A STUDY OF SELECTED MICHIGAN ELEMENTARY AND SECONDARY TEACHERS' AND PRINCIPALS' ATTITUDE TOWARD COMPUTER ASSISTED INSTRUCTION By Carlton Porterfield Robardey, Sr. A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1971 PLEASE NOTE: Some Pages have i n d i s t i n c t p rin t. F i l m e d as r e c e i v e d . UNIVERSITY MICROFILMS ABSTRACT A STUDY OP SELECTED MICHIGAN ELEMENTARY AND SECONDARY TEACHERS' AND PRINCIPALS' ATTITUDE TOWARD COMPUTER ASSISTED INSTRUCTION By Carlton Porterfield Robardey, Sr. The purpose of this study was to examine the relation­ ship between the dependent variable attitude and several independent variables with respect to computer assisted instruction (CAI). (1) knowledge; These independent variables included: (2) age; (3) level in the educational field; i.e., elementary-secondary; (4) classification of the school district employing the sample member; rural-urban; i.e., (5) position in the education field; i.e., principal-teacher; and secondary teachers; (6) subject matter taught by i.e., English/social studies-science/ mathematics. Three instruments were developed to gather data in order to answer questions pertaining to the purpose of the study. The instruments were an attitude scale, an instrument to measure knowledge, and a background questionnaire. Carlton Porterfield Robardey, All of the instruments were pretested using subjects similar to the population of the study. In addition, the construct and face validities were established for the attitude instrument and its reliability computed based on the pretest data and the face validity of the modified instrument established. These instruments were then distributed to a random sample of teachers and all of the principals in the population of interest (Washtenaw County, M i c h i g a n ) . the 276 sets of instruments mailed, Of 256 sets of instru­ ments were returned for a 92.0% response. Seven null hypotheses were tested in an attempt to answer questions relative to the purpose of the study. All of the hypotheses were tested at the .05 level of significance by either the one-way analysis of variance, Pearson-product moment correlation, or the F-test. addition, In internal consistency reliabilities were cal­ culated for the attitude scale and knowledge instrument. Thus, it was found that the following statement is supportable; There is a statistically significant positive relationship between knowledge and attitude with respect to CAI. DEDICATION To Mom and Dad It was you who instilled in me the courage to always strive for that which is beyond the easy reach and in the reflection of time, the example, you both set as parents will always remain a goal to be achieved in my lifetime. That this dissertation was attempted can be attributed to you both. ACKNOWLE DGMEN TS There are a great man y people who contributed significantly to the completion of this dissertation. The writer would like to acknowledge their help and encouragement. Special thanks goes to Dr. Dale V. Alam, Chairman of the doctoral committee. Likewise, gratitude is extended to the other members of the guidance c o m m i t t e e , Dr. Jane E. Smith and Dr. Norman T. Bell, for their help and understanding. To a good friend and colleague, Dr. Robert L. Trezise of the Michigan Department of Education goes much appreciation for his constant encouragement and construc­ tive criticism from the beginning to the end of the study. Recognition is given to Dr. Robert Scrivens and the INDICOM staff for their help in field testing the instru­ ments used in this study. Acknowledgement is given to Dr. Harvey Long, Instructional Systems Consultant, International Business Machines Corporation; Dr. Hal Wilson, Director, Instruc­ tional Systems, Harcourt Brace Jovanovich, Inc., Dr. Ronald Christopher, Director of the CAI Laboratory, iii The Ohio State University; Dr. Ivan Wagner, Director of Data Systems, Ann Arbor Public Schools, Ann Arbor, Michigan; Dr. Ronald Arnold, Director of Project INDICOM, Waterford Public Schools, Waterford, Michigan; and Mr. John Grate, Director, CAI Project, Cincinnati Public Schools, Cincinnati, Ohio for giving freely of their time to act as judges for the establishment of face validities for the cognitive and affective instruments used in the study. The writer also wishes to acknowledge the help and encouragement of all the members of the General Education Services Area of the Michigan Department of Education. However, a special note of thanks is due Mr. Don Goodson (deceased), the late director of E.S.E.A. Ill, whose help and advice were invaluable in the early stages of the study and to the writer's secretary, Mrs. Carolyn Terrill, who constantly kept him on the "straight and narrow." Thanks also goes to Dr. Patricia Carrigan, Mrs. Marilyn Field and Mrs. Carolyle Towers of the Office of Research, Ann Arbor Public Schools, Ann Arbor, Michigan, and Dr. Wiley Brownlee, Principal of Willow Run High School, Ypsilanti, Michigan, for their assistance in implementing the study. Much appreciation goes to Bob Wilson, fellow student and consulting statistician for his insightful criticism and his tact in expressing it and to Howard Heitzeg, iv friend and neighbor who's humor made the dissertation effort bearable. Last but far from least, to m y wife Peggy, for contributions far too numerous to recall and many times taken for granted and to m y children Michaelle, Port and Millette for their patience, love and understanding in accepting the absence of dad during much of the time the study was in progress, goes m y deepest appreciation. Without their support the entire doctoral program would have been impossible. v TABLE OF CONTENTS CHAPTER I Page INTRODUCTION . . . ........................... 1 Need for the S t u d y .............................. 2 Purpose of the S t u d y ...........................15 H y p o t h e s e s ...................................... 16 General Procedures ........................... 17 A ss u m p t i o n s...................................... 18 Delimitations....................................19 Limit a t i o n s ...................................... 19 Definition of Terras............................. 21 Organization of the S t u d y ............... 22 II REVIEW OF THE L I T E R A T U R E ...................... 24 A t t i t u d e ........................................ 24 Measurement of Attitudes .................... 25 Achievement and A t t i t u d e ...................... 28 Pupil Achievement in Relation to Teacher A t t i t u d e s ...................................... 31 Attitude Toward Computer Assisted I nstruction.................................... 32 Computer Assisted Instruction..................35 Potentials of Computer Assisted Instruction. ................................ 46 S u m m a r y .......................................... 49 III METHODS AND P R O C E D U R E S ........................ 51 Development of the Attitude S c a l e ............. 51 Attitude Scale V a l i d i t y ........................ 54 Attitude Scale Reliability ................. 60 Development of the Test for Knowledge of C A I ........................................ 62 Test V a l i d i t y .................................... 66 Development of the Background Question­ n a i r e ...........................................68 Geographic Area of the S t u d y ..................71 Description of the S a m p l e ...................... 72 Sampling P r o c e d u r e s............................. 73 Collection of D a t a ............................. 76 Analysis and Preparation of the Data . . . . 78 S u m m a r y .......................................... 84 vi CHAPTER IV V Page ANALYSES OF THE D A T A ....................... 86 Analysis of H y p o t h e s e s ............... Summary of Hypotheses .................... Analysis of the Knowledge Instrument. . . Analysis of the Attitude S c a l e ........... S u m m a r y .................................... 87 89 96 96 97 SUMMARY AND CONCLUSIONS 99 .................. S u m m a r y .................................... F i n d i n g s .................................... Delimitations of the S t u d y ................ Limitations of the S t u d y .................. C o n c l u s i o n s ................................ Di s c u s s i o n.................................. Implications for Further S t u d y ........... 99 100 101 102 103 104 106 BI B L I O G R A P H Y ......................... 107 vii LIST OF TABLES TABLE 1 2 3 Page Face Validities of the Attitude Scale I t e m s ............................................. 55 Attitude Scale Construct Validity S t a t i s t i c s ...................................... 59 Range of Scores for Two Groups Given A CAI Attitude Scale .................................. 60 4 Attitude Scale Reliability Statistics. 5 Face Validities of Knowledge Test Items. 6 School Districts Included in the Study . . . 73 7 Numbers and Percentages of Questionnaires Returned by Categories ....................... q 78 Pearson-product Moment Correlation Between Knowledge and Attitude with Respect to CAI . 87 8 9 10 11 12 13 . . . . . 61 68 Table of Means of Attitude Scores for Elementary-Secondary Teachers and P r i n c i p a l s ...................................... 89 One-Way Analysis of Variance on Attitude Scores of Elementary-Secondary Teachers and P r i n c i p a l s ...................................... 89 Table of Means of Attitude Scores for RuralUrban Teachers and P r i n c i p a l s.................. 90 One-Way Analysis of Variance on Attitude Scores for Rural-Urban Teachers and P r i n c i p a l s ...................................... 90 Table of Means of Attitude Scores for Young (age < 30) Old (age > 30) Teachers and P r i n c i p a l s ...................................... 91 viii TABLE 14 Page One-Way Analysis of Variance on Attitude Scores of Young (age < 30) Old (age > 30) Principals and T e a c h e r s ....................... 91 Table of Means of Attitude Scores for Principals and Teachers....................... 92 One-Way Analysis of Variance on Attitude Scores of Principals and T e a c h e r s ........... 93 Table of Means of Attitude Scores for Secondary Science/Mathematics-English/ Social Studies Teachers....................... 93 One-Way Analysis of Variance on Attitude Scores on Secondary Science/MathematicsEnglish/Social Studies T e a c h e r s .............. 94 19 Summary of Hyp o t h e s e s .......................... 94 20 Knowledge Instrument Summary Statistics. . . 96 21 Attitude Scale Summary S t a t i s t i c s............ 97 15 16 17 18 ix LIST OF APPENDICES APPENDIX Page A Attitude S c a l e .................................. 115 B Letter to CAI E x p e r t s ......................... 118 C Indices of Difficulty and Discrimination for Knowledge Instruments A and B ...........120 D Knowledge Instrument Used in the Survey E Background Questionnaire...................... 138 F Map of Michigan by C o u n t i e s .................. 139 G Cover Letters and Follow-up L e t t e r ........... 140 H Item Analysis of the Knowledge Instrument Used in the S u r v e y ............................. 144 I Derived Weights for Attitude Scale I t e m s .................. J . .132 157 Item Response Pattern for the Attitude S c a l e .......................................... 159 x CHAPTER I INTRODUCTION Education in the United States, true to its philos­ ophy, is being called on to provide for the needs of all its citizens and not just the few. However, in recent years certain critical problems have become evident which makes the pursuit of this philosophy difficult. These problems include the need to lower the drop-out rate, the need to re-educate segments of our adult population, the realization that students learn in different ways, and the demand and right of disadvantaged groups for a quality education are some of the factors that have created a dual problem of increasing enrollments and rising costs. Increasing costs, the need to educate many more students at all ages, but with different backgrounds, while using limited resources— all of these create demands for a new approach to the problem. Educational tech­ nology and computers specifically offer one hope for a solution. What are the special characteristics of this particular medium which offer some solutions to mass education? The traits of self-pacing, interaction, presentation of instructional sequences based on prior 2 responses and available past information, diagnosis of weakness in skills and abilities that are often overlooked by human evaluation, and the ability to employ different media for basic and remedial sequences are some primary examples. In addition, the computer, or rather the computer terminal which interfaces the student and machine, race, is completely neutral toward the learner's religion or economic background. This in itself is of prime importance to some disadvantaged g r o u p s .^ If the computer follows the same growth pattern as television and air conditioning (and there is no reason to think that it will n o t ) , the cost will continue to decline. With all of the potentials that the computer holds for education, it is of prime importance that the computer does not go the w a y of other highly touted educational media, but rather becomes a vital part of education in the decade of the seventies. Need for the Study During the past decade or so the field of instruc­ tional technology has grown in importance in education ^This point was emphasized by John Grate, Director of the Computer Assisted Instruction Project, Cincinnati Public Schools, Cincinnati, Ohio, and Dr. Hal Wilson, Director of Instructional Systems, Harcourt, Brace and Jovanovich, Inc., during the course of a conference on computer applications to learning which the present researcher coordinated. 3 from the use of the tape recorder, 16 m m movie projector, and educational television to the use of computers for administrative data processing and classroom instruction. The rate of growth of instructional technology, unfortunately, has not always been accompanied by sufficient research as to its effectiveness. Yet, the need for research in the area of instructional technology, particularly in computer assisted instruction (CAI), becomes clear when the present use of computers in today's society is realized and, even more so, when the p r o j e c ­ tions for the future use of computers for instruction are taken into account. The projected importance of the computer in education is almost startling. For example, Dr. Patrick S u p p e s , Director of the Institute for Mathematical Studies in the Social Sciences of Stanford University, stated that: ...both the processing and uses of infor­ mation are undergoing an unprecedented technological revolution. N o t only are machines now able to deal with man y kinds of information at high speed and in large quantities, but also it is possible to manipulate these quantities of information so as to benefit from them in entirely novel ways. This is perhaps no more true than in the field of education. One can predict that in a few more years millions of school children will have access to wha t Philip of Macedon's son, Alexander, enjoyed as a royal prerogative: the personal services of a tutor as well ^ informed and responsive as Aristotle. 2 PatrickSuppes, "The Uses of Computers m Education," Scientific American, Vol. 215, No. 3, 1966, p. 207. 4 Thus, Suppes predicts an "Aristotle" for millions of students in the near future. Further indications of the impact computers will have on the whole establishment of education are indicated by Alexander Schure: The use of the computer will alter the face of education, and indeed of civili­ zation. The computer will be imbedded as a prime foundation stone in the schools, education centers, and universi­ ties of tomorrow. It will be a tool used locally within the classroom as well as a management device to administer large regional school systems.-^ Therefore, the question of whether the computer will become a part of education in the United States apparently can no longer be debated. In support of this statement one has only to converse with the directors of several of these projects and examine statistics on the number of computing installations in the United States in 1950 (10-15) and then compare that number to the number of installations in operation in 4 1970 (50,000). This increase alone is indeed impressive, but at the present time there are at least 30 operational CAI projects at the public school level in the United 3 Robert Marker, ed., Computer Concepts and Educa­ tional A dmi n i s t r a t i o n , University of Iowa: Iowa Educational Information Center, 1968, p. 130. 4 Computers in Higher E d u c a t i o n , Report of the P r e s i ­ dents Science Advisory Committee, Donald F. Hornig, Chairman, Washington, D. C . : Government Printing Office, 1967, p. 58. 5 States. This number was arrived at by personal conver­ sation with the directors of several of these projects. These directors included Dr. Ronald Carruth of Macomb, Mississippi; John Grate of Cincinnati, Ohio; Dr. Sylvia Charp of Philadelphia, Pennsylvania; Waterford, Michigan; and others. Ronald Arnold of Between 1966 and 1969, at seven universities or related installations, there have been put into operation 34 CAI programs giving instruction in chemistry alone; and this list is by no means comprehensive.^ There seems to be no question, then, that the growth of the electronic computer during the past 25 years has been remarkable. However, there are some people who believe that even though the use of the computer in education has grown rapidly over the last few y e a r s , its use is still in its infancy when the extent that it will g probably be used in the years to come is considered. The question seems to be, then, not will CAI become a part of education, but what will be the nature of its role m the years to come? 7 ^Fredrick D. Tabbutt, "Computers in Chemical E d u c a ­ tion," Chemical and Engineering N e w s , Vol. 48, No. 3, pp. 53-54. g Computers in Higher E d u c a t i o n , p. 1. 7 Patrick Suppes and Max Jerman, "Computer-Assisted Instruction," The Bulletin of the National Association of Secondary School Pr i n c i pa l s , Vol. 54, No. 343, p. 27. o Lawrence Grayson of the National Center for E d u c a ­ tional Research and Development, United States Office of Education, has suggested that there are at least five factors that have to be considered before computer assisted instruction can begin to play a really major role in education. 1. Adequate hardware 2. The cost of telephone lines for remote terminals 3. Instructional and computer software 4. Proving the educational effectiveness of CAI 5. The attitudes toward CAI on the part of e d u c a t o r s . This last factor is what this investigator will con­ sider in this study. In other words, the role the computer will play in the future in schools m a y be very much influenced by the attitude of educators toward it. The literature indicates that attitudes are related to the acceptance and success of any instructional method.^' ^ In addition, there is evidence, based o Lawrence P. Grayson, "A Paradox: The Promises and Pitfalls of CAI," E D U C O M , Vol. 5, No. 2, pp. 2-3. 9 Equality of Educational O p p o r t u n i t y , Washington, D. C . : U. S. Government Printing Office, 1966, p. 22. ■^John F. O'Toole, "Teachers' and Principals' A t t i ­ tudes Toward Programmed Instruction in the Elementary School," A - V Communications R e v i e w , Vol. 12, No. 4, p. 431. ^ J o s h i Vidaya, "Attitude Toward Reception of Technology," Journal of Social Psychology, Vol. 58, 1962, p. 7. No. 1, 7 on research, that indicates the way graduate students learn in a computer assisted instruction course in statistics is influenced by their preconceived attitudes. 12 The importance of attitudes toward CAI in determining one's acceptance of and success in using computers in education is stressed by the opinions of resear c h e r s , experts and professional organizations in the field of computer assisted instruction and education in general. For example, the importance of teacher attitudes toward computer assisted instruction is expressed by Charles F. Hoban: The attitude of the classroom teacher tow:ard any instructional innovation— technological or othe r w i s e — is of par a ­ mount importance.13 The success of students in a CAI situation is also influenced by teachex attitude as is stated by Jerman and Anastasiow; The attitude of the teacher is a very important factor in determining the attitudes students will bring to their work, on the terminals. 14 12 Paul A. Stieman, A Normative Evaluation of a Computer-Assisted Instruction Laboratory in Statistical Inference, unpublished Master's Thesis, University of Pittsburgh, 1969, p. 45. 13 "Man, Ritual, The Establishment and Instructional Technology," Educational T e c h n o l o g y , Vol. 8, No. 20, p. 6. 14 Nicholas J. Anastasiow and Max Jerman, "Intro­ duction to Computer Based Drill and Practice in Arithmetic," H a n d b o o k , L. W. Singer Co., 1968, p. 13. 8 Another indication of teacher attitude was pointed out by a participant in a seminar on computer applications to education: This haggling about machine failing to substitute for teachers is exactly the kind of attitude I run into all the time with our computer-aided instruction projects. We find two predominant attitudes among the t e a c h e r s . O n e , the teachers are afriad that the machine will replace them; or two, they are so c o n ­ cerned about their students performing poorly on this system that they stand over the children and say, "There is the F; push that button."15 Grayson has also expressed a concern about acceptance of CAI by educators based on attitude: Advocates of CAI must contend with faculty conservatism and their tradition­ al reliance on textbooks and lectures as the principle vehicles of teaching. If computer assisted instruction is to become widely adopted, a change in teacher attitudes will have to occur.1 6 Reaction to CAI isn't limited to the average class­ room teacher. Lawrence Stolurow, a noted authority in the area of computer applications to education and the director of the CAI laboratory at Harvard University has stated: As a teacher it frightens me to think that the use of CAI will reveal to the 15 The Computer in E d u c a t i o n , An I/D/E/A Occasional Paper, a publication of the Institute for Development of Educational Activities Inc., Dayton, Ohio: 1970, p. 27. 16 Grayson, CAI," p. 3. "A Paradox: The Promises and Pitfalls of 9 world that I m a y have been wrong about teaching principles to which I am c u r ­ rently committed and believe in.l? This negative reaction isn't voiced by teachers alone. Administrators share some of the same fears. For example, the American Association of School Admini­ strators has concluded that: There is considerable fear, anxiety, or an unhealthy sense of awe that suggests that the computer is some kind of allpowerful, incomprehensible, infallible, and independent " s u p e r b r a i n . " 1 8 Another indication of administrator attitude toward and fear of the computer was brought out by McDonald when he said: The fears and negative speculation engaged in by education administrators m a y n o t be valid, but they are real at the present time. If computers and the solution they tend to offer education for the problems outlined in the introduction are to become a reality, CAI must be accepted by the educational community. This lack of understanding and 17 Lawrence M. Stolurow, "Computer Assisted Instruc­ tion," Education Automation Monograph S e r i e s , American Data Processing, Inc., Detroit, 1968, p. 8. 18 American Association of School Administrators' Committee on Electronic Data Processing, EDP and the School A dm i n i s t r a t o r s , Washington, D. C . : 1967, p. 10. 19 Joseph B. Maroline and Marion R. Misch, Education in the 70's , Final Report of Educational Policy Project, George Washington University, Washington, D. C., Autumn, 1967, p. 289. 10 mistrust by educators is clearly stated by Dwight Allen and Glenn Hawkes along with the need for education to accept the computer and place it in its proper ed u c a ­ tional p e r s p e c t i v e : Computers, of course, are a major medium for information storage and retrieval, but u n f o r t u n a t e l y , educators have not yet learned to treat them (and all that they symbolize in terms of new media) as they should be treated, in their proper place. We remain basically suspicious of the new technology, thus becoming part of an unnecessary "two cultures" self-fulfilling prophecy. We think and act in relation to the computer as if it were some kind of personal foe. We say, "The computer sent me an inaccurate bill last m o n t h ," or "Look at the mistake that the computer m a d e ;" and because we do not see the computer in its place as a tool of human invention for human use, we resist bringing it into our educational process except in the m ost peripheral ways, like keeping attendance records. We think of the typewriter as a tool, and we utilize it; we do not say, "My typewriter typed an unsatisfactory letter;" rather we say^* "I made a mistake in typing the letter." Thus research directed at variables which may be critical in determining whether computer assisted instruc­ tion can be successfully implemented is needed. This need was clearly indicated by Tobias. In informal conversations with personnel involved in the implementation of auto­ ma t e d devices in the classroom, the 20 Dwight W. Allen and Glenn W. Hawkes, "Reconstruc­ tion of Teacher Education and Professional Growth Programs or How the Third Little Pig Escaped the Wolf," Phi Delta K a p p a , Vol. 52, No. 1, p. 10. 11 teacher's fear of and resistance to these new devices is frequently noted. It is surprising, therefore, that there is little systematic research regarding such fears among teachers.2^ Further support for research involving the accept­ ance of computer assisted instruction is given by Richard T. Bueschel, President of Time Share Corporation, H a n o v e r , New H a m p s h i r e : Perhaps the biggest problem to be o v e r ­ come in exploiting time-sharing within the school is the computer mystique. This is usually manifested in a subtle fear of what m ost laymen and teachers regard as a complex electronic m a c h i n e — the com puter.2 2 In addition, Robert T. Filep, director of three computer assisted instruction projects, Education Systems Projects, System Development Corporation, Santa Monica, California, stated: If the technological innovation of computeraided instruction is to have any impact on the problem of mass education, some feasibility studies will have to be c o n ­ ducted to determine if terminals made available in remote operational locations will be used, and with w hat results.22 21 Sigmund Tobias, "Effects of Attitudes to Programmed Instruction and Other Media on Achievement from Programmed Materials," A - V Communications Review, Vol. 17, No. 3, p. 299. 22 Richard T. Bueschel, "Time-Sharing, A Progmatic Approach in the School," Educational T e c h n o l o g y , Vol. 10, No. 3, p. 23. 23 Robert T. Filep, "Individualized Instruction and the Computer: Potential for Mass Education," A-V Communications Review, Vol. 15, No. 1, p. 109. 12 A study of this nature is certainly needed when it is realized, also, that very little if any research has been conducted concerning e d u c a t o r s 1 attitudes as they exist in the field rather than in an experimental room or demonstration project. class­ O'Toole emphasized this need when he reported a study concerning teaching machines and programmed instruction: ...data reported in this study suggest the need for additional research in­ volving faculty attitudes if programmed instruction is to be introduced success­ fully and used in classroom settings. Based upon the history of other innovations, t e a c h e r s 1 acceptance of a new audio-visual aid or teaching method is largely dependent upon their attitude. ^ In addition to opinions such as those cited above, there are several research studies indicated in the l i t ­ erature which deal specifically with attitudes toward CAI. However, these studies have been concerned with attitude upon completion of a computerized instruction unit. Only three studies have been concerned with 24 O'Toole, "Teachers' and Principals' Attitudes Toward Programmed Instruction," p. 438. 13 attitude before treatment. 25 ' 26 ' 2V Yet even these were more concerned with changes in attitude than variables associated with the attitude as originally measured. In other words, very little research has been done in the area of determining attitudes toward CAI before teachers and administrators have been involved in CAI p r o g r a m s . Since attitudes are related to the acceptance and success of any instructional method, it is important to look at those variables which are associated with attitude. One of the variables that seems to be related to attitude is knowledge. Therefore, it is logical to. expect attitudes of teachers and administrators, i.e., principals who have a knowledge of CAI, to be positive in relation to those who do not have this knowledge. There is support for this assumption based on past 25 George R. Christopher, The Influence of a Computer Assisted Instruction Experience Upon the Attitudes of School Admin i s t r a t o r s , unpublished doctoral thesis, The Ohio State University, 1969, pp. 70-71. 26 Arthur Mathis, Timothy Smith and Duncan -Hansen, "College Students' Attitudes Toward Computer-Assisted Instruction," Journal of Educational Psychology, Vol. 61, No. 1, 1970, pp. 46-47. 27 Stieman, Evaluation of a Computer Assisted Instruction Laboratory, pp. 1-3. 14 research. Christopher 28 in research conducted at Ohio State University in 1969, indicated that a relationship existed between knowledge of computer applications education and attitude toward CAI. to His findings indicated this relationship to be in a positive direction; i.e., the more knowledge an individual has the more favorable the attitude of the individual. However, there is some confusion indicated by the literature on this point. Tobias in a study dealing with attitudes of teachers toward teaching machines and programmed instruction indicated that: ...the negative reaction to automation appears to be only slightly affected by information the teachers possess c o n ­ cerning programmed instruction.29 In addition, Christopher's study was confined to administrators involved in a CAI program as students and not as educators in a typical school setting. In summary, the need for this research is seen from the following observations. 1. Computers and computer assisted instruction possibly hold the solution to some of the most urgent educational problems. Yet there is 28 Christopher, Influence of Computer Assisted In­ struction Upon the Attitudes of Admi n i s t r a t o r s , p. 69. 29 Sigmund Tobias, "Teaching Machines and Programmed Instruction," A-V Communications R e v i e w , Vol. 14, No. 1, p. 108. 15 little research dealing with those variables which could influence the acceptance of this form of instruction. 2. Studies of educators' attitudes toward computer assisted instruction in the on-going educational environment of the school are for the most part nonexistent. 3. Attitudes are important in the acceptance of an instructional method; yet the relationship of knowledge to attitude is unclear in the litera­ ture. Purpose of the Study In light of the literature, attitudes are important in considering the acceptance and success of computer assisted instruction in the coming years. Therefore, the purpose of this study is to examine the variables that m a y be related to attitude toward C A I , particularly the variable of knowledge, other variables will, however, be considered as well. These variables are age, position and level held in the education field, the classification of the school district the sample member is employed by (rural-urban) and the subject matter taught in the case of secondary teachers. The specific questions for which answers were sought included: 1. Is there a relationship between knowledge of CAI and attitude toward CAI? 16 2. Is there a relationship between elementary level educators and secondary level educators with respect to attitude toward CAI? 3. Is there a relationship between urban educators and rural educators with respect to attitude toward CAI? 4. Is there a relationship between educators under 30 years of age and educators over 30 years of age with respect to attitude toward CAI? 5. Is there a relationship between principals and teachers with respect to attitude toward CAI? 6. Is there a relationship between secondary science and mathematics teachers and secondary English and social studies teachers with respect to attitude toward CAI? Hypotheses In an effort to find answers to the preceding questions the following research hypotheses were f o r m u l a t e d : 1. There is a strong positive relationship between knowledge of CAI and attitude toward CAI. 2. Elementary principals and teachers, as a group, are significantly different in attitude toward CAI when compared to secondary principals and teachers as a g r o u p . 3. Urban principals and teachers, as a group, are significantly different in attitude toward CAI whe n compared to rural principals and t e a c h e r s , as a group. 4. Principals and teachers under 30 years of age, as a group, are significantly different in attitude 17 toward CAI when compared to principals and teachers 30 years of age and older, as a group. 5. Elementary and secondary principals, as a group, are significantly different in attitude toward CAI when compared to elementary and secondary teachers as a group. 6. Secondary science and mathematics teachers as a group, are significantly different in attitude toward CAI when compared to secondary English and social studies teachers, as a group. General Procedures In order to investigate the preceding h y p o t h e s e s , it was necessary to carry out the following procedures. Three instruments were constructed. First, a twenty-item attitude scale was developed to measure attitudes toward computer assisted instruction. Second, an instrument to measure knowledge of CAI was developed from a pool of multiple choice items submitted by known experts in the field of CAI. Third, a background questionnaire was developed to gather selected biographical information. These three instruments were field tested using subjects similar to the sample of the study and modified where necessary. In addition, the validity for both the attitude scale and knowledge instrument were established. These field testing and instrument construction procedures will be discussed in greater detail in Chapter III. Elementary and secondary teachers and principals in Washtenaw County (Michigan) were selected as the target population of the study. Washtenaw County was selected 18 because it has all of the attributes that were considered necessary for the s t u d y . T h e s e attributes included: 1) all five community types of school districts located in the county; 2) no school districts less than K-12; 3) the county was geographically small enough to allow for personal contact with sample members; and 4) lease telephone lines were available for the researchers use. The sample for this study was selected by random methods in the case of teachers. However, the entire population of principals was included due to their small total numbers in the population. A packet of materials was sent to the identified sample members. questionnaire, This packet included a background attitude scale, measure of knowledge, and directions for taking and returning the materials. Follow-up was by letter and telephone call. Thus, the returned responses provided information relevant to the testing of the previously stated hypotheses. Assumptions Certain assumptions have been made with regard to this study, as follows: 1. Computer assisted instruction will remain a part of education. 2. Teachers and principals are the key personnel in determining the acceptance and success of CAI in the public schools. 19 3. Teacher and principal attitudes are important in determining the success or failure of CAI. 4. The instrument (p.133) used to measure knowledge of CAI covers the important aspects of this form of instruction. 5. The instrument used to measure principal and teacher knowledge of CAI will be sufficiently reliable for the purposes of this study. Delimitations This study is restricted to: 1. Elementary grades defined as grades defined as (K-6) and secondary (7-12). 2. Principals and teachers. 3. School districts with grades K-12. 4. Public schools. 5. Washtenaw County (Michigan). Limitations Readers should be made aware of certain limitations of this study. These limitations include the following: 1. In this study, the attitudes of teachers and principals toward computer assisted instruction are considered of prime importance. However, there is doubt as to what an attitude scale, such as the one used in this study, actually measures. What the scale does measure, at best, is the 20 expressed reaction of a person responding to written statements. W h a t relationship exists between this expressed reaction and "true" attitude is not known. 2. The traditional classification of rural-urban is based on the livelihood of the population. However, in this study the classification of rural-urban is dependent upon the population, size of a community, and livelihood is not considered. 3. It is possible the the instrument used to m e a ­ sure knowledge of CAI (p. 133) does not cover all of the important aspects of computer assisted instruction. 4. It is possible that the reliability of the measure used to indicate teacher and principal knowledge of computer assisted instruction is not sufficient to be of a high predictive value. 5. All subjects in the survey were from a county that had one out of every six persons in the population enrolled in a program of higher education. Thus, the sample members could have been influenced by particular requirements and experiences in an institution of higher learning. 21 Definition of Terms For purposes of this study, Attitude: these terms are defined: "Attitude entails an existing predis­ position to respond to social objects which, in interaction with situational and other dispositional variables, guides and directs the overt behavior of the individual." (Cardno, 1955) Computer Assisted Instruction: "30 Computer assisted instruc­ tion, abbreviated CAI, is usually taken to mean a method of allowing the writer of programmed instruction material to use a computer system as a means of writing and presenting instructional material to a number of students individually. Hardware: The equipment or "machinery" used in a computer system. This would include the computer itself and any device used to relay instruction to the s t u d e n t s . Interface: The common boundary between the computer and the student; Remote T e r m i n a l : i.e., the terminal. Any device which is not at the actual site where the computer is located but is some distance away connected by some type 30 Marion E. Shaw and Jack M. Wright, eds., Scales for the Measurement of A t t i t u d e s , N e w York: Disney, McGrawHill Book Co., 1967, p. 2T 22 of communications system and is used to give instruction to students. Software: The programs, routines, and other written information used to give instruction. This would also include all of the personnel needed to write and maintain the programs and hardware. Terminal: Any device used which allows communications between a computer and the student by means of a communication line. Organization of the Study This study consists of five chapters, a selected bibliography and an appendix. Chapter I includes the introduction to the study, the need for the study, purpose of the study, hypotheses relevant to the study, general procedures, delimitations, assumptions, limitations, definition of terms and the organization of the study. Chapter II contains a review of the literature related to the study undertaken. This includes attitudes, measurement of attitudes, achievement and attitude, pupil achievement in relation to teacher a t t i t u d e s , attitudes toward computer assisted instruction, computer assisted instruction, potentials of computer assisted instruction and a summary. 23 Chapter III describes and explains the methods and procedures of the study. are reviewed: In this chapter the following development of the attitude scale, attitude scale validity, attitude scale reliability, development of the test of knowledge of CAI, test validity, development of the background questionnaire, geographic area of the study, description of the sample, sampling procedures, collection of data, analysis and preparation of the data, and a summary. Chapter IV contains the analysis of the data, the results and a summary. Chapter V contains the summary, findings, conclusions and implications for further research. CHAPTER II REVIEW OF THE LITERATURE In this chapter is presented a review of selected literature related to the study undertaken. The litera­ ture reviewed includes those studies related to attitude, measurement of attitude, pupil achievement in relation to teacher attitude, attitudes toward computer assisted instruction, computer assisted instruction, of computer assisted instruction, the potential and a summary. No attempt has been made to include a large number of studies in this review of literature. Rather, it has been the aim to include only those studies which have a bearing on the problem under study. Attitude There are a number of definitions of attitude that m ay be cited as examples of common usage of the term. English and English define attitude a s : An enduring learned predisposition to behave in a consistent way toward a given class of objects.^ Horace B. English and Ava C. English, A Comprehen­ sive Dictionary of Psychological And Psychoanalytic Terms: A Guide to U s a g e , N e w York: McKay, 1958, p. 50. 24 25 Krech feels that attitudes are best defined a s : An enduring system of positive or negative evaluations, emotional feelings, and pro or con action tendencies with respect to a social object.2 Allport gives this definition of attitude: A mental and neural state of readiness, organized through experience, exerting a directive or dynamic influence upon the individual's response to all objects, and situations with which it is related. Katz and Stotland components. 4 consider that attitudes have three major The information component forms the foundation on which the attitude is built. involves feelings, The affective component and it is this component which attitude scales attempt to measure. The third component, the action component, represents the extent to which the attitude has habits of action associated with it. Measurement of Attitudes One of the first studies on attitude research was by Thurstone and Chave.^ 2 They developed an attitude scale David Krech, Richard S. Crutchfield and Egerton L. Ballachey, Individual in Society, N e w York: McGraw-Hill, 1962, p. 177. 3 Gordon W. Allport, "The Historical Background of Modern Social Psychology," Handbook of Social Ps y c h o l o g y , Vol. 1, Cambridge, Mass.: A ddison-Wesley, 1954, p. 45. 4 Daniel Katz and Ezra Stotland, "A Preliminary State­ ment to a Theory of Attitudinal Structure and Change," Psychology; A Study of a Sci e n c e , Edward S. Koch, New Y o r k : M c G r a w - H i l l , 1 9 5 9 , p. 399. 5 Louis L. Thurstone and Ernest J. Chave, The M e a s u r e ­ m e n t of A t t i t u d e s , C h i c a g o : The University of Chicago Press, 1929, pp. 16-17. 26 which consisted of a series of statements of opinion each of which is allocated to a particular point on a continuum, which ranges from an extreme point of view on one end to the opposite extreme on the other end. The method of constructing this scale involves formulating a large number of nonmonotone items related to the object of the attitude in question, i.e., items that will cause diagree­ me n t between persons placing the items on the attitude continuum. These items are then sorted by a sizable number of judges into 11 piles or categories which appear to the judges to be equally spaced in terms of the degree to which the items reflect the underlying attitude. The piles thus formed are then numbered from 1 to 11, and a scale value is determined for each item by computing the median of the position given the item by all judges. interquartile range or Q value, A is then computed which is a measure of interjudge variability, and all items for which there is muc h disagreement are rejected. A small number of items for the final scale are then selected so that they are spread more or less evenly along the attitude continuum. In using the scale, the respondent is asked to check each item wit h which he agrees. His score is the median of the scale values of all the items checked. Thurstone, thus conceived of attitude in the singular as being a point on the scale and a broader concept of an attitude to be represented by an area along the scale. 27 This early work by Thurstone removed much doubt about the use of opinions as indices of attitude. g Likert modified the Thurstone procedure for c o n ­ struction of attitude scales by introducing a system of arbitrarily assigning values to the attitude scale. This is commonly known as the method of summated r a t i n g s . Scale construction requires the selection of a large number of monotone items, i.e., items having the character­ istic that the more favorable the individual's attitude toward the attitude object, for the item. the higher his expected score The multiple response allows a respondent to indicate his reaction to the item by rating it in one of five categories: strongly agree, agree, undecided, disagree, and strongly disagree. The categories are scored by assigning values of 5, 4, 3, 2, and 1, respec­ tively. This scoring procedure is reversed for negatively worded items. Thus one-half of the items are designed in such a way that "agree" would represent a favorable response, while selection of "disagree" in the remaining one-half would also indicate favorability. In using the scale the respondent uses the five-point rating scale to respond to the items in the final scale, and his score is computed as the sum of his individual item scores. g Rensis A. Likert, "A Technique for the Measurement of Attitudes," Archives of Psychology, Vol. 22, No. 140, 1932, p. 42. 28 The attitude scale used in this study is of the Likert type, i.e., employing the method of summated r a t ings. The Likert and Thurstone scales are probably the most commonly used scales to measure attitude. The advantage of the Likert scale is that it is easier to construct and score. It is difficult to compare these two methods of attitude scale construction, but one study 7 by Edwards indicates, that both scales have similar reliabiliti e s . Achievement and Attitude There have been a number of studies carried out dealing wit h attitude as it relates to achievement. Q Neale, Gill and Tismer report significant positive correlations between attitude and achievement. Their 7 Allen L. Edwards, Techniques of Attitude Scale Construction, N e w York: Appleton-Century-Crofts, 1957, pp. 167-168. g Daniel C. Neale, Noel Gill and Werner Tismer, "Relationship Between Attitudes Toward School Subjects and School A c i e v e m e n t ," The Journal of Educational Research, Vol. 63, No. 5, 1970, p. 235. 29 findings are supported by Buckeye, Shatlein.^ 9 Young, 10 and However, all of this research deals with attitude and achievement in the specific subject area of mathematics. Doty and Doty 12 show that achievement appears to be related to a series of personality character­ istics, one of which they define as attitude. They found a high positive correlation between grade point average and attitude toward the instructional mode. Shimabukuro and Woodruff 13 the pupil became negative, Frey, found that as the attitude of there was a marked decline in 9 Donald A. Buckeye, "The Effects of a Creative Classroom Environment on the Creative Ability of P rospec­ tive Elementary Mathematics T e a c h e r s ," Dissertation A b s t r a c t s , Vol. 29, No. 6, p. 1801-A. ^Norman E. Young, "Effectiveness of Three Approaches to the Teaching of a Methods Course in Mathematics for Pre-Service Elementary Teachers as Related to Understanding and Attitude," Dissertation Abstracts, Vol. 29, No. 9, 1968, p. 290 4 -A. 11 Stephen D. Shatlein, "A Study of the Change of Attitudes Toward Mathematics of Prospective Elementary Teachers," Dissertation Abstracts, Vol. 29, No. 9, 1968, p. 2904-A. 12 Barbara A. Doty and Larry A. Doty, "Programmed Instruction Effectiveness in Relation to Certain Student Characteristics," Journal of Educational Psychology, Vol. 55, No. 6, 1964, pp. 334-338. 13 Sherman Frey, Shinkichi Shimabukuro and A. B. Woodruff, "Attitude Change in Programmed Instruction Related to Achievement and Performance," A-V Communi­ cations Review, Vol. 15, No. 2, 1967, p. 204. achievement. In a study by Hedlund and Neidt 14 it is reported that attitude becomes more closely related to achievement durxng the period of instruction. reported that undergraduate college students' Tobias 15 attitudes toward programmed and automated d e v i c e s , and especially their attitudes toward traditional d e v i c e s , were highly related to their actual achievement from an instruction program. Thus, persons with highly positive attitudes toward traditional devices, such as exercise books and workbooks, tend not to learn as much from new media, such as programmed instruction. Contrary findings do exist. Jackson 16 concluded that nearly all investigations of the matter have found no statistically significant relationship between attitudes toward school and school achievement. However, the studies reviewed by Jackson concentrated on measuring general satisfaction with school, while the previous studies measured attitude toward specific school subjects. Thus, while there is some difference in the literature, in general there would seem to exist a positive correlation "^Dalva C. Hedlund and Charles 0. Neidt, "The Relationship Between Changes in Attitude Toward a Course and Final Achievement," Journal of Educational Research, Vol. 61, No. 2, 1967, pp. 56-58. 15 Tobias, "Attitudes to Programmed Instruction," p. 304. 16 Philip W. Jackson, Life in Cl a s s r o o m s , N e w York: Holt, Rinehart and Winston, 1968,p. 177. 31 between attitude and academic achievement, at least in specific subject areas. Pupil Achievement in Relation to Teacher Attitudes There is conflicting research as reported in the literature on the effect of teacher attitude on pupil achievement. Pinetz 17 and Cheikin 18 report no significant relationship between t e a c h e r s 1 attitude and the academic achievement of pupils. However, Mastin 19 found that elementary teachers with a positive attitude toward their pupils and the subject matter being taught brought about a positive attitude in their pupils and enhanced their ability to acquire factual knowledge. Also, Tobias 20 suggests that teacher attitudes are likely to affect pupil achievement. Further, O'Toole 21 reports that in a 17 Midlred C. Pinetz, "The Relationship Between Teachers' Attitudes and Effectiveness in the Classroom," Dissertation A b s t r a c t s , Vol. 24, No. 6, 1963, p. 2340. 18 Martin L. Cheikin, "An Investigation of the Effect of Measured Teacher Attitude on Selected Eight Grade Students," Dissertation A b s t r a c t s , Vol. 28, No. 10, 1967, p. 4042-A. 19 Victor E. Mastin, "Teacher Enthusiasm," Journal of Educational Research, Vol. 56, No. 7, 1963, pp. 385386. 20 p. Tobias, "Attitudes to Programmed Instruction," 305. 21 O'Toole, "Teachers' and Principals' Attitudes Toward Programmed Instruction," p. 431. 32 field study using programmed mathematics textbooks, programmed materials without the teacher was found to be superior to programmed material used by the teacher with a negative attitude. Although the literature fails to show a causal relationship between teacher attitude and pupil achieve­ ment, it does reveal that teacher attitude is significantly related to pupil attitude. This would suggest that positive teacher attitude is important in developing positive pupil attitude. The significance of this is seen from the research dealing with the relationship of achievement and attitude cited earlier. Attitude Toward Computer Assisted Instruction It is evident from the literature that studies c o n ­ cerned with attitude toward computer assisted instruction are not abundant. existence. However, there are some studies in Christopher 22 found that public school administrators, when exposed to a Computerized Instruc­ tion Unit (CIV), exhibited a favorable attitude toward computer assisted instruction. In addition, Christopher found a relationship existing between knowledge of computer applications in education and favorable attitude 22 Christopher, Influence of Computer Assisted Instruction Upon the Attitudes of Administrators, pp. 7071. 33 toward computer assisted instruction. Hansen 23 Mathis, Smith and report that college students involved in com­ puter assisted instruction were more positive in attitude toward computer assisted instruction than the control group. This study also emphasized the fact that those individuals whose attitudes tended to be negative were those students who made many errors while being instructed by the computer. Suydam and Mitzel 24 reported that the attitude of Applachia elementary teachers toward mathematics was generally positive after seven weeks of mathematics instruction by computer. In this same report, Borman 25 found that an overwhelming majority of all the Applachia elementary teachers completing the computer assisted instruction course of in-service mathematics education obtained scores which indicated that they were favorably disposed toward computer assisted instruction. 23 Mathis, Smith and Hansen, "College Students' Attitudes," pp. 50-51. 24 Marilyn N. Suydam and Harold E. Mitzel, "Evaluation of Attitudes Toward Mathematics" in Inservice Mathematics Education via Computer-Assisted Instruction for Elementary Schools Teachers in A p p a l a c h i a , Final Report No. R-26, University Park, Pennsylvania: Computer Assisted Instruc­ tion Laboratory, The Pennsylvania State University, January, 1970, pp. 15-18. 25 . . Karl G. Borman, "Expressed Student Opinion Toward Computer-Assisted Instruction," Inservice Mathematics Education via Computer-Assisted Instruction for Elementary School Teachers in A p p a l a c h i a , Final Report No. R-26, University Park, Pennsylvania: Computer Assisted Instruc­ tion Laboratory, The Pennsylvania State University, January 1970, p. 31. 2g Stieman found that college students exposed to a computer assisted instruction laboratory in statistical inference showed measurable positive change in attitude toward the computer. What is more significant in this study is evidence showing that preconceived attitudes seemed to have a bearing upon the way an individual learns from the course. 27 Long 28 found that high school through college age students who were taught computer programming via the computer had positive attitudes toward the course, the hardware and the system e nviron­ ment. In g e n e r a l , the literature indicates that exposure to the computer and computer assisted instruction tends to foster a positive attitude toward this mode of instruction. 26 Stieman, Evaluation of a Computer Assisted Instruc­ tion Labora t o r y , p. 37. ^ I b i d . , p. 45. 28 Harvey S. Long, A Determination of the Relation of the Total Time for Course Completion to the Duration of the Study Interval in Teaching Via Computer Assisted In­ struction, unpublished Doctor of Philosophy Thesis, New York University, 1968, pp. 86-90. 35 Computer Assisted Instruction Feldhusen and Syabo 29 credit the teaching machines and programmed instruction movement, which grew out of the work of B. F. Skinner, 30 as being the major new educational development of the late 1950's and early 1960's. They also believe that the corresponding major development of the late 1 9 6 0 's and early 1 9 7 0 's will be computer assisted instruction. The increasing numbers of published reviews of the literature 31 ’ 32 readily attest to the rapid development of computer assisted instruction. Atkinson and Wilson 33 attribute the rate of growth of computer assisted instruction to the following: 1) the potential of computer assisted instruction to individualize instruction; 2) the development of 29 John Feldhusen and Michael Syabo, "The Advent of the Educational Heart Transplant, Computer-Assisted Instruction: A Brief Review of Research," Contemporary Ed u c a t i o n , Vol. 40, No. 6, 1969, p. 265. 30 York: B. F. Skinner, The Technology of T e a c h i n g , New Appleton-Century-Crofts, 1968. 31 Helen A. Lekah, Ed., Index to Computer-Assisted Instruction, Milwaukee: Instructional Media Laboratory, University of Wisconsin, 1969. 32 Albert E. Hickey, Computer-Assisted Instruction: A Survey of the Literature, Newburyport: ENTELEK Inc., 1968. 33 Richard C. Atkinson and Hal A. Wilson, ComputerAssisted Instruction: A Book of R e a d i n g s , New York, London: Academic Press, 1969, pp. 3-4. 36 programmed instruction; 3) the tremendous growth of electronic data processing in general; and 4) the increasing aid to education by the federal government. Stolurow 34 has identified five basic modes of instruction using the computer which are: 1. Problem s o l v i n g . This mode allows a student to use the computer as a t o o l . and enters the data. He writes a program The computer then processes the data for the correct solution to the problem. 2. Drill and P r a c t i c e . The computer presents learning materials such as spelling or arithmetic drills utilizing the same sequence and format giving the student repeated opportunities for response. The student uses his natural language with the objective being to build skills. 3. Inquiry M o d e . The student uses his natural language as in Drill and Practice to address questions to the computer. The computer, using key words and search algorithms will retrieve an answer. 4. Simulation and G a m i n g . Using his natural lan­ guage the student is given the option of varying 34 Lawrence M. Stolurow, "Some Factors in the Design of Systems for Computer-Assisted Instruction," In ComputerAssisted Instruction: A Book of R e a d i n g s , eds. Richard C. Atkinson and Hal A. Wilson: New York, London: Academic Press, 1969, pp. 81-82. 37 the input with the computer quickly reporting the outcome of his decision. 5. Tutorial. This m ode n o t only involves dialogue bu t also other modes. For example, the response to a student's question might be Drill and Pra c ­ tice or Simulation and G a m i n g . In addition a number of variations are possible within the various modes. In other words the system will not only have to select the proper mode as well as variation but do so depending upon the particular student and his past performance. According to Stolurow, this mode can be viewed as a form of artificial intelligence. Gerard 35 identified these five benefits that computer assisted instruction will bring to the student. They are: 1) better and faster learning since the student can time his learning at his own convenience, go at his own pace, and catch up missed time; 2) better teaching at many levels and in m a n y areas; 3) automatic measurement of progress; 4) personalized tutoring; and 5) the opportunity to work with rich materials and sophisticated problems. 35 R. W. Gerard, "Computers: Their Impact Upon Society," Journal of American Information Processing So cieties, Vol. 27, No. 1, 1965, pp. 33-40. 38 Hansen and Harvey 36 predict the role of the teacher in a computer assisted instruction environment. They based their predictions on a pilot study by Hill and Furst 37 and the development pattern of computer assisted instruction. They predict that: 1. The teacher will perform much less of the information presentation function presently found in the classroom. The teacher will become m ore involved in the managerial and strategy functions found in the sequencing and evaluation of instruction. 2. Teachers will play less of the corrective role in terms of their questioning and evaluative behaviors. This undoubtedly will offer a significant step forward in teacher-student relationships in that much of the negative verbal behavior observed in classrooms will now be shifted to a more individualized and private interaction within C A I . 36 Duncan N. Hansen and W. L. Harvey, "Impact of CAI on Classroom Teachers," Educational T e c h n o l o g y , Vol. 10, No. 2, 1970, pp. 47-48. 37 Russell A. Hill and Norma Furst, "Teacher Behavior in CAI Classrooms," Educational T e c h n o l o g y , Vol. 9, No. 2, 1969, pp. 60-62. 39 3. Teachers will become much mor e concerned with the host of individual characteristics important in designing an instructional strategy; thus the array of instructional resources and the decision making found in employing these resources will become more complex and also more frequent in terms of teacher behaviors. 4. The teacher will have a greater involvement in guiding individual students rather than in m a i n ­ taining classroom discipline. With the computer relieving the teacher of the information p r e ­ sentation tasks, she will be able to devote the time usually expended in group communication to individual counseling and advising. 5. Teachers will have to perform a wider range of discussion techniques involving a richer oppor­ tunity to affect the social and emotional behavior of students. Teachers will have to have greater skill and understanding of human behavior, viewed in the broadest terms. This requirement may in part be aided by CAI system's information retrieval capability, which may monitor the patterns and rates of student develop­ ment. 6. It is clear that the teachers will have a greater array of differentiated professionals joining 40 them in the team effort to provide optimal in­ struction. Some teachers m a y become experts in the guidance process, while others m a y become more competent in the application of technological procedures and functions for the fullest employ­ ment of computer technology. 7. Teachers m a y take on many more of the diagnostic assessment and prescriptive functions presently assigned the school psychologist. Teachers may, in fact, utilize more group interactive proce­ dures in an attempt to develop latent social and creative talents within their students. As Hansen and Harvey point out, these predicted professional roles of the classroom teacher offer a more creative and interesting world. However, they also challenge the teacher training institutions to anticipate the changing role of the teacher rather than following behind the classroom practice. Hansen 38 emphasized that the cost of the machinery is not nearly as important as the training of professionals to use this very powerful resource in a wise way. T h u s , it is clear that for computer assisted in­ struction to reach its full potential a new concept of the 38 Duncan N. Hansen, "Myths That Need to be Destroyed and Myths That We Ought to Create," Speech given at the National Conference on Computer Applications to Learning, Bloomfield Hills, Michigan, July 9, 1970. 41 teacher will have to emerge. It is also apparent that if there is to be any direction to the emerging role of the teacher in a CAI environment it will have to come from the institutions of teacher training. According to Long, 39 the large numbers of articles appearing in the literature discussing computer assisted instruction suggest that this form of instruction has been in existence for a long period of time. Further he indicates this would also imply that computer assisted instruction has been subjected to extensive research, but this has not been the case. The review of the literature dealing with CAI tends to support Long's statement and this investigator could only identify the following studies dealing with this instructional method. Uttal 40 identifies Gustave Roth and Nancy Anderson 41 as being the first researchers to use the computer for an 39 L o n g , Study Interval in Teaching via Computer Assisted Instr u c t i o n , p. 21~ 40William R. Uttal, "My Teacher Has Three Arms!!!" I.B.M. Research Report RC-788, 1962. 41 Gustave J. Roth, Nancy S. Anderson and R. C. Brainerd, The IBM Research Center Teaching Machine Project in Automated Teaching: The State of the A r t , Ed., E. Galanter, N e w York: John Wiley and Sons, 1959, pp. 117-130. 42 instructional tool. According to Long 42 this activity, which occurred in 1958, was not much different than the more sophisticated research which has been carried out to date. Schwartz and Long 43 report that field engineers receiving required training through remote computer assisted instruction completed the course of instruction in considerably less time than a self-study group. was also supported by Schwartz and Haskell. Feldman and Sears 45 This 44 in an exploratory study found that learners in a computer assisted instruction class­ room became more academic in their classroom behavior, while the non-CAI children increased their nonacademic behavior the course of the year. Gilman and Moreau 46 42 L o n g , Study Interval in Teaching via Computer Assisted In s t r u c t i o n , pp. 21-22. 43 H. A. Schwartz and H. S. Long, "A Study of Remote Industrial Training," Journal of Applied Psychology, Vol. 51, No. 1, 1967, p. 11. 44 H. A. Schwartz and R. J. Haskell, "A Study of Computer-Assisted Instruction in Industrial Training," Journal of Applied P s y c h o l o g y , Vol. 50, No. 5, 1966, p. 360. 45 David H. Feldman and P. S. Sears, "Effects of Computer-Assisted Instruction on Children's Behavior," Educational Te c h n o l o g y , Vol. 10, No. 3, 1970, p. 13. 46 D. A. Gilman and N. A. Moreau, "Effects of Reducing Verbal Content in Computer-Assisted Instruction," A -V Communication Review, Vol. 17, No. 3, 1969, p. 297. 43 reported that by reducing the verbal content of a CAI program, the learning by the student did not decrease. However, it was found that the instructional time was reduced. They also reported a lower correlation b e t ­ ween intelligence and learning on the part of the students who studied the low verbal content program. O'Neil, Spielberger and Hansen 47 used a CAI program on complex numbers and compound fractions which could be completed in one sitting with 26 students who were given anxiety-inducing messages when they arrived. gram was divided into a hard and easy part. The p r o ­ They found that students responded to difficult CAI materials with an increase in self-reported and physiologically measured anxiety (blood p r e s s u r e ) . High-anxiety students also made more errors on the difficult portion of the program than low anxiety s t u d e n t s , while low-anxiety students made more errors on the easy part of the program. 47 H. F. O'Neil, C. D. Spielberger and D. N. Hansen, "Effects of State Anxiety and Task Difficulty on ComputerAssisted Learning," Journal of Educational P s y c h o l o g y , Vol. 60, No. 5, 1969, pp. 345-350. 44 Hall, Adams and Tardibuono 48 studied the effects of providing feedback in the form of the full correct response when an error was made or of pointing out matching letters of the alphabet between the correct answer and that given by the student. College undergraduate students learned states and capital c i t i e s . The group that received full response feedback took significantly less time to com­ plete the program, but the amount learned did not differ significantly between groups. Diamond 49 reported that no difference in learning was found between high school and junior high school students taught biology by CAI and students in traditional classes. However, Scrivens 50 reported that fourth and fifth grade students receiving English lessons by CAI during a four-month period showed a gain of seven months while the control groups showed a gain of three m o n t h s . 48 Keith A. Hall, Marilyn Adams and John Tardibuono, "Gradient- and Full-Response Feedback in Computer-Assisted Instruction,11 Experimentation With Computer-Assisted Instruction in Technical E d u c a t i o n , Semi-annual Progress Report R - 6 , University Park, P e n n s y l v a n i a : Computer Assisted Instruction Laboratory, The Pennsylvania State University, June, 1967, p. 11. 49 James J. Diamond, A Report on Project Grow; Philadelphia's Experimental Program in Computer-Assisted Instruction , Philadelphia, Pennsylvania; Office of Research and Evaluation, The School District of Philadelphia, August, 1969, p. 11 of Section 2. 50 Robert W. Scrivens, Evaluation Monograph N o . 1, Waterford, Michigan: U.S.O.E. Grant 67-04301-0, Waterford Township School District, February, 1970, p. 59. 45' In this same report it was noted that significant gains were recorded in mathematics for grades three and four. Probably a more significant result was that s t u d e n t s 1 attitudes toward mathematics appeared to have been enhanced by CAI. Schurdak 51 found that college students taught the computer language FORTRAN by CAI scored significantly higher on achievement tests than equal groups taught either by programmed test or a conventional textbookworkbook combination. Bitzer and Boudreaux 52 reported that student nurses instructed by the PLATO system did not show significant gains in learning over the control group, but there was a significant reduction of instruction time. They also report that favorable student response to CAI seemed to increase with exposure to the system. 51 John J. Schurdak, "An Approach to the Use of Com­ puters in the Instructional Progress and an Evaluation," The American Educational Research Journal, Vol. 4, No. 1, 1967, pp. 71-73. 52 Maryann D. Bitzer and Martha C. Boudreaux, "Using a Computer to Teach Nursing," Reprint from Nursing F o r u m , Vol. 8, No. 3, 1969. 46 Potentials of Computer Assisted Instruction Alpert and Bitzer 53 provide an excellent discussion of advances in the field of computer assisted instruction. They point out that not only is the computer a powerful tool for -drill and practice in basic courses in welldefined content areas, b u t the PLATO II and III systems have also demonstrated the feasibility of using a diver­ sity of strategies in a variety of courses. Slavens 54 has utilized the computer to develop and test materials to train reference librarians. Author- controlled linear programmed exercises were tested via the computer. These programs were then revised, validated and transferred to a standard printed format for student use. In addition to these programs, the computer simulated a library user and the student acted as a reference librarian. Thus, the computer terminal acted as a means of communication or interface between the student and computer. The student librarian interacted with the computer as in a real life situation or in other words as a problem was presented by the computer, the student 53 D. Alpert and D. L. Bitzer, "Advances in ComputerBased Education," Sci e n c e , Vol. 167, No. 3917, 1970, pp. 1582-1590. 54 Thomas S l a v e n s , Development and Testing of Materials for Computer-Assisted Instruction in the Education of Reference L i b r a r i a n s , U. S. Department of Health, Education and Welfare, Project No. 80560, 1970. 47 librarian would proceed to ask questions which would classify the user's problem. In the final outcome, the student librarian would recommend the best reference work for the user's particular need. The use of games and simulation has been developed by Wing 55 to teach economics to sixth grade students. r c. Castleberry and Lagowski report the simulation of laboratory experiments in chemistry at the University of Texas. Mortensen and Penick lation games for chemistry. 57 have also developed simu­ These programs enable the student to take part in experiences that would be impossible in a normal school setting because of the factors of time and/or safety. Kemeny and Kurtz 58 describe how the Dartmouth T i m e ­ sharing system utilizes the computer effectively in college teaching. This system can serve many users at the same time, at many locations, on and off campus, by 55 R. L. Wing, "Two Computer-Based Economics Games for Sixth Graders," American Behavioral S c i e n t i s t , Vol. 10, No. 1, 1966, pp. 31-36. 56 S. Castleberry and J. J. Lagowski, "Individualized Instruction Using Computer Techniques," Journal of Chemi­ cal E d u c a t i o n , Vol. 47, No. 2, 1970, p p . 91-96. 57 Earl M. Mortensen and Ronald J. Penick, "Computer Animation of Molecular Vibrations: Ethane," Journal of Chemical E d u c a t i o n , Vol. 47, No. 2, 1970, pp. 102'-104. 58 John G. Kemeny and Thomas C. Kurtz, "Dartmouth Time-Sharing," S c i e n c e , Vol. 162, No. 3850, 1968, pp. 223-228,. 48 means of ordinary telephone lines. In addition, students and faculty are able to learn how to write and correct their own programs in a short period of time and with little inconvenience. This is primarily due to the simple interactive BASIC language used by the system. Another interesting feature of the system is the ability of the computer to check the program as it is written by the student or faculty member and give him hints when errors are made. Another application of the computer to education is in the management of instruction. In this m ode the computer is not limited to the role of an interactive device for the student but rather controls all aspects of his education. This would range from working out an individual daily schedule for the student, keeping track of all his learning activities and informing him as to what particular learning sequence he should start next in mathematics based on his past performance in this area, to giving him instruction using one of the instructional strategies as outlined by Stolurow. 59 At this time most studies utilizing the computer to manage instruction are still being evaluated. 59 However, Stolurow, "Systems for Computer-Assisted Instruc­ tion," pp. 81-82. 49 Gallagher 60 has reported that student selected sequence groups demonstrated superior performance over computer assigned sequence g r o u p s . It should be emphasized that the list of computer applications to education are growing, and these examples are but a small part of the total. Nevertheless, the examples do serve to show the wide range of subject matter and instructional strategies to which the computer can be applied. Summary Although the computer, and especially computer assisted instruction, has not been in existence for very long, it is evident from the literature that the growth of this instructional strategy is accelerating. It would also seem that in general there exists a positive correla­ tion between attitude and academic achievement, at least in specific subject areas. While no causal relationship between teacher attitudes and pupil attitudes has been found the relationship does seem to be positive. It is also evident that exposure to a computer and CAI tends to foster a positive attitude toward them. 60 Paul D. Gallagher, An Investigation of Instructional Treatments and Learner Characteristics in a ComputerManaged instruction C o u r s e , Tech. Report No. 12, Tallahassee, Florida: Computer-Assisted Instruction Center, Florida State University, 1970, pp. 56-59. 50 The present research on computer assisted instruction would lead to the conclusion that CAI can do as well for better than traditional teaching methods, but in less time. This in itself m a y be a significant factor in the near future wit h education's problem of an expanding population and a corresponding need for extended education. There can be no doubt that a redefining of teacher roles will be necessary to insure that computer assisted instruction will become an effective and efficient tool of education. This, in turn, stresses the importance for institutions of higher learning to recognize the challenge of teacher education for tomorrow and m eet the challenge today. CHAPTER III METHODS AND PROCEDURES In this chapter the procedures and methods used in conducting the study are presented. This includes construction of the measures used to indicate teacher and principal knowledge of and attitude toward computer assisted instruction. Also the methods used to establish the reliability and validity of these two measures are discussed. In addition the development of the background questionnaire, a description of the geographic area of the study, a description of the sample, sampling p r o c e ­ dures, methods used to collect and analyze the data, and a summary are presented. Development of the Attitude Scale A twenty-item attitude scale (Appendix A) was c o n ­ structed to determine teacher and principal attitude toward computer assisted instruction. The items making up this instrument were taken from two previously constructed attitude scales. Thirteen items were chosen from an attitude scale constructed by G. R. Christopher^" Christopher, Influence of Computer Assisted Instruc­ tion Upon the Attitudes of A d m i n i s t r a t o r s , pp. 111-114. 51 52 to measure public school administrator attitudes toward computer assisted instruction. tained forty-seven items. The original scale con­ However, only thirteen items were used from the instrument and not the entire questionnaire for the following reasons: Christopher's original instrument was designed for administrators in the public schools. However, the thirteen items selected by this researcher were considered to be appropriate to both teachers and principals who constitute the sample of this study. Also research has shown that the return rate for surveys increases as an instrument decreased in length. 2 Thus, since m the present study, a sufficiently large response was needed, it seemed impor­ tant to use as short an instrument as possible to increase the return rate. The remaining seven items were taken from a twenty3 item attitude scale developed b y H. A. Pearson. The Pearson instrument was designed to measure attitude toward programmed instruction, therefore, only some of the items of this instrument were applicable to the present study, 2 Larry L. Leslie, "Increasing Response Rates to Long Questionnaires," The Journal of Educational R e s e a r c h , Vol. 63, No. 8, 1969, pp. 347-350. 3 Harry A. Pearson, Programmed Instruction for Groups of Teachers in Remote Locations: Prototype Development, unpublished doctoral thesis, Michigan State University, 1969, pp. 176-177. 53 and these items could be made appropriate to the present study only with minor modifications, such as changing the word "programmed" to "computer assisted." The rationale for selecting these seven items from the Pearson instru­ me n t was that these particular items covered areas not included in the Christopher instrument. Further, these seven items were considered appropriate to both teachers and principals who constitute the sample of this study. The rationale for changing the word "programmed" to "computer assisted" was that since this study focuses specifically on CAI rather than programmed instruction, the term "programmed" would not be appropriate. This change was mad e under the assumption that this is accept­ able on the basis of consultation with Dr. William Schmidt of the Educational Psychology Department, Michigan State University, and also on the basis of research reported by Mathis, Smith and Hansen, 4 where an attitude scale was modified in a very similar way to obtain data for that study. Ten of the items selected reflected negative or unfavorable attitudes toward computer assisted instruction such as: computer assisted instruction is an impersonal teaching approach. The remaining ten items reflect a 4 Mathis, Smith and Hansen, "College Students' Attitudes Toward Computer Assisted Instruction," pp. 4651. 54 positive or favorable attitude toward computer assisted instruction such as: computer assisted instruction will improve instructional programs. Attitude Scale Validity The face validity of each item and the whole scale was estimated by asking three persons with experience in computer assisted instruction to rate the face validity of each item and the whole scale on a five-point scale from 0 for "no apparent validity" through 4 for "very high face validity." An index for each item and the whole scale was calculated by adding all ratings for each item and the whole scale and dividing by twelve. This divisor of twelve was derived by multiplying the maximum rating possible by the total number of judges. Thus, the maximum face validity index possible was 1.00 and the minimum was 0.00. The face validity index computed for the whole scale was .750. The item indices for the attitude scale are given in Table 1 below. 55 Table 1. Face Validities of the Attitude Scale Items* Face Validity Item Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 .500 .750 .917 .583 .833 .833 .750 .583 .500 .583 .667 .917 .750 .583 .917 .667 .667 .667 .667 .750 *Calculated from rating by three persons on a five-point scale with a validity index o f : 1.000 = very high face validity 0.750 = high face validity 0.500 = medium face validity 0.250 = low face validity 0.00 0 = very low face validity Thus, it can be seen that according to this procedure, the constructed scale, as a whole, had "high face validity." To establish construct validity, a variation of the known-groups technique was employed. This technique assumes that two or more groups should hold different attitudes toward a given object. It would also be expected that a valid scale to measure the attitude in question 56 should produce different scores for these groups. Thus, one might validate a scale to measure attitude toward Sunday observance by showing that Baptists score higher (have a more positive attitude toward Sunday observance) than do Seventh Day Adventists. 5 The details of the variation of the known-groups technique used in this study are given in the following pa r a g r a p h s . It was necessary to identify two groups with which to validate the attitude instrument: one group would be assumed to have positive attitudes with respect to CAI, the other group would be assumed to have a variety of attitudes ranging from the positive to the negative. Assuming the attitude instrument does reflect atti­ tudes toward CAI, the first group should score higher on the instrument than the second group. It would also be expected that the range of scores for the first group to be less than the range of scores for the second group. The first group, identified for this purpose, con­ sisted of fifty-two participants in a pre-conference seminar held at Mott Riverside High School, Waterford, Michigan. These participants were considered similar to the target population, however, there were more 5 Shaw and Wright, e d s . , Scales for the Measurement of Attitudes, p. 19. 57 administrators in this group than t e a c h e r s . The conference itself was one dealing with computer appli­ cations to learning with the main theme being computer assisted instruction (CAI). This seminar was designed to provide basic information to those participants who had indicated prior to the conference that they had no knowledge of CAI. As part of the pre-conference registration all participants were asked to indicate place of employment; i.e., elementary, secondary or higher education, and position held; or teacher. i.e., administrator This information was used to insure that no participant employed by a school district in the target population of this study would be included in this validation procedure. However, considering the technique used to establish construct validity, nine of these participants w ere excluded from the testing. These individuals were employees of the Michigan Department of Education who were attending the seminar on assignment, and not voluntarily. Therefore, no assumption could be made in regard to their attitude toward CAI. This left forty-three subjects to be tested in this group. In addition, all instruments were administered and collected prior to conducting the seminar. The second group to which the attitude scale was administered consisted of two general curriculum classes of ten and twenty-eight graduate students each at Michigan 58 State University. Care was taken to insure that members of either class employed by any of the schools in the target population of interest were not included in the validation procedure. This was done by consulting with the instructors of the classes prior to testing. Both instructors had collected some biographical information about their students which included teaching or administra­ tive experience and place of employment. These precautions were taken with both groups for the following reasons: first, this insured the validity of the field testing procedure, and second, this insured that the population of this study would not be contaminated. Thus, the assumption was made that the pre-conference seminar group would have more favorable attitudes toward CAI than the general curriculum group and this difference would be significant. This assumption was stated in the form of a null statistical hypothesis such that: There will be no difference between mean attitude scores for the seminar group and the m ean attitude scores for the curri­ culum group Symbolically: Ho: yi = \iz H i : |ii ^ y 2 L e g e n d : yi = seminar group mean y 2 = curriculum group mean Decision rule: Reject Ho if t > 2.33 at a = .01 It was also assumed that the range of scores for the pre-conference group would be less than for the general curriculum group. 59 In order to test this hypothesis and assumption and attempt to establish construct validity, the attitude scales were scored for each individual in both g r o u p s . For scoring purposes the attitude scales were weighted a priori by the researcher so that items reflecting a favorable attitude had a weight of 5 for strongly agree to 1 for strongly disagree and items reflecting a negative attitude had a weight of 1 for strongly agree to 5 for strongly disagree. Thus, the possible range of scores was from 20 to 100 with the higher the individual score the more positive the attitude. attitude scales, After scoring the the t-test, as given by Hays, 6 for unequal variances and sample sizes, was used to determine if the null hypothesis could be rejected. The results are given in Table 2 below. Table 2. Attitude Scale Construct Validity Statistics Pre-conference mean score 75.49 General curriculum m e a n score 64.90 Estimated standard deviation Corrected number of degree of freedom Computed t statistic 2.76 58 3.84* *Significant at the .01 level ^William L. Hays, Statistics, N e w York, Chicago, San Francisco, Toronto, London: Holt, Rinehart and Winston, 1963, pp. 317-322. 60 The computed t statistic indicates that the differ­ ence between the mean scores for the two groups is certainly significant. Therefore, the null hypothesis was rejected and the alternative hypothesis accepted. In addition, as Table 3 indicates, the range of scores for the pre-conference seminar group is m uch more narrow than the general curriculum group, having a spread of 36 points as compared to 62 points for the second group. This would agree with the previously stated assumption. Thus, on the basis of the t statistic and range of scores computed for both groups, the attitude scale is considered to have construct validity. Table 3. Range of Scores for Two Groups Given a CAI Attitude Scale (Possible Range 20-100) Group N Range Pre-conference 43 57-93 General Curriculum 38 36-98 Attitude Scale Reliability In order to establish the reliability of the instru­ ment, subjects that were as similar as possible to the subjects to be sampled from the target population were used. The two curriculum classes of ten and twenty- eight students each that were used to establish construct validity were considered appropriate for this task. These students were all practicing or experienced teachers 61 and administrators. Also, both the secondary and elementary levels of instruction were represented by both groups. In addition, the previously described precautions were taken to insure that none of the individuals were employed by any of the schools in the geographic area delimited to this study. The same a -priori weighting procedure previously discussed in establishing construct validity was employed. The weighted responses were key punched and verified. These responses were then analyzed by the Control Data Corporation's (CDC) 3600 computer at the Michigan State University Computer Center using the 7 FORTAP program which computed the Hoyt's reliability g coefficient. The attitude scale reliability statistics are given in Table 4 below. Table 4. Attitude Scale Reliability Statistics (N = 38) Standard Error Hoyt's Reliability Coefficient 3.2553 .9487 7 David J. Wright, FORTAP: A Fortran Test Analysis Package by F. B. Baker and T. J. M a r t i n , Occasional Paper No. 10, East Lansing: Office of Research Consultation, Michigan State University, 1970, pp. 8-13. 8 Cyril J. Hoyt, "Test Reliability Estimated by Analysis of Variance," in Principles of Education and Psychological Measurement: A Book of Selected R e a d i n g s , ed. by William Mehrens and Robert L. Ebel, Chicago: Rand McNally & Co., 1967, pp. 108-111. 62 This reliability is considered adequate for the purpose of the study. Development of the Test for Knowledge of CAI A test to measure knowledge of computer assisted instruction was developed from a pool of multiple choice test items submitted by known experts in this field. The researcher also modified submitted items to meet the criterion of multiple choice. This group of experts was identified by the University of Wisconsin listing of computer assisted instruction project directors. 9 The majority of these identified experts were contacted by mail. The letter (Appendix B) included a statement of the purpose of this study and a stamped, self-addressed envelope to be used for their reply. One hundred and one letters were mailed to these identified experts on com­ puter assisted instruction in the United States and Canada. A total of twenty-one people responded for a return of 20.8%. From these twenty-one responses a total of twenty-seven test items were submitted. All twenty- seven items were of the true-false type which were then changed to the multiple choice form. Personal contact was made with some practitioners in the field of CAI. This was possible because of the work 9 Lekan, Index to Computer Assisted Instruction, pp. 283-295. 63 that the researcher was engaged in as an interm with the E.S.E.A. Title III Office of the Michigan Department of Education. The primary responsibility of this internship was the coordinating of a national conference on computer applications to learning, the main theme of which was computer assisted instruction. This conference was held in cooperation with project INDICOM,'1'^ a computer assisted instruction project in the Waterford Public Schools, Waterford, Michigan. This afforded the researcher an opportunity to become acquainted with the entire staff, as well as with Mr. Ronald Arnold, project. the director of the At the request of the researcher, Mr. Arnold and twelve staff members selected by him were asked to submit multiple choice test items to be used in the construction of the final test instrument. These staff members were selected by Mr. Arnold on the basis of experience and expertise with computer assisted instruc­ tion. A total of forty-five items of a multiple choice form were submitted by this group. From this total pool of seventy-two items, items were selected to be field tested. forty These items were selected because they wer e judged by the writer to represent a range of important aspects about CAI and no two questions pertained to the same aspect. These forty ■^Computer Based Individualized Instruction Program 64 items were put into two separate test forms, A and B, for purposes of field testing. The reason for using two forms was due to the time constraint imposed by the nature of the groups available for field testing. The same groups that were used to establish the attitude scale validity and reliability were also used to field test the two forms of the knowledge instrument. addition, In all groups were administered both the attitude scale and the instrument to measure knowledge concurrently. Since the groups used for field testing were gathered for other reasons, limited. the time allocated for this purpose was Therefore, two short forms of the test were necessary rather than one long form. In all, three separate groups were used to field test the knowledge instrument, the fifty-two participants to the pre-conference seminar and the thirty-eight graduate students in the general curriculum classes at Michigan State University, previously described. number of ninety subjects, Out of this total forty-six individuals were administered form A and forty-four individuals received form B. However, the responses of two persons receiving form B were discarded because of their being incomplete. In addition, precautions which have been previously described were taken to insure that none of the individuals involved in the field testing were employed by any of the school districts that were part of this study. 65 The eighty-eight response sets resulting from the field testing were divided into two groups according to test form. The individual responses for each group were coded onto optical scanning forms in preparation for scoring and item analysis by the Office of Evaluation Services, Michigan State University. The tests were scored by the IBM 1230 Optical Scanner with the individual scores and item responses stored on magnetic tapes. These tapes were then sent to the Data Processing Department of Michigan State University where the item analysis was computed by an IBM 360/40 computer using a program written for this purpose by the Office of Evaluation S e r v i c e s . The selection of items to be included in the final instrument was made on the basis of the index of difficulty and the index of discrimination resulting from the item analysis. The index of difficulty reported in this study is the proportion of the total group who got the item wrong. Thus, a high index indicates a difficult item and a low index indicates an easy item. The index of discrimination reported in this study is the d i ffer­ ence between the proportion of the upper 27% of the total group who got an item right and the proportion of the lower 27% of the total group who got the item right. Those items which had the highest index of discrimination and an indix of difficulty as close to 50 as possible were selected. In addition, some of the selected items were 66 modified on the basis of the operation of the item distractors as determined by the item analysis. example, For if an item h a d four responses or distractors, ideally the respondents who answered the item incorrectly should select each incorrect option in roughly equal proportions rather than concentrating on a single incorrect option."^ Those selected items which had distractors which were judged as ineffective were changed in an attempt to correct this weakness. Test forms A and B along with item difficulty and discrimination indices can be found in Appendix C. on the procedures described, Based the final instrument used to measure knowledge of computer assisted instruction in this study was derived from the original forty selected items. This instrument has twenty-three items each of which has four distractors with only one being considered correct (Appendix D ) . Test Validity It is assumed that the items used in constructing the instrument have content validity by virtue of the experts who submitted them. 11 Item A n a l y s i s , East Lansing: Office of Evaluation Services, Michigan State University, Mimeographed, October, 1970, p. 2. 67 The face validity of each item and the whole scale was estimated by asking three persons with experience in computer assisted instruction to rate the face validity of each item and the whole scale on a five-point scale from 0 for "no apparent validity" through 4 for "very high face validity." An index for each item and the whole scale was calculated by adding all ratings for each item and the whole scale and dividing by twelve. Twelve is equal to the maximum rating possible multiplied by the number of judges. Thus, the maximum face validity index possible was 1.00 and the minimum was 0.00. The face validity index computed for the whole knowledge instrument was .750. The item indices for the knowledge instrument are given in Table 5 below. 68 Table 5. Face Validities of Knowledge Test Items* Item Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Face Validity .750 .833 .583 .750 .833 .833 .750 .500 .667 .750 .667 .917 .833 .667 .917 .833 .667 .750 .667 .750 .750 .750 .417 *Calculated from rating by three persons on a five-point scale with a validity index o f : 1.000 = very high face validity 0.750 = high face validity 0.500 = m e dium face validity 0.250 = low face validity 0.000 = very low face validity Thus, it can be seen that according to this procedure, the scale, as a whole, has "high face validity." Development of the Background Questionnaire A background questionnaire (Appendix E) was developed in order to gather selected bibliographical information 69 about the respondents sampled. This information was used to determine if the variables of age, professional position (teacher-principal), educational community type (rural-urban), subject matter responsibility if a secondary teacher, and educational level (secondary-elementary) are related to attitude toward computer assisted instruction. This background questionnaire was reviewed by several teachers and administrators and the format modified based on their criticism. The rural-urban classification system used in this study is based upon the one used by the Michigan Department of Education m the state assessment program. 12 This program categorizes Michigan school districts on the basis of the following community t y p e s : Type 1 Metropolitan Core: One or mor e adjacent cities with a population of 50,000 or more which serve as the economic focal point of their e n v i r o n s . Type 2 City: Community of 10,000 to 50,000 that serves as the economic focal point of its e n v i r o n s . 12 Activities and Arrangements for the Michigan A s s e s s ­ me n t of Ed u c a t i o n , Assessment Report Number Two, Prepared in the Bureau of Research, Michigan Department of Education, December, 1969, pp. 10-11. 70 Type 3 Town: Community of 2,50 0 to 10,000 that serves as the economic focal point of its environs. Type 4 Urban Fringe: A community of any population size that has as its economic focal point a m e t r o ­ politan core or a city. Type 5 Rural Community: A community of less than 2,500. For the purposes of this study, the Type 1, 2, and 4 school districts were treated collectively as urban, and Type 3 and 5 districts were treated collectively as rural. However, it should be recognized that the classi­ fication of rural and urban used in this study is a far cry from wha t is usually characterized as such. According to W e b s t e r : The test whether a tenement is rural or urban is not the place where the property is situated, but the use to which it is devoted. The report used by this- writer used the criteria of population and economic focal point of the inhabitants to establish the community type to which the school district bebngs. 13 In this respect, the classification system used Webster's New International Dictionary of the English L a n g u a g e , Second Edition, 1956. 71 in this study is limited to the definition of rural-urban used by the Michigan Department of Education state assessme n t report number 2. 14 For the purpose of this study, administrators and teachers who worked with students in grades K-6 were classified as elementary while secondary educators were classified as individuals who worked with students in grades 7-12. Geographic Area of the Study Washtenaw County, Michigan (Appendix F ) , was selected as the geographic unit to which this study was delimited. This county was considered as having all of the attributes necessary for the study. There were all five community types of school districts in the county none of which are less than K-12. The county was geographically small enough to allow the researcher to make personal contact with the sample members and lease telephone lines, which were available to the researcher, were connected to all of the target schools in the study for follow-up. Washtenaw County is a member county in Michigan's southeastern economic region. population center. 14 This region is the state's On an area representing only 8% of Activities and Arrangements for Michigan, pp. 10-11. 72 Michigan's land mass, the southeastern region supports over 55% of the state's total population. Washtenaw County has a semi-independence from m e t r o ­ politan Detroit. There is considerable industry directly related to Detroit's manufacturing activities, but the product, not the workers, travel to Detroit. Within Washtenaw County, between 1954 and 1964, both farm and farm incomes wer e becoming larger. number of farms was declining. However, the The breakdown of o c cupa­ tions for the county, based on the 1960 census, reveals that the job structure is predominantly non-agriculture with professional and technical occupations constituting the largest group in the population (23%). Thus, higher education is the county's predominant product. In 1966 one out of every five persons living in the county was a student following an education program beyond the high school level. 15 Description of the Sample The sample of this study was delimited to randomly selected secondary and elementary teachers and all of the principals from the ten school districts located in Washtenaw County, Michigan (see Table 6). The sample members worked with kindergarten through twelfth grade 15 Washtenaw County Planning Commission, Recent Economic Trends in Washtenaw C o u n t y , Ann Arbor, Michigan, July, 1966, pp. 1-29. 73 students. Thus, teachers and principals whose primary responsibility were grades K-6 were considered as elementary, and a secondary person was considered as an individual whose primary responsibility was working with students in grades 7-12. Table 6. School Districts Included in the Study* School District Location ul. Ann Arbor Public Schools Ann Arbor, Michigan R2 . Chelsea Schools Chelsea, Michigan R 3. Dexter Community Schools Dexter, Michigan R4 . Lincoln Consolidated Schools Ypsilanti, Michigan R5 . Manchester Public Schools M a n c h e s t e r , Michigan R 6 . Milan Area Schools Milan, Michigan R 7. Saline Area Schools Saline, Michigan R 8. Whitmore Lake Public Schools Whitmore Lake, Michigan U 9. Willow Run Public Schools Ypsilanti, Michigan U10. Public Schools of Ypsilanti Ypsilanti, Michigan *Those schools whose name is preceded by the symbol U were treated as urban while those w i t h the symbol R preceding their name were treated as rural. Sampling Procedures In order to carry out the study, it was necessary to I obtain lists of teachers and principals by school districts and educational level (secondary-elementary). The writer obtained these lists by calling each of the respective school districts included in this study. The writer was very fortunate in being employed by the Michigan Department of Education at the time this study was conducted. The director of E.S.E.A. Title III, Don Goodson, gave the researcher permission to make this request under the title of his office. This made the process of getting teacher and principal lists by level within the school districts relatively easy. In addition, this insured cooperation by the ten school districts involved. On the advice of Dr. Maryellen McSweeney of the Educational Psychology Department, Michigan State Univer­ sity, it was decided to stratify the sample on the variables of position and level held in education as well as community type. Age was not included because this information would not be available until the data had been gathered and the background questionnaire analyzed. The other variables were taken from the teacher lists that were furnished by the ten local districts and the state assessment report number 2. X6 In regard to the sample size needed to determine if a relationship existed between knowledge and attitude, it was determined that if 25% of the estimated variance of attitude was accounted for by knowledge, then a correlation of .50 would be significant at the .05 confidence level using a sample size of 40. ^ Activities and Arrangements for Michigan, pp. 10-11. 75 The technique of proportional sampling of teachers was employed in this study. Thus, 10% of the teachers that were classified under each variable on which the sample was stratified were randomly selected. All of the principals were included in the sample because of their small number in the total population of interest. This procedure gave a sample size that was considered adequate to test all of the hypotheses of this study. The lists of teachers from those school districts classified in this study as urban were randomly put together and those lists from districts considered rural were also randomly compiled. The researcher then con­ secutively numbered from 1 through X those teachers who were classified as elementary and the same procedure was used for those teachers considered as secondary. This method was employed for both the rural and urban classified schools. Then using a table of random numbers, 17 pro­ portional numbers of teachers were selected based on the variables used to stratify the sample. The matrices below show the numbers of principals and teachers by category in the sample of this study. 17 Sidney J. Armore, Introduction to Statistical Analysis and Inference, New York, London, Sidney: John Wiley and Sons, Inc., 1966, pp. 498-499. 76 TEACHERS Rural PRINCIPALS Urban Rural Urban Elementary 30 76 Elementary 11 38 Secondary 32 68 Secondary 12 11 Thus, the total number of subjects in the sample of this study was 2 78, or 206 teachers and 72 principals. Collection of Data The final seventy-two packets of materials were mailed to the principals in the study. The questionnaires were then distributed by the principals to the teacher(s) in their schools who were part of the study. In addition, a questionnaire was included for each principal wit h instructions to complete and return it with the others from his school. In some cases, due to the random selection procedure, a principal would not have any teacher(s) who were part of the sample of this study. in his building In that case he would receive only one questionnaire to be completed and returned by him. Each packet contained the following: 1. Cover letter to the principal* 2. Self-addressed, stamped, return envelope *This letter varied somewhat, dependent upon whether there were teachers who were part of the sample in the principals' building or not. 77 3. List of teachers to receive the questionnaire(s)** 4. Questionnaire(s) which included A. Cover letter B. Background instrument C. Attitude scale D. Test of CAI knowledge In an effort to increase the number of responses, the following additional techniques were employed: 1. The responses were requested to be returned to the E.S.E.A. Title III Office, Michigan Department of Education. 2. Each cover letter to the principal was personally signed. 3. A date for the return of the forms was given in the letters for both teachers and a d m i n istrators. 4. Follow-up letters were mailed promptly to those principals who had not returned the forms by the designated date. 5. A week after the follow-up letter was sent, those principals who had not responded were contacted by phone. * 6. The respondents were requested to remain a n o n y m o u s . **Inclusion of this list was dependent upon the random sampling procedure used. 78 7. The results of the study were promised to those schools and individuals involved in the study. A copy of the cover letters and the follow-up letter are included in Appendix G. Of the 27 8 survey instruments mailed, 256 instru­ ments were returned for a 92.0% response rate. Four of these instruments were eliminated from the analysis because they were incomplete. Therefore, 252 instruments were used for the analysis of the data which represented a 90.6% response. The breakdown of this return by categories is given below in Table 7. Table 7. Numbers and Percentages of Questionnaires Returned by Categories Number Mailed Number Returned Percent Returned Secondary urban teachers 68 67 98.6 Secondary rural teachers 32 32 Elementary urban teachers 76 65 85.5 Elementary rural teachers 30 28 90.3 Secondary urban principals 11 10 90.0 Secondary rural principals 12 10 83.3 Elementary urban principals 38 30 78.9 Elementary rural principals 11 10 90. 3 Category 100 Analysis and Preparation of the Data Each of the returned sets of instruments was prepared for analysis in the following way. All of the returned 79 instruments were checked for completeness. Incomplete instruments were not considered usable and were discarded. The background instrument had a section that was marked for dat a processing. This consisted of a series of boxes along the right side of the questionnaire arranged vertically. Those boxes that corresponded to a particular question were placed directly across from that question. Under each of these boxes were printed numbers which corresponded to columns on an eighty-column data processing card. Thus, the information could be coded by hand from the background questionnaire onto the box or boxes corresponding to that question. For example, if a school was classified as urban, a 2 would be placed in the corresponding box or a 1 if rural. The keypunch operator could then punch the code written in the box(es) onto the column in the card identified by the number under each box. In addition, there were boxes for identification I>! number, attitude score, and knowledge test score, i.e., total number of correct responses. This procedure p r o ­ vided for mos t of the data to be keypunched directly from the background questionnaire. There were six research hypotheses of the study that made predictions as to possible relationships that existed between the dependent variable attitude and several independent variables. .The acceptance or rejection of 80 these research hypotheses depended on the acceptance or rejection of their underlying null statistical hypotheses. Thus, data gathered from the statistical hypotheses were used to determine if the research hypotheses were accepted or rejected. These research hypotheses, along with their underlying null statistical hypotheses, are listed below. Research Hypothesis 1: There is a strong positive relationship between knowledge of CAI and attitude toward CAI. Statistical Hypothesis 1: The correlation between the dependent variable attitude and independent variable knowledge will be less than or equal to 5. Symbolically: Legend: Ho: r £ .50 Hi: r > .50 r = Pearson-product moment correlation coefficient. Research Hypothesis 2: Elementary principals and teachers, as a group, are significantly different in attitude toward CAI when compared to secondary principals and teachers, as a group. Statistical Hypothesis 2: There will be no difference between mean attitude scores for elementary principals and teachers and mean attitude scores for secondary principals and teach e r s . Symbolically: Ho: yi = y 2 yi Legend: ? ya yi = elementary group mean; y 2 = secondary group mean. Research Hypothesis 3: Urban principals and teachers, as a group, are significantly different in attitude toward CAI when compared to rural principals and t e a c h e r s , as a g r o u p . 81 Statistical Hypothesis 3: There will be no difference between mean attitude scores for urban principals and teachers and mean attitude scores for rural principals and teachers. Symbolically: Ho: yi = y 2 Hi: yi f y 2 Legend: yi = urban group mean; Research Hypothesis 4: y2 = rural group mean. Principals and teachers under 30 years of age, as a group, are significantly different in attitude toward CAI when compared to pri n ­ cipals and teachers 30 years of age and older, as a group. Statistical Hypothesis 4: There will be no difference between mean attitude scores for principals and teachers under 30 years of age and mea n attitude scores for teachers and principals 30 years of age and older. Symbolically: Legend: Ho: yi = y 2 Hi: yi ? y 2 yi = young (age < 30) group mean; y 2 = old (age > 30) group mean. Research Hypothesis 5: Elementary and secondary p r i n c i p a l s , as a group, are significantly different in attitude toward CAI when compared to elementary and secondary teachers, as a group. Statistical Hypothesis 5: There will be no difference bet­ ween mean attitude scores for elementary and secondary principals and mean attitude scores for elementary and secondary t e a c h e r s . Symbolically: Legend: Ho: yi = y 2 Hi: yi f y 2 yi = principals' group mean; y 2 = teachers' group mean. 82 Research Hypothesis 6: Secondary science and mathematics teachers, as a group, are signifi­ cantly different in attitude toward CAI when compared to secondary English and social science teachers, as a group. Statistical Hypothesis 6: There will be no difference between mean attitude scores for secondary science and mathematics teachers and m e a n attitude scores for secondary English and social studies teachers. Symbolically: Legend: Ho: yi = y 2 Hi: yi ^ y 2 yi = science-mathematics group mean; y 2 = Englishsocial studies group mean. To analyze the data collected during the study, the researcher, in consultation with members of the Research Consultation Office, Michigan State University, selected the following statistical treatments as appropriate for purposes of testing the statistical hypotheses stated above. The Pearson-product m o ment correlation coefficient was computed using the BASTAT 18 program. This coefficient was used to determine whether a significant relationship existed between the dependent variable, independent variable knowledge. attitude, and the Statistical Hypothesis 1 was tested in this way. 18 B A S T A T , East Lansing: Agricultural Experiment Station, Michigan State University, Mimeographed, October, 1969, pp. 2-5. The one-way analysis of variance as given by Finn 19 was used to determine if the mean attitude scores were significantly different for: elementary teachers and principals as compared to secondary teachers and p r i n ­ cipals, rural teachers and principals as compared to urban teachers and principals, teachers and principals under 30 years of age as compared to teachers and p r i n ­ cipals over 30 years of age, principals as compared to teachers, and secondary science and mathematics teachers as compared to secondary English and social studies teachers. Statistical Hypotheses 2 through 6 were tested in this way. The use of the analysis of variance in this study assumes that the individuals have been selected on the basis of random sampling from a normally distributed population. It is also assumed that the variance of each group is homogeneous and the individuals comprising each group are independent. A confidence level of .05 was selected as the critical level for statistical significance of the hypotheses investigated in this study. Thus, a hypothesis was rejected when an F-test or correlation coefficient 19 David J. Wright, Jeremy D. Finn's MultivarianceUnivariate and Multivariate Analysis of Variance and Covariance: A FORTRAN IV P r o g r a m , Occasional Paper No. 89, East Lansing: Office of Research Consultation, Michigan State University, 1970, pp. 1-8. indicated that the results obtained could occur by chance alone no more than five times in one hundred. The reliability of the attitude scale was determined by the procedure explained previously in the section on attitude scale reliability. The reliability and item analysis of the instrument to measure knowledge was computed by the Office of Evaluation Services, Michigan State University. The reliability coefficient computed is the Kuder-Richardson #20. Statistical treatments of the data in this study were conducted through the use of the facilities of the Computer Laboratory and the Office of Evaluation Services, Michigan State University. The data were processed through the use of the Control Data Corporation 3600, and the International Business Machine (CDC) (IBM) 360/40 computers. Summary The purpose of this study was to examine the r e l a t i o n ­ ship between the dependent variable attitude and several independent variables with respect to computer assisted instruction (CAI). All of the teachers and principals working in the ten public school districts located in Washtenaw County (Michigan) were chosen for the population of this study. 85 Three instruments were developed to gather data in order to answer questions pertaining to the purpose of the study. These instruments were: an instrument to measure knowledge, an attitude scale, and a background questionna i r e . All of the instruments were pretested using subjects similar to the population of the study. In addition, the construct and face validities were established for the attitude instrument and its reliability computed based on the pretest data. The A and B forms of the knowledge instrument under­ we n t an item analysis and a final instrument devised on the basis of this analysis. The face validity of this final instrument was also established. These instruments were then distributed to a random sample of teachers and all of the principals in the population of interest. The Pearson-product m o ment correlation and the o n e ­ way analysis of variance were used to analyze the survey data. The assumptions underlying the one-way analysis of variance model were examined and found appropriate for this study. CHAPTER IV ANALYSES OF THE DATA Presented in this chapter are the analyses of the data gathered from the responses of 60 principals and 192 teachers to the three instruments that were developed for the study. These three instruments include a twenty- item attitude scale, a twenty-three-item instrument to measure knowledge, and a six-item background question­ naire. There were six research hypotheses formulated by the researcher that made predictions as to the relation­ ships that existed between the dependent variable attitude and several independent v a r i a b l e s . Whether the six research hypotheses were accepted or rejected depended on whether the statistical hypothesis of each was accepted or rejected. All of the null statistical hypotheses were tested at an alpha level of .05 by either the Pearson-product moment correlation, the one-way analysis of variance, or the F-test. 86 87 Analysis of Hypotheses Research Hypothesis 1: There is a strong positive relationship between knowledge of CAI and attitude toward CAI. Statistical Hypothesis 1: The correlation between the dependent variable attitude and the independent variable k n o w ­ ledge will be less than or equal to .50. Symbolically:" Legend: Ho: r £ .50 Hi: r > .50 r = Pearson-product moment correlation coefficient. Decision rule: Table 8. Reject Ho if r > .50 at a = .05 Pearson-product Moment Correlation Between Knowledge and Attitude with Respect to CAI Variable Knowledge Attitude X SD N Pearson-r 9.80 3.04 252 .19 63.29 12.29 On the basis of the computed statistic, hypothesis was not rejected. the null Although the correlation between knowledge and attitude is positive, it is not of a sufficient magnitude to be considered significant in this study. Therefore, the research hypothesis is not supported. However, it was considered desirable to further test the computed correlation coefficient in order to determine if the relationship between knowledge and attitude was statistically significant. Thus, the following hypotheses 88 were formulated and an F-test run on the correlation to determine if it was significantly different from zero.'*’ If the probability was small that chance variations might have produced the results, then the null hypothesis would be rejected. Research Hypothesis 1A: There is a statistically signifi­ cant relationship between knowledge and attitude with respect to C A I . Statistical Hypothesis 1A: The correlation between the dependent variable attitude and the independent variable kno w ­ ledge will be equal to or less than z e r o . Symbolically: Legend: Ho: r < 0 Hi: r > 0 r = Pearson-product moment correlation coefficient. Decision Rule: Reject Ho if F > 3.89 at a = .05 F = 9.03 On the basis of the computed statistic, the null hypothesis was rejected. accepted, The alternative hypothesis was suggesting a statistically significant relation­ ship between knowledge and attitude. In addition, the computed F statistic was also found to be significant at the .01 level of confidence. Research Hypothesis 2: Elementary principals and teachers, as a group, are significantly different in attitude toward CAI when compared to secondary p rin­ cipals and teachers, as a group. ^Bernard Astie, Statistics in R e s e a r c h , Ames, Iowa: The Iowa State University Press, 1963, pp. 174-176. 89 Statistical Hypothesis 2: There will be no difference between m ean attitude scores for elementary principals and teachers and m ean attitude scores for secondary principals and teachers. Symbolically: Legend: Ho: y 1 = y 2 Hi: yi ? y 2 yi = elementary group mean; y 2 = secondary group mean. Decision Rule: Table 9. Reject Ho if F > 3.84 at a = .05 Table of Means of Attitude Scores for ElementarySecondary Teachers and Principals N Attitude Elementary 133 62.74 Secondary 119 63.89 Pooled Mean 63.32 Pooled S. D. Table 10. Source Between Within 252 12.49 One-Way Analysis of Variance on Attitude Scores of Elementary-Secondary Teachers and Principals MS 82.54 156.06 OF F-Ratio 1 0.5289 250 On the basis of the computed statistic, the null hypothesis was not rejected. hypothesis is not supported. Therefore, the research 90 Research Hypothesis 3: Urban principals and teachers, as a group, are significantly differ­ ent in attitude toward CAI when compared to rural principals and teachers as a group. Statistical Hypothesis 3: There will be no difference between mean attitude scores for urban principals and teachers and mea n attitude scores for rural principals and t e a c h e r s . Symbolically: Legend: Ho: vu = ii2 Hi: y l ? U 2 yi = urban group mean; y 2 = rural group mean. Decision Rule: Table 11. Reject Ho if F > 3.84 at a = .05 Table of Means of Attitude Scores for RuralUrban Teachers and Principals N Attitude Rural 80 63.96 Urban 172 62.97 Pooled Mean 63.47 252 Pooled S. D. Table 12. Source Between Within 12.50 One-Way Analysis of Variance on Attitude Scores of Rural-Urban Teachers and Principals MS 53.69 156.18 DF F-Ratio 1 0.3438 250 91 On the basis of the computed statistic, hypothesis was not rejected, therefore, the null the research hypothesis is not supported. Research Hypothesis 4: Principals and teachers under 30 years of age, as a group, are significantly different in attitude toward CAI when compared to p rin­ cipals and teachers 30 years of age and o l d e r , taken as a g r o u p . Statistical Hypothesis 4: There will be no difference between m e a n attitude scores for principals and teachers under 30 years of age and mean attitude scores for teachers and principals 30 years of age and older. Symbolically: Legend: Ho: yi = y 2 Hi: yi f y 2 yi = young (age < 30) group mean; y 2 = old (age > 30) group mean Decision Rule: Table 13. Reject Ho if F > 3.84 at a = Table of Means of Attitude Scores for Young (Age < 30) - Old (Age > 30) Teachers and Principals Attitude N Young Old (age < 30) (Age > 30) 101 61.80 151 64.28 Pooled Mean 63.04 D. Pooled S . : Table 14. Source .05 252 12.45 One-Way Analysis of Variance on Attitude Scores of Young (Age < 3 0 ) - Old (Age > 30) Principals and Teachers MS DF F-Ratio 2.3955 Between 371.07 1 Within 154.91 250 92 On the basis of the computed statistic, the null hypothesis was not rejected, therefore, the research hypothesis is not supported. Research Hypothesis 5: Elementary and secondary p r i n ­ cipals, as a group, are significantly different in attitude toward CAI when com­ pared to elementary and secondary teachers taken as a group. Statistical Hypothesis 5: There will be no difference between m e a n attitude scores for elementary and secondary pri n ­ cipals and mean attitude scores for elementary and secondary teach e r s . Symbolically: Legend: Ho: yx = y 2 Hi: y 1 ¥ Vz y 1 = principal group mean; y 2 = teacher group mean. Decision Rule: Table 15. Reject Ho if F > 3.84 at a = .05 Table of Means of Attitude Scores for Principals and Teachers N Principal Teacher 60 64.95 192 62.77 Pooled Mean Pooled S. D. Attitude 63.86 252 12.47 93 Table 16. One-Way Analysis of Variance on Attitude Scores of Principals and Teachers Source MS OF F-Ratio 0.2375 Between 218.13 1 Within 155.52 250 On the basis of the computed statistic, the null hypothesis was not rejected, therefore, the research hypothesis is not supported. Research Hypothesis 6: Secondary science and mathematics teachers, as a group, are signifi­ cantly different in attitude toward CAI when compared to secondary English and social studies teachers as a group. Statistical Hypothesis 6: There will be no difference between mean attitude scores for secondary science and mathematics teachers and mean attitude scores for secondary English and social studies teachers. Symbolically: Legend: Ho: yi = y 2 Hi: yi ? y 2 y 1 = science and mathematics group mean, y 2 = English and social studies group mean. Decision Rule: Table 17. Reject Ho if F > 3.84 at a = .05 Table of Means of Attitude Scores for Secondary, Science/Mathematics-English/Social Studies Teachers N Attitude Science/Mathematics 44 63.30 English/Social Studies 45 60.80 Pooled Mean Pooled S. D. 62.05 89 13.48 94 Table 18. One-Way Analysis of Variance on Attitude Scores on Secondary Science/Mathematics-English/Social Studies Teachers MS OF F-Ratio Between 138.54 1 0.7623 Within 181.73 250 Source On the basis of the computed statistic, the null hypothesis was not rejected, therefore, the research hypothesis is not supported. Table 19. Summary of Hypothesis Research Hypotheses Level at which the null h y p o ­ thesis was accepted or rejected Statement of accept or reject 1. There is a strong positive relationship between knowledge of CAI and attitude toward C A I . r = .19 The research hypothesis was not supported. 1A.There will be a statistically significant relationship between knowledge and attitude with respect to C A I . F = 9.03 The research hypothesis was supported. 2. Elementary principals and tea c h e r s , as a group, are signifi­ cantly different in attitude toward CAI when compared to secondary principals and teachers as a group. F = 0.53 The research hypothesis was not supported. 95 Table 19 (Cont'd) Research Hypotheses Level at which the null hypothesis was accepted or rejected Statement of accept or reject 3. Urban principals and teachers, as a group, are significantly different in attitude toward CAI when com­ pared to rural principals and teachers as a group. F = 0.34 The research hypothesis was not supported. 4. Principals and teachers under 30 years of age, as a group, are signifi­ cantly different in attitude toward CAI when compared to principals and teachers 30 years of age and older as a group. F = 2.40 The research hypothesis was not supported. 5. Elementary and secondary principals, as a group, are signi­ ficantly different in attitude toward CAI when compared to elementary and secondary teachers taken as a group. F = 0.24 The research hypothesis was not supported. 6. Secondary science and mathematics t e a c h e r s , as a group, are signi­ ficantly different in attitude toward CAI when compared to secondary English and social studies teachers as a group. F = 0.76 The research hypothesis was not supported. 96 Analysis of the Knowledge Instrument The instrument used to measure knowledge in this study underwent an item analysis based on the responses to the instrument taken from the survey data. In addition, the Kuder-Richardson #20 reliability coefficient was computed. The summary statistics for the knowledge instrument are given in Table 20 below, while the detailed item analysis can be found in Appendix H. Table 20. Knowledge Instrument Summary Statistics (N = 252) Mean 9.80 Standard deviation 3.04 Variance 9.26 Mean item difficulty 57.00 Mean item discrimination 32.00 4605 Kuder-Richardson reliability #20 Analysis of the Attitude Scale The sample data was analyzed to determine the attitude instrument's reliability coefficient, to derive a set of optimum weights for each item, and to establish the item response pattern based on the sample of the study. The RAVE section of the FORTAP 2 program, using a 2 Wright, FORTAP, A Fortran Test Analysis P a c k a g e , pp. 8-13. 97 method of reciprocal averages as given by W r ight and Porter, 3 gave an optimum weighting scale for each item from 0 through 5. The assigned weights and item response pattern can be found in Appendix I and J respectively. Table 21 contains the summary statistics of the attitude scale obtained from the sample data. Table 21. Attitude Scale Summary Statistics (N = 252) Mean 63.2850 Standard deviation 12.4806 Standard error 3.0982 Hoyt's reliability coefficient 0.9332 Summary Seven null hypotheses were tested in an attempt to answer questions pertaining to the relationship between attitude and other variables with respect to computer assisted instruction. All of the hypotheses were tested at the of significance. .05 level The findings are summarized in Table 19, which lists the hypothesis, the level at which it was accepted or rejected, and a rejection or acceptance statement. 3 David J. Wright and Andrew C. Porter, An Adaptation of Frank B. Baker's Test Analysis P a c k a g e , Occasional Paper No. 1, East Lansing: Office of Research Consultation, Michigan State University, 1968, p. 13. 98 Thus, it is concluded that the following statement is support a b l e : There is a statistically significant positive relationship between knowledge and attitude wit h respect to CAI. The instrument to measure knowledge was analyzed to determine its reliability and to compute an item analysis based on the sample data. The instrument to measure attitude toward CAI was analyzed to determine its reliability, item response pattern, and a set of optimum weights for each item based on the sample data. CHAPTER V SUMMARY AND CONCLUSIONS Summary The purpose of this study was to examine the relation­ ship between the dependent variable attitude and several independent variables with respect to computer assisted instruction (CAI). (1) knowledge; field; These independent variables included: (2) age; (3) level in the educational i.e., elementary-secondary; (4) classification of the school district employing the sample member; rural-urban; i.e., (5) position in the education field, i.e., principal-teacher; and secondary teachers, (6) subject matter taught by i.e., English/social studies-science/ mathematics. Three instruments were developed to gather data in order to answer questions pertaining to the purpose of the study. The instruments were an attitude scale, an instrument to measure k n o w l e d g e , and a background questionna i r e . All of the instruments were pretested using subjects similar to the population of the study. In addition, the construct and face validities were established for the attitude instrument and its reliability computed based on 99 100 the pretest data and the face validity of the modified instrument established. These instruments were then distributed to a random sample of teachers and all of the principals in the population of interest (Washtenaw County, M i c h i g a n ) . Of the 276 sets of instruments mailed, 256 sets of instru­ ments were returned for a 92.0% response. Seven null hypotheses were tested in an attempt to answer questions relative to the purpose of the study. All of the hypotheses were tested at the .05 level of significance by either the one-way analysis of variance, Pearson-product m o ment correlation, or the P-test. In addition, internal consistency reliabilities were calculated for the attitude scale and knowledge instrument. T h u s , it was found that the following statement is supportabl e : There is a statistically significant positive relationship between knowledge and attitude with respect to C A I . Findings ^ H o , : The correlation between the dependent variable atti­ tude and the independent variable knowledge will be less than or equal to .50. Hypothesis accepted. HO 2 : There will be no difference between mean attitude scores for elementary principals and teachers and mean attitude scores for secondary principals and teachers. Hypothesis accepted. 101 Ho^: There will be no difference between mean attitude scores for urban principals and teachers and mean attitude scores for rural principals and t e a c h e r s . Hypothesis accepted. H o ^ : There will scores for of age and principals accepted. be no difference between mean attitude principals and teachers under 30 years mean attitude scores for teachers and 30 years of age and older. Hypothesis Ho,.: There will be no difference between mean attitude scores for elementary and secondary principals and mean attitude scores for elementary and secondary teachers. Hypothesis accepted. H O g : There will be no difference between mean attitude scores for secondary science and mathematics teachers and mean attitude scores for secondary English and social studies teachers. Hypothesis accepted. It was decided to further analyze the data in an attempt to clarify the findings with regard to the relationship between knowledge and attitude since there is some confusion indicated in the literature in this regard (see page 14). Thus, the following additional hypothesis was formulated and tested: Ho: The correlation between the dependent variable attitude and the independent variable knowledge will be equal to or less than zero. Hypothesis rejected. Deliminations of the Study This study is restricted to: 1. Elementary grades defined as grades defined as (K-6) (7-12). 2. Principals and teachers. 3. School districts with grades K-12. and secondary 102 4. Public schools. 5. Washtenaw County (Michigan). Limitations of the Study 1. In this study, the attitudes of teachers and principals toward computer assisted instruction are considered of prime importance. However, an attitude scale, such as the one used in this study, actually measures. there is doubt as to what What the scale does measure, at best, is the expressed reaction of a person responding to written statements. What relationship exists b e t ­ ween this expressed reaction and "true" attitude is not known. 2. The traditional classification of rural-urban is based on the livelihood of the population. However, in this study the classification of rural-urban is dependent upon the population size of a community, and livelihood is not considered. 3. It is possible that the measure for knowledge does not cover all of the important aspects of computer assisted instruction. 4. It is possible that the reliability of the measure used to indicate teacher and principal knowledge of computer assisted instruction is not sufficient to be of a high predictive value. 103 5. All subjects in the survey were from a county that had one out of every six persons in the population enrolled in a program of higher education. Thus, the sample members could have been influenced by particular requirements and experiences in an institution of higher learning. Conclusions Within the limitations and delimitations of this study, the following conclusions were supported with respect to computer assisted instruction: 1. The correlation between knowledge and attitude was less than .50. 2. There was no significant difference between m ean a t t i ­ tude scores for elementary principals and teachers and m ean attitude scores for secondary principals and teachers. 3. There was no significant difference between mea n att i ­ tude scores for urban principals and teachers and mea n attitude scores for rural principals and teachers. 4. There was no significant difference between m e a n atti­ tude scores for principals and teachers under 30 years of age and mean attitude scores for principals and teachers 30 years of age and older. 5. There was no significant difference between m ean atti­ tude scores for elementary and secondary principals and mean attitude scores for elementary and secondary teachers. 104 6. There was no significant difference between mea n atti­ tude scores for secondary science and mathematics teachers and mea n attitude scores for secondary English and social studies teachers. 7. There was a statistically significant positive corre­ lation between knowledge and attitudes. Discussion In summary, this study revealed that many assumptions that have been made with respect to computer assisted instruction should possibly be questioned. In talking with individuals using CAI, the writer has been told that: A. Science and mathematics teachers would be more positive in attitude toward CAI than English and social studies t e a chers. B. Secondary educators would be less positive in attitude toward CAI than elementary educators. C. Young educators (age < 30 years) would be more positive in attitude toward CAI than older educators (age > 30 years). In addition, the writer assumed that: D. Principals would be more positive in attitude toward CAI than teachers. E. Rural principals and teachers would be less positive in attitude toward CAI than urban principals and teachers. 105 However, on the basis of this study, these assump­ tions have not been justified. In fact, the mean attitude score for teachers and principals whose age is less than 30 years, (x = 61.80), is lower than the mean attitude score for principals and teachers whose age is 30 years or greater, (x = 64.28). statistically significant, While this difference is not it is nevertheless interesting in that the means are contrary to expectations in light of the assumption. With regard to the attitude of principals and teachers in the field toward CAI, the findings of this study indi­ cate that it is generally favorable. This is reflected in Table 21 which shows a mea n attitude score of 63.29. In general, however, the knowledge of teachers and principals concerning CAI was found to be quite low. In looking at Table 20, one can see that the m ean knowledge score was 9.80. While the relationship between knowledge and attitude was not as strong as expected; i.e., r = .50, further analysis of the data supported the notion that this relationship is both positive and statistically significant. T h u s , the findings of this study tend to support the r e ­ search reported by Christopher."*■ ■^Christopher, Influence of Computer Assisted Instruc­ tion Upon the Attitudes of A d m i n istrators, p. 69. 106 The general conclusion made from the study is that of the six variables taken into consideration, only k n o w ­ ledge seems to be related to attitude; i.e., knowledge in relation to the specific attitude object. Implications for Further Research Some of the questions posed in this study for future research are: 1. Would a study replicating the present study verify the findings obtained? 2. Would the results from this survey study be supported by an experimental study, with subjects randomly assigned to treatment and control groups where the treatment would be designed to increase knowledge? 3. In schools involved with C A I , what is the relationship between teacher attitude, the variables investigated in this study, and pupil achievement with respect to CAI? 4. What effect does the type of administrative leadership, sex, years of teaching and/or administrative experience, and type of degree held have on teacher-principal attitude with respect to CAI? BIBLIOGRAPHY BIBLIOGRAPHY Books Armore, Sidney J . , Introduction to Statistical Analysis and I n f e r e n c e , New York, London, Sidney: John Wiley and Sons, Inc., 1966. Atkinson, Richard C. and Wilson, Hal.A. eds., ComputerAssisted Instruction: A Book of R e a d i n g s , New York, London: Academic Press, 1969. Edwards, Allen L., Techniques of Attitude Scale Construc­ t i o n , N e w York: Appleton-Century-Crofts, 1957. English, Horace B. and English, Ava C., A Comprehensive Dictionary of Psychological and Psychoanalytic T e r m s : A Guide to U s a g e , New York: McKay, 1958. Hays, W illiam L., S t a t i s t i c s , N e w York, Chicago, San Francisco, Toronto, London: Holt, Rinehart and Winston, 1963. Hickey, Albert E., Computer-Assisted Instruction: A Survey of the L i t e r a t u r e , Newburyport: ENTE LEK, I n c . , 1968. Jackson, Philip W . , Life in Cl a s s r o o m s , New York: Rinehart and Winston, 1968. Holt, Krech, David, Crutchfield, Richard S. and Ballachey, Egerton L., Individual in Society, New York: McGrawHill, 1962. Lekan, Helen A., Index to Computer-Assisted I n s t r u c t i o n , Milwaukee: Instructional Media Laboratory, University of Wisconsin, 1969. Marker, Robert, e d . , Computer Concepts and Educational A d m i n i s t r a t i o n , University of Iowa: Iowa Educational Information Center, 1968. Ostle, Bernard, Statistics in R e s e a r c h , Ames, Iowa: Iowa State University Press, 1963. The 108 Roth, Gustave J., Anderson, Nancy S. and Brainerd, R. C . , The IBM Research Center Teaching Machine Project in Automated Teaching; The State Of the A r t , e d . , E. Galanter, N e w York: John Wiley and Sons, 1959. Shaw, Marvin E. and Wright, Jack M., eds., Scales for the Measurement of Attitudes, N e w York: Disney, McGrawHill Book Co., 1967. Skinner, B. F . , The Technology of T e a c h i n g , New York: Appleton-Century-Crofts, 1968. Thurstone, Louis L. and Chave, Ernest J . , The Measurement of A t t i t u d e s , Chicago: The University of Chicago Press, 1929. Articles Allen, Dwight W. and Hawkes, Glenn W . , "Reconstruction of Teacher Education and Professional Growth Programs or H o w the Third Little Pig Escaped the Wolfe," Phi Delta K a p p a n , Vol. 52, No. 1, 1970. Allpart, Gordon W . , "The Historical Background of Modern Social Psychology," in Handbook of Social P s y c h o l o g y , Vol. 1, Cambridge, Mass.: A d d i s on-Wesley, 1954. Alpert, D. and Bitzer, D. L . , "Advances in Computer-Based Education," S c i e n c e , Vol. 167, No. 3917, 1970. Anastasiow, Nicholas J. and Jerman, Max, "Introduction to Computer-Based Drill and Practice in Arithmetic," in H a n d b o o k , L. W. Singer Co., 1968. Bitzer, Maryann D. and Boudreaux, Martha C., "Using a Computer to Teach Nursing," Reprint from Nursing Form, Vol. 8, No. 3, 1969. Borman, Karl G., "Expressed Student Opinion Toward ComputerAssisted Instruction," in inservice Mathematics Ed u c a ­ tion Via Computer-Assisted~Instruction for Elementary School Teachers xn A p p a l a c h i a , Final Report No. R - 2 6 , Bueschel, Richard T . , "Time-Sharing, A Programatic Approach in the School," Educational Technology, Vol. 10, No. 3, 1970. Castleberry, S. and Lagowski, J. J . , "Individualized Instruc­ tion Using Computer Technqies," Journal of Chemical E d u c a t i o n , Vol. 47, No. 2, 1970. 109 Doty, Barbara A. and Doty, Larry A . , "Programmed Instruc­ tion Effectiveness in Relation to Certain Student Characteristics," Journal of Educational P s y c h o l o g y , Vol. 55, No. 6, 1964. Feldhusen, John and Szabo, Michael, "The Advent of the Educational Heart Transplant, Computer-Assisted Instruction: A Brief Review of Research," Contemporary E d u c a t i o n , Vol. 40, No. 6, 1969. Feldman, David H. and Sears, P. S., "Effects of ComputerAssisted Instruction on Children's Behavior," Educational T e c h nology, Vol. 10, No. 3, 1970. Filep, Robert T . , "Individualized Instruction and the Computer: Potential for Mass Education," A-V Communications R e v i e w , Vol. 15, No. 1, 1967. Frey, Sherman, Shimabukure, Shinkichi and Woodruff, A. B., "Attitude Change in Programmed Instruction Related to Achievement and Performance," A-V Communications R e v i e w , Vol. 15, No. 2, 1957. Gerard, R. W . , "Computers: Their Impact Upon Society," Journal of American Information Processing S o c i e t i e s , Vol. 27, No. 1, 1965. Gilman, D. A. and Moreau, N. A., "Effects of Reducing Verbal Content in Computer-Assisted Instruction," A - V Communications Review, Vol. 17, No. 3, 1969. Grayson, Lawrence P., "A Paradox: The Promises and Pitfalls of CAI," E D U C O M , Vol. 5, No. 2, 1970. Hall, Keith A., Adams, Marilyn and Tardibuono, John, Gradient and Full Response Feedbacks in ComputerAssisted Instruction," in Experimentation With Computer-Assisted Instruction in Technical E d u c a t i o n , Semi-Annual Progress Report R - 6 , University Park, Pennsylvania: Computer-Assisted Instruction Laboratory, The Pennsylvania State University, June, 1967. Hansen, Duncan N. and Harvey, W. L . , "Impact of CAI on Classroom Teachers," Educational Tec h n o l o g y , Vol. 10, No. 2, 1970. Hedlund, Dalva C. and Neidt, Charles 0., "The Relationship Between Changes in Attitude Toward a Course and Final Achievement," Journal of Educational R e s e a r c h , Vol. 61, No. 2, 1967. 110 Hill, Russell A. and Nurst, Norma, "Teacher Behavior in CAI Classrooms," Educational Technology, Vol. 9, No. 2, 1969. Hoyt, Cyril J . , "Test Reliability Estimated by Analysis of V a r i a n c e ," in Principles of Education and Psycho­ logical Measurement: A Book of Selected R e a d i n g s , ed. by William Merhrens and Robert L. Ebel, Chicago: Rand McNally & Co., 1967. Katz, Daniel and Stotland, Ezra, "A Preliminary Statement to a Theory of Attitudinal Structure and Change," in Psychology: A Study of a S c i e n c e , ed. by Edward S. Koch, N e w York: McGraw-Hill, 1959. Kemeny, John C. and Kurtz, Thomas C., "Dartmouth Tim e ­ sharing," S c i e n c e , Vol. 162, No. 3850, 1968. Leslie, Larry L . , "Increasing Response Rates to Long Questionnaires," The Journal of Educational Research, Vol. 63, No. 8, 1969^ Likert, Rensis A . , "A Technique for the Measurement of Attitudes," Archives of Psychology, Vol. 22, No. 140, 1932. "Man, Ritual, the Establishment and Instructional Tech­ nology," Educational Technology, Vol. 8, No. 20, 1968. Mastin, Victor E., "Teacher Enthusiasm," Journal of Ed u c a ­ tional R e s e a r c h , Vol. 56, No. 7, 1963. Mathis, Arthur, Smith, Timothy and Hansen, Duncan, "College Students' Attitudes Toward Computer-Assisted Instruc­ tion," Journal of Educational Psychology, Vol. 61, No. 1, 1970. Mortensen, Earl M. and Penick, Ronald J . , "Computer Animation of Molecular Vibrations: Ethane," Journal of Chemical E d u c a t i o n , Vol. 47, No. 2, 1970. Neale, Daniel C . , Gill, Noel, and Tismer, Werner, Relationship Between Attitudes Toward School Subjects and School Achievement," Journal of Educational R e s e a r c h , Vol. 63, No. 5, 1970. O'Neil, H. F., Spielberger, C. D. and Hansen, D. H., "Effects of State Anxiety and Task Difficulty on Computer-Assisted Learning," Journal of Educational P s y c h o l o g y , Vol. 60, No. 5, 1969. Ill O'Toole, John F., "Teachers' and Principals' Attitudes Toward Programmed Instruction in the Elementary School," A - V Communications Review, Vol. 12, No. 4, 1964. Schurdak, John J . , "An Approach to the Use of Computers in the Instructional Progress and an Evaluation," The American Educational Research Journal, Vol. 4, No. 1, 1967. Schwartz, H. A. and Haskell, R. J., "A Study of ComputerAssisted Instruction in Industrial Training," Journal of Applied P s y c h o l o g y , Vol. 50, No. 5, 1966. Schwartz, H. A. and Long, H. S., "A Study of Remote Industrial Training," Journal of Applied Psychology, Vol. 51, No. 1, 1967. Stolurow, Lawrence M., "Computer-Assisted Instruction," Education Automation Monograph S e r i e s , D e t r o i t : American Data Processing Inc., 1968. _________, "Some Factors in the Design of Systems for Computer-Assisted Instruction," in Computer-Assisted Instruction; A Book of Readings, ed. by Richard C. Atkinson and Hal A. Wilson, N e w York, London; Academic Press, 1969. Suppes, Patrick, "The Uses of Computers in Education," Scientific A m e r i c a n , Vol. 215, No. 3, 1966. _________ , and Jerman, Max, "Computer-Assisted Instruction," The Bulletin of the National Association o f Secondary School Principals, Vol. 54, No. 343, 1970. Suydan, Marilyn N. and Mitzel, Harold E., "Evaluation of Attitudes Toward M a t h e m a t i c s ," in Inservice M a t h e ­ matics Education via Computer-Assisted Instruction for Elementary School Teachers in A p p a l a c h i a , Final Report No. R-2 6, University Park, Pennsylvania; Computer-Assisted Instruction Laboratory, The Pennsylvania State University, January, 1970. Tabbutt, Frederick D . , "Computers in Chemical Education," Chemical and Engineering N e w s , Vol. 48, No. 3, 1970. Tobias, Sigmund, "Effects of Attitudes to Programmed In ­ struction and Other Media on Achievement from Programmed Materials," A-V Communications Review, Vol. 17, No. 3, 1969. 112 ________, "Teaching Machines and Programmed Instruction," A - V Communications R e v i e w , Vol. 14, No. 1, 1966. Vidaya, Joshi, "Attitude Toward Reception of Technology," Journal of Social P s y c h o l o g y , Vol. 58, No. 1, 1962. Wing, R. L, "Two Computer-Based Economics Games for Sixth Graders," American Behavioral Scientist, Vol. 10, No. 1, 1966. Dissertation Abstracts Buckeye, Donald A., "The Effects of a Creative Classroom Environment on the Creative Ability of Prospective Elementary Mathematics T e a c h e r s ," Dissertation A b s t r a c t s , Vol. 29, No. 6, 1968. Cheikin, Martin L., "An Investigation of the Effect of Measured Teacher Attitude on Selected Eighth Grade Students," Dissertation Abstracts, Vol. 28, No. 10, 1967. Pinety, Mildred C . , "The Relationship Between Teachers' Attitudes and Effectiveness in the Classroom," Dissertation A b s t r a c t s , Vol. 24, No. 6, 1963. Shatlein, Stephen D . , "A Study of the Change of Attitudes Toward Mathematics of Prospective Elementary T e a c h e r s ," Dissertation A b s t r a c t s , Vol. 29, No. 9, 1968. Young, Norman E., "Effectiveness of Three Approaches to the Teaching of a Methods Course in Mathematics for Pre-Service Elementary Teachers as Related to Under­ standing and Attitude," Dissertation Abstracts, Vol. 29, No. 9, 1968. Theses Christopher, George R . , The influence of a Computer Assisted instruction Experience Upon the Attitudes of School A d m i n i s t r a t o r s , Unpublished doctoral thesis, The Ohio State University, 1969. L o n g , Harvey S ., A Determination of the Relation of the Total Time for Course Completion to the Duration of the Study Interval in Teaching Via Computer Assisted I nstr u c t i o n } Unpublished doctoral t h e s i s , N e w York University, 1968. 113 Pearson, Harry A., Programmed Instruction for Groups of Teachers in Remote L o c a t i o n s ; Prototype D e v e l o p m e n t , Unpublished doctoral thesis, Michigan State Univer­ sity, 1969. Stilman, Paul A., A Normative Evaluation of a ComputerAssisted Instruction Laboratory in Statistical In ­ ference , Unpublished masters thesis, University of Pittsburgh, 1969. Reports Activities and Arrangements for the Michigan Assessment of E d u c a t i o n , Assessment Report No. 2, Prepared in the Bureau of Research, Michigan Department of Education, December, 1969. American Associated of School Administrators Committee on Electronic Data Processing, EDP and the School A d m i n i s t r a t o r s , Washington, D. C . : 1967. Computers in Higher E d u c a t i o n , Report of the President's Science Advisory Committee, Donald F. Hornig, Chairman, Washington, D. C . : Government Printing Office, 1967. Diamond, James J . , A Report on Project Grow: Philadelphia's Experimental Program in Computer-Assisted Instr u c t i o n , Philadelphia, Pennsylvania: Office of Research and Evaluation, The School District of Philadelphia, August, 1969. Equality of Educational O p p o r t u n i t y , Washington, D. C . : U. S. Government Printing Office, 1966. Gallagher, Paul D., An Investigation of Instructional Treatments and Learner Characteristics in a ComputerManaged Instruction C o u r s e , Tech. Report No. 12, Tallahassee, Florida: Computer-Assisted Instruction Center, Florida State University, 1970. Maroline, Joseph B. and Misch, Marion R . , Education in the 7 0 's, Final Report on Educational Policy Prject, George Washington University, Washington, D. C . : Autumn, 1967. Scrivens, Robert W . , Evaluation Monograph No. 1 , Waterford, Michigan: U.S.O.E. Grant 67-04301-0, Waterford Township School District, February, 1970. 114 S l a v e n s , T h o m a s , Development and Testing of Materials for Computer-Assisted Instruction in the Education of Reference Li b r a r ia n s , U. S. Department of Health, Education and Welfare, Project No. 80560, 1970. Uttal, Wi l l i a m R . , "My Teacher has Three Arms!!I" I.B.M. Research Report RC-788, 1962. Washtenaw County Planning Commission, Recent Economic Trends in Washtenaw C o u n t y , Ann Arbor, Michigan, July, 1966. Others B A S T A T , East Lansing: Agricultural Experiment Station, Michigan State University, Mimeographed, October, 1969. Hansen, Duncan N . , "Myths that Need to be Destroyed and Myths that we Ought to Create," Speech given at the National Conference on Computer Applications to Learning, Bloomfield Hills, Michigan, July 9, 1970. Item A n a l y s i s , East Lansing: Office of Evaluation Services, Michigan State University, Mimeographed, October, 1970. The Computer in E d u c a t i o n , Dayton, Ohio: an I/D/E/A Occasional Paper, a publication of the Institute for Development of Educational Activities, Inc., 1970. Webster's New International Dictionary of the English L a n g u a g e , Second Edition, 1956. Wright, David J., F O R T A P : A Fortran Test Analysis Package b y F. B. Baker and T. J. M a r t i n , Occasional Paper No. 10, East Lansing: Office of Research Consultation, Michigan State University, L970. , Jeremy D. Finn's Multivariance-Univariate and Multivariate Analysis of Variance and Covariance^ A FORTRAN IV P r o g r a m , Occasional Paper No. 89, East Lansing: Office of Research Consultation, Michigan State University, 1970. _______/ and Porter, Andrew C . , An Adaptation of Frank B. Baker's Test Analysis P a c k a g e , Occasional Paper No. 1, E a s t Lansing: Office of Research Consultation, Michigan State University, 1968. APPENDICES Appendix A Attitude Scale Appendix A Attitude Scale In the Attitude Scale, the term "Computer Assisted Instruction" refers to a meth o d of instruction in which subject matter is presented by a computer. The person is instructed to make responses by means of a "terminal"; usually a device similar to an electric typewriter. There are 20 statements about Computer Assisted Instruction. Consider each statement separately and indicate the extent to which you agree or disagree with it by circling the appropriate symbol to the right of the statement. The symbols used are: SA A N D SD - Strongly agree Agree No opinion Disagree Strongly disagree All responses will be treated confidentially. 115 116 Appendix A (Cont'd) 1. I am very interested in learning about computer assisted instruction. 1. SA A N D SD 2. Teaching machines can individualize instruction more effectively than other methods. 2. SA A N D SD 3. Computer assisted instruction is an impersonal teaching a p p roach. 3. SA A N D SD 4. Computer assisted instruction will improve instructional programs. 4. 5. Computer assisted instruction challenges the student to do his best. 6. I would prefer to take a course by computer rather than by conventional instruction. 7. Use of teaching machines causes students to feel isolated. 5. 6. SA A N D SD SA A N D SD SA A N D SD 7. SA A N D SD 8. Use of the computer for data processing activities is more important than use of the computer for instruction. 8. SA A N D SD 9. Computer assisted instruction is based on the same principles as good classroom teaching. 9. SA A N D SD 10. I am uneasy about the use of computers for teaching youth. 10. SA A N D SD 11. Computer assisted instruction can develop problem-solving techniques. 11. SA A N D SD 12. Teaching machines are inflexible mediums. 12. SA A N D SD 13. Most elementary students would be adversely affected by computerized instruction. 13. SA A N D SD 14. I would prefer to take a course by conventional instruction rather than by computer assisted instruction. 14. SA A N D SD 15. Teaching by machine will tend to dehumanize the curriculum. 15. SA A N D SD 117 Appendix A (Cont'd) 16. The advocates for computer assisted instruction should press harder for its adoption. 16. SA A I D SD 17. By using computer assisted instruction, a teacher will probably become a better teacher. 17. SA A I D SD 18. Computer assisted instruction threatens the t e a c h e r 1s r o l e . 18. SA A t D SD 19. Educators will find computer assisted instruction techniques successful. 19. SA A f D SD 20. Computer assisted instruction hinders the social development of the student. 20. SA A IS D SD Appendix B Letter to CAI Experts Appendix B Letter to CAI Experts May 4, 1970 Dear Sir: Literature indicates that attitudes are important for the success of failure of any new instructional method. With a n e w expensive instructional strategy such as CAI, attitudes of the educators and people involved may be critical to the project's success. If attitudes can be influenced by some variable, then it m a y be possible to create positive attitudes by manipulating this variable. In view of this, I am in the process of developing a test to measure knowledge of CAI. This test will be used in combination with an attitude scale measuring attitudes to ­ ward C A I . The purpose of the test and attitude scale will be two-fold: (1) to evaluate part of a national conference and (2) to gather information for m y dissertation. I am sure you are aware of the national conference to be held in Michigan this summer, which I hope you will attend. It is at this conference that I propose to field test these instruments using those Michigan educators present. It was suggested by Ron Arnold, the director of INDICOM, that I gather as many test items as possible from experts in the field. I know from working with Ron and the project that your schedule is very crowded, but if you could find a moment either to submit some test items of a multiple choice type or give m e information concerning where I might find these kinds of items, it would be very m u c h appreciated. Since time is extremely critical to the study and your r e ­ sponse is equally vital, I have enclosed a self-addressed, air mail envelope to insure a quick return. 118 119 Appendix B (Cont'd) I am looking forward to seeing you this summer at the conference. Sincerely, Carlton P. Robardey, Sr. Coordinator for: National Conference on "Computer Applications to Learning" Encl. Appendix C Indices of Difficulty and Discrimination for Knowledge Instruments A and B Appendix C Indices of Difficulty and Discrimination for Knowledge Instruments A and B Form A In this test select the best possible answer for each question. Place the letter of the correct answer (found under each question) in the blank to the left of the number of that question. There is only one best answer for each question. PLEASE ANSWER ALL OF THE QUESTIONS If you are in doubt as to the correct answer, put down the response that you think is best. 120 121 Appendix C (Cont'd) DIFF 68 _____ DISC 50 1. In terms of instruction the m o s t efficient method for storing Computer Assisted in­ struction curriculum, as students interact with the computer, is by means of: *A. B. C. D. DIFF 33 _____ DISC 33 2. Computer Assisted Instruction is considered by man y experts as a part of: A. *B. C. D. DIFF 61 _____ DISC 50 3. 4. Computer Computer Computer Computer Assisted Testing Managed Instruction Based Vocational Guidance Managed Administration Most Computer Assisted Instruction systems use: A. B. *C. D. DIFF 8 DISC 33 Magnetic disc Magnetic tape Magnetic card Punched card Student Student Student Student terminals terminals terminals terminals real time up time on line off line In a tutorial Computer Assisted Instruction program, a student knows if his response is correct: A. By the time he completes his lesson *B. Within a few seconds C. As soon as the teacher informs him D. Within a few minutes DIFF 32 _____ DISC 33 5. Computer Assisted Instruction can be used to individualize instruction: A. If the programs are of a branching type B. If the programs are of the Skinnerian type *C. If the curriculum writer plans for individualization D. If there are enough terminals for each student ♦Indicates the correct response 122 Appendix C (Cont'd) DIFF 85 DISC 9 6. The best classification of Computer Assisted Instruction programs is: A. B. C. *D. DIFF 67 DISC 33 7. At the present time, which of the following can't a Computer Assisted Instruction system do: A. *B. C. D. DIFF 76 DISC 0 Analyze Analyze Analyze Analyze typed responses spoken responses tactil responses optical responses 8 . Most languages used for Computer Assisted Instruction curriculum production do not have: *A. B. C. D. DIFF 59 DISC 17 Linear Branching A and B None of the above 9. Efficient computational modes Adequate vocabulary Adequate response time The ability to produce branching programs Cathode Ray Tubes are used: A. In the operation of the Buffer Storage Unit *B. For the display of information C. In the operation of the Data Reduction Unit D. A and C DIFF 54 _____ DISC 25 10. Effective Computer Assisted Instruction programs are very difficult to construct: A. Because they are very expensive B. Because very little is known about how students learn *C. Because they require detailed planning by the author D. Because of the tremendous problem of coding them into machine language 123 Appendix C (Cont'd) DIFF DISC 9 6 11. in its best application Computer Assisted Instruction is a: A. *B. C. D. DIFF 33 _____ DISC 17 Testing and evaluation tool Supplement to classroom instruction Replacement for classroom instruction Record keeping system for the teacher 12. From the standpoint of effective learning, the greatest advantage to the student of using the computer i s : A. The repetition gained in drill and practice B. The creative use of programmed instruc­ tion *C. The immediate feedback to the student D. The automatic sequencing of instruc­ tional frames DIFF 62 _____ DISC 34 13. Which of the following is mos t often true of students taking lessons from a computer: A. They have a difficult time learning to use the complex equipment *B. They get bored quickly with poor in­ struction C. They tend to vandalize the equipment D. It takes a long time for them to adjust to the new instructional situation DIFF 72 _____ DISC 33 14. When students are working at the terminals their teacher: A. Is usually present supervising and assisting the students with the material B. Can be present but usually is not C. Is present but acts only as a supervisor for social control *D. No generalization can be made DIFF 16 _____ DISC 25 15. Which of the following types of programs is the easiest for curriculum writers to construct: *A. Drill and Practice B. Simulation C. Tutorial D. Dialogue 124 Appendix C (Cont'd) DIFF 37 DISC 16 16. One of the primary advantages of Computer Assisted Instruction curriculum design to the teacher i s : A. Teachers have a chance to become better acquainted with each learner B. Instruction is automatically individual­ ized C. It is not as much work as regular class­ room unit design *D. It forces teachers to look carefully at purposes and procedures of instruc­ tion DIFF 28 DISC 17 17. Computer Assisted Instruction, in its present form, is able to present material to students by means of: A. Printed information on paper B. Projected information on a screen or by a television type device C. Auditory information D. A and B * E . All of the above DIFF 57 DISC 25 18. Which of the following types of Computer Assisted Instruction programs is designed to allow the learner to manipulate the machine rather than the machine manipulating the l e a r n e r : A. B. *C. D. DIFF 34 DISC 17 19. Simulation Tutorial Problem solving Drill and Practice In a well designed instructional program to be given by computer, a person could follow the sequence of steps the learner could go through by looking at t h e : A. B. *C. D. Address registers Autochart Flowchart Flow direction 125 Appendix DIFF 59 DISC 33 (Cont'd) 20. If a computer for a Computer Assisted Instruction project is said to have 35,000 positions of core storage, this would be referring to t h e : A. Buffer storage unit *B. Central processing unit C. Data reduction unit D. External storage unit 126 Appendix C (Cont'd) Form B In this test select the best possible answer for each question. Place the letter of the correct answer (found under each question) in the blank to the left of the number of that question. There is only one best answer for each q u e stion. PLEASE ANSWER ALL OF THE QUESTIONS If you are in doubt as to the correct answer, put down the response that you think is best. 127 Appendix C (Cont'd) DIFF 60 _____ DISC 0 1. The time it takes a computer to respond to a student's inquiry is affected by: *A. B. C. D. DIFF 33 _____ DISC -9 2. The sophistication of the program Cycling time per student Differential analysis reaction time Frequency response analysis The objective of a good Computer Assisted Instruction program is to: A. Standardize instruction in each school district *B. Individualize instruction C. Compensate for poor instruction D. All of the above DIFF 34 _____ DISC 37 3. Computer Assisted Instruction is an effective means of instruction because: A. Large segments of learning can be covered quickly B. A student's assignment can be predeter­ mined by his past performance on given concepts C. Instruction can be individualized * D . All of the above DIFF 31 _____ DISC 55 4. At this point in time the greatest disadvantage of the computer in terms of instruction i s : A. The computer can only communicate with the student by typed words or projected images B. The computer can only be used to teach supplementary materials *C. The limited curriculum materials available for use in a Computer Assisted instruction program D. The absence of direct teacher observa­ tion makes it diifficult to determine pupil progress ♦Indicates the correct response 128 Appendix C (Cont'd) DIFF 73 DISC 28 5. Instructional programs which are loaded into the computer: A. Can be constructed by the teacher B. Can be purchased from a commercial company C. Are furnished by the manufacturer of the computer *D. All of the above E. A and B DIFF 76 _____ DISC 18 6. Which of the following computer languages is not used for Computer Assisted In­ struction as a curriculum author language: A. B. C. *D. DIFF 46 _____ DISC 37 7. The largest operational Computer Assisted Instruction system can serve at the same time: A. B. *C. D. DIFF 31 DISC 64 8. Basic Focal Mentor RPG A maximum A maximum More than Less than of 150 student terminals of 100 student terminals 150 student terminals 100 student terminals At this point in time, the single most common type of Computer Assisted Instruc­ tion program, in terms of numbers of students served, is: A. Problem solving * B . Drill and Practice C. Tutorial D. Simulation DIFF 74 _____ DISC 37 9. The lack of curriculum exchange between users of Computer Assisted instruction can be attributed for the most part t o : A. A fear of standardizing the curriculum across the country *B. Absence of the same kind of computer and/or author language C. The tremendous cost w h i c h prohibits such an exchange D. No generalization can be made at this time. 129 Appendix C (Cont'd) DIFF 65 ______ 10. DISC -9 In a Computer Assisted Instruction project total software is considered to be: A. Curriculum or instructional programs B. System programs to operate the curriculum C. Teacher, administrators, and technical personnel involved irk the project *D. All of the above E. A and B DIFF 43 DISC 55 11 . In a Computer Assisted Instruction pro­ ject involving man y students at the same time, the computer: A. Has to be at the location of the student user B. Can be miles away or at the location of the student user C. Has to have time-share capabilities *D. B and C DIFF 53 DISC 73 12. The fastest response to a student's question or answer by the computer occurs when the curriculum is stored on: *A. B. C. D. DIFF 35 _____ DISC 55 13. M ost Computer Assisted Instruction systems use a computer of the: A. B. *C. D. DIFF 57 _____ DISC 18 14. Magnetic disc Magnetic tape Magnetic card Punched card Analog type Hybrid type Digital type Special purpose type Computer Assisted instruction involving man y students at the same time was not possible until the advent o f : *A. B. C. D. Time-share computers Third generation computers Analog computers Frequency response analysis 130 Appendix C (Cont'd) DIFF DISC 7 _____ 15. At this point in time, Computer Assisted 9 Instruction is mos t widely used in the subject area o f : A. B. *C. D. DIFF 91 _____ DISC 9 16. W h i c h of the following computer languages would be best for mathematical computation by students in a Computer Assisted Instruc­ tion program: A. B. *C. D. DIFF 55 DISC 55 17. 18. Autocoder Simple APL RPG A carefully planned Computer Assisted Instruction program should reduce the number o f : A. *B. C. D. DIFF 84 DISC 27 English History Mathematics Science Teachers Failures The work hours of teachers B and C A t this point in time, the single m o s t expensive aspect of computer Assisted Instruction i s : A. The hardware *B. The software C. The maintenance of the system D. Technical personnel DIFF 33 _____ DISC 46 19. Computer Assisted Instruction curriculum authors generally produce: A. Instructional assemblers * B . Instructional software C. Instructional compilers D. Instructional program controllers 131 Appendix DIFF 59 DISC 55 (Cont'd) 20. In a drill and practice Computer Assisted Instruction program, a second grade student wo u l d average using the terminal between: A. B. *C. D. 1 to 1 to 1 to More 5 minutes 10 minutes 15 minutes than 15 minutes Appendix D Knowledge Instrument Used in the Survey Appendix D Knowledge instrument Used in the Survey In this test select the best possible answer for each question. Place the letter of the correct answer (found under each question) in the blank to the left of the number of that question. There is only one best answer for each q u e s t i o n . PLEASE ANSWER ALL OF THE QUESTIONS. If you are in doubt as to the correct answer, put down the response that you think is best. 132 133 Appendix D (Cont'd) 1. In terms of instruction the mos t efficient method of storing Computer Assisted Instruction curriculum, as students interact with the com­ puter, is by means of: *A. B. C. D. 2. Magnetic disc Magnetic tape Punched card Punched tape Computer Assisted Instruction is considered by many experts as a part of: A. Computer Assisted Tresting *B. Computer Managed Instruction C. Computer Assisted Guidance D. Computer Managed Administration 3. Most Computer Assisted Instruction systems use: A. B. *C. D. 4. Student Student Student Student terminals terminals terminals terminals real time up time on line off line Computer Assisted Instruction can be used to individuali ze ins t r u e t i o n : A. If the programs are of a branching type B. If audio and graphic display devices are used *C. If the curriculum writer plans for individualization D. If there are enough terminals for each student 5. At the present time, which of the following can't a Computer Assisted Instruction system d o : A. Analyze typed responses *B. Analyze spoken responses C. Analyze tactile responses D. Analyze optical responses 6. Cathode Ray Tubes are u s e d : A. In the operation of the Central Processing Unit *B. For the display of information C. In the operation of the Data Reduction Unit D. For the storage of information ♦indicates correct response 134 Appendix D (Cont'd) 7. Effective Computer Assisted Instruction programs are very difficult to construct: A. Because they are very expensive B. Because very little is known about how students learn *C. Because they require detailed planning by the author D. Because of the tremendous problem of coding them into machine language 8. Which of the following is most often true of students taking lessons from a computer: A. They feel isolated from their fellow students *B. They get bored with poor instruction C. It takes a long time for them to adjust to the new instructional situation D. They have a difficult time learning to use the terminal 9. When students are working at the terminals their teacher: A. Is usually present supervising and assisting the students with the material B. Can be present but usually is not C. Is present but acts only as a supervisor * D . No generalization can be made 10. Which of the following types of Computer Assisted Instruction programs is designed to allow the learner to manipulate the machine rather than the machine manipulating the learner: A. B. *C. D. 11. Dialogue Tutorial Problem solving Drill and practice If a computer for a Computer Assisted Instruction project is said to have 35,000 positions of core storage, this would be referring to the: A. Buffer storage unit *B. Central processing unit C. Data reduction unit D. External storage unit 135 Appendix D 12 . (Cont'd) Computer Assisted Instruction is an effective means of instruction because: A. The student will receive immediate feedback to his response B. A student's assignment can be predetermined by his past performance on given concepts C. Instruction can be individualized *D. All of the above 13. At this point in time the greatest disadvantage of the computer in terms of instruction is: A. The computer can only communicate with the student by typed words or projected images B. The present computers were designed for industry and not education *C. The limited curriculum materials available for use in a Computer Assisted Instruction program D. The absence of direct teacher observation makes it difficult to determine pupil progress 14. The largest operational Computer Assisted In­ struction system can serve at the same time: A. B. *C. D. 15. A max i m u m A max i m u m More than Less than of 150 student terminals of 100 student terminals 150 student terminals 100 student terminals At this point in time, the single mos t common type of Computer Assisted Instruction program, in terms of numbers of students served is: A. Problem solving * B . Drill and Practice C . Tutorial D. Simulation 136 Appendix D (Cont'd) 16. The lack of curriculum exchange between users of Computer Assisted Instruction can be attributed for the m o s t part to: A. A fear of standardizing the curriculum across the country *B. Absence of the same kind of computer and/or author language C. The tremendous cost which prohibits such an exchange D. A lack of communications between Computer Assisted Instruction projects 17. In a Computer Assisted Instruction project in­ volving m a n y students at the same time, the computer: A. Has to be at the location of the student user B. Can be miles away or at the location of the student user C. Has to have time-share capabilities * D . B and C 18. The fastest response to a student's question or answer by the computer occurs when the curriculum is stored on: *A. B. C. D. 19. Mos t Computer Assisted Instruction systems use a computer of the: A. B. *C. D. 20. Magnetic disc Magnetic tape Magnetic card Punched card Analog type Hybrid type Digital type Special purpose type Computer Assisted Instruction involving m a n y students at the same time was not possible until the advent of: *A. Time-share computers B. The differential analysis of latency response time C. Analog computers D. Frequency response analysis 137 Appendix D (Cont'd) 21. A carefully planned Computer Assisted Instruc­ tion program should reduce the number o f : A. *B. C. D. 22. Computer Assisted instruction curriculum authors generally produce: A. *B. C. D. 23. Teachers Failures The work hours of teachers B and C Instructional Instructional Instructional Instructional assemblers software compilers program controllers In a drill and practice Computer Assisted In­ struction program, a second grade student would average using the terminal between: A. B. *C. D. 1 to 1 to 1 to More 5 minutes 10 minutes 15 minutes than 15 minutes Appendix E Background Questionnaire Appendix E Background Questionnaire Please respond to all of the following questions by either placing a check in the appropriate box, or by filling in the blank. 1. What is the name of the School District you are employed by? These boxes are for Data Processing and are NOT to be marked. [ ] I ] 2 3 [ ] 4 t ] 10 2. What position do you hold with your school district? [ ] Principal [ ] Teacher [ ] 11 3. What is your teaching or adminis­ trative level? [ ] Elementary [ ] Secondary (K-6) [ 1 12 (7-12) 4. What is your current age? [ ] 13 [ ] 14 5. If you are a secondary teacher, what subject matter area are you responsible for? 6. Do you see your subject matter area as related more to science and mathematics or English and humanities? [ ] 15 [ ] Science and Mathematics [ ] English and Humanities [ ] Neither 21 [ ] 22 [ ) 24 [ ] 25 [ ] 138 [ ] 26 Appendix F Map of Michigan by Counties 139 BARAGA .«“* * ! I I _j wee \uAROUETTE U - JI RON 14 \juckiied % i T" |ALGER | JfiMMflflM j [S C H O O LC R A FT* _____ j \~dtm--1 p s s ^ i L >=C) • f£MMer\cnf» preso ueisle ^ h ar lev 6H T\ I antrWT ! ! ^ i . I I f^ ^ ^ U f/lK A L K A S ^ C R A W F O R o tO S C O D A H i j a n iste e x ^ifFO R Ft*i^lM }^nsc6i^ I liMI NON .... f~ALCONa I ! • I Tosco' " I J._. .1__L ._!_ H a s o n T l a k e " \ o s c e o la '\ ~ c l a r e \ gladw in T a r e n a c S i____ K , j ____ ij______i!_____j ibatj