effectiveness of compensatory education programs?" He concluded there was no evidence that children in these special programs learn more than children-not exposed to these remedial services (Savage, 1987). Yet, Passow (1986) after reviewing the literature to 1986, had a strongly opposing view. It was evidenced since the

PAGE 83

66 1960s that efforts to provide compensatory programs and services had met with a variety of successes and failures. The 30,000 Title I projects, for example, reflected multiple goals and multiple treatments which are not easily converted into overarching objectives or successful program models. This multiplicity has complicated any effort to evaluate compensatory programs, but twenty years of compensation programs have taught us that the immediate goal should be to raise the achievement levels of disadvantaged students. (Passow, 1986, abstract) He concluded that one of eight areas in the lessons of compensatory education which justifies increased expenditures is the pull-out design vis-a-vis regular classroom instruction. Further programs in the 1980s did.look at raising achievement as an immediate goal. For example, a profile of Oregon's Chapter I programs during the 1982-83 school year (Oregon State Department of Education, 1984) looked at the percentage of all Oregon students who participated in Chapter I programs. Twenty-one percent of these belonged to an ethnic minority. The pull-out design received most of these Chapter I students for instruction in reading (59%), math skills (24%), language arts (17%), and/or other subjects. In general, the Chapter I programs were effective in improving the achievement level of participating students. Further, Oregon's results compared favorably with other states, especially in reading, at most grade levels. Yap reminded the readers again that in 1985, while there was little direct evidence regarding the lack of efficacy of the pull-out setting, negative perceptions had been expressed by

PAGE 84

67 some researchers. However, results from the Yap reanalysis in 1985 concluded that the pull-out setting produced the highest achievement gains and the most favorable cost effectiveness ratio (Yap, 1985). Yet, the other negative perceptions of researchers were not to be ignored. A study using longitudinal data and a cross-lagged panel correlation ruled out a causal relationship between self-concept and achievement (Pottebaum, Keith, & Ehly, 1985). Byrne (1984) had concluded that selfconcept is a multidimensional construct having one general factor and several specific facets, one of which is academic self-concept. Pottebaum et al. (1985) suggested that achievements were caused by a third variable and concurred with other research suggesting that social class and ability may be predominant over both academic achievement and self-concept. The sample in the Pottebaum et al. study was huge and nationally representative, but sampled only high school students. However, findings were consistent with much of the well done research in this area (Caslyn, 1974). The important implication of the research was that educators should not focus on the development of general self-concept as a means to improve academic achievement. The two domains would best be dealt with in an educational setting as two and unrelated constructs (Pottebaum et al., 1985, p. 143). Therefore, while research has demonstrated that pull-out programs produce higher achievement gains than other designs (e.g., Yap, 1985), the

PAGE 85

self-concept factor required further research when comparing the two program designs, pull-out and in-class. 68 Passow gave clear direction to compensatory education, stating that the immediate goal should be to raise the achievement levels of disadvantaged students. Pull-out programs studied, in the majority of cases, had accomplished this goal. However, researchers and legislators were not satisfied that achievement levels were the only target of Chapter I funding. Two studies described the mood of the nation regarding the desired direction of compensatory education. First, Fretchling and Others [sic] (1985) summarized the current literature as quite consistent in suggesting instructional value in that there appeared to be direct and indirect effects of the pull-out approach that impede rather than foster the goals of schools. Fretchling and Others cited loss of instructional time, resegregation and social isolation among the outcomes believed to result from the pull-out setting. "Like ability grouping, its merits as an instructional approach have not been affirmed for low achieving students" (Fretchling and Others, 1985, p. 66). Based on the findings aimed at increasing minority achievement in Montgomery County Schools Fretchling et al. recommended class size reduction to 15 or less, student team learning and teacher training programs and further analysis of the variables shown to enhance low-achieving students' performance.

PAGE 86

69 While school districts and educators were calling for another look at variables affecting students in compensatory education, legislators and evaluators were doing the same. Kennedy (1985) reported on the National Assessment of the Chapter I Program. In the NIE mandate to study the reauthorization of Chapter I, of the seven items of information requested by Congress, six concerned implementation and only one concerned effectiveness: o The recipients o The services o The background and training of teachers and staff o Coordination of services o How funds are allocated among schools o Effectiveness o How certain key provisions are implemented. The most frequently asked questions by legislatures concerned whether and how a proposed program could be implemented. Thus, the 1987 reauthorization hearings would focus on four implementation questions: administration, targeting, services, and program design. Looking at program design certainly did occur often in 1985. Researchers, school districts and other evaluators sought to develop the in-class program design as an alternative to the pull-out model.

PAGE 87

70 To consider a series of studies that aimed at replacing pull-out compensatory program designs with the in-class model, the question of program design affecting outcomes was considered. In the first study under consideration, the 1984 study by Fretchling and others, the report concluded that it is not possible to draw linkages between program features and student outcomes. This study does demonstrate, however, that White, Black and Hispanic students all seemed to benefit from the in class design, implying that socioeconomic status (SES) and/or ethnicity may not need to be a factor when considering program design. In the Fretchling study, schoolwide in-class programs in Montgomery public schools, Rockville, Maryland, were provided which eliminated and replaced pull-out instruction, providing a pupil/teacher ratio of 15:1 or less. The lower ratio was accomplished through the use of Chapter I and supplemented district funds. Fretchling and Others reviewed the literature on special programs and called for cautious optimism regarding findings on factors of special programs which have proven valuable in enhancing the performance of low achieving students, at least "in the short term": o reductions in class size to 15:1 or less o use of student team learning and of teacher training programs such as Teacher Expectations and Student Achievement o mastery learning programs

PAGE 88

71 According to research studies on class size dating back to 1900, the effects of class size on achievement have been inconclusive despite common sense suggestions that small class size leads to greater academic achievement (Swan and Others [wic], 1985). The Fretchling and Others study indicated that reduced class size was a factor leading to the success of the in-class program. However, Swan and Others warned against this approach due to financial costs and political consequences of reducing class size, which have generally prohibited the adoption of class size reduction alternatives. This warning was issued despite overwhelming evidence of gains in basic reading and mathematics scores for small size classes (18 students or fewer) in the 1984-85 school year. Students as subjects in 13 studies were first grade regular education students in three elementary schools in Princeton, Indiana. Therefore, it would seem that although reduced class size has been demonstrated to have positive outcomes in both regular and specific programs, there are cautions prohibiting its adaptation--financial and political. Class size reduction was a significant factor in the Fretchling et al. study, with an emphasis on more efficient use of time, decreased discipline problems, better use of class time, an increased closeness between teachers and students, and higher teacher morale. These factors were reported to influence achievement gains in the Texas study. The program in Texas was one of three types of

PAGE 89

72 program interventions and described as an in-class model. A pull-out program entitled Project Umbrella aimed at enhancing reading achievement was also described. The low achieving students served under Project Umbrella also appeared to be making gains. In the third alternative model, an Academic Goals Program, the performance of Black, Asian, Hispanic and White students had improved. Also, the difference in performance between schools serving higher and lower concentrations of minority students had decreased. Fretchling concluded, however, in the programs described above that linkages between specific practices and outcomes could not be made (Fretchling et al., 1984). The Fretchling (1984) study would seem to indicate that the in-class model could provide compensatory education that would benefit all social groups. However, the reading achievement of Blacks, Asians, Hispanics and Whites also showed significant gains in a pull-out model. Yet, Rodriguez (1985), looking at achievement of Hispanic students, recommended that ability/achievement grouping or tracking of all students be eliminated, calling for assistance to be provided in the regular classroom with extra or pull-out instruction available. Rodriguez called for heterogenous grouping in the regular classroom, and pull-out provided for extra instruction. Pull out was to be avoided if it sustained membership in low achievement groups. Imperatives to effective education were

PAGE 90

73 high expectations for academic and social behavior of Hispanic students in particular, and all students in general. While Rodriguez did not favor pull-out as a long-term method of remediation, he did not favor the in-class program design if it assigned Hispanics to low groups within the classroom. His views, which directly disagreed with the Fretchling findings, can be summarized from a monograph considering the problem of low achievement among Hispanic youth in elementary and secondary schools (approximately 75% score below grade on California Achievement Tests (CAT) given after third grade and remain underachievers throughout their school careers). Organizationally, low expectations are implicitly communicated in such ways as in assigning students to low cluster groups, low tracks_, or low achievement groups under the mistaken belief that structure will provide for individual differences. But such assignments turn out to be a self-fulfilling prophecy. Goodlad reported that students performing in the bottom group show up less well on indices of personal and social adjustment and are students most likely to have high expectations for further failure. He also found that in achievement/ability grouping the upper group experiences a richer curr.iculum than low or middle groups. But when heterogeneously grouped, such classes were more similar to high than to low track classes in teaching methods, what was studied, and teacher-student interactions. The practice of subgrouping within classrooms has a common sense appeal. However, the long-term effects were not anticipated or even acknowledged by most. The fact is that instead of meeting individual needs as intended, it more likely tends to decimate academic initiative among those assigned to low groups. (Rodriguez, 1985, p. 5) Richard Allington (in press) conducted a study on the nature of remedial reading which also implied a lack of high teacher expectations. The study examined the focus of remedial

PAGE 91

instruction and the relationships of this instruction to the classroom reading program. Results showed that individual teachers used little instructional variety in a model which used pull-out instruction whereby students from five different first through fourth grade classrooms were pulled out for remedial instruction. The fifth program was an in-class model in which instruction was delivered by a Chapter I aide. Allington and Others concluded that there was little evidence of the use of clear cut goals or of monitoring of student advancement toward goals. It would be difficult to assume high teacher expectations without clear-cut goals to define expectations. 74 Factors were defined in the Allington study that were not correlated to program design, but did indicate that student achievement could be improved in either remedial setting, in class or pull-out. Observers selected four schools in four districts with classroom and remedial teachers volunteering to participate. Twenty-seven remedial students were observed in five separate classrooms working with five different classroom and remedial teachers (the in-class program provided instruction by an aide in the regular classroom). The five classroom teachers were experienced, 7 to 20 years of experience with permanent certifications, a bachelor's degree, and some graduate work. The remedial teachers and aides tended to be less experienced generally, with 1 to 10 years in

PAGE 92

teaching and had taught remedial reading for 1 to 7 years. All had a bachelor's degree, four had New York State Reading Certification, and all had completed some graduate work in reading. The factor of teacher experience seemed to indicate that reading teachers were only somewhat less experienced. 75 The Chapter I teachers reported using a greater variety of methods and materials than did classroom teachers. However, observers did not witness much instructional variety within individual teachers. Teacher methods were not defined. Allington reported not knowing of an optimal number of methods and/or materials that remedial teachers should possess, but he believed that these resources should construct a well-defined program using clear goals in an individualized program designed to improve students' reading and evidence of monitoring toward these goals. He saw no evidence of these factors of individualized instruction, but rather a general "allpurpose" approach to remediation (Allington and Others, 1985, p. 10). Johnston, Allington and Afflerbach (1985) argued that when classroom and remedial instruction were not congruent, less than optimal achievement will be the result. Allington et al. found relatively few instances of curricular congruence, no matter how defined. Similar materials were employed in both instructional settings in less than one quarter of the observations while the same teaching strategies were emphasized in both instructional settings in less than one lesson in 20.

PAGE 93

All teachers, however, stressed the need for communication between classroom and remedial teachers, and most indicated more communication was needed in order to plan optimal instruction for target students in both settings. Teachers indicated a lack of clear understanding of the instructional focus in the other setting. Repeating research concluded that the quantity of instruction received by Chapter I students was not greater than the quantity received by non-Chapter I students, Allington did further study through simultaneous observa-76 tion. The data supported previous research and suggested that with additional travel time to pull-out sites, social meeting time and so on, Chapter I may actually have less time for instructional activities. Remedial students most often missed independent seatwork time while away from the regular classroom. Allington proposed that lack of quantity of instruction could be offset by the quality of instruction provided by Chapter I. Time to learn and time on task were factors indicating quality of instruction to improve achievement. The instructional focus in remedial reading should be greater time allocations, time on task and related to the kinds of tasks the student is asked to perform. Students learn what they practice. The observations of this study found that roughly onethird of student time was spent in direct reading activities,

PAGE 94

one third in indirect reading and one third in management, waiting, out-of-room activities and other non-academic activities. Allington et al. concluded that quality of reading instruction, as well as quantity, did not appear optimal in the Chapter I programs. However, remedial programs varied widely; Allington et al. felt that he captured the variation. Each of the programs offered needed and useful services, but each needed improvement. The improvement, Allington et al. summarized, most often necessarily stems from complete program redesign, not just greater effort by individual teachers. Higher teacher expectation of student behavior through clear goals and strict monitoring would not suffice. Although it would be an improvement, another look at program design was required. 77 Thomas Neumann agreed with Allington and Others that remedial reading programs were as varied as the students they served in an article on "Class Remedial Instruction: An Alternative to Pull-out Programs" (French, 1985). Chapter I programs, reading laboratories, clinics, separate reading classes, and programs all aimed to provide remedial reading instruction beyond what the regular classroom teacher can provide. Further, much data exists, according to Neumann, to support the effectiveness and advantage of each of these types of reading classes. The disadvantage was restated that despite the documented success of these programs, the disadvantage

PAGE 95

remained that the student must be removed from the regular classroom to receive special help since pull-out remained the norm in 1985. 78 Extending the findings of Allington and Others, Neumann maintained that the goal was not only to improve reading skills, but to transfer skills taught in the "reading room" to the content of the classroom. "Unfortunately this goal is not always met because the separation of the student from the content room also means the separation of the reading teacher from the content teacher" (French, 1985, p. 26). Since the desired objective, close consultation between the reading teacher and the content teacher, was infrequent or nonexistent, a lack of knowledge about what was taught in the reading room is a common complaint of content teachers. To meet the above-stated goal, the New London School, in the Wisconsin School District, developed an in-classroom remedial reading program to provide services to students in grades 4, 5 and 6. The program consisted of two remedial reading teachers to work with teachers and students in the district's four elementary schools. Reading teachers met daily with students selected by teacher referral and diagnostic testing. Reading teachers traveled from room to room and provided instruction from a chair at the student's desk or a small table at the back of the classroom. Using one.-to-one instruction, the student's content materials and texts

PAGE 96

79 frequently became the reading text. Emphasis was placed on viewing reading as a means for learning. This emphasis seems a logical extension of the reminder of Allington and Others that students learn what they practice. Neumann (1985) summarized the success of this program: Although the program is still being studied to fully measure its effectiveness, comments from the content teachers indicate an overwhelming approval of this in-class approach. All agree that they are definitely kept informed about what help their students are receiving from the reading professional. As one fifth grade teacher wrote, help is provided "without any interference or interruption of the regular class in session. Reading instruction is given with the materials and skills the child used that day. This provides immediate clarification and support for what the child missed earlier and, therefore, the child is more ready to meet the next day's class." As with any educational program, there are several disadvantages to this in-class remedial approach. First, because the reading teachers move from room to room, (and in New London also from building to building), the reading teacher must be highly organized and flexible enough to work well in several teaching environments. A reading professional who needs the security of the self-contained classroom would encounter numerous problems. Second, because the reading teacher sees the content teachers daily and works directly within their classroom, he/she must possess the personality and diplomacy for working closely with colleagues. Third, and perhaps most significantly, a program of this nature must be locally funded. Because state or federal aides do not exist for such a remedial service, many local school boards may be reluctant to pay for the complete cost of the program. (Neumann, p. 27) Weighing the advantages and disadvantages, Neuman concluded that teachers, students, and parents affected learned that advantages outweighed the disadvantages. Content teachers could see first hand what help the remedial reading teacher provided. Although no achievement data were provided, students were reported to have learned that reading has meaning for all

PAGE 97

80 subjects, not just for reading class. Again, the conclusion is supported that attitudes of teachers and students toward a particular program design may be of primary importance when considering the degree of success of the program. Neumann declared that the trend in 1985 was to promote reading in the content areas along with individual attention in reading instruction. Thus, the in-class remedial approach could prove to be the most successful, declared Neumann. Bean and Eichelberger (1985) also thought the trend towards in-class programs was an important one that would have a great deal of impact on reading achievement of students. Recognizing that there were limited empirical data about the effectiveness of such programs or their strengths and weaknesses, these researchers undertook a study, changing the role of reading specialists from pull-out to in-class programs. The opportunity to study the role of the specialist in an in-class program arose in a large school district when the switch from pull-out to in-class provided data to compare teacher perceptions of the two program types. Questionnaires were distributed near the end of the first year of the in-class program and returned by 74 reading specialists and 411 classroom teachers from 105 elementary, middle, and secondary schools. Ninety percent of the questionnaires were also returned by teachers and reading specialists in 22 parochial schools. Thirty descriptors in the questionnaire of the reading

PAGE 98

81 specialist's role were rated on a scale of the importance of function or activity from 0 to 3. Four categories of descriptors included student-related functions, resource role functions, administrative functions and curriculum development. In addition, the questionnaire included open-ended questions requesting views about the most and least important roles for reading specialists and their perceptions about the strengths and weaknesses of the in-class program and follow-up interviews. School administrators selected representative schools in which to conduct the interviews. Results interpreted from consistency of responses and the high percentage of questionnaires gave credence to the following conclusions: 1. Reading specialists changed the manner in which they functioned when they worked in the classroom. Specialists focused more on reinforcing skills taught by the classroom teachers than on diagnosing skill needs. The instructional program then was more focused, despite some concern about the appropriateness of teaching strategies and materials. There was also more emphasis on giving feedback to and working with teachers. 2. Teachers and specialists both valued the specialists' role as it is currently defined, that is, an instructional position in which specialists focus on working with students with reading problems. Neither group saw the reading specialist primarily as a resource to the classroom teacher, although both groups, valued having interaction about specific children. 3. Problems in instituting an in-classroom program fell into two categories. Conceptually, both specialists and teachers had difficulty with two instructors in one room, and the problem of leadership, or control, became a real issue. The notion of teaming did not appear to be easy to accept. 4. Program implementation also created problems, especially in the areas of definition of role functions

PAGE 99

for teachers and specialists, scheduling, transportation of materials, and use of space. 82 5. Because the in-class program has potential for developing a more consistent, focused program for students and for increasing the efficiency and effectiveness of instruction, we support the notion. However, an in-class program is apparently not easy to institute . . Further, in-class programs may require alterations in the training of reading specialists. Continued investigation of the roles of the specialists, ways in which they are effective, and the relationship between those various roles and student learning must accompany the implementation of a new role for the reading specialist. (Bean & Eichelberger, 1985, p. 652) Finally, Bean and Eichelberger discussed recommendations for preparation of staff, degree of implementation, and instructional concerns, including a schedule for joint planning. In regard to degree of implementation, Bean recommended a most compelling alternative. Although the in-class program was reported to have many advantages for students, Bean suggested that some balance between an in-class program and pull-out program might be useful. Examples included specialists needing the flexibility to conduct diagnosis and student benefits derived from intensive one to one or small group instruction away from the classroom. Seven other in-class alternative approaches were described in a study by Harpring (1985), "Inclass Alternatives to Traditional Chapter I Pullout Programs." The alternative in-class programs which evolved over time were developed to overcome problems such as classroom disruption, scheduling difficulties, student movement disturbances, fragmentation of instruction, loss of time and the stigmatization of Chapter I

PAGE 100

students. Although there was little evidence that changing to in-class caused a drastic reduction in these problem areas, this study made efforts to take a close look at districts seeking alternatives to the pull-out model to solve the problems. Harpring reported that although some problems were alleviated, others surfaced. 83 Seven programs were selected in Phase III of the District Practices Study (Advanced Technology 1983) after extensive pre-site screening, document reviews, assurances that programs had operated for at least 12 months, were supported by those affected (administrators, teachers, parents and students) and conformed to Chapter I level requirements. The seven programs included two replacement models, two aid-based delivery service designs, an in-class reading program for the secondary level, an in-class elementary level program, an in-class laboratory model, and a traditional in-class program which evolved over time. There appeared to be few problems for the districts with the aide-based Chapter I in-class design. The aides were highly trained, over one-third having college degrees, extensive training, and a lead aide or experienced aides to help or train others. A lead Chapter I teacher monitored the program, but the role is primarily administrative. The aides traveled from classroom to classroom with materials in a canvas bag that were not readily available to non-Chapter I teachers or students.

PAGE 101

84 Space was defined and provided within the classroom. The mobility of the aides was said to weaken the tendency for principals or teachers to become possessive about particular aides. However, this design was not totally in-classroom. Students were pulled out of the classroom from time to time to reduce noise or to improve the learning environment. Consideration was given to older children in the fifth and sixth grades who sometimes felt self-conscious about receiving remedial help in the classroom setting. The question remained whether or not the in-class approach improved self-concept or reduced stigmatization. Design two developed in a district where resource teachers at one time provided services, but numbers of eligible students increased and time could not be found to help all identified as needy. An approach developed featuring aides who were in open classrooms with added reinforcements of skills provided by a master teacher in a laboratory setting. Activities of the aides were directed by classroom teachers who consulted with the (lab) resource teacher regarding the specific remediation of skills to be provided in the pull-out laboratory setting. In order to ensure effective working relationships, this district: o scheduled commitment meetings of the total staff in each school to agree upon rules for teacher/aide and pupil conduct that would be applied consistently o involved the teachers and aides in "leadership circles" --open-classroom level planning groups and building level planning groups.

PAGE 102

o maintained detailed records on all students, usually in the form of folders showing where the student stands regarding the elements of the curriculum. (Harpring, 1985, p. 15). The replacement models discussed were alternative approaches that eliminated the need for students to leave their regular classroom, except that their regular classroom was composed entirely of Chapter I students (approximately 20), a Chapter I teacher, a regular education teacher, and an aide. The class 85 was further divided into two groups for reading, mathematics and language arts. This design was said to be a pure replacement design involving a totally redesigned educational program for a full day, as Allington seemed to recommend. The second type of replacement model involved team teaching for two-hour periods. Again, all students in the class were eligible for Title I. Each day the Chapter I aide went into the regular classroom for a two-hour language arts period while teaming with the regular classroom teacher. The same concept provided math instruction during a designated two-hour period. The teachers divided the class according to ability levels and the Chapter I teacher was responsible for grading Chapter I students. Students could change groups as they demonstrated progress. No information was provided regarding the self-concept of students who were grouped according to ability and grouped as a special class within the regular school setting. It was recommended, however, that for replacement models to be successful, the philosophy and teaching styles of the

PAGE 103

86 instructional team had to be compatible and adequate planning time also had to be provided. There were variations among districts in replacement model design. In general, however, districts contributed teaching time rather than money to the project. Problems encountered by one district were finding enough space for the smaller classes, finding adequate planning time, and ensuring that everyone understood the replacement model. Chapter I funds took care of Chapter I teachers and aide salaries and materials. Both the in-class elementary-level program and the in-class laboratory evolved in the same district. The Chapter I teachers and/or paraprofessionals worked in the classroom under the guidance of the regular classroom teacher who was responsible for instructional planning and who graded the students. A clear understanding of the respective roles was said to be critical to the success of the program. Cooperative efforts to keep the noise level down had to be maintained. Chapter I teachers had to work only with Chapter I participants. The necessity for regular meetings of teachers/Chapter I aides was stressed. The in-class laboratory model required movement of both the regular teacher and the entire class to a laboratory facility. The Chapter I students, teachers and aide were present along with the 'regular classroom teacher and students. Students were assigned to small groups in the classroom. There was an expectation that the classroom teacher received training in the

PAGE 104

87 use of a variety of equipment and materials, the diagnostic approach, and learning contracts. Chapter I teachers and aides worked exclusively with the Chapter I students from each class. No achievement data were presented. However, scheduling time for the laboratory was a problem here as in other program described. Variations from school to school existed for the traditional in-class design. The dominant design, however, used instructional aides to reinforce basic skills instruction in elementary schools. The methods and approaches used by the aides were diverse, but one-on-one tutoring and working with small groups were common. Space for aides was a table or desk in each classroom for a work area. Individual work folders for each student were common. Aides had usually been with the district for several years so were familiar with the classroom teachers, their responsibilities and materials used. In a district with a large secondary Chapter I program, the Chapter I teacher essentially spent approximately one period per week in each of the content areas designated to receive services. The teacher would work with an individual student, a small group, or on occasion present a lesson to the entire class. In some instances the Chapter I teacher functioned more as a resource person, reviewing texts for readability, selecting instructional materials, preteaching vocabulary and developing study guides. Among the problems which had to be overcome with

PAGE 105

88 this model were finding time to coordinate services and getting regular teachers to accept a second teacher in the classroom. Among the critical elements for successful implementation of all the in-class strategy described above, as identified by district Chapter I staff, were retention by the classroom teacher for instructional planning, diagnosis, prescription, introduction of new material, and student evaluation. In her closing comments, Harpring (1985) summarized the value of implementing in-class strategies as alternatives to the pull-out programs: Several alternatives to pullout programs have been presented. That some are not dramatically different from others is evident. All have required forethought and planning and continued dedication on the part of the individuals involved in order to survive. For some school districts, the benefits may not be worth the effort. Others may believe the pullout program is still the most workable for them.. As one teacher put it, "I like having students pulled out; those remaining in my classroom benefit for a period of time too." Districts contemplating an alternative approach are advised to consider a number of factors including numbers of students requiring services, costs of additional teachers and/or aides, space and materials. Each alternative model calls for inservicing for Chapter 1 personnel as well as school district personnel. Ensuring compliance with federal guidelines can be more difficult if there are not clear policies on roles and responsibilities. Monitoring and evaluation procedures may also require revisions in current practices. As more and more states develop state remediation programs and also attempt to implement the characteristics of effective schools as defined by research, a Chapter 1 approach which rejects fragmentation becomes increasingly attractive. (p. 11) In a recent Phi Delta Kappan article, Robert E. Slavin (1987) projected that the future of Chapter I was presently

PAGE 106

89 based on a growing consensus that Chapter I works. For the sum of about $500 per student the reading achievement of the lowest-performing minority students has increased significantly in recent years, according to the National Assessment of Education. Slavin summed the present conditions of Chapter I: The School Improvement Act of 1987 (H.R.S) has reauthorized Chapter I. It states that 'Chapter I is the bedrock on which federal aid to elementary and secondary schools has been built. (p. 110) Despite the vote of confidence in the effectiveness of traditional Chapter I programs in the $3.9 billion annually authorized by the House, Slavin stated that traditional Chapter I programs are not adequate for the job they are supposed to do. Slavin's main findings supported the hypothesis that there are no differences in program design based on setting, achievement, and effectiveness. 1. doesn't matter. For more than a decade, researchers have criticized the use of Chapter I pull-out programs. Consequently, an increasing number of schools (though still a minority) have begun to use in-class models, in which a Chapter 1 teacher or aide works in the classroom along with the regular teacher. However, research that compares pull-out and in-class models has found few differences. It seems not to matter where Chapter 1 services are provided; what matters is the program itself. 2. Traditional diagnostic-prescriptive pull-outs make little difference. The typical Chapter 1 pull-out or in-class program makes little lasting difference in student achievement. For example, a study in Oklahoma City found that in comparison to similar but unserved students, Chapter 1 students served in a program using a traditional pull-out model gained only two percentage points more than unserved students in reading and three in

PAGE 107

math. Even smaller differences between Chapter 1 and similar unserved students were found in two studies in Baltimore. 90 In our search for effective Chapter 1 programs, we wrote to every one of 116 Chapter 1 programs identified as "exemplary" by the U.S. Department of Education. All the programs that sent us multi-year data showed significant gains in percentile ranks for fall-to-spring testing, but by the following fall theseogains had disappeared in almost every case. 3. Effective programs differ markedly from traditional models. With few exceptions, the programs we located that presented consistent and convincing evidence of instructional effectiveness looked completely different from traditional pull-out or in-class Chapter 1 models. In fact, the majority of the most effect and most costeffective models could not be completely funded in most Chapter 1 schools, either under existing Chapter 1 legislation or under the new reauthorization. (Slavin, 1987, p. 111) Cost continues to be a major factor in a time when school districts are looking to the states for new sources of funding. Districts with low income student families depend on Chapter I as a major source of federal funding. Other program models involve greater cost and would cause districts to be ineligible for Chapter I funds. Alternative programs are classroom designs while Chapter I funds can service only students eligible for Chapter I due to the factors of low income and poor achievement within a Chapter !-eligible school. Changing traditional program designs on a national scale would require extensive changes in legislation, funding, and practice. The probability of such changes in the near future is not great. Therefore, the question of a favorable program design is still a matter of current research. Which program design, pull-out,

PAGE 108

in-class, or a combination model must be a decision based on research data (Slavin, 1987). 91 In the first finding of the Slavin article, "setting doesn't matter," Slavin indicated that the rise of in-classroom programs was due to the past decade or more of continued criticisms of pull-out programs. Summary of the Literature Chapter I began as Title I, the first federally funded compensatory education effort, when the President signed into law the Elementary and Secondary Education Act (ESEA) in 1965. Through the '70s the federal government provided more specific directions on how Title I funds were to be spent. With direction and additional financial support from Congress, Title I programs increased in quality and sophistication while program outcomes became increasingly positive. Reenactment of Title I as Chapter I of the Education Consolidation and Improvement Act of 1981 (ECIA) reduced many of the requirements of Title I that had caused the pull-out program design to be the predominant serVice model for delivering Chapter I services. The first requirement that had provided impetus for pull-out designs was the supplement-not-supplant requirement for Chapter I services, which did not supplant regular education programming. The second requirement, the excess costs provision of Title I programs, provided that Title

PAGE 109

92 I funds could be used only to fund the excess costs of Title I programs and projects. In the summer of 1981, Congress repealed the excess costs provisions and a new provision was added. Pull-out projects could not be required to show compliance with the new ECIA. Further, Congress did not require any particular educational strategies, and also went so far as to direct the Office of Education to tell administrators, through regulations, how to operate in-class and pull-out programs. With these new directions for program designs, two results occurred. Pull-out programs were criticized and in-class pro-grams were tried as alternatives. In a study of pull-out pro-grams, the following criticisms were made by Glass and Smith (1977): 1. Pulling Title I eligible pupils out of regular classrooms for compensatory instruction is virtually universal. 2. The "pull out" procedure per se has no clear academic or social benefits and may, in fact, be detrimental to pupils' progress and adjustment to school. 3. The "pull out" procedure is used by schools niore to satisfy Title I regulations than because it is judged by teachers to bea sensible and beneficial plan. (p. 46) However, Glass and Smith surveyed neither Chapter I teachers nor coordinators before drawing this conclusion. In fact, of the 23 persons interviewed by Glass and Smith, only three were employed by public schools, titles unknown. This study has utilized the chapter of the Glass and Smith study on its nature and effects of "pull-out" programs to form the basis of the Teachers and Coordinators Survey used in this

PAGE 110

93 study (see Appendix B). Criticisms contained in the Glass and Smith study were put into question form to elicit response from teachers and coordinators, the Chapter I service providers were not interviewed in the Giass and Smith study. Incidence of in-class program design models increased following the criticisms of pull-out programs, beginning with the criticisms of Glass and Smith (1977) in the summary of the literature. As in-class programs developed and were evaluated, a review of the literature shows that both program designs, inclass and pull-out programs, demonstrated positive outcomes and detrimental effects to the delivery of compensatory education services. Tables 3 and 4 summarize the positive and negative effects of these program designs as reported in studies of Chapter I designs.

PAGE 111

Table 3 Studies of Pull-Out Designs Positive Effects, Benefits Allington and Others 1985 French. 1985 Fretchling and Others. 1984 Whites, Blacks, and Hispanics all gained academically. Gaffney and Schember. 1982 Title I evaluation and informal assessment important reason for increasing use of pull out. A third of directors chose the program due to ease of administration. Negative Effects Little instructional variety. Little curricular congruence. 94 Need for communication between teachers. Teachers indicated a lack of understanding of the instructional focus in the other setting. Less time for instructional activities due to travel to pull out site. Missed independent seatwork. Only 1/3 of time spent in direct reading and activities. Close consultation is infrequent or non-existent. Loss of instructional time. Resegregation. Social isolation. No correlation with district size and pull out Subject matter missed while pulled out. Many of the regular teachers wanted full responsibility for their students -did not know what they did in Title I. Conflicts of philosophies between regular and Chapter I teachers.

PAGE 112

Table 3 (continued) Positive Effects, Benefits Good 1982 Hayes. 1983 More positive student response than to in-class design, resulting in -reduced embarrassment for students fewer distractions The best learning environment for remediation. Povner and Others. 1981 More individualized instruction for non-participants while served students received compensatory instruction. More appropriate instruction for non-participants. Improved non-participant attitudes toward school, self, and other students. Negative Effects Students miss class so frequently unable to implement state-mandated curriculum. Special categorical programs replaced core programs. Confusion of students due to conflicting approaches. Teaching materials incompatible in many cases. 95 Students with poorest time management skills with schedule demands and make-up requirements. Reentry disrupts regular classroom activity. Misses knowledge regarding implicit norms of classroom behavior. Conflicts with students and teachers inherent and therefore affect student self-concept in a negative way. Non-participants suffered from not being able to receive benefits of CE programs. Principals have positive attitudes toward CE programs due to decisions regarding design and implementation.

PAGE 113

Table 3 (continued) Positive Effects, Benefits Poyner and Others 1981 (continued) Regular teachers had positive attitudes toward CE programs. Issue of stigmatization was of little importance in the opinions of principals, teachers, and students. Vasguez-Nutall Associates, 1982 Yap, 1983 A more favorable impact on achievement than other instructional settings. Effective setting with students performing as well or better than in other settings. Superior when grade level not taken into account. Smaller class size produces greater achievement and greater instructional support to students. Negative Effects Too disruptive. Too difficult for children to make up missed work from the regular classroom. 96

PAGE 114

Table 3 (continued) Positive Effects, Benefits Yap 1985 Oregon results, most commonly used design. Highest achievement gains. Most favorable cost effectiveness ratio. 97 Negative Effects

PAGE 115

Table 4 Studies of In-Class Design Positive Effects, Benefits Allington and Others, 1985 Bean and Eichelberger 1985 Reinforcement of classroom skills by reading specialist. More feedback to classroom teachers. Specialist's role as an instructional position. Potential for developing a more consistent, focused program. Potential for increasing efficacy and effectiveness of instruction. Fretchling and Others, 1984 Blacks, and Hispanics all seemed to benefit. Gaffney and Schember, 1982 Use of in-class may increase if appropriations continue to decline. High correlation with .district size. Larger districts utilize in-class. 758 Halasa 1983 Significant reading comprehension gains influenced by: the model used the grade level supportive attitude of teacher.s Negative Effects Lack of curricular congruence. Need for communication between teachers. Conceptually difficult with two instructors in one room. Leadership or control a real issue. Notion of teaming not easy to accept. Program implementation problems -definitions of role functions, scheduling space, and transportation of materials. Two adults in the same classroom was distracting, resulting in classroom discipline and management problems. Chapter I .teachers came from pull out, enjoyed their own 'turf,' responded negatively to pilot. Distracting Conflict over workload. Scheduling class activities to accommodate another class forced more upon regular teacher. 98

PAGE 116

Table 4 (continued) Positive Effects, Benefits Harpring 1985 Program rejects fragmentation. Hayes 1983 More feedback on student progress. A better opportunity for communication between the classroom teacher and the remedial teacher. Levine and Stark. 1981 Elimination of pull out designs alleged to be fragmented in operation and impact on achievement. Attempt to eliminate pull out to aid desegregation. Facilitated substantially improved instruction in some schools. Success with homogeneous grouping for essential skills greater than basal reader programs. Newnann 1985 Regular teacher kept informed of the help the students are getting. Negative Effects Forethought, planning and continued dedication required for programs to survive. Requires inservice for personnel as well as school district personnel. Ensuring compliance w-ith federal guidelines can be more difficult. Monitoring and evaluating procedures may require revisions in current practices. Problems inherent in pull out could be overcome or minimized in inner city schools even in the absence of participation in a project. Extensive cost and financial limitations forced elimination of project. Reading teacher who needs security of self-contained room has many problems. Requires local funding. Local school boards reluctant to pay. 99

PAGE 117

Table 4 (continued) Positive Effects, Benefits Vasguez-Nutall Associates, 1982. 1983 Negative Effects No advantages found for in-class model. Too much distraction to run cwo instructional programs in the same classroom. Inadequate space--"turf" struggles between teachers. Inability to sufficiently individualize. 100

PAGE 118

CHAPTER III RESEARCH DESIGN AND PROCEDURES Overview The problem of this study was to compare the effectiveness of the Chapter I pull-out program design with a Chapter I inclass design by comparing student academic achievement scores, as measured by NCE (Normal Curve Equivalent) pretest and posttest scores, in each type of delivery system in selected schools. Also, scores from self-concept subtests, as measured by the student Self-Description Questionnaire (SDQ), were compared by program design in-class and pull-out. Surveys of teachers and coordinators were utilized to research the subproblem of the study. The regular education and Chapter I teachers of these students were surveyed to determine their opinions on the effectiveness of the program design. The Chapter I coordinators from the State of Colorado were also surveyed to determine their opinions of the effectiveness of program designs. Selection of Districts The Coordinator of Chapter I Services for the State of Colorado, Virginia Plunkett, was very supportive of the

PAGE 119

102 research and aided in selecting the sample. She provided information on districts utilizing both program designs, inclass and pull-out. Six districts utilizing both program designs were requested to participate in the study. Chapter I coordinators in the school districts were originally contacted by Plunkett to determine their willingness to participate in the study. The coordinators from three school districts, referred to as Districts A, B, and C, were willing to participate. They provided the time, energy and cooperation required to facilitate administration of the Self-Description Questionnaire by Chapter I teachers who also contributed pupil data and pretest and post-achievement scores. The researcher communicated regularly with district coordinators and Plunkett through meetings, letters to coordinators and teachers, and by frequent telephone conversations. The coordinators met with their teachers and distributed testing and survey materials and gathered data after the test administration. All data regarding students remained confidential and teachers and coordinators mailed survey data directly to the researcher to maintain confidentiality. Methodology The study compared students experiencing two delivery methods. The sample selected was based on the willingness of districts contacted to cooperate. Third, fourth and fifth

PAGE 120

grade students were selected as samples. Students had been diagnostically placed within pull-out programs or in-class programs, and the Chapter I program provided remedial instruction to the students for all of the 1986-1987 school year. The pretest of student achievement was administered in September of 1986, and the posttest was administered to the same group of subjects in May of 1987. 103 Scores reported by the were NCE scores and were standardized scores. Teachers administered standardized tests which the Colorado State Department of Education accepted as valid and reliable for determining pupil gain in reading and mathematics. Tables provided for each of the tests allowed the teachers to convert raw scores to NCE scores for annual evaluations by the State Department of Education. The NCE scores were standardized scores. Therefore, the scores could be compared across districts. These standardized scores were used by the State Department of Colorado for Chapter I program evaluation. The SDQ, Self-Description Questionnaire, was administered to all specified subjects in May, 1987, during the same month as the administration of the posttest. Students enrolled in Chapter I were selected by school districts according to state-mandated criteria. There was no comparable group of Chapter I students who did not receive services to provide for random assignment; therefore, there was no control group. This study compared two groups which

PAGE 121

experienced different delivery designs in their Chapter I program. Survey Design 104 From October 1986 to the end of December an instrument to survey teacher opinions was developed, piloted and finalized by the researcher. Survey questions in the Teacher Questionnaire (Appendix G) were developed from the research questions, findings, and resulting criticisms of the Glass study (1977). A pilot study of the Teacher Questionnaire was conducted to test the face validity of the instrument. Upon approval of participating school districts, the questionnaire was administered to all of the regular classroom teachers and Chapter I teachers of students included in the study. The researcher then contacted the coordinators in six of the districts that were recommended by Plunkett. Central administrators in four of the six districts contacted agreed to participate in the study at the request of the coordinators. By December, 1986, the researcher completed formal requests to use students as subjects in the school districts and gave samples of materials proposed for use in the study, including the teacher surveys. Three districts agreed to allow student subjects to be used for research. Upon written approval by the districts, the researcher proceeded to administer materials to the subjects through the Chapter I coordinators in three school districts (see Appendices G and J).

PAGE 122

lOS To gain a broader data base, a coordinator survey was developed in February, 1987. Questions applicable to coordinators were taken directly from the teachers' survey. Coordinators were in direct contact with teachers of Chapter I students, but not with the students. As program evaluators, their opinions of program design effectiveness provided a statewide source of data regarding the positive and detrimental effects of in-class versus pull-out programs; therefore, coordinator surveys were mailed in May, 1987. Plunkett submitted names and work addresses of all Chapter I coordinators in the state of Colorado. Dr. Bob L. Taylor, Professor at the University of Colorado, Dr. Alan Davis of the Technical Assistance Center in Denver, Colorado, and Plunkett gave their support for the study in a cover letter to teachers, Chapter I coordinators and directors (see Appendices G, H, and I). Surveys were returned to the researcher in self-addressed, stamped envelopes. Teacher surveys and coordinator surveys were collated and analyzed to determine professional opinions of the positive and detrimental effects of program design. The survey results were analyzed to respond to the subproblem of the study. The Sample School Districts Three districts from the State of Colorado participated in the study.

PAGE 123

106 District A, a large, rural city, serviced a state institution. Lower class to upper middle class populations were located in District A and there were approximately 11,100 students enrolled. Chapter I students totaled 961 in the district in grades K to 12. The racial/ethnic composition of the district showed a higher percentage of Hispanic students in Chapter I (64.3%) than in the total district. The total district contained 29.1% Hispanic, 68.8% Caucasian, .4% Black, 1% Asian, and .3% American Indian. Total percentage of Chapter I students by racial/ethnic group in District A were: .1 American Indian or Alaskan Native .7 Asian or Pacific Islander .8 Black, not of Hispanic origin 64.3 Hispanic 34.0 White, not of Hispanic origin There were 23 Chapter I teachers servicing District A. Of these, six and one-half full-time teachers taught in an inclass design and 16 1/2 taught in the pull-out design. District B, a small rural city, served a state institution. The population was primarily working class with professionals in service roles. Approximately 3,600 students were enrolled at the time of the study. Teachers reported approximately 531 students in Chapter I. The district racial/ethnic population of Whites (92.5%) and Hispanics (7%) closely reflected the Chapter I racial/ethnic composition.

PAGE 124

The district also serviced .1% Black, .2% Asian and .1% American Indian students. The total percentage of Chapter I students in District B by racial/ethnic group was as follows: .0 -American Indian or Alaskan Native .8 -Asian or Pacific Islander .8 -Black, not of Hispanic origin 6.0 -Hispanic 92.4 White, not of Hispanic origin 107 Four teachers provided services in a pull-out program design, while one teacher reported utilizing an in-class model. Chapter I serviced grades 2 through 6. District C, a small suburban area, was located near a large metropolitan area. The population was largely working class, but contained a large transient sector of the population. The socioeconomic status of the population was somewhat depressed, including predominately lower to lower middle classes. Of the approximately 5,800 students enrolled in the district, 350 were serviced by Chapter I at the time of the study. The ethnic composition of the district was 55% White, 38% Hispanic, 3% Black, 7% Asian, and 2.6% American Indian students. The ethnic composition of Chapter I students was approximately equal to district percentages.

PAGE 125

108 Students Descriptive data regarding individual students were tallied and analyzed using information from Pupil Data Sheets. (See Appendix E.) Chapter I teachers completed a Pupil Data Sheet and attached it to the SDQ for each student completing the self-concept measure. Descriptive data on each student included sex, grade, and ethnicity as classificatory variables. Also included were NCE pretest and posttest data to determine if academic gains were similar for both sexes by grade and ethnicity. Academic achievement scores for students, as analyzed by an ANOVA comparing group means, showed the effects of the independent variables of program type, in-class and pull-out, on the dependent variables of NCE pretest and posttest. Interaction of these variables by the classifications of grade, sex and ethnicity were analyzed for Districts A, B, and C combined. Districts A, B, and C each contained a major number of Caucasians, a sizable minority of Hispanics, and a small percentage of Blacks, Asians, and American Indians. District A contained approximately equal numbers of students of both program types, in-class and pull-out, allowing for withindistrict comparisons. Results of descriptive statistics from pupil data sheets and NCE pretest and posttest data are shown in the following discussion of Instrumentation, Descriptive Statistics for Academic Achievement.

PAGE 126

109 Student data from the NCE pretests and posttests and from the Self-Description Questionnaire (SDQ) were analyzed to show the relationship of academic student achievement scores and self-concept scores from the student population. Subtest scores from the SDQ measuring student self-concept yielded eight dependent variables included in a MANOVA from data for student scores. The effect of program type on student self concept by grade, sex and ethnicity was analyzed to determine if there was a negative interaction of student self-concept and program type. Results of the descriptive statistics combining classificatory variables from pupil data sheets and the dependent variables from the SDQ are shown in the following discussion of Instrumentation, Descriptive Statistics of Self Concept. Teachers The Chapter I teachers servicing grades 3 to 5 in Districts A and B were surveyed with the Teacher Questionnaire. A single Chapter I teacher from District C was similarly surveyed. Regular education teachers of the same sample of students serviced by the Chapter I teachers were surveyed with the same Teacher Questionnaire. The surveys were distributed through the Chapter I coordinators to all teachers involved in the study.

PAGE 127

110 Eighty teachers responded to the Teacher Questionnaire including all of the Chapter I teachers and regular education teachers. Teachers from in-class and pull-out designs in all districts responded. Twenty-eight Chapter I teachers, or 35 percent, responded while 52 classroom teachers, or 65 percent., responded to the survey; Ten teachers from District A were labeled as teachers of a mixed or alternative model program design since they occasionally pulled out in-class students for small group instruction. As shown in Table 5, teachers from the combined districts, 54 teachers, or 67.5%, reported that they provided services in Table 5 Teacher Questionnaire: Building Program Design. "Pull-Out." "In-Class" Frequency Percent Pull-Out 54 67.5 In-Class 16 20.0 Other 10 12.5 a pull-out program, while 16 teachers, or 20.0%, provided in-class Chapter I services. Others (10 teachers, or 12.5%) included teachers with a mixed program design, in-class instruction with students pulled out for remedial instruction by Chapter I teachers.

PAGE 128

111 Coordinators Coordinator/Director Questionnaires were mailed to all of the 109 Chapter I coordinators as listed by the Colorado Department of Education. Included were coordinators of 99 school districts, eight boards of cooperative services representing an additional 76 districts, and two state-administered institutions. Fifty-one coordinators, or 47%, returned the surveys. Instrumentation for Academic Assessment Student achievement scores were analyzed to test resesarch Hypothesis I. Hypothesis I -There was no significant difference between academic achievement scores of Chapter I students in the pull-out program and academic achievement scores of Chapter I students in the in-class program. Standardized Achievement Tests Academic achievement for Chapter I students from grades 3 to 5, from Districts A, B, and C was measured to provide data to respond to research question number one (Is the pull-out design more effective than the in-class design for improving student academic achievement?). Gains in student achieve-ment were measured in NCE's (Normal Curve Equivalents). These scores were used in the national Chapter I evalu-ation and reporting system. The tests used were norm

PAGE 129

112 referenced, standardized achievement tests using equal interval scales; therefore, scores could be compared between tests (Buros, 1983). The Instruments The following tests were administered in the districts identified. Standardized Test by District for Chapter I Student District A assessed students for NCE pretesting and posttesting with the California Test of Basic Skills. District B administered Stanford Achievement Tests and Gates-MacGinitie Reading Tests. District C administered the Stanford Achievement Test for the pretest and posttest. The resulting NCE scores used to report student gains were standardized scores with a mean of 50 and a standard deviation of 21.07 (Buras, 1978). These scores were used in the national Chapter I evaluation and reporting system. As standard scores, they can be aggregated across different nationally normed tests, as they have approximately equal interval properties. According to Gay (1981), most variables measured in education form normal distributions, such as intelligence, aptitude and psychological measures if enough subjects were tested (p. 290). The California Test of Basic Skills (CTBS),

PAGE 130

113 Stanford Achievement Test and Gates-MacGinitie Reading Tests are commercial tests that are nationally normed and are standardized. These tests yield a standard score and a standard deviation. According to Gay (1981), "A standard score is a derived score that expresses how far a raw score is from a mean in terms of deviation units" (p. 437). "A standard deviation is the most stable measure of variability which takes into account each and every score in a distribution" (p. 437). Standard scores are based on deviation units. Assuming that the variables measured by these three tests are normally distributed since they yield normal curve equivalency scores, or standard scores, each raw score can be compared across tests according to its distance from the mean (X) or its standard deviation (SD). Research Design A pretestjposttest nonequivalent comparison group design was used to compare achievement data for two comparison groups--one pull-out and one in-class. All Chapter I students, grades 3 to 5, included in the study received services in either a pull-out or in-class design for one school year. Students experienced pretest and posttest measures and the SDQ to compare achievement and self-concept. Research has not demonstrated an unequivocable correlation between self-concept and achievement. These variables were considered as two

PAGE 131

114 important indicators of program design effectiveness that were not tested for direct correlation. Data Processing for Academic Achievement Analysis of Variance, or ANOVA, was used in this study to equate homogeneous groups of Chapter I students, grades three to five, on the academic achievement variables, pretest and posttest NCE scores. Initial differences on the pretest were adjusted to determine if there were differences in gains in achievement between the in-class and pull-out students as measured by performance on the posttest. The ANOVA was applied in this study, as recommended by Gay (1981), to equate groups on one or more variables. The ANOVA has been used for causal comparative studies as well as experimental studies (p. 204). Demographic Data for Achievement Analysis Comparison of student academic achievement for students from Districts A, B, and C combined was based on a sample of total number of students of 235 Chapter I students. All third through fifth grade students were included in the study. The total sample contained 126 males and 108 females with a racial/ethnic composition: American Indian (1), Asian Pacific (3), Blacks (1), Hispanic (108), and Whites and others (122).

PAGE 132

115 Hypothesis I Hypothesis I determined the differences in student academic achievement that may have resulted from program design--pull-out versus in-class, and the academic achievement for the total student sample was analyzed. Differences in academic gains that resulted by the classificatory variables of sex, grade, or ethnicity cannot be considered as differences, due to the pull-out design as opposed to differences due to the in-class design as independent variables. Therefore, differences in the student sample by classificatory variables were compared to NCE pretest and posttest differences, or gains in achievement. Descriptive statistics include a breakdown of the student sample with respect to ethnicity so that the nature of the sample was clear. Results of the descriptive statistics show similar gains on the NCE posttests as compared to NCE pretests for the students by sex, grade and ethnicity. There were differences by these classificatory variables on the NCE pretests. Two ethnic groups were compared, Whites and Others, and Hispanics, due to the large number of Hispanics contained in the sample. Whites and Others scored consistently higher in all three grade levels. Females from both ethnic groups scored higher than males. However, when posttest scores were compared with pretest scores, females from both ethnic groups made approximately the same gains. Males from both ethnic groups made greater gains than the females. Overall, posttest gains

PAGE 133

116 showed a pattern of increase in academic achievement scores similar to the pattern of scores shown by the NCE pretest. Scores for students on the NCE pretests are reported in Table 6 by grade and by ethnicity, as shown below: American Indian 1 Asian Pacific 2 Black 3 Hispanic 4 White and Others 5 Table 6 NCE Pretest. Ethnicity by Grade Level Ethnicity Grade 1 2 3 4 5 3 X 36.00 0.00 7.00 26.23 32.98 n (1) (0) (1) (57) (47) 4 X 0.00 30.00 0.00 24.21 34.57 n (0) (1) (0) (34) (37) 5 X 0.00 29.00 0.00 28.38 33.56 n (0) (1) (0) (13) (25) Table 6 displays a comparison of cell means. American Indian, Asian Pacific and Blacks are not discussed, as the number for each was comparatively small, with 0 or 1 of each ethnic group at a given grade level. The Whites and Others scored consistently higher than Hispanics in all three grades. However, mean difference (M.D.) between the two groups in grade

PAGE 134

117 5 was somewhat less than in grade 3, M.D. = 6.75, grade 4 =10.36, and grade 5 = 5.18. When mean scores were compared for ethnicity by sex, Table 7 indicates that females from the Hispanic and White populations scored higher on the NCE pretests. NCE posttest scores in Table 8 show mean scores by grade and ethnicity. Table 7 NCE Pretest. Ethnicity by Sex Ethnicity Sex 1 2 3 4 5 Male x 36.00 29.00 0.00 24.60 32.34 n (1) (1) (0) (57) (58) Female X 0.00 30.00 7.00 27.34 35.14 n (0) (1) (1) (47) (51) Table 8 NCE Posttest. Ethnicity by Grade Level Ethnicity Grade 1 2 3 4 5 3 X 32.00 0.00 35.00 32.96 37.09 n (1) (0) (1) (57) (47) 4 X 0.00 34.00 0.00 23.47 37.49 n (0) (1) (0) (34) (37) 5 X 0.00 36.00 0.00 29.92 40.04 n (0) (1) (0) (13) (25)

PAGE 135

118 NCE posttest scores show greater gains in cell means for Whites than for Hispanics, grade 3 M.D. = 4.23, grade 4 M.D. 14.02, grade 5 M.D. 10.12. The gains indicate a pattern of increase similar to theNCE pretest gains. Table 9 shows NCE posttest comparisons by sex and ethnicity. The NCE posttest scores show only slightly more gain for females than males for both Hispanics and Whites. Table 9 NCE Posttests. Ethnicity by Sex Ethnicity Sex 1 2 3 4 5 Male x 32.00 36.00 0.00 28.81 37.74 n (1) (1) (0) (57) (58) Female X 0.00 34.00 35.00 30.17 38.08 n (0) (1) (1) (47) (51) Table 10 shows gains through mean differences. One Black, third grade student made an exceptional gain of 28 NCE's. The only group showing no gain were Hispanic fourth graders who, in fact, dropped 0.74 NCE's. Comparisons of the students in Table 11 showed gains by difference in cell means for sex and ethnicity. Over all, Hispanic and White females made approximately the same gains. Hispanic and White males made greater gains than

PAGE 136

119 Table 10 NCE Pretest to Posttest Gains Ethnicity Grade 1 2 3 4 5 3 X 4.00 0.00 28.00 6.63 4.11 n (1) (0) (1) (57) (47) 4 X 0.00 4.00 0.00 0.74 2.92 n (0) (1) (0) (34) (37) 5 X 0.00 7.00 0.00 1.54 3.93 n (0) (1) (0) (13) (25 Table 11 NCE Pretest to Posttest Sex Ethnicity Sex 1 2 3 4 5 Male X 4.00 7.00 0.00 4.21 5.40 n (1) (1) (0) (57) (58) Female X 0.00 4.00 28.00 2.83 2.94 n (0) (1) (1) (47) (51) females, with White males and one Asian male showing the greatest gains, 5.40 NCE's and 7.0 NCE's, respectively.

PAGE 137

Instrumentation for Self-Concept Assessment Student self-concept was assessed through the administration of the Self-Description Questionnaire to test research Hypothesis II. Hypothesis II stated that there was no significant difference in the self-concept of Chapter I students, as 120 measured by the Self-Description Questionnaire in the pull-out program and the self-concepts of Chapter I students in the in-class program. Self-concept data from the student sample were analyzed to respond to research question 2 (Is the pull-out design more effective than the in-class design for improving student self-concept?). Research Design Static group comparison was the form of non-equivalent control group design used for the SDQ measure, for no pretest was given. Self-concept of the student was measured at a point in time near to the posttest, or the end of the treatment. Subjects who left the program before the administration of the NCE posttest and the SDQ administration were not included (n = 1). Variables Pull-out program design and in-class program design were the programs compared. The program designs were compared to

PAGE 138

121 determine if there was a significant difference by program design in academic achievement or student self-concept. Classificatory variables for students by achievement and selfconcept measures were sex, grade, and ethnic origin. Student achievement and student self-concept were compared with the same classificatory variables. The SDQ, measuring students' self-concept, included eight subtest scores: physical ability, physical appearance, peer relations, parent relations, reading, mathematics, general school, and general self. Marsh (1983) gave directions for hand scoring the SDQ for each of the subtest scores. According to Marsh's directions, these subscores combined into two further scores, total academic and total nonacademic scores. As shown in Table 12, the total academic score was combined with responses by students on their view of self in relation to reading, math, and general school. These two variables combined, yielded to a total score variable. Selection of the Instrument The Self-Description Questionnaire (SDQ) was recommended by Dr. Alan Davis of the Technical Assistance Center in Denver, Colorado. The questionnaire was selected due to the multidimensional design of the instrument. The hierarchical model of self-concept hypothesized by Shavelson (Marsh, 1984), describing the seven facets of self-concept, was considered the appropriate model to measure the self-concept of Chapter I

PAGE 139

Table 12 Variables from Data for Combined Districts Variable Mean Std.Dev. Minimum Maximum Valid N Label Id 117.000 67.983 0 234 235 Age 9.885 1.209 8 13 235 Grade 3.821 .944 1 6 235 Sex 1.472 .525 1 4 235 Ethnicity 4.487 .602 1 5 234 Ethnicity Adprog .145 .353 0 1 235 Additional programs taken Time 11.706 2.024 8 15 235 Time of day Hrwk 165.298 34.858 120 250 235 Minutes per week District 3.085 .966 2 4 235 School 6.357 2.760 1 15 235 Type 1. 753 .432 1 2 235 Three 32.357 5.995 11 40 235 One 28.800 7.940 8 40 235 Seven 29.749 7.355 8 40 235 Five 34.468 5.837 12 45 235 Four 29.757 7.126 8 40 235 Six 30.464 9.118 7 40 235 Two 25.923 7.903 7 40 235 General 31.638 6.432 11 40 235 TN 29.271 5.168 13 40 235 Total non-academic score TA 31.455 5.354 13 40 235 Total academic score TS 30.519 5.192 17 62 235 Total score NCEPRE 30.026 11.746 1 60 234 NCE pre-program NCEPOST 34.021 12.275 1 66 234 NCE post-program score N N

PAGE 140

123 students. Due to the low socioeconomic status of Chapter I schools and the students they serve, and the low academic achievement of Chapter I students, a multifaceted instrument was required. A letter describing the study and asking permission to use the SDQ was sent to Dr. Herbert Marsh by the researcher in December, 1985. Prior permission to use the SDQ for educational research in the United States with American subjects had been granted to Dr. Davis. Dr. Marsh had requested that information regarding the studies be sent to him in Australia as he was revising the SDQ. Similar permission and the same request were made for this study (see Appendix C). Dr. Marsh granted permission and was interested in receiving the results of this study. Background of the SDQ The Self-Description Questionnaire (SDQ) used to measure student self-concept was developed by Herbert W. Marsh, Ian D. Smith, and Jennifer Barnes from the University of Sydney, Australia (Marsh, 1983). This instrument was selected due to the recommendation by Dr. Marsh that the construction of the instrument overcame two basic problems with development and l selection of an instrument to adequately measure self-concept. First, definitions of self-concept are imprecise. Secondly, few of the more commonly used instruments have been adequately validated (Crowne & Stephens, 1961; Marx & Winne, 1978;

PAGE 141

Shavelson & Bolus, 1982; Shavelson, Hubner, & Stanton, 1976; Wylie, 1974, 1979). 124 According to the definition of self-concept presented by Shavelson (Shavelson & Bolus, 1982; Shavelson et al., 1976), "Self concept is an individual's perception of self, and is formed through experience with the environment, interactions with significant others, and attributions of his or her own behavior" (Marsh, 1983, p. 335). Despite a pervasive, implicit assumption by most theorists that self-concept is multifaceted, empirical support for the assumption has been modest, according to Marsh. He further stated that most attempts to demonstrate the multidimensionality of self-concept have relied on factor analysis or multitrait-multimethod (MTMM) analyses. Marsh (1983) established the construct validity of the SDQ through the application of factor analysis and multitraitmultimethod analysis (MTMM). The SDQ was designed to measure seven dimensions of self-concept. According to the definition of self-concept presented by Shavelson and Bolus, 1982, the organization of self-concept is multifaceted and hierarchical, with perceptions moving from inferences about self in the sub areas of reading, mathematics, physical ability, physical appearance, peer relationship, parent relationship, general school and general self, moving to broader areas of academic and nonacademic, and finally to general self-concept. Empirical confirmation of the facets of self-concept is

PAGE 142

us demonstrated through application of factor analysis. Marsh (1983) also explained the use of factor analysis in the development of the SDQ to eliminate ambiguity of interpretation. Factor analytic studies typically combine exploratory and confirmatory modes of the approach. In the exploratory mode, the researcher simply factor analyzes responses and tries to identify the factors that emerge. In the confirmatory mode, the attempt is to demonstrate empirical support for the set of dimensions that the instrument was designed to measure. If the match between the hypothesized and obtained factors is reasonably good, then there is support for both the construct validity of the particular instrument and the multidimensionality of self-concept. Typically there is not a clear match and then the interpretation is ambiguous. This ambiguity is particularly likely when factor analysis has not been used in the development of the instrument. (p. 335) Factor Analysis Marsh randomly selected a group of teachers in Sydney, Australia, to participate in an evaluation of the SDQ. Teachers were asked to evaluate student self-concept on each of the seven dimensions measured by the S.D.Q. Marsh stated that teacher ratings were .reasonable indicators of student self-concept because elementary school teachers (a) have considerable contact with their students; (b) are exposed to a wide variety of children; (c) have experience in making professional decisions about individual students; and (d) are likely to observe behaviors relevant to each of the SDQ dimensions (with the possible exception of Relations with Parents, which is the dimension that shows the least student-

PAGE 143

126 teacher agreement. Further, previous research (e.g., Perkins, 1958; Phillips, 1963-1964) had reported substantial agreement between teacher ratings and self-reports by students (Marsh, 1983, p. 353). Marsh stated that the importance of an MTMM analysis of agreement between student and teacher ratings of students' self-concepts was to support both the convergent and divergent validity of the self-concept dimension. The MTMM analysis supported both the convergent and divergent validity of the self-concept dimensions. Student-teacher agreement on the self-concept dimensions was significant, and agreement on each dimension was relatively independent of ag;eement on other dimensions. Finally, student and teacher ratings of the seven dimensions of self-concept demonstrated similar and predictable correlations with attributions of academic achievement, student sex, and reading achievement. Taken together, these findings provide strong support for the multi-dimensionality of self-concept and the construct validity of interpretations based on the SDQ. (Marsh, 1983, p. 353) Reliability and Construct Validity The multidimensional instrument was designed to measure the seven facets of self-concept hypothesized in Shavelson's hierarchical model (Physical Ability, Physical Appearance, Peer Relationship, Parent Relationship, Reading, Mathematics, and General Self). The pattern of small correlations among the student self-concept dimensions was generally consistent. The correlations varied from close to zero to .42 (Mnr = .24). The reliabilities of the scales were in the .80's and .90's. The distinctiveness of the various dimensions was an advantage in

PAGE 144

127 determining the probability that the reading factor was significantly different in either of the two program designs as compared to the other six dimensions. There was a near zero correlation between mathematics and reading self-concepts in the MTMM analysis. This correlation was an exception. Shavelson's hierarchical model predicts substantial correlations among the three academic factors (Reading, Math, and General School) between the two social factors (Peers and Parents), and between the two physical factors (Abilities and Appearance; thus, the multitrait, multimethod matrices derived from the data can be summarized with an ANOVA model. A study was designed to establish the construct validity of interpretations based on the SDQ. A group of students completed the SDQ, and their teachers also filled out an SDQ on each student. Student and teacher ratings of the seven dimensions of self-concept demonstrated similar and predictable correlations to strongly support the construct validity of the instrument. Construct validity was confirmed by a factor analysis of the students' ratings of self-concepts, investigation of an MTMM analysis of agreement between student and teacher ratings of students' self-concepts, and an investigation of the pattern of correlations among the selfconcept dimensions and other variables. The MTMM analysis supported both the convergent and divergent validity of the self-concept dimensions.

PAGE 145

128 There seemed to be a possible problem with the validity of the instrument with the third to eighth grade Chapter I samples. First, the instrument had not been validated with a U.S. population. Second, the SDQ was analyzed using a fifth and six grade sample. Therefore, the instrument was piloted with. third grade students from a Front Range school district, one not included in the research sample, to determine use at the third grade level and possible cultural bias. Pilot Study The SDQ, Self-Description Questionnaire, was developed and normed in Australia. Dr. Herbert Marsh was interested in the use of the instrument for research with populations from the United States (see letter of support for research, Appendix C). The SDQ required a pilot study in this study for two reasons. First, the appropriateness of language and face validity of the instrument was to be determined. Second, detailed instructions to teachers to provide a consistent way of administration to anticipate difficulties with student comprehension was a necessary precaution to avoid possible testing bias. Forty-one Chapter I; third grade students comprised the pilot population for completing the SDQ. Third graders were sampled due to the probability that the youngest students in the research sample of third to fifth graders would have the greatest difficulty with comprehension and understanding directions. Teachers were asked to attach a data sheet for

PAGE 146

each child along with each SDQ, including the age, grade, sex and ethnic origin of the child. (Questions regarding the programming of the child were included to determine each child's qualifications for inclusion in the pilot sample.) Teachers were then requested to relate information concerning the dates and time of test administration and problems which students experienced with the test. (See Appendix D.) 129 Results of the pilot study indicated problems in two areas. First, the Australian word for J was "tick" in the directions to the child, to put a "tick" in the box indicating the degree of agreement with an SDQ statement. The word "tick" had to be eliminated and replaced with a J. Secondly, teachers identified difficulties that the students had with vocabulary. The vocabulary words causing difficulty were listed in additional directions to teachers for the test administration. Also, a revised Pupil Data Sheet was developed to improve the ease of completion and provide for greater clarity of data. (See Appendix E). The teachers were alerted to problems with comprehension of words and phrases, including "raised me," "featured," "athlete," "mathematics," and "school subjects." The directions for administering the test to a population of Colorado Chapter I students were developed as a result of pilot teachers reporting the student problems with the test. (See Appendix E).

PAGE 147

130 The SDQ, as administered in the pilot program, demon-strated that the Chapter I students were able to comprehend the directions and the meaning and content of the statements as presented for their response. Vocabulary was appropriate for their comprehension. Descriptive Statistics for Self-Concept The descriptive statistics are included in the data for a MANOVA to test Research Hypothesis II: There was no significant difference in the self-concepts of Chapter I students, as measured by the Self-Description Questionnaire in the pull-out program and the self-concepts of Chapter I students in the in-class program. The total population of 235 Chapter I students, grades 3 through 5, completed the SDQ. Empty cells caused totals for the MANOVA to show 218 students. Factors comprising the scores from the SDQ were assigned the following numbers: Physical appearance 1 General school 2 Physical ability 3 Reading 4 Parent relations 5 Mathematics 6 Peer relations 7 General self G

PAGE 148

Scores were combined to yield a total academic score, TA, a total nonacademic score, TN, and a total score, TS. 131 Students are described on each of the above factors of the SDQ. Differences of all means greater than 1.50 on subtest scores listed above are discussed. Differences of less than 1.5 were not considered as meaningful and are described as of little or no difference. Results of the descriptive statistics for student selfconcept are discussed below by subtest. Results showed that there were meaningful differences among students in every factor of self-concept by at least one classificatory variable, sex, grade, or ethnicity. The MANOVA tested for the significance of these differences at the a = .OS level of significance. Physical appearance (1): As shown in Table 13, fifthgrade white students, male and female, had a lower self-concept score on physical appearance with a mean of 26.56 for fifth graders and a mean of 27.48 for white males. The fifth-grade white males' score was lower than that for females. General School (2): As shown in Table 14, Hispanic third-, fourth-, and fifth-grade students scored higher than white students on general school as an academic sub-test score. White, fifth grade males showed the lowest self-concept in relation to general school with a mean of 23.33. Females at

PAGE 149

132 Table 13 Ethnicity Grade 1 2 3 4 5 3 X 25.00 0.00 40.00 29.07 28.68 n (1) (0) (1) (58) (47) 4 X 0.00 36.00 0.00 28.65 29.16 n (0) (1) (0) (34) (37) 5 X 0.00 13.00 0.00 33.38 26.56 n (0) (1) (0) (13) (25) Sex Male x 25.00 13.00 0.00 29.22 27.48 n (1) (1) (0) (58) (58) Female X 0.00 36.00 40.00 29.77 29.35 n (0) (1) (1) (47) (51) all grade levels indicated higher scores in relation to general school. Physical Ability (3): As shown in Table 15, the mean for physical ability for males was higher than for females. There was no difference by grade level for ethnicity. Reading (4): As shown in Table 16, there was no meaningful increase in any grade by ethnicity. However, both males and females showed a slight decline in reading self-concept scores, as grade level increased. The exception was females in fourth grade showing a slight increase over third grade, from a mean score of 31.28 to 32.03.

PAGE 150

133 Table 14 General School Grade 1 2 3 4 5 3 X 32.00 0.00 32.00 27.19 25.17 n (1) (1) (1) (58) (47) 4 X 0.00 40.00 0.00 27.32 25.27 n (0) (1) (0) (34) (37) 5 X 0.00 23.00 0.00 25.54 23.88 n (0) (1) (0) (13) (25) Sex Male x 32.00 23.00 0.00 26.19 23.33 n (1) (1) (0) (58) (58) Female x 0.00 40.00 32.00 28.06 26.71 n (0) (1) (1) (47) (51) Table 15 Physical Ability Grade Sex 3 4 5 Male X 34.58 34.26 31.08 n (64) (35) (19) Female X 29.79 31.08 28.50 n (43) (37) (20)

PAGE 151

134 Table 16 Reading Grade Sex 3 4 5 Male X 29.64 28.20 27.68 n (64) (35) (19) Female X 31.28 32.03 30.10 n (43) (37) (20) Parent Relations (5): As shown in Table 17, third to fourth grade males and females both showed a slight decrease in the parent relations variable. In fifth grade, males dropped meaningfully, from 34.95 to 33.0, while females showed a marked increase from 33.34 to 34.95. White males scored much lower (34.57) than White females (32.94). Yet, Hispanic females score somewhat higher (36.68) than Hispanic males (34.33) or White males (34.57). Mathematics (6): As shown in Table 18, Hispanic and White males and females scored equally well on mathematics, showing no difference by sex on ethnicity. The pattern by grade and ethnicity was variable, with Whites scoring higher than Hispanics in grades 3 and 5, with a marked difference in grade 5, 35.36 and 28.15 respectively.

PAGE 152

135 Table 17 Parent Relations Sex by Grade Grade Sex 3 4 5 Male x 34.73 33.34 35.11 n (64) (35) (19) Female x 35.37 34.95 33.00 n (43) (34) (20) Sex by Ethnicity Sex 1 2 3 4 5 Male x 37.00 24.00 0.00 34.33 34.57 n (1) (1) (0) (58) (5.8) Female X 0.00 40.00 30.00 36.68 32.94 n (0) (1) (1) (47) (51) General Self (G): As shown in Table 19, there was no meaningful difference on this self-concept indicator by sex or ethnicity. However,.by grade, Whites showed a decrease from 31.49 to 30.24, while Hispanics showed an increase from 31.21 to 33.31.

PAGE 153

Table 18 Mathematics Grade 1 3 X 25.00 n (1) 4 X 0.00 n (0) 5 X 0.00 n (0) Table 19 General Self Grade 1 3 X 36.00 n (1) 4 X 0.00 n (0) 5 X 0.00 n (0) 2 3 4 5 0.00 10.00 28.55 30.47 (0) (1) (58) (47) 40.00 0.00 32.35 29.70 (1) (0) (34) (37) 30.00 0.00 28.15 35.36 (1) (0) (13) (25 2 3 4 5 0.00 40.00 31.10 32.64 (0) (1) (58) (47) 35.00 0.00 31.21 31.49 (1) (0) (34) (37) 29.00 0.00 33.31 30.24 (1) (0) (13) (25) Data Processing and Analysis of Self-Concept Data To test Research Hypothesis II, a multiple analysis of variance (MANOVA) and a repeated measures ANOVA were used to compare the effects of sex, grade, ethnicity, and type of 136 program on academic achievement and self-concept scores on all

PAGE 154

Chapter I students described in the sample of students completing the SDQ. The significance level (alpha) for rejecting the null hypothesis was set at a = .05. 137 The multiple analysis of variance (MANOVA) allows for determination of the effect of the independent variable and control variable, both separately and in combination. Therefore, it can be determined whether there was an interaction between the independent, dependent, and descriptive variables, separately or in combination. This allows for the academic variable derived from scores to be compared with each of the seven self-concept variables of the SDQ alone and combined in a total score or combined in subscores of total academic self-concept and total non-academic self-concept scores. Factor analysis of variance considers all of the variables in the academic achievement and self-concept comparisons individually, and with each variable compared to each of the others for the in-class versus pull-out programs. Summary of NCE's and SDQ In summary, research shows that academic achievement was generally correlated with self-concept, and even more highly correlated with measures of academic self-concept (Shavelson & Bolus, 1982; Shavelson et al., 1976; Wylie, 1979). According to Marsh, this relationship is particularly strong when students' self-concept was determined by asking them to rank

PAGE 155

138 themselves against some comparison group (classmates) in terms of academic achievement being measured. Such correlations contribute to the construct validity of self-concept. The correlation between theNCE's and the SDQ can be determined, and if the correlation is high but there is no significant difference between the two program designs, it can be concluded that pull-out programs are not more detrimental to student self-concept than are in-class programs. Likewise, in-class programs are no more detrimental than pull-out programs. Instrumentation for Teachers' and Coordinators' Survey--Subproblem Two questionnaires, a Teacher Questionnaire and a Coordinator Questionnaire, were developed by the researcher to survey opinions regarding the subproblem of the study. Subproblem: A subproblem of this study was included to determine the support among school personnel for the two approaches (pull-out and in-class). This could have long-range implications for future decisions on which approach will become established practice. Which program, and for what reasons, was favored by the Chapter I teachers, the Chapter I coordinators, and the regular classroom teachers? Pilot Study A pilot of the Teacher Questionnaire was administered to a class of 24 students at the University of Colorado, Denver

PAGE 156

139 campus. The students were enrolled in a class in educational administration. The instructor requested that the students, most of whom held professional positions in the field of education, critique the questionnaire. All 24 students contributed critiques, raised questions or commented on the questionnaire. (See Appendix F.) Revisions included the heading "Teacher Questionnaire" with the sentence "Please check building design ______ Pullout Inclass _____ Other, specify." Also the wording of some of the questions was revised, directions were added, and the format changed for some questions. For the revised Teacher Questionnaire, see Appendix G. Data Collection and Analysis for Subproblem The questionnaire was administered to the teachers involved in the study to measure their attitudes toward the program designs. Regular education teachers and Chapter I teachers of the students included in the sample were surveyed from Districts A, B, and C. The questionnaire was designed to determine the positive and negative detrimental effects of the pull-out design as compared to the in-class design. Teacher responses addressed the subproblem of the study. The coordinators' survey was designed with questions taken directly from the original teachers' survey. It excluded those questions which required direct contact with Chapter I stu-

PAGE 157

140 dents. (See Appendix H.) Responses to the coordinator survey were collated to tally with the original teacher survey; hence, responses to the coordinator survey could be compared to responses from the teacher survey. The questionnaire results provided data to respond to the subproblem of the study. Responses on the teacher and coordinator surveys were tallied for each question and presented in table form. Results were compared to positive and negative effects of the in-class versus the pull-out program designs as summarized in Tables 3 and 4, pages 89-92. Demographic Results of Coordinator Survey Coordinator responses indicated that all 51 coordinators had programs containing the age population of this study, Chapter I students in grades 3 to 5. The mean student population of the districts of the 51 coordinators was 4,978. Thirty-nine of the districts had total enrollments of fewer than 5,000 students. District A had a total enrollment of 11,100 students. District B had a total enrollment of 3,600 students. District C had a total enrollment of 5,800 students. Districts B and C were not divergent from the average popu-lations of the coordinators' districts. District A was more than twice the size of the average population and was repre-sentative of the larger rural populations in the state.

PAGE 158

141 The mean Chapter I total student enrollment of the districts of the coordinators was 323 students, or 6.5% of the mean total student enrollment. District A enrolled 8.7% of students in Chapter I. Enrollment of Hispanics in Chapter I by percentage, 64.3%, was more than double that of the total district enrollment, 29.1%. The racial/ethnic percentages of Chapter I students in Districts B and C reflected the percentages of the total populations. District B enrolled 14.8% of the students in Chapter I, while District C enrolled 6% of the students in Chapter I. District B had a larger number of students enrolled in Chapter I, but the greater percentage was Yhite (92.4%), indicating a relatively large economically dis-advantaged Yhite population. However, in the combined sample, the larger numbers of Yhite or non-Hispanic students acted to counterbalance the larger numbers of Hispanic students in Chapter I in District A. Overall, the combined district sample was racially/ethnically a more balanced sample. Teachers and coordinators were asked about the Caucasian population of their Chapter I program to determine if minority students were more likely to be placed in Chapter I programs than Caucasian students. Question: Does the Caucasian population of your Chapter I pupils reflect the population of the district in number? Two-thirds of the coordinators (66.7%) responded that numbers of Caucasians, or Yhites, in Chapter I reflected the district

PAGE 159

population. Fifty-one percent of the 66.7% coordinators had pull-out programs, as compared to 15.7% with in-class. 142 Teachers agreed with the coordinators, with a response of "same as" with 44.4% for pull-out and 25% for in-class.

PAGE 160

CHAPTER IV DATA ANALYSIS AND RESULTS Overview Chapter IV presents the data collected from student tests and teacher and coordinator surveys. The survey responses were analyzed to determine frequency of responses. Problem of the Study The study sought to compare the effectiveness of the Chapter I pull-out program as the predominant service model to the in-class program design. The subproblem of the study was included to determine the support among school personnel for the two approaches (pull-out and in-class). This could have long-range implications for future decisions on which approach will become established practice. Which program, and for what reasons, was favored by the Chapter I teachers, the Chapter I coordinators, and the regular classroom teachers?

PAGE 161

144 Research Question 1 1. Is the pull-out Chapter I program design more effective than the in-class design for improving student academic achievement? To investigate Research Question 1, the student academic achievement scores were analyzed. Student achievement data utilized scores from Chapter I testing of 235 students in one school year, 1986-1987, comparing NCE (Normal Curve Equivalency) pretest and posttest scores. An Analysis of Variance (ANOVA) was employed to test research questions for District A, the district with approximately equal numbers of in-class and pullout students. A Multiple Analysis of Variance (MANOVA) was utilized for the academic achievement scores of the combined districts' sample to determine that the total sample was academically similar. Research Question 2 2. Is the pull-out design more effective than the in-class design for improving student self-concept? To investigate Research Question 2, results of a second MANOVA were employed. Self-concept data from the population of students in Districts A, B, and C were analyzed in a MANOVA. Scores from the SDQ and the NCE pretests and posttests were analyzed to determine if there were significant differences in the students' academic achievement as compared to the students'

PAGE 162

14S self-concepts for the two program designs, pull-out versus in class. A decision was made to set a = .OS to reject or accept the hypotheses tested at the a< .OS level. The hypotheses and decisions made are contained in the final sec.tions of this chapter. Subproblem A subproblem of the study utilized survey data to determine support among school personnel for the two program designs. The study attempted to determine long-range implications for future decisions.on which approach or program design will become established practice. The subproblem of the study asked the following question: Which program, and for what reasons, was favored by the Chapter I teachers, the Chapter I coordinators, and the regular classroom teachers? Responses of the survey data were tabulated in frequency tables and analyzed to answer the subproblem of the study. Discussion of statistical findings and survey results are summarized in Chapter IV and conclusions and recommendations are presented in Chapter V. Analysis of Research Question 1 Hypothesis I: There will be no significant difference in the academic achievement of Chapter I students in the pull-out program as compared to the academic achievement of Chapter I students in the in-class program.

PAGE 163

An ANOVA was utilized to investigate Hypothesis I. A MANOVA compared the achievement scores for students of the combined districts' sample to determine if the combined districts' samples were academically similar. District A ANOVA District A contained all of the in-class programs in the samples, with the exception of one in-class program from District B. Using the single district, District A, sample provided approximately equal numbers of in-class and pull-out students for statistical measures. In addition, numbers of students at each grade level were similar with respect to variables of sex and ethnicity. Finally, the students experienced the same pretest and posttest instruments. The sample of students with NCE pretest and posttest scores totaled 119. Major Findings for Academic Achievement Tables 20 and 21 show subjects compared in the combined 146 districts in a MANOVA to determine if there was variance within groups on the dependent variables of NCE pretest and NCE posttest. Table 20 shows that Chapter I in-class students scored lower as a group on the NCE pretest, 25.8 mean score as compared to the pull-out students 29.8, at the a= .05 level. Table 21 shows that there was no significant difference on the NCE

PAGE 164

147 posttest at the a = .05 level. Therefore, Tables 20 and 21 show no difference in academic achievement by program design. In-class students, although scoring lower on NCE pretests than pull-out students, made gains in achievement that were approximately equal to the gains of the pull-out students. Table 20 District A: NCE Pretest Between Group Comparison Number Mean In-class 58 25.7931 Pull-out 61 29.8197 Total 119 27.8571 Source of variation between groups: Mean F F Standard Deviation 13.8915 10.1858 12.2497 Degrees of Freedom Sum of Squares Squares Ratio Probability 1 482.0378 482.0378 3.2743 .0729 Table 21 District A: NCE Posttest Between Group Comparison Number In-class 58 Pull-out 61 Mean 30.0000 33.3770 Standard Deviation 15.1866 10.7054

PAGE 165

Table 21 (Continued) Source of variation between groups: Degrees of Sum of Mean Freedom Squares Squares 1 12.5428 12.5428 In-class number 58 Pull-out number 62 Combined Districts MANOVA for Academic Achievement F Ratio .0971 F Probability .7559 A non-repeated measures MANOVA showed the effects of the NCE pretest as compared to the NCE posttest to measure the 148 academic achievement of 218 Chapter I students from the combined district sample. Of the total population of 235 students, 16 cases were rejected due to out-of-range factor values. One case was rejected due to a missing NCE posttest. The result was 16 non-empty cells with 218 cases. The first test of between-subjects effects using UNIQUE sum of squares was to determine any source of variance between subjects by the independent variables of grade by sex by ethnicity. At the a = 0.05 level there was no variance between subjects with the significance ofF= .71. When the independent variables were considered as a source of variance using the unique sum of squares involving NCE pretest and posttest differences, there was no significant difference by within-sub-jects effects, F-.810. See Table 22.

PAGE 166

Major Findings for Achievement by Combined Districts There was no difference found in the combined district samples on the classificatory variables of sex, grade and ethnicity on the dependent variables of NCE pretest and posttest. Students made similar gains on the NCE posttests with no variance by the classificatory variables Therefore, students from combined districts made similar academic gains Table 22 Tests of Between-and Within Subjects Effects 149 Source of Variance Sum of Degrees of Squares Freedom Mean Square Sig. of F F Grade by sex ethnicity Grade by sex by ethnicity by NCE test gains 148.60 2 16.87 2 74.30 34 711 8.43 .17 .840 when students were combined without regard to program type, in-class versus pull-out. All pull-out students from Districts A, B, and C made gains comparable to the in-class students. Analysis of Research Question 2 Hypothesis II: There was no significant difference in the self-concepts of Chapter I students, as measured by the

PAGE 167

150 Self-Description Questionnaire in the pull-out program and the self-concepts of Chapter I students in the in-class program. To assess the self-concepts of Chapter I students, results of the subtests of the SDQ were analyzed in an ANOVA for District A and a MANOVA for the combined districts. District A contained approximately equal numbers of students in both program designs. Significant differences were found in three subtests for the ANOVA for District A, peer relations, general self and total academic score. However, when the variables for the District A ANOVA were tested in a MANOVA for the combined districts sample, where larger numbers of pull-out students were compared to the in-class students, a significant difference was found for one subtest only, physical ability. Results of District A ANOVA for Self-Concept Table 23 shows there is no significant difference at the a = .05 level in self-concept scores between in-class and pull-out students, with total score, F = .0682. While there is no significance at the a = .05 level, the pull-out students had a significantly higher self-concept than the in-class students at the a= .10 level. Pull-out students scored significantly higher at the a = .05 level than in-class students on three variables--peer relations (.0045), general self (.0096), and total academic score (.0213). See Table 23 for a comparison of group means.

PAGE 168

151 Tables 23 and 24 show that in-class students scored lower on two subtest scores, peer relations and general self. Results show that these students are negatively affected by participating in Chapter I in the in-class program in the area of peer relations. Further, students experiencing lower self-concept scores in the areas of peer relations and in general self-concept seem to have a significantly lower view of themselves as learners, as seen in the total academic mean score of 30.6 as compared to the pull-out students with a mean of 32.6, resulting in the significant F probability F .0213. Overall, pull-out students scored significantly higher in three subtest areas within a single district sample. Results of Combined Districts MANOVA for Self-Concept In order to determine whether the total sample of pull-out students would score higher than the total sample of in-class students, self-concept scores were compared from the combined districts. A MANOVA, Table 25, shows all Chapter I students from combined districts and looks at effects of sex, grade, ethnicity and type of program on self-concept scores. Table 25 shows the result of the MANOVA that compared the classificatory variables of grade, sex and ethnicity on the dependent self-concept variables shown in the table. There was

PAGE 169

152 Table 23 District A ANOVA Self-conceRt as Measured bl the SDQ, Self-DescriRtion Questionnaire. Variance Between In-class and Pull-out Programs F Sum of Mean F Proba-Self-concept Squares Squares Ratio bility Physical appear-ance 139.1164 139.1164 2.6835 .1041 General school 4.0623 4.0623 .0720 .7889 Physical ability 78.9818 78.9818 2.2741 .1342 Reading 38.8034 38.8034 .7109 .4009 Parent relations 5.4532 5.4532 .2019 .6540 Mathematics 131.3981 131.9981 1.7586 .1874 Peer relations 324.7549 324.7549 8.3751 .0045* General self 252.5806 252.5806 6.9354 .0096* Total non-academic score 32.4234 32.4234 1.3133 .2541 Total academic score 126.5678 126.5678 5.4486 .0213* Total score 66.0818 66.0818 3.3867 .0682 In-class number = 58 Pull-out number =62 shows significance at a = .OS.

PAGE 170

Table 24 District A Self-Concept Variables Showing Significance at the a .05 Level Peer Relations In-class Pull-out General Self In-class Pull-out Total Academic Score In-class Pull-out Table 25 Mean 28.2241 31.5161 30.0 32.9032 30.5828 32.6379 Combined Districts MANOVA Grade, Sex, Ethnic and Variable Physical appearance General school Physical ability Reading Parent relations Mathematics Peer relations Total score Standard Deviation 6.5295 5.9306 6.6912 5.3492 4. 7411 4.8920 Standard Error .8574 .7532 .8786 .6793 .6225 .6213 for Effects of Self-Concept Program F Probability Ratio ofF 1.35140 .248 .15351 .696 .03312 .856 .02966 .864 .17355 .678 .28446 .595 .15356 .696 .35054 .555 153

PAGE 171

154 no significant effect of the independent variable of program type, in-class and pull-out, on the dependent and self-concept variables. Results of the effects of each variable on each of the eight self-concept variables in the MANOVA were too extensive to be shown in Table 25. Therefore, the effects of the interactions of the independent variables of program type and classificatory variables with each self-concept variable is shown in Table 25. The probability of F shows that all of these interactions were not significant at the a = .OS level of significance. No significant effect of the independent variable of program type, in-class and pull-out, on the dependent and classificatory variables was found. Program type does not show a negative interaction with student selfconcept. The variable of physical appearance which had a depressed F = .248, as compared to the other variables shows a marked difference for program type by ethnicity. Factors from the MANOVA for combined districts of program type by ethnicity contained a significant F score of .007 for students' concept of their physical appearance. A comparison of average group means, Table 26 shows that students in the in-class program, as a group, have a lower self-concept score in the area of physical appearance than students in the pull-out program. Non-Hispanic in-class students showed a lower mean average, 29 .8, than nonHispanic pull-out students, 31.73.

PAGE 172

Table 26 District A MANOVA Comparison of Group Means by Ethnicity and Type of Program Hispanic X Non-Hispanic X F Probability = .007 Major Findings of MANOVA for Self-Concept In-Class Pull-Out 32.16 32.38 29.79 35.22 The MANOVA for self-concept was. utilized to show whether pull-out students (178) would score higher than in-class 155 students (58) on the self-concept variables. In-class students scored significantly lower than pull-out students on one self-concept variable only, physical appearance. Non-Hispanic in-class students showed a lower mean average, 29.8, than non-Hispanic pull-out students, 35.22. Differences were not significant for Hispanic students. When physical appearance was combined with other non-academic variables as measured by the SDQ, no significance was found by program type. Therefore, there was no significant effect by program design, in-class versus pull-out, on student self-concept for the total student sample.

PAGE 173

Summary Statement of Problem Analysis Students making similar gains academically would be 156 expected to have similar self-concepts as learners unless some other variable, such as program design, was impacting the student's progress or self-concept. The study was designed to assess whether students' self-concept was affected by being placed in a program design that would negatively affect the student's ability to perform academically. The student sample was shown to have made similar academic gains on the NCE posttests for both program designs. Students from District A in-class program scored lower than pull-out students in three subtest areas--peer relations, general self, and total academic self-concept. Further, in the combined districts' MANOVA, in-class students, particularly non-Hispanics, scored significantly lower on the subtest of self-concept of physical appearance. Results indicate that in-class students from the single district sample and from the combined districts' sample tended to have more areas of self-concept where they scored lower than pull-out students. It cannot be determined if in-class students compare themselves to the regular education students in the mainstreamed regular education classroom while pull-out students compare themselves to their peers in Chapter I. Since both groups do equally well academically, differences in self-concept subtests may be

PAGE 174

157 attributed to the type of Chapter I program design, in-class or pull-out design, that delivers the services. The low subtest score of peer relations is of special import when considering the self-concept of a Chapter I student in the mainstreamed situation who may feel singled out from his(her regular education peers. Therefore, results seem to indicate that while both program designs were effective in improving student academic achievement, subtests indicate that the pull-out program students were positively affected by the design, as shown by higher subtest scores in the areas of physical appearance, peer relations, general self and total academic score. Results of subtests, however, were only indicators of possible areas of differences in the effectiveness of program design for improving student self-concept. Results of total tests showed no significant difference in the self-concept of students in either program design, pull-out or in-class. No significant difference was found in the academic achievement or the self-concept of Chapter I students due to program design, pull-out or in-class. Summary Chapter I students were assessed to determine if either program design, pull-out or in-class, negatively affected students' academic achievement or self-concept. Findings supported Research Hypothesis I and Research Hypothesis II. No

PAGE 175

158 significant difference was found in student academic achievement or in student self-concept at the a = .OS level of significance. Research Hypothesis I: There was no significant difference in the academic achievement of Chapter I students in the pullout program and the Chapter I students in the in-class program. In-class students scored significantly lower as a group on the NCE pretest, indicating within-subjects differences in the selected sample at the time of the pretest. However, both groups, in-class students and pull-out students, made approximately equal gains on the NCE posttests at the end of the school year. Results show that students from both program designs improved academically. Therefore, no significant differences were found by program type in the effectiveness of the Chapter I programs for improving student academic achievement. Research Hypothesis II: There was no significant difference in the self-concept of Chapter I students in the pull-out program, as measured by the Self-Description Questionnaire, and the self-concept of the Chapter I in-class students. Although a direct correlation between student academic achievement and student self-concept has not been demonstrated through research, the relationship between the achievement of the student and hisfher self-concept is one that can be tested. Findings of student academic achievement tests showed no difference by program design, pull-out or in-class. Similarly,

PAGE 176

159 no differences were found by program design in student selfconcept. Although in-class students scored lower in the selfconcept areas of physical appearance, peer relations, general self and total academic score, differences were not significant. Therefore, both program designs, in-class and pull-out, were equally effective in improving academic achievement and student self-concept for the Chapter I students: that. were contained in the sample. Analysis of the Subproblem Two program designs, pull-out and in-class, were compared in this study to determine both the effects of the designs on students and to determine if the effects were corrobo-rated by the opinions of teachers utilizing the design. A statewide survey of Chapter I coordinators supplemented and enhanced the view of the teachers regarding the benefits and detrimental effects of the two program designs. Teacher and coordinator survey results were utilized to address the subproblem of the study. A subproblem of this study was included to determine the support among school personnel for the two (pull-out and in-class). This could have long-range implications for future decisions on which approach will become established practice. Which program, and for what reasons, was favored by -.

PAGE 177

160 the Chapter I teachers, the Chapter I coordinators, and the regular classroom teachers? No significant differences were found in the academic achievement or self-concept of the Chapter I students in the pull-out program as compared to the in-class program. Program decisions based upon the effect of design on Chapter I students cannot be recommended as a result of the research findings for Hypothesis I and Hypothesis II. To develop a further approach of design comparison, the teachers of the student sample were included in the study. Survey show differences in reasons to support or reevaluate a Chapter I program design as a method of service delivery to the students. The subproblem of the study focused on which program design was favored by the teachers of the student sample population. Coordinators of the State of Colorado were also surveyed to determine which program design was favored, and the positive and detrimental effects of both program designs were described by teachers and coordinators to determine the reasons for favoring the pull-out or the in-class design. Analysis of these effects were reported for teachers and coordinators separately, and then the results of both surveys were compared. Relevant Questions of Teacher Survey Results Eighty teachers responded to the Teacher Question-naire. All of the Chapter I teachers and regular education

PAGE 178

16l teachers of the student sample population responded. Twentyeight Chapter I teachers responded while 52 classroom teachers responded to the survey. A small sample of teachers from District A were labeled as a mixed or alternative model program design since they occasionally pulled out in-class students for small group instruction. Written responses by the teachers were categorized and tabulated. According to congruity of teacher statements, a category may be one generalized statement or a combination of phrases from teachers explaining or describing the category. Written and tabulated responses are displayed in the following tables. Questions are stated and responses are discussed incorporating the corresponding tables. The total number of teachers who responded to a given question is labeled "total" in the tables. Responses by coordinators employing in-class designs in their districts are labeled "in-class" in the tables, while those using no "in-class designs" are labeled as "pullout." Questions 3 to 11, 19 and 23 of the teacher questionnaire were analyzed regarding the status of the Chapter I program. Questions 7 to 12, 15 and 19 of the Coordinators Questionnaire were also analyzed. They were discussed in the survey results. Then questions from both questionnaires were compared and analyzed. Results and a discussion of the remaining survey questions are contained in Appendix K.

PAGE 179

Question 1. Who decided on the program design in your building, "pull out" versus "in classroom" instruction? As seen in Table 27, the responses to this question occurred with in-class teachers who reported that 75% of the 162 time administrators made this decision, while principals did so 62.5% of the time. Teacher involvement in the decision regarding program design was shown to be minimal, with in-class teachers at 31.3% as compared with pull-out teachers at 44%. Chapter I teachers reported that principals (61.5%) and teachers (57.1%) were more often involved in program design decisions as compared to classroom teachers, with 34.6% and 34.6%, respectively. Therefore, Chapter I teachers, especially in-class teachers, stated that administrators more often .. made the decision without including the teachers who are affected by the decision and who would be providing the services. Question 4. For what reasons was the design selected? As shown in Table 28, unlike responses to Question 3, when comparing in-class and pull-out teachers, over half agreed on two major reasons for selection of program design--professional judgment about instruction and teacher preference. When compar-ing Chapter I teachers with regular classroom teachers,

PAGE 180

163 Table 27 Persons 'Who Decided Program Design in Building Chapter I Classroom Pull Out In Class Other Number of teachers 28 52 54 16 10 Central 3.6% 11.5% 11.1% 75% 10% Administrator (1) (6) (6) (12) (1) (Superinten-dent, Director of Curriculum) Principal 61.5% 34.6% 29.6% 62.5% 80% (16) (18) (16) (10) (8) Chapter I Coordinator/ 21.4% 21.2% 24.1% 12.5% 20% Director (6) (11) (13) (2) (2) Teacher/ 57.1% 34.6% 44.4% 31.3% 40% Teachers (16) (18) (24) (5) (4) (Chapter I teachers, Regular education teachers, Teaching staff) Parents 7.1% 32.7% 27.8% 25% (2) (17) (15) (4) Tradition 32.1% 25% (0) (17) (0) (4) School 7.1% 32.7% 27.8%

PAGE 181

164 Table 28 Reasons for Selection of Design Chapter I Classroom Pull Out In Class Other Number of teachers Time of instruction Space available Research findings Varying Remedial Needs Professional judgment about instruction Number of staff Number of pupils Preference of teachers Professional judgment about curriculum To cut excess costs 28 39.3% (11) 57.1% (16) 21.4% (6) 35.7% 82.1% (23) 10.7% (3) 39.3% (11) 67.9% (19) 42.9% (2) (0) 28.6% (8) 52 38.5% (20) 40.4% (21) 4.6% (5) 44.2% SO% (26) 7.7% (4) 36.5% (19) 44.2% (23) 15.4% (8) 36.5% (19) 36.5% (19) 54 44.4% (24) 40.7% (22) 9.3% (5) 51.9% 57.4% (31) 9.3% (5) 38.9% (21) 51.9% (28) 20.4% (11) 31.3% (17) (0) 16 10 12.5% SO% (2) (5) 18.8% SO% (3) (5) 31.3% 90% (5) (9) 18.8% 20% 68.8% 70% (11) (7) 6.3% 90% (1) (9) 37.5% 70% (6) (7) 56.3% 50% (9) (5) 37.5% 30% (6) (3) 18.8% (3) (0) 18.8% 30% (3) (3) aother was checked, but responses were either not explained or responded to with "I don't know."

PAGE 182

165 an overwhelming 82.1% of Chapter I teachers chose professional judgment about instruction, while 50% of classroom teachers believed in this decision. Further, over two-thirds of Chapter I teachers, 67.9%, stated that preference of teachers was a reason for design selection. A smaller number of classroom teachers, 44.2%, indicated selection by teacher preference. Professional judgment about curriculum was a greater issue for Chapter I teachers (42.9%) as compared to a small number of classroom teachers (15.4%). Findings indicated that professional judgment about instructional issues and the issue of professional judgment about curriculum and curricular congruence were of greater concern for Chapter I teachers than for regular teachers, while cost was not an issue for Chapter I teachers, 0.0%. If space available can be considered as a cost effectiveness item, then space for the program was considered as a posi tive, or beneficial, reason for program design selection by Chapter I teachers (57.1%) Most of these Chapter I teachers were pull-out teachers (40.7%) as cqmpared to in-class (18.8%). Teachers of students in pull-out programs (44.4%) were also far more concerned about time of instruction than in-class teachers (12.5%) as a cost effectiveness issue. Since in classroom teachers serviced students during regularly scheduled class times, it was reasonable that time of instruction would not be a concern. However, scheduling time when students could

PAGE 183

be pulled out of the regular classrooms remained an issue for nearly half of the teachers in a pull-out design. Question 5. What are the most beneficial effects of the Chapter I program? All of the responses to Question 5 are represented as 166 categories in Table 29. Each response was made by at least one teacher. There were 16 categories of beneficial effects as stated by regular and Chapter I teachers from both program designs. Every response was considered important, although all responses are not discussed. Results, as compared to research findings shown in Tables 3 and 4 of this study, will be dis-cussed in a comparison of teacher and coordinator surveys. Table 3 and 4 of this study show both beneficial and negative effects of Chapter I program designs, as stated in the research reviewed in Chapter II. Small pupil-to-teacher ratios were considered by Chapter I teachers (46.4 percent) in all program designs to be.a bene-ficial effect of the Chapter I program. This included 50% of pull-out teachers, 37.5% of in-class teachers, and 60% of the mixed program teachers. Programs with small pupil-to-teacher ratios also have the opportunity to individualize instruction

PAGE 184

Table 29 qucbtiou 5r-The Most Beneficial Effects of Chapter I Program Teachers Chapter I Classroom Pull Out In Class Other

PAGE 185

Increased self-concept/ self-confidence/self-worth/ self-image peers jealous of Chapter I students/want to get into Chapter I) Individualization to meet student need (increased/intense instruction in specific learning area/ low reader receives extra teaching to student strengths to improve areas of weaknessi increasing sense of language without pressure of regular education repetitioni reinforcement of skills for slower studenti reinforcement of reading strategie&i additional help in content math instruction begun at concrete. manipulative level) 32.1% (19) 60.7% (17) 5.8% ()) 5.8% ()) 14.8% (8) 44.4% (24) 25% (4) 50% (8) 30% (3) 70% (7) t-' 0'\ 0)

PAGE 186

Qualified teachers providing 14.3% the best possible instruction (4) (certified reading teachers/ experts; knowledgeable/expert/ positive attitude; creative/ ". f)exible/positive attitude; creative/flexible/outstanding; cooperative. with other teachers/ sharing materials; Chapter I teacher models effective instruction; assures students receive best education) Constant & profitable communication/regular teacher 21.4% & Chapter I teacher (6) (closer coordination of program) 5.8% 9.3% (J) (5) 5.8% 7.4% (3) (4) 31.3% (5) 25% (4) 10% (1) 10% (1) ..... 0\ \0

PAGE 187

Improved academic success 21.44 5.8% 27.8% 25% 30% of students (6) (3) (15) (4) (3) (results generalized to regular classroom; basic skills improved to grade level/ vocabulary; Jonger budget than regular classroom; improved reading/math ability; test of program during year; remediation of reading deficiency/use of grade level objectives; improved NCE scores; improved content area information/ reading Independent fLnding/self-5.8% supporting programs (0) (3) (0) (0) (0) Instruction baaed on research 10.7% 5.8% 5.6% 18.8% (diagnosis of specific (J) (J) (J) (3) (0) learning problems; students test in and out of program; goals set for each student) ....., 0

PAGE 188

School work at student ability 5.8% 1.9% level/lndlviduallzed (0) (3) (1) (0) (0) (extra time on specific skills) Lack of peer 5.8% 3.7% (time for talking with (0) (3) (2) (0) (0) students/discussion of problems in regular classroom, no labeling for students leaving program) Different teaching styles/ approaches to reinforce student 25% 5.8% 14.8% 18.8% 10% skills (7) (3) (8) (3) (1) (oral discussions/movement activities/oral reading/reading leveled/increased motivation; computer time/large amounts) Increase In support 3.6% 5.8% 6.3% (from Chapter I coordinator/ (1) (3) (0) (1) (0) excellent support; from parents/ jn-claasroom 1-rogram; conferencing with parents throughout year) ...... -...I ......

PAGE 189

Table 29 (Continued) More etudente serviced 11in-classroom11 setting (Chapter I aid goes into regular classroom1 less mobility of pupils throughout building) Increased communication (in-classroom/student problems or progress discussed daily; Chapter I and regular teacher planning together/plan units; Chapter I teacher supplements/reteaches/ activities of rLgular program; opportunity for team teaching in classroom) Chapter I Jo.n. (J) 35.7% (10) Classroom 5.8% (J) 5.8% (J) Teachers Pull Out 3.7% (2) 1.9% (1) In Class 12.5% (2) 37.5% (6) Other (0) 40% (4)

PAGE 190

173 to meet student need. The greatest number of Chapter I teachers (60.7%) indicated that individualization in all program designs was a most important beneficial effect of the Chapter I program, pullout 44.4%, in-class 50.0%, and mixed 70.0%. Regular classroom teachers, however, did not appear to recognize the benefits of either a small pupiltoteacher ratio (5.8%) or individualization to meet student need (5.8%) as positive aspects of Chapter I design. Differences in response occurred in an area of interest to this study-communicationthat combined two categories. First, Chapter I teachers (21.4%), or six teachers, stated that constant and profitable communication between the regular and Chapter I teachers was a benefit. Numbers were approximately equal, with three classroom teachers, four pull-out teachers, and four in-class teachers making this claim. Only one teacher from the mixed program design responded in this category. However, when increased communication was considered as a second category, an additional 35.7%, or 10 Chapter I teachers, responded. A larger percentage of in-class teachers responded (37.5%), or six teachers, as compared to pull-out (1.9%), or one teacher, and 40%, or four teachers, for the mixed program. In the communication categories, in-class Chapter I teachers stated benefits in teacher communication that were not recognized by regular classroom teachers.

PAGE 191

174 Differences in percentages of teachers occurred in three categories of response. However, actual numbers were equal or approximately equal. In the category of research as a basis for the instructional design, 18.8% of the in-class teachers (three) and 5.6% of the pull-out teachers (three) responded, while three of these were classroom teachers and three were Chapter I teachers; no teachers from the mixed design responded. A larger percentage of in-class teachers (31.3%) responded that qualified teachers were providing the best possible instruction --a total of five teachers. The same number of pull-out teachers (five) or 9.3%, similarly responded, and of these teachers, four were Chapter I and three were regular classroom teachers. One teacher from the mixed design also .responded in this category. A very interesting response occurred in the category, stating that more students were serviced in an inclassroom setting. In-classroom teachers (12.5%), or two, and Chapter I teachers especially (10.7%), or three, responded positively. However, two teachers from the pull-out program, or 3.7%, also commented that the more students serviced in an in-class setting, the more positive or beneficial effect of program design. The two categories of student academic success and student self-concept were of particular interest to this study. Approximately equal percentages of pull-out teachers (27.8%), in-class teachers (25%) and mixed program teachers (30%) stated

PAGE 192

175 that students in their program experienced improved academic success. Student sample results show that students in pull-out and in-class designs made approximately equal academic gains. Teachers in both program designs recognized student improvement; however, only 5.8% of classroom teachers mentioned academic success. In the category of improved student self-concept, Chapter I teachers (32.1%), or 19 teachers, agreed that Chapter I improves student self-concept. Only 5.8%, or three classroom teachers, responded, and three teachers (30.1%) of mixed program teachers. In-class teachers at 25%, or four, as compared with pull-out teachers (14.8%), or eight, are not remarkable in number. Results show that Chapter I teachers from both program designs can recognize improved self-concept in Chapter I students. Question 6. What are the most detrimental effects of the Chapter I program? As shown in Table 30, the most detrimental effect of the Chapter I program, according to regular classroom teachers (51.9%) and to Chapter I teachers (25%), was time missed from the regular classroom instruction and activities. This time missed was a greater problem for teachers in the pull-out program (57.4%) than in-class teachers (12.5%). Teachers com-plained that students missed subjects of interest and special

PAGE 193

176 events resulting in negative feelings of students toward their program. Finding time for Chapter I instruction and scheduling problems were also detrimental to teachers in the pull-out program (14.8%), but were not reported as a problem for in-class teachers (0). Teachers' schedules, as well as students' schedules, were a problem for two pull-out Chapter I teachers (3.7%) as compared to in-class (0.0%). Another difficulty that seemed to be experienced by three pull-out teachers, but no in class teachers, was that the Chapter I program was only supplemental in design and did not coordinate with regular education curriculum, objectives, expectations, or materials. In-classroom teachers experienced detrimental effects of program designs different from those of pull-out teachers. Chapter I teachers (17.9%) reported difficulty with regular classroom instruction being interrupted by the Chapter I program being conducted simultaneously. Further, lack of teaming, communication, and coordination required for the in classroom program was reported by 14.3% of Chapter I teachers, one classroom teacher, and one regular education teacher, but no pull-out teacher reported a communication problem. Direct detrimental effects as experienced by students were minimal. Students' self-concept may have been negatively affected by the student being embarrassed by one-on-one

PAGE 194

Table 30 Question 6"": The Most DetrJmental Effects of the (:hapter I Program Teachers Chapter I Classroom Pull Out In Class Other

PAGE 195

Extensive paper work (required by federal regulations; requiring more time to prove what is done than time on actual project) Student embarrassed by one-on-one instruction (student desires to be able to do "everything") Labeled as different from the norm (stigma of being a special reader11 by a few teachers/peers regardless of performance; identified as special/11dumb"/older students in front of other classmates; discriminated against; causes self-esteem and self-image problem&; students in grades viewed as behavior problems/not serviced in Chapter I) 14.3% (4) (0) (0) 13.6% 1.9% (1) (1) (0) 10.77. 11.57. 11.1% (3) (6) (6) (0) 6.3% (1) 6.3% (1) (0) 10% (0) 207. (2) t-' -...1 CD

PAGE 196

Chapter I program only supplemental 3.67. 3.8% 5.67.. (coordinated with regular (1) (2) (3) (0) (0) education curriculum; not effective; should/does not support classroom objectives and expectations1 students not pulled for reading or mathematics repeat subject/different materials/ assures failure) Loss of staff 3.67. 15.4% 6.3% (due to exemplary student (1) (8) (0) (1) (0) performance. ncerls assessment. cuts) Time is limiting/problems due 14.37. 7.7% 14.8% to scheduling (4) (4) (8) (0) (0) (for instruction/as little as 30 minutes/15 minutes; need more time in basal reading program; wfth regular teacher/ time limited; A.M. preferred to teach reading/not always possible; leas time for work in class) t-' ....... \C

PAGE 197

No remediat:l on (learning gaps in informa-tion for succeeding years; students remaining in program grades 1-6; lack of student ability/slow learner/little chance of improvement; less success/miss spelling every day; lack of motivation; poor reading skills due to self-fulfilling prophecy Teacher schedule impacting effectiveness (moving among classrooms for in-class programs; changing grade level instructor; transporting lesson plans and supplies) 1. 1% 7.7% (2) (4) 10.7% (3) (0) 9.3% (5) (0) J. 7'%. (2) (0) 107. (1) 107. (1) ...... co 0

PAGE 198

Table 30 (Continued) Interruption of regular class instruction (in-class/two lessons/one class simultaneously) Chapter I 17.9% (5) Classroom (0) Teachers Pull Out In Class Other (0) (0) (0)

PAGE 199

182 instruction. One regular education teacher and one Chapter I teacher, in-class, reported such an effect, while no pull-out students were reported as negatively affected academically. Lack of academic remediation was reported by 9.3% of the pull-out teachers, regular and Chapter I teachers, and not reported by in-class teachers. Effects on students from external causes were that students who remained in the program from one to six years showed remediation insufficient to exit the program. One teacher stated that poor reading skills were due to the self-fulfilling prophecy effect on students. Another teacher reported gaps in student information and learning due to successive years in the Chapter I program. Causes for student lack of remediation were lack of motivation, lack of student ability, and little chance for improvement in the slow-learner student. A small number of teachers (9.37% of the pull-out teachers, no in-class teachers, and no teachers in the mixed design group) made strong statements regarding detrimental effects of the pull-out program design. The majority of teachers reported no direct detrimental effects for student achievement or for self-concept. Question 7. What problems are impacting your success with the program? All problems identified in the study were considered important as viewed by the teachers. Therefore, all problems, as frequency of responses to Question 7, are described in Table

PAGE 200

183 31. No major differences were noted between the program and in-class designs. However, two categories of response were of particular interest when comparing the two program designs. First, only Chapter I pull-out teachers (14%) responded that there were problems with student mobility resulting from students being pulled out of class or students moving within the building or out of district. Secondly, teachers in all categories responded that the noise level with the in-class program impacted program success, and pull-out as well as in-classroom teachers stated that the impact was significant enough to require change from in-class to the pullout program design. Differences between Chapter I teachers and regular classroom teachers are also of interest to this study. Problems of students missing regular classroom instruction were identified by 11.5% of regular classroom teachers as a serious problem. More pull-out teachers (six or 11.1%) _explained difficulties with students missing regular classroom instruction to experience Chapter I services as compared to in-class teachers (three or 18.8%). Chapter I teachers, however, stated problems outside of the classroom setting as impacting their success with the programs that were not stated by regular classroom teachers. For example, limited parent involvement or response was noted by 17.9% of Chapter I teachers and 1.9% of

PAGE 201

Table 31 -Question 7:-Problema Impacting Success with the Program Teachers Chapter I Classroom Pull Out In Class Otl'!er

PAGE 202

Lack of planning time 14.3'7.. 407. (to conference with regular (4) (0) (0) (0) (4) teachers, two school assign-ment&i for diagnostic work) I.arge total pupil/Chapter I 14. 37. 13.5% 11.17. 25% 10% teacher ratio (4) (7) (6) (4) (1) (many students with needi lack of Chapter I teachers/ students on waiting Jisti lack of teacher time, 3/4 time math teacher with full time number of students) Limited parent involvement/ 17.9% 1.9% 5.67. 18.87. 30% re&llOn&e (5) (I) (3) (3) (3) (parent dissatisfaction with placementi little help and support for students at home, lack of convenient accPss to phone for parent contact) t-' 00 V1

PAGE 203

Lack of Chapter I at middle school level Noise level of regular classrooms/ limite activities in-class (two teachers teaching in one classroom/dl fficult; nightmare due to noise in one fifth grade classroom/change from in-class to pull-out needed) Conflict education teacher and Chapter I teacher (in-class confusion/how to assist wiLh regular instruction; pull-out/regular education teacher lacks instruction about pull-out program; Chapter I teacher in-class/asked to do aclivities with the whole class; classroom teachers who do not support Chapter I must be pressured by principal) 7.1% (2) 21.4% (6) 3. 7% (10) (0) 4. 77 .. (4) 9.6% (5) 1.9% (1) 5.6?. (3) 7.4% (4) (0) 31.3% (5) 12.5% (2) 10% (1) 20% (2) 10% (1) ...... CD 0\

PAGE 204

Table 31. (Continued) Time is limiting for Chapter I teacher blocksa limits activities that should occur daily; too much time on discipline; detracts from instr.uct ion) of parental support (too much television) Negative impact on regular classroom (teacher change of style so children leaving do not miss reading instruction; cannot remove students from in-classroom instruction/lack of space in building for an activity with more noise; in-class computer work and special projects difficult; difficult to develop individual rapport with students/in-class) Chapter I (0) (0) 10.7% (J) Classroom 3.8% (2) 1.9% (l) 1.9% (l) Teachers Pull Out 3.7% (2) (0) 1.9% (l) In Class (0) 6.3% (l) 12.5% (2) Other (0) (0) 107. (l) I-' 00 ......

PAGE 205

188 classroom teachers. Chapter I teachers (14.3%) complained of lack of planning time and 40.0% of teachers from the mixed program design were primarily impacted. Question 8. My students miss no academic subject since remedial instruction occurs during study periods in the regular class. It may be of interest that most of the classroom teachers (94.2%) responded to this question. As shown in Table 32, more than twice as many classroom teachers replied in the negative (63.5%) as replied in the affirmative (30.8%) that their students did miss academic subject instruction. An overwhelming number of pull-out teachers (39, or 72.2%) responded that students missed regular classroom instruction to receive Chapter I services. Pull-out teachers strongly supported the criticism of the pull-out program that time missed from the regular academic was a detrimental effect of the design. A minority of pull-out teachers, 12 or 22.2%, replied in the affirmative. However, in-classroom teachers had the opposite response, with twice as many responding that their students did not miss academic subject instruction--50.0% in the negative as compared to 25.0% in the affirmative. Results show that classroom teachers (63.5%), especially teachers in the pull-out design (72.2%) stated that students miss academic

PAGE 206

Table 32 Question 8: Students Miss No Academic Subject Chapter I Teachers Classroom Teachers Pull-Out Teachers In-Class Teachers Other Yes 47.9% (5) 30.8% (16) 22.2% (12) 50.0% (8) 10% (1) No 35.7% (10) 63.5% (33) 72.2% (39) 25.0% (4) (0) 189 Missing 46.4% (5) 5.8% (5) 5.6% (3) 25.0% (4) 90% (9) instruction in the regular classroom setting to receive Chapter I remedial instruction.

PAGE 207

Question 9. Do your students miss regular classroom instruction to receive Chapter I reading? 190 Responses, as indicated in Table 33, show that classroom teachers (61.5 percent) and a large majority of in-class teachers (93.8 percent) stated that students did not miss social studies as regular classroom instruction to receive Chapter I reading instruction. However, although almost half of the pull-out teachers (48.1%), or 26, responded "no," an alarming 37.0%, or 20 teachers, responded "yes." Results show that most of these teachers were probably classroom teachers, since 26.9%, or 14, of classroom teachers replied "yes" as compared to 25% of Chapter I teachers, or seven. The same pattern of responses occurred for science as for social studies, showing that classroom teachers (52.1%) and a large majority of in-class teachers (93.8%) stated that students did not miss science to receive Chapter I reading instruction (see Table 33). Yet, half of the pull-out teachers responded "no," students did not miss regular instruction, while 37.0 percent of pull-out teachers, or 20, responded "yes." This was the same number that responded to the question regarding social studies. Results again show that most of these teachers were probably classroom teachers, since 61.5%, or 32 teachers, responded "yes." When asked if students missed classes other than social studies and science, an alarming majority of classroom teachers, more specifically pull-out teachers, responded

PAGE 208

191 "yes," 44.2%, or 23, and 55.6% or 30, respectively. However, Chapter I teachers were approximately equally divided on the question, with 35.7% responding "yes" and 39.3% responding "no." In-class teachers (81.5%) overwhelmingly stated that their students did not miss any other class to receive Chapter I reading instruction (see Table 33). With approximately two-thirds of the teachers reporting time missed from classes, regular classroom teachers reported more time missed from class than was reported by Chapter I teachers. Teachers are in agreement that 30 minutes was the mode for amount of time missed, with 42.9% of Chapter I teachers and 38.5% of classroom teachers agreeing on amount of time missed. Question 10. Do your students miss regular classroom instruction to receive Chapter I math? As shown in Table 34, two classroom teachers (3.8%) and two pull-out teachers (3.8%) responded that students miss regular instruction to receive Chapter I math instruction. Results show that Chapter I math instruction does not supplant or replace regular education social studies.

PAGE 209

192 Table 33 Students Miss Regular Classroom Instruction to Receive I Reading Yes No Missing Students miss social studies Chapter I 25% 23.6% 21.4% Teachers (7) (15) (6) Classroom 26.9% 61.5% 11.5% Teachers (14) (32) (6) Pull-Out 37.0% 48.1% 14.8% Teachers (20) (26) (8) In-Class 6.3% 93.8% Teachers (1) (15) (0) Other 60% 40% (0) (6) .(4) Students miss science Chapter I 21.5% 57.1% 24.4% Teachers (6) (16) (6) Classroom 23.1% 61.5% 15.4% Teachers (12) (32) (8) Pull-Out 33.35% 50.0% 16.7% Teachers (18) (27) (9) In-Class 93.8% 6.3% Teachers (0) (15) (1) Other 60.0% 40.0% (0) (6) (4)

PAGE 210

193 Table 33 (Continued) Students miss other classes Chapter I 35.7% 39.3% 25% Teachers (10) (11) (7) Classroom 44.2% 38.5% 17.3% Teachers (23) (28) (9) Pull-Out 55.6% 25.9% 18;5% Teachers (30) (14) (10) In-Class 6.3% 81.5% 12.5% Teachers (1) (12) (2) Other 20.0% 40.0% 40.0% (2) (4) (4) Classes missed Time per day Minutes N Percent Chapter I 22 1 3.6 Teachers 25 1 3.6 30 12 42.9 40 1 3.6 45 2 7.1 Missing 11 39.3 Total 28 100.0 Classroom 25 1 1.9 Teachers 30 20 38.5 35 3 5.8 40 5 9.6 45 5 9.6 60 2 3.8 Missing 16 30.8 Total 52 100.0

PAGE 211

194 Table 33 (Continued) Minutes Frequency Percent Pull-Out 30 25 45.3 Teachers 35 3 5.6 40 6 11.1 45 7 13.0 60 2 3.8 Missing 11 20.4 Total 54 1oo.o In-Class 30 6 37.5 Teachers Missing 10 62.5 Total 16 100.0 Other 22 1 10.0 25 2 20.0 30 1 10.0 Missing .2. 60.0 Total 10 100.0 The same number of classroom teachers (3.8%), or two, and pull-out teachers (3.7%), or two, responded that Chapter I students miss regular education science to receive Chapter I math instruction. The majority of teachers responding stated that students did not miss science instruction. Similarly, teachers responded that Chapter I students did not miss other regular education classes to receive Chapter I math instruction.

PAGE 212

195 Table 34 Students Miss Regular Classroom Instruction to Receive ChaRter I Math Yes No Missing Students miss social studies Chapter I 21.4% 78.6% Teachers (0) (6) (22) Classroom 3.8% 51.9% 44.2% Teachers (2) (27) (23 Pull-Out 3.7% 50.0% 46.3% (2) (27) (25) In-Class 37.5% 62.5% (0) (6) (10) Other (0) (10) 100.0% Students miss science Chapter I 21.4% 78.6% Teachers (0) (6) (32) Classroom 3.8% 50.0% 46.2% Teachers (2) (26) (24) Pull-Out 3.7% 48.1% 48.1% Teachers (2) (26) (26) In-Class 37.5% 62.5% Teachers (0) (6) (10) Other (0) (0) 100.0%

PAGE 213

196 Table 34 (Continued) Yes No Missing Students miss other classes Chapter I 21.4% 78.6% Teachers (0) (6) (22) Classroom 3.8% 50.0% 46.2% Teachers (2) (26) (24) Pull-Out 3.7% 48.1% 48.1% Teachers (2) (26) (26) In-Class 37.5% 62.5% Teachers (0) (6) (10) Other (9) 100.0% 100.0% Classes missed are No additional responses are recorded Time :ger day Minutes Frequency Missing Chapter I 0 (0) 100.0% Teachers Classroom 30 1. 9% 98.1% Teachers (1) (51) Pull-Out 30 1. 9% 98.1% Teachers (1) (53) In-Class 0 (0) 100.0% Other 0 (0) 100.0% One classroom teacher (1. 9%) and one pullout teacher (1. 9%) responded that students missed 30 minutes of regular education classes to receive Chapter I math instruction. The missing

PAGE 214

197 references (98%) indicated that no other students miss regular classes due to Chapter I remediation in math. Question 11. Does the Chapter I program supplant (replace) or supplement the regular education program? Table 35 indicates that three classroom teachers (5.8) in the pull-out program (5.6%) stated that Chapter I instruction supplanted the regular education program. There were no Table 35 The Chapter I Program Supplants or Supplements the Regular Education Program Teachers Supplant Supplement Missing Chapter I 57.1% 42.9% (0) (16) (12) Classroom 5.8% 80.8% 13.5% (3) (42) (7) Pull-Out 5.6% 83.3% 11.1% (3) (45) (6) In-Class 75% 25.0% (0) (12) (4) Other (10% 90% (0) (1) (9)

PAGE 215

198 instances of supplanting found with the in-class teachers or by Chapter I teachers in either program design. Results show that classroom teachers in the pull-out program do not agree with Chapter I teachers that there are no instances of supplanting with Chapter I services. Question 18. The Chapter 1 program design in my building is judged by teachers to be a sensible and beneficial plan. As shown in Table 36, one or two teachers in each category either disagreed or strongly disagreed with the statement above. The exception was the case of the Chapter I math program, as three Chapter I teachers (7.1%) disagreed with the statement. The majority of teachers in all categories agreed or strongly agreed, with a larger percentage of in-class teachers (68.8%) strongly agreeing with the' statement regarding Chapter I reading. Pullout teachers were nearly equally divided, with 44.4% strongly agreeing and 46.3% agreeing. Results of the study clearly show that in most buildings where Chapter I teachers provide services, the teachers judge the design, whether in-class or pull-out, to be a sensible and beneficial plan for students.

PAGE 216

199 Table 36 I Program Design in Building Is Judged Teachers to be a Sensible and Beneficial Plan Strongly Strongly Teachers Agree Agree Disagree Disagree Missing Chapter I Reading 64.3% 21.4% 3.6% 10.7% (18) (6) (1) (0) (3) Math 1.1% 3.6% 7.1% 82.1% (2) (1) (3) (0) (23) Both 3.6% 3.6% 92.9% (1) (1) (0) (0) (26) Classroom Reading 44.2% 44.2% 3.8% 1. 9% 5.8% (23) (23) (2) (1) (3) Math 1. 9% 3.8% 94.2% (1) (0) (2) (0) (49) Both 1. 9% 1. 9% 94.2% (1) (1) (0) (0) (49) Pull-Out Reading 44.4% 46.3% 1. 9% 1. 9% 5.6% (24) (25) (19) (33) (2) Math 3.7% 1. 9% 5.6% 94.4% (2) (0) (1) (0) (51) Both 37.0% 1. 9% 1. 9% 92.6% (21) (1) (1) (0) (SO) In-Class Reading 68.8% 12.5% 6.3% 12.5% (11) (2) (1) (0) (2) Math 6.3% 6.3% 87.5% (0) (1) (1) (0) (14) Both 100.0% (0) (0) (0) (0) (16) Other Reading 60.0% 20.0% 10.0% 10.0% 10.0% (6) (2) (1) (1) (1) Math 10.0% 10.0% 70.0% (1) (0) (2) (0) (7) Both 10.0% 90.0% (0) (1) (0) (0) (9)

PAGE 217

200 Question 23. Overall, do you have any worries or concerns about the program or Chapter I students in relation to program design in your building? Funding was a primary concern for teachers in every cate-gory (see Table 37). Teachers are concerned that as funding for Chapter I decreases, Chapter I services will be curtailed. The exception was four classroom teachers (7.7%) who believed that more teaming and communication between teachers was a priority. Only 3.8% of the classroom teachers (3.8%) and no Chapter I teachers were concerned about how to better meet the needs of students. The concerns over selection of program design are evident in responses to question 23. Chapter I (7.1%), or two, and classroom teachers (3.8%), or two, in the pull-out program design were concerned about the need to change program design as student needs change. Due to problems with the pull-out program model, a change to the in-class model may be more beneficial to students, according to these teachers. Another classroom teacher in the pull-out model was concerned that the pull-out program was not effective and would prefer direct instruction in the classroom. Pull-out teachers (7.4%) stated that Chapter I regulations were preventing improvement in program design.

PAGE 218

Table 37 Question 23z Worries or Concerns About Program or Chapter I Student6 in Relation to Program Design in Building Teachers Chapter I Classroom Pull Out In Class Other

PAGE 219

Chapter I regulation preventing improvement in design (material strictly regulated; supplanting requirement prevents improvement/students miss other classes/spelling. handwriting) More teaming/communication needed (prevented by strenuous Chapter I schedule; to group students; to focus student strengths and weaknesses; to correlate schedules with regular classes) Control issue (Chapter I coordinator needing control; less control by building administrator needed) 3.6% 5.8% 7.4% (1) (3) (4) 7.7% 5.67. (0) (4) (3) 1.9% (0) (0) (1) (0) 6.3l (1) 6.3% (1) (0) (0) (0) 1\.l 0 1\.l

PAGE 220

Funding decreases/services curtailed (many districts with elementary reading only/need for Chapter I math; not enough money for a full year funding in a facility; teachers needed; no aides; no materials/need more books/ trade books; increased numbers needing Chapter I, Jl to 50 percentile range; teacher aide model up to 50 students per day; prior model 25 to JO Jter day; students scoring Jl to 50 could benefit from Chapter I not served; increased special education decrease program effectiveness) 17.97. (5) 5.87. (J) 7.47. (4) 18.87. (3) 10.07. (1) N 0 w

PAGE 221

Need to change program as student needs change/in class model may be 1.n. 3.8% 7.4% more beneficial to students (2) (2) (4) (0) (0) (time spent in pull-out design is a problem; time restriction is limiting) Prefer assistance that students 10.7% 3.7% 10.07. receive in pull-out model (3) (0) (2) (0) (1) (students do better/reading is essential, need to convince students. staff and principal for the future; 90% of time spent on reading/application in regular classroom needed) Expectations by regular education 1.9% 1.9% teachers (0) (1) (1) (0) (0) (too high for Chapter I teacheri remediation time/not sole responsibility of Chapter I teacher) N 0

PAGE 222

Pull out progra is not effective (prefer reading aides/direct. Instruction in classroom with lowest level students; students subjected to different teaching styles/rules) Concerns about in-class program (Chapter I teacher not teaching remedial reading; the need team teach; classroom teacher needs excellent management and sJmJlar philosophies of teachers/how to teach adequate space within the classroom) (0) 7.1% (2) 1. 97. (1) 1.9% (1) 1. 97. (1) (0) (0) 12.5% (2) (0) 10.0'7. (1) I'.) 0 V1

PAGE 223

206 In-class teachers had other concerns. Two Chapter I teachers (7.1%) and one classroom in-class teacher (1.9%) expressed the problems with the in-class program as described in Table 37 as "concerns about the in-class program" and one teacher (10.0 percent) from the mixed design shared these concerns. One in-class teacher and one pull-out teacher expressed concern over control. It seems that the Chapter I coordinator or the building principal needing control was an issue for these teachers. Relevant Questions of Coordinators/Directors Survey Analysis Using an adaptation of the basic instrument, the coordinators/directors of Chapter I programs in Colorado were surveyed. The results of this survey are analyzed here by questions. Question 7. Who decided upon program design(s) in your district, "pull out" versus "in classroom" instruction? As shown in Table 38, the Chapter I coordinator was most likely the person to decide upon the program design in the dis-trict for both pull-out (50.5%) and in-class (22.0%). The

PAGE 224

207 Table 38 Person(s} Who Decided Program Design in Your District Responses Total In-class Pull-out Central Administrator 60.8% 17.6% 43.2% (31) ( 9) (22) Principal 60.8% 19.6% 29.8% (31) (10) (21) Chapter I Coordinator Director 72.5% 22.0% 50.5% (37) (11) (26) Given for "Other" Teacher/Teachers 39.2% 13.7% 25.5% (Chapter I teachers; (20) ( 8) (13) Regular education teachers; Teaching staff) Parents 11.8% 3.9% 7.9% ( 6) ( 2) ( 4) Tradition School Board 2.0% 2.0% ( 1) ( 0) ( 1) pull-out program design was almost as frequently administration (43.2%). Principals (19.6%) and central administrators (17.6%) selected the in-class design nearly as often as the Chapter I coordinators. An interesting result of this study was that teachers' decisions as to program design were their responses to "other" by almost twice as many pull-out program coordinators (25.5%) as in-class coordinators (13.7%). With the exception

PAGE 225

208 of parent input, teachers of the in-class program, or the staff in the building where the in-class program services the Chapter I students, have the least involvement in decisions. Therefore, teachers in the in-class program may have less commitment to the program design than teachers in the pull-out model. Question 8. For what reasons was the design(s) selected? Research findings as documented in Chapter II of this study reported that the pull-out design was primarily implemented nationally for two reasons--to cut excess costs and to comply with federal regulations to supplement and not supplant regular education curriculum. As shown in Table 39, results of Question 8 show that Chapter I coordinators in all instances did not respond to federal regulations as a reason for program design selection, and in only one instance (2%) was a program design selected to cut excess costs. Similar results were found for the in-class design with one coordinator (2%) responding that federal regulations were the reason for design selection and no coordinators responding to cut excess costs. Approximately half (51%) of coordinators with only pull-out programs responded that professional judgment about instruction was a reason for the design selection. A larger percentage of coordinators with in-class programs (23.5%) also selected professional judgment about instruction as the

PAGE 226

209 Table 39 Reason Design Was Selected Responses Total In-Class Pull-Out Time of instruction 51% 15.6% 35.40% (26) ( 8) (18) Space available 37.3% 15.6% 21.7% (19) ( 8) (11) Research findings 21.6% 7.8% 13.8% (11) ( 4) ( 7) Varying remedial needs 49% 11.7% 37.3% (25) ( 6) (19) Professional judgment 74.5% 23.5% 51% about instruction (38) (12) (26) Number of staff 29.4% 12.75% 16.7% (15) ( 4) (11) Number of pupils 31.4% 17.6% 13.8% (16) ( 9) ( 7) Professional judgement 41.2% 13.7% 27.5% about curriculum (21) (7) (14) To cut excess costs 2% 2% ( 1) ( 0) ( 1) Res!!onses given for "Other" Student choice for 2% 2% in-class/not to ( 1) ( 1) ( 0) miss favorite classes Federal regulations 2% 2% ( 1) ( 1) ( 0)

PAGE 227

210 reason the in-class design was selected. The reasons listed in Question 8 were all selected more frequently by pull-out coordi-nators than by coordinators with in-class programs, with the exception of number of pupils where 13;8% selected pull-out and 17.6% selected in-class. Therefore, results indicated that numbers of pupils that can be serviced by the Chapter I program was somewhat more of a concern for in-class program coordi-nators. A larger percentage of the coordinators with the pull-out design expressed a greater variety of reasons for program design selection. Question 9. What, in your are the most beneficial effects of the Chapter I program? The most interesting result of responses to Question 9 was found in the first category of responses in Table 40. Coordi-nators of pull-out programs (41.6%) cited that the most beneficial effects of the pull-out program were the result of a small pupil-to-teacher ratio. Coordinators of the in-class program (9.8%) reported far less frequently the benefits of extra time with students, one-on-one individualized instruction and small groups of students who experience success. Individualization to meet student need was cited by an additional 27.5% of the coordinators of pull-out programs as compared to 9.8% of in-class coordinators. Pull-out coordinators cited beneficial effects in every category, with a

PAGE 228

211 Table 40 Question 9. Most Beneficial Effects of Chapter I Program Small pupil to teacher ratio (extra time with student; one-on-one instruction; small groups; individualized instruction) Supplemental program/ materials Success oriented (positive reinforce-ment; resulting in a positive attitude toward reading and math; student feelings of success generalized to other areas) Increased self-concept/ self-confidence; self-image/self-esteem Individualization to meet student need (increased/ intense instruction in specific learning area; teaching to student strengths to improve areas of weakness; increasing a sense of language without the pres-sure of regular education Qualified teachers providing the best possible instruc-tion (certified reading teachers; knowledgeable/ expert; cooperating with other teachers; assuring that students don't fall "through the cracks") Total 41.2% (21) 7.8% ( 4) 9.8% ( 5) 23.5% (12) 37.5% (19) 15.7% ( 8) Responses In-Class Pull-Out 9.8% ( 5) 2.0% ( 1) 2.0% ( 1) 3.9% ( 2) 9.8% ( 5) 3.9% ( 2) 41.7% (20) 5.9% ( 3) 7.8% ( 4) 19.6% (10) 27.5% (14) 11.8% ( 6)

PAGE 229

Table 40 (Continued) Support from the State Department Improved academic success of students (basic skills improved to grade level; improved reading/ math ability; remediation of reading deficiency; improved NCE scores) Independent funding/selfsupporting programs Instruction based on research (diagnosis of specific learning problems; close monitoring of student progress; non-graded achievement) Schoolwork at student level/individualized Lack of peer pressure (time for talking with students; time to build teacher/student friendships) Different teaching styles/ approaches to reinforce student skills Total 3.9% ( 2) 35.3% (18) 5.9% ( 3) 8.9% ( 4) 7.8% ( 4) 5.9% ( 3) 7.8% ( 4) 212 Responses In-Class Pull-Out ( 0) 7.8% ( 4) 2.0% ( 1) 2.0% ( 1) ( 0) ( 0) 2.0% ( 6) 3.9% ( 2) 27.5% (14) 3.9% ( 2) 5.8% ( 3) 7.8% ( 4) 5.9% ( 3) 5.8% ( 2)

PAGE 230

213 greater percentage of responses; however, since the in-class numbers were significantly smaller than pullout, a similar variety of responses cannot be expected by a smaller number of respondents. However, in two areas of particular interest to this student academic achievement and self-concept, the percentages show notable differences. Coordinators of the pull-out designs (19.6%) stated that improved self-concept was a benefit of the program design, as compared to 3.9% of the in-class coordinators. Similarly, 27.5% of the pull-out coordinators reported improved academic success as a benefit of program design, as compared with 7.8% with in-class. Results show, therefore, that coordinators of pull-out programs more often reported improved student academic achievement and self-concept as benefits of the program design than did coordinators of in-class programs. Question 10. What, if any, are the most detrimental effects of the Chapter I program? Two categories show detrimental effects of the pull-out design as compared to the in-class program design (see Table 41). The first category was the time missed by pull-out students from regular classroom subjects and activities (17.7% for pull-out and 3.9% for in-class Chapter I coordinators). Secondly, when students are pulled out for

PAGE 231

214 Table 41 Most Detrimental Effects of the Chapter I Program Responses Total In-Class Pull-Out Time missed from subject 21.6% 3.9% 17.7% area/activities -(11) ( 2) ( 9) Lack of teaming of Chapter 3.9% 3.9% I teacher and regular (2 ) ( 0) ( 2) education teacher (misunderstanding of Chapter I teacher by regular teacher) Extensive paperwork 9.8% 5.8% 4% (required by federal ( 5) ( 3) ( 2) regulations; excessive compared to revenue received; requiring more time to prove what is done than time on actual project) Student dependency on 3.9% 3.9% one-to-one instruction ( 2) ( 0) ( 2) (student relief from pressure of preparing for an academic subject) Labeled as different from 21.6% 11.7% 9.9% the norm (stigma of (11) ( 6) ( 5) needing extra help; by a few teachers less of performance; identified as special/ singled out at the secondary levels; cause of self-image problems.

PAGE 232

215 Table 41 (Continued) Responses Total In-Class Pull-Out Chapter I program only 3.9% 2.0% 1. 9% supplemented (coordinated ( 2) ( 1) ( 1) with regular education curriculum though not effective; students not pulled for reading or mathematics repeated subject/different materials; assures failure) Loss of staff 2.0% 2.0% (due to exemplary ( 1) ( 0) ( 1) student performance; due to needs assess-ment; due to cuts) Time is limiting 5.9% 5.9% (for instruction/as ( 3) ( 0) ( 3) little as 30 minutes; for planning student program/Chapter I teacher with regular teacher) Chapter I instruction, time is limiting for the Chapter I teacher to provide the instruction, plan for the instruction, and to plan with the regular classroom teacher, as reported by three coordinators (5.9%) as compared to in-class coordinators. Further detrimental effects were expressed by pull-out program coordinators that were not expressed by coordinators of the in-class program. Two coordinators responded that the regular education teacher and Chapter I teacher did not team and that the regular education teacher misunderstood the Chapter I

PAGE 233

216 teacher. One coordinator reported a loss of staff. Students who perform well academically are dropped from the program, resulting in loss of teachers to cut costs. Finally, two coordinators reported that students became dependent on one-on-one instruction. Students pulled out of a regular class are relieved of the pressure of preparing for an academic class. However, an area of interest to this study was the effect of program design on student self-concept and academic achievement. Coordinators report detrimental effects of program design on academic achievement in one instance for pull-out (1.9%) and one instance for in-class (2.0%). While both coordinators reported the cause to be that the Chapter I program could be only supplemental to the regular education curriculum, the Chapter I program may be coordinated with regular educa-tion curriculum but was not effective. A larger percentage of coordinators reported detrimental effects to student self-concept due to the labeling of Chapter I students as different from the norm. The effect was reported for both program designs, 11.7% for in-class and 9.9% for pull-out programs. One coordinator stated that students were labeled by a few teachers regardless of academic performance. Other coordinators reported that students were identified as special or singled out, especially at the secondary level. Students suffered the stigma of extra help or labeling was considered as a cause of self-

PAGE 234

217 image problems for students in Chapter I programs of both designs. Question 11. What problems are impacting the success of the program? Problems were expressed by coordinators of the pull-out design that are not expressed by coordinators of the in-class program. It should be noted, as for previous questions eliciting a written response, that variety of problems or responses by coordinator may be attributable, in part, to the larger number of districts with pull-out programs (51) as com-pared to 13 with in-class design. All problems are considered important and are listed in Table 42, although those problems common to both program designs will not be discussed. Problems that are stated for pull-out teachers with regular classroom teachers are not expressed by in-class coordinators. Lack of teaming, communication and coordi-nation of Chapter I with regular education are problems impacting the success of the pull-out program for two coordinators, or 3.9%. The regular education teachers resist having students leave the classroom to receive Chapter I services, according to three coordinators (5.9%). In addition, when conflicts arise regarding Chapter I services, five

PAGE 235

218 Table 42 Problems Impacting Success of the Program Responses Total In-Class Pull-Out Lack of teamingjcoordi-3.9% 4.3% 3.9% nation/communication; ( 2) ( 2) ( 2) Chapter I teacher with regular education teacher) Lack of funding 27.5% 2.0% 25.5% (to meet individual (14) ( 1) (13) needs; to do job efficiently; to broaden and expand programs; allocations based on 1980 census/not fair to growing districts; not reflective of the income level of dis-tricts for past 6 years) Regular education 5.9% 5.9% teachers (resistance ( 3) ( 0) ( 3) to student participa-tion in Chapter I (resist students leaving class; pile on missed work; reluctant to participate/in class-room instruction) Lack of parental consent/ 9.8% 2.0% 7.8% support/interest/involve-( 5) ( 1) ( 4) mentjconcern; to imple-ment program)

PAGE 236

Table 42 (Continued) Chapter I confused with special education (dumping ground for problem students; parents sometimes feel Chapter I is special education; confusion regarding differentiation between the two programs) Extensive paper work Federal regulations, limiting/inflexible/ rigid/time consuming; (require programs to be only supplemental; scheduling difficulties due to requirement not to supplant a class in reading or mathematics; programs limited to reading, mathematics, and language arts; student out of classroom a maximum of 25% of time; minority students with no language skills needing more oral language development) Home environment unstable Total 11.8% ( 6) 2% ( 1) 21.6% (11) 2% ( 1) 219 Responses In-Class Pull-Out 5.9% ( 3) 2% ( 1) 13.7% ( 7) ( 0) 5.9% ( 3) ( 0) 7.9% ( 4) 2.0% ( 1)

PAGE 237

220 Table 42 (Continued) Responses Total In-Class Pull-Out Numbers 9.8% 9.8% (too many students with ( 5) ( 0) ( 5) need and can only serve part; too many students per teacher/per aide; increased teacher cost causing program to be reduced to 166 students in one district) Administrative control at 2.8% 2.0% the building level ( 1) ( 0) ( 1) (trying for too much control of program; control of student movement) Political influence 9.8% ( 0) 9.8% (issues decided on ( 5) ( 5) political rather than educational basis) coordinators (9.8%) report that issues are decided on a political basis rather than by educational rationale. Issues regarding numbers of students and funding to adequately service large numbers of students were expressed by coordinators of the pull-out programs. Too many students with need and the increased cost of supplying teachers and/or aides were problems impacting the success of the program for five coordinators (9.8%). Lack of funding to efficiently meet student need or to broaden or expand program was expressed by an additional 13 coordinators (25.5%). When viewed as

PAGE 238

221 problems that influence program success, lack of funding to adequately service large numbers of students in the pull-out program model was a significant problem for 18 of 51 coordi-nators, or 35.3% of the respondents. Respondents from both program designs reported problems with federal regulations as shown in Table 42, and problems with confusion of Chapter I and special education. Apparently, Chapter I service providers feel strongly that Chapter I should not be a 'dumping ground' for problem students. Part of the confusion between the two programs appeared to be the need to clarify what 'kind' of students should be placed in Chapter I so that problem students remain the domain of special education. Question 12. Do the Chapter I students miss regular classroom instruction to receive Chapter I reading or math? Table 43 shows that there is validity to the criticism of the Chapter I pull-out program design that students miss regular classroom instruction to receive Chapter I services. More coordinators (20 or 30.9%) responded that students did miss regular classroom instruction than responded that they did not (13 or 25.5%). One coordinator responded that students did miss instruction, but not during reading or math time. Another pull-out coordinator stated that missing regular classroom instruction was a greater problem for secondary students.

PAGE 239

Table 43 Do Chapter I Students Miss Regular Classroom Instruction to Receive Chapter I Reading or Math? Responses Total In-class Pull-Out Yes 52.9% 13.7% 39.2% (27) ( 7) (20) No 39.2% 13.7% 25.5% (20) ( 7) (13) Yes, but not during 2% reading or math ( 1) ( 0) More problems at the 2% secondary level, ( 1) ( 0) junior and senior Missing 2% ( 1) 222 Other 2.0% ( 1) 2.0% ( 1) However, of interest to this study was the finding that an equal number of in-class coordinators (13.7% or seven) responded that students missed regular instruction, as compared to 13.7%, or seven, that responded that they did not miss such instruction. Surprisingly, students miss regular classroom instruction as reported by coordinators of districts that have in-class programs. However, since these districts may have both program designs, it was not possible to conclude that students in the in-class program in these districts were directly affected by program design due to missing regular classroom instruction. Results were not considered conclusive.

PAGE 240

223 Question 15. The Chapter I program design in my district is judged by most teachers to be a sensible and beneficial plan. As shown in Table 44, the majority of teachers in both Chapter I program designs, in-class and pull-out, reported that the Chapter I program design in their district was judged by most teachers to be a sensible and beneficial plan. However, one coordinator in the in-class program disagreed that the in-class program was a beneficial plan for students in Chapter I math. One coordinator in the pull-out program also disagreed that the pull-out program design was a beneficial plan for Chapter I students in reading, in math, and in both. Since the number disagreeing was relatively small (2.0%) for each program design, results show that coordinators with both program designs, pull-out and in-class, believed that most teachers judge the Chapter I program design to be a sensible and beneficial plan. Question 19. Overall, do you have any worries or concerns about the program or the Chapter I students in relation to program design in your district? All worries or concerns expressed by coordinators were con-sidered important and are reported in Table 45. More concerns were expressed by coordinators with the pull-out designs, but

PAGE 241

224 Table 44 I Program Design Judged to be Sensible and Beneficial Strongly Strongly Agree Agree Disagree Disagree Missing Reading 47.1% 15.7% 2.0% 35.3% (24) ( 8) ( 1) ( 0) (18) Math 13.7% 5.9% 3.9% 76.5% (7) ( 3) ( 2) ( 0) (34) Both 27.5% 9.8% 2.0% 60.8% (14) ( 5) ( 1) ( 0) (31) In-class Reading 15.6% 12.8% ( 8) ( 4) ( 0) ( 0) Math 5.9% 3.9% ( 3) ( 2) ( 1) ( 0) Both 5.9% 3.9% ( 3) ( 2) ( 0) ( 0) Pull-out Reading 31.5% 2.8% 2.0% (16) ( 4) ( 1) Math 2.8% 2.0% 2.0% ( 1) ( 1) ( 1) ( 0) Both 21.6% 5.9% 2% (11) ( 3) ( 1) ( 0)

PAGE 242

Table 45 Worries or Concerns About Program or Chapter I Students in Relation to Program Design Responses 225 Total In-Class Pull-Out How better to meet student needs (numbers in programs get smaller as students get older; a way to help high school students) How to improve results (importance of applying as many reading and math skills as possible to classroom objectives) Teacher turnover Chapter I regulations preventing improvement in design (supplanting requirement preventing improvement) More teaming/communication needed (prevented by strenuous Chapter I schedule) Control issue (Chapter I coordinator needing control; less control by building administrator) 5.9% ( 3) 3.9% ( 2) 2.0% ( 1) 7.8% ( 4) 3.9% ( 2) 2.0% ( 1) 3.9% ( 2) 2.0% ( 1) ( 0) 3.9% ( 2) ( 0) ( 0) 2.0% ( 1) 2.0% ( 1) 2.0% ( 1) 3.9% ( 2) 3.9% ( 2) 2.0% ( 1)

PAGE 243

226 Table 45 (Continued) Responses Total In-Class Pull-Out Funding decreased/services curtailed (many districts with elementary reading only; not enough money for full-year funding in a facility; no aides; no materials; increased numbers may result in decreased achievement; teacher aide model up to 50 students per day, prior model 25 to 30 students per day) Need to change program as needs change/in-class model may be more benefit to students; (to reflect more of a mainstreamed approach and less pull-out) Prefer assistance in pull-out model; students do better; need to convince students, staff and principal for future) 3.9% ( 2) 2.0% ( 1) 2.0% ( 1) ( 0) ( 0) ( 0) 3.9% ( 2) 2.0% ( 1) 2.0% ( 1) these coordinators were more in number so that responses were more extensive and varied. Of interest to this study was the concern that Chapter I regulations prevented improvement in design due to the federal requirement that Chapter I not supplant regular education curriculum. It was important to note that two coordinators from each program design, in-class and pull-out, stated the

PAGE 244

227 same concern. There was no difference shown by program design (see Table 45). However, the need for more teaming and communication with the Chapter I teachers and regular education teachers was expressed by two coordinators of the pull-out program. The strenuous schedule of the Chapter I teachers was reported as a reason for lack of teaming and communication. This problem was not expressed by coordinators with in-class designs. Comparative Analysis of Teachers' and Coordinators' Surveys Program design decisions. Teachers and coordinators were asked two questions to determine who selected the program design through which they provided services and for what reasons. The teachers provided direct services and must operate in the setting that has been selected. The two questions were asked to determine if teachers preferred a program that they had decided upon and supported, and the reasons for the design decision. Question: Who decided on the program design in your "Pull-out" versus "in-classroom" instruction? Teachers of the pull-out program responded that teachers (Chapter I teachers, regular education teachers, and/or teaching staff) more frequently decided, 44.4% or 24.

PAGE 245

228 Teachers of the in-class program responded that administrators (central administrators (75% or 12, and principals, 62.5 percent or 10) more frequently decided. Coordinators from both program designs responded that administrators, especially coordinators (50.5%, or 26, for pull-out and 22%, or 11, for in-class) decided more frequently than teachers (25.5%, or 13, for pull-out and 13.7%, or 8, for in-class). Overall, teachers were more likely to decide upon the pull-out design, while teachers and coordinators agreed that admini-strators most frequently decided upon the in-class program. Therefore, of those who directly provided services to students, the in-class teachers were least likely to decide upon the in-class program design. Question: For what reason(s) was the design selected? Teachers and coordinators overwhelmingly judged their program design decision to be based upon professional judgment that the design was the preferred service to deliver instruc-tion and to meet curricular concerns. Coordinators responded with 51% pull-out and 23.5% for in-class, the largest number of responses for in-class regarding instruction. Teachers responded with 57.4% for pull-out and 68.8% for in-class. Teachers (42.9%) and coordinators (41.2%) from both program designs also agreed that the program design was selected due to

PAGE 246

229 professional judgment about curriculum, either pull-out or inclass. Teachers and coordinators agreed on three reasons for the selection of the pull-out design to meet concerns that were more frequent concerns largely for pull-out respondents. 1. Finding time for Chapter I instruction. 2. Finding available space for the program. 3. Meeting the varying remedial needs of students. Teachers from both program designs stated teacher preference as a reason for program design selection. Coordinators did not cite teacher preferences as a reason for design selection. However, coordinators (39.2%) did respond previously that the teachers had participated in the program design decision. It is an important finding that in-class teachers (56.3%) and pull-out teachers (51.9%) considered teacher preferences as the reason for the program design that was selected. Numbers of staff was a greater concern for coordi-nators (29.4%) than for teachers (10.7%), but numbers of pupils was a concern for respondents equally for both program designs --coordinators (31.4%) and teachers (39.3%). In-class teachers (31.3%) and pull-out coordinators (13.8%) were the only respondents interested in research findings as a reason for program design selection.

PAGE 247

230 Coordinators and teachers agreed that the reasons for program design selection were professional judgment about instruction and curriculum. A little over half of the teachers of both program designs cited teacher preference as a reason for the selection. Problem-solving as a reason for design selection was more frequently cited by pull-out respondents. The pull-out program was more likely to be selected when there were problems with time for instruction, space available, and varying remedial needs of students. Benefits of both program designs. The benefits of both program designs, as explained by coordinators and teachers, were individualization to meet student need, small pupil-toteacher ratio, and the improved academic success of the student. Respondents stated that both Chapter I program designs benefitted students due to the individualized instruction and attention to the variety.of student remedial needs. Responses from the teachers (44.4% pull-out and 50% in-class) were: Individualization to meet student need (increased/intense instruction in specific learning area/reading; low reader receives extra help/attention; teaching to student strengths to improve areas of weakness; increasing sense of language without pressure of regular education; repetition; reinforcement of skills for slower student;

PAGE 248

reinforcement of reading strategies; additional help in content areas; math instruction begun at concrete, manipulative level) 231 Responses summarized from coordinators (27.5% pull-out and 9.8% in-class) were: Individualization to meet student need (increased/ intense instruction in specific learning area; teaching to student strengths to improve areas of weakness; increasing a sense of language without the pressure of regular education. A small pupil-to-teacher ratio was cited by all respondents, but was a somewhat greater benefit for pull-out teachers (50%) as compared to in-class (37.5%), and for pullout coordinators (41.7%) as compared to in-class (9.8%). Responses by teachers were: Small pupil-to-teacher ratio (extra help/attention/ time with student; one-on-one instruction; eliminating distractions/noise; small groups/consistent; emotional support/concern/caring). Responses by coordinators were: Small pupil-to-teacher ratio (extra time with student; one-on-one instruction; small groups; individualized instruction). Improved academics was the third largest category of responses, with approximately equal percentages of pull-out

PAGE 249

232 teachers (27.8%) and in-class teachers (25%) reporting this benefit to students. More pull-out coordinators (27.5%) than in-class (7.8%) cited this benefit. Teachers' responses were: Improved academic success of students (results generalized to regular classroom; basic skills improved to grade level/ vocabulary; longer budget than regular classroom; improved reading/math ability; test of program during year; remediation of reading deficiency/use of grade level objectives; improved NCE scores; improved content area information/reading. Coordinators' responses were: Improved academic success of students (basic skills improved to grade level; improved reading/math ability; remediation of reading deficiency; improved NCE scores). Benefits of pull-out design. Coordinators (19.6%) of the pull-out programs and teachers of pull-out programs (14.8%) more frequently cited improved self-concept as a benefit to Chapter I students, as compared with in-class coordinators (3.8%) and pull-out coordinators (5.8%). Coordinators were asked to describe ways in which the Chapter I program design affected the self-concept of students. There were five categories of response where larger percentages of pull-out coordinators than in-class coordinators identified effects of the program design on student self-concept. The largest percentage of coordinators (37.3%, 7.8% in-class)

PAGE 250

233 stated that student self-concept was improved due to Chapter I programs being success oriented. A small number of teachers (11.1%, or 6 for pull-out, as compared to 18.8%, or 3 in-class teachers), mentioned that Chapter I programs were success oriented as a beneficial result of the program. According to coordinators, the success experienced by students improved self concept, and since students feel special, they progress. Smaller numbers of coordinators responded in three categories that students "love" or "feel" good about the Chapter I program and the teachers. Outstanding staff members--caring, supportive, sensitive teachers--give special attention to student needs. Thus, students want and choose to be in the Chapter I program. Results show that pull-out programs may be more beneficial to improving student self-concept than in-class. Benefits of the in-class program. It was a benefit to students in the in-class program that they did not miss regular classroom instruction to receive Chapter I services. Only one teacher responded that students missed social studies, and no teachers responded that students missed science. However, four teachers (or 25.0%) responded that the students did miss some academic subjects. Coordinators were divided equally, with 13.7% saying "yes" students missed and "no" students did not miss. Overall, numbers responding that students missed academic subjects were small, especially when compared to pull-out

PAGE 251

234 numbers. Students were more likely to receive academic instruction in the regular classroom with the in-class design. Qualified teachers with expertise and knowledge related to remedial instruction provided the best possible instruction for students was more frequently cited for the in-class program but not by coordinators. Constant and profitable interchange between Chapter I teachers and regular education teachers was also cited by 25% of the in-class teachers and 7.4% of the pullout. However, the number of teachers (4) was small and improved communication was not mentioned by coordinators. Therefore, results of the benefits of in-class programs were inconclusive. Detrimental effects of the pull-out design. The most frequently mentioned detrimental effect of the pull-out design was time missed by students from the regular program. Teachers responded that students missed subject area instruction as well as subjects of interest and special events or activities. Smaller numbers of pull-out teachers and coordinators cited further detrimental effects that were not cited by more than one or two in-class respondents. These detrimental effects were: 1. Lack of teaming of the Chapter I teachers and regular classroom teachers. 2. Problems due to scheduling so that time for instruction was limited.

PAGE 252

3. A lack of remediation as reported by 9.3% of the teachers was characterized by a small but adamant number of teachers: No remediation (learning gaps in information for succeeding years; students remaining in program grades 1-6; lack of student ability/slow learner/ 235 little chance of improvement; less success/miss spelling every day; lack of motivation; poor reading skills due to self-fulfilling prophecy. Students miss classroom instruction to receive Chapter I instruction. Chapter I coordinators (39.2%) reported that students missed regular classroom instruction to receive Chapter I reading or math instruction, as compared to 25.5% responding that they did not miss classroom instruction. A large percentage of teachers (72.2%), especially classroom teachers (63.5%), responded that students miss academic subjects to receive instruction. Students missed social studies, according to 37% of the teachers. Students missed science according to 33.4% of the teachers. Although 48% responded that students did not miss social studies and 50% responded that students did not miss science, that students were missing instruction was a concern. Those coordinators and teachers that responded verified a criticism of the pull-out design, that Chapter I students miss regular classroom instruction to

PAGE 253

236 receive remedial services and therefore may fall further behind in missed academic areas. Detrimental effects of in-class design. There were no detrimental effects of the in-class design reported in greater frequencies than for the pull-out design by coordinators and teachers. However, the noise level with the in-classroom program was reported as a detriment by 31.3% of the in-class teachers in an open-ended question. Detrimental effects of both program designs. Coordinators from both program designs, pull-out (9.9%) and in-class (11.7%) cited the labeling of Chapter I students as "different from the norm" as a detrimental effect of both program designs. Only one in-class teacher found labeling to be a concern, while four pull-out teachers (7.5%) so responded. Although numbers were small, there was some concern with both program designs that labeling students caused stigmatization and resulted in self-esteem/self-image (self-concept) problems. Opinions regarding program design. No difference by program design was found regarding the academic and social benefits of the Chapter I programs. Results show that teachers and coordinators overwhelmingly agreed that Chapter I program design, whether in-class or pull-out, clearly has academic and social benefits to students. Further, the majority of coordinators and teachers judged the design, whether in-class

PAGE 254

237 or pull-out to be a sensible and beneficial plan for students. Both program designs were in fact clearly not judged to be detrimental to pupil progress and adjustment to school by coordinators and teachers. Open-Ended Questions Five categories of problems impacting the success of the Chapter I program for teachers were mentioned by small numbers of teachers from both program designs. These were extensive paperwork, excessive student mobility, large pupil-to-ChapterI-teacher ratio, limited parent involvement, conflicts with regular classroom teacher and Chapter I teacher, and the negative impact of the Chapter I program on the regular classroom teacher (see Table 31). Two differences resulted by program design that support findings of detrimental effects of program design. First, 31.3% (5) in-class teachers responded that noise in the regular classroom was limiting classroom success. Secondly, 11.1% (6) pull-out teachers (the only ones who commented) responded that students missing regular classroom instruction caused difficulty for the Chapter I students and the teachers. Three categories of problems of a more political nature were described by coordinators in small numbers from both program designs. The first and most frequently mentioned problem, pull-out (21.6%) and in-class (13.7%), was the limitations and inflexibility of the program due to federal

PAGE 255

238 regulations. Chapter I was confused with special education and therefore becoming a "dumping ground for problem students" was a problem for both program designs. Also, lack of training, coordination and communication by Chapter I teachers and regular teachers was also mentioned. Three of problems were mentioned by at least five coordinators of the pull-out design for each category. Lack of funding to meet individual needs, to do the job efficiently, and to broaden and expand programs was an issue for 25.5% (13) pull-out coordinators. Issues decided on a political rather than an educational basis was mentioned by pull-out but not in-class coordinators. Three categories of problems of a more political nature were described by coordinators in small numbers from both program designs. The first and most frequently mentioned problem, pull-out (21.6%) and in-class (13.7%), was the limitations and inflexibility of the program due to federal regulations. Chapter I being confused with special education and therefore a "dumping ground for problem students" was a problem for both program designs. Also, lack of teaming, coordination and communication by Chapter I teachers and regular teachers was mentioned. Three categories of problems were mentioned by at least five coordinators of the pull-out designs for each category. Lack of funding to meet individual needs, to do the job

PAGE 256

239 efficiently, and to broaden and expand programs was an issue for 25.5% (13) pull-out coordinators. Issues decided on a political rather than an educational basis was mentioned by pull-out but not in-class coordinators. A final category was too many students with need, large numbers, and only part can be served. Although more problems were mentioned by pull-out coordi-nators, their numbers were larger than in-class. A greater variety of responses was possible. Therefore, it can not be determined if these problems were greater issues because the pull-out design was selected. Teachers and coordinators had no worries or concerns about the Chapter I program design that showed important differences by program design. Major Findings The major findings of the study answered the two research questions and the subproblem of the study. Research Question 1 Is the pull-out Chapter I program design more effective than the in-class design for improving student academic achievement? 1. Results of an ANOVA for a single district and a MANOVA for combined districts supported Research Hypothesis I. There was no significant difference in the academic achievement of

PAGE 257

240 Chapter I students in the pull-out program and the academic achievement of the Chapter I students in the in-class program. 2. Results of the District A ANOVA showed no significant differences in academic achievement by program design, pull-out or in-class. 3. Results of a combined districts' MANOVA showed that the combined sample was academically similar without regard to program design, pull-out or in-class. No within-subject effects were found by the classificatory variables of sex, grade, and ethnicity. Research Question 2 Is the pull-out Chapter I program design more effective than the in-class design for improving student selfconcept? 1. Results of a District A ANOVA showed no significant effects by program design, in-class versus pull-out, for the total test score for student self-concept. 2. Results of a subtest of the District A MANOVA showed that in-class students were negatively affected by participation in the in-class program in the area of peer relations. 3. Results of three subtests of the District A ANOVA showed that pull-out students scored higher in three areas--peer relations, general self, and total academic self-concept. 4. Results of the combined districts' MANOVA supported Research Hypothesis 2. There was no significant difference found in the self-concept of Chapter I students in the pull-out

PAGE 258

241 program and the self-concepts of Chapter I students in the inclass program for the total test. 5. Results of the combined districts' MANOVA showed significance in one subtest only. In-class students, particularly non-Hispanics, scored lower in the area of selfconcept of physical ability. Subproblem The subproblem of the study asked the following question: Which program, and for what reasons, was favored by the Chapter I teachers, the Chapter I coordinators, and the regular classroom teachers? 1. Teachers were more likely to decide upon the pull-out design. Teachers and coordinators agreed that administrators most frequently decided upon the in-class program design. Therefore, of those who directly provided services to students, the in-class teachers were least likely to decide upon the inclass program design. 2. Teachers and coordinators overwhelmingly agreed that the reason the program design was selected, whether in-class or pull-out, was that the design decision was based upon professional judgment about the preferred service model to deliver instruction and to meet curricular co,ncerns. 3. Teachers and coordinators agreed on three reasons for the selection of the pull-out design: a. Finding time for Chapter I instruction.

PAGE 259

b. Finding available space for the program. c. Meeting the varying remedial needs of the students. 4. The benefits of both program designs agreed upon by 242 teachers and coordinators were individualization to meet student need, small pupil-to-teacher ratio, and the improved academic success of the student. 5. Pull-out programs were more beneficial to improving only 12 of 51 student self-concept than in-class, according to coordinators. 6. A benefit to students of the in-class program as reported by teachers more frequently was that students did not miss regular classroom instruction to receive Chapter I services. 7. The most frequently mentioned detrimental effect of the pull-out design was time missed from the regular program. Other detriments were: a. Lack of teaming of the Chapter I teachers and regular classroom teachers. b. Problems due to scheduling so that time for instruction was limited. c. Lack of remediation. 8. There were no detrimental effects of the in-class design reported in greater frequencies than for the pull-out design. However, noise level in the regular classroom was reported as a detriment by in-class teachers only.

PAGE 260

243 9. Coordinators cited the labeling of Chapter I students as "different from the norm" as a detrimental effect of both program designs. There was some concern with both program designs that labeling students caused stigmatization and resulted in self-esteem/self-image problems (self-concept).

PAGE 261

CHAPTER V SUMMARY, CONCLUSIONS AND RECOMMENDATIONS Chapter V presents statements of the problem of the study, the research questions and hypotheses, and the sub-problem of the study. The major findings, or results of the study, are summarized and conclusions and recommendations are drawn from the major findings of the study. The Chapter I compensatory education program is a federally funded program targeting the need to improve the educational opportunity of economically disadvantaged children. Clearly, the $500 per child provided by the Federal Government through Chapter I cannot provide the entire intervention needed to assist the targeted children. Hence, there is a need to determine the most effective, creative, and responsible service model to meet the needs of these children in the face of the limited funds and the greater expectations of our nation. Problem of the Study The problem of the study sought to compare the effectiveness of the Chapter I pull-out program as the predominant service design to the in-class program design by

PAGE 262

245 comparing student achievement and student self-concept scores from students in each delivery system. Research Questions This study investigated the following research questions: 1. Is the pull-out Chapter I program design more effective than the in-class design for improving student academic achievement? 2. Is the pull-out design more effective than the in-class design for improving student self-concept? Hypotheses To investigate the problem of this study, two hypotheses were tested. Hypothesis 1: There will be no significant difference in the academic achievement of Chapter I students in the pullout program as compared to the academic achievement of Chapter I students in the in-class program. Hypothesis 2: There will be no significant difference in the self-concept of Chapter I students in the pull-out program as compared to the self-concept of Chapter I students in the in-class program. Subproblem A subproblem of the study utilized survey data to determine support among school personnel for the two program designs.

PAGE 263

246 The study attempted to determine long-range implications for future decisions on which approach or program design will become established practice. The subproblem of the study asked the following question: Which program, and for what reasons, was favored by the Chapter I teachers, the Chapter I coordinators, and the regular classroom teachers? Research Design and Procedures Students In-class and pull-out designs were compared to determine their effectiveness for impr.oving student academic achievements without negatively affecting student self-concept. Students and their Chapter I teachers and regular classroom teachers from three.districts were used in the sample. Also, a survey of all of the coordinators from the state of Colorado was used. To test the first hypothesis of the study, that there was no significant difference between the academic achievement scores of Chapter I students in the pull-out program and achievement scores of Chapter I students in the in-class program, NCE scores were analyzed. To test the second hypothesis of the study, that there was no significant difference in the self-concepts of Chapter I students, as measured by the Self-Description Questionnaire (SDQ) in the pull-out program and the self concepts of Chapter I students in the in-class program. Students in the sample

PAGE 264

247 identified above were tested with the SDQ; the resulting scores were analyzed using multiple analysis of variance. Teachers All of the Chapter I teachers and regular classroom teachers of the students included in the student sample participated in the study and returned the Teacher Questionnaires. Also all of the coordinators of the state of Colorado (n=l04) received the Coordinators Survey mailed by the researcher. Fifty-one coordinators returned the completed survey. The coordinators' survey was designed with questions taken directly from the original teachers' survey so that responses from the coordinators' survey could be compared with responses from the teachers' survey. Major Findin&s The major findings of the study answered the problem of the study, two research questions and the subproblem of the study. 1. Results of an ANOVA for a single district and a MANOVA for combined districts supported Research Hypothesis I. There was no significant difference in the academic achievement of Chapter I students in the pull-out program and the academic achievement of the Chapter I students in the in-class program.

PAGE 265

248 2. Results of the District A ANOVA showed no significant differences in academic achievement by program design, pull-out or in-class. The results supported Hypothesis I. 3. Means for NCE pretests and posttests were compared for the students of the in-class and pull-out designs. Results for 119 Chapter I students showed that Chapter I in-class students scored lower as a group on theNCE pretest, 25.8 mean score as compared to 29.8 for the pull-out students. However, there was no significant difference on the NCE posttest. There was no difference in student academic performance by program design. 4. Results of a combined districts' MANOVA showed that the combined sample was academically similar without regard to program design, pull-out or in-class. No within-subject effects were found by the classificatory variables of sex, grade, and ethnicity. 5. A non-repeated measures MANOVA showed the effects of the NCE pretest as compared to the NCE posttest to measure the 218 Chapter I students from .the combined districts sample. There was no difference found in the combined sample on the classificatory variables of sex, grade and ethnicity on the dependent variables of NCE pretest and posttest. Therefore, students made similar gains on the NCE posttests with no variance by ctassificatory variables.

PAGE 266

6. Results of a District A ANOVA showed no significant effects by program design, in-class pull-out, for improving student self-concept. 249 7. Results of a subtest of the District A MANOVA showed that in-class students' self-concepts were negatively affected by participation in the in-class program in the area of peer relations. 8. Results of three subtests of student self-concept for the District A ANOVA showed that pull-out students scored higher in three areas--peer relations, general self, and total academic self-concept. 9. Results of the combined districts' MANOVA supported Research Hypothesis II. There was no significant difference found in the self-concept of Chapter I students in the pull-out program and the self-concepts of Chapter I students in the inclass program for the total test. 10. Results of the combined districts' MANOVA showed significance in one subtest only. In-class students, particularly non-Hispanics, scored lower in the area of selfconcept of physical appearance. 11. Results of this study indicated that pull-out programs were not more detrimental to student self-concept than in-class programs. Similarly, in-class programs were not more detrimental.

PAGE 267

12. Both program designs, in-class and pull-out, were favored by teachers and coordinators of each design due to beneficial effects that were common to both programs. 250 The benefits of both program designs were individualization to meet student need, a small pupil-to-teacher ratio, and the improved academic success of the student. The results of this study indicated that both pull-out students and in-class benefit similarly from both Chapter I program designs. 13. Teachers were more likely to decide upon the pull-out design, while teachers and coordinators agreed that administrators most frequently decided upon the in-class program design. 14. Teachers and coordinators overwhelmingly agreed that the reason the program design was selected, whether in-class or pull-out, was that the design decision was based upon professional judgment about the preferred service model to deliver instruction and to meet curricular concerns. 15. Teachers and coordinators agreed on three reasons for the selection of the pull-out design: a. Finding time for Chapter I instruction. b. Finding available space for the program. c. Meeting the varying remedial needs of the students. 16. The teachers and coordinators agreed that individualization to meet student need, small pupil-to-teacher

PAGE 268

ratio, and the improved academic success of the student were benefits of both program designs. 251 17. Pull-out programs were more beneficial to improving student self-concept than in-class, according to coordinators. 18. Teachers and coordinators of the in-class program preferred the in-class design due to benefits of the program. o Students do not .miss regular classroom instruction to receive Chapter I services. o Highly qualified teachers with expertise and knowledge related to remedial instruction provided the best possible instruction to students. 19. A benefit to students of the in-class program as more frequently reported by teachers was that students did not miss regular classroom instruction to receive Chapter I services. 20. The most frequently mentioned detrimental effect of the.pull-out design identified by the teachers was time missed from the regular program. Other detriments were: a. Lack of teaming of the Chapter I teachers and regular classroom teachers. b. Problems due to scheduling so that time for instruction was limited. c. Lack of remediation. 21. There were no detrimental effects of the in-class design reported in greater frequencies than for the pull-out

PAGE 269

design. However, noise level in the regular classroom was reported as a detriment by in-class teachers only. 252 22. Fewer detrimental effects were found with the in-class than with the pull-out program. 23. Coordinators cited the labeling of Chapter I students as "different from the norm" as a detrimental effect of both program designs. There was some concern with both program designs that labeling student caused stigmatization and results in self-esteem/self-image problems (self-concept). Conclusions The major conclusions of the study are stated below. The documentation for these conclusions was presented in earlier chapters of the dissertation. 1. There were no significant differences by program design in the academic achievement scores and in the self-concept scores of the Chapter I students from pull-out and in-class programs. 2. Both pull-out and in-class program designs were beneficial to Chapter I pupil progress when they were taught by competent and committed teachers. 3. The acceptance and support by the implementing teachers of the program design being used was critical to the success of the Chapter I program.

PAGE 270

253 4. The decision with respect to what program design was used in the Chapter I program usually was made by administrators. The teachers in the program usually did not make the decision. 5. Chapter I teachers and coordinators shared opinions concerning the value of Chapter I program designs--pull-out versus in-class--which were not always shared by their administrators. 6. Curriculum congruence was a problem in the implementation of the Chapter I program, for students encountered divergence between the Chapter I program and the regular classroom curriculum. 7. Both pull-out and in-class designs had detrimental effects which had to be overcome by regular and Chapter I teachers in the implementation of a Chapter I program. Recommendations Based on the results of this study, these recommendations are justified. 1. The benefits and detrimental effects of each program design should be considered when the selection of a program design is an issue for a district, a building or another Chapter I setting. The detrimental effects of a program design can be .minimized, whether in-class or pull-out, if teachers and administrators are involved in the selection and planning of

PAGE 271

the program. It is recommended that teachers, coordinators, and researchers continue profitable. communication and cooperation to maximize the expected beneficial outcomes of both Chapter I program designs. 254 2. Program design decisions regarding selection of design, in-class or pull-out, can be made with regard to solving problems that are impacting the efficiency and effectiveness of Chapter I programs since both program designs benefitted students similarly. 3. Self-concept studies need to be conducted to determine if students in either program design, pull-out or in-class, experience an improved self-concept as a result of the Chapter I program. This study was not designed to determine whether or not the self-concept of students actually improved as a result of Chapter I services. The study demonstrated only that one program design did not have a greater effect or lesser effect on student self-concept than the other design. No self-concept studies over a period of a year or more were found in the literature regarding the self-concept of Chapter I students in in-class programs. Therefore, it remains to be determined if the self-concept of students can be improved by being enrolled in either the Chapter I pull-out or in-class program. 4. Opinions of teachers who have experience with both program designs should be compared to determine which program design is favored. This study compared opinions of teachers

PAGE 272

255 working in a single program design, either in-class or pullout. No conclusions were made regarding which program design was favored by teachers since the teachers did not have personal experience with both program designs as a basis for comparison. 5. Further research needs to be done to determine the effect of teacher decisions about program design on the overall efficiency and effectiveness of the program if the teacher selects the design. This study was not designed to determine how positive attitudes of the teacher regarding the design of the program can positively affect student outcomes. 6. Since teacher opinion is a primary factor causing the program design to be beneficial to students, teachers favor program designs that they decide upon or with which they are involved in the decision or choice. Administrators who involve teachers in program design decisions can cause the program design to be more beneficial in the opinions of teachers and coordinators. 7. Overall, this study presented a comparison of student academic achievement and self-concept in two Chapter I program designs, in-class and pull-out. Further study needs to be done to conclude that both program designs are equally effective in improving the progress and self-concept of Chapter I students.

PAGE 273

REFERENCES Advanced Technology. (1983). Local operation of Title I. ESEA. 1976-1982: A resource book. McLean, VA: Advanced Technology. Allington, R. L. (in press). Policy constraints and effective compensatory reading instruction: A review. In J. Hoffman (Ed.), The effect of teaching of reading from research to practice. Newark, DE: International Reading Association. Allington, R., & Johnston, P. (1986, June). Coordination. collaboration, and consistency: The redesign of compensatory and special education interventions. Paper presented at the Conference on Effects of Alternative Designs in Compensatory Education, Washington, D.C. Allington, R., &'Others. (1985, February 7-9). What is remedial reading? A descriptive study. Paper presented at the annual meeting of the Colorado Council of the International Reading Association, Denver, CO. Archambault, F. X. (1987). Pullout versus in-class instruction in compensatory education. Paper presented at the annual meeting of the American Educational Research Association, Washington, D.C. Archambault, F. X., & St. Pierre, R. G. (1980). The effect of Federal policy on services delivered through ESEA Title I. Education Evaluation and Policy Analysis, l, 33-46. Ascher, C. (1986). Creating racial integration in a desegregated magnet school. ERIC/CUE Digest. (29). Bean, R. T., & Eichelberger, R. T. (1985). Changing the role of reading specialists: From pullout to in-class programs. Reading Teacher, 38(7), 648-653. Breglio, V. J., Hinckley, R. H., & Beal, R. S. (1978). Students' economic and educational status and selection for compensatdry education (Tech. Rep. No. 2). Santa Monica, 'cA: System Development Corporation.

PAGE 274

257 Brooks, M. G. (1981). and mathematics program funded by Title I and CEP. 1980-81 (Report No. 16-1). Atlanta, GA: Atlanta Public Schools, Division of Research, Evaluation, and Data Processing. Buras, 0. K. (Ed.). (1978). Mental measurements yearbook. Lincoln, Nebraska: University of Nebraska Press. Byrne, B. M. (1984). The general/academic self-concept nomological network: Review of construct validation research. Review of Educational Research, 54, 427-456. Carter, L. F. (1984). The sustaining effects study of compensatory and elementary education. Educational Researcher, 13, 4-13. Caslyn, R. J. (1974). The causal relation between selfconcept, locus of control, and achievement: A crosslagged panel analysis. (Doctoral dissertation, Northwestern University, 1973). Dissertation Abstracts International, 42, 4076a. Cooley, W. W. (1981). Effectiveness in compensatory education. Educational leadership, 38, 98-301. Cooley, W. W., & Leinhardt, G. (1980). The instructional dimensions study. Educational Evaluation and Policy Analysis, 1(1), 7-25. Coulson, J. E., Ozenne, D., Hanes, S., Bradford, C., Doherty, D., Duck, G., & Hemenway, J. (1977). The third year of the Emergency School Aid Act (ESAA) implementation. Santa Monica, CA: System Development Corporation. (ERIC Document Reproduction Service No. ED 154 952) Davis, M. (1982, April). A survey of teacher attitudes towards in-class versus pull-out compensatory reading programs. Unpublished master's thesis, Kean College, Union, NJ. Douthitt, F. (1985). One to grow on. chapter one in Ohio: Education consolidation and improvement act. Columbus, OH: Ohio State Department of Education. Ehly, S. W., & Larsen, S. C. (1980). Peer tutoring for individualized instruction. Boston, MA: Allyn and Bacon. English, F. (1984, May). whole-class teaching? Pull-outs: How much do they erode Principal, 63(5), 32-36.

PAGE 275

258 Felix, N., Hertlein, F., McKenna, & Rayborn, R. (1987, June). Combining categorial program services can make a major difference. Phi Delta Kappan, 68(10), 787-788. Fretchling, J. A., & Hammond, P. A. (1978). Policy implica tions of the instructional dimensions study. (ERIC Document Reproduction Service No. ED 157 205) Fretchling, J. and Others. (1984, July). Montgomery public schools. Rockville, MD: Department of Educational Accountability. French, M. P. (1985, Spring). TI: Content area reading. Wisconsin State Reading Association Journal, 29(3). Gaffney, M. J., & Schamber, D. M. (1982). The effects of the Title I supplement-not-supplant and excess costs provisions on program design decisions. McLean, VA: Advanced Technology, Inc. (ERIC Document Reproduction Service No. ED 243 238) Gay, L. R. (1981). Educational research: Competencies for Analysis & Application (2nd ed.). Columbus, OH: Charles E. Merrill Publishing Company. Glass, G. V., & Hopkins, K. D. (1984). Statistical methods in education and psychology (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall, Inc. Glass, G V., & Smith, M. L. (1977). 'Pullout' in compensa tory education. Paper prepared for the Office of the Commissioner, U.S. Office of Education. Unpublished manuscript, University of Colorado. Glass, G. V., & Smith, M. L. (1987). Pull out in compensatory education. Washington, DC: Department of Health, Education, and Welfare. Good, T. L. (1982, May 25). What is learned in schools: Responding to school demands in grades K-6. Paper presented at meeting of the National Commission on Excellence in Education, Washington, DC. Griswold, P A., Cotton, K. J., & Hansen, J B. (1986). Effective compensatory education sourcebook. Washington, DC: U.S. Government Printing Office. Halasa, 0. (1983, August). Reading strategy project. ECIA Chapter 1. evaluation report. 1982-83. Cleveland, OH: Cleveland Public Schools Department of Research and Analysis.

PAGE 276

259 Harpring, S. A. (1985, April). Inclass alternatives to traditional Chapter 1 pullout programs. Paper presented at the annual meeting of the American Educational Research Association, Chicago. Hayes, M. V. (1983). A survey of teacher attitudes towards in-class versus pullout compensatory reading programs. (ERIC Document Reproduction Service No. ED 228 619) Henderson, A. T. (1986, April). Chapter 2: For better or worse? Phi Delta Kappan, 67(8), 597-602. Henerson, M. E., Morris, L. L., & Fitz-Gibbon, C. T. (1978), How to measure attitudes. Beverly Hills, CA: Sage Publications. Johnston, P., Allington, R., & Afflerbach, P. (1985, March). The congruence of classroom and remedial instruction. Elementary School Journal, 85. Kennedy, G. C. (1978). School setting and learning. In Perspectives on the instructional dimensions study: A supplemental report from the National Institute of Education (pp. 33-39). (ERIC Document Reproduction Service No. ED 174 728). Kennedy, M. M. (1985, March 31-April 4). The national assessment of the Chapter 1 program. Paper presented at the annual meeting of the American Educational Research Association, Chicago. Kimball, G. H., Crawford, J., & Raia, F. (1985). Evaluation of the 1984-85 Oklahoma City Public Schools Chapter I Program. Oklahoma City: Oklahoma City Public Schools. Kimbrough, J., & Hill, P. T. (1981). The aggregate effects of federal education programs; Santa Monica, CA: The Rand Corporation. Leinhardt, G., & Pallay, A. (1982). Restrictive educational settings: Exile or haven? Review of Educational Research, 52(4), 557-578. Levine, D. U., & Stark, J. (1981). Instructional and organizational arrangements and processes for improving academic achievement at inner city elementary schools. A study of the Chicago Mastery Learning Reading Program and other school-wide approaches for improving reading at selected schools in Chicago. Los Angeles. and New York. Paper presented at the annual meeting of the American Educational Research Association, New York (ERIC Document Reproduction Service No. ED 213 814)

PAGE 277

Lignon, G. D., & Doss, D. A. (1982, April). Some lessons we have learned from 6.500 hours of classroom observations. Paper presented at the annual meeting of the American Educational Research Association, Montreal. Madhere, S. (1981). Issues in program evaluation. (ERIC Document Reproduction Service No. Ed 217 098) Marsh, H. W., Barnes, J.,Cairns, L., & Tidman, M. (1984). The Self Description Questionnaire (SDQ): Age and 260 sex effects in the structure and level of self-concept for preadolescent children. Journal of Educational Psychology, 76, 940-956. Marsh, H. W., Smith, I. D., & Barnes, J. (1983). Multitraitmultimethod analyses of the Self Description Questionnaire: Student-teacher agreement on multidimensional ratings of student self-concept. American Educational Research Journal, 20, 333-357. Morehead, A., & Loy, R. Webster dictionary. Chicago: New American Library, 1972. National Institute of Education. (1977). The effects of services on student development. Washington, DC: Author. Neumann, T. (1985). In-class remedial instruction. An alternative to the pullout programs. Wisconsin State Reading Association Journal, 29(3). Noddings, N. (1978). Overview of the Instructional Dimensions Study Conference. In National Institute of Education (Ed .. ), Perspectives on the Instructional Dimensions Study (pp. 5-14). Washington, DC: National Institute of Education. (ERIC Document Reproduction Service No. ED 174 728). Oregon State Department of Education. (1984). Chapter 1 in Oregon. Facts and figures. 1982-83. Salem: Oregon State Department of Education. Passow, A. H. (1986). Educating the disadvantaged--the task school districts face. (ERIC Document Reproduction Service No. ED 267 138) Plunkett, V. R. L. (1985, April). From Title I to Chapter 1: The evolution of compensatory education. Phi Delta Kappan, 66(8), 533, 537. Plunkett, V. R. L. (1987, October). Conversation.

PAGE 278

261 Pottebaum, S.M., Keith, T. Z., & Ehly, S. W. (1985, September October). Is there a causal relation between self-concept and academic achievement? Journal of Educational Research, 79(3), 140-144. Poyner; L. H. and Others. (1981). A description of compensatory services in high-poverty schools. Sustaining Effects Study Technical Report No. 18. Santa Monica, CA: Systems Development Corp. Rodriguez, J. H. (1985). Toward effective education for Hispanic youth: A leadership challenge. Monograph. San Diego: County Office of Education. Savage, D. G. (1987, April). Why Chapter I hasn't made much difference. Phi Delta Kappan, pp. 581-584. Shavelson, R. J., & Bolus, R. (1982). Self-concept: The interplay of theory and methods. Journal of Educational Psychology, 74, 3-17. Shavelson, R. J., Hubner, J. J., & Stanton, G. C. (1976). Self-concept: Validation of construct interpretations. Review of Educational Research, 46, 407-441. Slavin, R. (1987, October). Making Chapter I make a difference. Phi Delta Kappan, pp. 110-119. Swan, E. and Others. (1985, August). The educational effects of a state supported reduced class size program. A comprehensive evaluation of Indiana's Project PRIME TIME at the North Gibson School Corporation. Terre Haute, IN: School of Education. Taylor, B. (1985, September). Let's pull out of the pullout program. Principal, 65(1), 52-54. Tobias, S. (1982). When do instructional methods make a difference? Educational Researcher, 11(4), 4-9. Van Ecko, J. J., & Ames, with Archambault, H. F. (1980). Who benefits from federal education dollars? Cambridge, MA: Abt Books. Vasquez-Nuttall Associates. (1982, 1983). Chapter 1 elementary reading programs. Interim evaluation report. (ERIC Document Reproduction Service No. ED 238 968)

PAGE 279

Wylie, R. C. (1979). The self-concept: Theory and research on selected topics (Vol. 2). Lincoln: University of Nebraska Press. 262 Yap, K. 0. (1983). Effects of instructional setting and approach in compensatory education: A statewide analysis. (ERIC Document Reproduction Service No. ED 230 745A) Yap, K. 0. (1985). A cost analytic approach to determining Chapter 1 program impact: Some preliminary findings. (ERIC Document Reproduction Service No. ED 254 564)

PAGE 280

APPENDICES

PAGE 281

APPENDIX A THE SELF-DESCRIPTION QUESTIONNAIRE

PAGE 282

265 SELF DESCRIPTION QUESTIONNAIRE Narne ........................................................................................................... Boy .......... Girl .......... .. Age ................................. Schooi. ................................................. Teacher .................................... : ................. This is a chance to look at yourself. It is not a tast. There are no right answers and everyone will have different answers. Be sure that your answers show how you feel about yourself. PLEASE DO NOT TALK ABOUT YOUR ANSWERS WITH ANYONE ELSE. We will keep your answers private and not show them to anyone. When you are ready to begin, please read each sentence and decide your answer. (You may read quietly to yourself as I read aloud.) There are five possible answers for each question "True", "False", and three answers in between. There are five boxes next to each sentence, one for each of the answers. The answers are written at the top of the boxes. Choose your answer to a sentence and put a tick ( 1 I in the box under the answer you choose. DO NOT say your answer out loud or talk about it with anyone else. Before you start there are three examples below. Somebody named Bob has already answered two of these sentences to show you how to do it. In the third one you must choose your own answer and put in your own tick ( I ). SOME TIMES MOSTLY FALSE, MOSTLY FALSE FALSE SOME TRUE TRUE TIMES TRUE EXAMPLES 1. I like to read comic books .................................... 2. (Bob put a tick in the box under the answer "TRUE". This means that he really likes to read comic books. If Bob did not like to read comic books very much, he would have answered "FALSE" or "MOSTLY FALSE".) In general, I am neat and tidy ............................... 2 [:::::::J [:YJ 2 (Bob answered "SOMETIMES FALSE, SOMETIMES TRUE" because he is not very neat, but he is not very messy either.) 3. lliketowatchT.V .............................................. 3 [:::::::J c::J 3 (For this sentence you have tQ choose the answer that is best for you. First you must decide if the sentence is "TRUE" or "FALSE" or somewhere in between. If you really like to watch T.V. a lot you would answer "'TRUE" by putting a tick in the last box. If you hate watching T.V. you would answer "FALSE" by putting a tick in the first box. If your answer is somewhere in between then you would choose one of the other three boxes.) If you want to change an answer you have marked you should cross out the tick and put a new tick in another box on the same line. For all the sentences be sure that your tick is on the same line as the sentence you are answering. You should have one answer and only one answer for each sentence. Do not leave out any of the sentences. If you have any questions put up your hand. Tum over the page and begin. Once you have started, PLEASE DO NOT TALK. H. W. Manh and I. 0. Smith, The Univenitv of Sydney 1981

PAGE 283

!OMI TIMES 266 lo'OSTLV FALSI, MOSTLY FALSI FALU SOME TAUI TAUI TIMU TnUE 1. I am goad looking................................................. CJ CJ c:::J c:::J C] 1 2. I'm good at all SCHOOL SUBJECTS.................... 2 CJ CJ 0 0 C] 2 3. lcanrunlast ........ ............................................... 3c::J CJc:JCJCJ3 4. lgetgaadmarksin READING ............................. 4 CJ CJ c:::J 0 CJ 4 5. My parent! undentand "'' ................................... 5 C] 0 0 0 0 5 6. lhateMATHEMATICS ........................................ 6 C] CJ 0 0 06 7. lha'tllol!ollriends ............................................. 7CJ CJ c::J D D 7 B. llikethtwayllook ............................................. aC]OODDa 9. I enjoy doing work in all SCHOOL SUBJECTS.... 9 CJ CJ c:::J CJ c::::::::J 9 10. I like to run and play hard .................................... 10 CJ CJ CJ CJ c::::::::Jto 11. I like READING ................................................... 11C] CJ c:::J c::::::::J c::::::::JII 12. My parents are usually unhapny or dis"nnointl!li withwhatldo ...................................................... 120 CJ c:::J 0 012 13. .............. 130 0 W c::::::::J 013 14. I make lrlendseasily ............................................. CJ :::1 :::1 W14 15. lhaveapleasantlaakingt!'ce ............................... 15CJ CJ 0 0 1::]15 16. I get good marks in all SCHOOL SUBJECTS ........ 16CJ CJ c::::::::J 0 018 17. lhalespomandgames ........................................ 170 CJ :::1 c::::::::J 1::]11 18. l'mgoodatREAOING ......................................... 180 CJ 0 c::::::::J c::::::::J1a 19. I like my parents .................................................. 190 CJ 0 0 019 20. llooklorwardtoMATHEMATICS ...................... 200 CJ 0 0 c::::::::J20 21. Mastkidshavemorefrio.ndsthanldo ................. 21C] CJ 0 0 021 22. I am a nice looking .................................... 220 CJ 0 c::::::::J 022 23. lhateallSCHOOLSUBJECTS ............................. 230 0 W 0 023 24. lenjoysportsandgarnes ...................................... 0 CJ 0 024 25. I am interetted in READING ................................ 25 c::J D 0 c:::J 02s 26. My narents like mo .................................. ............ 'G c::J CJ W CJ CJ2s

PAGE 284

!OM! IIMl' P.'tflr.tlV FAUI, MO!ii\,Y 267 FALS! FALS! SOMF. TAUI TAUI tiMES '"'" 27. .................... 21C] CJ c:::J C] W21 28. .......................... 2110 0 c:::J C] [:::J28 29. ldolotsolirnJtortairtthin!l......................... 1!1W 0 c:::J 0 (::J29 30. larnugly............................................. 300 0 [:::J C] [:::J3o 31. llearnthingtQuicklyinoiiSCIIOOLSIJUJECIS. 310 C] c:::J 0 CJ31 32. I have good musolrs .............................................. 37 C] C] [:::J C] c:J 32 33. lamdumbatAEADINO ..................................... 3JO CJ c::::J c::::J C]33 34. If I h:tVII! r.;hildrron "' ll1y' f'I'.-.JII I WiHll In lttillfllhron up like my par .. nll rnisd me .............................. 1::] 1::] c:J 0 c:::Jl4 35. lamlnterestedinMATIIEMAliCS ...................... J51:::J Q CJ C] (:=J35 38. I am eny to like.................................. J(i[::J [::J [::J 0 [::J36 37. Ovaralllanrnognocl .......................... 31c:::J [::J 0 0 CJ37 39. Other kirh think I ""' qonrllnukinq ...... .......... .. Jn[::J r:::=J 0 r:::=J 039 39. lanrlnl!relle!linaiiSCIIOOLSIJUJEr.rs .......... J91:::J CJ CJ CJ (::J39 40. larngoodatsporrt................................... C] [:::J c:J c:J40 41. 410 0 c:J [:::J C]4t 42. Myparentsandlrnrlalntnltirntrurr"rr c:J CJ 0 CJ C]42 43. I Jearn things in MA Ill EMil III:S [::J [::J 0 0 [::J43 44. Ollrerkidswantrnetnhthirhirnrl......... 4t0 c:J CJ c:J 044 45. In geno.rall like hrinq tho w.w I""'........... c:J 0 0 CJ45 46. I hava a good lnnkiroq hndy ................. ... ......... 4r.[::J [::J CJ CJ C)46 47. lamdumbinaiiSCIIOOLSUIIJECI!l ................. 41( Jl ( c::J 0 047 48. I ti1n run alonq w:e.,wilhtat 0 [:::J [:::J [:::J 48 49. Work in REIIDINO isro
PAGE 285

110111 mEI IIOSTLY 'ALII, III:IST\Y 268 PALII PALII TRUI TIIUI nMU TRUI 53. Overall I have I lot to be proud of --... 53 c::::::J CJ c:J CJ CJ53 '54. l'mbelllrlookin!lthanmonofmyfriendl ....... 54c::::::J CJ c:J CJ CJ54 55. I look forward to all SCHOOL SUBJECTS ..... s5c::::::J D D D D 56. lamagoodthlele ................................ ............. ssc::J CJ D D c:J5e 57. I look forward to READING ................................ 57r:::J c::J c:J c:J c:J57 sa. I get along well with my parentt .. ........ ............. sa r:::J c::J D D Dsa 59. l'mqoodatMATHEMATICS ............................... sgr:::J D D D Ds9 60. lampopularwithkidsolmyowna!J! ................. soc::::J D D D Dso 61. 1 can't do anything right ...................................... 61 c::::J D c:J CJ CJ61 62. I have nice feature like nose, and eyes. and nair .. 62 c::::J c:J c::J CJ CJ62 63. Work in all SCHOOL SUBJECTS is easy for me ... 63r::=:J D c:J 0 063 64. l'mgoodattnrowingaball .................................. 64r::=:J D c:J 0 064 65. 1 nate READING .................................................. 6sc::::J D c:J CJ 065 66. My oarenn and 1 naw a lot ollun together .......... 66CJ CJ c:J CJ CJ66 67. I can do things as well as molt other people ......... 67 CJ c::::J CJ c:J CJs7 68. I en,oydoingwork in MATHEMATICS ............... sac::::J CJ D D CJ6B 69. 'Aost other kids like me ........................................ 69 c::::J c:J c:J CJ 0 69 70. people think I am a good person ................. 70 [:::J CJ c:J CJ 0 70 71. I like all SCHOOL SUBJECTS .............................. 71 r::=:J c:J c::::J c:J C]11 72. A lot of about me are good ........................ 72 [:::J .CJ c::::J c:J 012 73. I learn things quickly in READING ...................... 73 c::::J CJ c::::J 0 073 74. I'm as good as moll other people .......................... 74 r::=:J c::J c::::J CJ CJ74 75. lamdumbatMATHEMATICS ............................ 75r::=:J W c::J c:J W75 76. When I do something, I do it well ......................... 76 r::=:J W c::J CJ c:J 76

PAGE 286

APPENDIX B INSTRUCTIONS FOR HAND SCORING THE SDQ SCORING SHEET

PAGE 287

270 APPENDIX II -INSTRUCTIONS FOR USING THE HAND-SCORING SHEET The number in the lower left-hand corner of each bo:: corresponds to the SDO question with the same number. The response to each SDQ question is to be translated 1nto a number and put into the appropriate box. For the positive items --those boxes that do not have an "XXX" --the responses should be translated so that l=False; 2=Mostly False, 3=Sometimes False, Sometimes True, 4=Mostly and 5=True. For the negative items --those boxes that do have an "XAX" --the responses should be translated so that l=True, 2=Mostly True, .3=Sometimes False, Sometimes True, False, and S=False.
PAGE 288

271 SDQ 68 APPENDIX II -Scorin9 Shee.t For Hanc1 Scoring Responses to the SDQ lABILITY APPEAR RELATION RELATION MATICS SCHOOL SELF : --------:--------:--------:--------:--------1--------:--------:--------: . . . . 1 I I I I I I I : : : : : XXX : : : 3 1 : 7 : s : .4 : 6 : 2 : : --------:--------:--------:--------:--------:--------:--------:--------1 : : : I I I ; : XXX : : I I : 10 B 14 12 : 11 I 13 : 9 : I --------: -----:-------:-------:-----:-------: ---------: : : : : : : : : XXX : : XXX I : : : I : 17 : 1:5 : 21 : 19 I 18 : 20 : 16 I : --------:--------;--------:--------:--------:--------:--------:--------: : : : I : : : : I I I XXX I iv i I I I I I I I t t I I I I I I I : XXX & : : XXX : : XXX I 32 : 30 : 36 : 34 : 33 : 35 : 31 : 37 : --------:--------:--------:--------:--------:--------:--------:--------: I I I I I I I I I I I I I I I I : : I : : : : I 40 : 39 : 44 : 42 : 41 : 43 : 39 : 45 : I I I I I I I I I I I I I I I I I r : : I XXX : : 48 : 46 : 52 : 50 : 49 : 51 : 47 : SJ : --------:--------:--------:--------:--------:--------:--------:--------: I I I I I I I I I I I I I I I I : I : : XXX : 56 : 54 : 60 : 58 : 57 : 59 : 55 : 61 : --------:--------:--------:--------:--------:--------:--------:--------: I I I I I I I : XXX 64 : 62 : 69 : 66 : 65 : 69 : 63 : 67 : --------:--------:--------:--------:--------:--------:--------:--------: I I I I I I I I I I I I : XXX : : f I : I : 73 : 75 : 71 : 70 : --------:--------:--------:--------:--------:--------:--------:--------: ' ' : 72 : :--------: NAME---------------------------------------------------' ' ' . : 74 : AGE _______ SEX _______ Grade/ Year -----------:-------; ' ' ' ' TEACHER---------------------SCHOOL __________________ : 76 : ;-------' SCALE SCORES. Sum responses to B positive items in each column --those boxes not marked with an xxx --and put the total in box (below) at the bottom of the column. :--PHVS--:--APPR--:--PEER--:--PRNT--:--READ--:--MATH--:--SCHL--!--GENL--1 I I I I I I I I I I I I I I I I I I I I I t I I I I I I I I I I I I I I I I I I I I I 1 I I I I I I I above and divide by the number of scales as indicated below. TOTAL NONACADEMIC = < PHYS + TOTAL ACADEMIC = TOTAL + READ + + AF'F'R + PEER + PRNT + + I'IATH + SCHL. l I 3 + l I :; = SELF =

PAGE 289

APPENDIX C LETTER FROM DR. HERBERT W. MARSH IN SUPPORT OF RESEARCH

PAGE 290

Dear Colleaguea Dr. llerbert W. Faculty of University of Sydney Sydney 2006 Australia IS January, 1986 273 1 Mal pleased to hear of your Interest ln my self-concept research and the reprint which you requested is enclosed. I have also dane ather research In this area which 1 would ike to share Hith you but cost and time limitations do nat allow far thls. Instead( 1 have made a copy at the abstracts from each of the atudles. Please do nat quote d rect.ly from those Hhich are "accepted far publlcatton or n review. Nearly all of my research has been conducted In Australia, and I eager to collaborate Nlth reaearchers In other settings. Consequenuy1 lf you would like to Ull! the SDQ1 please write to me about rour research and l wl cooperate In whatever May 1 Also, I would like to rece ve copies of your research ln this area. Funding far research In Educational Psychology Is almost nonexistent here, and I have no budget far responding to reprint. requests. I estimate that my cast Is t2/paper to cover the cast of prlnt1ng, handling, postage, ate., and I have no funds for this. Yours truly, Herbert W. Marsh

PAGE 291

APPENDIX D PILOT STUDY OF THE SELF-DESCRIPTION 'QUESTIONNAIRE INTRODUCTORY LETTER INSTRUCTIONS TO THE TEACHER PUPIL DATA SHEET SAMPLE OF COMPLETED TEACHER RESPONSE

PAGE 292

275 Darlene M. Pestello Baber Chapter I Study for C. U. Dissertation Chapter I Third Grade Teacher Schools Pilot Test Project March 31, 1986 Dear ----------------------Work: 428-7503 Home: 499-3340 Thank you for agreeing to be a part of this Chapter I study. The purpose of the study is to compare "pull-out" programs to in-classroom Chapter I programs. Time is critical since I plan to collect the data this spring. This pilot of the instrument is a critical first step. The instrument is entitled Self Description Questionnaire and is recommended by Dr. Alan Davis from the Technical Assistance Center. The questionnaire has good validity and reliability for sixth grade and up. I need to know how well third graders can respond to the design. You may also be interested in your students' responses for your own purposes. I would be happy to give you any results that you would request. I sincerely appreciate all of your time and cooperation.

PAGE 293

FOR THE TEACHER: Please complete a data sheet for each child. Match the letter on the cover sheet to the letter on the data sheet. Do not use names. Confidentiality is an issue. Please answer these questions: Dates and times of test administration Time required for administration What is the total population of your school? Describe all problems that the students had with the test. 276

PAGE 294

277 PUPIL DATA SHEET About the child: Age Birthdate -----------------Grade----Sex ----Ethnic origin. ____________________________ List other programs this child may attend. How many hours does the child spend in the regular education classroom? ---------------------------------How many hours or minutes does the child spend in Chapter I? --------------------What time of day does Chapter I instruction occur? How many total months has this child. been in Chapter I? Were there any difficulties in providing the above information? ----------------------------------------

PAGE 295

FOR THE TEACHEB.: Please complete a data sheet for each child. Match the letter on the cover sheet to the letter on the data sheet. Do not use names. Confidentiality is an issue. Please answer these questions: Dates and times of test administration F,G,H.I fd t:-ltJ;:r,KJ,,/ol @ '/:.s-4 Time required for administration IS' z:.; .:/.!" l?V ;i,Q '.,1, ioe d I What is the total population of your school? Describe all problems that the students had with the test. 278

PAGE 296

APPENDIX E REVISED SELF-DESCRIPTION QUESTIONNAIRE INSTRUCTIONS TO STUDENT INSTRUCTIONS TO TEACHERS FOR ADMINISTRATION OF THE S.D.Q. PUPIL DATA SHEEET REVISED

PAGE 297

280 SELF DESCRIPTION QUESTIONNAIRE SLudcnt r:o
PAGE 298

281 PUPIL DATA SHEET -TO BE ATTACHED TO EACH SELF DESCRIPTION QUESTIONNAIRE About the child: Age __ Bi rthda te __________ Grade ---Sex __ Ethnic origin ______________ List other programs this child may attend. What time does the student school day begin? --------What time does the student school day end? ________ What days does the student receive Chapter I ihstruction? M T W T F At what hour does Chapter I instruction begin? A.M. 7 8 9 10 11 p.m. 12 1 2 3 How many hours and minutes does the child spend in Chapter instruction per week? ------hours minutes ------How many total months has this child been in Chapter I?

PAGE 299

282 To the Teacher: HELPFUL HINTS FOR STUDENT SELF QUESTIONNAIRE (SDQ) ADMINISTRATION Dear Teacher, The SDQ has good validity and reliability for sixth grade and up._ However, there was some uncertainty about the readability of the SDQ for third and fourth graders. A pilot study was completed at this time last year in another Front Range School District. Six teachers and 54 third grade students participated. The results have been summarized in the HELPFUL HINTS listed below. Again, thank you for your participation. Teachers reported that the SDQ required 20 to 25 minutes to administer. LOCATION Teachers of students with an "in classroom" program chose to take students to a quiet location to reduce distractability. CONTENT VALIDITY Some children complained about having already answered a question. ouestions are restated to insure consistency of responses. It may help to explain that some questions may seem alike, but to try to answer each one with a thoughtful response. STUDENT CHOICES One very helpful teacher shared her description of questionnaire administration. "I drew and labelled the 5 boxes on the chalkboard and emphasized which end was true and which false by pointing it out after every 6 or 8 items. A child stopped me on #6 thinking he had gotten confused on this. I said "If you hate Math, Mark T, if not, Mark F" and pointed to the boxes. He decided he was doing it right. One child required a marker to stay on the right line. Another got off the row also at one ooint." VOCABULARY There were a few problems with vocabulary. Certain words may need to be explained. One third grade teacher had to explain "raised me" in question 34. Also, "featured" had to be explained in question 62. One child asked for the meaning of "athlete" in 56. Teachers the students that "mathematics" meant math and that "school subjects" meant classes.

PAGE 300

APPENDIX F PILOT STUDY OF THE TEACHER QUESTIONNAIRE INTRODUCTORY LETTER SURVEY

PAGE 301

Dear 284 3195 Heidelberg Drive Boulder, CO 80303 Work phone: 428-7503 Home phone: 499-3340 I greatly appreciate your participation in the Chapter I study. This part of the study is STRICTLY CONFIDENTIAL. It will require about twenty minutes of your time. As a teacher, I know the sacrifice in time that this questionnaire requires. Your contribution will effect Chapter I programs nat i ona 11 y. This study has the support of the Chapter I Office of the Colorado State Department and of the Chapter I Technical Assistance Center. PURPOSE OF THE STUDY The Chapter I Office of the State Department of Colorado has requested that a study be done to compare two Chapter I program designs for reading instruction and remediation, The first design is labeled "pull out11 and is structured so that Chapter I students are grouped and removed from the regu lar education setting for a specified period of time each day. The second design is labeled 11in classroom11 whereby the Chapter I teacher teams with the regular education teacher to plan and implement instruction. The Chapter I teachers and regular education teachers are being asked to contribute their expert opinions on program design within a specific site and program effects on Chapter I students. DIRECTIONS -Please fill out the questionnaire as candidly as possible. -Return it in the self addressed enclosed envelope to insure confiden-ti a 1 i ty. -Please feel free to add your own feelings or observations. Sincerely, Darlene M. Baber

PAGE 302

TEACHER QUESTIONNAIRE 1. How many students are in your building? 2. What percent are Chapter I students? 3. Who decided upon the program design in your building, "pull out" versus in "in classroom" instruction? 285 Central Administrator Principal Chapter I --Title: ---Coordinator 4. For what reasons was the design selected? __ time of instruction number of staff __ space available __ research findings __ varying remedial needs professional judgment --about instruction __ number of pupi 1 s __ preference of teachers professional judgment --about curriculum to cut excess costs 5. What are the most beneficial effects of the Chapter I program? 6. What are the most detrimental effects of the Chapter I program? 7. What problems are impacting your success with the program? 8. Do your students miss regular classroom instruction to receive remedial reading? My students miss social studies My students miss science My students miss other classes Yes No The classes missed are _______________

PAGE 303

286 TEACHER QUESTIONNAIRE page 2 9. Do your students miss regular classroom instruction to receive remedial math? Yes No My students miss social studies My students miss science My students miss other classes The classes missed are 10. My students miss no academic subject since remedial instruction occurs during study in the regular class. Yes No 11. Does the Chapter I program supplant or supplement the regular education program? 12. Are you a remedial subject matter specialist? 13. What are your qualifications and/or certifications for your position? 14. Do you receive extra pay for your role in instructing remedial pupils? Yes No 15. Does the caucasian population of your Chapter I pupils reflect the population of the school? more than __ __; ____ same as fewer than ----16. How well do you expect your Chapter I students to perform academically as compared to regular education students at their grade level? better than ---_____ about the same less well ---Yes 17. Does your program utilize peer tutoring? ----___ No 18. The Chapter I program design in my building has no clear academic or social benefits and may. in fact. be detrimental to pupils progress and adjustment to school. __ Strongly Agree __ Agree __ Disagree __ Strongly Disagree 19. The Chapter I program design in my building is used more to satisfy Title I regulations than because it is judged by teachers to be a sensible and beneficial plan. __ Strongly Agree __ Agree __ Disagree __ Strongly Disagree 20. The Chapter I program design in my building is used primarily to satisfy Chapter I regulations. __ Strongly Agree __ Agree __ Disagree __ Strongly Disagree

PAGE 304

287 TEACHER QUESTIONNAIRE page 3 21. The Chapter I program is judged by administrators to be a sensible and beneficial plan. __ Strongly Agree _Agree __ Disagree _Strongly Disagree 22. Chapter I students are unconsciously neglected in the regular classroom. __ Strongly Agree __ Agree __ Disagree __ Strongly Disagree 23. Overall, do you have-any worries or concerns about the program or the Chapter I students in relation to program design in your building? 24. Would you please describe the Chapter I program in the space below or on the back. Thank you so much for your time. Name. ___________ Bu i1 ding _________ District. _________ Position ----------

PAGE 305

APPENDIX G TEACHER QUESTIONNAIRE INTRODUCTORY LETTER TEACHER SURVEY

PAGE 306

Dear 2.89 3195 Heidelberg Drive Boulder, CO 80303 Work phone: 428-7503 Home phone: 499-3340 I greatly appreciate your participation in the Chapter I study. This part of the study is STRICTLY CONFIDENTIAL. It will require about twenty minutes of your time. As a teacher, I know the sacrifice in time that this questionnaire requires. Your contribution will effect Chapter I programs nat i ona 11 y. This study has the support of the Chapter I Office of the Colorado State Department and of the Chapter I Technical Assistance Center. PURPOSE OF THE STUDY The Chapter I Office of the State Department of Colorado has requested that a study be done to compare two Chapter I program designs for reading instruction and remediation. -The first design is labeled "pull out" and is structured so that Chapter I students are grouped and removed from the regular education setting for a specified period of time each day. The second design is labeled "in classroom" whereby the Chapter I teacher teams with the regular education teacher to and implement instruction. The Chapter I teachers and regular education teachers are being asked to contribute their expert opinions on program design within a specific site and program effects on Chapter I students. DIRECTIONS -Please fill out the questionnaire as candidly as possible. -Return it in the self addressed enclosed envelope to insure confiden-tiality. -Please feel free to add your own feelings or observations. Sincerely, Darlene M. Baber .

PAGE 307

290 TEACHER QUESTIONNAIRE Please check Building design __ "Pull out" "In class" Other --Specify 1. How many students are enrolled in your building?----2. What percent are Chapter I students? ___ 3. Who decided upon the program design in your building, "pull out" versus "in classroom" instruction? Central Administrator __ Principal __ Chapter I Coordinator --Title: Other: Please explain. _________________ 4. For what reasons was the design selected? Check as many as apply. time of instruction __ space available __ research findings __ varying remedial needs professional judgment -about instruction number of staff __ number of pupils __ preference of teachers professional judgment --about curriculum to cut excess costs __ other: Please explain ________________ 5. What are the most beneficial effects of the Chapter I program? 6. What are the most detrimental effects of the Chapter I program? 7. What problems are impacting your success with the program?

PAGE 308

291 TEACHER QUESTIONNAIRE page 2 8. My students miss no academic subject since remedial instruction occurs during study periods in the regular class. Yes No 9. Do your students miss regular classroom instruction to receive Chapter I reading? y N es o My students miss social studies. My students miss science. My students miss other classes. The classes missed are. __________________ How much time per day? __________________ 10. Do your students miss regular classroom instruction to receive Chapter I math? Yes No My students miss social studies. My students miss science. My students miss other classes. The classes missed are ------------------------How much time per day? __________________ 11. Does the Chapter I program supplant (replace) or supplement the regular education program? 12. Are you a remedial subject matter specialist? 13. What are your qualifications and/or certifications for your position? 14. Do you receive extra pay for your role in instructing remedial pupils? Yes No 15. Does the caucasian population of your Chapter pupils reflect the population of the school in number? more than same as fewer than Please specify if you can school population i. ___ Chapter I population ,; ___ 16. How well do you expect your Chapter I students to perform academically as compared to regular education students at their grade level? better than about the same less well 17. Does your program utilize peer tutoring? ___ Yes No

PAGE 309

TEACHER QUESTIONNAIRE page 3 For questions 18 to 20 please check reading __ math __ or both __ 18. The Chapter I program design in my building has no clear academic or social benefits and may, in fact, be detrimental to pupils' progress and adjustment to school. 292 Reading Strongly Agree Agree ___ Ohagree ___ Strongly Disagree Math ----Both --19. The Chapter I program design in my building is judged by teachers to be a sensible and beneficial plan. Reading Strongly Agree Agree ___ Disagree ___ Strongly Disagree Math ----Both --20. The Chapter I program design in my building is used primarily to satisfy Chapter I regulations. __ Strongly Agree __ Agree __ Disagree __ Strongly Disagree 21. The Chapter I program is judged by administrators to be a sensible and beneficial plan. __ Strongly Agree __ Ajree __ Disagree __ Strongly Disagree 22. Chapter I students are unconsciously neglected in the regular classroom. __ Strongly Agree __ Agree __ Disagree __ Strongly Disagree 23. Overall, do you have any worries or concerns about the program or the Chapter 1 students in relation to program design in your building? 24. Would you please describe the Chapter I program in your building in the space below or on the For example, are there aspects of your program that may be unique to other Chapter I programs? Are there aspects that may affect the results of this study? Thank you so much for your time. Name. _____________________ __ District'-------------------Address ----------------Bu i1 ding, _________ Position _________ I would like to thank you personally. After your building responses are collated, I will destroy all names.

PAGE 310

APPENDIX H CHAPTER I COORDINATOR/DIRECTOR QUESTIONNAIRE INTRODUCTORY LETTER COORDINATOR SURVEY

PAGE 311

Dear 294 3195 Heidelberg Orive Boulder, CO 80303 Work phone: 428-7503 Home phone: 499-3340 I am a teacher in Adams County School District 50, Westminster. I am conducting a Chapter I study as part of the research for my doctoral dis sertation. I need your help to answer questions regarding program design. The enclosed survey is not an evaluation of your Chapter I program. However, the opinions of the Chapter I coordinators/directors in the State of Colorado are crucial to the research design. I greatly appreciate your participation in the Chapter I study. This part of the study is STRICTLY CONFIDENTIAL. It will require about twenty minutes of your time. I know the sacrifice in time that this questionnaire requires, especially at this time of year. Your response is critical. I would greatly appreciate receiving this questionnaire by June 15th. I know you will agree that the question of program design is an important issue. This study has the support .of the Chapter I Office of the Colorado State Department and of the Chapter I Technical Assistance Center. PURPOSE OF THE STUDY The Chapter I Office of the State Department of Colorado has requested that a study be done to compare two Chapter I program designs for reading in struction and remediation, and mathematics where applicable. The first design is labeled "pull out" and is structured so that Chapter I students are grouped and removed from the regular education setting for a specified period of time each day. The second design is labeled "in classroom" whereby the Chapter I teacher teams with the regular education teacherto plan and implement instruc tion. The Chapter I teachers and regular education teachers are being asked to contribute their expert opinions on program desiqn within a specific site and program effects on Chapter I students. DIRECTIONS To insure confidentiality Please fill out the questionnaire as candidly as possible. -Return it in the self addressed enclosed envelope to insure confidentiality -Please feel free to add your own feelinqs or observations. Sincerely, Darlene M. Baber

PAGE 312

295 CHAPTER I COORDINATOR/DIRECTOR QUESTIONNAIRE School District----------------------------------------------------1. What grades in your district are served by Chapter I? 2. How many Chapter I students are enrolled? __________________ __ 3. What is the total district enrollment? ______________________ 4. How many Chapter I teachers have a "pull out" program design? ______________________________________________________ __ 5. How many Chapter I teachers have an nin classroom" program design? ______________________________________________________ __ 6. If there are teachers utilizing other I designs, please specify how many. ________ Please describe these program alternatives. 7. Who decided upon the program design(s) in your district, pull out" versus "in classroom" instruction? Check those that apply Central Administrator Title ------------------Principal Chapter I -----Coordinator Other: Please explain __________________________________ __

PAGE 313

COORDINATOR/DIRECTOR QUESTIONNAIRE page 2 B. For what reasons was the design(s) selected? Check as many as apply. _____ time of instruction _____ space available _____ research findings _____ varying remedial needs professional judgment -----about instruction number of staff number of pupils _____ preference of teachers professional judgment -----about curriculum to cut excess costs _____ other: Please explain. __________________________________ __ 9. What, in your opinion, are the most beneficial effects of the Chapter I program? 10. What, if any, are the most detrimental effects of the Chapter I program? 11. What problems are impacting the success of the program? 12. Do the Chapter I students miss regular classroom instruction to receive Chapter I reading or math? Yes No

PAGE 314

297 COORDINATOR/DIRECTOR QUESTIONNAIRE page 3 13. Does the caucasian population of your Chapter I pupils reflect the population of the district in number? ___ more than same as fewer than Please specify percentages if you-have this data. For 14 & 15 please check reading___ math____ or both 14. The Chapter I program design(s) in my district has no clear academic or social benefits and may, in fact, be detrimental to pupils' progress and adjustment to school. Reading Strongly Agree ___ Agree ___ Disagree ___ Strongly Disagree Math Both 15. The Chaoter I program design in my district is judged by most teachers to be a sensible and beneficial plan. Reading Strongly Agree ___ Agree ___ Disagree ___ Strongly Disagree Math Both 16. The Chacter I program design(s) in my district is used primarily to satisfy Chapter I regulations. ___ Strongly Agree ___ Agree ____ Disagree ____ Strongly Disagree 17. The I program is judged by most administrators to be a sensible and beneficial plan. ___ Strongly Agree ____ Agree ____ Disagree ____ Strongly Disagree 18. Chacter I students are unconsciously neglected in the regular classroom. ___ Strongly Agree ____ Agree __ Disagree ___ Strongly Disagree 19. Overall, do you have any worries or concerns about the program or the Chapter I students in relation to program design in your district?

PAGE 315

APPENDIX I LETTER OF SUPPORT FROM DR. BOB L. TAYLOR

PAGE 316

299 Universit v nf Color ad.' .at I ...... I !IMI l.&lh "\U:..:I I .IIIII""' ltu\ lrtt1 l1n, ,., l ,,,,,,,,,lo ,:,,, I February 16, 1987 Dear Administrator: This is a letter of support for Darlene M. Baber's request to conduct a doctor's dissertation study using a sample of Chapter I program teachers and students from your school district. The "Pull Out Program in compensatory education is a practice surrounded by controversy; hence, it continues to be investigated. This study is being conducted with the cooperation of the Chapter I Office of the Colorado State Department of Education and the Chapter I Technical Assistance Center. The study is concerned with two Chapter I programs designed for reading and math instruction and remediation. These are the "Pull out approach and the "In Classroom" approach. The intent is to compare these two programs with respect to their impact on academic performance and self concepts of students involved. This is an important research problem. We hope that your examination of the research proposal will lead to your approval for the use of a sample of teachers and students from your school district. Sincerely yours, Bob L. Taylor Professor of Education

PAGE 317

APPENDIX J COMPLETE MATERIALS GIVEN TO TEACHERS ADMINISTERING THE S.D.Q.

PAGE 318

Dear 301 3195 Heidelberg Drive Bou 1 de_r. CO 80303 Thank you so much for your support with conducting the Chapter I study in your school district. I am most eager to work with you this month so that the data for the study can be collected in March, 1987. I realize that the post test for the N.C.E.'s will not be done until April and that part of the data can be gathered whenever you have the infonnation. Please find enclosed: A brief description of the study A Teacher Questionnaire which is subject to revision. A student Self Description Questionnaire also subject to some minor rivision and, An attached pupil data sheet. Please call me to let me know how we can proceed or if you have any questions. Work: 428-7503 Home: 499-3310 The best times to reach me are: A.M. 10:30 to 11:05 P.M. 12:15 to 1:15 At horne after 4:00 Thank you for your time and support. Sincerely. Darlene M. Baber

PAGE 319

To the Teacher: HELPFUL HINTS FOR STUDENT SELF QUESTIONNAIRE (SDO) ADMINISTRATION Dear Teacher. The SDQ has good validity and reliability for sixth grade and up. However. there was some uncertainty about the readability of the SDQ for third and fourth graders. 302 A pilot study was completed at this time last year in another Front Range School District. Six teachers and 54 third grade students participated. The results have been summarized in the HELPFUL HINTS listed below. Again. thank you for your participation. TIHE Teachers reported that the SDQ required 20 to 25 minutes to administer. LOCATION Teachers of students with an "in classroom" program chose to take students to a quiet location to reduce distractability. CONTENT VALIDITY Some children complained about having_already answered a question. Questions are restated to insure consistency of responses. It may help to explain that some questions may seem alike, but to try to answer each one with a thoughtful response. STUDENT CHOICES One very helpful teacher shared her description of questionnaire administration. "I drew and labelled the 5 boxes on the chalkboard and emphasized which end was true and which false by pointing it out after every 6 or 8 items. A child stopped me on #6 thinking he had gotten confused on this. I said "If you hate Math, Mark T, if not, Mark F" and pointed to the boxes. He decided he was doing it right. One child required a marker to stay on the right line. Another got off the row also at one ooint." VOCABULARY There were a few problems with vocabulary. Certain words may need to be explained. One third arade teacher had to explain "raised me" in question 34. Also, "featured" had to be explained in question 62. One child asked for the meaning of "athlete" in 56. Teachers reassured the students that "mathematics" meant math and that "school subjects" meant classes.

PAGE 320

303 PUPIL DATA SHEET TO BE ATTACHED TO EACH SELF DESCRIPTION QUESTIONNAIRE About the child: Age __ Birthdate Grade "---------------------Sex. __ Ethnic List other programs this child may attend. What time does the student school day begin? -----------------What time does the student school day end? ________ What days does the student receive I instruction? M T W T F At what hour does Chapter I instruction A.M. 7 8 9 10 11 p.m. 12 1 2 3 How many hours and minutes does the child spend in Chapter instruction per week? hours minutes --------------------How many total months has this child been in Chapter I?

PAGE 321

304 Pupil Code Sheet Please check one. Teacher Name ------------------------"In-Classroom" Program __ Building __________________________ "Pull-Out" Program Directions to Teacher: Please send a copy to me as soon as possible and retain a copy until post tests, N.C.E., are recorded. CONFIDENTIAL -The pupil names will be used only to assure proper matching of data. Please assign a number to each student; MATCH The number to the pre and post test scores for the Chapter I N.C. E. It is CRITICAL that the Student Self Description Ouestionnaire code and the code on this sheet can be matched with the N.C.E. data. Student Code Number-Same as on Questionnaire Student Name Pre-test Score N.C.E. THANK YOU SO MUCH FOR YOUR HELP Post-Test Score N.C.E.

PAGE 322

305 SELF DESCRIPTION QUESTIONNAIRE Student. r.ode .......................................................................................... Boy . Girl . Gr:de ...... . Age ....................... Sdloui.. ............. M ................................. Teacher ..................................................... .. This is a chance to look at rourself. It il not 11 teL There are no riyhl answers and everyone -.ill huvP. different answers. Be sure thai your answers show how you feel about yourself. PLEASE 00 NOT TALK ABOUT YOUR ANSWI!RS WITH ANYONE I:LSE. We will keep your answers private and not show them to anyone. When you are real.ly tu l.legin, please read ead1 sentance and decide your answer. (You may read quietly to yourself as I read aloud.) There are liv1 possible answers for eadl question -''True", "False", and three answers in bc:twcen. There are five boxes next to eadl sentence, one lor each of the answers. The answers are written at the lOJI of lha boxes. Choose your answer to a senlenca and put a ,/ in the bolt untJer the answer you choose. 00 NOT say your answer out loud or lalk about it with anyone else. Be lore you start there are three examples below. Somebody named Bob has already answered two of these sentences to show you how to do it. In the third one you must dloose your own answer and put in your own v' SOME TIMES MOSTLY FALSE, MOSTLY FALSE FALSE SOME TRUE 'TRUE TIMES TRUE EXAMPLES 1. I like to read comic books ................................... .. 2. (Bob put a ,/ in the bo" under the answer "TRUE". This means that ha really likes to read comic books. If Bob did not l1ke to read comic boolu very much, he would have answered "FALSE" or "MOSTLY FALSE".) In general, I am neat and lidy ............................... 2 CJ CJ m Cl c::J 2 !Bob answeretJ "SOMEliMES FALSE, SOMETIMES TRUE" because ha is not very neat, but he is not very messy c;lher.) 3. I like to watch T.V .............................................. 3 CJ CJ CJ CJ CJ 3 1For this '{OU have tO choose the answer that is best for you. First you must decitJe if the sentence is "TRUE" or "FALSE" or somewhere in between. II you really like to watch T.V. a lot you wo,ltl ''TRUE" by cutting a o/ in the last box. If you hate watching T.V. you woult.l answer "FALSE" IJy puting a \/.in the first bolt. If your answer is somewhere in between then you would chouse one of the: ther three boxes.) If yow .,ant To chenJ'' you have marked you should cross out the \/' and put a new V ;, on other bOlt on the line. f'or all the sentences be sure that your V is on the same line as the sentence are answering. You should have one answer and only one answer for each sentence. Do nol leave uu1 any <1 I the senlences. I r you ha1e ant out l'P your hand. Turn over the page and IJeyin. Once you have started, PLI.:.ti:;E [IQ NOT TALK.

PAGE 323

SOMl TIMU MOSTLY '-"LSI, -TLY 306 '-"LII '-"LSI SOMI TAUI TIIUI TOMEI TAUI 1. lamgoocllooklng ..................................... .......... I 0 c:J 0 CJ 01 2. I'm good at all SCHOOL SUBJECTS.................... 2 0 0 0 0 0 2 3. lcanrunlast .... ..... 3c=Jc:JDD03 4. lgetgoodmarksinREADING ...... ................. 4 0 0 0 0 04 5. My perentsundeRtand me................................... 5 0 c::J 0 CJ c::J 5 6. lhateMATHEMATICS ........................................ 6 c::J c::J c:J c:J 06 7. lhavelotsoflriends ............................................. 70 c:J 0 c::J c:J 7 B. llikethewayllook ............................................. sc:J DO CJc::Js 9. I enjoy doing work in all SCHOOL SUBJECTS.... 9 c:J c::J 0 c::J c::J 9 10. I like to run and play hard .................................... 100 0 c::J c::J c::Jto 11. llikeREADINO ..................................... 110 0 0 c:J 011 12. My parems are usually unhappy or disappointed withwhatldo ...................................................... 120 c:J c::J CJ 012 13. WorkinMATHEMATICSiseasylorme .............. 13c:J c:J 0 CJ c::J13 14. lmakelriendseasily ................................... : ......... 140 c::J c::J CJ c:J14 15. lhaveapleasantlookingl!'ca ............................... tsO 0 c::J CJ 015 16. lgetgoodmarksinaiiSCHOOLSUBJECTS ........ t6c:J c:J c::J CJ W16 17. lhatesPortsandgames ........................................ 11c:J c::J c::J 0 011 18. l'mgoodatREADING ........ ............................... sO 0 c:J 0 01a 19. llikemyparenu ................. : ................................ 190 0 0 CJ 019 20. llooklorwardtoMATHEMATICS ...................... 200 0 0 CJ C]2o 21. Most kids have more friends than I do ................. 210 c:J 0 CJ 021 22. I am a nicalooklng penon ............ ...................... 220 D c:J CJ 022 23. lhateaiiSCHOOLSUBJECTS ............................. 23W D c::J CJ 023 24. I enjoy sports and games ...................................... 240 0 0 CJ 0 24 25. I am internted in READING ................................ 25 D c::J 0 CJ 025 26. My parents like me ............................................... 26 W 0 D CJ [:=J26

PAGE 324

SOME TIMES MOSTLY F"'LS!, MOSTLY 307 FALSE FALSE SC"'E TAU& TAUI TIME! tAU! 27. lgetgaodmarksinMATHEMATICS ................... 270 0 0 0 021 28. I get along with other kids easily ......................... 280 0 c:J 0 028 29. ldolotsolimportanttnings ................................. 290 c:J c:J 0 029 30. lam ugly .............................................................. 300 0 c:J D 03o 31. llearnthingsquicklyinaiiSCHOOLSUBJEClS. 310 c:J c:J 0 031 32. I have goad muscles .............................................. 320 0 c:J c:J 032 33. lamdumbatREAOING ...................................... 330 0 0 0 033 34. II I nave children ol my own I want tn hring '""'" up like my parenu raised me ............................... 34 0 0 0 c:J 034 35. I am Interested in MATHEMATICS ...................... 35 c:J E::J E::J CJ c:J35 38. lameasytolike ................................................... 360 0 CJ 1::] 03& 37. Ovaralllamnogood ........................... ............... 370 c::J CJ 0 037 38. Otherkidsthinklamgoodlooking ...................... 311c::J Q c::J.CJ 038 39. I am Interested in all SCHOOL SUBJECTS .......... J!lC] c:J C] c:J c:J39 40. lamgoadatsports ............................................... 400 0 0 0 040 41. lenjoydalngworkinREAOINO .................... : .... 410 0 0 0 041 42. My parents and I spend a lot oltitn ...... 42 l::J 0 0 0 042 43. llearnthingsquicklylnMATHEMATICS ............ CJ 0 0 043 44. Other kids want me to bo their lriend .................. 4 0 0 0 c:J 044 45. lngeneralllikeheingthewavlam ...................... 450 0 c:J 0 C]4s 46. lhaveagoodlookinghody .................................. 4r.C] c::J c:J 0 04& 47. I am dumb in all SCHOOL SUBJEClS ................. 47c:J c::J 0 c:::J c:J 47 48. I can run a long woy without Slotroirg ................. 0 0 c:::J 0 48 49. Work in READING is v lor mo..................... 4!1 CJ c::J CJ c:::J 0 49 50. Myparentsareeasytolalkto .............................. soCJ D D D [::=Jso 51. I like MATHEMATICS ......................................... 51 CJ D C] CJ C]s1 52. ............. 520 0 0 0 052

PAGE 325

308 lOIII nwu rAUl, rALU rALII IIIIMI TIIUI T11U1 niG 1111111 53. Ovenlll haw a lot to be proud of .... 53c::::J [:::J CJ 0 053 "54. I'm betttr looking than most of my lriendl ..... _. 54 c::::J c::::J CJ D 054 ss. 1 look lorv.ard to au SCHOOL 55(:=J c:J 0 0 0 56. I am a good athletl ............................................ ssc::::J c::::J CJ 0 D 58 57. I look lorv.ardtoREADING ....... _______ ,_, .. S7(:=J [:::J CJ 0 c:Js7 58. I get along well my parents ........................... 58 c::::J [:::J CJ 0 Osa 59. I'm good at MATHEMATICS ............................... sgc::J [:::J D 0 Dsg 60. I am popular with kids of my own age................. 60 c::::J [:::J c:J D D 60 61. I can'tdoanythingright ..................................... 61 c::J c:J CJ D 0&1 62. I have nice features like nose, and eyll'l, and hair .. 62 c::::J c:J CJ 0 0&2 63. Work in all SCHOOL SUBJECTS is easy lor me ... 6:lc::::J c:J c:J D 0&3 64. I'm good at throwing a ball ....... ......................... 64(:=J c::::J c:J D 0&4 ss. 1 hate READING .................................................. ssc::J c::J c::J DOss 66. My garents ard I haw a lot of fun together .......... ssc::J D D D Des 67. I can do things as well as most other people ......... 67 CJ c:J c::J D 0&7 68. I enjoy doing work in MATHEMATICS ............... sac::J c::J D D 068 69. Most other kids like me ........................................ 69 c::J c::::J c::J D D 69 70. Other people think I am a good person ................. 70 c::J c::J c:J D D10 71. .............................. 11c::J c::J 0 DOn 72. Alotolthingsaboutmearegood ........................ nc::J D D D 072 73. I leam things Quickly in READING .................... 73 c::J c::J c:J c:J W73 74. l'masgoodasmostotherpeople .......................... 74c::J c::J 0 c:J [:J74 75. lamdumbatMATHEMATICS ............................ 1sc::J c::J c::J 0 [:J7S 76. When I do something, I do it well ......................... 76 c::J c::J c::J c:J [:J 76

PAGE 326

Dear 309 3195 Heidelberg Drive Boulder, CO 80303 Work phone: 428-7503 Home phone: 499-3340 I am a teacher in Adams County School District 50, This Chapter I study is part of the research for my doctoral dissertation. I need your help to answer questions regardinq proqram desi9n. The enclosed survey is not an evaluation of your Chapter I program. I greatly appreciate your participation in the Chapter I study. This part of the study is STRICTLY CONFIDENTIAL. It wi 11 require about twenty minutes of your time. As a teacher, I know the sacrifice in time that this questionnaire requires. Your response is critical since only teachers in certain school districts have been asked to participate. This is not a random sample. This study has the support of the Chapter I Office of the Colorado State Department and of the Chapter I Technical Assistance Center. PURPOSE OF THE STUDY The Chapter I Office of the State Depntment of Colorado has requested that a study be done to compare two Chapter I program designs for reading in struction and remediation, and mathematics where applicable. The first design fs labeled "pull out" and is structured so that Chapter rstudents are grouped and removed from the regular education setting for a specified period of time each day. The second design is labeled ''in classroom" whereby the Chapter I teacher teams with the regular education teacher to plan and implement instruc tion. The Chapter I teachers and-re9ular education teachers are being asked to contribute their expert opinions on program design within a specific site and program effects on Chapter I students. DIRECTIONS To insure confidentiality Please fill out the questionnaire as candidly as possible. -Return it in the self addressed enclosed envelope to insure confidentiality. -Please feel free to add your own feelings or observations. Sincerely, Darlene M. Baber

PAGE 327

310 TEACHER QUESTIONNAIRE Please check Building design __ "Pull out""In classY Other --Specify 1. How many students are enrolled in your building?----2. What percent are Chapter I students?----3. Who decided upon the program design in your building, "pull out" versus "in classroom" instruction? Central Administrator __ Principal __ Chapter I Coordinator --Title: Other: Please explain _________________ 4. For what reasons was the design selected? Check as many as apply. time of instruction __ space available __ research findings __ varying remedial needs professional judgment --about instruction number of staff __ number of pupils __ preference of teachers professional judgment --about curriculum to cut excess costs __ other: Please explain ________________ 5. What are the most beneficial effects of the Chapter I program? 6. What are the most detrimental effects of the Chapter I program? 7. What problems are impacting your success with the program?

PAGE 328

311 TEACHER QUESTIONNAIRE page 2 8. My students miss no academic subject since remedial instruction occurs during study periods in the regular class. _Yes _No 9. Do your students miss regular classroom instruction to receive 10. Chapter I reading? Yes No My students miss social studies. My students miss science. My students miss other classes. The classes missed are __________________ How much time per Do your students miss regular classroom instruction to receive Chapter I math? My students miss social studies. My students miss science. My students miss other classes. Yes No The classes missed How much time per day? __________________ 11. Does the Chapter I program supplant (replace) or supplement the regular education program? 12. Are you a remedial subject matter special fst? 13. What are your qualifications and/or certifications for your position? 14. Do you receive extra pay for your role in instructing remedial pupils? Yes No 15. Does the caucasian population of your Chapter I pupils reflect the population of the school in number? more than same as fewer than Please specify if you can school population __ Chapter I population __ 16. How well do you expect your Chapter I students to perform academically as compared to regular education students at their grade level? better than about the same less well 17. Does your program utilize peer tutoring? __ Yes No

PAGE 329

TEACHER QUESTIONNAIRE page 3 For questions 18 to 20 please check reading __ math __ or both __ 18. The Chapter I program design in my building has no clear academic or social benefits and may, in fact. be detrimental to pupils' progress and adjustment to school. 312 Reading Strongly Agree Agree ___ Disagree ___ Strongly Disagree Math ----Both -19. The Chapter I program design in my building is judged by teachers to be a sensible and beneficial plan. Reading Strongly Agree Agree __ Disagree __ Strongly Disagree Math ----Both -20. The Chapter I program design in my building is used primarily to satisfy Chapter I regulations. __ Strongly Agree __ Agree __ Disagree _Strongly Disagree 21. The Chapter I program is judged by administrators to be a sensible and beneficial plan. _Strongly Agree __ Asree __ Disagree __ Strongly Disagree 22. Chapter I students are unconsciously neglected in the regular classroom. __ Strongly Agree __ Agree __ Disagree __ Strongly Disagree 23. Overall, do you have any worries or concerns about the program or the Chapter I students in relation to program design in your building? 24. Would you please-describe the Chapter I program in your building in the space below or on the pack. For example, are there aspects of your program that may be unique .to other Chapter I programs? Are there aspects that may affect the results of this study? Thank you so much for your time. Name ______________ __ District. ___________ Address. __ Bufl ding _________ Position. ___________ I would like to thank you personally. After your building responses are collated, I will destroy all names.

PAGE 330

APPENDIX K QUESTIONS REGARDING THE STATUS OF THE CHAPTER I PROGRAM

PAGE 331

Additional Teacher Survey Results Number of Students Enrolled in Building Question 1. How many students are enrolled in your building? 314 Of the 80 teachers questioned, 23 did not respond to this question. One teacher reported having three students, and one teacher reported having 60 students. The values of responses of the remaining 55 teachers were 229 to 597: Mean, 453.789; Mode, 575.000; Kurtosis, .859; Standard Error, 18.866; Standard Deviation, 142.436; Standard Error of Kurtosis, .623; Median, 494.000; Variance, 20288.098; Skewness, -1.128. The distribu-tion of building populations is a negatively skewed normal distribution. Building populations are a representative sample of elementary school size with two atypical cases. Therefore, the elementary Chapter I student sample can be as a representative sample by building size. Percent of Chapter I Students? Question 2. What percent are Chapter I students? As shown in Table 27, of the 80 teachers, 34, or 42.5 per-cent, did not respond to the question. However, of the 46 teachers responding, 25, or a little over half, reported having 13 to 15 percent of the students in their building enrolled in Chapter I. Approximately 74 percent of the teachers had 15

PAGE 332

315 percent or fewer students enrolled in Chapter I. Two percentages appear to be atypical scores, 90 percent and 55 percent. The researcher verified school enrollments, and the 90 percent was an error and should have been recorded as 10 percent. Other scores reflect a normal distribution of Chapter I students in a school building.

PAGE 333

316 Table K-1 Question 2: Percent of Cha11ter I Students in Buildinga Student Valid Cumulative Percentb Frequency Percent Percent Percent 6 3 3.8 6.5 6.5 7 1 1.3 2.2 8.7 10 3 3.8 6.5 15.2 11 1 1.3 2.2 17.4 12 1 1.3 2.2 19.6 13 15 18.8 32.6 52.2 14 1 1.3 2.2 54.3 15 9 11.3 19.6 73.9 18 3 3.8 6.5 80.4 19 2 2.5 4.3 84.8 20 1 1.3 2.2 87.0 25 1 1.3 2.2 89.1 30 2 2.5 4.3 93.5 33 1 1.3 2.2 95.7 55 1 1.3 2.2 97.8 90 1 1.3 2.2 100.0 34 42.5 missing Total 80 100.0 100.0 Mean 17.326 Mode 13.000 Kurtosis 18.714 Standard Error 2.015 Standard Standard Deviation 13.664 E. of Kurtosis .688 Median 13.000 Variance 186.714 aof the 80 teachers, 34, or 42.5% did not respond. The range was from 6% to 90%. bstudent percent refers to the percent of students in the building serviced by Chapter I.

PAGE 334

Questions Regarding the Status of the Chapter I Program Teachers as Remedial Subject Matter Specialists Question 12. Are you a Remedial Subject Matter Specialist? 317 The majority of Chapter I teachers (67.9 percent), or 19, of the Chapter I teachers responded that they were remedial subject matter specialists, while only 7.7 percent, or four, of the regular classroom teachers similarly responded. Of the 67.9 percent of teachers responding "yes" to Question 12, 20.4 percent, or 11, were pull-out teachers and 37.5 percent were in-class teachers. Results show that both pull-out and in-class Chapter I teachers consider themselves as remedial subject matter specialists (see Table L-2).

PAGE 335

318 Table K-2 Question 12: Teacher;s Remedial Subject Matter S12ecialists Teachers Yes No Missing Chapter I 67.9% 28.6% 3.6% (19) (8) (1) Classroom 7.7% 86.5% 5.8% (4) (45) (3) Pull Out 20.4% 70.4% 5.6% (11) (40) (3) In Class 37.5% 62.5% (6) (10) (0) Other 60.0% 30.0% 10.0% (6) (3) (1)

PAGE 336

Qualifications and/or Certification for Position 319 Question 13. What are your qualifications and/or certifications for your position? Results show that only one Chapter I teacher (3.6 percent) in the pull-out program was endorsed in special education while four regular education teachers in the pull-out program (7.7 percent) and no in-class teachers were similarly endorsed. There were no Chapter I teachers stating less than five years teaching experience, while 28.8 percent, or 15, of the regular classroom teachers claimed to have one to five years experience in the classroom. Chapter I teachers state that they do more remedial reading classwork (25 percent), or seven, than classroom teachers (3.8 percent), or two. These teachers are both in-class (18.8 percent), or 3, and pull-out (11.1 percent), or six. Also, two more Chapter I teachers (32.1 percent), or nine, have a Master's degree in elementary education or reading than do classroom teachers (13.5 percent), or seven. Of these teachers with advanced degrees, 24.1 percent, or 13, are in the pull-out design as compared to in-class . with 12.5 percent, or two. Percentages are similar, showing that Chapter I teachers in both program designs are as likely or more slighly likely than regular classroom teachers to have an advanced degree. Classroom teachers are more likely to have a Bachelor's degree (19.2 percent), or 10, and an elementary teaching certificate

PAGE 337

320 (36.5 percent), or 19, as compared to Chapter I teachers, with 10.7 percent, or three, and 7.1 percent, or two, respectively. However, classroom teachers, especially in the pull-out design (7.4 percent), or four, seem to be the veteran teachers with more than 15 years teaching experience. Results show, therefore, that Chapter I teachers tend to have more than five years teaching experience, and have more advanced preparation either in remedial reading classwork or in an advanced degree program than regular classroom teachers, overall (see Table K-3). Teacher Receives Extra Pay for Role in Instructing Remedial Pupils? Not a single teacher responded that extra pay was received for instructing remedial students (see Table K-4).

PAGE 338

Table K-3 Teacher's Qualifications and/or Certification for Position WbatAre the Teacher's Qualifications and/or Certifications for His or Her Position? Chapter I Classroom Teachers Pull out In Class other

PAGE 339

M.A. degree 32.U 13.5\ (elementary education; (9) (7) reading) CAA approved 3.6 (1) (0) Teaching certificate/elementary 7.1% 36.5\ education (2) (19) (Type B) Bilingual education/secondary 1.9\ Spanish (0) (1) None 3.8\ (O) (2) More than 15 years teaching 3.6\ 7.7\ experience (1) (4) TypeD 1.9l (0) (1) 24.U (13) 1.9\ (1) 31.5\ (17) 1.9\ ( 1) J.n (2) 7.4\ (4) 1.9\ (1) 12.5l (2) (0) 25 (4) (0) (0) 6.3% ( 1) (0) lOt (1) (0) (0) (0) (0) (0) (0) w N N

PAGE 340

Table K-3 (Continued) Missing Chapter I 42.9\ (42) Classroom 28.8\ (15) Teachers Pull out 22.2\ (12) In Class 37.5\ (6) Other 90.0\ (9)

PAGE 341

324 Table K-4 Teacher Receives Extra for His or Her Role in Instructing Remedial Pu];!ils? Teachers Yes No Missing Chapter I 57.1% 42.9% (0) (12) Classroom 67.3% 32.7% (0) (35) (17) Pull Out 72.2% 27.8% (0) (39) (15) In Class 68.8% 31.3% (0) (11) (5) Other 10.0% 90.0% (0) (1) (9) Peer Tutoring In-class teachers were evenly divided, with 37.5 percent answering "yes" and 37.5 percent answering "no." However, a higher percentage of in-class teachers than pull-out teachers (16.7 percent) utilized peer tutoring. A slightly higher percentage of Chapter I teachers (28.6 percent) than class-room teachers (19.2 percent) responded that peer tutoring was utilized. Therefore, results show that in-class Chapter I teachers may utilize peer tutoring slightly more often than other categories of teachers (see Table K-5).

PAGE 342

Table K-5 Peer Tutoring Teachers Yes No Missing Chapter I 28.6% (8) Classroom 19.2% (10) Pull Out 16.7% (9) In Class 37.5% (6) Other 30.0% (3) Chapter I Program Judged by Administrators to be Sensible and Beneficial 53.6% (15) 59.6% (31) 63.0% (34) 37.5% (6) 60.0% (6) Question 21. The Chapter I program is judged by administrators to be a sensible and beneficial plan. 17.9% (5) 19.2% (10) 20.4% (11) 25% (4) 10.0% (1) 325 The overwhelming response of teachers in all categories was either to agree or to strongly agree with the statement above (see Table K-6). Results of the study show conclusively that Chapter I programs are judged by administrators to be a sensible and beneficial plan for both program designs, in-class and pull-out.

PAGE 343

Table K-6 Cha11ter I Program Is Judged Administrators be a Sensible and Beneficial Plan Strongly Teachers Agree Agree Disagree Chapter I 67.9% 32.1% (19) (9) Classroom 28.8% 35.8% (15) (29) Pull Out 37.0% 55.6% (20) (30) In Class 50.0% 25.0% (8) (4) Other 60.0% 40.0% (6) (4) Cha11ter I Program Design(s) Used to Chapter I Regulations (0) (0) (0) (0) (0) 326 to Strongly Disagree Missing (0) (0) 15.4% (0) (8) 7.4% (0) (4) 25.0% (0) (4) (0) (0) Question 16. The Chapter I program design in my building is used primarily to satisfy Chapter I regulations. As shown in Table K-7, a total of 49 percent of the coordinators in the pull-out category disagreed or strongly disagreed that the Chapter I program design was used primarily to satisfy Chapter I regulations. However, 21.6 percent of the pull-out coordinators agreed with the statement above and another pull-out coordinator strongly agreed. For coordinators with in

PAGE 344

327 class programs, a total of 17.7 percent disagreed or strongly disagreed, while 7.8 percent agreed with the statement. Results show that slightly more than twice as many coordinators with both program designs disagreed that the Chapter I program was used primarily to satisfy Chapter I regulations. Table K-7 Chapter I Program Design Used Primarily to Satisfy Chapter I Regulations Total In-class Pull-out Strongly Agree 2.0% ( 1) ( 0) 2.0% ( 1) Agree Disagree 29.4% 39.2% (15) (20) 7.8% 11.8% ( 4) ( 6) 21.6% 27.4% (11) (14) Strongly Disagree 27.5% (14) 5.9% ( 3) 21.6% (11) Missing 2.0% ( 1) It was of interest to compare Table 49, reasons for the selection of program design, with Table K-2, showing reasons for the program designs' continued use by a district. Only one coordinator with the in-class program, as shown in Table 49, reported that the program design was selected to satisfy Chapter I federal regulations. No coordinators with pull-out designs exclusively responded that selection was due to federal

PAGE 345

328 regulations. Yet, almost one third of the coordinators in both program designs reported that the design that was in use con-tinued mostly as a result of federal Chapter I regulations. although there is no support for a conclusion that Chapter I coordinators would change the program's designs that were originally selected if it were not for perceptions about federal regulations, one result is clear. Perceptions about Chapter I regulations influence both program designs equally, and the pull-out design was not continued as the predominant service model primarily to satisfy federal regulations any more than was the in-class program design. Chapter I Program Judged to Be Sensible and Beneficial Question 17. The Chapter I program design in my building is judged by teachers to be a sensible and beneficial plan. One pull-out program coordinator (2.0 percent) responded negatively to the above statement. Results show that a sig-nificant majority of coordinators support the conclusion that administrators from both program designs, in-class and pull-out, judge the program to be sensible and beneficial (see Table K-8).

PAGE 346

329 Table K-8 Cha12ter I Program Judged bl Most Administrators to be Sensible and Beneficial Strongly Strongly Agree Agree Disagree Disagree Missing Total 60.8% 37.3% 2.0% (31) (19) ( 1) ( 0) ( 0) In-class 17.6% 9.8% ( 9) ( 5) ( 0) ( 0) Pull-out 43.2% 22.8% (22) (14) ( 1) ( 0) The Caucasian Po12ulation Question 15. Does the Caucasian population of your Chapter I pupils reflect population of school in number? Results shown in Table K-9 indicate that minority students are enrolled in Chapter I either in the same numbers as in the regular education population or in greater numbers. Where Caucasians are fewer in Chapter I populations, it is assumed that minorities are greater. Minority populations are the same pro-portion in Chapter I as in regular education populations for both program designs, but somewhat larger numbers are reported in the pull-out design (44.4 percent), or 24, as compared to the in-class design (25 percent), or four. However, minorities seem to be in

PAGE 347

330 greater numbers in the in-class programs when percentages of teachers responding are compared (56.3 percent), or nine, as compared to the pull-out design (35.2 percent), or 19. Table K-9 The Caucasian Po11ulation More Than Same As Fewer Than Teachers Population Population Population Missing Chapter I 3.6% 46.4% 50.0% 0.0% (1) (13) (14) (0) Classroom 5.8% 40.4% 30.8% 23.1% (3) (21) (16) (12) Pull-Out 5.6% 44.4% 35.2% 14.8% (3) (24) (19) (8) In-Class 6.3% 25% 56.3% 12.5% (1) (4) (9) (2) Other 60.0 20.0% 20.0% (6) (0) (2) (2) Percentage of Caucasian Cha11ter I Students Range No. of Standard Teachers Students Mean Deviation Median Skewness Chapter I 84 52.1% 33.8 46.0% .226 Classroom 93 40.8% 25.4 46.0% .541 Pull-Out 93 53% 26.6 46.0% .462 In-Class 47 27.7% 18.4% 15.0% 1.311

PAGE 348

331 Other 9 cases missing, inadequate data. Overall, numbers of minorities in Chapter I programs are not reduced when a pull-out model is employedas seen in Table K-9. In fact, when the percentages of Caucasian students enrolled in Chapter I was compared to those in total building populations of regular education students, the means of confirmed student members is of interest. The mean Caucasian students in the pull-out design (53 percent) was almost twice as large as the mean of in-class students (27.7 percent). Therefore, numbers of minority students were actually reduced with the pull-out program. Expectation of Academic Performance Question 16. How well are Chapter I students expected to perform academically? As shown in Table K-10, not a single teacher responded that Chapter I students were expected to perform better academically than regular education students. However, a surprising number in the pull-out design (48.1 percent) expected Chapter I students to perform academically about as well with a somewhat smaller percentage of responses (37.5 percent). Somewhat smaller percentages of teachers in all categories expected Chapter I students to perform less well academically than regular educa-tion students. It was of interest that more teachers in every category expected Chapter I students to perform as well as of classroom teachers (50.0 percent), particularly teachers in regular students rather than to perform less well academically.

PAGE 349

332 Table K-10 Ex)2ectations of Academic Performance Less Well Better Than About the Same Than Regular as Regular Regular Teachers Classroom Classroom Classroom Missing Chapter I 25% 0) (7) Classroom 50.0% 0) (26) Pull-Out 48.1% (0) (26) In-Class 37.5% (0) (6) Other 10.0% (0) (1) Cha)2ter I Program Has No Clear Academic or Social Benefits and May Be Detrimental 25% 50.0% (7) (14) 38.5% 11.5% (20) (6) 40.7% 11.1% (22) (6) 31.3% 31.3% (5) (5) 90.0% (0) (9) Question 18. The Chapter I program design in my building has no clear academic or social benefits and may, in fact, be detrimental to pupils' progress and adjustment to school. Only one in-class teacher (6.3 percent), Chapter I (3.6 percent) agreed with statement number 18 in the area of math. As shown in Table K-11, the majority of teachers strongly disagreed with the statement. Results of the study show that teachers overwhelmingly agree that Chapter I program design, whether in-class or pull-out, clearly has academic and social benefits. The

PAGE 350

333 Chapter I program was not viewed as detrimental to pupil progress. Table K-11 I Program Design in Building Has No Clear Academic or Social Benefits and be Detrimental to Progress Strongly Strongly Teachers Agree Agree Disagree Disagree Missing Chapter I Reading 3.6% 78.6 17.9% (0) (0) (1) (22) (5) Math 3.6% 3.6% 14.3% 78.6% (0) (1) (1) (4) (22) Both 26% 92.9% (0) (0) (0) (2) (51) Classroom Reading 40.4% 55.8% 3.8% (0) (0) (21) (29) (2) Math 3.8% 3.8% 92.3% (0) (0) (2) (2) (48) Both 3.8% 1. 9% 94.2% (0) (0) (2) (1) (49) Pull-Out Reading 35.2% 61.1% 3.7% (0) (0) (19) (33) (2) Math 1. 9% 5.6% 92.6% (0) (0) (1) (3) (SO) Both 3.7% 3.7% 92.6% (0) (0) (2) (2) (50) In-Class Reading 18.8% 62.5% 18.8% (0) (0) (3) (10) (3) Math 6.3% 6.3% 81.3% 81.3% (0) (1) ( 1) (13) (13) Both 100.0% (0) (0) (0) (0) Other Reading 80.0% 20.0% (0) (0) (0) (8) (2) Math 10.0% 20.0% 70.0% (0) (0) (1) (2) (7) Both 10.0% 90.0% (0) (1) (0) (0) (9)

PAGE 351

Chapter I Program Design Used Primarily to Satisfy Chapter I Regulation Question 20. The Chapter I program design in my building is used primarily to satisfy Chapter I regulations. It is an interesting result of this study that an 334 approximately equal percentage of the teachers in each category agreed and disagreed with the statement above. However, as presented in Table K-12, more teachers in each category strongly disagreed than strongly agreed. The results of this study show that the tendency is to select a Chapter I design in a building for reasons other than to satisfy Chapter I regulations; how-ever, more teachers in both in-class (27.5 percent total), and Table K-12 Question 20: Chapter I Program Design in Building Is Used Primarily to Satisfy Chapter I Regulations Strongly Strongly Teachers Agree Agree Disagree Disagree Missing Chapter I 10.7% 21.4% 39.3% 25.0% 3.6% (3) (6) (11) (7) (1) Classroom 5.8% 30.8% 30.8% 19.2% 13.5% (3) (16) (16) (10) (7) Pull-Out 5.6% 27.8% 37.0% 20.4% 9.3% (3) (15) (20) (11) (5) In-Class 12.5% 25.0% 31.3% 18.8% 12.5% (2) (4) (5) (3) (2)

PAGE 352

335 Table K-12 (continued) Strongly Strongly Teachers Agree Agree Disagree Disagree Missing Other 10.0% 30.0% 20.0% 30.0% 10.0% (1) (3) (2) (3) (1) pull-out (33.4 percent total), either agree or strongly agree that the Chapter I program design was used primarily to satisfy Chapter I regulations. Therefore, the criticism that pull-out programs are designed primarily as a result of Chapter I regulations appears to be equally true, or untrue, for in-class program designs. Chapter I Students Unconsciously Neglected Question 22. Chapter I students are unconsciously neglected in the regular classroom. Two Chapter I teachers (7.2 percent total), one class-room teacher (1.9 percent, one pull-out teacher (1.19 percent) and one in-class teacher (1.9 percent) agreed that students are unconsciously neglected in the regular classroom (see Table K-13). The mixed program design had a surprisingly large percentage (SO percent) agreeing with this statement and may account for four of the teachers that agreed since no in-class or pull-out teachers strongly agreed with the statement. The

PAGE 353

336 majority of teachers in all categories disagreed with the statement, and pull-out teachers showed the strongest disagreement, 64.8 percent as compared to in-class, 43.8 percent. Results of the study show that Chapter I students are not unconsciously neglected in the regular classroom in either program design, regular or in-class design. Table K-13 Question 22; I Students Are Neglected in the Regular Classroom Strongly Strongly Teachers Agree Agree Disagree Disagree Missing Chapter I 3.6% 3.6% 46.4% 46.4% (1) (1) (13) (13) (0) Classroom 1. 9% 30.0% 65.4% 1. 9% (0) (1) (16) (34) (1) Pull-Out 1. 9% 33.3% 64.8% (0) (1) (18) (35) (0) In-Class 6.3% 43.8% 43.8% 6.3% (0) (1) (7) (7) (1) Other 10.0% 40.0% 50.0% (1) (4) (0) (5) (0)

PAGE 354

Aspects of Chapter I Program Unique to Other Chapter I Programs and Which Would Affect Results of This Study 337 Question 24. Would you please describe the Chapter I program in your building in the space below or on the back. For example, are there aspects of your program that may be unique. to other Chapter I programs? Are there aspects that may affect the results of this study? Responses indicated that there were no aspects of program design that would affect the results of the study. Additional Coordinator Survey Results Grades Served by Chapter I Question 1. What grades in your district are served by Chapter I? The majority of the 51 Chapter I programs service grades kindergarten to eighth grade (see Table K-14). No major dif-ferences were noted between districts with in-class and pull-out programs in the distribution. This study was concerned with students primarily in the third through fifth grades. Responses indicate that all coordinators have programs contain-ing the age population of interest to this study.

PAGE 355

Table K-14 Question 1. Grades in District Served by Chapter I Grades In-class K-4 1 K-5 0 K-6 0 K-8 2 K-9 1 K-12 2 1-4 0 1-5 0 1-6 0 1-8 4 1-9 0 1-12 2 2-5 1 2-8 1 2-12 2 Missing: 1 in-class, 1 pull-out Total 51 districts responding Number of Chapter I Students Enrolled Pull-out 0 1 2 5 0 1 2 4 6 7 1 3 0 0 1 Question 2: How many Chapter I students are enrolled? 338 As shown in Table K-15, the range of students enrolled was 11 to 7,013 Chapter I students from the 51 districts, which is a very divergent sample from within the state of Colorado. The mean or average size of the Chapter I program was a 323-student enrollment. The variance in size is extremely large at 992199.077 due to five districts with student Chapter I popula-tions of 400 to 7,013, an atypical number. The majority of

PAGE 356

339 Chapter I students were in districts (44) with populations of less than 400 Chapter I students. Table K-15 Number of Chapter I Students Enrolled Number of Students Frequency 1-100 101-200 201-300 301-400 401-500 501-600 601-700 701-800 801-900 901-1000 1001-7,013 Missing Total Range 11 to 7,013. 25 8 6 5 1 0 1 1 0 1 1 2 51 Mean -323.082; Std. Error -142.299; Median 89.000; Mode 80.00; Std. Deviation -996.092; Variance -992199.077; Kurtosis -44.947; S E Kurtosis -.668; Skewness -6.577. Total District Enrollment Question 3. What is the total district enrollment? The range was from 508 to 60,256 students enrolled in the 51 school districts responding. The distribution reflects the distribution in Table K-16. When Table K_l7 is compared to Table K-16, there are similarly positively skewed distributions and the variance of the total population is also very large, 1014700124.12. The mean for the total enrollment was 4,978

PAGE 357

Table K-16 Question 3. Total District Enrollment Number of Students 1-1,000 1,0001-2,000 2,001-3,000 3,001-4,000 4,001-5,000 5,001-6,000 6,001-7,000 7,001-8,000 8,001-9,000 9.001-10,000 10,000-20,000 20,"001-30,000 30,001-(60,256) Missing Total Range from 508 to 60,256. Frequency 16 9 6 3 5 1 1 0 1 2 0 3 1 3 51 Mean-4977.667; Std. Error-1453.946; Median1480.000; Mode 195.00; Std. Deviation 10073.232; Variance 101470012; Kurtosis 0; S E Kurtosis -.674; Skewness 4.120. 340 students. Of the 51 districts responding, 39 had total enroll-ments of fewer than 5,000 students. Results indicate that in Colorado most districts were not urban or metropolitan areas (see Table 45). This study contained student samples from districts that were not urban or metropolitan. Results of the study may be similar to those that could be expected from the majority of the districts from which Coordinators completed and returned surveys.

PAGE 358

Number of Chapter I Teachers Having Pull-out Program Design 341 Question 4. How many Chapter I teachers have a "pull-out11 program design? Table K-17 indicates that all 51 districts in the sample utilize the pull-out design. Total teachers using the pull-out design were 271, with two districts missing. Number of Chapter I Teachers Having In-class Program Design Question 5. How many Chapter I teachers have an "in classroom" program design? Table K-18 indicates that 33, or 66 percent, of the 51 dis-tricts contain no in-classroom programs. Fourteen districts, or Chapter I service providers utilize in-class programs totaling 271 teachers from all districts. Question 6. Number of Teachers Utilizing Other Chapter I Designs Coordinators responded that five districts, or approxi-rnately 9.8 percent of the 51 districts responding, had teachers utilizing other program alternatives, described as follows. 6.1 Chapter I offered as a high school elective course to supplement the language arts program. 6.2 Preschool staff deliveree direct and consultative services at children's horne and used parents as service providers. 6.3 Service provided under "delinquent or neglected" to

PAGE 359

students under 21 years of age and functioning academically below the 8.0 grade level in an 'adult correctional' facility. Table K-17 Teachers Utilizing a Pull-Out Design Number of Teachers Frequency 1 12 2 9 3 5 4 5 5 1 6 6 7 2 8 2 9 1 16 1 17 1 25 1 68 1 Missing 2 Total 51 Percent 23.5 17.6 9.8 9.8 2.0 11.8 3.9 3.9 2.0 2.0 2.0 2.0 2.0 3.0 100.0 Mean -5.531;. Std. Error -1.460; Mode 1.000; Std. Deviation -10.223; Kurtosis -30.117; S E Kurtosis -.668 6.4 One Chapter I teacher had as pull-out program, 342 grades 1-6, and the aide worked in an "in-class program" in grade K for reading and math in grade 2.

PAGE 360

6.5 At the middle school level, the teachers provided some 'in-classroom' instruction during the content area classes. Table K-18 Number of Chapter I Teachers with "InClassroom" Program Design Number of Teachers Frequency 0 33 1 13 3 1 4 1 15 1 61 1 Total 51 Percent 64.7 25.3 2.0 2.0 2.0 2.0 100.0 Mean -1.920; Mode .000; Kurtosis -43.659; Std. Error -1.245; Std. Deviation 8.806 Kurtosis -43.659; S E Kurtosis -.662. Note: Seventeen districts or Chapter I service providers utilize in-class programs totaling 96 teachers from all districts. 343 6.6 "Block instruction" involved the classroom'teacher and two Chapter I teachers providing instruction during reading time; students rotated from teacher to teacher. Chapter I services were provided in a physical setting outside of the regular classroom.

PAGE 361

Caucasian Population of Chapter I Pupils Reflects Population of the District 344 Question 13. Does the Caucasian population of your Chapter I pupils reflect the population of the district in number? Results in Table K-19 show that equal numbers of coordinators with both program designs, in-class and pull-out Table K-19 Does Caucasian Population in Chapter I Pupils Reflect Population of District Responses Total In-ClassPu11-0ut More than 7.8% 7.8% ( 4) ( 0) ( 4) Same as 66.7% 15.7%51% (34) ( 8) ( 26) Fewer than 23.5% 11.8%11.8% (12) ( 6) ( 6) Missing 2% 2.0% ( 1) ( 0) ( 1) (11.8 percent, or six) report the Caucasian population of Chapter I pupils in the district as fewer in number than the total district population. Conversely, no more than 11.8 percent of coordinators with pull-out or in-class programs report that numbers of minority students were greater than the total district Caucasian population. The majority of coordinators, 51 percent with pull-out and 15.7 percent for in-

PAGE 362

345 class, report that Chapter I student populations are approximately the same in numbers of minority and Caucasian students as the districts that contain the Chapter I programs. Chapter I Program Design Has No Clear Academic or Social Benefits and May Be Detrimental Question 14. The Chapter I program design(s) in my district has no clear academic or social benefits and may, in fact, be detrimental to pupils' progress and adjustment to school. All of the coordinators with the pull-out program design disagreed with the above statement (Table K-20). The majority of teachers in both program designs strongly disagreed that the Chapter I program design in their district had no clear academic or social benefits and could in fact have been detrimental to pupils' progress and adjustment to school. Coordinators found no detrimental effects to pupil progress in reading or math. One in-class teacher (2.0 percent) strongly agreed with the above statement for both reading and math, but the number was not large enough to conclude that there was a difference attributable to program design.

PAGE 363

346 Table K-20 Question 14; ChaRter I Program Has No Clear Academic or Social Benefits and Be Detrimental Strongly Strongly Agree Agree Disagree Disagree Missing Reading 2.0% 7.8% 52.9% 37.3% ( 1) ( 0) ( 4) (27) (19) Math 2.0% 7.8% 17.6% 74.5% ( 1) ( 0) ( 4) ( 9) (38) Both 2.0% 7.8% 33.3% 56.9% ( 1) ( 0) ( 4) (11) (29) In-class Reading 2.0% 2.0% 17.6% ( 1) ( 0) ( 1) ( 9) Math 2.0% 5.9% 3.9% ( 1) ( 0) ( 3) ( 2) Both 2.0% 3.9% 7.8% ( 1) ( 0) ( 2) ( 4) Pull-out Reading 5.9% 35.3% ( 0) ( 0) ( 3) (18) Math 2.0% 13.7% ( 0) ( 0) ( 1) ( 7) Both 3.9% 25.5% ( 0) ( 0) ( 2) (13)

PAGE 364

Chapter I Students are Unconsciously Neglected Question 18. Chapter I students are unconsciously neglected in the regular classroom. 347 As shown in Table K-21, two coordinators with in-class pro-grams agreed with this statement (3.9 percent) and one coordinator with the pull-out program design strongly agreed with the above statement. The majority of coordinators with the pull-out program design disagreed that Chapter I students were unconsciously neglected in the regular classroom, while another 29.4 percent of the pull-out coordinators strongly disagreed. Similarly, 19.6 percent of the in-class coordinators strongly disagreed with unconscious neglect of Chapter I students, while another 5.9 percent of the in-class coordinators disagreed. Coordinators of the in-class programs, comparatively, appeared to feel more strongly about the statement, with a larger percentage strongly disagreeing. However, results show that Chapter I students are not neglected in the regular classroom setting, according to Chapter I coordinators with both Chapter I program designs, in-class and pull-out.

PAGE 365

348 Table K-21 Chapter I Students Unconsciously Neglected in Regular Classroom Strongly Agree Agree Disagree Total 2.0% 3.9% 45.1% ( 1) ( 2) (23) In-class 3.9% 5.9% ( 0) ( 2) ( 3) Pull-out 2.0% 39.2% ( 1) ( 0) (20) Method in Which Chapter I Instruction is Coordinated with Regular Education Program Strongly Disagree 49.0% (25) 19.6% (10) 29.4% (15) Missing ( 0) ( 0) ( 0) Question 20: How, if at all, is Chapter I instruction coordinated with the regular education program? The largest percentage of responses for both program designs, 43.1 percent for pull-out and 19.6 percent for in-class, was that instruction was coordinated through constant communication and coordination, as shown in Table K-22.

PAGE 366

349 Table K-22 Chapter I Instruction Coordinated with Regular Education Program Constant communication/ coordination/formal and informal (between Chapter I and regular teacher, staff meetings; discussions of needs/ concerns/progress of individual students/ student placement in program; progress reports/written reports/ 9-week reports at secondary' parent teacher conferences; staff development activities) Coordinated by central administration Teachers work together/ cooperate (Chapter I /Regular education teachers/special education teachers; develop and reinforce objectives/skills/goals/ activities/homework assignments) Chapter I follows district guidelines (supplements district curriculum; using same objectives; supplement/supports/ expands reading; one district/Houghton Mifflin reading series) Total 62.7% (32) 3.9% ( 2) 43.1% (22) 17.6% ( 9) Response In-Class Pull-Out 19.6% (10) 2.0% ( 1) 7.8% ( 4) 3.9% ( 2) 43.1% (22) 2.0% ( 1) 35.3% (18) 13.7% ( 7)

PAGE 367

350 Table 59 (Continued) Response Total In-Class Pull-Out Use of alternative methods/ 17.6% 3.9% 13.7% styles/materials for ( 9) ( 2) ( 7) instruction (regular education does not work/ for the non-traditional learner; alternative A.V. materials/computers; review of materials before use in regular education program; use of test results; test-taking strategies taught) Schedule building 5.9% 2.0% 3.9% (effort made to rotate ( 3) ( 1) ( 2) classes; Chapter I an elective at high school) Extra planning time/ 2.0% 2.05 Chapter I staff ( 1) ( 0) ( 1) (accomplishes close communication) In-class observation and 7.8% 3.9% 3.9% assistance (for qualified ( 4) ( 2) ( 2) students; teacher works with students in same classroom during reading time/grades 1 to 5, one district) Coordination by staff 3.9% 3.9% (building level/teacher/ ( 2) ( 0) ( 2) principal; cooperation/ development/decision making/participation)

PAGE 368

Chapter I teachers meet with regular education teachers and staff formally and informally as coordinators explained in their responses. 351 A category closely related to constant communication and coordination was teachers working together, Chapter I teachers cooperating with regular education teachers. Pull-out coordi-nators (35.3 percent) stated that teachers work cooperatively together and in-class coordinators, 7.8 percent) reported teachers working together cooperatively to coordinate Chapter I instruction with the regular education program. Two categories with equal percentages of responses were also related, with 13.7 percent of pull-out coordinators and 3.9 percent of in-class coordinators responding in each category. First, coordinators responded that Chapter I follows district guidelines in supplementing district curriculum and in using the same objectives as the regular education program. Secondly, coordinators responded that for Chapter I, use of alternative methods, teaching styles and materials for instruction, were possible through coordination and familiarity with the regular education program. Other responses were considered of interest to the study and are reported in Table K-22.

PAGE 369

Unique Aspects of Chapter I Program Design Question 21. Would you please describe the aspects of 352 the Chapter I program design(s) in your district that may be unique to Chapter I programs? Are there any aspects that may affect the results of this study? Please use the space below and/or write on the back. There were no aspects of program design stated by coordinators that related to research questions of this study. However, there were unique aspects of program design as described by coordinators that were of interest to the study. No important differences by program design, in-class as compared to pull-out design, were noted. All responses are recorded in Table K-23. Ways in Which Chapter I Program Designs Affect Self-Concept of Students Question 22. This question is very important as student self-concept is a key issue. Could you please describe how the Chapter I program design(s) in your district, in your opLnLon, affects the self-concept of Chapter I students? There were five categories of response where larger percentages of pull-out coordinators than in-class coordinators identified effects of the program design on student self-concept. The largest percentage of coordinators, 37.3 percent pull-out and 7.8 percent in-class coordinators, stated that student self-concept was improved due to Chapter I programs being success oriented. The success experienced by students improved self-concept, and since students feel

PAGE 370

Table K-23 Describe Aspects of Chapter I Program Designs Unique to Your Area Response 353 Total In-Class Pull-Out Individualization (complete/ 1 to 3 students; 4 to 5 students 30 minutes per day; Junior H.S. 6 to 7 students 5 times per week/ 45 minutes) Program based on developmental age/physical, social, emotional (as opposed to chronological age; pre-kindergarten students) Extended day kindergarten (2 1/2 hours extra time and lunch; to compensate for developmental delays; preschool program at a BOCES) Parent involvement (parents a part of inservice/monthly with special education teachers; regarding planning and implementation; parent volunteers at H.S. rather than direct reading help; parent permission for student placement in program) 5.9% ( 3) 2.0% ( 1) 3.9% ( 2) 2.0% ( ) 2.0% ( 1) 2.0% ( 1) 2.0% ( 1) ( 0) 3.9% ( 2) ( 0) 2.0% ( 1) 2.0% ( 1)

PAGE 371

Table K-23 (Continued) Improved methods for reading instruction (reading aloud much of the time; novels used as approach to reading; good literature; speed reading) Key issue/reading and math well coordinated with regular education program (language experience approach/ bring hobbies/read and write about them) Grouping/homogeneous populations in Chapter I, regular classrooms (for pullout) Problems with team teaching/in-class design Total program based on needs assessment/ continual adjustment to needs of students Information given to parents by teachers/awards (home visits to explain Chapter I program/to ease minds about test results/to explain what Chapter I can do; in primary grades high flyer awards given; parent per= mission for career shadowing program at intermediate level; student of the week Total 7.8% ( 4) 5.9% ( 3) 2.0% ( 1) 2.0% (1) 2.0% ( 1) 3.9% ( 2) 354 Response In-Class Pull-Out 2.0% ( 1) 2.0% ( 1) ( 0) ( 0) 2.0% ( 1) ( 0) 5.9% ( 3) ( 0) 2.0% ( 1) 2.0% ( 1) ( 0) 3.9% ( 2)

PAGE 372

355 Table K-23 (Continued) Response Total In-Class Pull-Out Small rural schools great distances apart present Improved public relations (students as ambassadors of good will; invite a friend program/friends into class Fridays on a rotating basis; end of year ice cream social) 2.0% ( 1) 2.0% ( 1) ( 0) ( 0) 2.0% ( 0) 2.0% ( 1) special, they progress. Further responses in this category are summarized in Table K-24. Pull-out coordinators (11.8 percent) and one in-class coordinator (2.0 percent) reported that students choose to be in Chapter I. Therefore, the Chapter I program was a privilege. Another 11.8 percent of the pull-out coordinators, or six, responded that the special attention given to student needs and the individualized instruction and reinforcement to students in Chapter I improved their self-concept. In addition to special attention, outstanding staff members were viewed as caring, supportive, sensitive teachers able to inspire students and "make kids feel special." Pull-out coordinators (9.8 percent) and in-class coordinators (5.9 percent) recognized the ability of outstanding Chapter I staff members to positively

PAGE 373

356 affect the self-concept of those Chapter I students privileged to receive services (see Table K-24). Table K-23 Way in Which Chapter I Program Designs Affect Self-Concept of Chapter I Students Response Total In-Class Pull-Out Small class size, low pupil-teacher ratio Success oriented (success improves selfconcept; success oriented activities; students feel special, progress; more success in regular class results; results in development of oral language and reading skills; grades not given) Special attention to student needs (individualized instruction; reinforcement) Outstanding staff members (Caring, supportive sensitive teachers; build, create, inspire; make kids feel special; part-time counselor) Students want/choose to be in Chapter I; a privilege; students choose to read for pleasure/choose materials; computers in Chapter I only. 7.8% ( 4) 45.1% (23) 11.8% (6) 15.7% ( 8) 13.7% (7) 2.0% ( 1) 7.8% ( 4) ( ) 5.9% ( 3) 2.0% ( 1) 5.95 ( 3) 37.3% (17) 11.8% ( 6) 9.8% ( 5) 11.8% ( 6)

PAGE 374

Table K-24 (Continued) Total Positive parent support/ (close contact with home; parents attend sessions with students) Students love/feel good about Chapter I program/ teacher Improves self-concept/ self-confidence/ emotional development very much; through praise, success; through improved instructional methods/materials; coordination with other classes; transfer of knowledge from Chapter I class to other classes) Pull-out hurts self-concept (Kids hate it; 50% at Junior H.S. embarrassed to come at first; Decreased dropout risk/ improved attendance Stigma for special help of of any kind (few students not progressing due to Chapter I as remedial program; identification as poor reader) Work not piled on for classes missed (students not penalized; positive attitude about Chapter I results) Response In-Class Pull-Out 7.8% ( 4) 9.8% ( 5) 49% (25) 9.8% ( 5) 2.0% ( 1) 2.0% ( 1) 2.0% ( 1) 17.6% ( 9) 2.0% ( 1) 3.9% ( 2) 3.9% ( 2) ( 0) 5.9% ( 3) 7.8% ( 4) 31.4% (16) 7.8% ( 4) ( 1) ( 1) 2.0% ( 1) 357

PAGE 375

358 More teachers from both programs expected the Chapter I students to perform about as well academically as the regular education students at the same grade level than did those who expected a poorer performance. Pull-out teachers (48.1 percent) and in-class teachers (37.5 percent) expected the students to perform about the same. These percentages were higher than would be expected, considering that the students were remedial populations with NCE scores usually in the lowest quartile of the total building population. Results indicate that these teachers had high expectations for the Chapter I students in the sample for this study. In comparison, pull-out teachers (40.7 percent) and in-class teachers (31.3 percent) expected the Chapter I students to perform less well than the regular education students. Therefore, fewer teachers expected Chapter I students to perform less well academically, and more of these teachers were regular classroom teachers than were Chapter I teachers for both program designs. Chapter I students were not unconsciously neglected in the regular classroom setting in either program design. Teachers and coordinators either disagreed or strongly disagreed with the criticism of unconscious neglect of Chapter I students. Students were not viewed as being neglected in the regular classroom since they were considered to be the singular responsibility of the Chapter I teacher.

PAGE 376

359 Students in both program designs in the student sample of this study were found to have made approximately equal academic gains. Teachers and coordinators of the pull-out program agree that students made academic gains in both program designs. A question was asked of coordinators to describe how Chapter I instruction was coordinated with the regular education program. The largest percentage of responses for both program designs, 43.1 percent for pull-out and 19.6 percent for in-class, was that instruction was coordinated through constant communication and coordination. Chapter I teachers meet with regular education teachers and staff, formally and informally, as coordinators explained in their responses. A category of responses by coordinators closely related to constant communication and coordination was that of teachers working together, Chapter I teachers cooperating with regular education teachers. Pull-out coordinators (35.3 percent) stated that teachers work cooperatively together, and in-class coordinators (7.8 percent) reported teachers working cooperatively to coordinate Chapter I instruction with the regular education program. The second largest category of responses by coordinators was the improved self-concept of students. Pull-out coordinators (31.4 percent) and in-class coordinators (17.6 percent) responded that Chapter I programs improved the self-concept and

PAGE 377

self-confidence of students and helped them to develop emotionally. 360