"‘7: A STUDY OF COMPUTER SCORED GROUP IIOLTZMAN .INKBI.0T VARIABLES AS RELATED TO STU-DENT TEACHING SUCCESS, MAJOR TEACHING FIELDS, AND SEX Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY B. BRADLEY WEST ' ' 1:959 A w- ‘1- van-1mm E V inaeas L 1' B [if .4 R Michigv tare Univcm’y This is to certify that the thesis entitled A STUDY OF COMPUTER SCORED GROUP HOLTZIVIAN INKBLOT VARIABLES AS RELATED TO STUDENT TEACHING SUCCESS, MAJOR TEACHING_ FIELDS, AND SEX presented by B. Bradley West has been accepted towards fulfillment of the requirements for _.Eb_._D__ degree in_EdUCati on Wflwflmflg/ Majéy professor Date November 141 1969 0-169 ABSTRACT A STUDY OF COMPUTER SCORED GROUP HOLTZMAN INKBLOT VARIABLES AS RELATED TO STUDENT TEACHING SUCCESS, MAJOR TEACHING FIELDS, AND SEX By B. Bradley West Purpose of the Study For years researchers have attempted to identify the personality correlates of successful teachers. This study sought to deal with this problem in terms of per- ceptual measures. Specifically, the purpose of the study was to determine the relationship between measures of perception as identified by Holtzman Inkblot variable scores and success in student teaching. Sub-problems studied the relationship between perceptual measures and student teachers grouped according to sex and major teaching field. Methodology The group version of the Holtzman Inkblot Test was administered to 802 Michigan State University student teachers during Winter Term, 1968, in each of the geo— graphical centers where students are assigned for student teaching. Forty—five inkblot slides were exposed to all the student teachers for one minute each and the subjects B. Bradley West responded with a short written description of what each inkblot "looked like" on the specially designed group answer sheet. The answer sheets and identifying data were keypunched, loaded onto magnetic tape and computer scored by the Holtzman Inkblot Scoring Service in Perry Point, Maryland. Output from the Scoring Service in- cluded a print out and a score summary card for each subject in the sample on seventeen Holtzman variables: Location, Rejection, Form Definiteness, Color, Shading, Movement, Integration, Human, Animal, Anatomy, Sex, Abstract, Anxiety, Hostility, Barrier, Penetration and POpular. Findings of the Study When the sample of student teachers was grouped on the basis of success ratings at the conclusion of their student teaching, a study of the mean Holtzman scores for the above seventeen variables showed that only one vari- able score, Rejection, was significantly different among the groups. But the unequal variance of the R variable among the groups suggested that even this single signifi- cant finding was questionable. A re-grouping of the student teachers by sex and then success ratings failed to identify any other significantly different variable scores. When the student teachers were grouped by sex alone with no additional sub-grouping, eleven of the B. Bradley West seventeen variable scores significantly differentiated between males and females. Thus, the Holtzman variable scores would appear to be of some usefulness in discrimi- nating between male and female student teachers and suggested the importance of analyzing the data for the sex influence. When the sample of student teachers was grouped into thirteen major teaching areas, nine of the seventeen variables differentiated among the various teaching areas. However, when the sample was first grouped according to sex and then sub-grouped into major teaching area, one variable significantly differentiated among the males while eleven variables were significantly different among females. Thus, the Holtzman scores would appear to be of little or no value in discriminating among male student teachers grouped according to major teaching fields but would be of some, but restricted, value when applied to females grouped according to major teaching area. The study also generated perceptual norms on the Holtzman for Michigan State University student teachers and determined the feasibility of administering and com- puter scoring a projective technique with a large (N = 802) pOpulation. A STUDY OF COMPUTER SCORED GROUP HOLTZMAN INKBLOT VARIABLES AS RELATED TO STUDENT TEACHING SUCCESS, MAJOR TEACHING FIELDS, AND SEX By B. Bradley West A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Student Teaching Office 1969 GC/72’7 4» 2770 ACKNOWLEDGMENTS The investigator wishes to express his gratitude and thanks to the many colleagues, faculty members, students and friends who gave their time, cooperation, and effort to this study. The experimenter is indebted to Dr. W. Henry Kennedy, Committee Chairman and Dissertation Director, for his counsel, guidance, thought provoking questions and encouragement as the study was conceived and under- taken. With his scholarly guidance and practical advice, the study was strengthened and considerably improved. To Dr. Donald Gorham, Chief, Research Psychology, at the Veterans Administration Hospital in Perry Point, Maryland, the author offers special thanks for his interest, help and cOOperation in computer scoring the hundreds of test protocols; To my wife, Dorothy, for her support, critical reading and unfailing encouragement to get the job done; To Dr. and Mrs. Kent Gustafson who contributed significantly at the initial planning stages; To Dr. Albert Rabin, who first suggested the computer-scoring approach; 11 To Dr. Wayne H. Holtzman, Dean, College of Educa- tion, University of Texas, for providing the inkblot slides; To Drs. George Myers, John Suehr, and Arthur Seagull, members of the Guidance Committee, for their interest and critical evaluation of the dissertation; To the Student Teaching Coordinators for their splendid cooperation in arranging testing dates, times and locations; To The College of Educational Research Bureau and particularly to Mr. David Wright for assistance in the statistical analysis; To the staff of the Computer Laboratory who spent many hours loading the 40,000 keypunched cards onto magnetic tape; To The Student Teaching Office Graduate Affairs Committee for financial assistance in conducting the research; And finally, to the hundreds of unnamed student teachers who willingly took part in the study; To all of these, the investigator expresses his sincerest appreciation and gratitude. iii TABLE OF CONTENTS Page ACKNOWLEDGMENTS . . . . . . . . . . . . ii LIST OF TABLES. . . . . . . . . . . . . vi LIST OF APPENDICES . . . . . . . . . . . viii Chapter I. INTRODUCTION. . . . . . . . . . 1 Statement of the Problem. . . . . . 1 Significance of the Study . . . . . 1 Theoretical Basis of the Study. . . . 6 Importance of the Study . . . . . . 13 Definition of Terms . . . . . . . 13 Hypotheses . . . . . . . . . . 1A Assumptions . . . . . . . . . . 15 Limitations . . . . . . . 16 Procedures for Collecting Data. . . . 16 Description of the Sample . . . . . 18 The Instrument . . . . . . . . 18 The Computer Scoring System. . 21 Reliability and Validity of the Computer Scoring System . . . . . 23 Rating of Success . . . . . . . . 26 II. REVIEW OF RELATED LITERATURE . . . . . 31 Problems of Personality Definition . . 32 Criteria of Teacher Effectiveness Problems . . . . . . . . . . 34 Problems of Rating Teacher Effective- ness . . . . . . . . . . . “1 Problems in the Prediction of Teacher Effectiveness . . . . . . A6 Research of the Holtzman Technique . . 75 iv Chapter Page III. PRESENTATION AND DISCUSSION OF THE DATA. . 85 Presentation of the Data. . . . .~ . 88 Summary of the Findings . . . . . . 112 Discussion of the Findings . . . . . 115 IV. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS . 131 Summary . . . . . . . . . . . 131 Conclusions . . . . . . . . . . 133 Discussion . . . . . . . . . . 135 Recommendations. . . . . 139 BIBLIOGRAPHY . . . . . . . . . . . . . 1A3 APPENDICES . . . . . . . . . . . . . . 151 Table 10. 11. LIST OF TABLES Distribution of sample population by teaching field and sex 0 O O O O O I O O O A comparison of hand scoring and computer scoring for 17 HIT variables . . . . A comparison of computer scoring with three hand scorers . . . . . . . Odd-even reliability coefficients for hand and computer scoring. . . . . . . . . Distribution of success rating numbers. . . Means and standard deviations for 15 variables from individual inkblot technique and a similar set of 15 variables from group ink- blot technique. . . . . . . . . . . Correlation coefficients between hand scored individual and group HIT records and computer scored individual and group HIT records . . Overall mean scores and standard deviations of the Holtzman Inkblot variable scores for the student teacher population . . . . Selected Holtzman variable scores and corresponding percentiles of Michigan State University student teachers . . . . . Means, standard deviations, F statistic and its approximate significance of Holtzman variables among student teacher groups claSsi- fied according to rated success . . . . Means, standard deviations, F statistic and its approximate significance of Holtzman Inkblot variable scores among student teachers grouped according to sex . . . . vi Page 19 2A 25 26 29 81 8A 86 87 89 91 Table 12. 13. 1A. 15. 16; 17. 18. Means, standard deviations, F statistic and its approximate significance of Holtzman Inkblot variable scores among male student teacher groups classified accOrding to rated success . .- . . . . . . . . Means, standard deviations, F statistic and its approximate significance of Holtzman Inkblot variable scores among female student teacher groups classified according to rated success . . . . . . . The student teacher population (excluding junior high school student teachers) classi- fied according to major teaching field and sex 0 O O O O O O I O C O O 0 Means, standard deviations, F statistic and its approximate significance of Holtzman variable scores of student teachers Classi- fied according to major teaching field. . . Means, standard deviations, F statistic and its approximate significance of Holthan variable scores of male student teachers classified according to major teaching field. Means, standard deviations, F statistic and its approximate significance of Holtzman variable scores of female student teachers classified according to major teaching field. Summary of findings . . . . . vii Page~ 97 99 101 102 107 110 113 Appendix LIST OF APPENDICES Description of the 17 Computer—Scored HIT variables . . . . . . Success Rating Chart . . . . . Cover Letter to Student Teachers HIT Response Sheet. . ._ . . . . Sample of Input - Sample of Output Cover Letter to Staff Success Rating Form . . . . viii Page 153 155 157 159 161 163 165 CHAPTER I INTRODUCTION Statement of the Problem The major problem of this study was to determine the relationship between computer scored group Holtzman Inkblot variables and success in student teaching. Sub- problems included study of the relationships between Holtzman variables, and major teaching fields, and sex of the student teachers. Significance of the Study For more than a generation educational researchers have attempted to identify the personality correlates of successful teaching behavior. The question of how to describe the successful teacher personality has been the object of hundreds of research inquiries. Contemporary literature in personality and teaching success provided ample support for the belief that personality factors contribute significantly to successin teaching. Getzels and Jackson, for example, reported conclusions in a major compilation of teacher personality research studies, and stated that, "the personality of the teacher is a significant variable in the classroom. Indeed, some would argue that it is the most significant variable."l Symonds, a prolific writer and researcher on teacher effectiveness, reported, "it is now well known that per— sonality is an important factor in the effectiveness of the teacher."2 Arthur Combs, a leading spokesman in teacher education, summed up the literature when he erte, "A good teacher is primarily a personality."3 Administrators of teacher education programs also accept personality as an important factor in admission to programs. Studies by Stout!"l and Magee5 confirmed that teacher training program administrators consider per- sonality the single most important selection factor; Willcox and Beigel6 wrote that such measures could serve 1J. W. Getzels and P. W. Jackson, "The Teacher's Personality and Characteristics," Handbook of Research on Teaching, ed. by N. L. Gage (Chicago: Rand McNally and Company, 1963), p. 506. 2Percival M. Symonds and Stephanie Dudek, "Use of the Rorschach in the Diagnosis of Teacher Effectiveness," Journal of Projective Techniques, XX (June, 1956), 227. 3Arthur W. Combs, "The Personal Approach to Good Teaching," Educational Leadership, XXI (March, 196A), 327. “Ruth Stout, "Selective Admissions and Retention Practices in Teacher Education," Journal of Teacher Edu- cation, VIII (December, 1957), A225A32. 5Robert M. Magee, "Admission—Retention in Teacher Education," Journal of Teacher Education, XII (March, 6Isabel Willcox and Hugo G. Beigel, "Motivations in the Choice of Teaching," Journal of Teacher Education, IV (June, 1953), 106-109. as selection criteria for admitting candidates into teacher education programs. And the profession at large, speaking through the Teacher Education and Professional Standards Commission (TEPS) of the National Education Association recommended that, Screening should include observation of per— sonality traits and emotional stability . . and colleges should devise and use a measure of personality.7 Furthermore, teacher employers are reluctant to hire candidates without some specific comments about the per- sonality factor, as Charles8 suggested. But there is a gap between Opinion and practice. Inlow,9 studying thirty-eight midwestern colleges offering teacher education, reported that only eleven of the thirty- eight institutions used any standard procedure of per- sonality evaluation and, of these, all but two relied entirely upon teacher-student interviews for accomplishing their purpose. 7National Commission on Teacher Education and Pro- fessional Standards, The Education of Teachers: Con- siderations in Planning Institutional Programs (Washington, D.C.: National Education Association, 1960), p. 6. 8Harvey Charles, "The Use of a Projective Technique in Teacher Selection," National Catholic Educational Association Bulletin, LVII (August, 19617, 172-173. 9Gail M. Inlow, "Comparative Study of Student Teaching Practices in 38 Midwestern Institutions," Journal of Experimental Education, XXVIII (June, 1960), 337-3H9. It appears that although there is considerable Opinion favoring the use of personality as a method of selection, administrators are reluct- ant to make further use of this approach until the body of factual data concerning teacher characteristics has been enlarged and sub- stantiated.10 ll 12 13 Barr, Mitzell, Getzels and Jackson, and Ryans,]‘Ll extensively reviewed recent efforts to enlarge and substantiate teacher personality data and concluded that no consensus exists on what combination of per- sonality factors are characteristic of a successful teacher. In the studies reported by Morsh and Wilder15 researchers attempted to investigate successful teaching using administrator ratings, peer ratings, self-ratings 10Thomas K. Allen, "Personality as a Predictor of Student Teaching Success," Journal of the Student Per- sonnel Association for TeacherTEducation, VI (Fall, 1966), 12. llArvil S. Barr, "Wisconsin Studies of the Measure- ment and Prediction of Teacher Effectiveness: A Summary of Investigations," Journal of Experimental Education, XXX (September, 1961), 5—156i 9 1“Harold E. Mitzel, "Criteria of Teacher Effective- ness," Encyclopedia of Educational Research, ed. by C. W. Harris TMacMillian Company, 1960), pp. 1A81-1A86. 13 1“David G. Ryans, "Assessment of Teacher Behavior and Instruction," Review of Educational Research, XXXIII (October, 1963), AlS-lAAl. Getzels and Jackson, op. cit., pp. 506-582. 15Joseph E. Morsh and Eleanor w. Wilder, Identify- ing the Effective Instructor: A Review of the Quantita- tive Studies, 1900-1952 (Research Bulletin, AFPTRC—TR 5H:AA, San Antonio: United States Air Force Personnel Training Research Center, 195“), pp. 1—151. and student gains criteria. The researchers also attempted to relate such variables as intelligence, age, scholarship, education and a host of other variables to successful teaching.l6 But most findings were inconclusive. The studies were hampered and compounded by situation, cri— terion and predictor limitations, teacher shortages, theoretical issues and difficulty in constructing instru- ments which predict teaching success. Thus, despite the voluminous research concerning this problem, very little evidence has emerged which might be generally useful in the selection of teacher education candidates and in future on-the-job predictions. In effect, this constituted a demand for information clarify- ing the relationship between personality and teacher success. Further research is needed that will lead to the discovery of specific and distinctive features of the successful teacher personality. Perhaps no one personality factor will ever be found to predict success in teaching, but rather a pattern of personality factors. As Lamke pointed out, success in teaching may require a balance of personality factors. Teachers are successful . . . because there is a kind of "balance" among their personality l6Simeon J. Domas and David V. Tideman, "Teacher Competence: An Annotated Bibliography," Journal of Experimental Education, XIX (December, 1950), 101-218. traits. Another kind of balance means they will probably be poor teachers. Lacking either, they will likely be average.l Lamke further suggested that success prediction demands an understanding of what is required for this balance. Study of the association of traits, one by one, with success will not suffice. Theoretical Basis of the Study However, within the course of this prodigious effort there has appeared a small but important group of studies which utilized projective techniques to study teacher personality which suggested an interesting phe- nomenon. These studieslB’lg’2O suggested, each in its own way, that teachers characteristically view their world--give it structure and meaning-~in ways that dif- ferentiate them from non-teaching groups. Further, this 17Thomas A. Lamke, "Personality and Teaching Suc- cess," Journal of Experimental Education, XX (December, 1951), 2D7. 18Florence L. Goodenough, Elizabeth M. Fuller, and Edna Olson, "The Use of the Goodenough Speed-of- Association Test in the Pre-service Selection of Nursery School-Kindergarten-Primary Teachers," Journal of Educa- tional Psychology, XXXVII (September, 19A6),'335-3H6. 19Thomas L. Hilton, "Ego Involvement in Teaching: Its Theory and Measurement by a Word Completion Technique" (unpublished Ph.D. dissertation, Harvard University, 1955). 2OMerle M. Ohlson and Raymond E. Schultz, "Pro- jective Test Response Patterns for Best and Poorest Student Teachers," Education and Psychological Measure- ment, XV (Spring, 19557: 18-27. perceptual phenomenon also appeared to distinguish teachers rated successful from those rated unsuccessful. The fact that students in training showed this perceptual tendency suggested, too, that it is not merely a product of experience. Since perception is a phenomenon of personality, it was reasonable to hypothesize that the personality struc- ture of successful teachers was significantly different from that of unsuccessful teachers. Arthur Combs wrote that, Whether an individual will be an effective teacher depends upon the nature of his private world of perceptions.21 Since direct research on personality, competencies, and other major variables has been unable to identify any common trait or practice of successful teaching, and in view of the positive indications of studying teacher personality through perception, the present investigation attempted to identify perceptual modes of student teachers classified according to (1) success rating, (2) sex and (3) major teaching field. Other possible classifications such as age, intelligence, and OCCUpational level also seemed appropriate, but since these are common background characteristics of the sample (i.e., all in the sample were in the same approximate age group, all met academic 21Arthur W. Combs, The Professional Education of Teachers (Boston: Allyn and Bacon, Inc., 1965), p. 19. requirements for student teaching and all were training to be teachers) it was assumed these characteristics would be of no significant effect. The effects of other background factors such as religion, race, or socio-economic status, however, were unknown and remained unspecified. Since most research studies draw samples of subjects that are biased in varying degrees with regard to some background characteristics, some understanding of the kind or amount of the influence of these factors needed to be developed. The most obvious of the background factors that could have influenced the results was the sex of the test subjects. The sex characteristic was shown to have significant influence on M.M.P.I. test behavior, for example. So pronounced was the effect of the sex char- 22 found it neces— acteristic, that Hathaway and McKinley sary to furnish separate norms for men and women on almost all of the basic clinical scales. Thus, the present study also attempted to identify perceptual modes of males and females to determine the effect of sex grouping on the results. 223. R. Hathaway and J. C. McKinley, The Minnesota Multiphasic Personality Inventory Manual (rev. ed.; New York: The Psychological Corporation, 1951), pp. 1A-16. \L.) The literature also suggested the appropriateness of a third classification. Allen23 reported that students who plan to teach different subjects scored differently on the California Psychological Inventory scales. Student teacher groups in English, for example, scored differently on the C.P.I. than those in social studies, physical edu- cation, or math and science, who also scored differently from each other. Allen's findings supported the conclu- sion that no single combination of characteristics de- fined all student teachers without regard to choice of teaching major. Thus, the present study also attempted to utilize a projective technique to identify perceptual modes of student teachers classified according to major teaching field. The term projective technique has been applied to a wide variety of data-gathering procedures ranging from such classic instruments as the Rorschach and the TAT to such less widely used procedures as the sentence completion, word association tests, and picture drawing. While the "correct" usage of the term may occasion theoretical argument, general usage labels "pro- jective" those techniques that require a person to bring structure to an unstructured test stimu- lus without knowing how his response will be evaluated.2 23Thomas K. Allen, "Personality Patterns of Student Teachers Planning to Teach Different Subjects," Journal of the Student Personnel Association for Teacher Educa— tion, IV (December, 1965), 92. 2“Getzels and Jackson, op. cit., p. 55A. is The general theoretical framework within which all projective techniques are fashioned is that there is a relationship between an individual's structuring of an ambiguous stimulus and his personality. How an individual structures or organizes meaningless stimuli into meaning— ful perceptions reveals basic aspects of his personality. Inkblots are most appropriate as stimuli because they are relatively meaningless and unstructured and permit a wide variety of individual responses. The subject, for ex- ample, when asked what he sees in an inkblot, must react in a personal, individual and unlearned fashion since there are no correct answers. And although the char- acteristics of the blot may condition the scope of the subject's response somewhat, his perceptions are selected and organized in terms of his fundamental personality structure. L. K. Frank wrote, The Rorschach method offers a procedure through which the individual is induced to reveal his 'private world' by telling what he 'sees' in the several cards upon which he may project his mean- ings, significance, and feelings, just because they are not socially standardized objects or situations to which he must give culturally pre- scribed responses. The Rorschach method is essentially a procedure for revealing the personw ality of the individual as an individual, as contrasted with rating or assessing him in terms of his likeness or conformity to special norms of action and Speech. It is just because a subject is not aware of what he is telling and 11 has no cultural norms behind which to hide himself that the Rorschach and other projective methods are so revealing.25 Although Frank wrote specifically about the Rorschach in the above quotation, his remarks could also apply to other inkblot tests or techinques. If, in realistic situations, a person's relation- ships with peOple are unsatisfactory, he may be unable or unwilling to perceive human forms in an inkblot. He may see an animal, a biological specimen or a forest from a tower. If, in his response, the subject uses the whole blot, major parts of it, or only a small detail, whether he uses the color, what he actually sees or is unable to see, what determines his response—~all are held to reveal some of his personality structure. However, direct personality hypotheses from pro— jective test responses can only be tentative. Klopfer wrote that, Direct behavioral parallels may not always exist . . . and may not be the most important information revealed. The behavior of an individual in the . . . situation differs in one very important respect from the same indi- vidual's reaction to any unfamiliar situation in life. In life situations, a person tends to behave in a more or less socially acceptable way. He has learned, if he is a 'normal' 25L. K. Frank, "Projective Methods for the Study of Personality," Journal of Psychology, VIII (1939), 389, quoted in Bruno Klopfer and Helen H. Davidson, The Rorschach Technique: An Introductory Manual (New York: Harcourt, Brace and World, Inc., 1962), pp. 1A—15. 12 individual, to control himself, to appear pleas- ant and kindly rather than aggressive and hos- tile . . . and to conceal from public view those traits that are not prized. Thus a person's outward, observable behavior may 8ften not reveal his true feelings and attitudes.2 In a projective testing situation a person does not know what the examiner is looking for or the typical way to respond. He reacts in his own particular manner and in so doing unconsciously reveals himself, even that part of himself of which he is not entirely aware. Therefore there is both an advantage and a dis- advantage in projective testing. There may be only Slight correspondence between actual personality and behavior and predictions of personality and behavior be- cause actual behavior is partly determined by the culture, environment and learned responses. Thus, any specific predictions of classroom personality and behavior of a student teacher must allow for learned responses to the environment. Since an analysis of the teaching environ— ment was not a part of this study, direct personality predictions were not made. However, if perceptual trends (e.g. successful student teachers see more humans than less successful student teachers) were related to successful student teaching, there would be sufficient 26Bruno Klopfer and Helen H. Davidson, The Rorschach Technique: An Introductory Manual (New York: Harcourt, Brace and World, Inc., 1962), p. 15. factual data to relate perceptual trends to observed classroom behavior. Importance of the Study The results of the study could have a significant impact on pre-training selection policies. In particular, if the null hypothesis (pages 14—15) tested were re- jected, there would be cause for review of present per— sonality assessment techniques. Certainly there would be cause to view teacher personality in a new light. In any case, it is not being argued that measures of perception could become the answer everyone is seeking, but only that they might provide one more significant piece of information for use by administrators of teacher education programs. Definition of Terms 1. Student teaching--that period of undergraduate guided teaching experience provided by Michigan State University as part of its teacher education program. 2. Student teachers—~those students enrolled in Education A36 (Student Teaching) at Michigan State Uni- versity during the Winter Term of the 1967—68 academic year. 3. University coordinators-~those staff members employed by Michigan State University who coordinated 14 the programs and supervised students enrolled in student teaching. A. Student teaching center—-one of the seventeen geographical clusters of school districts in Michigan where Michigan State University placed student teachers. 5. Group Holtzman Inkblot variable--one of the seventeen variables measured by the group Holtzman Ink— blot technique. 6. Rating of succeSS--any one of the seven rating numbers assigned to a student teacher at the conclusion of his student teaching experience by his university coordinator. Success rating numbers are operationally defined in Appendix B. Hypotheses The following hypotheses were made concerning the relationship of computer scored group Holtzman Inkblot variables and the sample of student teachers. 1. There will be no significant differences among the mean scores on each of the following Holtzman Inkblot variables for student teachers grouped according to rated success in student teaching by the University coordinators. (L) Location A B. (R) Rejection C (FD) Form Definiteness D (C) Color l5 (Sh) Shading (M) Movement (I) Integration (H) Human (A) Animal (At) Anatomy (Sx) Sex (Ab) Abstract (Ax) Anxiety (Hs) Hostility (Br) Barrier (Pn) Penetration OWOZZWXQHHZQ'IIW (P) Popular 2. There will be no significant differences among the mean scores on each of the above Holtzman variables for the male student teacher group and the female student teacher group. 3. There will be no significant differences among the mean scores on each of the above Holtzman variables for student teachers in various major teaching fields. Assumptions 1. It was assumed that personality is an important variable in successful teaching. 2. It was assumed that perception is a function of personality. It was assumed that the group Holtzman Inkblot variables measure perception. It was assumed that the University coordinator's ratings of success on the basis of performance discriminated between student teachers who were more successful than others in their student teaching experience. Limitations As designed and conducted, the framework of this study possessed several recognized limitations. They were: 1. The study was limited in that it was not within the scope of the research to predict probable success as a teacher on—the—job. The study con- cerned the success of the student teacher in student teaching. The study was also limited in that direct personality factors, as such, cannot be pre- dicted without specifying environmental situ- ations. Finally, the study was limited in that no attempt was made to generalize to other popu- lations. Procedures for Collecting Data The data for this study were collected in the following manner: 17 l. The group Holtzman Inkblot Test (HIT) was admin— istered during a regular seminar meeting of the student teachers at each student teaching center. Each University coordinator distributed a letter (Appendix C) to his stu- dent teachers prior to the testing date and requested all the student teachers to be present. Nineteen separate administrations were given in the first weeks of the student teaching experience. In each administration, the test manual's directions to the examiner were closely followed. The screen projection method was used at each administration of the instrument. Unusual or adverse testing conditions were minimized and the administrations took place under normal conditions. 2. After the Holtzman records were collected using a Specially modified answer Sheet (Appendix D), the infor- mation was keypunched and verified by comparing the data card with the original response (Appendix E), loaded onto magnetic tape and computer scored. Computer output in— cluded a listing and a summary card of the Holtzman vari- able scores for each subject in the sample (Appendix E). 3. The rating of success of each student teacher was collected at the end of the term from each University coordinator who supervised any student teacher (Appendix G) and the success rating number entered on the summary card. A. The final summary cards were then submitted for statistical analysis. Description of the Sample The Holtzman Inkblot Test was administered to each student enrolled in student teaching during the Winter Term of 1968. Of the 81A subjects tested, 12 did not receive terminal grades in student teaching and were dropped from the study. Useable Holtzman protocols were obtained from the remaining 802 subjects. These 802 student teachers were fulfilling the requirements of student teaching leading to certifica— tion. Further description of the student teachers who served as the sample in this study is presented in Table 1. The Instrument The Holtzman Inkblot Test is an instrument designed, to study personality through perception. The HIT group form was selected as the instrument for measuring per— ceptual trends because of an available computer scoring system which allows efficient and objective scoring of a large number of test protocols. Computer scored HIT protocols yield scores on perceptual modes described in Appendix A. The group Holtzman Inkblot Test consisted of dis- tributing the HIT RESPONSE SHEET (Appendix D) to each subject, directing the subjects to complete the mom u z Heooe Hmo u z moHHEos HsH u 2 moan AOH H “mHaooe new 3mm omH “mHnooe new AH m eoosaomm sHHnoeoz m mH moHpmeteonz oooce>o< H o esohepmHm HHHeeoHoosm OH 0 onomsonpmz meHeeHmom m o mcHamom do ennm\ento o mH nose HHHApnsecH s H ocsoooEom\poHoaHno m: o mOHEocoom msom oooeme mm mm moHesom HmHoomxsaoomHm H o HHHnHosed\eeHHm m m ooeoHom Hnsoeoo coHpmospm HmHooqm COHpmcHnEoo a m anHwam\noHosom HmHoom m o andahwooo mm NH emHHwem H H eHonH H H monsnd\shooHEono m m cmHnmsm OH 0 AHmHHMDmHoomV mmochsm OH m echadm OH mH aonon H 0 season om a one AH H houses 0 mH ossoHsloma s o COHpooAmoo.£omoam H : COHpmospm o>HpanHpmHQ mH m cooomm mpoonodm szUCOOom COHpMosom AH o HmoHasnd m.HsHo mm H eoHoooshoo nomodm eoHonoaem mm m an» 0 ON HeoHnsnd n.som m: m oIm H H oHnsz Hmoo> as H Hum m a onsz HancosssoncH mHH H mIHIMII mpomnmsm manpcooom mousse depcoEon monEmm monz pHon wcHnomoB monEom monz pHon wCHnomoB .xmm paw pHon wzHcomop an QOHpmHsmoa mHQEMm mo COHuannpmHQII.H mqmp3 00 ms» 00 020030300303 000.0 030.0 00.0 03.0 00.0 00.0 30.0 30.0 03.0 03.0 00.0 00.0 0 030.0 300.0 00.0 00.0 00.0 00.0 33.0 00.0 03.0 00.0 03.0 00.0 20 003.0 000.0 30.0 00.03 30.0 00.0. 03.0 00.0 03.0 00.03 30.0 00.0 00 000.0 003.3, 00.3 00.0 30.3 00.0 00.3 00.0 00.0 00.0 03.0 00.0 00 000.0 000.3 00.0 03.0 33.0 03.0 30.3 30.0 30.3 30.0 00.0 30.0 00 033.0 000.3 00.0 30.0 03.3 33.3 00.3 00.3 00.3 30.3 00.3 00.3 00 000.0 003.0 00.3 00.0 03.3 30.0 30.3 00.0 00.3 00.0 00.0 30.0 00 000.0 303.0 00.0 00.0 00.0 00.0 03.0 00.0 00.0 00.0 00.0 30.0 00 300.0 000.0 00.0 00.00 00.0 00.00 03.0 00.00 00.0 00.00 00.0 03.00 a 000.0 030.3 00.0 00.00 00.0 30.00 00.0 00.00 00.0 00.00 03.0 00.00 0 000 0 330.3 00.0 30.0 00.3 00.0 30.3 00.0 00.3 00.3 00.3 00.3 3 030.0 300.0 00.33 03.00 00.0 00.00 00.33 00.00 00.33 00.00 03.33 03.00 2 000.0 030.0 00.0 00.3 33.0 00.0 00.0 03.3 00.0 33.3 00.0 00.3 00 000.0 000.0 00.3 30.33 00.0 03.03 00.0 00.03 33.0 30.03 00.0 03.03 0 000.0 000.0 00.03 00.00 30.0 03.00 00.03 00.00 33.03 30.00 00.0 00.00 00 *300.0 000.3 30.0 00.0 30.3 03.3 00.0 00.3 30.3 00.0 00.3 00.3 0 000.0 300.0 00.03 00.00 00.03 00.30 00.03 00.30 03.33 00.00 30.33 00.30 3 .000000 0000 .0>0 303mm m0< 0000 02302000000 "020000 .mmmoosm copay op moapmooom 0030000030 mazopw monommp powwowm wcoEm moanmasm> poaost swampaom mo moc00000c030 oudSonmoom mua 0cm 03uwfip0pm m «m203p00>00 opmoc0um .mcmmzll.03 03009 (m L) Since the R score was the only probable significant finding, the null hypothesis was rejected for the R variable but may not be rejected for the remaining six- teen variables. Hypothesis Two The second hypothesis was that there will be no significant differences among each of the mean Holtzman variable scores for the male student teacher group and the female student teacher group. Table 11 presents the mean scores, standard deviations, and F statistic and its approximate significance, of student teachers classi— fied according to sex. An examination of Table 11 shows that the mean Holtzman scores on nine variables significantly differed between the sexes. Males and females scored significantly different on the R (rejection), FD (form definiteness), M (movement), 1 (integration), A (animal), At (anatomy), H (human), Sx (sex) and P (popular) variables. As previously indicated, the R (rejection) score was the total number of inkblots to which the subject did not give a scorable response. From the data in Table (1‘ 11, i is evident that males gave a significantly higher number of responses classified as rejections than did females, subject, of course, to the assumption of equal variances. 91 TABLE 11.--Means, standard deviations, F statistic and its approximate significance of Holtzman Inkblot variable scores among student teachers grouped according to sex. MALES FEMALES F Approx. variable N = 171 N = 631 Statistic Signifi- Mean S.D. Mean S.D. cance L 32.05 13.01 31.45 13.18 0.281 0.603 R 1.71 2.54 0.81 1.93 25.310 0.0005* FD 78.46 13.49 83.54 11.36 24.758 0.0005* c 9.73 5.50 10.62 5.80 3.228 0.069 Sh 4.38 2.67 4.21 2.49 0.546 0.466 M 25.78 10.53 30.67 11.64 24.627 0.0005* 1 4.36 1.91 5.19 1.91 25.010 0.0005* H 26.02 6.85 27.74 6.46 ' 9.298 0.003* A 26.30 6.11 28.82 5.89 24.130 0.0005* At 3.22 2.95 2.41 2.28 14.531 0.0005* Sx 1.01 1.30 0.62 0.94 17.997 0.0005* Ab 1.49 1.66 1.43 1.56 0.146 0.702 Ax 8.38 4.16 8.19 4.19 0.263 0.614 Hs 8.73 4.07 8.83 4.25 0.082 0.767 Br 9.59 3.69 10.11 3.41 2.942 0.083 Pn 3.56 2.26 3.33 2.08 1.593 0.204 P 7.42 2.48 8.94 2.40 52.614 0.0005* *Significant at the 5% level. \O [U The sexes also differed significantly on FD (form definiteness) scores which were responses that were definite in form, regardless of the goodness of fit to the inkblot. For example, responses such as "splattered paint" or "a cloud" were concepts that have an indefinite or non-specific form and received a lower Form Definite- ness score. A highly structured concept such as "a boy riding a bicycle" received a higher FD score. Form Definiteness was scored on a five point scale, ranging from zero (concepts that were completely formless or lacking in specificity) to four (concepts that were highly definite in form). From the data presented in Table 11 it is evident that females, as a group, scored significantly higher than males on the Form Definiteness variable. Females also scored significantly different from males on the M (movement) score. The Movement variable reflects the degree of movement, tension or dynamic energy projected into the inkblot. The scoring system for Movement is a five point scale as follows: 0 — no movement, or static potential for movement. 1 - static potential for movement as indicated by such participles as sitting, looking, resting, lying. 2 - casual movement, such as walking, talking, climbing, reaching. 93 3 - dynamic movement such as lifting, dancing, running, weeping. 4 - violent movement such as whirling, exploding.l Responses such as "two men reclining" and "a dog gazing into the sky" received Movement scores of 1; responses such as "whirling dervishes" or "an atomic bomb blast" received Movement scores of 4. Since females scored significantly higher (mean of 30.67) than males (mean of 25.78) one may conclude that females characteristically projected more responses that received Movement scores than males. Besides scoring differently on R, FD and M, the sexes also differed significantly on the I (integration) scores. Holtzman wrote that, The organization of two or more adequately per— ceived blot elements into a larger whole is necessary for a score of 1 on Integration. The particular elements that are integrated may in— volve either identical or non—identical contents and the resulting concept need not utilize the whole inkblot.2 Although Phillips, Kaden and Waldman3 distinguished among several types of Integration (i.e., functional, collective, positional, structural) credit for Integration lHoltzman, et al., Inkblot Perception and Per— sonality, p. 51. 2Ibid., pp. 64-65 3L. Phillips, 5. Kaden, and M. Waldman, "Rorschach Indices of Developmental Level," Journal of Genetic Psychology, LXXXXIV (1959), 267—285. 94 was either given (0) or not given (1), thus permitting a theoretical range of 0 to 45 across the forty-five ink- blots. Responses such as "roosters fighting," "a collec- tion of shells," "a candle stuck in the neck of a bottle" and "a bee standing up" received a score of 1. Females had a significantly higher mean score (5.19) than males (4.36) so it is evident that females characteristically gave more integrated responses than males. Table 11 also shows that the sexes differed signifi- cantly in the content scores of H (humanx A (animal), At (anatomy), and Sx (sex). For each of these content areas a three point weighting system permitted the scaling of content as follows: 0 — No reference to humans, animals, penetration, or sex. 1 - References to parts of humans, animals, X rays, medical drawings and socially accepted sexual activity and expressions (thighs, bust, pregnant woman, kissing, embracing, etc.). 2 - References to whole humans, animals, visceral and crude anatomy responses, blatant sex (penis, vagina, anus, breast, etc.). From an analysis of the content scores presented in Table 11, it is evident that females gave significantly more responses with H (human) and A (animal) content than did males. On the other hand, males scored significantly ”Holtzman, et al., Inkblot Perception and Per- sonality, p. 68. \O \J 1 higher than females in At (anatomy) and Sx (sex)con— tent. Males also scored significantly different from females on the P (pOpular) variable. A Popular response is one which occurred frequently enough in the original HIT normative samples to be clasSified as popular among normal subjects. For a response to be judged Popular, it must have occurred in at least one out of seven pro- tocols in the original normative protocols. This arbi- trary cutting point was chosen, Holtzman reported,5 be— cause it provided a sufficient number of inkblots with Populars to yield a variable with good psychometric quali— ties. Several inkblots so closely resemble human or, animal form, for example, that most people perceive these concepts. In other words, the inkblot has a "pull" for a certain content response; responding like most people do yields a score of 1 on the Popular variable. In the group Holtzman Test, there are twenty-five inkblots scorable for Popular resulting in a theoretical POpular range of 0 to 25. Table 11 shows that females characteristically per- ceived significantly more POpulars than males. In sum, males and females differed significantly on nine of the seventeen Holtzman variable scores: R (rejec- tion), FD (form definiteness), M (movement), 1 (integration), 5Ibid., p. 80. 96 A (animal), At (anatomy), H (human), Sx (sex) and P (pOpular). Thus, the null hypothesis was rejected for these variables but was not rejected for the remaining eight variables of L (location), 0 (color), Sh (shading), Ab (abstract), Ax (anxiety), Hs (hostility), Br (barrier) and Pn (penetration). Since there were significant variable score dif— ferences between the sexes, it was logical for the in- vestigator to undertake a re-examination of the first hypothesis to determine if initial grouping by sex and. then an analysis of mean scores related to success would alter the failure to reject the first null hypothesis. In other words, Hypothesis One stated: There will be no significant differences among each of the mean Holtzman variable scores of student teachers classified according to rated success, ‘ and was modified to: (1,A) There will be no significant differences among each of the mean Holtzman variable scores of MALE student teachers classified according to rated success, and, (1,B) There will be no Significant differences among each of the mean Holtzman variable scores of FEMALE student teachers classi- fied according to rated success. Table 12 presents the means, standard deviations, and F statistic and its approximate significance, of male student teachers grouped according to rated success. An examination of Table 12 shows that there are no statis- tically significant differences among sixteen of seventeen .30>03 00 000 00 00003030030* 000.0 000.0 I- nu 00.0 00.0 03.0 00.0 03.0 00.0 00.0 00.0 0 300.0 000.0 .. s. 00.0 00.0 00.0 00.0 00.0 03.0 00.0 03.0 :0 000.0 300.0 .. I- 00.0 00.0 00.0 00.0 30.0 00.03 00.0 00.0 00 300.0 000.0 u. n. 03.3 30.0 00.3 00.0 00.0 33.0 00.0 00.0 00 000.0 000.3 .. n. 03.0 30.0 03.3 00.0 33.3 33.0 00.0 03.0 30 000.0 000.3 .. n. 30.3 03.3 03.3 03.3 30.0 00.3 00.0 00.0 00 300.0 000.0 .. n: 00.3 .30.0 00.3 30.3 00.3 03.3 00.3 00.0 00 000.0 000.0 nu .: 00.3 03.0 03.0 00.0 00.3 33.0 00.0 03.0 00 030.0 000.0 I- 1. 00.0 00.00 30.0 03.00 30.3 00.00 00.0 03.30 0 030.0 300.0 .. .. 00.0 03.00 30.0 00.00 00.0 00.00 33.0 00.30 2 000.0 000.3 .. n. 00.3 00.3 00.3 00.3 30.3 00.3 03.0 00.0 3 000.0 000.0 I- I- 00.0 03.00 30.03 30.00 00.0 00.00 00.33 00.00 2 000.0 000.0 I. I- 00.0 00.0 00.0 00.3 00.0 00.3 30.0 00.0 20 033.0 000.0 .. I- 00.0- 00.0 03.0 00.03 00.3 00.0 00.3 00.0 0 300.0 300.0 I- I- 00.0 00.00 00.33 00.00 00.03 00.00 00.03 00.00 00 *000.0 030.3 I. I. 00.0 30.3 03.0 00.3 30.3 00.3 00.3 00.3 0 000.0 000.3 I- s- 03.0 30.00 00.03 00.00 30.03 00.00 00.03 00.00 0 -3mmmmM0 e3pmwpms0 .0.w ungz .0ww ugMpz .0mm ucM0z .0mm usM02 .0mm usmmz 0300300> .300000 0000 .0>< 20300 00000>< 0000 02302000300 ”003000 .0000050 00003 Op 003000000 0030300000 mgzomw 0050000 3:00:00 0302 0:020 003000 0300000> poabst c0200303 mo 002000000000.0p0swxo3000 000 0G0 000000000 0 «00o0000>00 03002000 .0s00zal.m3 mqmo3 00 000 00 000030300303 300.0 000.0 03.0 03.0 00.0 33.0 33.0 00.0 03.0 33.0 30.0 00.0 0 000.0 003.3 00.0 00.0 30.0 00.0 30.0 00.0 00.0 30.0 30.0 03.0 00 300.0 030.0 30.0 00.03 30.3 00.0 00.0 00.0 00.0, 03.03 03.0 03.0 00 003.0 000.3 00.3 00.0 00.0 00.3 00.3 33.0 00.0 00.0 30.0 30.0 00 033.0 300.3 00.0 03.0 03.0 00.0 00.3 03.0 00.0 00.0 30.3 00.0 30 000.0 033.3 00.0 30.0 00.3 00.3 00.3 00.3 30.3 00.3 30.3 00.3 00 330.0 300.0 03.3 00.0 03.3 30.0 30.0 00.0 03.0 00.0 03.0 00.0 00 003.0 303.0 00.0 00.0 03.0 00.0 30.0 00.0 03.0 00.0 00.0 00.0 00 333.0 000.0 03.0 00.30 00.0 00.00 03.0 00.00 00.0 00.00 33.0. 00.30 0 003.0 300.0 30.0 03.00 00.3: 33.00 00.0 00.30 00.0 00.30 00.0 00.00 0 030.0 000.0 00.0 30.0 00.3 33.0 30.3 30.0 00.3 03.0 00.3 30.3 3 003.0 330.0 00.33 03.00 30.0 00.00 03.03 00.00 03.33 00.30 00.33 00.00 0 003.0 00000 00.0 30.3 03.0 00.0 00.0 33.3 33.0 03.3 03.0 03.3 00 300.0 000.0 00.3 30.33 30.0 00.03 30.0 30.03 00.0 33.03 30.0 30.33 0 030.0 303.0 00.33 00.00 03.0 33.30 00.33 03.00 00.03 00.00 00.0 00.30 00 3030.0 000.0 30.0 30.0 03.3 00.0 03.3 03.0 33.3 00.0 30.0 00.3 0 000.0 000.0 00.03 00.00 00.33 30.30 00.03 30.30 00.03 03.30 00.33 03.30 0 -000. 2.0.... .220 2.2 .20.. .2.2 .me .202 .230 .20 .202 2...... .xo0000 0000 .0>< 20000 00000>< 0000 02302000000 "003000 .0000000 00003 00 003030000 0030300030 000030 3000000 0000000 030800 00050 003000 0300330> 0030003 00000300 00 000003030030 00083303000 003 000 030030000 0 .0003003>00 03000000 .0000za|.m3 03000 100 Inkblot variable scores of student teachers grouped according to major teaching fields or grade levels. To adequately test this hypothesis, the thirty- six subject/grade classifications presented in Table 1“ were collapsed into thirteen major subject/grade areas since in several of the thirty-six teaching fields there were few subjects. Table 13 illustrates the combined sample size for the major teaching areas. For example, student teachers in French (N=18), German (N=l), Spanish (N=l2), Russian (N=H) and Latin (N=2) were grouped under the major teaching area of language (N=37); student teachers in biology (N=25), chemistry and physics (N=S), beginning and advanced mathematics (N=38), and general science (N=5) were grouped in the major teaching area of science and mathematics (N=72). Junior high school stu- dent teachers were not included in the test of this hypo- thesis since student teachers at this grade level may be in any teaching area and as such would tend to confuse the data. Table 15 presents the means, standard deviations, and F statistic and its approximate significance, of the Holtzman variable scores of student teachers grouped according to major teaching field or grade level. An examination of Table 15 shows that the mean Holtzman vari- able scores on eleven variables differed significantly among the major teaching areas. That is, student teachers grouped according to major teaching field scored 101. TABLE lu.--The student teacher population (excluding junior high school student teachers) classified according to major teaching field and sex. Number Resultant Specific of Student Teachers Major Sample Size Teaching Field Teaching Field Male Female Total Male Female Total Elementary Grades: K,1,2, 1 112 113 Elementary 5 230 235 Grades: 3,h l 70 71 Education Grades: 5,6 3 MB 51 Physical Educ. 2O 17 37 Physical Educ. 2O 17 37 French 1 17 18 Languages 6 31 37 German 0 l ‘ 1 Spanish 2 10 12 Russian 2 2 u Latin 1 l 2 English 12 59 71 English 15 72 87 Speech l3 13 16 Biology 15 10 25 Science MO 32 72 Chem./Physics “ l 5 and Math Beginning Math 6 10 16 Advanced Math 13 8 21 General Science 2 3 5 Distrib. Educ. U l 5 Business A ll 15 Business (Sec'y) O 10 10 Education Home Economics 0 U5 US Home Economics 0 MS U5 Industrial Arts 13 O 13 Industrial Arts 13 O 13 Instrumental Mus. 7 2 9 Music 8 3 ll Vocal Music 1 1 2 Education Agriculture 15 O 15 Agriculture 15 O 15 Soc. Studies/ 3 7 10 History/ 28 61 89 Eng. Combined Social Studies History/Soc.St. 25 52 77 Geography 0 2 2 Art 6 3O 36 Art 6 30 36 Special Educ. ' Special 5 6H 69 Sph. Correct'n . Education Elementary l 23 2h Secondary O 7 7 Visually Handi- O u u capped ‘ Homebound 1 7 8 Deaf O 2 2 Emotionally Dis. 0 U u Mentally H'cap 3 17 20 TOTALS 165 596 761 H O\ \J’l \J1 ‘C’ C x \J C\ H TOTALS ”2 g I .59 ate 3 w ‘a-s CX Tu .3 .3 1; 111 or .c- no. r. O. a .... P 3 9.. f. E n. ”.1 rm .3 E 7.. e. u p. C 1. C U“ :1 . .2. a; 11 1.. n. a. .1. r“ P. 1.1%“ r“ 1. G. 1.; :1 ”1 A. nynll . ~13 .11 C :1 1.1“ . 1 1: :1. D. r“ 1.: 1... 111 . 1 .1 a; a. “a .1 .r“ A. T. .1” I I 11 2 a. “,1 _ 1 44 1.1 111 :2 h .1. n. .r“ C a. I .. 1.3 r: a. ".11 . 9 a... 2.11., «4,—qu a; . . u1; c.1r. 11L .1. . 7- .C rt 711‘ It... 1 .1 .11. —. a I I 1.14. .1 1. v1.. —.,.,r I I 1 1_ a .1 11.... 1.1.. I I 1.. a r— i. 5.2.4.. ‘ .zr. . I 1.151 1. 7..i FUR—v I I zr» «<1 1.. a- 1” \a. I I w 1.1 1‘ 1» I I 1. 11 11 11‘. .1 1. . I I 1.. 1 ,. 4 1.... I I .. 1 .' 1... .... I I ..1,. ,1... 1,. 1.. I I 1. ,1 1.“ . 11.11 e . v.1. 1. A. a, .C 1.9 «.1 I I I a... «1L «1 14. .._. . 11.1 1 I I I 1. _..1.1 .1 ..1 1.. 1 1. I I I 1.. _. 1 .. 1v q. . 1. 1. . I I I _. 111. .. . .11 1 .1. I I I 1. 1.... .1 .. 1 _. I I I 1. .111 1. 1. .1 1 1.1 . s o s .1 11,11 .. 1.1 ..,» A.“ /.1_. I I I .. 1.. .1 o r“ o 1,. 1.111 a L. o 1.. ”.1. . .. l 1 .. 1.... 7,. u. 7. 1C ,C 1‘. .u Q...:. 71 14. I 0 I I I I I 1.... I. 3,. «U «U l .i 1 a 1;. 2.1.5. «.32-. 1.. 1,11... 1111“. Li: I I I I I I I i ., ._ 7.1 1. .1 1L .1 1.1 1. 1.." 3.1.1 1.1-.. . .. I. 1. 1 I. o I a I O o o I, .111 a... 1 1.11. .. '1... 1.1., 1.4111 1. u. .. a. .c ‘ ..1 l I I I I I I n .11., .1.1 .171 _1 .1. 1 . T ,J . . 1 1 _. 1.11. I I I I I I I .1 11... ._. 111. . 1.1 1. u _ 1 n14 1 1.11 .r .. 1.. .. I I I I I I I 1 . 1.1111 1 .1. 11.»; 1. .vp. 11.7. uh F.. H u r 1. . H.217. I I I I I I I 1.. $11.1 1.1.. .1111 1y. 1 n u 1.1. ~ ‘ 241 1... 1.1111 1.11., 1..l1 I I I I I I I 1. 1,11; ”.1. 11.»; . r“ . C . .1.” . .1 S11. '11.. 911 o G. o a. o a. o .I. :2 fr. /0 I I /O 5 «NJ 31 c. KC WI 3.1-. «(J .u T¢1<_ C» u .51.. a: 11 1.-. u 1.1.1.. .M «1 ~,.. ,H 1.11.. 7.1,». Hi ya I. 1... a... 1... 7. -1 1,1111 AC a--. Cr. .4 a 1. e . u-“ 3 X fr. my C) Ow 5 .d R. W. 1.) r: NU an. 11.. I Q. 14. 7 3 AL. .19 I I Qv u 1... fr. 7. A 1.1:. 1....1... 1.. HI, I I 1. v 14‘. 1-. 1J1 1|. 11C 1.4. by 12) 1; 51./~I- C1 T... C11... n . a 1 . v.“ S w. 3“ 7 .CO 3314 «O O 2 3 NH ‘1. I.“ «<4 ./ 1.4 J 14.711 mean C-l I ‘4 I 16 7rd; CI 7,. /O A). 6 C], 8 dd “4). hug 1... n V Cr. 1/1 1H. l r. . rh. » C 3.. Au 9.. C , 1... I r5 7L 1l33 Music Agri- Social Special Educ. culture Sciences Art Educ. Statistic ggpgigigzfige N=ll N=15 N=89 N=36 N=69 8 30.u5 u0.06 32.98 22.97 27.31 3.8u3 0.0005* 12.21 10.22 12.52 13.89 12.05 1.36 1.66 1.89 0.77 0.81 3.101 0.0005* 2.01 2 2 3.99 1.u7 2. 2 81.5u 77.93 32.50 83.75 85.00 3.080 0.0005* 12. 2 10.93 13.60 2.51 10.75 8.28 9.80 9.56 11.88 10.89 1.015 0.u32 u.66 6 51 5.98 ' 6 75 5.69 .63 .60 3.82 3.27 u.63 2.086 0.016* 2 37 2.55 2.28 2.07 2.53 27.63 19.u0 28.89 35.72 36.13 6.160 0.0005* 8.78 7.35 12.09 13.83 11.10 u.27 3.96 5.39 5.61 5.65 3.3uu 0.0005* 1.92 1.80 2.09 2.32 1.58 26.09 21.80 28 55 30.02 27.60 2.u51 0.009* 5.91 7.0? 6 92 5.93 6.08 28 00 29 06 26.82 30.88 28.72 H.599 0.0005* 5 63 7 07 5.80 5.17 6.13 2.95 2 80 2 to 2.55 2.00 1.6u9 0.07u 2.5M 2 90 2 0 3.13 1.58 0.27 0 Mo 0.61 1.08 0.68 2.629 0.002* 0.06 0 63 0.89 1 78 0.86 2 7 0.93 1.53 1.66 1.65 1.309 0.211 2 2a 0.79 1.63 1.60 1.78 8.63 7.13 7.83 8.91 9.07 1.398 0.161 9.78 3.33 3.75 3.99 u.51 8.72 8.00 9.11 9.33 10.0u 1.753 0.052 3.52 3.52 u.11 u.57 9.88 10.5u 9.20 10.20 9.50 10.60 2.509 0.003* 3.36 3.62 3.78 3.90 3.29 2.63 3.06 3.05 3.72 3.60 0.791 0.711 1.68 1.98 2.0M 2.26 1.93 6.27 7.06 8.39 9.66 9.05 3.565 0.0005* 2.10 2.31 2.33 2.26 2.uu 104 significantly different on the L (location), B (rejection), FD (form definiteness), Sh (shading), M (movement), I (integration), H (human), A (animal), Sx (sex), Br (barrier), and P (popular) variables. Several of these variables (R, FD, M, I, H, A, Sx and P) were also significantly different between the male and female stu- dent teacher groups and were discussed on pages 88 through 97. The remaining variable scores (L, Sh and Br) although not significantly different between the sexes, were significantly different among student teachers grouped according to major teaching field. The L (location) variable score represented the tendency to break down the inkblot perception into smaller fragments. For example, a subject received an L score of zero each time all or most of the inkblot was used in the perception. If, on the other hand, the perception was a large area (an entire side, for example) but something less than the entire blot, the L score for that perception was one. If an area less than one—half of the inkblot was used, the L score was two. Thus, the total L score was computed by summing the individual L scores across the forty-five inkblots yielding a theoretical L range of O to 90. The Sh (shading) Holtzman variable score repre— sented the apparent primacy or importance of shading in the perception and was rated on a three point scale as follows: 135 O - shading not used as a determinant l - shading used only in a secondary manner as an elaboration of the percept reported (similar to Fe, FK, Fy in the Rorschach) 2 - Shading used as a primary determinant with no form present (similar to cF, c, K, YF, Y in the Rorschach) A response such as "mist or fog" uses shading as a primary determinant and, as such, received a Sh score of two. Thus, the theoretical scoring range of the Sh vari- able is O to 90. The concept of the Holtzman Br (barrier) variable referred to any protective covering, membrane, shell or skin that might be symbolically related to the perception of body—image boundaries. A score of one was given to each response where Barrier was present, a score of zero was given when Barrier was absent. Responses such as "woman in a high—necked dress" or references to things that are armored or much dependent on their own contain- ing walls for protection such as a "tank" or "man in armor" received a Barrier score of one. References to things being covered, surrounded or concealed also re- ceived a score of one. Since student teachers grouped according to major teaching field scored significantly different on L, R, FD, Sh, M, I, H, A, Sx, Br, and P, the conclusion is that null hypothesis III was rejected for these vari- ables, but was not rejected for the remaining variables 106 of C, At, Ab, Ax, Hs and P. Since, however, there were significant variable score differences between the sexes (hypothesis II, pages 96—97), the investigator undertook an examination of hypothesis III to determine if initial grouping by sex and then classification into major teach- ing fields would alter the findings for hypothesis III. In other words, hypothesis III stated: there will be no significant differences among each of the mean Holtzman Inkblot variable scores of student teachers grouped according to major teaching fields or grade levels and was modified to: (III,A) There will be no significant differences among each of the mean Holtzman Inkblot variable scores of MALE student teachers grouped according to major teaching field or grade levels, and, (III,B) There will be no significant differences among each of the mean Holtzman Inkblot variable scores of FEMALE student teachers grouped according to major teaching fields or grade levels. Hypothesis III-A Table 16 presents the means, standard deviations, and F statistic and its approximate significance, of male student teachers grouped according to major teaching field or grade level. An examination of Table 16 shows that there are no statistically significant differences among sixteen of the seventeen mean variable scores. At (anatomy) was the only variable that did reach statistical significance at the five per cent level. .107 TABLE l6.--Means, standard deviations, F statistic and its approximate significance of Holtzman variable scores of male student teachers classified according to major teaching field. Major Elem. Phys. Lang— Eng- Science Busi- Home Industrial Teaching Educ. Educ. uages lish & Math ness Econ. Arts Field N=5 N=20 N=6 N=15 N=90 N=9 N=0 N=13 Variable L Mean 33.20 37.70 21.16 32.96 29.55 28.75 29.89 S.D. 17.71 11.97 16.85 13.53 11.89 16.25 15.52 R Mean 1.60 2.2 2.00 0.80 1.75 3.00 1.89 8.0. 1.99 3.97 3.09 1.19 2.33 3.55 2.67 FD Mean 82.80 79.95 90.16 81.20 81.27 78.25 67.69 S.D. l3. 2 11.00 10.68 12.21 19.19 16.95 11.67 C Mean 8.20 9.95 8.00 10.3 9.95 12.25 11.15 S.D. 2.77 5.39 7.37 5.08 6.13 9 99 5.69 Sh Mean 9.20 3.80 9.66 9.73 3.90 5.00 6.07 S.D. 3.39 2.98 2.58 2.60 2.52 1.91 3.81 M Mean 23.90 23 70 31.16 28.66 26.67 29.25 21.76 S.D. 10.66 1 03 8.90 10.95 10.11 11.02 13.00 I Mean 3.80 3.60 5.16 9.6L 9.55 3.75 9.07 S.D. 1.'r 1 72 1.1“ 1 Q0 2 ‘2 1.50 2.96 H Mean 29.20 23.95 27 90 2/ 66 25.97 29 0 25.07 S.D. 9.59 8.09 r 17 ‘ L5 7.1“ » 2‘ 9.10 A Mean 28.00 “V 7 “7 up 29 33 .r 90 36 50 29.07 S.D. 5.83 55 5 ‘0 6 66 f 07 8 58 3.61 At Mean 7.80 9.15 9 83 2 r3 3 30 1. 5 3 30 8.0. 10.75 2.51 3 06 1 89 2 5 1.2r 2.92 Sx Mean 2.90 1.20 0.83 1 00 1 55 1 00 0.92 S.D. 9.27 1. 8 0.75 1 30 1 ,6 1 91 0.69 Ab Mean 1.60 1.25 1.83 1.80 1.37 2.50 1.23 S.D. 1.19 1.55 2.22 1.82 1 2 9.35 1 92 Ax Mean 9.60 7.95 8.33 10.06 8.92 9.00 8.61 S.D. 5.81 9.79 5.71 9.68 3 01 3.36 5.65 Hs Mean 6.90 7.75 8.50 10.90 9 10 50 8.00 S.D. 2.70 3.7 5.16 3.13 9 65 3.91 5.52 Br Mean 7.60 9.95 10.83 11.13 9.32 9.75 8.07 S.D. 1.67 3.18 9.62 2.29 3.89 1.70 9.90 P Mean 2.90 3.00 9.16 9.06 9.07 5.25 3.76 “ S.D. 1.39 2.55 9.07 1.79 2.91 2.21 3.00 P Mean 9.00 6.95 8.83 7.20 7.29 7 0 6.53 S.D. 2.73 2.35 1.99 1.89 2.86 9 2.29 *Significant at the 5% level. 108 Music Arri- Social Special 9 . F Approximate Educ. culture Sciences Art Educ. N=8 N=15 N=28 N=6 ”=5 Statistic Significance 28.00 90.06 32.78 31.50 23.20 1.893 0.051 13.58 10.22 10.39 16.37 6.26 1 5 1.66 1.85 1.83 1.20 0.399 0.979 2 25 2.12 3.39 1.83 0.83 78. 2 77.93 78.78 79.50 83.90 1.777 0.062 13.35 10 93 13 26 17.73 2.17 8.12 80 9 00 13.66 8 .0 0.636 0.796 5.11 L 51 5 92 9.89 3 36 9.00 9 60 9.19 3 33 9.80 0.865 0.576 2.61 2 55 2 39 2 06 3.56 29.12 19.90 25.75 31.66 33.00 1.522 0.128 9.90 7.35 9.30 15.29 11 76 9.50 3 96 f 10 9 Q3 9 20 1.301 0 229 1.19 1 £0 1 79 2 an 1 09 27.62 21.80 21.17 28.00 25.80 1.557 0.117 3.58 7 08 5.67 6 8 9 3 25 07 9.06 26 17 3 .66 2t.80 0.813 0.627 9 L 7 07 5 91 9.63 7 52 2 50 2.80 2.96 1.50 1.60 2.107 0.023* 2 82 2 90 1.81 1 22 1.19 0.37 0.90 0.60 0.50 0.80 1.806 0.057 0.51 0.63 0.56 0.59 1.30 2.50 0.93 1.71 2.00 1.2 0.799 0.695 2.56 0 79 1.78 1.26 0 E3 8.\2 7.13 7.53 7.16 9.80 0.796 0.692 9.97 3.33 3.19 3.12 2.09 9.50 8.00 8.39 8.50 10.90 0.808 0.632 3.66 3.52 3.39 9.89 2.90 10.37. 9.20 9.89 9.00 11.00 0.780 0.659 3.88 3.62 9.62 .52 1 58 3.12 3.06 3.00 3.33 3.90 0.968 0.977 1.69 1.98 1.98 1.63 0.59 6.50 7.06 8.21 8.16 7.20 1.015 0.936 1.85 2.31 2.72 2.31 2.86 109 The At variable is a content variable previously discussed since this variable score was also significantly different between the sexes (page 90). But this was the only significant finding and was not supported by signifi- cant differences among the other variables. Thus, null hypothesis III,A may not be rejected for the remaining sixteen variables and the conclusion is that mean Holtzman variable scores did not differ significantly among male student teachers grouped according to major teaching fields for sixteen of the seventeen variables. Hypothesis III-B Table 17 presents the means, standard deviations, and F statistic and its approximate significance, of female student teachers grouped according to major teach- ing field or grade level. An examination of Table 17 shows that there are ten variables, L (location), R (rejection), Sh (shading), M (movement), I (integration), A (animal), Sx (sex), Hs (hostility), Br (barrier), and P (popular) that are significantly different among the female student teachers grouped according to major teach- ing fields or grade levels. Each of these variables has been discussed previously in other hypotheses (L, loca- tion, page 109; R, rejection, page 88; Sh, shading, page 109; M, movement, page 92; I, integration, page 93; A, animal, page 99; Sx, sex, page 96; and Br, barrier, page 105) except for the Hs or Hostility variable. .110 TABLE 17.--Means, standard deviations, F statistic and its approximate significance of Holtzman variable scores of female student teachers classified according to major teaching field. Major Elem. Phys. Lang— Eng- Science Busi- Home Industrial Teaching Educ. Educ. uages lish & Math ness Econ. "Arts Field N=23 N=17 N=31 N=72 N=32 N=11 N=95 N=0 Variable L Mean 32.72 35.00 30.12 30.98 36.12 25.27 33.33 S.D. 12.60 11.01 15.02 12.87 13.10 19.05 19.55 R Mean 0.65 0.97 1.16 0.75 1 21 0.59 0.28 8.0. 1.57 0.62 1.71 1.18 2 93 1.03 0.69 FD Mean 83.60 83.91 82.22 89.27 89.15 76.09 82.69 S.D. 10.07 12.95 19.22 10.07 13.80 9.71 13.87 C Mean 10.72 9.58 11.51 9.97 11.03 11.72 11.37 S.D. 5.56 5.63 7.65 5.22 6.55 7.21 5.00 Sh Mean 9.28 9.58 9.32 3.83 9.00 36 9.97 S.D. 2.37 1 69 1.92 2 33 2.93 2 90 2.97 M Mean 30.63 28.52 23.51 30.22 26.06 32.00 32 2 ‘ S.D. 11.31 11.77 9.23 10.15 13.20 13.60 10.09 I Mean 9 93 99 9 83 5 22 5.13 5 36 5 11 s D 1 80 3 16 1 98 1.88 2.03 3 15 2.20 H Mean 26.92 29 (9 26 90 28.97 27 2, 27.90 28.80 S.D. 6.21 7.91 6 (9 6.73 7.r 6 09 6.73 A Mean 30.17 27.97 27 03 ‘7 2( 28.56 26 95 2t 60 S.D. 5.56 6. 8 5 21 5 10 6.83 9 13 6 59 At Mean 2.97 3 3 to 2.27 1 8 2 00 2 S.D. 2.09 2 31 93 1 7 1 90 2 "2 2 3Q Sx Mean 0.50 82 0.58 0.66 0 56 0.36 0.55 S.D. 0.70 1 00 0.67 0.76 0 80 0.67 0.96 Ab Mean 1.30 1.88 1.29 1.50 0 96 1 59 1 03 S.D. 1.50 1.79 1.39 1.59 1 09 1 21 1 88 Ax Mean 8.19 6.88 7.95 9.30 6.39 7.81 7.80 S.D. 9.12 3.37 9.79 9.03 3.98 9 75 3.62 H, Mean 8.65 6.91 '7.61 9.99 7.53 7.18 9.09 ° S.D. 3.95 2.98 3.79 9.19 9.63 3 68 9.56 Br Mean 9.59 10.70 10.83 11.00 8.89 11.36 10.75 S.D. 3.38 3.51 3.63 2 96 2 86 9 03 3.83 Pn Mean 3.29 3.17 3.91 3.55 2.65 2.90 3.28 S.D. 2.09 2.29 1.96 2.03 1.99 1.97 2.30 P Mean 8.80 9.91 8.00 9.29 9.28 9.09 8.71 8.0. 2.38 2.31 2.11 2.98 2.55 2.77 2.56 *Significant at the 5% level. lll. Music Agri— Social Special Educ. culture Sciences Art Educ. Statistic gppri§imate N=3 N=0 N=61 N=3O N=69 3“ cance 37.00 33.08 20.66 27.69 3.937 0.0005* 3 96 13 99 12.90 12.36 0.33 1.83 0.56 0.78 2.569 0.005* 0.57 3.59 1.33 2.19 89.33 89.21 89.60 85.12 0.809 0.625 7.76 13.52 11.91 10.73 8.33 9.81 11 53 11.00 0.799 0.639 9.16 6.19 7 08 5.81 2.66 3.67 26 9. 2 1.899 0.050” 1.52 2.29 2 11 2.97 23.66 3 39 36 53 36.37 9.907 0.0005* 3.21 13 92 13 16 11.11 3.66 52 5.76 5.76 1.877 0.096* 2.08 2 17 2.19 1.57 22.00 28.26 30 93 27.75 1.590 0.106 8.18 6.76 5 16 6.15 33.66 27.11 30 (0 28.8 3.363 0.0005* 2.51 5.91 5 39 6.05 2.33 2.92 2.76 2.03 1.005 0.937 2.08 2.35 3.37 1.72 0.00 0.62 1.2 0.67 2.078 0.029* 0.00 0.95 1.88 0 83 1.66 1.95 1.60 1.68 1.259 0.250 1.15 1.57 1.'7 1.83 8.66 7.96 8 66 9.01 1.779 0.062 6.65 9.02 9 8 9.65 6.66 9.99 9 50 10.01 2.269 0.013* 2.51 9.91 9.58 5 03 11.00 10.39 9.60 10.57 2.196 0.017* 0.00 3.36 9.03 3.39 1.33 3.08 3.80 I 3.62 1.137 0.331 1.15 2.09 2.38 2.00 5.66 8.97 9.96 9.20 2.329 0.011* 3.05 2.19 2.17 2.37 112 The Hostility variable score was based on a four point scale which considered symbolic, implicit or ex- plicit signs of hostility in the response. The more ex- explicit the hostility, the higher the Hs score, with a range of zero for no hostility to three for direct hos- tility among humans or animals seen in violent destructive action toward each other with elaboration of gore, in- juries, blood, death, etc. Since female student teachers grouped according to major teaching field scored significantly different on L, R, Sh, M, I, A, Sx, Hs, Br, and P, null hypothesis III,B was rejected for these variables but was not re- jected for the remaining variables of FD, C, H, At, Ab, Ax, and Pn. Summary of the Findings Table 18 presents a summary of the finding for each variable as it differentiates among student teachers grouped according to success (hypotheses I; I,A; l,B), sex (hypothesis II) and major teaching field (hypotheses III; III,A; III,B). With the entire student teacher sample grouped according to success (hypothesis I), only R (rejection) appeared to significantly differentiate among the success ratings. However, the unequal variance of the R variable (or the failure of the R variance to meet the assumption 113 mmma no Hm>ma name you m>fim 639 pm pGMOHMchHm mmma mo Hm>ma pcmo awn m>Hm 62» pm unmofiwficwfim poz * m2 camfim mcfinomoe genes on wcfiomooo¢ omasoau * mz * * m2 m2 * mz * mz 3* m * mz mz * * mhmcomme pamUSpm mHmEmm m.HHH 69696 wcfieomme ache: ou wcfiohooo< consoam mz mz mz mz m2 m2 mz * m2 m2 m2 mz mz mz m2 m2 m2 mmmcomme pcmvzpm mam: «.HHH camfim weapomme MOnmz on wcfionooo< * m2 * m2 m2 mz * mz * * * * * m2 * * * 6665686 66666669 pemezum HHH xmm on wcHUnooo< * m2 m2 mz m2 m2 * x * x * * mz mz * * mz Umosopw mmmnomme pcmozpm HH mmmoosm cmpmm o» wzfiomooo< coasoau mz m2 m2 m2 mz mz mz m2 m2 m2 m2 m2 m2 m2 mz * mz mhmhomme meGSpm mHMEmm m.H mmmoosm cmpmm on wcfiopooo< omdsoho m2 m2 mz m2 m2 m2 m2 m2 m2 m2 mz m2 m2 mz m2 * mz mpmnomme meGSpm 6H8: ¢.H mmmoosm ompmm op wcfiopoood m2 m2 m2 m2 mz mz mz mz m2 mz mz mz m2 m2 mz * mz condone mhmnomme uncoupm H m mm mm mm x< b< xm p< < m H 2 am 0 mm m A mommnpoamm moanmfinm> .mwsfiecfie no samsssmuu.ma mamas 119 of homoscedasticity for the F test) suggested that the significance of the R variable is doubtful. When the student teachers were first grouped according to sex (hypotheses I,A; l,B) and then according to success rating, the R variable was the only significant finding, although (and similar to the hypothesis I finding for R), the unequal variance suggested a questionable signifi- cant finding. When the student teachers were grouped according to sex (hypothesis II) and the variables studied for significant differences between males and females, nine variables (R, FD, M, I, H, A, At, Sx, and P) were found to be significantly different. Finally, when the sample was classified according to major teaching field (hypothesis III), eleven variables (L, R, FD, Sh, M, I, H, A, Sx, Br, and P) were found to be significantly different among teaching fields; when the student teachers were first classified according to sex and then teaching field (hypotheses III,A and III,B) only one variable, At, significantly differentiated among males classified according to teaching fields. For females grouped according to teaching fields, however, ten variables (L, R, Sh, M, I, A, Sx, Hs, Br, and P) were found to be significantly different. 115 Discussion of the Findings In the immediately preceding section of the present study, the major hypotheses were statistically analyzed and conclusions drawn. It has been the major focus of the study to determine what relationship, if any, exists between Holtzman Inkblot variable scores and success in student teaching, sex, and major teaching field. The hypotheses were necessarily limited in that, prior to this study, no relationships were known to exist. And in the investigation several significant findings were reported. To predict, hypothesize or guess what the significant differences mean, however, or to say what they are due to is, at best, speculative. The fundamental principle of clinical interpreta- tion consists of constant formulation of hypotheses, and as more evidence for a given hypothesis accumulates, more confidence is placed in it. As evidence appears to the contrary, the hypothesis is modified, changed or other- wise eliminated entirely from consideration. Evidence may come from clinical tests, case histories, diagnostic interviews and a wealth of other psychological techniques. The types of statements made in this section are based on the whole backlog of dynamic personality theory and psycho— analytic research and rely heavily on published Rorschach technique interpretative materials. The statements are not presented as facts, but simply as basic or first-level 116 hypotheses. If one looks upon the following discussion in this light, the findings are seen simply as hypotheses derived from an essentially group or non-clinical setting and should be considered as only gross or basic statements of assessment of the results. The writer feels compelled to make the point repeatedly of the dangers of inter- preting findings without other supporting clinical evi- dence. Further verification of the comments are necessary before one can minimize the tenuousness of any given statement. In other words, Rabin'su admonition of the "shoestring operation" (daring and detailed interpreta- tion without sufficient information) should be pointed out to those who would infer more than can be substantiated and is apropos of the ensuing discussion. In addition, even if individual diagnostic inter- views and case histories were available for comparative purposes, it should be noted that many hypotheses derived from projective techniques are not capable of being verified or discredited by knowledge of overt behavior. It has been pointed out by many clinicians that it is merely interesting if hypotheses from a projective tech- nique and a case history agree, but highly important when they do not. The implication is that projective “Albert I. Rabin, Varieties of Rorschach Interpre- tation, paper delivered in symposium on "The Uses and Misuses of the Rorschach," at the Michigan Psychological Association meetings in Detroit, May, 1958. 117 techniques tap levels of personality not revealed in the ordinary interview situation and as such make a real con- tribution to personality assessment. Hypothesis One--Discussion of Significant Findings Hypothesis one stated that there would be no signifi- cant relationships among each of the mean Holtzman vari— able scores of student teachers grouped according to rated success. For sixteen of seventeen variables, no signifi- cant differences existed. However, one variable, R (rejection), did significantly differentiate rated success in student teaching among the student teachers. However the significance of the R variable in this and the other hypotheses is questionable since the statistical test of significance among mean scores, the F test, assumes equal variances or homoscedasticity. Cursory investigation of the R variance (Table 10) shows a vari— ance range from 1.51 to 9.02 thus suggesting that the equal variance assumption for the F test is not justified in this instance. Nevertheless, a discussion of the R variable does seem appropriate. There are many methods of rejection ranging from no response at all (answer sheet left blank) to responses such as "just an inkblot" or "nothing at all." It is especially hazardous to interpret the meaning of a high rejection score on a group test. It could be that the 118 subject lost his place on the answer sheet; it could be that the subject was momentarily distracted, or, perhaps, that his pen simply ran out of ink while the test was in progress. If, however, one assumes that the subject saw the inkblot and failed to see anything at all, several tentative hypotheses, subject to further verification, are possible. High rejection scores are typically made by sub- jects who are relatively non-responsive to their environ- ments and/or are psychologically defensive. One possible interpretation, then, is that student teachers who score at the extreme ends of the rating scale are less defensive and more responsive to their environments than others. However, this single significant finding suggests that such an interpretation is very doubtful at best since it is not supported by additional significant differences in other variable scores. The mean R scores of outstanding and poor student teachers are approximately the same, thus, the possible significance of the R variable loses its meaningfulness in differentiating between outstand- ing and poor student teachers. Hypothesis Two--Discussion of Significant Findings Hypothesis two investigated the mean Holtzman scores of student teachers grouped according to sex and confirmed 119 that male and female variable scores differed signifi— cantly on nine variables: R, FD, M, I, H, A, At, Sex, and P. The R (rejection) variable, besides being related to success in student teaching, is also significantly related to the sex of the subject, although there is some question of whether the assumption of homoscedastic- ity was met. Males gave responses scored as Rejections twice as often as females and, considering the previous discussion on the Rejection variable, this suggests that males, as a group, may be less responsive or more guarded in their perceptions than females. On the other hand, it may simply be that the male student teachers were un- c00perative. On the FD (form definiteness) variable, females appeared to perceive concepts with definite structure more often than males, suggesting that females are more demanding than males for definite concepts and less given to perceiving abstract concepts that do not have structure. Such a notion is doubtful at best, however, since a subject who is critical and demanding with regard to acceptable definiteness may well reject an inkblot, and males have a significantly higher rejection score than females. On the other hand, if one assumes that the higher male rejection scores were not due to a demand for form 120 definiteness but to momentary distractions or other causes, one may hypothesize that, as a group, females are less given to perceiving abstract, free-form concepts than males. Such is certainly not a conclusion, but only a hypothesis based on an assumption. As a group, females also scored higher on the M (movement) variable than males. Psychologically, move- ment projection is typically related to creativity, fantasy, self control, imagination, good intelligence and perhaps the capacity for delay of gratification. Again, an interpretation is not warranted, but one may safely conclude that, as a group, males project less energy into the inkblots than females, although the exact meaning of a high or low M score lies in individual analysis and supporting clinical evidence. Besides scoring significantly higher in FD and M from males, females also scored higher on the I (integra— tion) variable. Integration scores represent the organi- zation of two or more perceived blot elements into a larger whole. For example, "two roosters fighting" or "a collection of shells" represent integrations or organized perceptions which are somewhat different from instantaneous perceptions. Psychologically, the subject sees the whole blot, analyzes it into components, sees a relationship between or among the parts and organizes the inkblot into a whole. Thus, the process is one of 121 higher analytic and organizational sophistication suggest- ing that female student teachers tended to synthesize related blot stimuli into organized perceptions more often than males. The remaining variables (H, human; A, animal; At, anatomy; Sx, sex; and P, popular) are primarily content oriented and are related to the response content. For example, "two people talking" has human content; "a large elephant" is animal content, etc. Assessing the significant differences in response content between any groups is extremely tentative, par- ticularly with results from a group test. The data are only meaningful in an actuarial sense and do not lend themselves to easy interpretation. If, for example, many fierce human figures are perceived by a particular group, ‘what does this mean? Are females acting out their aggression? Are they inhibiting it, repressing it? Per- haps they are having a reaction formation against it. It is ludicrous to suggest that females, as a group, act out their aggression. In addition, the meaning of seeing many humans, for example, is unclear unless one knows the type of human or what they are doing. "Two humans talking" has a different psychological meaning than "two humans kneeling." In a sense, then, one can only deduce content meaning from an individual analysis of a protocol, and group PLEASE NOTE: Not original copy. Some-pages have very light type. Filmed as received. University Microfilms 122 differences in content can only be commented upon in the most general sense. Thus, it appears that there is no easy method to assess the meaning of the content differences other than through individual comparisons of projection with con- scious attitudes such as may be obtained from case his- tories or from responses to other projective techniques. And even then the meaning may elude the clinician. What follows are comments about what one can say generally about the differences in content scores, although the meanings of the group differences lie beyond the scope of this study. In the Holtzman Inkblots, females, as a group, perceived significantly more human figures than males. Generally, one could say then that the female student teachers seem to be more "people oriented" than males, since perception of human figures is related to identifi- cation. On an individual basis, however, if a subject is afraid of people, he may not see many human figures. Pre-occupation with particular parts of the body may point to certain concerns or difficulties, or he may not be able to determine whether human figures he sees in the blots are males or females, revealing problems of sexual identification. But these comments are individual in nature, based on individual protocol analysis, and nothing of this nature can be said about the group as a whole 123 other than females appear to be more strongly identified with other pe0p1e than males. Much of the same can be said for Animal content. Females perceived more animal figures than did males, but its meaning lies in individual analysis. Animals are perceived by subjects of all ages, both normal and abnormal. These responses may represent potential ad— justive techniques. Even if the scores were to indicate what kind of animal, their meaning is still largely specu- lative, although there is some suggestion that seeing fierce animals such as tigers and lions, may refer to certain aggressive tendencies in the individual that the subject is trying to handle in some way. Passive animals, such as cows and sheep, and domestic animals such as dogs or cats may show passivity and dependent attitudes. Since the data show that females perceived more animals than males, but do not indicate what kind of animal, we can only cite the fact but cannot specify its meaning. The same interpretive difficulties occur in assess- ing Sexual and Anatomy responses. Males scored signifi- cantly higher in both of these areas than did females. The meaning attached to sex responses needs to be studied in relation to other factors in the individual's test protocol and with some knowledge of the subject's life history. It is impossible to say on the basis of sexual responses alone whether the individual is avoiding sexual 129 problems by naming them, or whether the subject is worried about sexual problems. If it is impossible to deduce the meaning on an individual basis without other confirming data, so much the more to comment on group data. Generally, though, one could say that the males are more "sexually oriented" and appeared to have more of an interest in sexual themes than females. Anatomy responses, too, are exceedingly difficult to interpret especially from a group test where individual case histories are not available. Typically, Anatomy responses are often given by subjects who are trying to show off intellectually as one way of covering up real feelings. Anatomy responses may also indicate feelings of intellectual inadequacy, especially when the responses avoid any commitment to definite form or specification. Thus, it could be speculated that males, as a group, are less prone to show their real feelings than females. The P (popular) variable, however, is somewhat easier to interpret than other content areas. The P variable is not comparable to other content area scores since the POpular variable indicates how often the subjects see things as other people do. On this basis, it can be said that female student teachers, when compared to male stu- dent teachers, see things more often as other people do. This may be related to sociological needs and pressure to conform. In other words, the Popular score reveals q 125 the group's tendency to view the world in the same way as most other people. Since females see significantly more popular perceptions than males, one can characterize the group as having a need to think as other people do, or at least more than male student teachers do. Hypotheses I,A and I,B Since the investigation of hypothesis II identified several variable scores which differentiated significantly between the sexes, and since hypothesis I did not identify any variable scores (other than R) which were signifi- cantly different among student teachers grouped according to success ratings, the investigator re-examined hypo- thesis I by grouping the student teachers by sex before seeking significant differences when classified according to success ratings. Hypothesis I,A stated that there would be no signifi- cant differences among each of the mean Holtzman variable scores of MALE student teachers grouped according to success rating; hypothesis I,B stated the same but for FEMALE student teachers. For sixteen of the seventeen variables in both hypotheses no significant differences were found. The single exception, however, was the R variable, which was discussed previously on page 117 and need not be re-discussed here. Hypothesis III Hypothesis III investigated the mean Holtzman scores of student teachers grouped according to major teaching field and identified eleven variables (L, location; R, rejection; FD, form definiteness; Sh, shading; M, move- ment; I, integration; H, human; A, animal; Sx, sex; Br, barrier; and P, popular) which were significantly different among the thirteen major teaching areas. Eight of these variables, R, FD, M, I, H, A, Sx and P were also signifi- cantly different between the sexes and were discussed on pages 117 through 125. L, Sh, and Br, however, were the additional vari— ables that differed significantly among major teaching fields. L scores of 0, l, and 2 correspond approximately to the Rorschach W (whole), D (large detail), Dd (small de- tail) scores. In Rorschach interpretation, emphasis is placed on the percentage relationship of W, D and Dd. As such, the Holtzman pooled L score is not directly com- parable to the W, D, and Dd but does suggest perceptual tendencies. In the Rorschach, a high W percentage score usually is a sign of organizational interest and ability. The subject possessing such ability is interested in and capable of viewing the relatively separate facets of his experience as an interrelated whole. A high D percentage score suggests interest and ability to differentiate 127 perceptually, with relatively little interest in inte- gration and organization. This could be interpreted as a practical, everyday, common-sense application of intelligence, an interest in the presented, obvious facts without much drive to seek relationships between presented facts of experience. A high Dd percentage score in the Rorschach is obtained by selecting for the response a small blot area and may represent not only a differentiated interest in factual things, but often is associated with an emphasis on accuracy, correctness and exactness or sometimes simply reflects a need for cer- tainty or minutiae of life. Thus, it would appear that the higher the Holtzman L score, the more the subject perceives the minutiae of life and, conversely, the lower the L score, the more apparent is the tendency to view life as interrelated. There is nothing in the traditional Rorschach scor- ing system which approximates the Holtzman Br variable. At best, the interpretation of a high Br score would suggest an interest in being protected from danger, harm, or the unknown, suggesting an underlying fear of some person, place, act or animal. Generally, the rationale of the use of shading is based on the general hypothesis that the way in which a person handles the shading aspects of a blot is related to the way the subject handles his primary security need 128 and derived needs for affection and belongingness. Since some frustration of affectional need is practically uni— versal and all persons have more or less anxiety from this source, the mere presence of shading in a response does not suggest anything out of the ordinary. However, the higher the Sh variable score, the more probable is the hypothesis that there is a significant amount of anxiety and frustration over affectional needs present. Hypotheses III,A and III,B As previously discussed, the investigation of hypothesis 11 identified several variable scores which ' significantly differed between the sexes and it was there- fore logical to re-examine hypothesis III by grouping the sample according to sex prior to investigating for significant differences among student teachers grouped according to major teaching fields. Hypothesis III,A stated that there would be no significant differences among each of the mean Holtzman variable scores of MALE student teachers grouped according to major teaching field; hypothesis III,B was similar but for FEMALE student teachers. The test of hypothesis III,A shows that of the seventeen variables only one, At (anatomy), reached statistical significance at the five per cent level. The At variable is a content score and as such presents 129 interpretive difficulties discussed previously on page 129. The same comments would apply here and would suggest the same speculative interpretation with no con- clusive interpretive remarks. The investigation of hypothesis III,B, however, revealed that ten variable mean scores (L, location; R, rejection; Sh, shading; M. movement; I, integration; A, animal; Sx, sex; Hs, hostility; Br, barrier; and P, popular) are significantly different among female student teachers grouped according to major teaching field. Six of these variable scores (R, rejection; M, movement; 1, integration; A, animal; Sx, sex; and P, popular) were also significantly different between the sexes (Table 10) and discussed on pages 118 through 125 and the re- maining variables of L (location), Sh (shading), and Br (barrier) were significantly different among major teach- ing fields and discussed on pages 126-128. The remaining significantly different variable, Hs (hostility) of females grouped according to major teaching field, was not significantly different in any other hypothesis. The Hs or Hostility variable concept score repre- sents symbolic, implicit or explicit signs of hostility. Although the mean Hs score does not differentiate be- tween the sexes of the student teachers, females, when considered as a group score significantly different 130 among each other when classified according to major teach— ing field. Why this variable is significantly different, or what interpretations can be given to this finding is unknown. And other than reporting the fact, little else can be said with any degree of certitude. Perhaps it is appropriate, in summary, to again call the reader's attention to the purpose of the study. Prior to the undertaking of this investigation, no rela— tionships of Holtzman variables to success in student teaching or major teaching fields were known to exist. At best, the study could have identified significant differences in such a way as to suggest group character- istics which could be studied in some later, more indi- vidual, or clinical inquiries. Failing to find clear significant differences in the variable scores when the sample was grouped according to various classifications did provide answers to the hypotheses posed and added more information which might be useful to other investi- gators and further studies in this area. CHAPTER IV SUMMARY, CONCLUSIONS AND RECOMMENDATIONS Summary The present research studied the relationship be- tween computer scored group Holtzman Inkblot variable mean scores and student teaching success (Table 9). The study also investigated the relationship between variable scores of student teachers grouped according to sex (Table 10) and grade/subject teaching fields (Table 19). The study also generated norms (Table 8) for student teachers on the group Holtzman Inkblot Test and determined the feasibility of administering and computer scoring a projective technique with a large sample. The group version of the Holtzman Inkblot Test was administered to a total of 802 Michigan State University student teachers during the Winter Term, 1968. The test was administered in each of the geographical centers where Michigan State places student teachers. Forty-five inkblots were exposed to each group for one minute each and the student teachers responded with a short written description of what each inkblot "looked like" on the specially designed group answer sheet (Appendix D). The answer sheets were alpha-numeric coded to identify the 131 132 sex and teaching field/grade level of the student teachers. After collecting all responses from the student teachers, the data and responses were keypunched according to the format specified by Gorham1 (Appendix E), loaded onto magnetic tape and, through special arrangements with Gorham, the tape was submitted for computer scoring to the Holtzman Inkblot Scoring Service in Perry Point, Maryland. Output from the scoring service included a print-out and a score summary card (Appendix E) for each student teacher on seventeen Holtzman variables: Loca- tion, Rejection, Form Definiteness, Color, Shading, Movement, Integration, Human, Animal, Anatomy, Sex, Abstract, Anxiety, Hostility, Barrier, Penetration and Popular. The study was limited to testing the following hypotheses: (l) Hypotheses I, I,A and I,B stated that there would be no significant difference on the Holtzman variable mean scores among (I——all; I,A—-ma1e; l,B-- female) student teachers classified according to rated success by the university coordinators. (2) Hypothesis 11 stated that there would be no significant differences on the Holtzman variable mean scores when the student teachers were classified according to sex, and 1Donald R. Gorham, The Development of a Computer Scoring System for Inkblot Responses, Report #1, Psycho- logical Research Reports, Research Laboratory, Psychology Service, Perry Point, Maryland, September, 1965. 133 (3) Hypothesis III stated that there would be no signifi— cant differences on the variable scores when (III--all; III,A—-male; III,B--female) student teachers were grouped according to teaching field or grade level. Conclusions Tables 9, 10, ll, 12, l3, l9 and 15 show the com— puted means, standard deviations, F statistic and its approximate significance for testing each null hypothesis. The analyzed data produced the following findings: Hypothesis I--Conclusions The statistical tests performed on hypothesis I show that, with the exception of one variable (R, rejec— tion), there are no significant differences among the mean Holtzman variable scores of student teachers grouped according to rated success. Thus, the conclusion is that the Holtzman scores are of little or no value in dis- criminating among the success ratings of the student teachers. Hypothesis I,A and I,B-—Conclusions The statistical tests performed on hypotheses I,A and I,B show that, with the exception of one variable (R, rejection) there are no significant differences among the mean Holtzman variable scores of male (I,A) or female (I,B) student teachers grouped according to rated success. 139 Thus, the conclusion is that the Holtzman scores are of little value in discriminating among the success ratings of either male or female student teachers. Hypothesis II+~Conclusions The statistics for testing hypothesis II show that eleven of the seventeen computer scored Holtzman variable mean scores significantly differentiate between student teachers grouped according to sex. Thus, the conclusion is that the Holtzman scores are of some but restricted value in discriminating between student teachers grouped according to sex. Hypothesis III--Conc1usions When the sample of student teachers (excluding those at the junior high school teaching level) were grouped into thirteen major teaching areas, the test of hypothesis III showed that nine of the seventeen Holtzman variable mean scores differentiate among the teaching areas. Thus, the conclusion is that the Holtzman scores are of some but restricted value in discriminating among student teachers grouped according to major teaching field. Hypothesis III,A and III,B-—Conglusions The statistical tests performed on hypotheses III,A and III,B show that when the student teachers are first grouped according to sex (male, III,A; female, III,B) and 135 then teaching field, there are no significant differences among the mean variable scores, with the exception of one variable, At (anatomy), for the male student teachers. Thus, the conclusion is that the Holtzman variable scores are of little or no value in discriminating among male student teachers grouped according to major teaching field. For females, however, grouped according to major teaching field (hypothesis III,B), eleven of the seventeen mean variable scores are significantly different. Thus, the conclusion is that the Holtzman variable scores are of some but restricted value in discriminating among female student teachers grouped according to major teaching field. Discussion The conclusions resulting from the statistical tests of the hypotheses are not claimed for any group other than the student teachers represented in this study. The data for the study were obtained entirely at Michigan State University during the Winter Term of 1968 and similar conclusions for other groups at other times can— not be inferred from this particular study. Perhaps as important, though, as statements regard— ing the generality of the conclusions, are comments on the assumptions and framework within which the study was con- ceived. 1. U.) 0 \ If, as Combs2 suggests, "whether an individual will be an effective teacher depends upon the nature of his private world of perceptions," then either this statement is an oversimplification of a highly complex reality or, insofar as the Holtzman Test measures perception, it is without supportive evidence at this point. However, even this may not be quite so. It was assumed in this study that personality is an important variable in successful teaching. This is well documented in the general literature and common sense suggests this is the case, but objective, empirical evidence is lacking. The problem stems, in part, from difficulties in personality definition, criteria and measurement. And the problem is exceedingly compounded by the relationship between reported perception and per- sonality, not to mention difficulties in criteria and measurement of successful teaching. If projective tech- niques measure perception, as most authorities would agree, their meaning is highly clinical or individual in nature and only the most general statements can be made on the basis of a group projective technique. A concept of a "group personality" is a contradiction in terms since personality as such has its meaning only in an individual sense. 2Combs, The Professional Education of Teachers, p. 19. 137 And then, too, it was assumed that the university coordinator could judge success in student teaching. It would be unwarranted to believe that subjective judg- ments of success, even with specific criteria, are in any way objective. Yet, the present state of teaching suc- cess research suggests that supervisory ratings are the most appropriate measure of success. Concepts of teach- ing success vary from one judge to another, and in this study the variance of success rating among coordinators is unspecified. What is success in student teaching as judged by one coordinator may not agree with another's judgment. Perhaps this is why no significant differences in mean Holtzman scores were found when student teachers were grouped according to rated success. In other words, perhaps the assumption that the coordinators' rating of success on the basis of performance discriminates among student teachers who were more successful than others in their experience is a questionable assumption and clouds over the discovery of significant mean score differences. The problem of rating success in not an easy one, but it cannot be simply ignored. It is caused by the absence of any generally agreed upon criteria of teaching success and objective methods of measurement. This study, or any similar study, would be dramatically improved by specification of teaching success criteria and objective 138 measurement, thus eliminating the subjective element in success judgments. In sum, it may be that the primary obstacle in dis- covering distinctive and specific features of the success— ful student teacher personality is due generally to the lack of an adequate theory of teaching and its resultant lack of success criteria, subjectivity in judgment even when the criteria are specified, problems in relating perceptual modes to individual personality from a group test, and the effect of subjectivity in all of these cases. In any event, the feasibility of such an under- taking has been evaluated. That is, it is technically and feasibly possible to administer and score a group projective technique to a large sample of subjects. The technique of computer scoring permits an endless variety of individual responses to be scored objectively (that is, the only subjective element is that which is built into the computer scoring program and which is a constant for the scoring of all protocols) with a minimum amount of time and effort. However, the interpretation of results should only be on an individual basis since a "group" interpretation is meaningless and only serves to describe the group in a most general sense. From the data collected, it would appear that the classification of all responses of all the subjects into seventeen 139 categories does not produce common perceptual modes. That is, the private world of perceptions, as measured by the Holtzman Inkblot variables, of successful student teachers does not differ significantly from the private world of perceptions of less successful student teachers. Recommendations From the presentation, analysis, and discussion of the data, the following recommendations are made: 1. A longitudinal study of the student teachers should be undertaken to follow-up their teaching success as a first year teacher. This would permit a determina- tion of the relationship between rated student teaching success and success on the job. 2. A study should be made to determine the vari- ance in success judgments among university coordinators. An ideal study in this area would have all coordinators judge the success of a sample of student teachers and in this fashion determine the amount of variance, if any, that exists in coordinators' success judgments. If a significant amount of variance does exist, there would be cause to re-evaluate the data in this, and other, similar studies. 3. It is also recommended that an in-depth study be made of the university coordinator to determine if personality is a significant factor in the university coordinator's evaluation of the student teacher. 190 9. It is recommended that the significance of the R (rejection) score when related to rated success be pur— sued with specific measures of "openness" or "psychologi— cal defensiveness" and its relationship to success in student teaching. 5. There should be a continued search for those personality characteristics related to successful student teaching. This should be done, however, by clinical or individual methods involving those student teachers at the extreme ends of the rating scale and comparing the individual protocols for systematic differences or simi- larities. 6. It is recommended that studies in the area of personality and student teaching success be less global in nature and utilize projective techniques that introduce an interactiOn or sociological dimension such as might be found in a teacher-type Thematic Apperception Test. 7. It is suggested that since the sexes scored significantly different on nine variables, future studies should continue to analyze differences in light of these findings and relate the data to cultural stereotypes of male and female temperaments. 8. It is also recommended that since females, when grouped according to major teaching area, scored signifi— cantly different on eleven variables, future studies 191 should analyze differences in the light of major teaching field and relate the data to teacher stereotypes of subject matter areas. 9. Finally, it is recommended that further work investigating personality and teaching success be under- taken when there is general agreement, either institu- tional or profession-wide, on what is successful teaching, derived from an overall theory of teaching. BIBLIOGRAPHY 192 BIBLIOGRAPHY Abt, Lawrence E., and Bellack, Leopold. Projective Psy- chology. New York: Alfred A. KnOpf, 1950. Alexander, T. "The Adult-Child Interaction Test: A Projective Test for Use in Research." Monograph Social Research in Child Deve10pment, Vol. XVII (No. 2, 1952). Allen, Thomas K. "Personality Patterns of Student Teachers Planning to Teach Different Subjects." Journal of the Student Personnel Association for Teacher Education, Vol. IV (December, 1965), I 89-93. "Personality as a Predictor of Student Teaching Success." Journal of the Student Personnel Association for Teacher Education, Vol. VI (Fall, 1966), 12—16. Allport, Gordon W. Pattern and Growth in Personality. New York: Holt, Rinehart and Winston, 1961. Barr, Arvil S. "The Measurement and Prediction of Teacher Efficiency, A Summary of Investigations." Journal of Experimental Education, Vol. XVI (June, 1998), 203-283. . "Pupil Changes and Teacher Ability," Journal of Experimental Education, Vol. XIV (September, 19553, 52-714. ”Wisconsin Studies of the Measurement and Pre- diction of Teacher Effectiveness: A Summary of Investigations." Journal of Experimental Educa- tion, Vol. XXX (September, 1961), 5-156. Biddle, Bruce J., and Ellena, William J. Contemporary Research on Teacher Effectiveness. New York: Holt, Rinehart and Winston, 1969. Charles, Harvey. "The Use of a Projective Technique in Teacher Selection." National Catholic Educational Association Bulletin, Vol. LVII (August, 1961), 172-173. 193 199 Clark, E. J. "The Mental Health of Elementary School Teachers as Measured by the Guilford-Martin Per— sonality Battery." Paper read at the National Council on Measurements Used in Education, Atlantic City, March, 1950. Cohan, Richard W. "The Holtzman Inkblot Test." The Sixth Mental Measurements Yearbook. Edited-By O. K. Buros. 3rd ed. Highland Park, New Jersey: The Gryphon Press, 1965. Combs, Arthur W. "The Personal Approach to Good Teach- ing." Educational Leadership, Vol. XXI (March, 1969), 269-377. . The Professional Education of Teachers. Boston: Allyn and Bacon, Inc., 1965. Cook, Desmond, et al. "A Factor Analysis of Teacher Trainee Responses to Selected Personality Inven- tories." Educational and Psychological Measure— ment, Vol. XXI (Winter;’l961), 865-872. Cronbach, L. J. "Statistical Methods Applied to Rorschach Scores: A Review." Psychological Bulletin, Vol. XXXXVI (September, 1999), 393—929. Dixon, W. R., and Morse, W. C. "The Prediction of Teaching Performance: Empathic Potential." Journal of Teacher Education, Vol. XII (September, 1961), 322-329. Domas, Simeon J., and Tideman, David V. "Teacher Com- petence: An Annotated Bibliography." Journal of Experimental Education, Vol. XIX (December, 195*), 101-218. Erickson, Harley E. "A Factorial Study of Teaching Ability." Journal of Experimental Education, Vol. XXIII (September, 1959), 1—39. Evans, Kathleen M. "A Critical Survey of Methods of Assessing Teacher Ability." British Journal of Educational Psychology, Vol. XXI (June, 1951): 89-95. Fattu, Nicholas A. "Effectiveness-~An Elusive Quality." The Education Digest, Vol. XXVII (January, 1962), 29-26. 195 Flanagan, Carroll E. "A Study of the Relationship of Scores of the MMPI to Success in Teaching as Indicated by Supervisory Ratings." Journal of Experimental Education, Vol. XXXIX (June, 1961), 329-359. Frank, L. K. "Projective Methods for the Study of Per- sonality." Journal of Psychology, Vol. VII (1939), 389-913.' Quoted in Klopfer, Bruno, and Davidson, Helen H. The Rorschach Technique: An Introductory Manual. New York: Harcourt, Brace and World, Inc., 1962. Gage, N. L. Address appearing in "Proceedings." Research Resume. Burlingame, California: California Teachers Association, 1960. Getzels, J. W., and Jackson, P. W. "The Teacher's Per- sonality and Characteristics." Handbook of Re— search on Teachin . Edited by N. L. Gage. Chicago: Rand McNallyIEnd Company, 1963. Goodenough, Florence L.; Fuller, Elizabeth M.; and Olson, Edna. "The Use of the Goodenough Speed of Associa- tion Test in the Pre-Service Selection of Nursery. School-Kindergarten-Primary Teachers." Journal of Educational Psychology, Vol. XXXVII (September, 1996), 335-395. Gorham, Donald R. The Development of a Computer Scoring System for Inkblot Responses. Report #1, Psy- chological Research Reports, Research Laboratory, Psychology Service, Perry Point, Maryland, September, 19 5. "Validity and Reliability of a Computer Based Scoring System for Inkblot Responses." Journal of Consulting Psychology, Vol. XXXI (February, 1967), 65-70. Gorham, Donald R.; Moseley, Edward C.; and Holtzman, Wayne H. "Norms for Computer-Scored Holtzman Inkblot Technique." Perceptual and Motor Skills, Vol. XXVI (June, 1968), 1279-1305. Gowan, J. C., and Gowan, May S. "The Guilford-Zimmerman and the California Psychological Inventory in the Measurement of Teacher Candidates." California Journal of Educational Research, Vol. VI (January, 1955;, 35-370 196 Gustafson, Kent L. "Simulation of Anxiety Situations and Its Resultant Effect on Anxiety and Classroom Interaction of Student Teachers." Unpublished PhéD. dissertation, Michigan State University, 19 9. Hadley, S. T. "A Study of the Predictive Value of Several Variables of Student Teaching Success as Measured by Student Teaching Marks." Teacher's College Bulletin, Vol. LX (No. 3, 1959), 1-10. Hamacheck, Donald. "Characteristics of Good Teachers and Implications for Teacher Education." Phi Delta Kappan, Vol. L, No. 6 (February, 1969), 391-399. Hampton, N. D. "An Analysis of Supervisory Ratings of Elementary Teachers Graduated from Iowa State Teachers College." Journal of Experimental Edu— cation, Vol. XX (December, 1951), 180-215. Hathaway, 8. R., and McKinley, J. C. Minnesota Multi— phasic Personality Inventory Manual. New York: The Psychological Corporation, 1951. Hertz, Marguerite R. "The Use and Misuse of the Rorschach Method." Journal of Projective Techniques, Vol. XXIII (March, 1959), 33-98. Hicks, William Vernon, and Blackington, Frank H. Intro- duction to Education. Columbus, Ohio: Charles E. Merrill Books, Inc., 1965. Hilton, Thomas A. "Ego Involvement in Teaching: Its Theory and Measurement by a Word Completion Technique." Unpublished Ph.D. dissertation, Harvard University, 1955. Holtzman, Wayne H., et a1. Inkblot Perception and Personality. Austin, Texas: University of Texas Press, 1961. . "Comparison of the Group Method and the Standard Individual Version of the Holtzman Inkblot Tech— nique." Journal of Clinical Psychology, Vol. XIX (October, 1963), 991-999. Holtzman, Wayne H. "The Holtzman Inkblot Technique." Projective Techniques in Personality Assessment. Edited by Albert Rabin. New York: Springer Pub- lishing Company, 1968. 197 Inlow, Gail M. "Comparative Study of Student Teaching Practices in 38 Midwestern Institutions." Journal of Experimental Education, Vol. XXVIII (June, 1960), 337-399. Jones, Margaret L. "Analysis of Certain Aspects of Teach- ing Ability." Journal of Experimental Education, Vol. XXV (December, 1956):’152-l80. Kemp, L. D. "The Prediction of Teaching Success." The Forum of Education, Vol. VI (August, 1997), 19. Klopfer, Bruno, and Davidson, Helen H. The Rorschach Technique: An Introductory Manual. New York: Harcourt, Brace and World, Inc., 1962. Lamke, Thomas A. "Personality and Teaching Success." Journal of Experimental Education, Vol. XX (December, 1951), 217—259. Leeds, C. H. "Teacher Attitude and Temperament as a Measure of Teacher-Pupil Rapport." Journal of Applied Psychology, Vol. XXXX (October, 1956), 333-337. I LePere, Jean M., and Cox, Richard C. Training Elementary Teachers: Comparisons of SeparatE and Block Methods Courses. East Lansing, Michigan: Bureau of EducationalIResearch Services, Michigan State University, 1969. Lewis, James N. "The Relationship of Attributes Measured by the Structured-Objective Rorschach Test and Success in Student Teaching." Unpublished Ed.D. dissertation, North Texas State University, 1966. Lindsey, Margaret. New Horizons for the Teaching Pro- fession. Washington, D.C.: National Commission on Teacher Education and Professional Standards, National Education Association, 1961. Lough, Orpha M. "Women Students in Liberal Arts, Nursing, and Teacher Training Curricula and the MMPI." Journal of Applied Psychology, Vol. XXXI (August, 1997), 937-995. ‘ Lynch, W. W. "Person Perception: Its Role in Teaching." Indiana University School of Education Bulletin, Vol. XXXVII (1962), 1-37. 198 Magee, Robert M. "Admission-Retention in Teacher Educa- tion." Journal of Teacher Education, Vol. XII (March, 1961), 81-85. McClendon, P. E. "Teacher Perception and Working Climate." Educational Leadership, Vol. XX (December, 1962), 109-108. Mitzel, Harold E. "Criteria of Teacher Effectiveness." Encyclopedia of Educational Research. Edited by C. W. Harris. New York: MacMillian and Company, 1960. Montross, Harold W. "Temperament and Teaching Success." Journal of Experimental Education, Vol. XXIII (September, 1959), 39—97. Moore, Clark H., and Cole, David. "The Relation of MMPI Scores to Practice Teaching Ratings." Journal of Educational Research, Vol. L (May, 1957), 711-716. Morsh, Joseph H., and Wilder, Eleanor W. Identifying the Effective Instructor: A Review of the Quanti- tative Studies, 1900-1952 (Research Bulletin, AFPTRC - TR 59T99). San Antonio: United States AiruForce Personnel Training Research Center, 195 . National Commission on Teacher Education and Professional Standards. The Education of Teachers: Considera- tions in Planning Institutional Progpams. ' Wazhington, D.C.: National Education Association, 19 0. Ohlsen,Merle M., and Schultz, Raymond E. "Projective Test Response Patterns for Best and Poorest Student Teachers." Education and Psychological Measure- ment, Vol. XV (Spring, 1955), 18-27. Osmon, Robert V. The Improvement of Secondary Teaching. St. Louis: Education Publishers, 1962. Pettinger, B. F. "Problems of Teacher Measurement." Journal of Educational Psychology, Vol. VIII (January, 1917), 103-110. Phillips, L; Kaden, S.; and Waldman, M. "Rorschach Indices of DevelOpmental Level." Journal of Genetic Psychology, Vol. LXXXXIV (1959), 267-285. 199 Rabin, Albert I. Varieties of Rorschach Inteppreta- tion. Paper delivered in symposium on "The Uses and Misuses of the Rorschach," at the Michigan Psychological Association meetings in Detroit, May, 1958. Ramsey, C. P. "Leadership Perception." Educational Leadership, Vol. XX (December, 1962), 151-159. Richter, H., and Winter, W. D. "Holtzman Inkblot Cor- relates of Creative Potential." Journal of Pro- eective Techniques and Personalipy Assessment, 0 . XXX (February, 1966), 62-67. Rolfe, J. F. "The Measurement of Teaching Ability." Journal of Experimental Education, Vol. XIV (September, 1995), 52-79. Ryans, David G. "The Criteria of Teaching Effectiveness." Journal of Educational Research, Vol. XLII (May, 1999), 690-699. . "Prediction of Teacher Effectiveness." Encyclo- pedia of Educational Research. Edited by C. W. Harris. 3rd ed. New York: MacMillian Company, 1960. - . Characteristics of Teachers. Washington, D.C.: American Council on Education, 1960. . "Assessment of Teacher Behavior and Instruc- tion." Review of Educational Research, Vol. XXXIII (October, 1963), 915-1991. Steeves, Frank L. "Crucial Issues in Student Teaching." Journal of Teacher Education, Vol. XVI (September, 1965), 307-310. Stevens, S. S. "Psychology and the Science of Science." Psychological Bulletin, Vol. XXXVI (April, 1939), 221-263; Stout, Ruth A. "Practices for Selection in Teacher Education." Teacher Education: The Decade Ahead. Washington, D.C.:T National Commission on Teacher Education and Professional Standards, National Education Association, 1955. "Selective Admissions and Retention Practices in Teacher Education." Journal of Teacher Education, Vol. VIII (December, 1957), 922—932. 150 Swartz, Jon D., and Holtzman, Wayne H. "Group Method of Administration for the Holtzman Inkblot Technique." Journal of Clinical Psychology, Vol. XIX (October, 19637. 933-991. Sultan, E. E. "A Quantitative Investigation of the Rorschach Inkblot Test as Applied to Student Teachers." British Journal of Social and Clinical Psychology, Vol. IV (September, 1965), 197-206. Symonds, Percival M. "Characteristics of the Effective Teacher Based on Pupil Evaluations." Journal of Experimental Education, Vol. XXIII (June, 1955), 289-310. Symonds, Percival M., and Dudek, Stephanie. "Use of the Rorschach in the Diagnosis of Teacher Effective- ness." Journal of Projective Techniques, Vol. XX (June, 1956), 227-239. Tarpey, Sister M. S. "Personality Factors in Teacher Trainee Selection." British Journal of Educational Psychology, Vol. XXXV (June, 1965), 190-199. VI Thetford, William H. "The Holtzman Inkblot Technique." The Sixth Mental Measurements Yearbook. Edited by O. K. Buros. 3rd ed. Highland Park, New Jersey: The Gryphon Press, 1965. Underhill, Robert G. "The Relation of Elementary Student Teacher Empathy (Affective Sensitivity) Change to Supervising Teacher Empathy and Student Teaching Success." Unpublished Ed.D. dissertation, Michigan State University, 1968. Willcox, Isabel, and Beigel, Hugo G. "Motivations in the Choice of Teaching." Journal of Teacher Education, Vol. IV (June, 1953), 10 —109. Witty, Paul. "Some Characteristics of Effective Teachers." Educational Administration and Supervision, Vol. XXXVI (April, 1950), 193-208. Zubin, J., "Failure of the Rorschach Technique." Journal of Projective Techniques, Vol. XVIII (September, 1959), 303-315. APPENDICES 151 APPENDIX A DESCRIPTION OF THE 17 COMPUTER—SCORED HIT VARIABLES 152 DESCRIPTION OF THE 17 COMPUTER-SCORED HIT VARIABLES '— m Variable ->-~ ___..--.. —— Brief Description . Score Range L A 3-point scale“ used to measure the tendency to fragment the inkblor into smaller areas. The greater the area‘used, the lower the score. 0-90 R Failure of S to give a scorable response. 0-45 FD A 5-point scale measuring the definiteness of the form of the concept reported. 0-180 C A 4-point scale measuring the primacy of bath chromatic and achromatic color in determining a response. 0-135 Sb A 3-point scale used to measure the primacy of shading as a response determinant. . 0-90 M A 5-point scale indicating the movement energy level 5 voluntarily ascribes to his percept regardless of its content. 0-180 1 A 2-point scale indicating the presence or absence of the organization of adquately perceived inkblot elements into a unified response. ' 0-45 H A 3-point scale used to measure the amount of human content seen. Score 1 is recorded for human details, and 2 for a whole human. 0-90 A A 3-point scale used to measure the amount of animal content seen. Score 1 is recorded for animal details and 2 for a whole animal. 0-90 At A 3-point scale used to measure the amount of anatomy content seen. Score 1 is recorded for bone structures, and 2 for visceral and crude anatomy responses. . 0-90 3:: A 3-point scale used to measure the amount of sexual content seen. Score 1 is recorded for socially acceptable sexual content and 2 for more blatant sexual responses. 0-90 Ab A 3-point scale characrerized by relative or tetal independence of the response from the formal b10t qualities. For instance; "happiness," "hope.” 0-90 Ax A 3-point scale used to measure the amount of anxiety refleCted by the response. 0-90 H: A 4-point scale used to measure the amount of hOStility elicited by the blot Stimulus. 0-135 Br A 2-point scale used to measure the presence or absence of body image boundary in a response. 0-45 P» A 2opoint scale used to measure the presence or absence of body image boundary diffusion or disruption in a response. 0-45 P A 2-point scale used to measure the presence or absence of a popular content. A response is judged popular which occurs at least once in seven prorocols for normal 55. 0-45 *In all other scales except R, the zero point represents an absence of the characteristic being measured. 153 APPENDIX B SUCCESS RATING CHART 159 .-¢ u. cocoon he. now auzomamom N20 Han—u VG; HO Maud—HUGO“ 03030. DHQQOKG Oi Ikfl‘OIOU anon-5- :. Ga 5 95qu guan< .ONZOLhmOL NOS-6:39.086 .uoueou use» cu «agenda» uneasy. auou as: onesooou unovaua age ea c neuucu emu-u< .cw4euouou you nanos- ecOocou uncuun> we 0.3.00: can : mausuuou o» vacuuoau uo: can zufioou hazy n ~«Qu cu axuao ~00u so» on: oucovau- onesu cu m ”and... :30: .02.:U¢n on: ouocu¢0u uneasy. ecosu ea a «canon anaco< Jauumuuuanmmwa Own we use mam usoan he one! .8 .2: 322 .N be d «wean-u we nouuuficav meuucauuuao sea :0:- uo: on can wcuuuu easy o>uouou vases. woo: .ouosodou ~53». Enema—86 3 m 9.2: .52: 43.332: Ono no use cum as... .5 «2 .cooo uo>u o>os so» unusuuou uneasy. anon ecu we use us: ca use ~ mafia-u cu anode Ouch tune: unovaun usu-nuouau ausuuc ash .oocqauOuuoa 9:53» ~83...- ozugamSo we. aEmSeBooa 65.33.. uuouosuuu>o= use d weuuuu easy buu~uau -uuo>o non-ad unseale- uo one.» 3 N 9.33 .532 .3333...» 5:2: .— weuunu emu-o- .uunovauu no uouuuo azaoaoam use auuuuuwm auuocou oeuua~¢=o floccuouououm nouuuuaaa uncoouum uguuuounz nauseous can 302% do c558 acuuusuuacH mcuwucaz ecuuunuuocH anaconda unusado Boouauo~o wa«;u«~nsuom «38a 5:: 9.23: mmmouam mo ¢u> 5: no 36 565.. SEE—«55H «BIS mush HD—flUQI Oh hqmflud mhllobhh hO HU