A STUDY OF THE RELATIONSHIP BETWEEN SELECTED BACKGROUND FACTORS OF STUDENT TUCHERS AND PUPIL OPINION OF CERTAIN TEACHING TRAITS Thesis for lee Degree ol ECl. D. MICHIGAN STATE UNIVERSITY William Randall Sleeper 1962 This is to certify that the thesis entitled A STUDY OF THE RELATIONSHIP BETWEEN SELECTED BACKGROUND FACTORS OF STUDENT TEACHERS AND PUPIL OPINION OF CERTAIN TEACHING TRAITS presented by William Sleeper has been accepted towards fulfillment of the requirements for Ed.D. . degree in Education a!» mu; Major professor W (C Jail-4 Date 2-5.62 0-169 LIBRAR Y1: Michigan State $4.? Umversxty «nu-coo w ABSTRACT A STUDY OF THE RELATIONSHIP BETWEEN SELECTED BACKGROUND FACTORS OF STUDENT TEACHERS AND PUPIL OPINION OF CERTAIN TEACHING TRAITS by William Randall Sleeper Statement of the Problem This study was conceived as a means of investigating the relationship between selected eXperiential background factors of secondary education students at Central Michigan University and pupil Opinion of certain teaching traits these students exhibit in student teaching. The background factors considered were pre-college in time of occurrence and social in nature. The selection of teaching traits investigated was based on studies of the reactions of high school boys and girls to certain teacher behavior. Pre-college education students tionnaire. Later, teaching, Opinions Procedure background factors of senior secondary were inventoried by means of a ques- as these students did their student were obtained from their pupils con- cerning certain exhibited teaching traits. WILLIAM RANDALL SLEEPER In this study an analysis was made of the degree of relationship between 19 background factors of the student teachers and 10 teaching traits of the student teachers as rated by their pupils. Findings Two hypotheses were posited to aid in the process of constructing answers to questions which pertained to the relationship between selected eXperiential background factors of student teachers and pupil Opinion of certain teaching traits. l. The first hypothesis was stated as follows: No relationship exists between pupil Opinion of certain teaching traits and selected background factors of student teachers. As a result of the evidence presented in this study there is little reason to invalidate or cast serious doubt upon the general null hypo- thesis. Though nine items were discovered where some significance was found, the relationships were entirely too small to permit usefulness of forecasting efficiency. No relationship exists between pupil Opinion of the student teacher's all-around teaching ability and certain combinations of home community size of both pupils and student teachers. WILLIAM RANDALL SLEEPER From the evidence reported in this study there is little reason to invalidate or cast serious doubt upon the general null hypothesis. If those persons involved with teacher education and teacher evaluation continue to believe in the importance of eXperiential background factors, then they must search out ways of measuring the effect of these factors upon teacher effectiveness. For this study has revealed a great deal of evidence to support the conclusion that pupil Opinion ratings, on an instrument such as the one utilized, do not discrim- inate sufficiently between the background factors studied to warrant attaching strong positive or negative values to any single factor. Recommendations Superintendents, directors of student teaching, and supervising teachers should use extreme caution in utilizing experiential background factors as instru- ments of evaluating the potential of future teachers. Teacher education institutions should continue the search for adequate instruments of prediction to use in screening future teachers, as the evidence dis- closed in this study does not support the use of eXperiential background factors for that purpose. Teacher education institutions which are basing con- siderable portions of their pre-student teaching eXperience program upon the experiential background WILLIAM RANDALL SLEEPER of the students, will have to continue to seek evidence to support their contention that these background factors make any significant difference in teacher effectiveness, at least, as rated by pupils. A STUDY OF THE RELATIONSHIP BETWEEN SELECTED BACKGROUND FACTORS OF STUDENT TEACHERS AND PUPIL OPINION OF CERTAIN TEACHING TRAITS by William Randall Sleeper A THESIS Submitted to the School for Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF EDUCATION Department of Teacher Education 1962 ACKNOWLEDGMENTS The completion of a study of this kind depends upon the assistance and cooperation of many persons. The inves- tigator wishes to thank the coordinators and supervisors of student teaching for their assistance in distributing and administering the questionnaire. He is also especially grateful for the valuable advice of Dr. Wilbur Harris on statistical procedures. The writer is deeply appreciative of the guidance and sincere encouragement he received from Drs. William V. Hicks, William Durr, Calhoun Collier, and Wilbur Brookover. Finally, this study would not have been possible without the ever-present help of the writer's partner Mrs. William Sleeper. W.R.S. ii Chapter 1. 1'1 . III. IV. TABLE OF CONTENTS INTRODUCTICN O O O O O O O O '0 O O O The Problem . . . . Null Hypotheses . . . . . . . . . . Definition of Terms . . . . . . . . Major Assumptions . . . . . . . . . Procedure . . . . . . . . . . Delimitation of the Study . . . . . Limitations of the Study . . . . . . Need for the Study . . . . . . . . . Organization of the Study . . . . . REVIEW CF THE LITERATURE . . . . . . Pupil Ratings of Teachers . . . . . Prediction of Teacher Effectiveness summarYOOOOOOOOOOOOOO METHOD OF INVESTIGATION . . . . . . Sources of Data . . . Instruments . . . . . . . . . Treatment of the Data . . . . . . . Summary . . . . . . . . . . . . . . PRESENTATION AND ANALYSIS OF DATA . Summary . . . . . . . . . . . . . . SUNMARY, CONCLUSIONS, AND RECOMMENDATIONS . Summary of Findings . . . . . . . . Conclusions . .1. . . . . . . . . . Recommendations .‘. . . . . . . . . BIBLICGRAPHY O O O O O O O O O O O 0 APPENDIX 0 O O O O O O O O O O O O 0 Appendix A: Questionnaire Form Used to C01- lect Student Teacher Background Appendix B: Introductory Requests, Direc- tions for Administration, and Student Cpinion Questionnaire . iii 86 89 LIST OF TABLES Table ' Page 1. Number of Participating Student Teachers Assigned in Each of the Nine Student Teaching Centers . . . . . . . . . . . . . . . 36 2. Number of Pupil Responses Giving Pupil Cpinion of Teaching Traits . . . . . . . . . . 38 3. Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background Factors and the Student Teachers' Know- ledge of the Subject Taught as Rated by pupils 0 O O O O O O O O O O O O O O O O O 47 4. Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background‘ Factors and the Student Teachers' Ability to Explain Things Clearly as Rated by pupiISooooooooooooooooeeoo5O 5. Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background Factors and the Student Teachers' Fair- ness in Dealing With Pupils as Rated by pupils O O O O O O O O O O O O O O O O O O 52 6. Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background Factors and the Student Teachers' Ability to Maintain Good Discipline as Rated by Pupils O O O O O O O O O O O C O O O O O O 54 7. Pearsonian Coefficients of Correlation(r) Between 15 Student Teacher Background Factors and the Student Teachers' Sympathetic Understanding as Rated by pupils O O O O O O O O O O O O O O O O O O O O 56 8. Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background Factors and the Amount Pupils are Learning in Class as Rated by Pupils . . . . . . . . . 58 iv Table 10. ll. l2. l3. l4. Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background Factors and the Ability of the Student Teacher to Make the Class Lively and Interesting as Rated by Pupils . . Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background Factors and the Ability of the Student Teacher to Get Things Done in a Business- Like Manner as Rated by Pupils . . . . . Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background Factors and the Value of the Subject to the Pupil as Rated by Pupils o o o o o Pearsonian Coefficients of Correlation (r) Between 15 Student Teacher Background Factors and All-Around Teaching Ability as Rated by Pupils . . . . . . . . . . Chi Square Tests of Relationship Between Certain Background Factors of Student Teachers and the All-Around Teaching Ability of the Student Teachers as Evaluated by Pupils . . Distribution of Pupil Ratings of Student Teachers from Communities of a Size Similar to that of Pupils Page 60 62 64 66 68 69 CHAPTER I INTRODUCTION This study was designed to add to the information in existence concerning a question that many educators adready believe to be answered. At least, some appear to function in their positions as if they so believed. The question is one dealing with the importance of the pre-college eXperiential background of prospective teachers. Through the years super- intendents have asked applicants for teaching positions to describe their home background, participation in high school activities, community participation, and other information of an eXperiential nature. Other things being equal these factors were apparently a deciding influence in the employ- ment of applicants. Superintendents are, of course, not the only persons who Operate under the assumption that pre-college social ex- periences are important determiners of an individual's readi- ness for teaching. Directors and supervisors of student teaching make distinct evaluations of these experiences in planning for the student teaching program. In fact, it is reported by some institutions that when a student's academic record is poor, his eXperiential background may be the de- ciding factor in permitting him to enroll in student teaching.1 lFligor, R. 3., Anal sis 2f the Evaluation, Use, and Value 2: Certain Competancies for Beginning the Student Teaching Experience, pp. 55-58. Though the evidence in existence, for the most part, fails to show any strong relationships between eXperiential factors and teacher effectiveness, the measurements have been made in ways differing from those employed in this study. The evidence collected in the present study should make a particularly meaningful contribution as it supports or rejects the evidence already in existence on this subject. If administrators are going to continue to evaluate the ex- periential background of applicants, then it is important that research continue in an effort to measure the influence of these factors on teacher effectiveness. The Problem This study was conceived as a means of investigating the relationship between selected eXperiential background factors of secondary education students at Central Michigan University and pupil opinion of certain teaching traits these students exhibit in student teaching. The background factors considered were pre-college in time of occurrence and social in nature. The selection of teaching traits studied was based on studies made of the reactions of some 30,000 boys and girls to certain teacher behavior.2 2Bryan, R. C., "Student Reactions and Merit Salary Schedules," Faculty Contributions, 4:12, July, 1958. Null Hypotheses 1. No relationship exists between pupil Opinion ratings of certain teaching traits and selected eXperiential background factors of student teachers. The variables to be tested in this hypothesis are listed below.3 Teaching traits Knowledge of subject taught Ability to explain clearly Fairness with students Maintains good discipline Sympathetic understanding Amount learned Makes class interesting Business-like manner Value of subject to pupils All-around teaching ability Experiential background factors Age Number of residence changes Size of town in which student was reared Number of younger brothers Number of older brothers Number of younger sisters 3Complete statements, as used in the questionnaire, may be found in the Appendix B. Number of older sisters Total siblings Socioeconomic status Church attendance Size of high school attended Number of varsity awards Number of intermural activities Number of organizations in which student participated Leadership score Member of Future Teachers Club Member of Student Council 2. No relationship exists between pupil Opinion ratings of all-around teaching ability and certain combina- tions of home community size of both pupils and student teachers. Definition of TermsA The term COOperating school is used to designate a school which is not controlled or supported by the college but which does provide facilities for student teaching in the teacher education program. The term coordinator is used to mean a person who serves as the college representative and who is responsible 4Where feasible definitions adOpted and approved by the Association for Student Teaching have been used. (J'‘ for supervising a group of student teachers, usually in a resident center. The term director 2; student teaching is used in this study to mean the person designated by the college with ad- ministrative responsibility for organizing and coordinating the college's program of professional eXperiences. The term eXperiential background is used to mean those pre-college experiences of a social nature which the secondary education students identify for this study on the Student 5 Teacher Questionnaire. The term laboratory school is used to mean a school which is controlled and supported by the college and which is organized as an integral part of the teacher education program to provide significant Opportunities to study and relate the various phases of the teacher's activities both in and out of school. The term EEEll is used to mean any boy or girl en- rolled in the junior or senior high schools, grades 7-12, which participated in this study. The terms pupil rating and pupil Opinion are used interchangably to refer to the Opinion of pupils concerning certain qualities of their teachers. This Opinion was 5A cOpy of the Student Teacher Questionnaire may be found in Appendix A. collected through the use of a questionnaire of the rating scale variety.6 The term socioeconomic status is used to mean the relative position assigned to the student teacher on the North-Hatt7 ranking of occupations. This position was de- termined by the occupation of the person contributing most to the support of the student teacher's family during his pre- college life. The term student is used to mean a person enrolled in the secondary education program at Central Michigan University. The term student teacher is used to mean the college student who is doing student teaching. The term student teaching is used to mean the period of guided teaching during which the student takes increasing responsibility for the work with a given group Of learners over a period of consecutive weeks. The term student teaching center is used to mean a community in which the student teacher lives and partici- pates in the community life and activities as a part of his assignment in student teaching. The term supervising teacher is used to mean one who teaches children or youth and who also supervises student teaching. 6A COpy of the Pupil Cpinion Questionnaire used in this study may be found in Appendix B. 7North, C. C., and Hatt, P. K., "Jobs and Occupations: A POpular Evaluation, "Sociological Analysis, pp. 464-474. The term teacher education institution is used to mean any school of higher learning where individuals may study for and be graduated with teaching certificates. The term teaching traits is used to mean a set of environmental conditions which form a part of the composite of the classroom learning situation. Specifically the term refers to those items rated by pupils on the Pupil Cpinion Questionnaire. Major Assumptions This study was based on the following assumptions: 1. That an investigation of the relationship between selected background factors and pupil Opinion of the teach- ing traits of student teachers constitutes a worthwhile study. This assumption appears reasonable in view of the facts that (a) even though some information exists on this tOpic, super- intendents of schools and supervisors of student teaching continue to give emphasis to the experiential background of their applicants and student teachers in evaluation, (b) additional information on this topic can serve to put these evaluations in their prOper perSpective, (c) if significant relationships were found, teacher education institutions could use the results to improve their measures of prediction of teacher effectiveness, and (d) these same institutions could use the results to analyze more carefully their pre—student teaching eXperience program. 2. That pupil Opinion is an important criterion when judging student teacher effectiveness. This assumption appears reasonable for, as Bush points out, The findings of this study suggest that the per- sonal liking of a pupil for his teacher is one of the most powerful factors in bringing about an effective learning relationship between the teacher and the pupil. The study shows clearly that those teachers who are most liked personally by their pupils tend to be the most competent. Pupil liking for the teacher is highly re- lated to pupil liking for the subject and the subject- matter achievement. There is a marked tendency for those pupils who mosé like the teacher to feel that they are learning more. 3. That pupils and student teachers respected the motives of the investigator and responded honestly to his questions. 4. That the techniques employed for collecting data were adequate for their intended purpose. 5. That the sample studied is typical of a larger universe and that the findings of this investigation will have application beyond that of the studied subjects. Procedure The following steps were taken in carrying out this study. 8Bush, R. N., The Teacher-Pupil Relationship, p. 189. l. The literature which is significantly related to this study was reviewed. 9 was constructed to collect 2. A questionnaire form specific information regarding certain background factors of student teachers: factors which, from such evidence as Fligor'slO were believed to be important considerations by directors of student teaching and supervising teachers. 3. A pupil Opinion questionnaire11 was adopted from among those carefully prepared by previous investigators. 4. A pilot study of four student teachers and their pupils was completed for purposes of improving the instru- ments and standardizing the procedure of administration. 5. Completed student teacher questionnaires were administered to secondary education students in their psy- chology and education classes. Pupil Opinion questionnaires were distributed to schools by coordinators, administered by supervisors, and then returned to the investigator by the coordinators. 6. Data gathered from the questionnaires were tab- ulated, analyzed, and interpreted. 7. Conclusions were drawn: and recommendations were made. 9See Appendix A. lOFlIgOI‘, 220 Cite pp. 55.58. 11See Appendix B. 10 Delimitation of the Study This study was delimited in the following ways: 1. It was limited to senior students in secondary education who were following Plan A12 at Central Michigan University during the 1959-1960 school year. 2. Pupil Opinion of each student teacher was limited to the responses from one regularly instructed class. 3. It was limited in a geographical sense, in that the individuals participating in the pupil opinion phase of the study were restricted to those public high schools where Central Michigan University has COOperative working relation- ships resulting in the establishment of a student teaching center.13 Limitations of the Study This study was subjected to certain limitations be- cause of the nature of the problem and restrictions on the investigator. These include: 1. An element of fear is aroused in many teachers and administrators when any type of rating of their effec- tiveness is attempted. This fear, particularly of pupil rating, may have been reflected in the presentation of the Pupil Opinion Questionnaire by the supervising teachers. 12Central Michigan University, Bulletin, 1959-1960, p. 97. 13A list of student teaching centers may be found in Appendix B. ll 2. Whenever a single criterion, such as pupil Opinion, is used to judge teacher effectiveness, there are certain limitations imposed. The comprehensiveness of teaching and the relative immaturity of the secondary school pupil are two such factors. 3. The fact that only one investigator with limited financial resources was involved in the direction of this study may have resulted in too little control of the ques- tionnaire administration. Need for the Study Will a varied eXperiential background make a teacher more effective in the classroom? The need for this study is based on the fact that peOple in strategic positions of authority are answering this question in the affirmative without substantial objective evidence to support this posi- tion. Corey referred to the manner in which every up-to-date superintendent of schools will appraise the many background factors, from temperance to outside activities, in order that he may be better able to judge the candidate's potential teaching abi1ity.14 Directors of student teaching and supervising teachers are currently among the strongest supporters of the impor- tance of pre-college eXperiences. The 1959 Yearbook of the 14Corey, S. M., "The Present State of Ignorance About Factors Affecting Teaching Success," Educational Administra- tion and Supervision, 18:481-490, October,‘l932. 12 Association for Student Teaching eXplains the importance of securing a great deal of information about the student teacher's background. It points out that the supervising teacher will want to know the student teacher's achievements, his attitudes and his enthusiasms. "He needs to know about the student's family and where he has lived; the circum- stances in his life and in his environment: the schools he has attended. . ."15 This is only one illustration of the many ways in which those in positions of advantage are stressing the importance of eXperiential background factors. Most books written for supervising teachers contain sample forms to be used in collecting such information; a prime ex- ample is Guiding Your Student Teacher by Curtis and Andrews.16 Fligor found evidence of this same concept and makes the following statement concerning it: "It seems that both directors and supervisors of student teaching evaluated their student teachers with reference to their background of ex- periences prior to entrance in college. In many instances this was a very subjective evaluation."17 Since importance is being given to eXperiential back- grounds, it is this investigator's intention to determine whether such backgrounds are significantly related to certain 15Association for Student Teaching, The Supervising Teacher, p. 44. l6Curtis, D. K. and Andrews, L. C., Guiding Your Stu- dent Teacher, pp. 345-346. l7F1igor, 22. cit., pp. 57-58. 13 teacher traits as perceived by pupils. If these factors are important it is necessary that efforts continue to be made to relate them to some measure of teacher effectiveness. Up to the present time investigations appear to reveal little or no relationship. Predictive possibilities. Teacher education institu- tions are continually searching for methods of predicting success of candidates for teaching certificates. If rela- tionships between certain background factors and teaching effectiveness were found to exist this would be a real asset in planning individualized programs for students to strength- en areas where shortcomings are discovered. It is also possible that failure in student teaching might be avoided by a better method of screening. Actually there are insti- tutions currently using their knowledge of the students ex- periential background to build pre-student teaching experi- 'ence programs to better prepare students for the teaching eXperience. Fligor18 discovered this in his survey and interestingly enough, Sinclair19 found no testably signif- icant difference between groups of student teachers prepared with and without the "eXperience-type" program. Central Michigan University. The need for this study has existed at Central Michigan University for some time. 18Ibid., pp. 57-58. 19Sinclair, W. W., Anal sis 2: Three Pre-Student leaching Expgriences Ig_the Preparation 2: Elementary School Teachers, 157 pp. 14 The pressures for up-dating programs have been particularly strong during recent years. The changes which have been contemplated, and in some cases made, have repeatedly in- volved eXperience-type programs. The theory behind these changes is much the same as that discussed above: that is, students with incomplete eXperiential backgrounds should be given Opportunities early in their college careers to fill the void. The intention of a program like this is that these experiences of a social nature are very important to effective teaching. This investigator hOpes to add signi- ficantly to the information available on the subject, so that wise judgments may be made in the future when change is considered. Organization of the Study The remainder of the study is presented in four chap- ters. Chapter II reviews the literature which is signifi- cantly related to this study. Chapter III outlines the method of investigation. Chapter IV contains an analysis of the relationship which exists between certain background factors and pupil Opinion of student teachers, and Chapter V includes a summary and discussion of the findings, con? clusions, and recommendations. 15 CHAPTER II REVIEW OF THE LITERATURE The related literature reviewed for this study was found to be concentrated largely in two major areas of writing-~that of pupil rating of teachers and its corre- lates, and that of prediction of teacher effectiveness. Though identifiable these two areas are not entirely separ- able, as there is considerable overlapping. However, the investigator has endeavored to organize the reviews in this way. Information gained from the pupil rating of teachers studies will be discussed first as they are more closely related to the problem investigated. Pupil Ratings of Teachers Many investigators have studied pupil rating as a measure of teacher effectiveness and instructor improvement. As might be eXpected, one of the first questions which fated these investigators concerned the reliability of pupil ratings. As early as 1922 and 1927, Knight1 and Guthrie2 concluded that there was considerable agreement among students concerning the abilities of their teachers. lKnight, F. B., "Qualities Related to Success in Teaching," Teachers College Contributions 12 Education, no. 120, Columbia University, 1922, 89 pp. ZSuthrie, E. R., "Measuring Student Cpinion of Teachers," School and Society, 25:175-176, February, 1927. 16 Using the chance half technique they received reliability coefficients as high as .91 and no lower than .56. Since the date of these studies their findings have been confirmed 4 and Amatora.5 There are several in- by Remmers,3 Bryan, vestigators who have reported findings which have a bearing on the reliability of student rating but in which correlation coefficients are not reported. Fritz6 discovered that 89 pupils had considerable difficulty in duplicating their judg- ments on two ratings of one teacher obtained on a seven-part scale when the rating periods were scheduled a week apart. While Porter7 found, in working with student teachers, that there was great variance in the leniency expressed by dif- ferent classes. Since he presented no statistical data in his report it is difficult to compare his study with others. Neither does he probe the possibility that some other factor, such as teacher merit, could be the reason for the variance in class reSponses. In summary, he points to the consistency 3Remmers, H. H., "To What Extent Do Grades Influence Student Ratings of High School and College Students' Judg- ments of Their Teachers," Journal g: Applied Psychology, 18:619-630, October, 1934. 4Bryan, R. C., "Reliability, Validity, and Needfulness of Written Student Reactions to Teachers," Educational Admin- istration and Supervision, 27:655-665, December, 1941. 5Amatora, S. M., "A Diagnostic Teacher-Rating Scale," Journal 2: Psychology, 30:395-399, October, 1950. 6Fritz, M. F., "The Variability of Judgment in the Rating of Teachers by Students," Educational Administration and Supervision, 12:630-634, December, 1926. 7Porter, W. A., "Pupil Evaluation of Practice Teaching," Journal of Educational Research, 35:700-704, May, 1942, 17 of pupil agreement concerning the best and poorest teachers, but finds judgments of the middle group quite varied, a common problem when using rating scales. Bryan makes the following statement with respect to the reliability of student reactions: From a statistical viewpoint, the responses of 30 pupils to a question dealing with sympathy, i.e., 'What is your Opinion concerning the sympathy shown the students by this teacher: excellent, good, average, be- low average, orlaoor?' will produce a reliability coef- ficient of .90. Student responses to the Bryan ques- tionnaire will produce reliability coefficients (chance half method) as high as or higher than those produced by the better standardized tests. Thus it may be said that there is much agreement among the Opinions students express concerning their teachers. All the many pub- lished studies concerning the reliability of student re- actions agree that they are adequately reliable for all practical purposes. l3Bryan, R. C., Pupil Ratinggf Secondary School Teachers, Contributions to Education, No. 708, Bureau of publicatigns, Teachers College, Columbia University, 1937, 96 pp. A second question around which a cluster of studies may be found is that of whether or not there is any correlation between pupil ratings and other measures of teacher effec- tiveness. Teacher ratings by administrators and peers have, in general, received very low coefficients of correlation when tested with pupil ratings of teachers. Varying criteria are used for purpose of comparison, but it is interesting to note that relative few studies have used pupil gains. 8Bryan, R. C., "Student Reactions and Merit Salary Schedules", Faculty Contributions, 4:21, July, 1958. 18 Self ratings by teachers were compared to pupil 9 and the ratings, on comparable scales, by Davenport, obtained coefficients of correlation were found to be very low, often approaching zero. He implies in his analysis that the teacher's philoSOphy and her actual practice may not be the same, and this through no fault of her own. Class size, for example, is not teacher controlled, but does have important effects on teacher procedure. As was mentioned previously, pupil gain has been infrequently used as the criterion of teacher effectiveness. The two studies to be cited here revealed only very small relationships. In a study made by Lins,lO he concludes that the small relationship may have been due to either the small sample or to the manner in which the students who were to rate each teacher were chosen. Differences in the following traits were found to be significant at the .01 level by Remmers:11 university student ratings of instructors and care of communal apparatus. 9Davenport, K., "An Investigation into Pupil Rating of Certain Teaching Practices,” Purdue University Studies in Higher Education, no. 49,1944,64 pp. lOLins, L. 3., "The prediction of Teaching Effi- ciency," Journal 2: Experimental Education, 15:2-60, Sep- tember, 1946. llRemmers, H. H.: Martin, R. D.: and Elliott, D. N., "Are Students' Ratings of Instructors Related to Their Grades?“ In H. H. Remmers(Ed. ). Student Achievement and Instructor Evaluation in Chemestry, Purdue University Studies in Higher Education, 66: 17- 26, July, 1949. 19 Significant at the .02 level were: university student ratings and supervision during tests, knowledge of chemistry, and returning test. There has been some investigation of a possible re- lationship between grades given and pupil rating of teachers. Certainly if a relationship were to be found it would have a significant bearing on the validity assigned to students' ratings of their teachers. Bryan12 found no significant correlation in his study of 22 senior high and 41 junior high teachers and their pupils. While Remmers,l3 however, found coefficients of substantial size, but in both positive and negative directions. He eXplains this contrast in terms of methodology. Morsh and Wilder sum up the influence of grades as follows: If one assumes that good students will approve of instructors who conduct their teaching at a high level (and over the heads of the poorer students), then, a positive correlation between student ratings and grades would result. Conversely, if the instructor pitches his teaching at the level of the weaker students, the brighter students will disapprove and a negative correlation will result. This hypothesis would account both for the range of coefficients obtained and for the fact that when correlations are not computed separately for each instruixor, coefficients of negligible magni- tude are found. 12Bryan, R. C., "Pupil Ratings of Secondary-School Teachers," Teachers College Contributions 33 Education, No. 708, 1937, 96 pp. 13Remmers, Martin, and Elliott, g2. cit., pp. 17-26. 14Morsh, J. E. and Wilder, E. W., Identifyigg the Ef- fective Instructor: A Review of anntitative Studies, 1900- I952, USAF Personel and Trainlng Research Center, 1954 p. 35. 20 A number of factors other than grades have been ex- amined in the light of their possible influence on pupil rating of teachers. Some of the factors which have been considered have been age and sex of teacher, length of ac- quaintance with pupils, pleasurable personal relationship between pupil and teacher, and whether or not the subject taught by the teacher was the pupils' favorite subject. Brookover has contributed two studies in this area. His first15 in 1940 was a study of 1139 pupils and 37 teachers, in which it was found that for 22 teachers the correlation between interaction and teaching effectiveness was .50 or higher, while only 15 were lower than .50. The correlation between the mean ratings of all pupils on the interaction scale with those of the teaching effectiveness scale was .639. There were no significant differences among sex, age, classification, age or sex of the teachers and the way pupils responded to either of the scales. Brookover's second study16 in 1945 was a study of 66 Indiana High School American History teachers in which selected social factors were correlated with both pupil rating and pupil gain. The conclusions which follow were 15Brookover, W. B., "Person-Person Interaction Be- tween Teachers and Pupils and Teacher Effectiveness," Journal gnyducational Research, 34:272-287, December, 1940. 16Brookover, W. B., "The Relation of Social Factors to Teaching Ability," Journal gf_EXperimental Education, 13:191- 205, June, 1945. 21 the result of a minor hypothesis in the study and bear direct relationship to the present study. 1. The nature of the pupil's personal relation- ship with their teachers affects their ratings of the teachers‘ abilities. The more friendly the personal relationship the higher the ratings of teaching ability . . although the students' relationship with their teachers was found to be negatively correlated with the extent of their learning, the students apparently feel that they learn more from the teachers with whom they have a close relationship than from those with whom they are less closely associated. 2. The pupils' ratings of teaching ability are positively related to the age of the teachers. The re- lation between length of acquaintance is also a positive correlation between pupils' ratings of ability and the length of time the teacher had taught in the school. Married teachers are more frequently rated high or low, while single teachers are more frequently given average ratings of ability. 3. The pupils ratings are not correlated with the frequency of the teachers' church attendance in the community. However, those teachers who do not parti- cipate in other than church activities are considered significantly better teachers by their pupils than those who do participate in such activities. 4. It seems that pupils are favorably impressed in their Opinions of a teacher's ability by long associ- ation with him. 5. Teachers who are better adjusted to their social situatign were considered better teachers by their pupils. In 1954 Drawhorne18 studied a group of eight student 17Ibid., p. 205. 18Drawhorne, C. L., "Relationship Between Pupil and Student-Teacher Interaction and Pupil Ratings of Teacher Effectiveness," Educational Administration and Supervision, 40:283-296, May, 1954. 22 teachers and 156 pupils in an effort to see if a relation- ship existed between pupil and student teacher interaction and pupil ratings of teacher efficiency. He used two rating scales, his own Person-Person Interaction Scale and Bryan's19 Teaching Effectiveness Scale. He found the following: 1. Reasonably high correlations between pupil ratings of interaction and those of teacher effective- ness, which indicates that the relation between the pupil and his student-teacher is predictive of how the pupil will rate his teacher on teaching effectiveness. 2. Pupils gave more positive than they did neg- ative ratings. 3. Pupils in the Laboratory School rated their student-teachers higher on interaction and teacher ef- fectiveness than those of the Northwest High School. Critical ratios of 4.08 and 3.64, respectively, were significant beyond the one per cent level. 4. The criteria used in this study reveal no reliable difference between boy and girl ratings of interaction and teacher effectiveness. 5. Even though high-achievers seem to rate higher interaction between themselves and their student-teachers than low-achievers the critical ratio of 1.50 is not significant. 6. Regardless of the fluctuation in pupil response from one student-teacher to another, the relationship of pupil ratings on interaction to those of teacher effect- iveness remains about the same. Correlations between the two variables for the two student-teachers rated by the same four pupils were .45 and .46, respectively. 7. The Laboratory School pUpils who rated them- selves less interested in the course, rated their student teachers as high on interaction and teacher effectiveness as those who expressed more interest in the course. 19Bryan, R. C., "Eighty-Six Teachers Try Evaluating Student Reaction to Themselves," Educational Administration and Supervision, 27:513-526, October, 1941. 23 8. As rated by themselves, more pupils in the Laboratory School rated themselves more interested in the course than pupils in the Northwest High School. 9. The Northwest High School pupils who were more interested in the course rated their student- teachers higher in interaction and teacher effectiveness than those who were less interested. Critical ratios of 4.04 and 3.25, reSpeEBively, are significant beyond the one per cent level. Pupil ratings during recent years have become an im- portant part of investigations directed toward develOpment of some type of workable evaluation as a basis for merit pay schedules. Symonds' study of the characteristics of the effective teacher based on pupil evaluations is a typical example of this group of studies. His study is two-fold. Part one describes a method of locating more effective teachers by having pupils rank their teachers on seven bases. He found that pupil rankings agree with each other, with co- efficients of correlation in the .70's, .80's, or low. 90's. This appears to indicate considerable halo effect in the rankings on the seven questions. Pupil rankings correlated with principal ratings of teacher discipline in the .60's, in the .70's for teacher-relationship with pupils, and in the .40's for teacher ability to secure pupil achievement. In the second part of the study based on the observation of teachers who were ranked high and those ranked low by pupils, 2ODrawhorne, _p. cit., p” 296. 24 the following three factors seemed to differentiate the teachers in the extreme groups:21 a. Superior teachers liked children; inferior teachers disliked children. b. Superior teachers were personally secure and self-assured; inferior teachers were personally insecure and had feelings of inferiority and inadequacy. c. Superior teachers were well integrated and possessed good personality organization; while the in§2 ferior teachers tended to be personally disorganized. In addition to the above mentioned research where correlations between pupil rating and the ratings of others were sought there have been a number of studies where ex- aminations have been made of the possible relationships be- tween pupil rating and various test scores, personality test scores in particular. The findings vary, but in general, 23 are typical. In his the results obtained by Rabinowitz study a large group (over 1600) of student teachers were given a number of personality and attitude tests during their senior year in college. Cbservations were conducted approximately one year later in the rooms of 49 of these subjects who were employed as elementary school teachers. 21Symonds, P. M., "Characteristics of the Effective Teacher Based on Pupil Evaluations," Journal gf Experimental Education, 23:289-310, June, 1955. 22Ibid. p. 310. 23Rabinowitz, William and Rosenbaum, Ira, "A Failure In the Prediction of Pupil-Teacher Rapport," Journal gf|Edu- cational Psychology, 49:93-98, April, 1958. 25 A measure of pupil teacher rapport based on pupil responses to questions about their teacher and class was also obtained An analysis showed none of the test measures correlated sig- nificantly with pupil-teacher rapport as measured. Only one of the 63 correlations between the test measures and class- room behavior measures proved significant. Manifest Teacher Hostility, a measure based on classroom observation of the teacher correlated significantly with rapport. COOper's study24 of quantitative Rorschach factors as indicators of teacher effectiveness also makes an interesting contribution at this point. The primary purpose of this study was to investigate the use of current methods of quantifying Rorschach data as a means for differentiating be- tween a group of teachers favorably rated by their pupils and a group of teachers less favorably rated by their pupils. The study further attempted to examine the relationship be- tween pupil ratings of their teachers and (a) the sex of the teacher rated: (b) the subject taught: (c) the marital status of the teacher: and (d) inservice and preservice status. A checklist develOped through a review of the litera- ture was administered to the pupils of 72 inservice teachers and 153 student teachers who had volunteered for the eXperi- ment. These teachers were then divided into two groups, those with less favorable ratings and those with favorable 24COOper, J. G. and Lewis, R. B., "Quantitative Rorschach Factors in the Evaluation of Teacher Effectiveness," Journal g: Educational Research, 44:703-707, May, 1951. 26 ratings. The Rorschach test was administered to each. Chi square was used as the method of determining significance.25 The conclusions relative to Rorschach as stated by COOper are: 1. Current methods of quantifying Rorschach data are not dependable as a means of differentiating between liked and less-liked teachers. 2. The presence of Maile and Harrower-Erickson psychoneurotic signs was associated with unfavorable pupil ratings. The extent of overlapping prevents in- dividual predictions. 3. No relationship was found between pupil ratings and personality maladjustment as measured by the Munroe Inspection Rorschach. 4. The percentage of the number and kind of deter- minants used bore no relationship to pupil ratings. 5. The median number of human movement responses was slightly higher for the well-liked teachers than the less-liked teachers. The difference was not statistically significant. 6. Introversiveness-extratensiveness was not re- lated to pupil ratings. 7. Emotionally impulsive persons were found equally among liked and less-liked teachers. 8. Emotionally constricted individuals were more often found among less-liked teachers than liked teachers. Conclusions 13 Respect tg Teacher Status 1. Preservice teachers were rated more favorably than inservice teachers. 2. Pupil ratings were affected by neither the teachers sex nor marital status. 25Ibid., pp. 703-707. 27 3. In some cases, the subject taught affects pupil ratings of teachers.2 Prediction of Teacher Effectiveness "The literature pertaining to investigations of the relationship between various hypothesized predictors and teaching effectiveness is extensive, but consists to a de- plorable degree of reports of researches which suffer parti- cularly from inadequate consideration of control and lack of replication, and which therefore yield questionable results."27 There have been many reviews of the literature con- cerned with predicting teacher effectiveness, but two of the 28 who have published most comprehensive are Barr and others their reviews every three years over the last twenty years, and Morsh and Wilder29 whose very thorough work in 1955 made a significant contribution. In general, these reviewers point first to a primary problem in studies of teacher effec- tiveness; that is determining the criterion by which effec- tiveness will be judged. Usually the approaches to this prob- lem evolve around the evaluation of either (1) teacher behavior, in process, (2) a product of teacher behavior, or 261bid., pp. 706-707. 27Ryans, D. G., “Prediction of Teacher Effectiveness," in The Encyclopedia gf_Educational Research, p. 1486. 28Barr, A. S., "The Measurement and Prediction of Teaching Efficiency: A Summary of Investigations," Review 9: Educational Research, 28:256-264, June 1958. 29Morsh and Wilder, 9p. cit., 150 pp. 28 (3) concomitants of teacher behavior. The number of predictors which investigators have studied in attempts to find significant relationships to effectiveness are many. Among the predictive char- acteristics more often studied and for which measurement has been attempted, are:3O Scores on tests of verbal and other cognitive abilities; Scores on tests of knowledge and understanding of general and special subject matter: Scores on tests of professional information: Course marks representing academic achievement: Course marks or ratings representing performance in student teaching; Amount of general and of professional education: Scores derived from inventories and/or projective devices develOped to measure various personality traits, and emotional and social adjustment: Scores on attitude scales and inventories devel- Oped to measure teacher-student relationships: Age: Experience: Sex; Marital status: Socio-economic status; Speech and voice characteristics: Factors influencing choice of teaching as a career; Social participation: Expressions of interest in, participation in, and preference for various sorts of activities.31 Progress toward an understanding of teacher behavior and the problem of teacher effectiveness and its prediction has proceeded slowly. One of the principal reasons reported for this is the lack of attention which has been devoted to theory develOpment, which, of course, restricts the gener— ation of hypotheses. 3ORyans, 2p. cit., p. 1488. 3llbio., p. 1488. 29 As research in prediction is examined, it is found that most of the studies producing information about teacher characteristics in relation to teaching have been derived from correlation studies. There has been little evidence produced which would aid in the understanding of cause and effect relationships. Examination of the literature concerning one of the possible predictors, that of tested intelligence, reveals that in the 60 indexed studies over the last 30 years where correlations have been run with various teacher ratings, there is tremendous variance in findings. The highest re- lationship, a correlation coefficient of .57 with student gains, was reported by Rostker32 for a group of 28 teachers. Among the 60 available studies in which correlations are re- ported between intelligence scores and various criteria of teacher effectiveness, the number of subjects is often so small, some with as few as six, that the correlation coef- ficients reported have little meaning.33 Morsh and Wilder point-up the short comings of intelligence as a predictive measure in the following way: Considering the more or less restricted range into which the intelligence of the public school teacher may be eXpected to fall (intelligence quotients with a range of 103 to 126 and an average of 114 as reported in findings with the Army Alpha): for all practical purposes 32Rostker, L. E., "The Measurement of Teaching Ability," Journal 2: EXperimental Education, 14:6-51, September, 1945. 33Morsh and Wilder, 9p. cit., pp. 60-61. 30 this variable is of little value as a single predictor of rated teacher successes, inasmuch as it would be used with a pOpugation already selected on the basis of in- telligence. 4 Efforts to use socioeconomic status of the instructor as a predictor is typically reflected in a study by Ullman.35 He used, among other measures, the Sims Score Card to deter- mine socioeconomic status of 116 junior and senior high school teachers with one semester experience. Coefficients resulting from correlations of socioeconomic status scores with social intelligence, general intelligence, knowledge of principles of teaching, knowledge of aims of secondary education, self- rating, academic marks, education marks, major subject marks, and practice teaching rating were near zero. Most of the studies where socioeconomic status was used as the factor to be tested for relationship with teacher effectiveness have used supervisory ratings as the criterion, and these ratings are very often found to be negatively cor- related with pupil gains. A number of investigators have studied the effect of the sex of teachers as it relates to their effectiveness. The conclusions reached in most cases are that no particular differences have been shown when the relative effectiveness of men and women teachers has been compared. However, 34Ibid., p. 65. 35Ullman, R. R., ”The Prediction of Teaching Success," Educational Administration and Supervision, 16:598-608, November, 1930. 31 36 Nemec in 1946 and COOper and Lewis in 195137 found small differences in favor of men teachers. Research outside the field of teacher education which is interestingly related to various parts of the problems herein discussed is represented by the work of J. J. Crowly.38 His study is a follow-up study of 485 graduates of ten con- secutive high school classes in a small town. The purpose being to determine the graduates' degree of leadership in their community. Cole's review of the study follows: Adult success in leadership was judged on the basis of general reputation, positions of trust (school superintendents, bank managers, judges, superior officials) or superior positions in business or industry, ownership of business, and election to chairmanship of community undertakings. In their high school days the 186 male graduates had shown the four levels or kinds of leader- ship in school life: 64 were prominent athletes, 22 played dominant roles in nonathletic student affairs, 23 were outstanding in both these classifications, and 77 had no record of any leadership. Nearly two thirds of the second and third groups became leaders in adult life. The student who was prominent in athletics but nothing else did not fare so well in later years. Only a few nonleaders in high school became leaders as adults. Among the 299 women graduates, only 59 had occupied pos- itions of leadership in school. Of these, 37 per cent held such positions as adults. Only 2 per cent of the 240 other women graduates, all nonleaders in higggschool, had had success as leaders in their communities. 36Nemec, L. G., "Relationship Between Teacher Certifi— cation and Education in Wisconsin: A Study of Their Effects On Beginning Teachers, "Journal 2: Experimental Education, 15:101-132, December, 1946. 37COOper and Lewis, pp. cit., pp. 703-707. 38Crowley, J. J., "High School Backgrounds of Success- ful Men and Women Graduates," School Review, 48:205-209, March, 1940. 39Cole, Luella, Psychology 2: Adolescence, pp. 424-425. 32 Summary Pupil rating of teacher effectiveness has been widely researched during the past 40 years. As a result of the re- cent growth of interest in merit pay the possible use of pupil rating has added considerable interest to studies of the subject. The reliability of such ratings appears to be high enough to make them usable for many purposes in today's high schools. The correlation Of pupil rating with ratings by others is low, and it is likewise low when the correlation is calcu- lated with most teacher background factors. The one place where there may be a significant relationship is between pupil opinion and pupil-teacher interaction. The study of predictors of teacher effectiveness has produced no emergence of a factor or factors strongly enough related to teacher effectiveness to be very useful. Ryans in his final summarizing paragraph in The EncyclOpedia 9: Education Research asserts: Certain of the above-named characteristics, then, do seem to be associated with certain dimensions of teacher behavior and teacher effectiveness, although the extent of obtained relationships frequently has not been high. It is important here to recall that relationships and differences which have been noted are in terms of averages for groups of teachers and any obtained relation- ship is limited by, and may be eXpected to vary with, conditions outlined above. The usefulness of research findings pertaining to the prediction of teacher effec- tiveness will be greatest when the results are considered 33 in an actuarial context, rather than in attempting highly accurate prediction for given individuals, and when variations in relationship found among different classifications of teachers and with the use of dif- ferent approaches to the pregactor criterion relation- ship are taken into account. 40Ryans, 22. cit., pp. 1490-1491. 34 CHAPTER III METHOD OF INVESTIGATION The purpose of this study was to investigate the relationship between selected background factors of student teachers and pupil opinion of certain teaching traits. In addition, the investigator proposed to study the possibility of relationships between various combinations of pupil's and student teacher's home community size and the teaching traits as rated by pupils. The steps which were taken in carrying the study through to its completion were outlined briefly in Chapter I. A more detailed account of the methods used in the procurement and treatment of the data follows. Sources of Data Central Michigan University, the teacher education institution to which this study was confined, has listed in its catalog two plans for preparing candidates for secondary school teaching. One plan, Plan A,1 is designed for the student who is continuously enrolled at the institution over a period of eight consecutive semesters. This program in- cludes a number of education courses which must be taken in sequence: these normally culminate in an eight week period of full-time student teaching in a public school. The other 1Central Michigan University Bulletin, l959-l9o0, p. 79. 35 plan, Plan B,2 is designed for the student who transfers from another institution or finds it necessary to enroll inter- mittently, possibly as a part-time student. For this student the course sequences are frequently broken, and other con- cessions made. In the present study the sample was drawn from Plan A students only, who were seniors during the 1959-1960 school year. It was necessary for purposes of this study that these students be enrolled in student teaching during their senior year. The total number eligible to participate was found to be 315. Of this number it was discovered that thirty were destined to teach in special education, e.g. speech correct- ion, mentally retarded, socially maladjusted. This elimin- ated them from the study because it was believed that the responses of their pupils would be quite difficult to obtain within the framework of this study. The remaining 285 students were intended to constitute the sample until it was discovered that one junior high school where student teachers are placed by Central Michigan University refused to COOperate in the study. The reason given by the school principal was that soliciting pupil Opinion concerning student teachers might negatively affect the attitudes of pupils toward all future student teachers. Since nineteen students were to teach in that school the sample for this study was limited to 266 seniors on Plan A at Central Michigan University during the 1959-1960 school year. 2Ibid., p. 79. 36 The 266 seniors were asked as a part of their psy- chology and education classes, which met on campus prior to their eight week student teaching period, to COOperate in this study. The investigator personally entered each psychology and education class and combined an eXplanation of the study with the administration of the student teacher background questionnaire.3 Student teachers at Central Michigan University are assigned to teach for an eight week period, and there are four eight-week student teaching periods during the school year. Students may be assigned to any one of nine student teaching centers. During the 1959-1960 school year the 266 student teachers in this study were assigned as shown in Table 1. TABLE 1. NUMBER OF PARTICIPATING STUDENT TEACHERS ASSIGNED IN EACH OF THE NINE STUDENT TEACHING CENTERS Student teaching centers * Number of participating student teachers assigned Bay City 21 Cadillac 24 Clare 19 Ludington 21 Manistee 24 Midland 35 Mt. Pleasant 75 Saginaw 34 Scottville 13 Total 266 3 A c0py of the student teacher background questionnaire may be found in Appendix A. 37 It can be seen from Table 1 that the distribution of student teachers among the centers presents a somewhat balanced picture with the notable exception of Mt. Pleasant which, because of its proximity to campus, has a greater number of student teachers than its size alone would warrant. Contacts with the school systems and supervisors of student teaching were initially made by the Director of Student Teaching who is also the Associate Dean of the School of Education at Central Michigan University.4 Each of the contacted parties agreed to COOperate in the study. The Director of Student Teaching and the investigator discussed the purposes of the study at some length with the student teaching coordinators and secured their help as distributors and collectors of the Pupil Cpinion Questionnaire. Each student teacher was rated by one class with which he had had regular contact for a period of seven weeks. All Pupil Opinion Questionnaires5 were administered to pupils by the student teacher's supervising teacher, with the student teacher absent from the room, on Tuesday morning of the seventh week of student teaching. Table 2 is presented to show the number of pupil responses obtained for this study. 4A COpy of the letters from the Director of Student Teaching may be found in Appendix B. 5See Appendix B. 38 TABLE 2. NUMBER OF PUPIL RESPONSES GIVING PUPIL OPINION OF TEACHING TRAITS Classrooms Pupils Mean Range Class size 266 7073 26.59 10-49 Table 2 indicates that the sample of pupil Opinion was very broad and that the mean class size was very near the state average for classes in Michigan secondary schools. Though the range of class size shown in Table 2 is 10-49, the investigator's examination of returns revealed a large cluster in the 20's and 30's. Instruments The Student Teacher Questionnaire was develOped to Obtain desired eXperiential background information about the student teachers participating in this study. The items re- present a number of years of thought, reading, conversation, and speculation on the part of the investigator. True, there are many other background factors which might have been in- cluded, but those selected are the ones which were most continuously brought to the foreground by authors Corey6 7 and Fligor, and which the superintendents interviewed 6Corey, S. M., "The Present State of Ignorance About Factors Affecting Teaching Success," Educational Administ- ration and Supervision, 18:481-490, October, 1932. 7Fligor, R. J., Anal sis g: the Evaluation, Use, and Value gf_Certain Competancies for Beginning the Student Teaching Experience, 218 pp. 39 informally in teacher education institution placement offices would defend as important. In addition, the examination of applications for student teaching assignments revealed the incorporation of many of these items. The form of the Student Teacher Questionnaire and some of the items were drawn from duplicated material from the Michigan State University Social Research Service.8 Most items are self eXplanatory in terms of content desired, and additional information concerning the more complex items will be given under the Treatment of Data section of the chapter. The Pupil Opinion Questionnaire used is one adapted 9 Dr. to this study from a form prepared by Roy C. Bryan. Bryan's questionnaire is one which was designed after very careful study of the research reviewed by Beecher, who makes the comment, "Attention is called to the consistency of findings in the pupil-reaction studies reviewed. If 30,000 boys and girls react similarly to certain teacher behavior, it must certainly follow that these strategic behaviors de- serve serious consideration by the teacher as well as by all 8The sixth draft of a student questionnaire by Dr. Wilbur B. Brookover, dated March 27, 1952. 9Bryan, R. C., ”Student Reactions and Merit Salary Schedules," Faculty Contributions, 4:1-67, July, 1958. Per- mission granted by Dr. Bryan for the use of the cited ques- tionnaire. 40 who attempt to evaluate the latter's effectiveness."lO Bryan also used the questionnaire extensively at Western Michigan University in Kalamazoo, Michigan. The investigator had two principal sources of help in perfecting the adaptation of the questionnaires for this study. One was a seminar in sociology taken with Dr. Wilbur B. Brookover in which the class members and instructor gave many helpful suggestions. The second was a pilot study of four student teachers and their classes at Mt. Pleasant, Michigan, Junior High School. This study made it possible to re-word some of the statements so that clarity would be better insured. The instructions at the beginning of each questionnaire received the most attention, and it is believed that the results of the pilot study gnntly facilitated later administration of both instruments. Treatment of the Data In this study an analysis was made of the degree of relationship, if any, between 19 background factors of student teachers and 10 teaching traits of the student teachers as rated by their pupils. A description of the procedures used to quantify these two types of data, and of the methods employed to compute the correlation coef- ficients between them, follow. lOBeecher, Dwight E., The Evaluation of Teachin , Syracuse University Press, New York, 1949, 165 pp. 41 The information concerning the background factors was translated to numerical data either by counting the number of characteristics in each category of a dichotomous factor or in each interval of a continuous scale. The numerical data were then coded in digit form as shown in the Student Teacher Questionnaire, with a few exceptions which are described be- low, and transcribed to tabulating cards by means of a punching machine. The socioeconomic background data were quantified by converting the reSponses to item 12 in the Student Teacher Questionnaire to numerical scores by first placing them on the North-Hatt Occupational Rating Scale11 and then dividing the scale arbitrarily into five equal categories. The oc- cupations were then coded into five classes from a low of one to a high of five and placed on punch cards. The background data concerning the degree of leader- ship demonstrated by the student teachers while they were in high school were quantified by counting the number of re- sponses listed in the third column of item 18 on the Student Teacher Questionnaire. The background information describing the number of high school organizations participated in by the student teachers was obtained by the number of entries in item 19 of the Student Teacher Questionnaire which had a value of fifty per cent or larger. 11North, and Hatt, gp. cit., pp. 464-474. 42 The 10 teaching traits of the student teachers were evaluated by pupils on a four-category scale as shown on the Pupil Opinion Questionnaire. These evaluations were quanti- fied by arbitrarily assigning the "below average" rating a value of one: the ”average" rating a value of two: the "good" rating a value of three: and the "excellent" rating a value of four. A median rating for each of the ten teaching traits was computed for each student teacher and placed on punch cards according to the following code: 1 1.00-1.49 5 3.00-3.49 2 1.50-1.99 6 3.50-3.99 3 2.00-2.49 7 4.00-4.49 4 2.50-2.99 8 4.50-4.99 From the above code it may be seen that the limits of the class intervals 1, 2, 3, and 4 were defined as follows: the interval ”1" extended from 1.00 to 1.99: the interval "2" from 2.00 to 2.99: the interval "3" from 3.00 to 3.99: and the interval "4" from 4.00 to 4.99. After the background and teaching trait variables were quantified, coded, and placed on punch cards, they were placed in electromechanical machines in order to obtain in- dices of the degree of correlation, if any, between them. Since 15 of the 19 background factors and all of the 10 teaching traits were in the form of metric data, 150 43 product-moment coefficients of correlation were calculated. These r's are presented in tables together with the .95 and .99 confidence intervals and the coefficients of forecasting efficiency. The tests of relationship between the teaching traits and those background factors which were described in terms of frequency data were made by the use of the chi square technique. The background factors falling into the latter category were sex, marital status, membership or non-mem- bership in student council, and membership or non-membership in Future Teacher clubs. Because preliminary analysis indicated a complete lack of relationship with other factors, the chi square analysis of relationship between the four dichotomous background variables and the teaching traits was based only on the all-around teaching factor. In addition, the all-around teaching ability factor was chosen because it is a summary item and was found to correlate to a rather high degree with other teaching trait items. For the purposes of the chi square tests, the all- around teaching factor was coded into three classes as follows: (1) below average and average, (2) good, and (3) excellent. The "below average" and "average" ratings were grouped together due to the small number of frequencies in the "below average" category. 44 Because other studies have shown significant rela- tionships between leadership in high school organizations and leadership in adult activities, the relationship between the student council membership background factor and all- around teaching ability was studied by means of the chi square technique. Finally, the chi square test was used to analyze the relationship between the different combinations of sizes of home communities of both the student teachers and of the pupils who evaluated their teaching traits on the one hand and all-around teaching ability on the other. The sizes of the home communities were divided into the three categories Of large, medium, and small and coded into nine classes as follows: Code Size of home community Size of home community number of student teacher of the pupils 1 large large 2 large medium 3 large small 4 medium large 5 medium medium 6 medium small 7 small large 8 small medium 9 small small All-around teaching ability was classified into three groups as described earlier. 45 Summary In this chapter the methods of the study have been described in some detail, and the persons participating in the study have been identified. The selection and develop- ment of questionnaires was outlined, and the procedures used in the analysis of data were discussed. 46 CHAPTER IV PRESENTATION AND ANALYSIS OF DATA As described in Chapter I, the two major purposes of this study were (1) to investigate the relationship between 19 background factors of student teachers and 10 teaching traits of student teachers as revealed through a pupil Opin- ionnaire: and (2) to investigate the relationship between certain combinations of home community size of student teacher and pupil background factors and all-around teaching ability of the student teachers as appraised by pupils. Chapter IV presents and analyzes indices of relationship between these variables of student teacher and pupil back- ground factors on the one hand and teaching traits of the student teachers as rated by pupils on the other hand. Since it was assumed that most Of the background factors and all of the teaching traits could be measured on a continuous scale and that any possible relationship between these variables would approximate linearity, Pearsonian co- efficients of correlation were used in most cases. In those instances in which a factor constituted a truly dichotomous variable, chi-squares were computed from contingency tables. Table 3 presents the 15 Pearsonian coefficients of correlation which resulted from testing the relationship be- tween 15 background factors and the student teacher's know- ledge of the subject taught as rated by pupils. Also included 47 .mocoAOApuo .4 Raucognd up HHfiM nH mnAvmeovou no paoAOAmmooo I m .OOApmHoAAoo I A endow on has mpOpomm ogdonmxoen cepmfieonnpde aIIIIIIIIIIIIIIIIIIIIIIIIIIII moo. $0.: on Hm. mo.l on ma. mo. onoom nAnonowog moo. no.1 op mm. :o.u op mA. so. odoppeupeewso po .oz Aoo. A.u op mA. :A.u op 0. mo.- aoApApppoo Aepsaeopep co .oz moo. m.u op no. 5A.- op o. mo.u oppose Append» po .oz Aoo. mA.- op RA. no.1 op MA. mo. Aoonoo nMAp po ouAm 0H0. Ho.l op mw. mo. op mm. :H. oocwccoppm £09520 moo. oa.l 0» 0m. 00.1 on we. mo. mapmpm opEocooooaoom Aoo. MA.- op RA. mo.- op MA. mo. owdAApAo AopoA 606. mA.- op mA. AA.- op AA. 06. oeopapu sovo po .oz moo. 50.- op mm. no.1 op dA. mo. aeopapa somdnoe po .oz 060. mA.u op mA. AA.- op AA. 66. oeonposp eocAo po .oz moo. om.u op 6A. SA.n op so. mo.- unopposp somesoe po .oz 606. mA.- op mA. AA.- op AA. 60. up posses drop po oepm Aoo. mA.- op NA. :A.u op mo. mo.u someone oodoepooe po .oz Aoo. :A.n op SA. 0A.- op NA. Ao. one mm. mm. smnopomu canopmxown m H lflafldlal 02538.80 .Aonomop pcooapm mAHmDm Mm amadm md HmobpOpcA oocoppmcoo c nonomop pnocnpm mAHmDm wm mmadm md Namquo mosza zH¢qum OH NBHAHmd .mmmmo¢ma HzmQDBm awe 92¢ wmoao4m QZDOm0M0¢m rampage azmapam mA zeaxawm AAV zoApaAmmmoo so mpzmAOAeemoo_zAAzommemm .: mqmda 51 Since r's as large as .11 and .15 may occur from pure chance fluctuations in samples containing 266 subjects, when the true pOpulation r is zero, 5 times in 100 trials and 1 time in 100 trials respectively, socioeconomic status is the only background factor which appears to be related to the ability to eXplain things clearly.4 And the degree of rela- tionship represented by an r of .16 is very small. As listed in Table 4, the coefficient of forecasting efficiency for an r of .16 is approximately .01. Table 5 presents the 15 Pearsonian coefficients of correlation which resulted from testing the relationship between 15 background factors and the student teacher's fairness in dealing with pupils as rated by pupils. In addi- tion Table 5 includes for each r the .95 and .99 confidence intervals as computed by use of Fisher's 2 function, and indices of forecasting efficiency. It may be observed in Table 5 that the background factors of the 266 student teachers are not closely related to the student teachers' fairness in dealing with pupils as rated by pupils. While the number of residence changes, size of town reared in, number of older brothers, and size of high school attended appear to correlate highest with the student teachers' fairness in dealing with pupils, the 4loid. p. 201. 52 .honOAoapmo mappmaoonop no pOOAOAMMOOO I m .aOppmAOAAOO I A .4 Hpcnonnd up AHOM up venom on has mpopomp unponmxowp copappoppnds «Amppopcp oocmcppdoo mAHmDm mHHS 02Hq4mm zH mmmzdeh mmmo¢ma Bzmnbam ma zmm3amm ARV ZoHadqmmmoo m0 mhzmHUHhmmoo z¢Hzowmdmm IIIIIIIIII4IIIIIIIIIIIIII A00. :A.l Op 0H. onoom nAnmAOOwoq H00. wH.I Op NH. :a.| Op mo. m0.| adOppmupnempo mo .02 N00. AA.| op 0H. 50.: Op mA. :0. mmppppppou Aennanopnp no .Oz moo dA.u op AA. mA.a op no. 16.- messed ppAoeop po .oz 0A0. A0.: op mm. :0.: op mm. A. Aoonom amp: no oupm A00. NA.| op 0H. 00.: Op :A. 0. OOAAscceppe nonsno ~00. AA.I Op 0H. 50.: Op mA. :0. mppmpm OAEOOOOOOAOom N00. AA.I Op 0A. 50.: Op mA. :0. mwOAAnAm Aepoa 000. mA.I Op ma. AH.I Op AA. 00. mAOpmAm nocAo no .Oz A00. mA.| Op NA. 00.: Op mA. N0. mpopmpm newcsoh MO .02 moo. 50.: Op mm. m0.l Op 0A. m0. wnonpopp AOOAO mo .Oz N00. 0A.: Op AA. mA.I Op 50. :0.: unoppopp nomnsom no .02 moo. m~.u op mo. mA.u op so. no.1 dA posses drop po oaAm 000. 0N.t op :0. mm.1 Op 00. AA.| mmmnmno ooOOOAmop no .Oz 600. mA.1 op mA. AA.- op AA. 06. owe 00. mm. ImAOpOOp canonmxomp m IIIIIIIIIIIIIII u ponowop pdoofipm mAHmDm Hm Omadm m4 .mmmmodma azmQDBm mma Qz< mmoaodh DZDOmGModm .m mqmaa 53 resulting coefficients of correlation are very small, not exceeding .14. Since r's as large as .11 and .15 may occur from chance fluctuations in samples containing 266 subjects, when the true pOpulation r is zero, 5 times in 100 trials and 1 time in 100 trials respectively, size of high school is the only background factor which appears to be related to the student teachers' fairness in dealing with pupils. And the degree of relationship represented by an r of .14 is very small. As listed in Table 5, the coefficient of forecasting efficiency for an r of .14 is approximately .01. Table 6 presents the 15 Pearsonian coefficients of correlation which resulted from testing the relationship be- tween 15 background factors and the student teacher's ability to maintain good discipline as rated by pupils. Also in- cluded in Table 6 for each r are the .95 and .99 confidence intervals as computed by use of Fisher's 2 function, and indices of forecasting efficiency. Table 6 indicates that the coefficients of correlation between 15 background factors of 266 student teachers and the student teacher's ability to maintain good discipline as rated by pUpils are small. Size of town reared in, socio- economic status, and size of high school attended are the background factors which appear to be correlated highest with the ability to maintain good discipline. The r's resulting from these correlations range from -.13 to .14. 54 .hocOpOpppo .4 upooooop op ppsp op mappmooonom Ho pGOpopmpooo n m .aOpproppoo I 9 union on has mpopomp onpopmxowp copopbopppdo A00. AA.: op 0A. 50.: Op mA. opoon npnmpmooop A00. NA.: Op mA. m0.: op :A. 0. maOppoupcwwpo no .02 ooo. mp.: op mp. pp.: op pp. 00. ooppppppoo popsapopop po .oz poo. op.: op :p. mp.: op op. po.: mopozo pppmpop po .oz opo. po.: op om. mo. op mm. :p. poonoo nap: po oopm ooo. mp.: op mp. pp.- op pp. oo. oooooooppo noppno M00. 0.: Op mm. :0.: Op A. no. mnpopm opaocooOOpoom poo. p.: op mp. mo.: op p. we. omoppopo popoa ooo. mp.: op mp. pp.: op pp. oo. opopopm poopo po .oz poo. mp.: op mp. mo.: op mp. mo. opopmpm Homoso» po .oz moo. mp.: op pm. mo.: op mp. oo. oponpopp poopo po .oz moo. pm.: op mo. mp.: op mo. mo.: oponpopp pomosop po .oz moo. mm.: op «0. :m.- op mo.: p.: op oopoop osop po oupm poo. mp.: op p. 0.- op :p. mo. mowoono ooooopmop po .oz poo. mp.u op p. p.: op mo. mo.- omp mm. mm. omHOpowH ccdopmxoon m .A umpo oooo 2ppazp42 op pppppmp mmmopma azmopam mp zmm39mm ppv 2oppppmmmoo mo mpzmpopmmmoo zHA mmdqo mma mad: OB mmmo Mme 02¢ mmoaodh Dzbomwmodm ammodma Hzmobam mA zmmzamm any 20Hadqmmmoo ho mazmHlohmoo z