OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. .. A SUMMATIVE EVALUATION OF A JUNIOR HIGH SCHOOL COMMUNICATION ARTS PROGRAM BY Leona Mae Wiest Barnhart A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Secondary Education and Curriculum 1979 <9 Copyright by LEOEUX DMAE WIEST BARNHART 1979 ABSTRACT A SUMMATIVE EVALUATION OF A JUNIOR HIGH COMMUNICATION ARTS PROGRAM BY Leona Mae Wiest Barnhart Purpose The purpose of this study was to do a summative evalua- tion of the Communication Arts Pilot (CAP) program at Tra- verse City (Michigan) Junior High School. Specifically, the purpose was to determine the effect of the program on the reading and associated abilities of students who were in grades seven to nine, 1972-78. The CAP program was intro— duced by the Communication Arts Department of the Traverse City Junior High School. Answers were sought to these questions: 1. Was there a difference between the effect of a new program (CAP) and the program it replaced, as measured by selected standardized tests? Four reading and associated ability areas were analyzed: reading comprehension, vocabu- lary, spelling, and language. 2. Was the effect of CAP the same for all groups who participated in the program? 3. Did the program have the same effect on students of all reading abilities? Leona Mae Wiest Barnhart CAP was described as a mastery learning type program 'with two major components: reading and writing. Students are pre- and posttested on reading and writing skills. Procedure Samples for the study were selected from the sixth grade classes of 1970, 1971, 1972, 1973, and 1974. Each sample became one group in the study, except Sample 2 (1971) which became Groups 2 and 6. Assignment to CAP or non-CAP for Groups 2 and 6 was random. The other groups were CAP (Groups 2, 3, 4, and 5) or non-CAP (Group 1) dependent only on the year the student began seventh grade English. Tests administered at five different grade levels were used in the study: Stanford Achievement Intermediate II (sixth grade), Gates-MacGinite Reading Test (seventh and ninth grades), and National Educational Developmental Test (tenth grade). The large design framework of the study was similar to Design 15, the Recurrent Institutional Cycle Design (Camp- bell, 1963, p. 57-61). Two factors allowed for the use of stronger designs when appropriate with their more powerful statistics: 1. Random assignment of Groups 2 and 6, treatment and control. ‘ Leona Mae Wiest Barnhart 2. Naturally occurring X. Campbell and Stanley (1963) Designs 4, 6, and 10 were used. Statistics used, where appropriate, were the t-test, analysis of variance, and analysis of covariance. Regres- sion lines were used to indicate trends. Conclusions Research Question 1: There was no statistically signi- ficance, as measured by standardized tests, between the ef- fect of CAP on the reading and associated abilities of stu- dents and the effect had by the program it replaced. Research Question 2: There was a statistically signi- ficant difference between the effect of CAP, as measured by standardized test, on the different groups who participated in the program. CAP appeared to be the most effective with Group 3, the second group to participate in CAP and the first to have CAP for two years. Research Question 3: CAP did not have a differential effect for students of varying abilities. The regression lines were positive. The trend was linear. Summary Whether CAP should be continued, discontinued, or modi- fied would depend on the goals of the school district. CAP is doing as well as the program it replaced, as measured by Leona Mae Wiest Barnhart standardized tests. It does appear to be influencing a de- cline in the effectiveness of reading comprehension, as mea- sured by standardized tests. CAP does not appear to be in- fluential in changing the relative performance level of students. DEDICATION To the wellspring of my joy; Tom, my husband, and our children, Tom, Terry, and LeAnne ii ACKNOWLEDGMENTS Sincere appreciation and gratitude are extended to many who have provided encouragement and support, not only during this study, but throughout my academic career. Among those acknowledged are: Dr. Peggy M. Riethmiller, committee chair and major professor, for the kindness, encouragement, guidance, and assistance provided throughout my doctoral pro- gram. Dr. Walter W. Scott, committee member, for his un- selfish gift of time and self. Dr. Samuel S. Corl and Dr. Daniel Jacobson, committee members, for their careful and percentive guidance. Dr. Charles A. Blackman for serving as a substitute at the defense of the dissertation. Dr. Barbara Aswad, Anthropology Department, Wayne State University, for encouraging me to pursue grad- uate study. Iota State, Delta Kappa Gamma Society International for the award of a scholarship to me in 1978 and 1979. ‘WOmen of Wayne, Wayne State University, for encour- aging women to return to school and for granting scholarships to me in 1967 and 1969. Traverse City Area Public School District for grant- ,ing permission to conduct this research. A special -thank you to Dr. Vernon L. Oxender, superintendent, It. Edward Johnston, director of personnel and curri- anlum, Maxine MacInnis, Dr. Richard Ayling, Paul Needham, Shirley Forton, Ella Moore, and Beverly Thornburg . My family and friends who have provided love and encouragement . Mary A. Ryan, my mother, for her constant faith, love and guidance. Rochus Wiest, my father, for iii his love and belief in me. William J. Ryan for his love and cheerful support. Mae MacNeill Barnhart and Joseph D. Barnhart, parents-in-law, for their love and support. Thomas M. Barnhart, my husband, who is without equal. Thomas M. Barnhart II, Terence M. Barnhart, and LeAnne J. Barnhart, our children, who are a job to my heart and a continual source of encour- agement. iv TABLE OF CONTENTS LiSt Of Tables 0 o o o o o o o o o o o o o o o o o o 0 ix LiSt Of Figures 0 o o o o o o o o o o o o o o o o o o X LiSt Of Graphs o o o o o o o o o o o o o o o o o o o o Xii CHAPTER I: THE PROBLEM Background . . . . . . . . . . . . . . l The Communication Arts Pilot (CAP) Program . . . 3 Purpose of the Study . . . . . . . . . . . . . . 7 Importance of the Study . . . . . . . . . . . . . 7 Assumptions of the Study . . . . . . . . . . . . 8 Limitations of the Study . . . . . . . . . . . . 9 Definitions of Terms . . . . . . . . . . . . . 10 Frequently Used Abbreviations . . . . . . . . . . 11 Overview of the Study . . . . . . . . . . . . . . 11 CHAPTER II: REVIEW OF THE LITERATURE Introduction . . . . . . . . . . . . . . . . . . 12 Need and Value of Educational Evaluations . . . . 12 Summary . . . . . . . . . . . . . . . . . . 15 Summative Evaluation . . . . . . . . . . . . . . 15 Summary . . . . . . . . . . . . . . . . . . . . 20 CHAPTER III: DESIGN OF THE STUDY Introduction . . . . . . . . . . . . . . . . . . 22 Restatement of Purpose and Research Questions . . 22 Summative Evaluation . . . . . . . . . . . . . . 23 Description of Communication Arts Pilot (CAP) . . 25 Population 0 O O O O O O C O O O O O I O O O O O 26 Samples . . . . . . . . . . . . . . . . . . . . . 28 Test Instruments . . . . . . . . . . . . . . . . 30 Data Needed . . . . . . . . . . . . . . . . . . . 34 Data Collection . . . . . . . . . . . . . . . . . 34 Description of Communication Arts Pilot (CAP) . . . . . . . . . . . . . . 34 Other Data . . . . . . . . . . . . . . . . . 35 Tests . . . . . . . . . . . . . . . . . . . 36 Design and Analysis . . . . . . . . . . . . . . . 37 Where This Study Expands Design 15 . . . . . . . 47 Hypotheses . . . . . . . . . . . . . . . . . . . 48 Research Question 1 . . . . . . . . . . 48 Research Question 2 . . . . . . . . . . . 51 Research Question 3 . . . . . . . . . . . 52 sumary O O I O O O O O O O O O O O O O O O O 53 CHAPTER IV: RESEARCH FINDINGS Introduction . . . . . . . . . . . . . . . . . . 56 Research Question 1: Hypotheses and Analysis . . 58 Null Hypothesis 1.1 . . . . . . 59 Alternate Hypothesis 1.1 . . . . . . . . 59 Null Hypothesis 1.2 . . . . . . . . . . . 60 Alternate Hypothesis 1.2 . . . . . . . . . . 61 Null Hypothesis 1.3 . . . . . . . . . . . 62 Alternate Hypothesis 1.3 . . . . . . . . . 63 Null Hypothesis 1.4 . . . . . . . . . . . . 64 Alternate Hypothesis 1.4 . . . . . . . . 65 Null Hypothesis 1.5 . . . . . . . . . . 66 Alternate Hypothesis 1.5 . . . . . . . . . . 66 Null Hypothesis 1.6 . . . . . . . . . . . . 67 Alternate Hypothesis 1.6 . . . . . . . . . 68 Null Hypothesis 1.7 . . . . . . . . . . . . 69 Alternate Hypothesis 1.7 . . . . . . . . . . 69 Null Hypothesis 1.8 . . . . . . . . . . . . 74 Alternate Hypothesis 1.8 . . . . . . . . . . 75 Null Hypothesis 1.9 . . . . . . . . . . . . 80 Alternate Hypothesis 1.9 . . . . . . . . . . 80 Null Hypothesis 1.10 . . . . . . . . . . . . 85 Alternate Hypothesis 1.10 . . . . . . . . . 86 Null Hypothesis 1.11 . . . . . . . . . . . . 87 Alternate Hypothesis 1.11 . . . . . . . . . 87 Null Hypothesis 1.12 . . . . . . . . . . . . 90 Alternate Hypothesis 1.12 . . . . . . . . . 90 Null Hypothesis 1.13 . . . . . . . . . . 92 Alternate Hypothesis 1.13 . . . . . . . 92 Null Hypothesis 1.14 . . . . . . . . . . . . 95 Alternate Hypothesis 1.1 . . . . . . . . . 96 Research Question 2: Hypotheses and Analysis . . 101 Null Hypothesis 2.1 . . . . . . . . . . . 102 Alternate Hypothesis 2.1 . . . . . . . . . . 102 Null Hypothesis 2.2 . . . . . . . . . . . . 103 Alternate Hypothesis 2.2 . . . . . . . . . . 103 Null Hypothesis 2.3 . . . . . . . . . . . . 104 Alternate Hypothesis 2.3 . . . . . . . . . . 104 Null Hypothesis 2.4 . . . . . . . . . . . . 106 Alternate Hypothesis 2.4 . . . . . . . . . . 106 Null Hypothesis 2.5 . . . . . . . . . . . . 107 Alternate Hypothesis 2.5 . . . . . . . . . . 107 Null Hypothesis 2.6 . . . . . . . . . . . . 108 Alternate Hypothesis 2.6 . . . . . . . . . . 108 Null Hypothesis 2.7 . . . . . . . . . . . . 109 Alternate Hypothesis 2.7 . . . . . . . . . . 109 Null Hypothesis 2.8 . . . . . . . . . . . . 110 vi Research Question 3: CHAPTER'V: Alternate Hypothesis Null Hypothesis 2.9 Alternate Hypothesis Null Hypothesis 2.10 Alternate Hypothesis Null Hypothesis 2.11 Alternate Hypothesis Null Hypothesis 2.12 Alternate Hypothesis Null Hypothesis 2.13 Alternate Hypothesis Null Hypothesis 2.14 Alternate Hypothesis Null Hypothesis 2.15 Alternate Hypothesis Null Hypothesis 2.16 Alternate Hypothesis Null Hypothesis 2.17 . Alternate Hypothesis 2 Null Hypothesis 2.18 Alternate Hypothesis 2 Null Hypothesis 2.19 Alternate Hypothesis Null Hypothesis 2.20 Alternate Hypothesis Null Hypothesis 2.21 Alternate Hypothesis Null Hypothesis 2.22 Alternate Hypothesis Null Hypothesis 2.23 Alternate Hypothesis Null Hypothesis 2.24 Alternate Hypothesis Null Hypothesis 2.25 Alternate Hypothesis Null Hypothesis 2.26 Alternate HypothesiS‘ Null Hypothesis 2.27 Alternate Hypothesis Null Hypothesis 3.1» Alternate Hypothesis 3.1 Null Hypothesis 3.2 N o o o Ho H0 \00 H No No No No No No N. o o H0 P4: No No No No N N .5 2.25 2.26 2.27 Hypotheses and Ana (I) Ho H. U) N0 N0 N0 N. N O .5 U! NH (a) 0 Alternate Hypothesis 3.2 . Summary . . . . . . . . . Research Question 1 and Research Research Question 2 and Research Research Question 3 and Research CONCLUSIONS AND RECOMMENDATIONS Introduction . . . . . . vii y i O O O C H. O O O O O O O O O O O O O O O O O C O O O C O O O O O O o o o I U) o o o o o o o a U) o o 0 Findings Findings Findings 110 111 111 113 114 116 116 118 119 120 121 123 123 125 126 127 128 129 129 130 130 131 132 132 133 134 134 135 135 136 136 138 139 139 140 141 142 142 143 143 149 149 150 150 158 158 159 160 161 Conclusions . . . . Research Question 1 Research Question 2 Research Question 3 Implications for Traverse City Public Implications for Public Schools in General Recommendations for Further Research Summary . . . . . . Appendix A . . . . . . Appendix B . . . . . . . Appendix C . . . . . . . Appendix D . . . . . . . Appendix E . . . . . . . Bibliography . . . . . . viii Schools 0 O O O 162 162 165 177 177 180 181 182 183 184 191 206 210 215 LIST OF TABLES Groups (CAP-nonCAP): Diagram of Treatment and Observations (Tests) of the Study . . . . . Comparability of Designs . . . . . . . . . . . . Groups (CAP-nonCAP), Tests, and When Administered Groups (CAP-nonCAP), Tests, and When Administered ix 46 51 54 58 LIST OF FIGURES The Relationship Between Evaluation Activities and the Stages of a Developing Program . . . . . . . 24 Untitled . . . . . . . . . . . . . . . . . . . . . . 77 Untitled . . . . . . . . . . . . . . . . . . . . . . 78 Untitled . . . . . . . . . . . . . . . . . . . . . . 82 Untitled . . . . . . . . . . . . . . . . . . . . . . 83 Untitled . . . . . . . . . . . . . . . . . . . . . . 89 Untitled . . . . . . . . . . . . . . . . . . . . . . 93 Untitled . . . . . . . . . . . . . . . . . . . . . . 95 Untitled . . . . . . . . . . . . . . . . . . . . . . 112 Untitled . . . . . . . . . . . . . . . . . . . . . . 113 Untitled . . . . . . . . . . . . . . . . . . . . . . 115 Untitled . . . . . . . . . . . . . . . . . . . . . . 115 Untitled . . . . . . . . . . . . . . . . . . . . . . 117 Untitled . . . . . . . . . . . . . . . . . . . . . . 118 Untitled . . . . . . . . . . . . . . . . . . . . . . 119 Untitled . . . . . . . . . .-. . . . . . . . . . . . 120 Untitled . . . . . . . . . . . . . . . . . . . . . . 122 Untitled . . . . . . . . . . . . . . . . . . . . . . 122 Untitled . . . . . . . . . . . . . . . . . . . . . . 124 Untitled . . . . . . . . . . . . . . . . . . . . . . 125 Untitled . . . . . . . . . . . . . . . . . . . . . . 127 Untitled . . . . . . . . . . . . . . . . . . . . . . 131 Untitled . . . . . . . . . . . . . . . . . . . . . . 138 Untitled......................141 xi LIST OF GRAPHS xii 70 71 72 73 97 98 99 100 144 145 146 147 148 152 153 CHAPTER I THE PROBLEM Background Questions about the effectiveness of educational pro- grams have been frequent in education. Rudolph Flesch brought to the fore in the mid-1950's the question of ef- fectiveness of programs on the reading ability of students (Why Johnny Can't Read, 1955). The questioning of educational programs continued when, in 1957, Sputnik appeared. The American public was shocked. The federal government provided funds to change the chal- lenged curriculum. Changes were brought about, primarily in the mathematics and the sciences (Goodlad, 1966, p. 11). In 1964 the Scholastic Aptitude Test (SAT) and the American College Testing Assessment (ACT) scores, which had been increasing since the early 1940's, began to decline ("Declines in Standardized Test Scores among Secondary School Students," 1976, pp. 4-5). Again, questions were raised about students and programs. Were the high school students of the 1960's and 1970's taking the SAT and ACT tests less able than the students of the '40's and '50's who had taken them? If the students were not less able, then were the school programs to be considered less effec- tive in areas represented on the examinations? More books decrying the schools appeared: How Children Fail by John Holt, 1964; Death at an Early Age by Jonathan Kozol, and 36 Children by Herbert Kohn, both in 1967. The nation's children were not receiving the education parents expected. The public concern about the problem of reading abili- ty of students had grown since Flesch's 1955 publication to the point that Richard M. Petre was to write in the Decem- ber, 1972, volume of Journal of Reading: Never before have the mass media reported so many confliciting stories about how "well" or how "poorly" Americans read. According to these reports, Americans appear to believe that they can hardly read, that large numbers of people are il- literate, that reading failures have reached epi- demic proportions, and that school systems neglect reading to such an extent that a national crisis exists (p. 200). Reading as a "subject" was being analyzed by many with- in the educational community. Publishers, reading special- ists, and teaching staffs were three groups who attempted to delineate reading skills. The skills identified by pub- lishers often varied. (For example, see the teacher's edi- tion cf More Sounds and Syllables: The Ginn Enrichment Program by Theodore Clymer, Thomas C. Barrett, and Lori E. Burmeister, 1969, pp. 3, 8; and the teacher's edition of Discoverinngreasure, 1973, pp. 320-325.) Objectives for identified reading skills were written at all levels of education: federal, state, and local (Gussion, 1979, pp. 1-2). They were frequently placed in a sequential order (Strang, 1967, pp. 131-144). English programs at the secondary level traditionally had been literature-oriented. English teachers had been trained in literature and composition, and English programs reflected that training (Dunning, 1975, p. 9). Reading as a "subject" was not taught. The Communication Arts Pilot (CAP) Program The junior high school communication arts department of a local school district which shifted the format of its program is the focus of this study. Its program was shifted from a traditional format to one emphasizing the teaching of reading and writing skills. The "traditional" junior high school program may be described as one using a classroom grammar text, a class- room literature text, heterogenous groupings, and vague ob- jectives. The "new" junior high school program may be de- scribed as one using no classroom text, heterogenous skill groupings, and specific objectives. The junior highschool is a part of the Traverse City Public School District, Traverse City, Michigan. The 1978- 79 district enrollment was 9,516 students. There are one senior high, one junior high, and fourteen elementary schools in the district. The junior high school contains grades seven to nine. The 1978-79 enrollment was 2,202 students. Each grade level had approximately 700 plus students. The junior high department chairperson described the Communication Arts Pilot (CAP) program asaimastery learning type program (Needham, 1979). It has beenauzvarious stages of implementation since 1972. In that year the program was introduced and conducted at the seventh grade level by a team of two classroom teachers and one reading teacher. CAP was expanded to the whole seventh grade communication arts program in 1973-74. In 1974 it was expanded to the eighth grade. In 1976 the Ninth Grade Project was started. This was to be a transitional program between CAP and APEX (Ap- propriate Placement for Excellence), the high school com- munication arts program. CAP, 1979, is still considered a pilot program. The program has been funded at the local level through Board of Education action. The department chairperson of the Traverse City Junior High School Communication Arts Department stated two rea- sons for the development of CAP: to provide data for an- swering the public concern over declining reading scores and to stabilize or improve student reading achievement (Needham, 1979). A second explanation for the develOpment of CAP was presented in a 1974 Master's thesis. In September, 1972, it was determined that students in the seventh grade com- munication arts classes ranged in reading ability from grades two to twelve, as measured by the Gates-MacGinite Reading Test. Two teachers expressed concern that they were not effectively teaching all the students within the reading ability range. The program deve10ped from that concern (Needham, 1974). The rationale for development of CAP provided in a de- partment memo distributed to parents and other interested persons, "CAP--Communication Arts Pilot" (undatedn is simi- lar to the explanation reported in the thesis. Also provided in the memo was a statement of the goal of CAP: To individualize instruction by: 1. Using pretesting to place students in temporary instructional skill groups, focusing on a particular writing or reading skill within a hierarchy of objectives. 2. Utilizing as many materials and approach- es as possible to teach the skill while responding to style of learning and in- terests. 3. Using post testing (frame sheets) to de- termine mastery (90%) and placement in the next skill group of the writing or reading design. Two tests are systematically administered to junior high communication arts students: the Gates-MacGinite Reading Test and the Croft Reading Comprehension Test. Ad- ministration of the tests was conducted by the communica- tion arts staff. The Gates-MacGinite Reading Test has been administered to all students in the seventh and ninth grade English classes since 1870. The seventh grade scores are used by the junior high counselors and the seventh and eighth grade communication arts staff for placement and grouping of students. The ninth grade scores are used by the ninth grade counselors and the senior high counselors in the de-_ velopment of a student's high school program. The ninth grade communication arts staff uses the ninth grade test scores for placement and grouping of students. The Croft Reading Comprehension Test is administered in the seventh grade. It is used for student placement in- to skill groups and as a pre-post test. The CAP program has two components, reading and writ- ing. The reading skills component was derived from the Croft Reading Comprehension Skills material. The CAP read- ing objectives are Croft reading objectives. The writing component of the program, on the other hand, is and has been teacher-developed. Students are pre—post tested on specific skills in reading and writing. Instruction is with large groups (class size is usually thirty-three). Students who do not pass the posttest for a skill may go through a skill cycle again. Students may go through a skill cycle up to three times in attempting to reach mastery. The number of oppor- tunities varies depending on the availability of staff, number of students needing the skill cycle, and the time of year. A skill cycle has six steps: 1. Performance objectives specified 2. Criterion test administered 3. "Teach" step implemented (attention, presen- tation, reinforcement, practice) - explain 4. Posttest 5. Application - give a sheet explaining kinds - length, extent 6. Recycle or on to the next ("Update on CAP Program," 1978) Purpose of the Study The purpose of this study was to do a summative eval- uation of the Communication Arts Pilot (CAP) program at Traverse City (Michigan) Junior High School. Specifically, the purpose was to determine the effect of the program on the reading and associated abilities of students who were in grades seven to nine, 1972-78. The CAP program was in- troduced by the Communication Arts Department of Traverse City Junior High School. The major research questions explored in this study were: 1. Is there a difference, as measured by standardized tests, between the reading and associated abilities of stu- dents who participated in CAP and students who did not? 2. Is the effect of CAP upon different student groups consistent over time, as measured by standardized tests? 3. Is there a difference in the effect of CAP for students of varying reading abilities? Importance of the Study Ralph Tyler states that evaluation is an important means of communicating school success to the public: Increasingly, we must expect to use evalua- tion procedures to determine what changes are actually taking place in students and where we are achieving our curriculum objectives and where we must make still further modifications in order to get an effective educational program (1949, p. 125). ' The Traverse City Public Schools implemented21communi- cation arts program in the junior high school in 1972. Money has been spent, human resources used, and for seven years students have participated in the program. What changes have occurred in the students? Malcolm Provus (1973) has stated that the "purpose of program evaluation is to determine whether to improve, maintain, or terminate a program" (p. 172). This study will provide data to help make that decision about the Communication Arts Pilot (CAP) program. In addition, the data and analysis presented in this study may be useful in three other areas of inquiry: l. Mastery learning as an effective method for improving reading achievement at the junior high school level 2. General studies of junior high school stu- dents' reading achievement 3. Impact of educational innovations on student outcomes Assumptions of the Study l. The tests used in the study are part of the Tra- verse City Public School's testing program. They are stan- dardized tests used to measure student achievement, reading achievement, and general aptitude. The school district is familiar with the tests. They have been used since prior to 1971. It is assumed that school district personnel find the tests acceptable and the results reliable. 2. The tests used in the study were administered by the Traverse City Public School staff. It is assumed the tests were administered in a consistent manner and accord- ing to the test developer's instructions. 3. The tests were scored by outside agencies, either the testing companies or the Intermediate School District. It is assumed the tests were accurately scored. 4. The test results were provided to the school dis- trict on computer printouts. It is assumed that the print- outs were accurate. 5. The test developers have conducted research to de- termine the content validity of their instruments. It is assumed the research was conducted in an adequate manner. It is also assumed the determination of content validity is accurate. Limitations of the Study 1. The researcher's lack of control over tests used, administration of the tests, scoring, recording and stor- ing of results are all limitations inherent in an ex post fagtg study. The limitations are those any district would encounter that wished to do a program evaluation for which they had not preplanned. 2. A study over time must contend with variables which may have a differential effect on the different 10 groups being studied. Information to help evaluate the im- pact of three such variables are provided: a. Demographic changes--demographic data for the Traverse City area have been provided. b. Differences in educational experiences of the six groups in the study--information about changes in the curriculum and materials of three Traverse City junior high departments--social science, science, and mathematics-—have been provided. c. Impact of history experienced by the six groups in the study--the seventh grade reading test results of the Michigan Educational Assessment Program for the Tra- verse City Public Schools and six other districts of the northwest region of Michigan are provided. It is assumed life experiences, such as mass media, are similar for the seven districts. It can be observed whether the position of Traverse City students relative to the students in the other six districts remained the same. Definition of Terms The following definitions are accepted throughout the study: glass: all the students in a grade level for a parti- cular year. Skill Cycle: a teaching procedure with six steps (ob- jectives stated, pretest, teach, posttest, student 11 application, recycle). Further described on page six of this paper. Summative Evaluation: a non-interfering outcome eval- uation of a stable program to help determine whether a pro- gram should be improved, continued, or discontinued. Frequently Used Abbreviations CAP: Communication Arts Pilot program Gates: Gates-MacGinite Reading Test Stanford or Stan: Stanford Achievement Test Inter- mediate II DAT: Differential Aptitude Test NEDT: National Educational Development Test Overview of the Study Chapter 11 consists of a presentation of the litera- ture which led to decisions made in the study. Chapter III includes the design of the study with a description of the population, the samples, instruments used, the data needed, collection of the data, statistical procedures, and analysis of the data to be used. Chapter IV contains an analysis of the data collected. Chapter V includes a summary, findings, conclusions, implications of the study for Traverse City Junior High School, and recommendations for further research. CHAPTER II REVIEW OF THE LITERATURE Introduction The purpose of this study was to do a summative eval- uation of a program innovation in a local school district. The review of the literature was directed to answering two questions: 1. Is an evaluation of an educational program per- ceived by educators as a valuable and needed activity? 2. When is it appropriate to do a summative evalua- tion? Need and Value of Educational Evaluations Many educators perceive evaluation of educational pro- grams as imperative. The following are quotations from educators who speak to this point. Careful evaluation has not been made of the innovations of the past, nor is it being made today. This failure to assess the effects of innovations against their total outcomes has been perhaps the cause of the fact that in American education curriculum revision proceeds by re- placing one scheme with another and one "approach" with another, not necessarily because objective evidence has demonstrated the merits of the one or the failures of the other, but merely because the new scheme or approach somehow has gained at- tention, is in "fashion" for the time being, or is championed by forceful leaders (Taba, 1962, p. 315). 12 13 There has been agreement, both within and without the ranks of education, that systematic investigation has much to offer. Indeed, there is agreement that massive, lasting changes in education cannot be made except on the basis of deep objective inquiry (Cronback and Suppes, 1969, p. 12). Accurate assessment of educational outcomes is essential for sound planning and effective stimulation of growth in our educational struc- ture. Assessment has always been an integral as- pect of curriculum development and is a major re- sponsibility of curriculum workers. This respon- sibility is especially critical in a time of awakened public concern, massive federal commit- ment and widespread professional reappraisal of our educational endeavors (Combs, 1967, p. v). Evaluation is one of the most widely dis- cussed but little used processes in today's edu- cational systems...despite these tacit trends toward accountability, only a tiny fraction of the educational programs operating at any level have been evaluated in any but the most cursory fashion, if indeed at all. Verbal statements about education and accountability? An abun- dance. Genuine evaluation of educational pro- grams? Unfortunately rare (Worthen and Sanders, 1973, p. 1). Until quite recently school systems have rarely been equipped or even motivated to attempt to evaluate the effects of their learning and teaching activities. There seems to have been an assumption that an innovation is good per se, be- cause it is new and unaccustomed. This idea is further reinforced if the innovation lasts, if it appears not to be doing a poorer job than the practice it replaced, and if it does not, at the same time, disturb too much the other activities which are in progress in the school (Moorish, 1978, p. 154). As public programs have become increasingly massive and expensive, it is not surprising that taxpayers and some of those who authorize expendi- tures of money have begun to ask questions about how much it buys. So, a new enterprise called program evaluation has come on the scene. Even though its record of results is still modest, pro- gram evaluation offers enough promise that it is being mandated and incorporated in the planning 14 for both large and small human-service inter- ventions (Anderson and Ball, 1978, p. 2). The need for and desirability of objective evaluation seems to have been a consistent concern \ Continual interchange of information (\é/ L\\\interchange. \‘<_____/—-/'l /' EVALUATION STAGES* PRE-FORMATIVE EVALUATION Needs Program assessment planning evaluation measures of current status, goal selection, pilot studies 1 FORMATIVE EVALUATION 1 Progress evaluation Implementation evaluation measures of program fidelity, student progress in individual units J, - SUMMATIVE EVALUATION Outcome evaluation Documentation evaluation unobtrusive, non-interfering no feedback to program *At each stage of the evaluation process, the following four phases occur: (1) (2) (3) ascertaining the decision areas of concern collecting and analyzing data in order to report summary information to decision makers 25 Description of Communication Arts Pilot (CAP) The Communication Arts Pilot (CAP) program has beenchr— scribed by the junior high department chairperson as a mas- tery learning type program (Needham, 1979). It has been at various stages of implementation since 1972. That year it was introduced and conducted in the seventh grade by two classroom teachers and one reading teacher. The following year, 1973-74, the whole seventh grade communication arts program was CAP. In 1974 CAP was expanded to the eighth grade. CAP has been funded by Traverse City Board of Educa- tion action. In the 1978-79 school year, CAP was still considered a "pilot" program, according to the department chairperson. A department paper showing the implementation process may be found in the appendix. The goal of CAP ("CAP--Communication Arts Pilot," un- dated) is: To individualize instruction by: 1. Using pretesting to place students in tem- porary instructional skill groups, focusing on a particular writing or reading skill within a hierarchy of objectives. 2. Utilizing as many materials and approaches as possible to teach the skill while re- sponding to style of learning and interests. 3. Using post-testing (frame sheets) to de- termine mastery (90%) and placement in the next skill group of the writing or reading design. 26 The CAP program has two components, reading and writ— ing. The CAP reading objectives are those provided in the Croft Reading Comprehension Skills material (see Appendix). The writing component of the program is and has been teach- er-developed. Students are pre- and posttested on specific reading and writing skills. Instruction is in large groups (class size is usually thirty-three). Students who do not pass a posttest for a skill may go through a skill cycle up to three times in attempting to reach mastery. The number of opportunities varies depending on the availabilitycfifstaff, numbers of students needing the skill cycle, and the time of year. The skill cycle has six steps: 1. Performance objectives specified 2. Criterion test administered . "Teach" step implemented (attention, presen- tation, reinforcement, practice) - explain 4. Posttest 5. Application - give a sheet explaining kinds - length, extent 6. Recycle or on to the next ("Update on CAP Program," 1978) Population The population of the study included three categories of Traverse City Junior High School students: 1. Those who had participated in the Communica- tion Arts Pilot (CAP) program from September, 1972, to June, 1977. 2. Students of seventh grade English teachers who did not participate in CAP in 1972-73, the first year of implementation of CAP. 27 3. Students in the seventh grade English classes of 1971-72, one year prior to any implementa- tion of CAP. Students who had reading improvement in seventh and/or eighth grade of junior high school duringtflmayears spanned by the study were not included in the population. The Read- ing Improvement program wasaiseparate English program prior to and after implementationmxfCAP; therefore, the exclusion does not make the CAP and non-CAP groups different. The population of the study is assumed to be normally distributed. The five sixth-grade classes from which the samples were drawn all have a large N: 1970 N = 644 1971 N = 668 1972 N = 717 1973 N = 714 1974 N = 691 A choice was made not to follow students past grade ten for two reasons. Grade ten is the last grade in the Traverse City Public Schools when a test is administered to all students. Tests are administered in the eleventh) and twelfth grades, Preliminary Scholastic Aptitude Test (PSAT), Scholastic Aptitude Test (SAT), and American Col- lege Testing Assessment (ACTh:but they are taken primarily by students planning to attend college. The second reason for not including grades eleven and twelve is the student drop-out problem. Students must at- tend school until they are sixteen years of age. For most 28 students sixteen years is reached during the tenth grade. At that time, students may drop out of school if they choose. Since this study was of a junior high school pro- gram, as many students as possible who had participated in the program should be included in the population from which the samples were drawn. If grades eleven and twelve were included, high school dropouts would have to be eliminated from the population of the study. For this study they were not eliminated. However, they were excluded from the study if they left school before being administered all of the test instruments being used in this study. Samples Initially five samples were selected from the popula- tion. A systematic sampling procedure was used. The first individual was selected at random using a random number table in Elementary Statistical Methods by Helen M. Walker and Joseph Lev (pp. 372-4). Subsequent individuals were selected at regular intervals. This is a standard proce- dure for selecting a sample from a list where there is no reason to believe that a bias exists in the arrangement of the list. The lists were in alphabetical order, and there was no reason to assume there was bias. The lists from which the sample selections were drawn were the Traverse City Public Schools' computer printouts of the test results of the Stanford Achievement Test Inter- mediate II. The test was administeredtx>all the sixth grade 29 students in the Traverse City School District in 1970, 1971, 1972, 1973, and 1974. 150 individuals were selected from each list. Individuals were eliminated from a sample if they (1) did not have complete test scores for the five tests used in the study, and/or (2) had reading improvement in seventh and/or eighth grade. Sample 1 was selected from the 1970 sixth grade class of Traverse City Public Schools. It is referred to as Group 1. None of the students participated in CAP. It is a control group. N=87. Sample 2 was selected from the 1971 sixth grade class of Traverse City Public Schools. It was further divided into two groups: those students who had had CAP and those who had not. The group which participated in CAP is refer- red to as Group 2. N=43. The group which did not parti- cipate in CAP is referred to as Group 6. N=52. Group 6 is a control group. Sample 3 was selected from the 1972 sixth grade class of Traverse City Public Schools. It is referred to as Group 3. All students participated in CAP. N=87. Sample 4 was selected from the 1973 sixth grade class of Traverse City Public Schools. It is referred to as Group 4. All students participated in CAP. N=87. Sample 5 was selected from the 1974 sixth grade class of Traverse City Public Schools. It is referred to as Group 5. All students participated in CAP. N=87. 30 The groups included in the study were: Group Description Years Status 1 Control group; 1970-74 Graduated 1970 6th grade 1977 6 Control group; 1971-75 Graduated 1971 6th grade 1978 2 1971 6th grade 1971-75 Graduated 7th grade CAP 1978 3 1972 6th grade 1972-76 1978-79 7th & 8th grade 12th grade CAP 4 1973 6th grade 1973-77 1978-79 same as #3 11th grade 5 1974 6th grade 1974-78 1978-79 same as #3 10th grade The N of each group is large: Group 1 N = 87 Group 2 N = 43 Group 3 N = 87 Group 4 N = 87 Group 5 N = 87 Group 6 N = 52 ...if N is reasonably large, the sampling distri- bution will tend to be normal; and this will be true even if the samples come from a decidedly skewed population...second, the mean of the sam- pling distribution of sample means will, in the long run, equal M, the population mean. This se- cond property of the sample mean permits it to be called an unbiased estimate of its parameter, the population mean (Klugh, 1974, p. 204). Test Instruments Test instruments used in this study were: (1) Stan- ford Achievement Test, Intermediate Level II, Forms Y, A, and B; (2) Gates-MacGinite Reading Test: Survey E; (3) . 31 Differential Aptitude Tests, Forms L and S; and (4) the Na- tional Educational Development Tests, Level 2. The four tests are reviewed in the Sixth Mental Mea- surement Yearbook (1965) and the Seventh Mental Measurement Yearbook (1972), edited by Oscar Krisen Euros. The tests have been widely used and accepted by educators. The Stanford Achievement Test, Intermediate Level II, is an achievement test given to students in grades 5.5-6.9. The developers consider it to be a power test. Three forms of the Stanford were used with groups in this study. Groups 1, 2, 6, and 3 used Form Y, 1964 edi- tion. Group 4 used Form A, and Group 5 used Form B of the 1973 edition. The 1964 edition, Form Y, had nine scores: word mean- ing, paragraph meaning, spelling, language, arithmetic com- putation, arithmetic concepts, arithmetic application, so- cial studies, and science. The 1973 edition, Forms A and B, had eleven scores: vocabulary, reading comprehension, word study skills, math- ematics concepts, mathematics computations, mathematics ap- plication, spelling, language; social science, science, and listening comprehension. Two new tests were included in the 1973 edition: word study skills and listening comprehension. A table of equi- valence of forms for corresponding tests of the successive editions has been prepared. It can be found in the Stgg— ford Research Report #5 (1975). 32 Information about major research programs, norms, va- lidity, reliability, and other technical information can be found in the Stanford Achievement Test, Manual Part W.Tech- nical Data Report (1975) and in Stanford Research Reports. The Gates-MacGinite Reading Test: Survey E is a gen- eral reading test for grades 7-9. It was first published in 1939. The test consists of three parts: speed and accuracy, vocabulary, and comprehension. The developers consider the test a power test. Information regarding the selection of items, reliabiL— ity, and other technical data can be found in the Technical Manual, Gates-MacGinite Reading Tests (1965). The Differential Aptitude Test is an aptitude test for grades 8-12 and adults. The first edition was published in 1947, the fourth edition (used with Group 1 of this study) in 1966, and the fifth edition (used with Groups 2, 3, 4, 5, and 6) in 1973. The Differential Aptitude Test has nine scores: ver- bal reasoning, numerical ability, total verbal reasoning and numerical ability, abstract reasoning, clerical speed and accuracy, mechanical reasoning, space relations, spel- ling, and grammar. Forms L (1966) and S (1973) were the forms from eachcflf the editions used in this study. Form 8 of the 1973 edition was largely based on Form L of the 1966 edition. Tables of interform correlations between Forms S and M are given in 33 the Fifth Edition Manual for Differential Aptitude Tests Forms 8 and T (1973), pp. 77-78. Information about norms and profiles, equivalence of forms, validity, reliability, correlation with other tests, and other technical information can be found in the Fifth Edition Manual for Differential Aptitude Tests Forms S and Z (1973). The National Educational Development Test, Level 2, is an achievement battery given to students in grades 9—10. It was first developed in 1959. It is a secured test with two new forms issued annually. The battery has six scores: English usage, mathematics usage, social studies reading, natural sciences reading, word usage, and total. Information about the technical data can be found in the Interpgetative Manual for Grades 9-10 and the Technical Report for each year of the test. The tests described are used in the Traverse City Pub- lic Schools and thus have been used in this study. Follow- ing is a chart with the tests and the months when they are administered: Grade Month . Test Type of Test 6 March Stanford Achieve- Achievement ment Test, Inter- mediate II 7 Sept. Gates-MacGinite Reading Reading Test 8 Oct. Differential Apti- Aptitude tude Test 9 Sept. Gates-MacGinite Reading Reading Test 10 Oct. National Educational Achievement Differential Test 34 Data Needed Data needed for this study were three-fold: 1. Data to describe the CAP program. 2. Data to describe other factors which may have impacted on test results and program effect: a. Demographics. b. Other educational experiences. c. Impact of history experienced by students in the same geographical area. 3. Standardized test results. Data Collection An April, 1979, interview was held with the Acting Di- rector of Personnel and Director of Curriculum and Instruc- tion of the Traverse City Public Schools to secure permis~ sion to do the study and to use test data collected by the school system. Permission was granted. A letter dated April 22, 1979, was provided to the district stating that random sampling would be used in the study and thatstudents would not be identified. Description of Communication Arts Pilot (CAP) Descriptive data of the CAP program was acquired through interviews with three primary informants: the junior high school communication arts chairperson (an ori- ginal teacher participant in CAP), the seventh and eighth grade reading teachers who had been involved with CAP from 35 the beginning, and through documents provided by these persons. CAP is herein after known as "X." "X" statistically has no meaning except as format for the whole program. Other Data Demographics. Demographic data were acquired from two sources: 1. Population Characteristics: Ten Counties, developed by Nancy C. Hayward, Director, Data Research Center, Traverse City, MI (June, 1979) 2. Traverse City Public Schools' fourth Friday counts for years covered by the study(1970-78) Other Educational Experiences. Information about cur- riculum and materials used in other subject areas at the Traverse City Junior High School was requested to determine changes in educational experiences of the groups studied. Interviews were held with the department chairpeople of the science, social science, and mathematics departments. They described the courses at the junior high school, which courses were required and which elective, and materials used since 1970. The department chairs said there were no cur- ricular changes in their departments since 1969-71. Information regarding materials was also acquired from the curriculum office of the Traverse City Public Schools. Impact of History Experienced fur Students in the Same Geographic Area. Michigan Educational Assessment Pro- gram results, years 9 and 10, were used as an indicator of 36 whether the effect of general life experiences had shown up on test results that all students in the area had taken. It was assumed that position of achievement on the tests would remain similar if there was no outside influence act- ing on the students. Results of the Reading Test adminis- tered to seventh grade students (statewide) were requested for seven school districts: Traverse City, Charlevoix, Manton, Elk Rapids, Sutton's Bay, Petoskey, and Mesick. Traverse City and Charlevoix provided their reports; the rest were requested from the Michigan Department of Educa- tion. The method of reporting the data has varied from year to year. The format has been consistent since year nine of the program. The data for the four years previous to year nine were included in the year nine report. Only data for those years appearing on the year nine and year ten reports were used in this study. To have sought information prior to 1974, which appeared on the year nine report, would have entailed effort beyond what the information would provide. gate. The tests used in the study were all administered by the staff of the Traverse City Public Schools. Scoring was done by the test publishers, except the Gates-MacGinite Reading Test, which was scored by the Traverse Bay Area In- termediate District. Computer printouts of the results were 37 provided to the Traverse City Public Schools by the scoring agents. The test data collected for the study were obtained, when available, from the computer printouts. Stanford. The computer printouts for the Stanford Achievement Test were located at the central office of the Traverse City Public Schools. The samples for the study were selected using the lists of the printouts. A systema— tic sampling procedure was used and is described in the "Sample" section of this chapter. Selection of Sample 3 and collection of the Stanford data for Sample 3were begun on May 4, 1979. 150 subjects were drawn for the sample. Stanford Form Y had been used- It was administered in the spring of 1973 to the sixth grade students of the district. Selection of Sample 3 and the collection of Stanford data for the sample were com- pleted on May 7, 1979. Selection of Sample 2 and collection of the Stanford data for this sample were begun on May 7, 1979. 150 subjects were drawn for the sample. Stanford Form Y had been used. It was administered in the spring of 1972 to the sixth grade students of the district. Selection of Sample 2 and the collection of Stanford data for the sample were completed on May 7, 1979. Selection of Sample 4 and collection of the Stanford data for it were begun May 7, 1979. Again, 150 subjects 38 were drawn for the sample. The Stanford Form A had been used, It was administered in the spring of 1974 to the sixth grade students of the district. Selection of Sample 4 and the collection of Stanford data for it were completed on May 8, 1979. Selection of Sample 5 and the collection of the Stan- ford data for Sample 5 werebegun on May 8, 1979. 150 sub- jects were drawn for the sample. Stanford Form B had been used. It was administered during the spring of 1975 to the sixth grade students of the district. Selection of Sample 5 and the collection of Stanford data for it were completed on May 8, 1979. Selection of Sample 1 and the collection of the Stan- ford data for Sample 1 were begun on May 8, 1979. Stanford Form Y had been used. It was administered Spring, 1971, to the sixth grade students of the district. 150 subjects were drawn for the sample. Selection of Sample 1 and the collec- tion of Stanford data for it were completed on May 16, 1979. Grade equivalents and percentiles were provided on the computer printouts and recorded for the study. The data provided in Form Y, used with Samples 1, 2 (which includes Groups 2 and 6 of the study), and 3, were for: word mean- ing, paragraph meaning, spelling, language, arithmetic com- putation, arithmetic concepts, arithmetic application, so- cial studies, science, and total test. 39 The data provided in Forms A and B, used with Samples 4 and 5, respectively, were for: vocabulary, reading com- prehension, word study skills, mathematics concepts, mathe- matics computations, mathematics application, spelling,lan- guage, social science, science, listening comprehension, total auditory, total reading, total math, and total test. Manuals for the Stanford Achievement Test,Intermediate Level II were provided by the curriculum office of the Tra- verse City Public Schools. National Educational Development Test. The National Educational Development Test (NEDT) and Differential Apti- tude Test (DAT) data were made available at the Traverse City Senior High in computer printouts, student CA 60's, or student permanent records. Collection of the NEDT data for Sample 3 was begun on May 16, 1979. The test was administered in the fall of 1976 to the tenth grade students of the district. Only the scores for the subjects in the sample were sought. If a subject did not have scores for this test, s/he was dropped from the sample. The collection of NEDT data for Sample 3 was completed on May 16, 1979. Collection of the NEDT data for Sample 4 was begun on May 17, 1979. The test was administered during the fall of 1977 to the tenth grade students of the district. Only the scores for the subjects in the sample were sought. If 40 a subject did not have scores for this test, s/he was drop- ped from the sample. The collection of NEDT data for Sam- ple 4 was completed on May 17, 1979. Collection of the NEDT data for Sample 5 was begun on May 17, 1979. The test was administered in Fall, 1978, to the tenth grade students of the district. Only the scores for the subjects in the sample were sought; consequently, if a subject did not have scores for this test, s/he was dropped from the sample. The collection of NEDT data for Sample 5 was completed on May 24, 1979. The collection of NEDT data for Sample 1 began on May 24, 1979, the test having been administered in the fall of 1974 to tenth grade students in the district. Only the scores of subjects in the sample were sought. If a subject did not have scores for this test, s/he was dropped fromtflue sample. The collection of NEDT data for Sample 1 was com- pleted on May 24, 1979. Collection of NEDT data for Sample 2 was begun on June 4, 1979. The test was administered during the fall of 1975 to tenth grades students of the district. Since only the scores for the subjects in the sample were sought, if a sub- ject did not have scores for this test, s/he was drOpped from the sample. The collection of NEDT data for Sample 2 was completed on June 12, 1979. National and local percentiles and standard scores were provided on the NEDT computer printouts and recorded fortflma 41 study. The data provided were for: composite, English usage, math usage, social studies reading, natural science reading, and word usage. The Interpretative Manual for Grades 9-10, 1970-71 and 1978-79 were provided by the counseling and guidance depart- ment of the school district. Differential Aptitude Test. Collection of the Differ- ential Aptitude Test (DAT) data for Sample 3 was begun on May 5, 1979. The test was Form S and was administered in the fall of 1974 to the eighth grade students of the dis- trict. Only the scores for the subjects in the sample were sought. If a subject did not have scores for this test, s/he was dropped from the sample. The collection of the DAT data for Sample 3 was completed on June 16, 1979. Collection of the DAT data for Samp1e4 was begun on May 17, 1979. The test was Form S and was administered in the fall of 1975 to the eighth grade students of the dis- trict. Since only the scores for the subjectsixmthe sample were sought, if a subject did not have scores for this test, s/he was dropped from the sample. The collection of DAT data for Sample 4 was completed on May 17, 1979. Collection of the DAT data for Sample 5 began on May 25, 1979. The test was Form S and was administered Fall, 1976, to the eighth grade students of the district. Only the scores for the subjects in the sample were sought. If 42 a subject did not have scores for this test, s/he was drop- ped from the sample. The collection of the DAT data for Sample 5 was completed on May 25, 1979. The collection of the DAT data for Sample 2 was begun on May 31, 1979. The test was Form 8 and was administered during the fall of 1973 to the eighth grade students of the district. Only the scores for the subjects in the sample were sought; consequently, if a subject did not have scores for this test, s/he was dropped from the sample. The col- lection of the DAT data for Sample 2 was completed on June 12, 1979. Collection of the DAT data for Sample 1 was begun on June 4, 1979. The test was Form L and was administered in the fall of 1972 to the eighth grade students of the dis- trict. Only the scores for the subjects in the sample were sought. If a subject did not have scores for this test, s/he was dropped from the sample. Collection of the DAT data for Sample 1 was completed on June 12, 1979. Raw scores and percentiles were provided on the compu- ter printouts and/or student record stickers on the perma- nent records and were recorded for the study. The data pro- vided in Form L, used with Sample 1, were for: verbal rea- soning, numerical ability, verbal reasoning plus numerical ability, abstract reasoning, clerical speed and accuracy, mechanical reasoning, space relations, spelling, and gram- mar . 43 The data provided in Form S, used with Samples 2, 3, 4, and 5 were for the same tests as named for Form L above. The Fifth Edition Manual for the Differential Aptitude Tests Forms S and T (1974) was provided by the department of guidance and counseling of the school district. Gates-MacGinite Reading Test. Computer printouts of Gates-MacGinite Reading Test: Survey E, administered to ninth grade students in the district, were available for Samples 2, 3, 4, and 5. One was not available for Sample 1. No computer printouts of the Gates administered to seventh grade students in the district were available. Ad- ministrative sources felt the printouts might have been de- stroyed when the students left the junior high school. Grade equivalent scores of the Gates administered in the seventh grade were available in permanent records for those students who had participated in CAP: Group 2 of Sample 2, Sample 3, Sample 4, and Sample 5. No information was available for the Gates seventh grade test for Sample 1 or Group 6 of Sample 2. Collection of the Gates data, administered in both the seventh and ninth grades, was begun on May 5, 1979, and was completed on June 16, 1979. Standard scores, percentiles, and grade equivalents were provided on the computer printouts for the Gates ad- ministered in the ninth grade and were recorded for the study. The data provided was for: vocabulary, comprehen- sion, and composite. 44 Grade equivalents were available for the Gates admin- istered in the seventh grade and were recorded for the study. The data provided was for vocabulary and comprehen- sion. Design and Analysis The large design/framework of the study is similar t0' Design 15, the Recurrent Institutional Cycle Design: a "Patched-Up" Design (Campbell, 1963): The design is appropriate to those situations in which a given aspect of an institutional pro- cess is, on some cyclical schedule, continually being presented to a new group of respondents. such situations include schools, indoctrination procedures, apprenticeships, etc. If in these situations one is interested in evaluating the effects of such a global and complex X as an in- doctrination program, then the Recurrent Institu- tional Cycle Design probably offers as near an answer as is available from the designs developed thus far (p. 60). This design combines the "longitudinal" and "cross-sectional" approaches commonly employed in developmental research...The cross-sectional study itself confounds maturation with selection and mor- tality. The longitudinal study confounds matura- tion with repeated testing and with history...The combination, perhaps with repeated cross-sectional comparisons at various times, seems ideal (p. 61). ...feature often characteristic of such designs is that the effect of X is demonstrated in several different manners. This is obviously an important feature where each specific comparison would be equivalent by itself (p. 57). 45 The model, as it would appear in most schools: Class A X 0 __-_-_--_l _______________________ Class Bl R 02 X 03 Class B2 R X 04 Class C O X Table 1 shows how the model is applied to this study. "X" represents the CAP program. This study expands Design 15 in two important ways. As shown in Table 1, Group 2 and Group 6 were randomly as- signed to treatment or no treatment. The study was ap- proached as an experimental, quasi-experimental study. De- signs which are used in both categories were used in this study. Random assignment is a necessary factor in true experi- ments: ...If you are told the selection was more or less random, keep questioning to find out what the "less random" part was...you could treat your data as though it came from a true control group design (Fitz-Gibbon, 1978, p. 49). ...randomization is conceived to be a process occurring at a specific time and is the all-purpose procedure for achieving pretreatment equality of groups, within known statistical limits (Campbell, 1963, p. 6). 46 acmecmwmmd Soccmm u Ammov uonm mane COwumowcsEEou u umma ucmEdon>wo.dmcofiumospm HocOAumz u Awpmum sumac a“ pmumumficflEpm. umme mcflpmmm wuwaOomzumwumo u umma wpsufiumm Hmwucmummmwo n Ampmum sucm>mm cw pououmH:aEpmv umma mcflpmmm muficfiuomz|mwumo n m HH mumapmfiumucH ucwEm>mw£o¢ ou0wcmum u 0 mo «0 x no x No Ho m mo eo x no x No Ho e mo «0 x mo x Nno Ho m mo eo no x No x Ho N mo eo mo m Ho e mo «0 Ho H meme puma when meme «has meme «has Heme oemfl macho sense may no Imummec mcowum>uwmno paw ucmEummuB mo Emummwa ”Ameoeocneeoc meeouo .H wanes mmmau woman mmmHO mmcau mmcau mmmao 47 Students were assigned to CAP or non—CAP classes in a random fashion. They were assigned by computer. The prob— lem of elective classes was not present for two reasons: there are very few electives in the seventh grade, and the CAP classes were taught all hours of the day, except the teachers' conference hour. The combination of these fac- tors makesidzvery unlikely a student would not have been able to be assigned to a CAP class. The researcher is sa- tisfied the assignment was random for Groups 2 and 6 of the study. Where This Study Expands Design 15 The other important factor is X, as referred to in this study, CAP. X was not introduced by the researcher. Ordinarily, this would preclude experimental or quasi- experimental designs; but, as Campbell states: Designs 7, 10, 12, 13 (but not 13a) and 14 would be applicable both for naturally occurring X's and for X's deliberately introduced by the experimenter (p. 64). Campbell and Stanley further state: Where we have pretests and where clear-cut determination of who were exposed and who were not is available, then Designs 10 and 14 may be convincing even without the randomization. But for a design lacking a pretest (imitating De- sign 6) to occur naturally requires very special circumstances, which almost never happen. Even so, in keeping with our general emphasis upon the Opportunistic exploitation of those settings which happen to provide interpretable data, one should keep his eyes open for them. Such settings will be those in which it seems plausible that exposure to X was lawless, arbitrary, uncorrelated with prior conditions (Campbell, 1963, p. 65). 48 The term longitudinal was used with this study based on Good's definition of the term: ...a study that follows a case or group of cases over a period of time; includes genetic stud- ies, follow-up studies, growth studies, and experi- mental growth studies; its purpose may be to gather normative data on growth, to plot trends (as of attitudes) or to observe the effects of special factors (1973, p. 565). Hypotheses The hypotheses tested and the models and statistics used in this study were: Research Question 1 Is there a difference,as measured by standardized tests between the reading and associated abilities of students who participated in CAP and students who did not? Null Hypothesis: no difference will be found, as mea- sured by standardized test scores, between the reading and associated abilities of students who participated in CAP and students who did not. Symbolically: H : M = M Legend: M1 = CAP group mean M2 - non-CAP group mean Model: Campbell and Stanley's Design 4 and 6 will be used for Groups 2 and 6 of the study. Random assignment is assumed. Design 4. Pretest-posttest control group design (Camp- bell, 1963, pp. 13-22). 49 R O O R = Random assignment 0 = Observation X = Treatment 0 Observation (p. 13) Stanford test results are used as the pretest, and tests subsequent to X are used as posttests. Statistics: analysis of covariance ...analysis of covariance with pretest scores as the covariant are usually preferable to simple gain-score comparisons (Campbell, 1963, p. 23). Design 6. Posttest-only control group design (Camp- bell, 1963, pp. 25-27). R X 0 R O (p. 25) While the pretest is a concept deeply embedded in the thinking of research workers in education and psychology, it is not actually essential to true experimental designs. For psychological rea- sons, it is difficult to give up "knowing for sure" that the experimental and control groups were "equal" before the differential experimental treat- ment. Nonetheless, the most adequate all-purpose assurance of lack of initial biases between groups is randomization (p. 25). Statistics: t-test, analysis of covariance. Design 6 is perhaps the only setting for which this test is Optimal...covariance analysis...can be used, thus providing an increase in the power of the significance test very similar to that pro- vided by a pretest. Identicalness of pretest and posttest is not essential...whether such a pseudo- pretest design should be classified as Design 6 or Design 4 is of little moment. It would have the advantages of Design 6 in avoiding an experi- menter-introduced pretest session and in avoiding 50 the "giveaway" repetition of identical or highly similar unusual content (Campbell, 1963, p. 26). ...if appropriate antecedent variants are available, they should certainly be used for block- ing or leveling, or as covariants. This recommen- dation is made for two reasons: first, the statis- tical tests available for Design 4 are more power- ful than those available for Design 6...second, the availability of pretest scores makes possible examination of the interaction of X and pretest ability level, thus exploring the generalizability of the finding more thoroughly (p. 26). Design 10 is used to make comparisons between CAP and non-CAP groups other than the Group 2-Group 6 comparison; i.e., CAP and non-CAP combinations: 2,3,4,5 - 1,6 Design 10: the Nonequivalent Control Group Design (Campbell, 1963, pp. 47-50). 0 X 0 O O (p. 47) --------- = no pre-experimental sampling equivalence ...the addition of even an unmatched or non- equivalent control group reduces greatly the equi- vocality of interpretation over what is obtained in Design 2,the One-Group Pretest—Posttest Design. The more similar the experimental and the control groups are in their recruitment, and the more this similarity is confirmed by the scores on the pre- test, the more effective this control becomes (pp. 47-48). Statistics: analysis of covariance 51 The Designs 4, 6, and 10 of Campbell and Stanley (1963) compare with Designs 1, 2, and 3 of Fitz-Gibbon and Morris (1978) and Designs 3, 4, and 7 presented in Borg and Gall (1971). These comparisons can be seen in Table 2. Table 2. Comparability of Designs Sources of Invalidity* Campbell Morris Borg Inter— Exter— nal nal Pretest-posttest ___ ——— control group Design 4 Design 1 Design 3 none Inter- action of Test- ing, X Posttest-only control group Design 6 Design 2 Design 4 mor- none tality Unequivalent control group Design 10 Design 3 Design 7 Inter- Inter- action action of se- of se- lec- lec- tion & tion & matur- test- ation, ing etc. 8 X *Borg, 1973, p. 376; adapted from Campbell, 1963. ResearchQuestion 2 Is the effect of CAP uponciifferent student groups con- sistent over time, as measured by standardized tests? Null Hypothesis: no difference will be found between the effect of CAP on one group of students than on another group of students. 52 Symbolically: HO : M1 = M2 Legend: Ml - CAP group (Groups 2, 3, 4, 5) M 2 - Any other CAP group (Groups 2, 3, 4, 5) Statistics: same as used for Research Question 1 Design 2, the One-Group Pretest-Posttest Design (Camp- bell, 1963, p. 7) would ordinarily fit this hypothesis, but there Was more powerful information available from the use of Designs 6 and 10, previously described. Graphsanuicharts will be used to present the information for this hypothesis. Research Question 3 Is there a difference in the effect of CAP for stu- dents of varying reading abilities? Null Hypothesis: no difference will be found in the effect of CAP for students of varying reading abilities. Symbol1cally: HO : M1 = M2 Legend: M1 CAP group (Groups 2, 3, 4, 5) M2 = any other CAP group (Groups 2, 3, 4, 5) Statistics: information acquired from statistics used in Research Question 1. Scattergrams wereldeveloped from the high and low values of Gates-MacGinite, administered in grades 7 and 9. Graphs were used to indicate trends. Regression lines were used to indicate difference in effect. The Statistical Package for the Social Sciences (SPSS) (Nie, Hull, Jenkins, Steinbrennen and Brent, 1975) was used 53 for data analysis. The analysis was done at the computer center at Michigan State University. Summary The purpose of this study was to do a summative evalua- tion of the Communication Arts Pilot (CAP) program of the Traverse City (Michigan) Junior High School. Specifically, the study was designed to determine the effect of the pro- gram on the reading and associated abilities of students,.as measured by standardized tests, who were in grades seven to nine in 1972-78. The program was introduced by the communi- cation arts staff of the Traverse City Junior High School. Samples for the study were selected from the sixth grade classes of 1970, 1971, 1972, 1973, and 1974. Each sample became one group in the study except Sample (1971) which became Groups 2 and 6. Assignment to CAP or non-CAP for Groups 2 and 6 was random. Groups 1, 3, 4, and 5 were non-CAP (Group 1) or CAP (Groups 3, 4, 5), dependent only on the year they began seventh grade English. Tests administered at five different grade levels were used in the study: Stanford Achievement Intermediate II (sixth grade), Gates-MacGinite Reading Test (seventh and ninth grades), Differential Aptitude Test (eighth grade), and National Educational Deve10pment Test (tenth grade). CAP was described as a mastery learning type program with two components: reading and writing. The objectives for the reading component are from the CROFT Reading 54 Table 3. Groups (CAP-non-CAP), Tests and When Administered Test and Grade Stanford Gates DAT Gates NEDT m a £5. 1 2 2 12 l 87 1970 1972 1974 6 52 1971 1973 1974 1975 2 43 1971 X 1972 1973 1974 1975 3 87 1972 X 1973 X 1974 1975 1976 4 87 1973 X 1974 X 1975 1976 1977 5 87 1974 X 1975 X 1976 1977 1978 Stanford = Stanford Achievement Test Intermediate II Gates = Gates-MacGinite Reading Test DAT = Differential Aptitude Test NEDT = National Educational Development Test X = Communication Arts Pilot (CAP) Comprehension Skills material. The writing component is teacher-developed. The students are pre- and posttested on reading and writing skills. Up to three opportunities may be provided for a student to reach mastery on a skill. The large design framework of the study is similar to Design 15, the Recurrent Institutional Cycle Design: a "Patched-up" Design (Campbell, 1963, pp. 57-61). Two fac- tors allowed for the use of stronger designs when appropri- ate with their more powerful statistics: 55 1. Random assignment of Groups 2 and 6, treatment and control. 2. Naturally occurring X. Campbell and Stanley speak to these factors and their use whenever the opportunity is presented (Experimental and Quasi-Experimental Designs for Research, 1963). Statistics appropriate to the described designs were ap— plied to the data. The SPSS programs were used. Chapter IV contains a presentation of the data col- lected. An analysis of the data is provided. Chapter V contains the conclusions of the study. Recommendations will be made. CHAPTER IV RESEARCH FINDINGS Introduction The purpose of this study was to do a summative evalua- tion of the Communication Arts Pilot (CAP) program at Tra- verse City (Michigan) Junior High School. Specifically, the purpose was to determine the effect of the program on the reading and associated abilities of students who were in grades seven to nine, 1972-78. The CAP program was in- troduced by the Communication Arts Department of the Tra- verse City Junior High School. The research questions explored in this study were: 1. Is there a difference, as measured by standardized tests, between the reading and associated abilities of stu- dents who participated in CAP and students who did not? 2. Is the effect of CAP upon different student groups consistent over time, as measured by standardized tests? 3. Is there a difference in the effect of CAP for students of varying reading abilities? Samples for the study were selected from the sixth grade classes of 1970, 1971, 1972, 1973, and 1974. Each sample became one group in the study except Sample 2 (1971) which became Groups 2 and 6. 56 57 Assignment to CAP or non-CAP for Groups 2 and 6 was random: Groups 1 (1970), 3 (1972), 4 (1973), and 5 (1974) were non-CAP (Group 1) or CAP (Groups 3, 4, 5), dependent only on the year they began seventh grade English. Tests administered at five different grade levels were used in the study: Stanford Achievement Intermediate II (sixth grade), Gates-MacGinite Reading Test (seventh and ninth grades), Differential Aptitude Test (eighth grade), and National Educational Development Test (tenth grade). CAP was described as a mastery learning type program with two components: reading and writing. The objectives for the reading component were those of the CROFT Reading Comprehension Skills material. The writing component was teacher-developed. The students were pre— and posttested on reading and writing skills. Up to three opportunities were provided for a student to reach mastery on a skill. Table 3 is reprinted here to provide information about groups, number in each group, tests, when tests were ad- ministered to each group, when and if students participated in CAP (indicated by the X). Blanks indicate information was not available for that group. On the following pages of this chapter, the data are presented and analyzed. The major divisions of the chapter are based on the three research questions. 58 Table 3. Groups (CAP-non-CAP), Tests and When Administered Test and Grade Stanford Gates DAT Gates NEDT ___eGrou .19. e .7. .2: _9_ 19. l 87 1970 1972 1974 6 52 1971 1973 1974 1975 2 43 1971 X 1972 1973 1974 1975 3 87 1972 X 1973 X 1974 1975 1976 4 87 1973 X 1974 X 1975 1976 1977 5 87 1974 X 1975 X 1976 1977 1978 Stanford = Stanford Achievement Test Intermediate II Gates = Gates-MacGinite Reading Test DAT = Differential Aptitude Test NEDT = National Educational Development Test X = Communication Arts Pilot (CAP) Research Question 1: Hypotheses and Analysis Research Question]; is there a difference,as measured by standardized tests, between the reading and associated abilities of students who participated in CAP and students who did not? ’ Groups 2 and 6 are the groups referred to in Hypothe— seslnl to 1.7 as CAP and non-CAP. ‘Students were randomly assigned to Groups 2 and 6. Group 2 was the experimental group (CAP). Group 6 was the control group (non-CAP). Ran- domly assigned control group designs were used with these groups, Design 4 and 6 (Campbell, 1963). 59 Null Hypothesis 1.1 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Ho : M1 2 Legend: M1 = CAP group mean (Group 2) = M M2 — non-CAP group mean (Group 6) Significance Level = .05 Alternative Hypothesis 1.1 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: H : M1 # M2 1 Legend: M1 = CAP group mean (Group 2) M; = non-Cap group mean (Group 6) Significance Level = .05 Scores are reported in two forms, percentiles and grade equivalents. Percentiles The t test was used to statistically test the differ- ence between the group mean vocabulary scores of CAP and non-CAP groups. The t-value was -.102. The two tailed probability level of .919 was ppp statistically significant at the .05 level. 6O Posttest--Gates (9) Vocabulary (PR) sig. at N Mean SD t-value t-probability .05 Group 2 43 45.67 24.89 -.102 .919 no Group 6 52 46.21 29.16 Grade Equivalents The t test was used to statistically test the differ- ence in the group mean vocabulary scores between the CAP and non-CAP groups. The t-value was -.052. The two tailed probability level of .959 was ppp statistically significant at the .05 level. Posttest-~Gates (9) Vocabulary' (GE) $19. at N Mean SD t-value t-probability -05 Group 2 43 8.97 2.33 -.052 .959 no Group 6 52 9.00 2.87 Analysis of covariance was used with the Stanford Achievement Intermediate II vocabulary test as the covari— ant. The F value was .801 with a significance of .373. It was not statistically significant at the .05 level of sig- nificance. The null hypothesis 1.1 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 1.2 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the comprehension section of the Gates- 61 Symbolically: H0 : M1 = M2 Legend: M1 = CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level = .05 Alternate Hypothesis 1.2 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the comprehension section of the Gates- MacGinite Reading Test, administered in the ninth grade. Symbol1cally: Hl : Ml # M2 Legend: M1 = CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level = .05 Scores are reported in two forms, percentile and grade equivalent. Percentiles The t test was used to statistically test the differ- ence between the group mean comprehension scores of CAP and non-CAP groups. The t-value was -.049. The two tailed probability level of .961 was not statistically significant at the .05 level. 62 Posttest--Gates (9) Comprehension (PR) sig. at .05 N Mean SD t-value t-probability Group 2 43 57.26 30.42 -.049 .961 no Group 6 52 57.56 31.83 Grade Equivalent The t test was used to statistically test the differ- ence between the group mean comprehension scores of CAP and non-CAP groups. The t-value of .083 with a probability of .934 was not statistically significant at the .05 level. Posttest-~Gates (9) Comprehension (GE) sig. at N Mean SD t—value t-probability .05 Group 2 43 10.02 2.75 .083 .934 no Group 6 52 9.98 3.13 Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F value was .316 with a significance of .575. It was ppp statistically significant at the .OSleNel of significance. The null hypothesis 1.2 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 1.3 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP 63 students on the spelling section of the Differential Apti- tude Test administered in the eighth grade. Symbolically: HO : M1 = M2 Legend: M1 = CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level = .05 Alternate Hypothesis 1.3 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the spelling section of the Differential Apti- tude Test administered in the eighth grade. Symbolically: Hl : M. # M2 Legend: M1 = mean of CAP group (Group 2) M2 - mean of non-CAP group (Group 6) Significance level = .05 Scores are reported in percentiles. The t test was used to statistically test the differ- ence between the group mean spelling scores of CAP and non- CAP groups. The t-value was -2.263. The two-tailed proba- bility level of .024 was significant at the .05 level. Posttest--DAT (8) Spelling (PR) sig. at N Mean SD t-value t-probability .05 Group 2 43 40.67 26.90 -2.263 .024 yes Group 6 52 53.50 24.79 64 Analysis of covariance was used with the Stanford Achievement Intermediate II Spelling Test as the covariant. The independent variable was "group." The F value waslu879 with a significance of .174. It was not statistically sig- nificant at the .05 level of significance. Figure 1 shows the ninety-five percent confidence in- tervals of Group 2 and Group 6 for the DAT Spelling Test. 3.0\ 3.5 4.0 32.4 ------- Group 2 ----- >4 .0 46.6----Group 6----60.4 There was a statistically significant difference (.024) on the t test, the difference between group means of CAP and non-CAP groups. There was ppp a significant difference (.174) for the F value (1.879) using the analysis of covari- ance. The analysis of covariance test is the stronger sta- tistical test (Campbell, 1963). The null hypothesis 1.3 was not rejected. Null Hypothesis 1.4 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the Language Usage section of the Differential Aptitude Test administered in the eighth grade. 65 Symbolically: H : M1 = M o 2 Legend: M1 = CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level = .05 Alternate Hypothesis 1.4 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the Language Usage section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Hl : Ml # M2 Legend: M1 = CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level = .05 Scores are reported in percentiles. The t test was used to statistically test the differ- ence between the group mean language usage scores of CAP (and non-CAP groups. The t-value was .130. The two-tailed probability level of .897 was pg; significant at the ~05 level of significance. Posttest--DAT (8) Language Usage (PR) sig. at N Mean SD t-value t-probability .05 Group 2 43 59.00 24.75 .130 .897 no Group 6 52 58.37 23.17 Analysis of covariance was used with the Stanford Achievement Intermediate II Language Test as the covariant. 66 The F value was 1.111 with a significance of .295. It was not statistically significant at the .05 level of signifi- cance . The null hypothesis 1.4 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 1.5 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the verbal reasoning section of the Differen- tial Aptitude Test administered in the eighth grade. Symbol1cally: Ho : M1 = M2 Legend: - CAP group mean (Group 2) non-CAP group mean (Group 6) Z M II Significance level = .05 Alternate Hypothesis 1.5 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the verbal reasoning section of the Differential Aptitude Test administered in the eighth grade. :leM2 Symbolically: H1 CAP group mean (Group 2) Legend: M1 = non-CAP group mean (Group 6) M2 Significance level = .05 Scores are reported in percentiles. 67 The t test was used to statistically test the differ- ence between the group mean verbal reasoning scores of CAP and non-CAP groups. The t-value was .197. The two-tailed probability level of .844 was ppp statistically significant at the .05 level. Posttest-~DAT (8) Verbal Reasoning GNU sig. at N Mean SD t-value t-probability .05 Group 2 43 63.32 26.54 .197 .844 no Group 6 52 62.27 27.80 Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The P value was .570 with a significance of .452. It was p95 statistically significant at the .05 level of significance. The null hypothesis 1.5 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 1.6 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the English usage section of the National Edu- cational Development Tests administered in the tenth grade. Symbolically: H 0 M1=Mz Legend: Ml CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level: .05 68 Alternate Hypothesis 1.6 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the English usage section of the National Edu- cational Development Tests administered in the tenth grade. Symbolically: Hl : M1 # M2 Legend: M1 = CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level = .05 Scores are reported in percentiles. The t test was used to statistically test the differ- ence between the group mean verbal reasoning scores of CAP and non-CAP groups. The t-value was ~.201. The two-tailed probability level of .841 was ppp statistically significant at the .05 level of significance. Posttest--NEDT (10) English Usage (PR) sig. at N Mean SD t-value t-probability .05 Group 2 43 43.35 27.13 -.201 .841 no Group 6 52 44.56 32.24 Analysis of covariance was used with the Stanford Achievement Intermediate II Language Test as the covariant. The F ratio was .284 with a significance of .595. It was ppp statistically significant at the .05 level. The null hypothesis 1.6 cannot be rejected. The .05 level of signficance was not reached. 69 Null Hypothesis 1.7 No difference will be found between the group mean scores of CAP students and the group mean scores of non-CAP students on the word usage section of the National Educa- tional Development Tests administered in the tenth grade. Symbol1cally: Ho : M1 = M2 Legend: M1 = CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level = .05 Alternate Hypothesis 1.7 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the word usage section of the National Educa- tional Development Tests administered in the tenth grade. Symbolically: Hl : Ml # M2 Legend: M1 = CAP group mean (Group 2) M2 = non-CAP group mean (Group 6) Significance level - .05 Scores are reported in percentiles. The t test was used to statistically test the differ- ence between the group mean word usage scores of CAP and non-CAP groups. The t-value was .249. The two-tailed pro- bability level of .803 was ppp statistically significant at the.05 level of significance. Grade Equivalents 11.0 10.8 10.6 10.4 10.2 10.0 9.8 9.6 9.4 9.2 9.0 8.8 8.6 8.4 8.2 8.0 7.8 7.6 7.4 7.2 7.0 6.8 6.6 6.4 6.2 6.0 70 GRAPH 4:1.1 Grade Equivalents-Group Means Group 2 (CAP) and Group 6 (Non-CAP) 2 Tests 2 (CAP) = 6 (Non-CAP) = __ __ Vocabulary f8.996 (6) 8.97 (2) T 9 Stanford Gates Vocabulary Vocabulary Percentile - Group Means sq 8% 86 84 82 80 78 76 74 72 7o 68 66 64 62 60 58 56 54 52 so 48 46 44 42 4o 38 36 34 32 3o 71 GRAPH 4:1.2 Percentiles of Group Means Groups 2 (CAP) and 6 (No-CAP) on Four Tests 2 (CAP) = 6 (Non-CAP) = __ __ Reading Comprehension 63.3 (2) fl / /’ .54. (6-No CAP) 1.9 (2-CAP) \ 47.7 (2) \46.3 (6) 6 7 8’ 9 10 Stanford DAT Gates NEDT Rdg. Comp. Verb.Reas. Rdg.Comp. Word Usage Percentile - Group Means 9o 88 86 84 82 80 78 76 74 72 7o 68 66 64 6.2 60 58 56 54 52 so 48 46 44 42 4o 38 36 34 32 3o 72 GRAPH 4:1.3 Percentile of Group Means Group 2 (CAP) and 6 (Non-CAP) on 2 Tests 2 (CAP) 6 (Non-CAP) Spelling .545. (5) __ __ .. _ 153.5 (6 - Non-CAP) JWZ)\ ~140.7 (2 - CAP) 6 7 8 9 10 Stanford DAT Spelling Spelling Percentile - Group Means 9o 88 86 84 82 80 78 76 74 72 7o 68 66 64 62 60 58 56 54 52 so 48 46 44 42 4o 38 36 34 32 3o 73 GRAPH 4:1.4 Percentile of Group Means Group 2 (CAP) and '6 (Non-CAP) on 3 Tests 2 (CAP) 6 (Non-CAP) __ __ __ Language \44.6 (6-NOCAP) 43.3 (2-CAP) 6 8* 10 Stanford DAT NEDT Language Language Usage English Usage 74 Posttest--NEDT (10) Word Usage (PR) _sig. at N Mean SD t-value t-probability .05 Group 2 43 47.79 27.60 .249 .803 no Group 6 52 46.37 28.34 Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F ratio was .934 with a significance of .336. It was ppp statistically significant at the .05 level. The null hypothesis 1.7 cannot be rejected. The .05 level of significance was not reached. Groups 2, 3, 4, and 5 and Groups 1 and 6 are the groups referred to in Hypotheses 1.8 to 1.14 as CAP and non-CAP. Students were not randomly assigned to the groups except students in Groups 2 and 6. Information was not available for Group 1 in all instances (see Table 3). Where it was not available, the analysis was conducted using only Group 6 as the non-CAP group. An indication is made when that occurs. A non-equivalent control group design was used with these groups, Design 10 (Campbell, 1963). Null Hypothesis 1.8 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP 75 students in the vocabulary section of the Gates—MacGinite Reading Test administered in the ninth grade. Symbolically: H : M = M o 1 2 Legend: M1 = CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Group 6) Significance level: .05 Alternate Hypothesis 1.8 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Hl : M1 = M2 Legend: M1 = CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Group 6) Significance level: .05 Scores are reported in percentiles and grade equiva- lents. Percentiles The t test was used to statistically test the differ- ences between the group mean vocabulary scores of CAP and non-CAP groups. The t value was -10.240. The two-tailed probability level of .0000 was statistically significant at the .05 level of significance. 76 Posttest--Gates Vocabulary (PR) sig. at N MEAN SD t-value t-probability .05 Groups 2,3,4,5 304 47.28 24.89 -10.240 .0000 yes Group 6 52 46.21 29.16 Analysis of variance was used with the F ratio as the test statistic to determine the difference among the sample means. The ratio is determined by dividing the within group sum of squares and the between group sum of squares by their appropriate degrees of freedom. If there really is a difference among the sample means...the value of F will be substan- tially larger than 1.0...If there is no signi- ficant difference among the sample means, the two estimates (the between and within group sum of squares) will be approximately equal and the value of F will be close to 1.0 (Herzow and Hooper, 1976, p. 393). The F value was .0775 with a significance level of .7808. It was not statistically significant at the .05 level of significance. Analysis of covariance was used with the Stanford Achievement Intermediate II vocabulary test as the covari- ant. The F value was .160 with a significance level of .689. It was not statistically significant at the .05 level of significance. The ninety-five percent confidence intervals for the means of Groups 2, 3, 4, 5, and 6 (those groups who had been administered the vocabulary section of the Gates- MacGinite Reading Test in the ninth grade) are presented in Figure 2. 77 35 60 N Mean 95% Confidence Interval *Group 2 43 45.67 38.01 to 53.34 *Group 3 87 50.67 45.40 to 55.94 *Group 4 87 48.62 43.14 to 54.10 *Group 5 87 43.34 38.22 to 48.47 Group 6 52 46.21 38.09 to 54.33 * = CAP There was overlap of the confidence intervals of each group with every other group. The group means of Groups 3, 4, and 5 lie outside of the overlap. If they were accepted as the true mean of the population, the null hypothesis would have to be rejected. The confidence interval pro- vides a range within which the "true" population mean might be with ninety-five percent confidence. Grade Equivalents The t test was used to statistically test for the dif- ference between the group mean vocabulary scores of CAP and non-CAP groups. The t value was -20.488. The two-tailed probability level of .000 was statistically significant at the .05 level of significance. 78 Posttest--Gates (9) Vocabulary (GE) sig. at N Mean SD t-value t-probability .05 Groups 2,3,4,5 304 9.17 2.40 _ -20.488 .0000 yes Group 6 52 9.00 2.87 Analysis of variance was used. The F value was .2081 with a significance level of .6485. It was pgp statisti- cally signficant at the .05 level of significance. Analysis of covariance was used with the Stanford Achievement Intermediate II vocabulary test as the covari- ant. The F value was .488 with a significance level of .485. It was ppp statistically significant at the .05 level of significance. The ninety-five percent confidence intervals for the means of Groups 2, 3, 4, 5, and 6 for the Gates vocabulary administered in the ninth grade are presented in Figure 3. N Mean 95% Confidence Interval *Group 2 43 8.97 8.25 to 9.69 *Group 3 87 9.46 8.96 to 9.97 *Group 4 87 9.33 8.80 to 9.87 *Group 5 87 8.80 8.30 to 9.29 Group 6 52 9.00 8.20 to 9.80 * = CAP 79 There was overlap of the confidence intervals of each group with every other group. The groups means of Groups 3, 4, and 5 lie outside of the overlap. If they were ac- cepted as the true mean of the population, the null hypoth- esis would have to be rejected. The confidence interval provides a range within which the "true" population mean might lie with ninety-five percent confidence. The t tests for the group mean scores recorded by both percentiles and grade equivalents produced t-values with probabilities which were statistically significant at the .05 level of significance. t-value teprobability sig. at .05 Percentile -10.240 .0000 yes Grade Equivalent -20.488 .0000 yes The analysis of variance for the data in percentiles and grade equivalents produced F values with significance levels which were not statistically significant at the .05 level of significance. F value sig, of F sig. at .05 Percentile .0775 .7808 no Grade Equivalent .488 .485 no The analysis of covariance for the data in percentiles and grade equivalents produced F values with significance levels which were not statistically significant at the .05 level of significance. 80 F value sig. of F sig. at .05 Percentile .160 .689 no Grade Equivalent .488 .485 no Figure 3 provided information about the ninety-five percent confidence interval for the means of Groups 2, 3, 4, 5, and 6. There was overlap of the confidence intervals of each group with every other group. The analysis of variance, analysis of covariance, and the confidence interval ranges provide stronger statistical evidence that the null hupothesis 1.8 cannot be rejected than does the t test statistic for rejection. The null hypothesis 1.8 cannot be rejected. Null Hypothesis 1.9 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the reading comprehension section of the Gates- MacGinite Reading Test administered in the ninth grade. Symbolically: H M = M o 1 2 Legend: Ml CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Group 6) Significance level: .05 Alternate Hypothesis 1.9 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP 81 students on the reading comprehension section of the Gates- MacGinite Reading Test administered in the ninth grade. Symbol1cally: H1 : Ml # M2 Legend: Ml CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Group 6) Significance level: .05 Scores were reported in percentiles and grade equiva- lents. Percentiles The t test was used to statistically test the differ- ence between the group mean comprehension scores of CAP and non-CAP groups. The t-value was -1l.44. The two-tailed probability level of .000 was statistically significant at the .05 level of significance. Posttest--Gates (9) Comprehension (PR) sig. at N Mean SD t-value t-probability .05 Groups 2,3,4,5 304 59.24 29.58 -11.022 .000 yes Group 6 52 57.56 31.83 Analysis <1f variance was used with the F ratio as the test statistic to determine the difference between the sample means. 'The F value was .1405 with a significance of .7081. It was 993 statistically significant at the .05 level of significance. 82 Analysis of covariance was used with the Stanford Achievement Intermediate II comprehension test as the co- variant. The F value was 1.197 with a significance level of .275. It was pep statistically significant at the .05 level of significance. The ninety-five percent confidence intervals for the means of Group 2, 3, 4, 5, and 6 for the Gates comprehension test administered in the ninth grade are presented below. 45’ 0 95% of Confidence Interval *Group 2 43 47.89 to 66.62 *Group 3 87 57.54 to 69.11 *Group 4 87 54.13 to 67.05 *Group 5 87 48.28 to 61.31 Group 6 52 48.70 to 66.42 * = CAP There was overlap of the confidence intervals of each group with every other group. The group mean of Groups 2, 3, and 5 lie outside of the overlap. The confidence inter- val provides a range within which the "true" population mean might lie with ninety-five percent confidence. Grade Equivalents The t test was used to statistically test the differ- ence between the group mean comprehension scores of CAP 83 and the non-CAP groups. The t-value was -20.925. The two- tailed probability level of .000 was statistically signifi- cant at the .05 level of significance. Posttest--Gates (9) Comprehension (GE) sig. at N Mean SD t-value t-probability .05 Groups 2,3,4,5 304 10.19 2.76 -20.100 .000 yes Group 6 52 9.98 3.13 Analysis of variance was used with the F ratio as the test statistic to determine the difference among the sample means. The F-value was .2583 with a significance of .6116. It was ppp statistically significant at the .05 level of significance. Analysis of covariance was used with the Stanford Achievement Intermediate II comprehension test as the co- variant. The F value was 1.434 with a significance of.232. It was p92 statistically significant at the .05 level of significance. The ninetyefive percent confidence intervals for the means of Groups 2, 3, 4, 5, and 6 for the Gates comprehen- sion test administered in the ninth grade appear below. \ 10.2 10 4 10.6 10 11.0 1.2 11.4 84 N Mean '95% of Confidence Interval *Group 2 43 10.02 9.18 to 10.87 *Group 3 87 10.61 10.07 to 11.15 *Group 4 87 10.22 9.62 to 10.82 *Group 5 87 9.82 9.20 to 10.43 Group 6 52 9.98 9.10 to 10.85 * = CAP There was overlap of the confidence interval of each group with every other group. The group mean of Groups 2, 3, 5, and 6 lie outside of the overlap. The confidence interval provides a range within which the "true" pOpula- tion mean might lie with ninety-five percent confidence. The t tests for the group mean scores recorded by both percentiles and grade equivalents produced t-values with probabilities which were statistically significant at the .05 level of significance. t-value t-probability sig. at .05 Percentile -11.022 .000 yes Grade Equivalent -20.100 .000 yes The analysis of variance for the data in percentiles and grade equivalents produced,F values with significance levels which were not statistically signficant at the .05 level of significance. F-value sig. of F sig. at .05 Percentile .1405 .7081 no Grade Equivalent .2583 .6116 no 85 Tha analysis of covariance for the data in percentiles and grade equivalents produced F values with significance levels which were not statistically significant at the .05 level of significance. F value sig. of F sig. at .05 Percentiles 1.197 .275 no Grade Equivalents 1.434 .232 no Figure 4 provided information about the ninety-five percent confidence interval for the means of Groups 2, 3, 4, 5, and 6. There was overlap of the confidence interval of each group with every other group. The analysis of variance, analysis of covariance, and the confidence interval ranges provide stronger statistical evidence that the null hypothesis 1.9 cannot be rejected than does the t—test statistic for rejection. The null hypothesis 1.9 cannot be rejected. Null Hypothesis 1.10 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the spelling section of the Differential Apti- tude Test administered in the eighth grade. Symbolically: H : M = M o 1 2 Legend: M1 = CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Groups 1,6) Significance level: .05 86 Alternate Hypothesis 1.10 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the spelling section of the Differential Apti- tude Test administered in the eighth grade. Symbolically: H1 : Ml f M2 Legend: M1 = CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Groups 1,6) Significance level: .05 Scores were reported in percentiles. The t test was used to statistically test the differ- ence between the group mean spelling scores of CAP and non- CAP groups. The t-value was 1.643. The two-tailed proba- bility level of .101 was pgp statistically significant at the .05 level of significance. Posttest--DAT (8) Spelling (PR) sig. at N Mean SD t-value t-probability .05 Groups 2,3,4,5 303 46.96 28.28 1.643 .101 no Group 6 139 50.33 26.18 Analysis of variance was used with the F ratio as the test statistic to determine the difference among the sample means. The F-value was 1.4196 with a significance of .2341. It was ppp statistically significant at the .05 level of significance. 87 Analysis of covariance was used with the Stanford Achievement Intermediate II spelling test as the covariant. The F-value was .3534 with a significance of .061. It was 22; statistically significant at the .05 level of signifi- cance. The null hypothesis 1.10 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 1.11 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the language usage section of the Differential Aptitude Test administered in the eighth grade. Symbolically: HO : M1 = M2 Legend: M1 = CAP group mean (Groups 2,3,4,5) 2 non—CAP group mean (Groups 1,6) M Significance level: .05 Alternate Hypothesis 1.11 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the language usage section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Hl : Ml # M2 1 CAP group mean (Groups 2,3,4,5) Legend: M II M2 non-CAP group mean (Groups 1,6) Significance level: .05 Scores were reported in percentiles. 88 The t test was used to statistically test the differ- ence between the group mean language usage scores of CAP and non-CAP groups. The t-value was -2.952. The two- tailed probability level of .003 was statistically signifi- cant at the .05 level of significance. Posttest--DAT (8) Language Usage (PR) sig. at N Mean SD t-value t-probability .05 Groups 2,3,4,5 303 59.65 23.92 -2.952 .003 yes Groups 1,6 139 50.54 23.84 The t-value which occurred with different combinations of CAP/non-CAP groupings also had significant levels of probability. Posttest--DAT (8) Language Usage (PR) N t-value t-probability sig. at .05 *Group 3 87 _ Group 1 87 4.395 .000 yes *Groups 2,3 174 _ Groups 1,6 139 3.026 .003 yes *Groups 2'3'4'5 303 -2.952 .003 yes Groups 1,6 139 * = CAP Analysis of variance was used with the F-ratio as the test statistic to determine the difference among sample means. The F-value was 13.8419 with a significance of 89 .0002. It was statistically significant at the .05 level of significance. Analysis of covariance was used with the Stanford Achievement Intermediate II Language Test as the covariant. The F-value was 7.493 with a significance of .006. It was statistically significant at the .05 level of significance. Analysis of covariance used with the independent vari- ables being Groups 1 and Groups 3, 4, and 5 and with the Stanford Achievement Intermediate II Language Test as the covariant showed a significant F. The F value was 17.041 with a significance level of .001. It was statistically significant at the .05 level of significance. ANCOVA-DAT Language Usage with Stanford Lang. N F-value sig. of F Sig. at .05 *Groups 3,4,5 261 Group 1 87 17.041 .001 yes *Groups 2,3,4,5 303 Groups 1,6 139 7.493 .006 yes * = CAP The ninety-five percent confidence intervals for the means of Groups 1, 2, 3, 4, 5, and 6 for the DAT language usage tests are presented in Figure 6. \w ( \ /"-—_—J 90 N Mean 95% of Confidence Interval Group 1 87 45.86 40.94 to 50.79 *Group 2 43 59.00 51.38 to 66.62 *Group 3 87 61.68 56.93 to 66.43 *Group 4 87 59.40 54.28 to 64.52 *Group 5 87 58.16 52.75 to 63.58 Group 6 52 58.37 51.91 to 64.82 * = CAP The range of the confidence interval for Group 1 (non- CAP) lies outside of the confidence intervals of each of the other groups. The confidence interval provides a range within which the "true" pOpulation mean might lie with ninety-five percent confidence. The null hypothesis 1.11 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 1.12 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the verbalreasoning section of the Differential Aptitude Test administered in the eighth grade. Symbolically: HO : M1 = M2 CAP group mean (Groups 2,3,4,5) Legend: Ml M = non-CAP group mean (Groups 1,6) 2 Significance level: .05 Alternate Hypothesis 1.12 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP 91 students on the verbal reasoning section of the Differential Aptitude Test administered in the eighth grade. Symbolically: H1 : Ml # M2 Legend: Ml CAP group mean (Groups 2,3,4,5) M non-CAP group mean (Groups 1,6) 2 Significance level: .05 Scores were reported in percentiles. The t test was used to statistically test the differ- ence between the group mean verbal reasoning scores of CAP and non-CAP groups. The t-value was -.837. The two-tailed probability level of .403 was not statistically significant at the .05 level of significance. Posttest--DAT (8) Verbal Reasoning (PR) sig. at N Mean SD t-value t-probability .05 Groups 2,3,4,5 303 62.25 25.75 -.837 .403 no Groups 1,6 139 59.51 26.67 Analysis of variance was used with the F ratio as the test statistic to determine the difference among the sample means. The F value was 1.0577 with a significance of .1793. It was not statistically significant at the .05 level of significance. Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F value was 1.025 with a significance level of .312. It was not statistically significant at the .05 level of significance. 92 The null hypothesis 1.12 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 1.13 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students in the English usage section of the National Edu- cational DevelOpment Test administered in the tenth grade. Symbolically: HO : M = M 1 2 Legend: M1 = CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Groups 1,6) Significance level: .05 Alternate Hypothesis 1.13 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the English usage section of the National Edu- cational Development Test administered in the tenth grade. Symbol1cally: Hl : Ml # M2 Legend: Ml CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Groups 1,6) Significance level: .05- Scores were reported in percentiles. The t test was used to statistically test the differ- ences between the group mean English usage scores of CAP and non-CAP groups. The t-value was —.201. The two-tailed probability level of .841 was not statistically significant at the .05 level of significance. 93 Posttest--NEDT English Usage (PR) sig. at N Mean SD t-value t-probability .05 Groups 2,3,4,5 304 45.60 28.76 -.085 .933 no Groups 1,6 139 45.18 29.89 Analysis of variance was used with the F ratio as the test statistic to determine the difference among the sample means. The F-value was .0200 with a significance of .8875. It was 293 statistically significant at the .05 level of significance. Analysis of covariance was used with the Stanford Achievement Intermediate II Language Test as the covariant. The F-value was 5.174 with a significance level of .023. It was statistically significant at the .05 level of signi- ficance. The ninety-five percent confidence intervals for the means of Groups 1, 2, 3, 4, 5, and 6 are presented in Fig- ure '7 for NEDT English usage. 56 N Mean 95% of Confidence Interval Group 1 87 45.55 39.46 to 51.64 *Group 2 87 43.35 35.00 to 51.90 *Group 3 87 48.24 42.09 to 54.39 *Group 4 87 45.47 39.08 to 51.87 *Group 5 87 44.21 28.14 to 50.28 Group 6 52 44.56 35.58 to 53.53 * = CAP There was overlap of the confidence intervals of each group with every other group. The group mean of each group lies within the overlap. The confidence interval provides a range within which the "true" population mean might lie with ninety-five percent confidence. The t test was used to statistically test the differ- ence between the group mean Stanford Language scores of the CAP and non-CAP students (the covariant used in the analy- sis of covariance above). The t-value was -2.050. The two-tailed probability level of .041 was ppp statistically significant at the .05 level of significance. Analysis of variance was used to determine the differ- ence among the sample means on the Stanford Language Test. The F-value was 6.3980 with a significance level of .0118. It was statistically significant at the .05 level of signi- ficance. The ninety-five percent confidence intervals for the means of Groups 1, 2, 3, 4, 5, and 6 are presented in Figure 8 for Stanford Language. ...... 5----- 58 60 62 64 66 ....... 4----- 8 70 72 N Mean 95% of Confidence Interval Group 1 87 48.22 42.55 to 53.89 *Group 2 87 42.51 35.59 to 49.43 *Group 3 87 45.21 39.45 to 50.97 *Group 4 87 62.41 57.13 to 67.70 *Group 5 87 61.78 56.36 to 67.20 Group 6 52 46.50 38.96 to 54.04 * = CAP The range of the confidence intervals for Groups 4 and 5 lie outside the confidence intervals of the other groups. The confidence interval and the group means are very simi- lar for Groups 4 and 5. The .05 level of significance was reached in the anal- ysis of covariance with the Stanford Achievement Intermedi- ate II Language Test as the covariant. There appears to be a difference. The null hypothesis 1.13 cannot be accepted. Null Hypothesis 1.14 No difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the word usage section of the National Educa- tional Development Test administered in the tenth grade. Symbolically: H03M1=M2 96 Legend: M1 = CAP group mean (Groups 2,3,4,5) M2 = non-CAP group mean (Groups 1,6) Significance Level: .05 Alternate Hypothesis 1.14 A difference will be found between the group mean score of CAP students and the group mean score of non-CAP students on the word usage section of the National Educa- tional Development Test administered in the tenth grade. Symbolically: H1 : M1 # M2 Legend: M1 = CAP group mean (Groups 2,3,4,5) non-CAP group mean (Groups 1,6) M2 Significance Level: .05 Scores were reported in percentiles. The t test was used to statistically test the differ- ence between the group mean word usage scores of CAP and non-CAP groups. The t-value was -.792. The two-tailed probability level of .442 was ppp statistically significant at the .05 level of significance. Posttest-~NEDT Word Usage (PR) sig. at N Mean SD tevalue tfiprobability .05 Groups 2,3,4,5 304 50.38 27.37 —.769 .442 no Groups 1,6 139 48.15 28.42 Analysis of variance was used with the F-ratio as the test statistic to determine the difference among the sample Grade Equivalent - Group Means 97 GRAPH 4:1.5 Grade Equivalents of Group Means on 2 tests All Groups CAP = Non-CAP = _____ __ 10.6 10.4 10.2 10.0 9.8 9.6 9.4 ,.9.46 (3) 9.2 9.33 (4) 9.00 (6) 9'0 ,’8.97 (2) 8.8 , 8.80 (5) 8.6 / ' / 8.4 8.2 / / 8.0 :.00 ’,CAP 7'8 .71 - AP) 7.6 / / 7.4 / 7.2 / . a (3 - CAP) / . * .. 7°C “7’21 -.99 (l - Non-CAP) 6.6 (2 - CAP) 6.4 6 2 *7.21 (6 - Non-CAP) 6.0 0 i 6 . 7 8 9 10 Stanford Gates Vocabulary Vocabulary Percentile - Group Means 9o 88 86 84 82 80 78 76 74 72 7o 68 66 64 62 6o 58 56 54 52 so 48 46 44 42 4o 38 36 34 32 3o 98 GRAPH 4:1.6 Percentiles of Group Means on Four Tests All Groups CAP = Non-CAP = __ __ __ Reading Comprehension 66 (3) 64.6 (5) as.6 (5) 63.3 (2 63 (3) \. ‘.§;:2 (4) ‘\ A2; ‘ 60. (4) ’ \ 5 5 (6) ,/ 57.8 (1) ‘\5.7.3 ,) 4/ (’3) 5} ( 251(6) \ \ ° 1 .8 (1) \ 52.8 (4) 1.9 (2) \\\ 52.5 (3) \.49.2 (1) 47.7 (2) '\ 47.1 (5) \46.3 (6) 6 7 8 9 10 Stanford DAT Gates NEDT Rdg. Comp. Verb.Reas. Rdg.Comp. Word Usage Percentiles — Group Means 90 88 86 84 82 80 78 76 74 72 70 68 66 64 62 60 58 56 54 52 50 48 46 44 42 4o 38 36 34 32 30 99 GRAPH 4:1.7 Percentile of Group Means on 2 Tests All Groups CAP = Non-CAP = Spellng—- _— 58.7 (5 - CAP) ~.53.5 (6) 52 (3) \ 48.9 (5) ‘ ~.48.4 (l) .9 (2 - CAP) \42.9 (4) 40.7 (2) 6 7 8 9 10 Stanford DAT Spelling Spelling 90 88 86 84 82 80 78 76 74 72 70 68 66 64 62 60 58 56 54 52 50 48 46 44 42 40 38 36 34 32 30 100 GRAPH 4:1.8 Percentile of Group Means All Groups On 3 Tests 2,3,4,5 (CAP) = 1,6 (Non-CAP) = __ __ __ Language é1°f‘(4 C9?) 61.7 (3) ’ ‘ ~9.4 (4) \ 6‘ (2) 58.3 »=. (5) \ .48/ ~Non-CAP) \\ 48.2 (3) I-. 6~Nop:CAP) \\b 45 (3-CAP)"”~~145.9_111 45.5 (1H6) 45.4 (4) 12.5 (2-CAP) .44.2 (5) 43.3 (2) 6 8 10 Stanford DAT NEDT Language Language Usage English Usage 101 means. The F-value was .6203 with a significance of .4314. It was ppp statistically significant at the .05 level of significance. Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F-value was 3.041 with a significance level of .082. It was not statistically significant at the .05 level. The null hypothesis 1.14 cannot be rejected. The .05 level of significance was not reached. Research Question 2: Hypothesis and Analysis Research Question 2: is the effect of CAP upon stu- dent groups consistent over time, as measured by standar- dized tests? Groups 2 and 3 are the groups referred to in Hypothe- ses 2.1 to 2.9. Groups 3, 4, and 5 were analyzed in Hy- potheses 2.10 to 2.18. Groups 4 and 5 were analyzed in Hypotheses 2.19 to 2.25. The analysis of covariance was used to determine dif- ference in the effect of CAP during different years of the program. The only group which had had students randomly assigned to it was Group 2. Group 2 was the first group of students to have CAP. They had the program for one year. Groups 3, 4, and 5 each had CAP for two years. All students in Groups 3, 4, and 5 were in CAP. 102 Null Hypothesis 2.1 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: HO : M1 = M2 one CAP group mean (Group 2) Legend: M1 M2 second CAP group mean (Group 3) Significance Level: .05. Alternate Hypothesis 2.1 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Hl : Ml # M2 Legend: “1 one CAP group mean (Group 2) M2 second CAP group mean (Group 3) Significance Level: .05 Scores were reported in two forms, percentiles and grade equivalents. Percentiles Analysis of covariance was used with the Stanford Achievement Intermediate II vocabulary Test as the 103 covariant. The F-value was .507 with a significance of .478. It was not statistically significant at the .05 level of significance. Grade Equivalents Analysis of covariance was used with the Stanford Achievement Intermediate II vocabulary test as the covari- ant. The F-value was .644 with a significance of .424. It was not statistically significant at the .05 level of sig- nificance. The null hypothesis 2.l cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.2 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the comprehension sec- tion of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: H ”1 = M2 0 Legend: M1 one CAP group mean (Group 2) M2 second CAP group mean (Group 3) Significance Level: .05 Alternate Hypothesis 2.2 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the comprehension 104 section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: H1 : Ml # M2 Legend: M1 = one CAP group mean (Group 2) 3 M II second CAP group mean (Group 3) Significance Level: .05 Scores were reported in two forms, percentiles and grade equivalents. Percentiles Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F-value was 1.218 with a significance level of .272. It was not statistically significant at the .05 level of significance. Grade Equivalents Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F-value was 1.824 with a significance level of .179. It was not statistically significant at the .05 level of significance. The null hypothesis 2.2 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.3 No difference will be found between the group mean score of students in one CAP group and the group mean score 105 of students in a second CAP group on the spelling section of the Differential Aptitude Test administered in the eighth grade. Symbolically: HO : M1 = M2 Lengend: M1 = one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 Alternate Hypothesis 2.3 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the spelling section of the Differential Aptitude Test administered in the eighth grade. Symbolically: H1 : M1 # M2 Legend: M1 = one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II spelling test as the covariant. The F-value was 3.107 with a significance level of .080. It was not significant at the .05 level of significance. The null hypothesis 2.3 cannot be rejected. The .05 level of significance was not reached. 106 Null Hypothesis 2.4 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the language usage section of the Differential Aptitude Test administered in the eighth grade. Symbolically: HO : M1 = M2 Legend: M1 = one CAP group mean (Group 2) M2 = aecond CAP group mean (Group 3) Significance Level: .05 Alternate Hypothesis 2.4 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the language usage section of the Differential Aptitude Test administered in the eighth grade. Symbolically: H1 : M1 # M2 Legend: M1 = one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II Language test as the covariant. The F-value was .113 with a significance level of .738. It was 22E statistically significant at the .05 level of signi- ficance. 107 The null hypothesis 2.4 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.5 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the verbal reasoning section of the Differential Aptitude Test administered in the eighth grade. Symbolically: HO : M1 = M2 Legend: M1 = one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 Alternate Hypothesis 2.5 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the verbal reasoning sec— tion of the Differential Aptitude Test administered in the eighth grade. Symbolically: H 0 M1 # M2 Legend: M1 = one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F-value was .037 with a significance 108 level of .847. It was not statistically significant at the .05 level of significance. The null hypothesis 2.5 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.6 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the English usage sec- tion of the National Educational Development Test adminis- tered in the tenth grade. Symbolically: H 0 M1 = M2 Legend: M1 = one CAP group mean (Group 2) M2 - second CAP group mean (Group 3) Significance Level: .05 Alternate Hypothesis 2.6 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the English usage section of the National Educational DevelOpment Test administered in the tenth grade. Symbolically: H 0 Ml # M2 Legend: M1 = one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford 109 Achievement Intermediate II language section as the covari- ant. The F-value was .564 with a significance level of.454. It was 22E statistically significant at the .05 level of significance. The null hypothesis 2.6 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.7 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the word usage section of the National Educational DevelOpment Test administered in the tenth grade. Symbolically: H M1 = M2 0 Legend: M1 one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 Alternate Hypothesis 2.7 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the word usage section of the National Educational Development Test administered in the tenth grade. Symbolically: HO : Ml # M2 Legend: M1 = one CAP group mean (Group 2) M2 second CAP group mean (Group 3) 110 Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II reading comprehension section of the National Educational Development Test as the covariant. The F-value was .194 with a significance level of .660. It was 293 statistically significant at the .05 level of signi- ficance. The null hypothesis 2.7 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.8 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: HO : M1 = M2 Legend: M1 = one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 - Alternate Hypothesis 2.8 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. 111 SymbolicallY: Ho M1 f M2 Legend: one CAP group mean (Group 2) 3 ...a II M2 = second CAP group mean (Group 3) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Gates—MacGin- ite Reading Test, vocabulary section, administered in the seventh grade, as the covariant. The F-value was .245 with a significance level of .622. It was pg; statistically sig- nificant at the .05 level of significance. The null hypothesis 2.8 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.9 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the comprehension sec- tion of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Ho M1 = M2 Legend: M1 one CAP group mean (Group 2) M2 = second CAP group mean (Group 3) Significance Level: .05 Alternate Hypothesis 2.9 A difference will be found between the group mean score of students in one CAP group and the group mean score of 112 students in a second CAP group on the comprehension section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Ho M1 # M2 Legend: Ml one CAP group mean (Group 2) 3 M II second CAP group mean (Group 3) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Gates—Mac- Ginite Reading Test, comprehension section, administered in the seventh grade, as the covariant. The F-value was 6.289 with a significance level of .013. It was statistically significant at the .05 level of significance. The mean of Group 2 on the Gates (7) comprehension section was 8.00. The mean of group 3 on the comprehension section was 7.6 The ninety-five percent confidence intervals for the mean of Groups 2 and 3 for the comprehension section of the Gates-MacGinite Reading Test administered in the seventh grade are shown in Figure 9. 113 N Mean 95% of Confidence Interval Group 2 43 8.00 7.23 to 8.78 Group 3 87 7.60 7.07 to 8.13 The mean of Group 2 (8.00) is near the upper end of the confidence interval for Group 3 (8.13). The ninety-five percent confidence interval for the means of Groups 2 and 3 on the comprehension section of the Gates-MacGinite Reading Test administered in the ninth grade are shown in Figure 10. 9.0 .2 9.4 9.6 9.8 10.0 10.2 10.4 10.6 13f8 11.0 kl.2 11.4 I __/ /' N Mean 95% of Confidence Interval Group 2 43 10.02 9.18 to 10.87 Group 3 87 10.61 10.01 to 11.15 The overlap of the ranges had changed. The null hypothesis 2.9 cannot be accepted. The .05 0 level of Significance was reached. Null Hypothesis 2.10 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section 114 of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: HO : M1 = M2 Legend: M1 = one CAP group mean (Group 3) M2 - second CAP group mean (Groups 4,5) Significance Level: .05 Alternate Hypothesis 2.10 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Hl : M1 # M2 Legend: M1 - one CAP group mean (Group 3) M2 second CAP group mean (Groups 4,5) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Stanford Achievement Intermediate II vocabulary section as the co- variant. The F-value was 21.465 with a significance level of .001. It was statistically significant at the .05 level of significance. The mean of Group 3 on the vocabulary section of the Stanford was 7.20. The mean for Group 4 was 7.99. The mean for Group 5 was 7.72 115 The ninety-five percent confidence intervals for the means of Groups 3, 4, and 5 for the vocabulary section of the Stanford are shown in Figure 11. 6.0 6.2 6.4 6.6 6.8 7.0 7.2 7.4 7. 7,8 8.0 8.2 .4 8.6 8.8 N Mean 95% of Confidence Intervals Group 3 87 7.21 6.90 to 7.52 Group 4 87 8.00 7.64 to 8.36 Group 5 87 7.72 7.32 to 8.11 The mean of Group 3 was outside the lower limits of Groups 4 and 5. The upper limit of the range of Group 3 was outside the lower limit of the range of Group 4. The ninety—five percent intervals for the same groups on the vocabulary section of the Gates-MacGinite, ninth grade, are shown in Figure 12. ....... 5------ -_- ------_ ............ 4-----__-_-- 9.6 l .0 ............... 3------_- -- N Mean 95% of Confidence Interval Group 3 87 9.46 8.96 to 9.97 Group 4 87 9.33 8.80 to 9.87 Group 5 87 8.80 8.30 to 9.29 116 There was a shift in position of the three groups in relation to each other. The null hypothesis 2.10 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 2.11 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group in the comprehension sec- tion of the Gates-MacGinite Reading Test adminsitered in the ninth grade. Symbolically: HO : M1 = M2 Legend: M1 one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance level: .05 Alternate Hypothesis 2.11 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the comprehension sec- tion of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: H 0 M1 # M2 Legend: M1 one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) 117 Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Stanford Achievement Intermediate II comprehension section as the co- variant. The F-level was 22.149 with a significance level of .001. It was statistically significant at the .05 level of significance. The mean of Group 3 on the Stanford was 7.14. The mean of Group 4 was 8.30. The mean of Group 5 was 8.35. The ninety-five percent confidence intervals for the means of Groups 3, 4, and 5 for the comprehension sec- tion of the Stanford are shown in Figure 13. 6.0 6.2 6.4 6.6 .8 7.0 7.2 7.4 7.6 7.8 8.0 8.2 8.4 8.6 8. N Mean 95% of Confidence Intervals Group 3 87 7.14 .6.78 to 7.50 Group 4 87 8.30 7.89 to 8.71 Group 5 87 8.35 7.90 to 8.80 There was no overlap of the confidence interval of Group 3 with those of Groups 4 and 5. The ninetyefive percent confidence intervals of Groups 3, 4, and 5 on the Gates-MacGinite comprehension section, ninth grade, are shown in Figure 14. 118 9.0 9. N Mean 95% of Confidence Intervals Group 3 87 10.61 10.07 to 11.15 Group 4 87 10.22 9.62 to 10.82 Group 5 87 9.82 9.20 to 10.43 There appears to have been a shift in relative posi- tions of the three groups, indicated by the range of the confidence intervals. The null hypothesis 2.11 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 2.12 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the spelling section of the Differential Aptitude Test administered in the eighth grade. - Symbolically: Ho : M1 = M2 Legend: M1 = one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance Level: .05 119 Alternate Hypothesis 2.12 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the spelling section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Ho : Ml # M2 Legend: M1 one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II spelling test as the covariant. The F-value was 22.149 with a significance level of .001. It was statistically significant at the .05 level of signi- ficance. The mean of Group 3 for the Stanford spelling test was 51.21. The mean of Group 4 was 54.09. The mean of Group 5 was 58.71. The ninety-five percent confidence intervals for the means of Groups 3, 4, and 5 for the spelling section of the Stanford are shown in Figure 15. 40 42 44 62 64 66 68 120 N Mean 95% of Confidence Intervals Group 3 87 51.21 45.59 to 56.82 Group 4 87 54.09 48.69 to 59.50 Group 5 87 58.71 52.83 to 64.60 The mean of Group 4 lies within the overlap of the in- tervals of the three groups. The mean of Group 3 lies be- low it and that of Group 5 lies above it. The ninety-five percent confidence interval of Groups 3, 4, and 5 on the spelling section of the Differential Ap- titude Test is shown in Figure 16. .......... 5-------_-- ---_ 52 54 60 62 64 ......... 3_____---- N Mean 95% Confidence Interval Group 3 87 52.12 46.28 to 57.97 Group 4 87 42.95 37.16 to 48.75 Group 5 87 48.92 46.60 to 60.40 There has been an apparent shift in position, but it is most evident between the Groups 4 and 5. The null hypothesis 2.12 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 2.13 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the language usage 121 section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Ho : M1 = M2 Legend: M 1 one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Alternate Hypothesis 2.13 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the language usage section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Ho : M1 # M2 Legend: M1 = one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II Language Test as the covariant. The F-value was 33.766 withaasignificance level of .001. It was statistically significant at the .05 level of signifi- cance. The mean of Group 3 for the Stanford language test was 45.20. The mean of Group 4 was 62.41. The mean of Group 5 was 61.78. 122 The ninety-five percent confidence intervals for the means of Groups 3, 4, and 5 for the language section of the Stanford are shown in Figure 17. ..... 5------__-- 58 60 62 64 66 8 ......... 4----— N Mean 95% of Confidence Intervals Group 3 87 45.20 39.45 to 50.97 Group 4 87 62.41 57.13 to 67.70 Group 5 87 61.78 56.36 to 67.20 The upper limit of the range of the confidence interval for Group 3 was below the lower limit of the range for Groups 4 and 5. There is pp overlap between Group 3 and Groups 4 and 5. The ninety-five percent confidence intervals for the means of Groups 3, 4, and 5 for the language section of the 7T0 52 5—5638 6o 62 64 66 68 '70? ....... 3----- -_ --__ DAT are shown in Figure 18. 123 N Mean 95% of Confidence Intervals Group 3 87 61.68 56.93 to 66.43 Group 4 87 59.40 54.28 to 64.52 Group 5 87 58.12 52.75 to 63.58 The shift in relative position of the range of confi- dence intervals is shown. In the covariant the ranges did not have an area of overlap (between Group 3 and Groups 4 and 5), In the posttest, the range of confidence inter- vals indicates quite an overlap and all of the grOup sam- ple means are within the overlap. The null hypothesis 2.13 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 2.14 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the verbal reasoning section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Ho : M1 = M2 1 one CAP group mean (Group 3) Legend: M M2 second CAP group mean (Groups 4,5) Significance Level: .05 Alternate Hypothesis 2.14 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the verbal reasoning 124 section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Ho : M1 # M2 Legend: M1 = one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F-value was 30.88 with a significance level of .001. It was statistically significant at the .05 level of significance. The mean of Group 3 for the Stanford reading comprehen- test was 54.87. The mean of Group 4 was 64.43. The mean of Group 5 was 64.60. The ninety-five percent confidence interval for the means of Groups 3, 4, and 5 for the reading comprehension section of the Stanford are shown in Figure 19. 46 48 72 74 N Mean 95% of Confidence Intervals Group 3 87 54.87 49.62 to 60.12 Group 4 87 64.43 59.32 to 69.53 Group 5 87 64.60 59.15 to 70.09 125 There was an overlap of the confidence intervals of the three groups. The ninety-five percent confidence interval for the means of Groups 3, 4, and 5 for the verbal reasoning sec- tion of the Differential Aptitude Test are shown in Figure 20. ........ 5------ ------- ____________ 4---------- ............ 3---- ------ 50 52 54 56 §:\\ 60 \\::Z/ 64 66 68 70 72 N Mean 95% of Confidence Intervals Group 3 87 66.00 60.97 to 71.03 Group 4 87 63.25 57.89 to 68.62 Group 5 87 56.92 51.02 to 62.82 Group 3 had shifted position in relation to Groups 4 and 5. In the figure showing means in intervals on the Stanford Reading Comprehsnion Test, the range of Group 3 was the lowest range. In Figure 20, it has the highest range. The null hypothesis 2.14 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 2.15 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the English usage sec- tion of the National Educational Development Test adminis- tered in the tenth grade. 126 0 M1 = M2 Legend: M1 - one CAP group mean (Group 3) Symbolically: H M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Alternate Hypothesis 2.15 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the English usage sec- tion of the National Educational Development Test adminis- tered in the tenth grade. Symbolically: Ho Ml’éMz Legend: M1 one CAP group mean (Group 3) M2 - second CAP group mean (Groups 4,5) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Acheivement Intermediate II Language Test as the covariant. The F-value was 40.109 with a significance level of .001. It was statistically significant at the .05 level of signi- ficance. See Figure 17 for information on the ninety-five per- cent confidence intervals for the means of Groups 3, 4, and 5 of the Stanford Language test. 127 The ninety-five percent confidence interval for the means of Groups 3, 4, and 5 for the English usage section of the NEDT, administered in the tenth grade, is shown in Figure 21. ----- 5----- --------- 4--- -- 36 2 54 56 58 6O 62 64 66 68 70 72 74 N Mean 95% of Confidence Intervals Group 3 87 48.24 42.09 to 54.39 Group 4 87 45.47 39.08 to 51.87 Group 5 87 44.21 38.14 to 50.28 The relative position of confidence intervals has changed. In Figure 17 there was no overlap between the con- fidence interval of Group 3 and those of Groups 4 and 5. The null hypothesis 2.15 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 2.16 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the word usage section of the National Educational Development Test administered in the tenth grade. Symbolically: Ho : M1 = M2 Legend: Ml one CAP group mean (Group 3) 128 M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Alternate Hypothesis 2.16 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the word usage section of the National Educational Development Test administered in the tenth grade. Symbolically: Ho : Ml # M2 Legend: M1 = one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F-value was 23.129 with a significance level of .001. It was statistically significant at the .05 level of significance. See Figure 19 for information on the ninety-five per- cent confidence intervals for the means of Groups 3, 4, and 5 of the Stanford Achievement Intermediate II Reading Com- prehension test. The null hypothesis 2.16 cannot be accepted. The .05 level of significance was reached. 129 Null Hypothesis 2.17 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: HO : M1 = M2 Legend: Ml one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Alternate Hypothesis 2.17 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Ho : M1 # M2 Legend: M1 one CAP group mean (Group 3) M2 = second CAP group mean (Groups 4,5) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Gates-Mac- Ginite REading Test, vocabulary section, administered in the seventh grade, as the covariant. The F-value was 1.63 with a significance level of .687. It was 22E statisti- cally significant at the .05 level of significance. 130 The null hypothesis 2.17 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.18 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the comprehension sec- tion of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Ho : M1 = M2 1 one CAP group mean (Group 3) Legend: M M2 second CAP group mean (Groups 4,5) Significance Level: .05 Alternate Hypothesis 2.18 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the comprehension sec- tion of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: HO M1 # M2 Legend: M1 - one CAP group mean (Group 3) M2 second CAP group mean (Groups 4,5) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Gates-Mac- Ginite Reading Test, comprehension section, administered in the seventh grade, as the covariant. The F-value was 131 11.781 with a significance level of .001. It was statis- tically significant at the .05 level of significance. The ninety-five percent confidence intervals for the means of Groups 3, 4, and 5 for the comprehension section of the Gates-MacGinite Reading test, administered in the seventh grade, are shown in Figure 22. KK ” KBJJ J 9° _ Mean 95% of Confidence Intervals Group 3 87 7.60 7.07 to 8.13 Group 4 87 8.21 7.60 to 8.82 Group 5 87 7.56 6.94 to 8.19 None of the group means is in the overlap area. The overlap is below the group mean of Group 4 and above the group mean of Groups 3 and 5. The null hypothesis 2.18 cannot be accepted. The .05 level of significance was reached. Null Hyppthesis 2.19 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. 132 Symbolically: Ho : M1 = M2 1 one CAP group mean (Group 4) 3 ll Legend: M2 = second CAP group mean (Group 5) Significance Level: .05 Alternate Hypothesis 2.19 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: Ho : Ml # M2 Legend: M1 = one CAP group mean (Group 4) M2 second CAP group mean (Group 5) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Stanford Achievement Intermediate II VOcabulary Test as the covari- ant. The F-value was 1.034 with a significance level of .311. It was pg; statistically significant at the .05 level of significance. The null hypothesis 2.19 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.20 No difference will be found between the group mean score of students in one CAP group and the group mean score 133 of students in a second CAP group on the reading comprehen- sion section of the Gates-MacGinite Reading Test adminis- tered in the ninth grade. Symbolically: Ho : M = M 1 2 Legend: M1 = one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Alternate Hypothesis 2.20 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the reading comprehen- sion section of the Gates-MacGinite Reading Test adminis- tered in the ninth grade. Symbolically: Ho : Ml # M2 Legend: M1 = one CAP group mean (Group 4) M2 second CAP group mean (Group 5) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Stanford Achievement Intermediate II Comprehension Test as the co- variant. The F-value was 2.513 with a significance level of .115. It was not statistically significant at the .05 level of significance. The null hypothesis 2.20 cannot be rejected. The .05 level of significance was not reached. 134 Null Hypothesis 2.21 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the spelling section of the Differential Aptitude Test administered in the eighth grade. Symbolically: H : M = M o 1 2 Legend: M1 = one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Alternate Hypothesis 2.21 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the spelling section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Ho : M1 # M2 Legend: M1 one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Score were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II Spelling Test as the covariant. The F-value was .676 with a significance level of .412. It was not statistically significant at the .05 level of significance. 135 The null hypothesis 2.21 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.22 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the language usage section of the Differential Aptitude Test administered in the eighth grade. Symbolically: H : M = M o 1 2 Legend: M1 = one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Alternate Hypothesis 2.22 A difference will be found between the goup mean scorecflfstudents in one CAP group and the group mean score of students in a second CAP group on the language usage section of the Differential Aptitude Test administered in the eighth grade. Symbolically: HO : M1 # M2 Legend: M1 = one CAP group mean (Group 4) M2 second CAP group mean (Group 5) Significance Level: .05 Scores were reported in percentiles Analysis of covariance was used with the Stanford Achievement Intermediate II Language Test as the covariant. The F-value was .058 with a significance level of .810. It 136 was not statistically significant at the .05 level of sig- nificance. The null hypothesis 2.22 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.23 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the verbal reasoning section of the Differential Aptitude Test administered in the eighth grade. Symbolically: H M = M o l 2 l — one CAP group mean (Group 4) Legend: M M2 - second CAP group mean (Group 5) Significance Level: .05 Alternate Hypothesis 2.23 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the verbal reasoning section of the Differential Aptitude Test administered in the eighth grade. Symbolically: Ho M1"!“2 Legend: M1 - one CAP group mean (Group 4) M2 - second CAP group mean (Group 5) Significance Level: .05 Scores were reported in percentiles. 137 Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F-value was 5.541 with a significance level of .020. It was statistically significant at the .05 level of significance. The ninety-five percent confidence intervals for the means of Groups 4 and 5 for the reading comprehension sec- tion of the Stanford test were shown in Figure 19- The in- tervals are very similar. The means of both groups are very similar: 64.43 Group 4 mean Group 5 mean = 64.60 The t-test was used to statistically test the differ- ence in the group mean Stanford Achievement Intermediate II Reading Comprehension scores between Group 4 and Group 5. The t-value was -.050. The two-tailed probability level of .960 was 223 statistically significant at the .05 level of significance. The t-test was also used to statistically test the dif- ference in group mean Differential Aptitude Test verbal reasoning scores between Group 4 and Group 5. The t-value was 1.605. The two-tailed probability level of .109 was pg; statistically significant at the .05 level of signifi- cance. The ninety-five percent confidence intervals for the means of Groups 4 and 5 for the verbal reasoning section of the Differential Aptitude Test are shown in Figure 23. 138 -------- 5------ ----------- --------------- 4--------- 50 52 54 56 60 66 68 70 72 N Mean 95% of Confidence Intervals Group 4 87 63.25 57.89 to 68.62 Group 5 87 56.92 51.02 to 62.82 The area of overlap of the confidence intervals in Figure 23 was reduced from the area of overlap shown in Figure 19. The groups had shifted from being similar to being not so similar. The null hypothesis 2.23 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 2.24 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the English usage sec- tion of the National Educational Development Test adminis- tered in the tenth grade. Symbolically: Ho : M = M 1 2 one CAP group mean (Group 4) Legend: M1 M2 = second CAP group mean (Group 5) Significance Level: .05 139 Alternate Hypothesis 2.24 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the English usage sec- tion of the National Educational Development Test adminis- tered in the tenth grade. Symbolically: H : M1 # M2 0 Legend: M1 = one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II Language Test as the covariant. The F-value was .057 with a significance level of .811. It was p93 statistically significant at the .05 level of sig- nificance. The null hypothesis 2.24 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.25 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the word usage section of the National Educational DevelOpment Test administered in the tenth grade. Symbolically: Ho : M1 = M2 140 Legend: M1 one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Alternate Hypothesis 2.25 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the word usage section of the National Educational Development Test administered in the tenth grade. Symbolically: H 0 M1 # M2 Legend: Ml one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Scores were reported in percentiles. Analysis of covariance was used with the Stanford Achievement Intermediate II Reading Comprehension Test as the covariant. The F-value was 5.777 with a significance level of .017. It was statistically significant at the .05 level of significance. The information about confidence intervals for Groups 4 and 5 was provided in Figure 19. The ranges were very similar. The means were very similar: Group 4 mean 64.43 Group 5 mean 64.60 The ninety-five eprcent confidence intervals for the means of Groups 4 and 5 for the word usage section of the 141 National Educational Development Test are shown in Figure 24. 48’ 60 62 N Mean 95% of Confidence Intervals Group 4 87 52.82 47.03 to 58.60 Group 5 87 47.10 40.94 to 53.27 The mean of Group 4 was just below the top of the con- fidence interval for Group 5. The mean of Group 5 was just above the bottom of the confidence interval for Group 4. The null hypothesis 2.25 cannot be accepted. The .05 level of significance was reached. Null Hypothesis 2.26 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: H : M = M2 0 1 Legend: M1 = one CAP group mean (Group 4) M = second CAP group mean (Group 5) 2 Significance Level: .05 142 Alternate Hypothesis 2.26 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. Symbolically: HO : Ml # M2 Legend: M1 = one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Gates-Mac- Ginite Reading Test, vocabulary section, administered in the seventh grade, as the covariant. The F-value was 1.934 with a significance level of .166. It was 22E statistical- ly significant at the .05 level of significance. The null hypothesis 2.26 cannot be rejected. The .05 level of significance was not reached. Null Hypothesis 2.27 No difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the reading comprehen- sion section of the Gates-MacGinite Reading Test adminis- tered in the ninth grade. Symbolically: HO : M1 = M2 143 Legend: Ml one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Alternate Hypothesis 2.27 A difference will be found between the group mean score of students in one CAP group and the group mean score of students in a second CAP group on the reading comprehen- sion section of the Gates-MacGinite Reading Test adminis- tered in the ninth grade. Symbolically: H 0 M1 # M2 Legend: M1 one CAP group mean (Group 4) M2 = second CAP group mean (Group 5) Significance Level: .05 Scores were reported in grade equivalents. Analysis of covariance was used with the Gates-Mac- Ginite Reading Test Comprehension Section, administered in the seventh grade, as the covariant. The F-value was .144 with a significance level of .705. It was pg; statistical- ly significance at the .05 level of significance. The null hypothesis 2.27 cannot be rejected. The .05 level of significance was not reached. Research Question 3: _Hypothesis and Analysis Research Question 3: is there a difference in the effect of CAP for students of varying reading abilities? Grade Equivalents 11.6 11.4 11.2 11.0 10.8 10.6 10.4 10.2 10.0 9.8 9.6 9.4 9.2 9.0 8.8 8.6 8.4 8.2 8.0 7.8 7.6 7.4 7.2 7.0 6.8 6.6 6.4 6.2 6.0 144 GRAPH 4:2.1 Grade Equivalent- Group Means on 2 tests CAP groups Vocabulary 1,9.5 (3) ,9.3 (4) / _,9.0 (2) /(,8.8 (5) 7 Gates Vocabulary 9 Gates Vocabulary 11.2 11.0‘ 10.8 10.6 10.4 10.2 10.0 9.8 9.6 9.4 9.2 9.0 8.6 8.4 8.2 7.8 7.6 7.4 Grade Equivalents 7.2‘ 7.0 6.8 6.6 6.4 6.2 6.0 808' 8.0. 145 GRAPH 4:2.2 Grade Equivalents-- Group Means on 2 Tests CAP Groups (2,3,4,5) Comprehension .10.6 (3) ’I I ,/ 1.10.2 (4) ,///.10.0 (2) /4(6 (3 and 5) Comprehension 9 Gates Comprehension 146 GRAPH 4:2.3 Percentiles--Group Means on 3 Tests CAP Groups Comprehension 80 78. 76' 74- 72 70‘ 68 66 64 62' 60 58 56 54_ 52 50 48 46 44 42, 40 38 36 34 32 30 \\.47.8 (4) (3) (2) (5) 6 8 710 Stanford DAT NEDT Comprehension Comprehension Word Usage Percentiles 82 80. 78 76 74 72 70 68 66 64 62 60 58 56 54 52 50 48 46 44 42 40 38- 36 34 32 30 147 GRAPH 4:2.4 Percentiles--Group Means on 2 Tests CAP Groups Spelling .58.7 (5) 52.1 (3) \ ~\ ‘.48.9 (5) w \43 (4) \‘\-.4o.7 (2) 6 8 Stanford DAT Spelling Spelling Percentiles 82 80 ‘78 76 74 72 70 68 66 64 62 60 58 56 54 52 50 48 46 44 42 40 38 36 34 32 30 0 148 GRAPH 4:2.5 Percentiles-—Group Means on 3 Tests CAP Groups Language - 2.4 (4) ,_61.7 (3) (4 ‘ 5am // (3) .4 .2 (3) (4) . / (5) .>.42.5 (2) '.43.3 (2) 6 8 ’10 Stanford DAT NEDT Language Language English Usage 149 Null Hypothesis 3.1 No difference will be found in the effect of CAP on students of varying reading abilities. No aptitude treat- ment interaction will be evident in the regression line drawn for each CAP group. Alternate Hypothesis 3.1 A difference will be found in the effect of CAP on students of varying reading abilities. An aptitude treat- ment interaction will be evident in the regression line drawn for each CAP group. The Gates-MacGinite Reading Test, vocabulary section, administered in the ninth grade, is the dependent variable. The Gates-MacGinite Reading Test, vocabulary section, ad- ministered in the seventh grade, is the independent vari- able. The regression line for each CAP group is shown in Graph 4:3.1. The linear trend is positive. The indication is that large values on the independent variable (Gates 7) are accompanied by large values on the dependent variable (gates 9). If a student received a high score on the Gates (7) vocabulary test, s/he would probably receive a high score on the Gates (9) vocabulary test. Likewise, a low score on the Gates (7) vocabulary test would probably be accompanied by a low score on the Gates (9) vocabulary test. The null hypothesis 3.1 cannot be rejected. The trend is linear and positive. 150 Null Hypothesis 3.2 No difference will be found in the effect of CAP on students of varying reading abilities. No aptitude treat- ment interaction will be evident in the regression lines drawn for each CAP group. Alternate Hypothesis 3.2 A difference will be found in the effect of CAP on students of varying reading abilities. An aptitude treat- ment interaction will be evident in the regression line drawn for each CAP group. The Gates-MacGinite Reading Test, comprehension sec- tion, administered in the ninth grade, is the dependent variable. The Gates-MacGinite Reading Test, comprehension section, administered in the seventh grade, is the inde- pendent variable. The regression line for each CAP group is shown in Graph 4:3.1. The linear trend is positive. The indication is that large values on the independent variable (Gates 7) are accompanied by large values on the dependent variables (Gates 9). The null hypothesis 3.2 cannot be rejected. The trend is linear and positive. The following chart provides the information for estab- lishment of the regression lines. Groups 2,3,4,5 H Ix I» Significance of r coefficient of determination correlation coefficient b (slope) a (y intercept) Gates (7) Vocabulary Independent Gates (9) Vocabulary Dependent Groups 151 .00001 .00001 .00001 .00001 .71 .50 .76 .57 .62 .38 .70 .49 .79 .91 .73 .79 l‘l‘v-lm O O MNQ'M (\V‘NO O \Dl‘l‘h NMQ‘I-fi Significance of r coefficient a of determination A correlation coefficient b (slope) a (y intercept) Gates (7) I“ Vocabulary Independent Gates (9) I» Vocabulary Dependent Groups .00001 .00001 .00001 .00001 .78 .60 .66 .43 .75 .57 .85 .72 .85 .67 .74 .83 Nmmxo O O (")me OWNW O Gl‘ml‘ 10.0 10.0 10.2 9.8 NMQ‘I-fi 152 GRAPH 4:3.1 CAP Groups 2,3,4,5 Regression Lines Gates (1)--Gates (7) Vocabulary GATES (7) VOCABULARY .5 1.5 2.5 3.54.5 5.5 6.5 7.5 8.5 9.510.5 11.5 12.5 13.5 13 w m m ...a D 3 Group 2 = .79 8 Group 3 = .91 > Group 4 = .73 -. Group 5 = .79 ‘31 m m 5 o 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 GATES (9) COMPREHENSION 153 GRAPH 4:3.2 CAP Groups 2,3,4,5 Regression Lines Gates (9)-~Gates (7) Comprehension GATES (7) COMPREHENSION .5 1.5 2.5 3.5 4.5 5.56.5 7.5 8.5 9.510.5 11.5 12.5 13.5 .85 .67 .74 .83 Group 5 O 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Nullj Hypothesis Rejected Significant at .05 ANCOVA Significance of F ANCOVA F-VALUE Significant at .05 ANOVA Significance of F ANOVA F-VALUE Significant at .05 t- Probability t-value Groups * = CAP Hypothesis NO NO .919 -.102 43 52 .373 NO NO .801 .959 NO -.052 43 52 *2 6 1.1 (GE) 154 NO NO .961 -.049 43 52 .575 NO NO 0 316 NO .934 .083 43 52 *2 6 (GE) .174 NO NO 1.879 -2.263 .024 YES 43 52 .295 NO NO 1.111 NO .897 .130 43 52 NO .452 NO .570 .844 NO .197 43 52 .595 NO NO .284 -.201 43 52 .336 NO NO .934 .803 NO .249 43 52 155 om- o.- -o- o: 6: moo. Hoo.m oa o-mo. momo. o: moo. ooo.- oom m.o.m.~. «H.- om- o.H -o- mm» mm» moo. ooa.m on mooo. oomo. 6: moo. moo.» oom m.o.m.~. m-.H on o; $- 0: 6: m-m. moo.- o: moss. oomot- 6: moo. ooo.- mom m.o.m.~. NH.- om- o.H -w- mm.» mm» 8o. 28.: 8» Nooo. 331.2 mm:- moo. «motor mom "14.2? 2.- oma o.- -w- o: 0: Hoo. ommm. 0: Homo. oo-o.- nx- Hod. moo.- mom m.o.m.~. o-.- mm 6 -mo- 6: o: mom. ooo.- oc oH-o. moon. 66s ooo. oo-.omu oom m.o.m.~. o.- . mm o -o on on mom. ooH.H oc Hooo. mooa. 66> ooo. ooo.-H- oom m.o.m.m. o.- Nm o 5o- oc on moo. ooo. on mooo. Hoom. mm» ,ooo. ooo.o~- oom m.o.m.~. o.- mm o -o- o: 6: ooo. oo-. o: oooo. mono. mm» ooo. oo~.oau oom m.o.m.m4 o.H lsd t5 AeF AE t5 AeF AE t5 . Y e N SM 8 lie n0 Vc VU n0 Vc VU n0 tt u P i ust a. Onf 0L a. Onf 0L a. .1 1 DC 5 N80 C Cao CA C Nao NA C 1 a 0 e he it NC NV it AC AV it i V r_- h tj fa Ai A- fa i - fa b - G t 08 .1 f F i f F i a t ..- 0 PR n i n i n b P Y 9 n g n g 0 Y H i g i g .1 r H S .1. S .1. S P S S 156 m H- m :11 LG H- :3 W o to M 'O P: :3 ”*0 o » m r6 P- m o ('1' I 3’ H-> m H: u.“- U M Q <32 0:: rfP- mzr m H >(3 0 or) o o mzz m (30 r*o rhoco - m ncnc P- 5:: C2< O‘C co: m Hwa m rap z 6:» 'ua>> mr+ Cbmld 2.1 *2 43 .507 .478 no no (%) *3 87 2.1 *2 43 .644 .424 no no (GE) *3 87 2.2 *2 43 1.218 .272 no no (%) *3 87 2.2 *2 43 1.824 .179 no no (GE) *3- 87 2.3 *2 43 3.107 .080 no no (%) *3 87 2.4 *2 43 .113 .738 no no (%) *3 87 2.5 *2 43 .037 .847‘ no no (%) *3 87 2.6 *2 43 .564 .454 no no (%) *3 87 2.7 *2 43 .194 .660 no no (%) *3 87 2.8 *2 43 .245 .622 no no (GE) *3 87 2.9 *2 43 6.289 .013 ‘yes yes* (GE) *3 87 157 :1: ‘< E * 71 6 5‘ II G) <2 (1) H >0 :2. 02 62 m gm: 2 u!» 2.10 *3 87 (GE) *(4,5) 174 21°465 2.11 *3 87 22.149 (GE) *(4,5) 174 2'12 *3 37 22.149 (%) *(4,5) 174 2-13 *3 37 33.766 (2:) *(4.5) 174 2.14 *3 87 30.88 (2s) *(4.5) 174 2.15 *3 37 40.109 m *(4.5) 174 2.16 *3 87 23.129 (%) *(4.5) 174 2.17 *3 87 1.63 (GE) *(4.5) 174 2.18 *3 37 11.781 *(4,5) 174 (GE) m P- m a H- m :3 Q *< P- :3 ”'0 ft P: m o H-> turn u-n o z r+H~ m:r 0 wt? (1 ram 2 H15'4,5 3 = 4,5 3 :> 4,5 3 >-4,5 = similar effect > greater effect < less effect Group 4 and Group 5 Comparison. Group 4 and Group 5 each participated in CAP for two years. Group 4 was the third group to participate. Group 5 was the fourth group. 1. Statistically significant differences a. There were conflicting results as to whether there was a difference in the effect of CAP on Groups 4 and 5 on reading ability (comprehension), as measured by stan- dardized tests. Reading comprehension was under consideration in hy- potheses 2.20, 2.23, 2.25, and 2.27. Null Hypotheses 2.23 and 2.25 were rejected; Null Hypotheses 2.20 and 2.27 were not. The posttest in both hypotheses 2.20 and 2.27 was the Gates-MacGinite Reading Test, comprehension section, admin- istered in the sixth grade. The covariant in hypothesis 2.20 was the Stanford Achievement Intermediate II, Reading Comprehension test, administered in the sixth grade. The covariant in hypothesis 2.27 was the Gates-MacGinite Reading Test, comprehension section, administered in the seventh grade. 173 Sig. Sig. Null Hypothesis . Posttest F—value _of F ,at.05 .Rejected 2.20 Gates (9) 2.513 .115 no no Comprehen- sion 2.27 Gates (9) .144 .705 no no Comprehen- sion The posttest in hypothesis 2.23 was the verbal reason- ing section of the Differential Aptitude Test administered in the eighth grade. The posttest in hypothesis 2.27 was the word usage section of the National Educational Develop- ment Test administered in the tenth grade. The covariant in both cases was the reading comprehension section of the Stanford Achievement Intermediate II administered in the sixth grade. The 1973 edition had been used for both Group 4 and Group 5. Sig. Sig. Null Hypothesis Posttest F-value of F at.05 Rejected 2.23 DAT (8) 5.541 .020 yes yes Verbal Reasoning 2.25 NEDT (10) 5.777 .017 yes yes Word Usage Graph 4:1.6 shows the relative positions of Groups 4 and 5 on the four reading comprehension tests. Gates (7) is not included in the graph because scores were only re- ported in percentiles. The path for each group is displayed. 174 In the sixth grade, the group means were very close. In the eighth grade (next test shown on the graph), the group means of the two groups were separated. The gap re- mained about the same in the ninth and tenth grades. The two groups started out very similar in the sixth grade and ended up dissimilar in the tenth grade. Conclusion: There was a statistically significant difference found, as measured by standardized tests, between the effect of CAP on the reading comprehension ability of students in Groups 4 and 5. CAP was a more effective program, in the area of reading comprehension, for Group 4 than for Group 5. b. No statistically significant difference was found, as measured by standardized tests, between the ef- fect of CAP on the vocabulary ability of students in Groups 4 and 5. Vocabulary ability was under consideration in hypothe- ses 2.19 and 2.26. The null hypotheses were 22E rejected. The posttest in each hypothesis was the vocabulary section of the Gates-MacGinite Reading Test administered in the ninth grade. The covariant for Hypothesis 2.19 was the vocabulary section of the Stanford Achievement Intermediate II adminis- tered in the sixth grade. The covariant for hypothesis 2.26 was the vocabulary section of the Gates-MacGinite Reading Test administered in the seventh grade. Hypothesis F-value of F at.05 Rejected 2.19 1.034 .311 no no 2.26 1.934 .166 no no 175 Conclusion: There was no statistically significant difference found, as measured by standardized tests, between the effect of CAP on the vocabulary ability of students in Groups 4 and 5. CAP was equally ef- fective in the area of vocabulary for Groups 4 and 5. 2. No statistically significant difference was found, as measured by standardized tests, between the effect of CAP on the spelling and language abilities of students in Groups 4 and 5. Spelling ability was under consideration in hypothesis 2.21. The null hypothesis was not rejected. Language ability was under consideration in hypotheses 2.22 and 2.24. The null hypotheses were not rejected. Hypoth- Sig. Sig. Null esis Covariant Posttest F-value of F at.05 Rejected 2.21 Stanford DAT (8) .676 .412 no no (6) Spel- Spelling ling 2.22 Stanford DAT (8) .058 .810 no no (6) Lan- Language guage Usage 2.24 Stanford NEDT (10) .057 .811 no no (6) Lan- English guage Usage Conclusion: There was no statistically significant difference found, as measured by standardized tests, between the effect of CAP on the spelling and lan- guage abilities of students in Groups 4 and 5. CAP was equally effective, in the areas of spelling and language, for Groups 4 and 5. CAP appears to have had a greater effect on Group 4 than Group 5 in the area of reading comprehension. The ef- fect of CAP appears to be similar for Groups 4 and 5 in the areas of vocabulary, spelling, and language arts. 176 Effect of CAP Reading Language Comprehension Vocabulary Spelling Usage 4>5 4:5 4:5 4:5 = similar effect :> greater effect Summary for Research ngstion 2. CAP appears to have had a similar effect on all CAP groups in the area of voca- bulary. Test results in the area of vocabulary ability showed no significant difference among groups. In the area of reading comprehension, CAP appearS‘to have been the most effective with Group 3. The effect ap- pears to have been greater for Group 3 than for Group 2 and Groups 4,5 combined. CAP appears to have had greater effect on Group 4 than Group 5. In the area of spelling, the effect for Groups 2 and 3 appears similar. The effect for Group 3 appears greater than for Groups 4,5. The effect for Group 4 appears greater than for Group 5. In the area of language, the arrangement of effect of CAP on the different groups appears similar to the effects described for spelling. Summary Table: Effect of CAP Reading Language Comprehension Vocabulary Spelling Usage 2‘4,5 3=4,5 3>4,5 3>4,5 4'>'5 4 = 5 4 = 5 4 = 5 = similar effect :> greater effect <1 less effect 177 Research Question 3 Is there a difference in the effect of CAP for students of varying reading abilities? Regression lines were drawn from scattergrams. The Gates-MacGinite Reading Tests administered in the seventh grade were used as the independent variable. The Gates- MacGinite Reading Tests administered in the ninth grade were used as the dependent variable. Scattergrams were created for 2323 CAP group using vocabulary and comprehen- sion scores. The regression lines were plotted on a graph to show the trend of all four CAP groups. The regression lines were positive. The trend was linear. (It appears that the CAP program did not have a differen- tial effect for students of varying abilities. If a student received a low score on the seventh grade Gates test, s/he probably received a low score on the ninth grade Gates test. Similarly, a high score on one probably meant a high score on the other. This is the usual case if an aptitude treat- ment interaction is not occurring.) Implications for Traverse City Public Schools It appears that there was no difference between the effect of CAP on students, as measured by standardized tests and the program it replaced. Four reading and associated ability areas were analyzed--comprehension, vocabulary, Spelling, and language. There was no statistically 178 significant difference found in any of the four areas. This study dealt only with the effects of the program that could be measured by student performance on standar- dized tests. That criterion is recognized by the public as a legitimate criterion by which to judge the effectiveness of an educational program. The outcry over apparent test score declines is an example of public acceptance of stan- dardized test scores as evaluation tools. The CAP program appears to be as successful in the area of student test performance as the program it replaced. If the school district was satisfied with the test perfor- mance¢m9 fldfiHHZEKHHEZW wmmm mNH coma mHHN mmmv 5mm mmo wan Han hum Hhma maoonom owanom mmu< >uwu mmum>mua "humfifism powuumfio Hmooq mavm ONH Hmwa Nmma mvmv are mmm mmm mmo ovm onma "wousom Huuoe ommm “mum Hmuo mam oH o m h o mam» \qm>mq 183 APPENDIX B METHODS OF PROGRAM EVALUATION AND SOURCES OF EVIDENCE .. , IL. erWDW ... I: a ‘1, # ..1,_ 1.. . .11...ng i -1- 1.1 1 ..1 .11. . .1 117.111-... . L .3! 1.. .1443}. .H. ...4 .. gm 1.; .1... .. l1- .. _H. Mr.“ 1.111%; ..m. F17. 1 .- 1.1.. 1; m..- ... u I. .I S V d .5 S n) S 3 1 .u m, D n i. s a n n n. u mu x J n c. u . .o J S 1 vl 1 P v ..I I. .. I. I u S u . on, s .4. u. I a 1. u .l ) . 1 ..I v J 1 d u U 0 1| la 5 cl 0 H. .J 00 U Q- n 1 D J 9 in V .1... n, I W. I .J . Q n n 1W «.1... 1S:- J Wu h J J v u n N I M U I p n C \I 6’ U 1533.42.— .n 7.35::- c-E _suczom .N H.313.— .— -:..-._-7 E:- 1:5:3. .3 sow-8?..— .--?:_..r. a. ...if-Sr. .n £12212 .N :51. .— 12373. 153.512.: .5...- .=-:3 133:. 7.5-S...- .,..-_x.--:7. oat-2.9:... 3. cam-2.: 5 3.3V .N :2... a--_.zn< .— -¢.C 7:553...— 53: ...-:_-.x_:-:. a. .3...- as- 5 its: .n 2:55 .u Sass-etaczr-c- .- 95.... :5» ...a-us:— -.:_-C .u 393.}. .s 2532?: .n .5282: .N 2.: zip-5.53- .32326-35 ...a-33:.- .-...- 3:: 356 .u taro-J. A 15335:. .a Gas-Ga...— .— 3.82 .2 .< -:-:__-..-m-- 522:... :52... 251-3.... 2 Baa-:5. c._. .55....- -:_--2_-a.=-: 2.3:.— .l I \ saw-3.5.5 Emma-i be 35:37. .2236 was Panic-..— ._ 635—. 184 185 J1 . 1.11. ..1. . ..... .I... 1 . P. H .n ..:... . 1. .. .v. .c . 1.11.1.1 v .H .. ... . w .u. . . . ..u :Iluoll... 1.1.1.!1E11 . .. . . b «no: ..aalnl‘ull o-| — . . I! ... Q "J .uu. 31.1.3.1..4 848173.11”... m .1 .0: ~ H. .II . H. .1‘ .\ c n . ..U U H . . ..ns 0’ Q Q. Q . . . .n ‘0 .4 ’- Lo .. Pu n. u .. 7 .. 1 a a I '1 SI " U I ll 1‘ I r‘ J ’ J — II .J I) r I“ . ..- J on I. I. u .l‘ 1 n .I I“ u .. I. . It. . ’0 S C“ '.u . — I C. . . ‘0 .l .l. .C 0! No I. .U ’0 u . ’l. v . I I J «0 ..v. [I .5 . z 14 .I n . ..u I .J J z x J . L I! “b. . _ .10 .Oro .. .1“ i U I.‘ l. — I) ( O... 0a s 5‘ n, u l “h l . I I. 0 . . .1. a. _ . 1 _ 1 ... v _ .. H _ .Ifll. . ) I 4 II . u . h r. 11.. ... A. . l. . to. U h .| . .fl . u .M . .M vi n . ”IF . I . _ J . .1 n n . ‘ x .u . _ u— 4 1.. . .b 4- :6 11171 .I . b-1151... oll. Q3 .110 1 I. .1 1 . 1 2:329:95 .n 1:55.: .5: 3.5:...— .m .25....— .— 9595; 1:: 152:3— ...— nr..:..>:....=.. :. :35... E .30 .n «.9»: 2:? z... 3 noisier. 3:15.53 3. :35... 5 =53 .N 7.3.. 2:322 .- 330 . a.) £95.. 3.7 .353.— .5259:— d 3.5:; .67. .3233: 1:555. .0 5.2.5....— N Erzéczm .— 75: 3.2.2.5.: :. 3.594....5... .330 .n 55¢ .9 335m A. 7.3.3:...— .5 .375...— .N EEC .u iii}. .a _E:.._.:?._ .a 5.837.; ._ .51.: M—__.::==C .< 2.: C» van 53:55.23... $53.3: ..::_:..:._..:...~: 5 ...Cfizmaxs .::.:a::_:=3 Effici 5.1: E3712. 2 2:21:59 «:1 :5:::::5 _ 931,—. 186 . ...... j”. . 9:35.33 aim .c .515... .- U -14 ......5......=. :.....U .... E .... . .. .. ...... .. ......3? .....:.........:...x .v .... 1.” ...............«m. .22.... ....p. u....§.. .... , .4....... .. -1. ...x.......: ............. ... 5......33...=..._ .N 1.11.1 1. .... 1. :... ......z::::. :..... . ... .55.... .. ..m...:.x.......: 53.2....— .0 3:51....” .....m ... ...a... -11..1 1.. . 3:53.52. 5.23 .... 1, .....-1 -1 31.11. 3:52... ... .1::.29.¢..»_ .3 . ..1 M 1.1.1. -..-.2111. ....ca........a .59.»... ... m...........n. -- . . .. H. 1...... :..:O . ... 3.5:.er .h :12! .... 11.. 5.32.1.5... ... ....r............5....._ ... 1 1 - , .. s w. . 5...... ... ...........u.....x. ... .........>.....x .... - A 92.3.2... ... 3.3.?— .v - 1- 11. 2.2.2.... .............- ... I-1..n-1--.mfl..r..__-1-1.-1 1. . 5......3: ... 5.2.5.3 .~. .111. ...-1 LL11 11 éfilz.1 1.1”“.-1 . 5* :..»... ......z 533...... ... ...—3.....2 .. . ..... 1 ...-I1 1.1-1 1. -_ 11! . 23...... 5:55... ... ..- H. EH11 .1... .........._....... =37. .... -...1. .11 .................. :.....”v .v 1 u -1 1. 1 1 1.1.1 .53.... .... ........$.....:.. .... . .. . ................. ............ .u . ......R .. 7.5.... 3......3... .... ......z ...... 3......5’ .- m............... 552...... .< :2... .51...: 5.2.4:... 5...... 2:33.... ... 8.3.3:... 3.. .... .9... .1.1111111. ...511 .| 1‘11. Iszl'll I .1111... I. 1. I411! l I .- :. ...) r19 .. .... \. \) x... .... l l ’ v. ‘ \ I ( l l D. .I-( . _ 4... ll- ] 9 I.. .... J ... l U _l 4 \A . :0 _.\ 0| p; \: ...: ‘ I .I ' 1' 1-.....- an}. ......“ ...... .... .....u..-..;... ......11. ._ u S - .\. i. u .1 ...5. u. .\. . nut. .... U . a. OI 0‘ . I . . _u...mi..n....u .r. .4. m_ . I .l v. ..u "I “J J n. _ \ . H: .c —.7.I ’ val ..u ”a I — 0 I“ J. — . U .1. .\. J .1 —.c U . .M u .o HUI — x \ - I . I'. l , . .p .. t! u . \I. _ v- \ ll . w 4. _ ...; ...! 1._ I. . .A” _ . rm. _ N. J 't u .00 m U. _ .... .... . n. ....- U. . I; . u. 5! _ ...: ...... .... . . _ . _ ' .11.! .II'I. i... ll I Ar D. fill. 1'11: .0 ll -.. 1-55 27..."...23; _ .....a —. 187 ,- -.lf ........ . A» .. . h u ‘ a: |.< ““7! *..LM i ., ...} .1. ...... 5a., a" H.-. .. h ‘. w .‘ flu... flu... H ... ._ .2. , .... 7...... ‘LV‘IIL,1. - A j? E , ..H “..l .....x.“ . trump.» .. . . U L» v hymn-w” III-.. . wnl O H. n) .‘I To .5 u. ...u S S .J .1340 «J to I U I 7 4\u :.. ‘4. n H” 0 -. ”H A u! c. I 0,. s s 1 a r ... ., .0 ~ t ' 5‘ OK n) ' ‘ . a ‘. J J 1‘ 1 I. no r . A. . 0‘ I ‘ .4’ u «\J Iv. U .1 s n!“ .t "I. a II .I n. ..o r. r 1. u L J. . 10 l l’. o: . “t n ‘1' m n u‘ U 9‘ I n ) RI .‘ ..A .3 .‘ I. a ( . . . ... u ... : .. 0. . A v u n as. .3 .17. cl I .l ) .\. 9. .5 .4, x. u d .... m a .u b .6 a v J a! .0 It 9: A. 40 '6 p I‘ '0 06 n we u o I c I .08. I . I l r,!“.lY'I'. ‘bltt C-llp 1.. I- 1 a. IIIIAIILW .l --- .II.-I' -IvI- ....r...:..cc: . .....sF ...-5.3.2:.— .3 .... 5......— ..:s .3331... ..— .s..:....- .< 95......» :..: .5 ...—“.23.. 2:23.»... 3.33.»... 55.... ...P :.. .7 ...... ...:tzarx 5.22.3.5". ...:z...:.:.. 2.. 55.5.5...— ... :3: 37¢. .N 23.523345. .n :.. .7 ...... 52.3.93...— ..:.:.....:u. ......z..:::.. .... ......a...a>m— G ......E:: 9.49.3.5 A. :E.._.....=... ...... 9.3.55. .3 :3. .N £322.33. .0 >3.» .. ...—:.....zmn— .1 r... .. ...:t...:....>.... .u ...." z ...........,.=... ...... .53....9m .A 5.55358 .3 5.5.5 .- ..a... ...... ...... ..........x. 22:35.... .... 3.2.55... ......1..7..:.:..§ .53...— .h . :.....m: .28 3:23... .53.... d ...a... ...... €5.33 .33....— .... 25.... 2.3.3.2 9.3:.— .v z.....£.....=..: ...a 6.2 RSV. .n viz—55.... :.....EE...=_.5 imbue... .N 3.1%.... .8.....:.=... .....E=_...=€< .— 3353 Sinai ... ..p— 188 2.9.3:? ...—Each. r . P H . . 5.3.1.5: .3 .............3...x. ...... Size... — m“. :.u a... ... 5...... . ... pins—5...... $5.52.. 388......5 .< 3 Brian 5:2: 5. 95:2. :..... 3:25.43.— .E 32:... .555...R.xu ...—......3 .0 8:5... 38..-. .I E sumnrmq aauvqf) Wm a)“ umwdo 112th? [.5 J 7 0) an ..H xw wa w .o ...m n. D.) :1. J vu U. S so U1) .4. 3:»... ... .5 m m. u.n.l 5, a .0 no a u um n w am a a m. W m rum So. m of SP 0” .. w ”J ...c..am..mo$._ ... €9.32 .3230 .5223 a... 5.3 23.23.... ...—283......— 85......“ .c .855. .25 8...? ... 3.95:5 .N 92¢. .>m......x:.5.:. 55.5.95 8.55:: ....— .....XJ. 1:53.332: ...........:_c_. ram.— 2.5.3.5.... .a.....2.....:._ .... 5.x... ...... 3.5:...- 3.2.3.. ”— .. C .U m < ’.....,..x:.. .73.. ... ........:a7...x.:: ...: ... 23.225... :.P .U .fl .< ...::.£.....: :..: 3 5.5.2... 2:12... .3.....:.E ...-.59 Oh. .> 189 '1 F1 ~5- H . H --..J._. H ..... ...- r—-‘ mm. w..- ID .3: ti-.. .---..»H..u-...-..-._....._ qutJ W m _ Hr... w... . .AMH.H.H.1.....-..H: HF. PM”. Q :.4'!‘ 3 .ydl I- ll|.l*. . Vo'll . i‘uflotl‘..‘ 4.1!. v Fina...4l . F .tw L fl.-w...i_!§..li .----..-...,..-.m.wfh til... -..... .IIH__..-....}.:m.M..‘ -...iw-_..w:.2.. :..... . . a -«nq---.-.n....£...j.. ..... 21...... ll: ....-.» Ir rrhrlrulilr Still-’5‘. !. -- lu-:..l ! ..ri-.u| . W ...- i. 32.3.9. 3.2.5.... .......:........... ...... :.....3... a ..n.......... ... .........x..... .....2......:... z .... ................:Ec....u. é. .........J....... :.....79: ...a......n..v. 95:63.». ES... .... ... .. ...w. .......:..... .... 37...... :..... ......E..:.m ... 5...... ......z...£:.. .c .7... .u........... .u.......:... .9 4. ...... .... .....x... ... 35¢... ...a .......é...é= 5...... ... 72:52.... ...a... £233.33... ..........:............... .2593... ... ...............m.x. ... I :..... ..........—..:.s.uc... ”2.2:... .2556... n...3...... .... :5..." £5.22... ...... 5.3.5.. ... 3.2.5:...2 ... 2.5.... ....,..x......: ... 1.33.1.5. .< 3...}. {1.7.3.2. ... .... . ......z.... .........._....C.v .: t 5...... .32.. ......n... .... ":....23 :.............¢.-.. .. ............... ... :..... ...... .r... 3...... .5. .. ...: ............. ...:x... :..... . z .........x. .......:......:.U .... .. ................. :..... :..... :..»... . 7:5... .... f..............,... r :..rdu... Z. ...... ... ......u.‘:.. .3733—12t..:m:.~:=‘.v.< ......» 2:57 .........-.......3..v 5.5.9.. .... c. .569... 5.. ...... e387... 2...... .... .3352? ... 53.3.3... .. 3.5.3.5 .U . 25......1... 5:895:25 Embraa ............. .... 9.23.5.9... :..... ... ......Enfiv .n 5.5.3... o... O. ...—3:370 .... ...... .... ......c ......a.........x. 3...... .... €936 .< ......7 .....8.=........u..nm:~v a 3.2.8.51... 3.8:... .... ...a“... .... a...» ......» 5...... . .... r. 5.... .........:... ......r......... .... 3.5 .U ...a .... 190 :.. 3 1:3 6.2—5...... :2... H w r H ~ H _ a g 55:. 215:2. .9 8:319:26 5.5a r:cv3:< .< >:O—.Em8 ho sat—“.33.? .958:— ._=> 333?. ...EEnz... .3: 3:32 2:235: 5 :..}... .32.? ..O scan—5:35 :5: 22.3 1:375:73 a: tsp—:55 .U ,.. :..; -~1 _ 93:23 253:3175 ... £3.38 _abmxc_:=u=5:2r~ .m r. ”1 hm “-4 r- P—I‘ ic—n 1 1 rd .1 . , - 2.: .1 E... 532.13... ......E..m.EaE m H.”_.mm..m.,_ m ”Whammy . H mmww.MMHHMMLwQ. H 4— .555:...2355v 3.5.935 #:552— 3 £3.52 .< >13: 8.3 5 .8750 .=> 38.78 333.5 W! N -_ M m 4 .3 EB. x .3 5.2%.... 2: 2 it, a: .0 :63“ U ...: 3:1,: 953.5... mu m d _ fl 4 h: 3.3.: LES—.2: aimskza 5.21:»: _c :..-.....— d .1. I. 7. u. S 3 D H 7 o L .fwwm m5 ... m. .0 1m .. u .u 1.4a 1 .... 0.. v u U. A .3 w: 2 ” .m. H m u w w. ,u w fl. \u m .r I .b S u w. w I w m. m. pl m u U m o. m M . ... u .1 ..Do I W m.» .u W ..u u U. ; .0... m m. mb :0 3 .6 \ c l w J. , s» m.» a m [It F I APPENDIX C GOALS AND OBJECTIVES OF CAP PROGRAM fitbzimrahab h>flmdflkum3 TRAVERSE CITY JUNIOR HIGH COMMUNICATION ARTS Reading Comprehension Objectives Grades 7,8) A-l: Recognize and select the detail taken directly from the paragraph that best completesaxsentence. A-2: Select a translation of a detail from the para- graph that best completes the sentence. A-3: Recognize and select the signal words relating to the paragraph pattern. A-4: Select the one response that best expresses the main idea of the paragraph. B-l: Complete a sentence by selecting a detail that is implied by the information given. B-Z: Select a statement that tells the relationship among the ideas in the paragraph. B-3: Select the main idea that may be inferred from information in the paragraph. C-l: Analyze a paragraph and select the question (main idea) that best states the problem. C-2: Select the hypotheses that best fit the situation C-3: Distinguish between relevant and irrelevant de- tails and select whichever is called for. D-l: Select the criterion that is the most appropriate basis for making a judgment. D—2: Select a judgment based on the criterion. 191 192 READING COMPREHENSION LEVEL III (Minimal: Grades 8-9) These skills are minimal grades 7-9 and should be main- tained above that grade level once mastered. If not mas— tered by the end of the 9th grade, the student should take courses that will allow him to master them before graduation Non-fiction: Journal, Essay, Biography, Autobiography, Newspaper Article Fiction: Short Story, Novelette, Novel Drama Poetry is dealt with as recreational reading and as craft- career ed. 1: The learner will recognize and select details taken directly from the fiction/nonfiction/drama material about charcters/setting. 2: The learner will recognize and select the signal words relating to the pattern of the material. 3: The learner will recognize and select signals for alliteration and simile. t‘b'wmtiI-it" 4: The learner will select the one response that best expresses the main idea of the material. 1: The learner will complete a sentence selecting a detail that is implied by the information in the material about characterization and setting. 2: The learner will select a statement that tells the relationship among the ideas in selections of fic- tion/nonfiction/drama. 3: The learner will identify the main idea of the plot that may be inferred from information in the fic- tion/nonfiction/drama material. H(3hier L J v Skill Cycle not successful go to Teacher #2 Reading Skill #1 ------------ ) MT ----- )Reading Skill #2 not successful f”’/’/”/ go to Teacher #3 4 A. D. 200 GOAL STATEMENTS 1977-78 8th grade Comm. Arts Speaking All students will complete one speech by the end of the school year. Reading 1. During the first semester, students will go through one teach cycle of Croft B and Croft C reading skills. Those who do not pass the teach cycle will enter a minimum of one reteach cycle. 2. During the first semester, all students will be taught one cycle of setting. 3. During the second semester, students will go through one teach cycle of Croft C and Croft D reading skills. Those who do not pass the teach cycle will enter a minimum of one reteach cycle. 4. During the second semester, all students will be taught one cycle of characterization. 5. A minimum of one written book report will be re- quired of each student by the end of the year. Writing 1. The 8th grade unit will give common pretests in writing. 2. Areas covered in writing must include: a. Parts of speech (nouns, verbs, adjectives, ad- verbs, pronouns, prepositions, conjunctions) b. Sentence structure (run-ons, fragments, complete sentences; subjects, predicates, objectives) c. Mechanics (commas, underlining, apostrophes, hyphens, quotation marks) All three areas of Communication Arts (reading, writing, and speaking) will be integrated whenever possible. 201 Progress of Program The CAP program began in 1972—73. A team approach was used which included the following: 310 students (7th grade only) 2 teachers 1 Reading Resource Teacher (The 3 teachers involved had a common planning period.) * In 1973-74 630 students (7th grade only) 4 1/5 teachers (common conference hour) 1 Reading Resource Teacher 1 Aide * In 1974-75 760 seventh graders 760 eighth graders 5 teachers (common) 5 teachers (common conference hour) conference hour) 1 Reading Resource Teacher 1 Reading Resource l Aide Teacher 1 Aide * In 1975-76. The Ninth Grade Project was established as a transition between the 7th - 8th grade pilot program and APEX. The structure during this year includes all jun- ior high students enrolled in communication arts. The struc- ture is as follows: 7th grade Teaching Unit (with common conference hour) 1 unit leader 1 Remedial Reading Teacher 8th grade Teaching Unit (with common conference hour) 1 unit leader 1 Reading Consultant 9th grade Teaching Unit (with common conference hour) 1 unit leader 1 Reading Consultant 2 aides assigned to Junior High Communication Arts Depart- ment to work with all grade levels. * 4 k-days of preservice in 1973-74 (7th grade), 1974—75 (8th grade), 1975-76 (9th grade). * lO k-days were provided (to the communication arts depart— ment) each year for inservice. 202 Example of a Skill Cycle OBJECTIVE: By target date teachers will have completed one complete cycle of: a. Teach b. Practice c. Feedback (student) d. Post-test Steps in Planning 1. thN 000 \IO‘U' o o C \Dm C Pre-planning (includes inservice time) of unit committees Set target date for completion Skill group Match up skill group with teacher strengths where possible Work out management system within teaching unit Begin cycle Evaluation during skill cycle by teaching unit dur- ing unit meetings Post-test Evaluation by teaching unit of skill cycle II. 203 UPDATE A. Goal Area: Continuity of Skills There is a sequential list of performance skills* in reading and writing upon which Junior High Communication Arts instruction is based. Students progress through these skills at the rate most appropriate for them but within constraints of grouping management. The skills have developmental clusters which are identified for each grade level. These clusters represent the point at which most students receive initial instruction in these skills. (*The skills list is a staff-developed adaptation, draw- from the K-12 Communication Arts Steering Committee list. The reading skills are based on the Croft Com- munication Reading Skills Design.) B. Needs Assessment --F1nalize minimal list of objectives with grade level clusters. --Solidify commitment to and understanding of common language; 1.e., teach model, grammar, reading. --Further develOpment of instructional techniques and suggested materials for special groupings in certain areas (skill group isolates, enrichment groups, man- agement needs). --Clarification of Teach Model Components C. Resources Necessary --A three year commitment for released time for Inser- vice and Program develOpment meetings on various top- ics (a minimum of six half days). A. Goal Area: DevelOpmental Groupings The unit goal is to place students in developmental groupings in reading based on Standardized Achievement Testing (Gates or other standardized tests) and skill cluster groupings in writing based on departmental skills list and criterion testing. Students are placed at their instructional level based on: 204 l.‘ Reading-~standardized achievement test scores or developmental levels (within a range of three developmental groups) 2. Writin --are placed in skills groups based on criterion testing. B. Needs Assessment --Cont1nue testing program --Work on continuous improvement of management sys- tem C. Resources Necessary --Inservice time --Continuing money for testing correction --One full day session per semester per unit to con- tinuing level skills grouping III.A. Goal Area: Units will operate with a shared decision-making ap- proach. Processes for arriving at decisions will be specified (majority; consensus; etc.) and unit members will be held accountable* for their goals and decisions. * 1. Unit establishes goals through shared decision making. 2. Unit goals approved by department head and prin- cipal. 3. Unit goals approved by director of instruction. Monitoring of implementation of goals is most effective first at the unit peer teacher level, but if problems cannot be resolved at that level, they will be handled through appropriate administrative channels. B. Needs Assessment --Continuing development of teaming skills necessary on the decision-making model. --Continuing development of leadership skills for Unit Leaders. C. Resources Necessary --Four Inservice Half Days. IV. 205 A. Goal Area: Individual student progress in specific reading and writing skills will be reflected in student record cards. B. Needs Assessment --Refinement of cards and system to suit needs for grouping and clarify what mastery "punch" indi- cates. C. Resources Necessary --Continuation of aide service to maintain records and grouping. APPENDIX D DEMOGRAPHIC INFORMATION 1960 33,490 (Census) 1 Change Townships Acme* Blair* East Bay* Fife Lake Fife Lake Village# Garfield* Grant Green Lake* Long Lake* Hayfield Paradise* Kingsley Village Peninsula* Traverse City* Union Hhite Hater Grand Traverse County' and Elmwood Twp. TABLE 34 Estimated Population Grand Traverse County Estimated* 1210. ..__1_9_7_9.___ 39,175 53,468 (Census) 17 .oz ' 36.5% (1960-70) (1970-79) Estimated** 1.91.0. 1979 1,662 2,612 1,677 4,590 3,356 5,726 638 894 (274) (304) 4,917 8,738 507 717 1,206 2,110 1,584 3,111 651 929 1,434 2,544 (632) (1,127) 2,642 3,330 18,048 16,710 57 101 796 1,356 41,415 56,766 Projected 1985 63,055 17.9% (1979-85) % Change 1970-1979 57. 174. 70. 40. (10. 77. 41. 37. 1% * County estimates and designated townships are based on average of three factors: electric meters, voter registration and school age children. ** Townships not designated by * are based on average of two factors: voter registration and electric meter. 1 Township population includes village papulation as shown in (). 2(16 Age Group Under 5 5 - 9 10 - 14 15 - 19 20 - 24 25 - 29 3O - 34 35 - 39 40 - 44 45 - 49 50 - 54 55 - 59 60 - 64 65 - 69 70 - 74 75 - 79 80 + TOTAL 2(17 TABLE 36 Estimated Population By Age Grogp 1970 Census 3,297 3,796 4,130 4,050 2,671 2,420 2,142 2,129 2,183 2,296 1,967 2,042 1,798 1,410 1,044 834 966 39,175 Grand Traverse County Spring 19 :01" 5 Total 8.4% \O .7% o—ov—o 00 0001 3423*! N N N (a) A 01 0'1 0'1 0'1 0'1 01 0% O5 0 O O O O O O O I O I O o N 100.0% 5 79 pring 1979 3 .900 4 .178 4,438 6,611 5,238 4 .465 3 .322 3,218 2,968 3 .000 2 .255 2 .422 2,060 2 .009 1,299 935 1.159 3,468 % of Total \I .3% \l .8% CD .3% .4% H N G) 32 .4% .2% .0% .6% .6% .2% .5% .9% .8% .4% 1.7% Nww-FémU'IO‘O’iQO 100.0% % Change 18.3% 10.1% 7.4% 6.3% 96.1% 84.5% 55.1% 51.2% 36.0% 30.7% 14.6% 18.6% 14.6% 42.5% 24.4% 12.1% 19.0% 36.5% 208 OFG— o mzzogu .9... lam—.25 :..??qu (EA lu+od.0_wnto .T I o m In +1 I o. o; I 9 id I ON ON I 0N +0 I on an I m0 +1... I 0.? or. I mt ...um I on am I mm .vo I 00 am I 00 4% I or m.» I on ... om Mgr man 0...... min... 0.24 Not 35.. ... s. @865 Ialhbb.mr b .Oouwndmwzfixb.nurmfiww and. as... One. ZOPPDQEQIPQQ UG< Median Family Income: Family Income, 1978: 0,000 5,000 10,000 15,000 20,000 25,000 30,000 + TOTAL $10,459 4,999 9,999 14,999 19,999 24,999 29,999 209 TABLE 38 Family Income Levels Grand Traverse County 1975 1978 Survey Survey $12,167 $18,500 Percentage 6.9% 13.4% 17.5% 17.9% 14.3% 15.7% 14.3% 100.0% APPENDIX E MICHIGAN EDUCATIONAL ASSESSMENT PROGRAM READING - SEVENTH GRADE PROPORTION ATTAINING 75-100% OF OBJECTIVES FOR SEVEN SCHOOL DISTRICTS IN NORTHWEST MICHIGAN DQKOH N Rank 90 80 70 60 50 40 30 69.4 68 2 1 68.6 66 . 611.3 1 I 58.5 T 35.6 TC A B C D E F (728) (153) (80) (73) (53)(244) (66) g 1 2 5 7 6 4 3 210 School 211 C M m H m o A m m a a a mm omm me as am mma amp 0 a z m m a o m a we m.ov I m.vm 8.8m e ~.~w - - m.mm v.o> a m.he ands mm>fluownno mo wooalmn mcwcwmuum cofiuuomoum momuo cucm>0m I mcwcmom zmmwomm BZMmemmmm qmonB