University of Southern California Dissertations and Theses

The Relationship Of Teacher Empathy And Student Personality To Academic Achievement And Course Evaluation

(USC Thesis Other)

The Relationship Of Teacher Empathy And Student Personality To Academic Achievement And Course Evaluation

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE RELATIONSHIP OF TEACHER EMPATHY AND STUDENT PERSONALITY TO ACADEMIC ACHIEVEMENT AND COURSE EVALUATION by Alice Frances Chang A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Psychology) August 1973 INFORMATION TO USERS This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted. The following explanation of techniques is provided to help you understand markings or patterns which m ay appear on this reproduction. 1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced in to the film along with adjacent pages. This may have necessitated cutting th ru an image and duplicating adjacent pages to insure you complete continuity. 2. When an image on the film is obliterated with a large round black mark, it is an indication th a t the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame. 3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is custom ary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below th e first row and continuing on until complete. 4. The majority of users indicate th at th e textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced. 5. PLEASE NOTE: Some pages m ay have indistinct print.. Filmed as received. X erox University Microfilms 300 North Z eeb Road A nn Arbor, Michigan 48106 I ;l I 74-9059 CHANG, Alice Frances, 1943- THE RELATIONSHIP OF TEACHER EMPATHY AND STUDENT PERSONALITY TO ACADEMIC ACHIEVEMENT AND COURSE EVALUATION. University of Southern California, Ph.D., 1973 Psychology, clinical I University Microfilms, A X E R O X Company, Ann Arbor, Michigan I ' /- © 1974 ALICE FRANCES CHANG ALL RIGHTS RESERVED THIS DISSERTATION HAS BEEN MICROFILMED- EXACTLY AS RECEIVED. UNIVERSITY O F SO U TH ER N CALIFORNIA TH E GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA S 0 0 0 7 This dissertation, written by Alice Frances Chang under the direction of h.&r.. Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillment of requirements of the degree of D O C T O R OF P H IL O S O P H Y Dean DateJ.j DISSERTATION COMMITTEE Chairman To all of those who have counseled, guided and supported me through the years of struggle to reach the threshold of my life's work. ACKNOWLEDGEMENTS I would like to thank the people who helped me in my data collections: Robert Arkin, Rex Beaber, Pamela Berg, Peter Chang, Robert Itatani, Michael Jarvis, Beverly Longstreet, Kathleen McCue, Laura Martin, Craig Muir, A1 Muscatel, and Ruth Ellen O'Brien. I wish to give addi tional thanks to those who helped me with the coding of the voluminous amounts of data: Suzy Liu, Laurie Ann, Susan and Warren Martin, and Art Wells. For my consultants in statistical matters I wish to thank Dr. Stephen E. Berger, Dr. David Burhans, and Mr. Alan Miller. I would also like to express my appreciation to the members of my committee who made every effort to be of help and assistance. In each person's graduate career there should be one person who rises above the others in assisting, guid ing, supporting, caring and sharing with the student in times of hardship, disappointment, achievement, joy, peace, and happiness. I wish to especially thank my chair man, Dr. Stephen E. Berger, for making my final years in graduate school pleasantly memorable, providing a growth environment for me both intellectually and emotionally, and for being that one special person who rises above, whom every graduate student needs as he shapes his future. In addition, I wish to thank those people who have expressed unending faith in ray ability to continue to grow from each experience both academically and personally: Kenneth A. Blinn, Barbara A. Henker, Harold Stone, and Howard E. Wilkening. TABLE OF CONTENTS Page LIST OF TABLES.................................... vii LIST OF FIGURES.................................. x£ Chapter I. INTRODUCTION.............................. 1 Empathy: Definition and Measurement Empathy: Relationship to Therapy Outcome Empathy: Relationship to Student Achievement Empathy: Relationship with Liking Student Personality: Self-Esteem Student Personality: Defensiveness and Social Desirability Student Course Evaluations Teaching Assistants The Present Study Structure of the Class Group Sampling Overall Grading System Teaching Assistants' Empathy Hypotheses II. METHODS.................................. 55 Subjects Procedures Measuring Instruments Objective Measurement of TA Empathy Students' Perception of TA Empathy Students' Course Evaluation Independent Assessment of Achievement Independent Assessment of Intellectual Functioning Personality Instruments Student Self-Esteem Student Defensiveness or Need for Social Approval Like Scale -V Chapter Page III. RESULTS 69 Academic Achievement Student Course Evaluations Liking Effects Intellectual Functioning Correlations Among the Variables Analysis of Covariance IV. DISCUSSION................................ 117 Evaluation of Hypotheses Relationship of Empathy to Academic Achievement and Course Evaluation Relationship of Student Personality to Academic Achievement and Course Evaluation Interaction of Student and Teacher Learning in the Classroom Limitations of This Study and Impli cations for Future Research APPENDICES 151 REFERENCES 165 vi LIST OF TABLES Table Page 1. Analysis of Variance of the Effects of Objec tive Empathy, Perceived Empathy, Defensive ness, and Self-Esteem on Academic Achieve ment as Measured by Performance on Course Examinations.............................. 70 2. Means and Standard Deviations of Objective Em pathy with Short Answer, Essay, Total Written Score, Total Exam Score, and Exam Grade.................................... 72 3. Means and Standard Deviations of Student Perceived Empathy with Multiple Choice, Short Answer, Total Written Score, Total Exam Score and Exam Grade................ 73 4. Means and Standard Deviations of Student's Defensiveness with Total Exam Score .... 74 5. Means and Standard Deviations of Student's Self-Esteem with Multiple Choice, Total Exam Score, and Exam Grade................ 75 6. Means and Standard Deviations of the Inter action Effects of Student Perceived Empathy and Defensiveness with Essay Score.................................... 75 7. Means and Standard Deviations of the Inter action Effects of Student's Defensiveness and Student's Self-Esteem with Short Answer Scores. . .............................. 77 8. Means and Standard Deviations of the Triple Interaction Effects of Objective Empathy, Perceived Empathy, and Defensiveness with Short Answer Scores..................... 77 9. Means and Standard Deviations of the Triple Interaction Effects of Objective Empathy, Perceived Empathy, and Self-Esteem with Short Answer Scores..................... 81 vii Table Page 10. Analysis of Variance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness, and Self-Esteem on Student Course Evaluations as Measured by Factors from the Course Comments Questionnaire . . 85 11. Means and Standard Deviations of Objective Empathy with Skill as Reflected in TA Knowledge................................ 88 12. Means and Standard Deviations of Perceived Empathy with Skill/Knowledge, Interpersonal Relationship, Course Difficulty, Skill/ Organization, Global Learning, Specific Learning, and Course Relevance .......... 89 13. Means and Standard Deviations of Defensive ness with Students' Perception of Global Learning................................ 90 14. Means and Standard Deviations of Self-Esteem with Students' Perception of Specific Learning ........................ 92 15. Means and Standard Deviations of Objective Empathy and Defensiveness with Interper sonal Relationships and Skill as Reflected in TA Organization...................... 92 16. Means and Standard Deviations of Perceived Empathy and Defensiveness with Specific Learning............ 94 17. Means and Standard Deviations of Objective Empathy, Perceived Empathy, and Defensive ness with Student-Teacher Interpersonal Relationships............................ 94 18. Means and Standard Deviations of Objective Empathy, Defensiveness, and Self-Esteem with Students' Perception of Course Difficulty.............................. 96 19. Analysis of Variance of the Effects of Objective Empathy, Perceived Empathy, De fensiveness, and Self-Esteem on the Like Scale and Intellectual Functioning .... 97 viii Table Page 20. Means and Standard Deviations of Perceived Empathy with Liking...................... 97 21. Means and Standard Deviation of the Quadruple Interaction Effects of Objective Empathy, Perceived Empathy, Defensiveness and Self- Esteem on Liking (the Means are Rank Ordered from Low to High)................ 98 22. Means and Standard Deviations of Defensive ness with Intellectual Functioning .... 100 23. Means and Standard Deviations of Perceived Empathy, Defensiveness, and Self-Esteem with Intellectual Functioning............ 100 24. Summary of Selected Correlations of Objec tive Empathy, Perceived Empathy, Intellec tual Functioning and Liking with Academic Achievement.............................. 102 25. Summary of Selected Correlations of Objec tive Empathy, Perceived Empathy, Intellec tual Functioning and Liking with Course Evaluations.............................. 103 26. Summary of Analysis of Covariance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness, and Self-Esteem with Academic Achievement as Measured by Performance on Exams................... 106 27. Summary of Means and Adjusted Means of the Effects of Objective Empathy, Perceived Empathy, and Defensiveness on Total Exam Scores.................................. 108 28. Summary of Analysis of Covariance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness and Self-Esteem with Course Evaluation................... 109 29. Summary of Means and Adjusted Means for the Interaction Effect of Perceived Empathy, Defensiveness and Self-Esteem on Multiple Choice.................................. Ill i*. Table 30. 31. Summary of Means and Adjusted Means for the Interaction Effects of Objective Empathy, Perceived Empathy and Defensiveness with Academic Achievement Variables, Multiple Choice, Written and Total Scores ........ Analysis of Covariance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness and Self-Esteem on Liking. . xx Page I 114 * 116 LIST OF FIGURES Figure Page 1. Graphic Presentation of the Three-Way Interaction Effect of Objective Empathy, Perceived Empathy, and Defensiveness, on Short Answer Questions.............. 78 2. Graphic Presentation of the Three-Way Interaction Effect of Objective Empathy, Perceived Empathy, and Self-Esteem on Total Exam Score . . .................. 82 3. Graphic Representation of the Relationship of Objective Empathy and Perceived Empathy with Academic Achievement Variables.............................. 83 4. Graphic Presentation, of the Effects of Perceived Empathy, Defensiveness, Self- Esteem on Academic Achievement as Represented by Multiple Choice ..... 115 xi CHAPTER I INTRODUCTION In the past decade and a half there have been In creasing research and concern in the area of academic course evaluation, especially at the college and university level. A few studies have pointed out the importance of teacher-student interaction or student feelings about course material. At the same time, growing out of the work of Rogers (1957) there have been studies such as those by Truax and his associates (1966, 1967) and the more recent work of Carkhuff (1967, 1969) regarding facilitative condi tions of "therapists" (i.e. teachers, parents and other helpers) and their relationship of these conditions to "clients'" (i.e. patients', students') behavior change. These studies have suggested the importance and relevance of empathy, respect and genuineness to the helping rela tionship (Carkhuff & Berenson, 1967). Much of the work regarding students and empathy has been conducted on the lower grade levels. Usually, outside raters judge the amount of empathy "radiated" by the therapist or teacher from segments of video-tape or audio-tape. Achievement has been measured as an outcome variable in these studies in addition to grades. 1 2 There has been very little empirical research re ported on the college level relating an independent assess ment system of achievement and learning to variables such as: empathy as rated by the students, evaluation of course subject matter, instructor skill and knowledge, and student perception of his own learning. In addition, when examin ing the amount of student outcome in terms of achievement (grades), a natural inclination would be to assume that the amount of liking or a single "liking" variable accounts for a great amount of change (i.e., if the student liked his teacher and/or his class he would perform better). Other personality characteristics such as student's self-esteem and need for social approval (defensiveness) also affect his performance in school. Further, teaching assistants are often used as instructors, particularly in the begin ning college courses. Yet the value of the teaching as sistant's role has often been overlooked or simply exploited while his facilitative processes have not been examined. The present study is important and necessary in that it offers a systematically designed evaluation of student achievement as related to the classroom environ ment. It explores the relationship between objectively assessed empathy of teaching assistants and student per ceived empathy of teaching assistants in relationship to various types of outcome measures (multiple choice, short 3 answer, and essay questions). In addition, this study is designed to make an assessment of how much a student's "liking" of a teaching assistant is related to grades and to empathy (both as perceived by his students and objec tively rated). This study further examines the effects of the empathy and the liking variables to conventional course evaluations. Also, the student's attitude toward himself is very important to his learning and therefore examinations:, of the relationships of the previous vari ables and the student's self-esteem and need for social approval are included. The process and results of the facilitation of learning or transference of knowledge is of utmost impor tance to the educative process. Since teaching assistants are commonly responsible for teaching undergraduate courses in the university and college systems, it is crucial to clearly demonstrate their ability to facilitate the learn ing process. This study is, therefore, an evaluation of the processes involved in teacher effectiveness using teaching assistants (graduate students) as teachers. This is a primary responsibility of the educational system, even at the university level, to examine its own processes. This responsibility makes it necessary and vital for the future of the university system and for the individual student to examine the variables which will enhance learn ing and facilitate the creative processes in each 4 individual. This study addresses itself to the examination of teacher-student relationships as they affect student learn ing (achievement as measured by grades and exam scores) and to the evaluation of that process in terms of both tradi tional evaluations of course appeal, instructor appeal, and course content, and to the specific amount of knowledge gained as perceived by the student. This study is designed to help define personality characteristics and attitudes of both student and teaching assistant, which will promote mutual growth in the creative process. One of the goals of most graduate schools is to produce adequate teachers, whether or not training experience is directly available to them. This study not only evaluates teaching assistants as teachers, but also is designed to show that teaching assistants can and are willing to participate in their own training program on a more active level. These data can therefore provide information and empirical evidence which can be used to plan a more effective learning environment for the future student based on objective data, as well as providing empirical evidence on which to base a training program for teaching assistants. Empathy: Definition and Measurement According to Pierce and Drasgow (1969) "empathy consists of the ability to recognize, sense and to 5 understand the feelings that another person has associated with his behavioral and verbal expressions, and to accu rately communicate this understanding to him." This con dition which involves both the person's sensitivity to current feelings and his verbal facility to communicate this understanding in a language attuned to the client's current feelings is properly referred to as "accurate empathy" by Truax and Carkhuff (1967). This does not nec essarily mean that one roust share the feelings that would require him to have the same emotions, but rather have an appreciation and awareness of those feelings. At a deeper level, accurate empathy involves one's having enough understanding of human behavior, feelings, and experience that he senses the feelings although the client only partially reveals them. There has been evidence in the literature which has correlated positive outcome in psychotherapy, teaching and child rearing with the personally facilitative dimen sions such as empathy, positive regard (respect), genuine ness, and concreteness (Carkhuff & Berenson, 1967; Truax & Carkhuff, 1967; Pierce & Drasgow, 1969). Berger (1969) has referred to these variables as style, which refers to the "therapist's manner of expression and his own mode of being in the therapeutic situation [p. 1]." In a recent review article Truax and Mitchell (1971) have referred to these variables as "interpersonal skills." 6 Several doctoral students in psychology and educa tion recently selected the variable of empathy to study. For example, Johnson (1971) found that, in adult learners, empathy and trust, as defined by Truax and Carkhuff (1967), were independent of each other when video tapes were rated by other students. Lawson (1971) also conducted a study in which students were to judge, via video tape, how teachers related to students on a teacher relatability scale. The sample was composed of 50 teacher interns and approximately 100 high school students. They reported that teachers who received a significantly higher score on "ability to re late" to students had many of the qualities one would find in a teacher with high empathy, while those rated lower on "ability to relate" to students had qualities commensurate with low empathy teachers. Three trained adults subse quently and independently rated the same video tapes and reported similar findings. Ingwell (1971) has shown in his study that the per ception of clients on the Truax-Carkhuff facilitative con ditions is a function of charismatic variables and/or of the external counseling conditions. The clients themselves were the subjects and they saw several "therapists" (who were trained to assume a certain role with the client) whom they rated after each session. These studies support the contention of Berger (1969) and Truax and Mitchell (1971) that empathy is a reliable personality characteristic. 7 In order to make an objective evaluation of the teaching assistant's accurate empathy, raters must be trained to discriminate between greater and lesser degrees of empathy as reflected in typed script and/or tape record ings (Carkhuff & Truax, 1965). In the rater training con ducted by Truax and Carkhuff (1967), the prospective raters were exposed to therapy excerpts selected by a variety of raters. These excerpts ranged across the various levels of the scales involved in order to insure a spread in the therapy process levels. The prospective raters were as signed to rate these according to the given criteria. The initial training sessions introduced the rater trainee to the research scale developed to measure accurate empathy, giving him theoretical background and practice in rating the samples mentioned above and teaching him to discrimi nate operationally between relative levels of emphatic un derstanding. Truax and Carkhuff (1967) , Carkhuff and Ber- enson (1967) and Truax and Mitchell (1971) have all cited studies in which they were able to train raters who main tained a high degree of inter-rater reliability among many different samples. This demonstrated that not only can training be successfully achieved, but also gives vast im plications for its use in training programs. In fact, with respect to therapy, Carkhuff and Berenson (1967) have dem onstrated that this training could take place in 100 hours or less of training using the previously described method. Empathy: Relationship to Therapy Outcome Carkhuff and Berenson (1967) cite evidence which strongly suggests that the variables relevant to success in psychotherapy are equally relevant to all human inter personal encounters. Therefore, a brief review of the psychotherapy literature on the empathy variable must pre cede the more specific review of the literature of the re lationship of the empathy variable to the educative process and consequent outcome. In reviewing studies on therapeutic outcome, it can be seen that theoretical orientation of the rater can make a difference in ratings. This is best demonstrated in the Hart and Halkides research cited by Truax and Carkhuff (1967) which showed that a less client-centered orientation (i.e. a therapeutic process which did not focus on client feelings of the rater) tended to diverge from findings that these facilitative conditions enhanced positive outcome. This was consistent with those findings reported by Roger (1959) when he outlined his theory and suggested that there was no evidence which would flatly deny or contradict the predictions from his theory. He cites research by Carr (1949) and Bruromond and John (1954) who had pre- and post therapy analyses of projective tests (Rorschach and The matic Apperception Test) completed by diagnosticians. They found little or no change in the degree of adjustment in the projective material. 9 In a series of 10 cases, the John ratings as dis cussed by Vargas (1954) had a significant negative corre lation with counselor ratings. When these same materials were analyzed "blind" by therapeutically oriented research ers, positive change was found, and the correlation with counselor ratings was significantly positive. An explanation for these seemingly contradictory statements was offered by Vargas (1954). He suggested that the diagnostician tends to think of adjustment as an open ness to experience, a more fluid expressiveness and adap tiveness. Therefore, what the diagnostician perceives as loss of control or even disorganization may be perceived by the therapeutically oriented person as progress toward a reduced defensiveness and greater openness to experience. According to Rogers (1959) the incompatible evi dence is not from research but more from a clinical point of view. "Very briefly stated, the Freudian group, on the basis of its experience, tends to see the individual is 'innately destructive' and hence in need of control [p. 248]." Rogers further points out that to hypothesize self-control to be natural to a person who is without de fenses is not tenable in this theoretical framework. This concept suggests that if one has no self-control, he pos sibly has no defenses. If he does not have defenses, then he is not open to experiential learning which is part of the basis for client-centered therapy. Rogers points out 10 that the concept of accurate empathy grew out of Freudian Theory. Where the followers of Freud became dogmatic in their procedures, Rogers found that expressing the client's feelings point for point (i.e., therapist often acting as an alter ego) thus helping him to gain insight into his feelings and behaviors was more beneficial to the client. Other research cited by Truax and Carkhuff (1967) has shown that in group psychotherapy empathy, warmth and genuineness were significantly associated with the pa tients' engagement in the process of therapy, self-revela- tion, and self-exploration. He found that of the three, accurate empathy and genuineness were the most important for the patient's behavior. Probably the experiments with the most historically important results were those referred to as the Wisconsin Schizophrenic Project. This project began in 1958 at the University of Wisconsin under the leadership of Rogers. The group of patients studied were 16 hospitalized schizo phrenic patients. The Rogers group of experimenters and the Truax group of experimenters, although using the same group of patients, dit not necessarily get the same results (Truax & Carkhuff, 1967). The discrepancies do not appear to be due to patient changes, but rather to experimenter changes. Truax and Carkhuff (1967) report that the re search between 1963 and 1967 using larger samples of thera pists and patients "appear to cross-validate the essential 11 findings that were originally reported, as well as a ma jority of the research reports from the .Wisconsin, program completed in 1962 and 1963 [p. 83]." A brief review of the Carkhuff and Truax findings follow: First, they found that psychotherapists whose pa tients improved on the personality tests administered, rated consistently higher on accurate empathy than those whose personality tests showed deterioration over time (P<.01). Ratings also showed that levels of therapist accurate empathy did not tend to vary throughout the six months of intensive psychotherapy. The patient sample in cluded four highly deteriorated and four improved schizo phrenic patients, as rated by personality tests. A second study included 14 of the 16 schizophrenic patients who were seen from six months to four and one-half years. Objective tape recorded segments from every fifth interview throughout treatment ware used. The raters were five naive undergraduate raters who were trained on a rating tape to a minimum of .50 inter-rater reliability. Comparisons were made between the mean level of accurate empathy offered by the therapist and the personality and behavioral changes observed in the patient. The correla tion between accurate empathy and outcome was .77 (p < .01). A positive correlation was obtained between accurate empathy and a personality change over time score which re flected in an evaluation which included in the MMPI and 12 Rorschach tests to be .48 (p<.05). This criterion was called the diagnostic evaluative measure. When three dif ferent raters had been trained to assess therapist genuine ness, the same data was analyzed for relationship of case outcome and therapist genuineness offered to the patient. The correlation of genuineness with the final Outcome Cri teria (which included psychological test change data, diag nostic evaluations of personality change, and a measure of time actually spent in the hospital since initiation of therapy) was .66 (p<.01). With the diagnostic evaluative measures (the blind evaluation made by expert diagnosti cians on the basis of change as seen in pre- and post administrations of the MMPI and Rorschach) correlation between their diagnostic evaluation and genuineness was .45 (p <.05). An identical procedure was used to train four more different raters in assessing the qualities of nonpossessive warmth (unconditional positive regard). For Final Outcome Criteria, the correlation was .73 (p <.01). Using diagnostic evaluative criteria, the correlation was .47 (p< .01). These results support the contention that these three therapeutic conditions tend to show similar relationships and suggests that there is a high intercor relation between the measures themselves. Truax and Carkhuff (1967,). in reviewing their findings show that accurate empathy measures correlated .54 with non possessive warmth measures and .49 with measures of therapist genuineness. Warmth and genuineness corre lated .25, indicating that between 6 and 30 percent 13 of the variation in one measure is due to the varia tion in another measure.[p. 86]. Spotts and Wharton (cited by Truax & Carkhuff, 1967) used the same patients from the Wisconsin Schizo phrenic Project for their analysis and found that the successful cases in therapy had received significantly higher levels of positive regard throughout the first 30 sessions of therapy when compared to the failure and in determinant outcome groups. Gendlin and Geist (cited by Truax & Carkhuff, 1967) used similar procedures to those of Spotts and Wharton in examining the effects of therapeutic genuineness. They suggested that the measurement of genuineness alone may be only an indication of deleterious effects when it is low. They found that instances of extremely low genuineness in the therapist tends to invalidate the effects of higher levels of other conditions offered by him, i.e., the low amount of communicated genuineness tends to cause the client not to perceive any of the other conditions. In another study using hospitalized and outpatient samples from the Stanford University and the University of Chicago, Truax found that the level of accurate empathy offered in therapy was significantly higher for successful cases than for failures (p c .01). He further found when taping full sessions and examining them for high and low periods of accurate empathy that a therapist is more 14 helpful by striving for deeper understanding, even at the risk of occasional misunderstanding; that occasional low moments of accurate empathy had no relationship to out come; and that outcome is affected by both the average levels and the very highest moments of accurate empathy. Cartwright and Lemer (1963) found that final ob jective empathy rating, not initial enpathy rating (level of emphatic understanding) of patients was related to im provement in therapy, after conducting systematic research on therapists' level of accurate empathy across time in therapy. This study suggests that it is the change in therapist's amount of empathy, rather than patient's per ception of accurate empathy that is the most critical variable. Another investigator, Van der Veen (1967), also examined the relationship of therapist facilitative condi tions over time. His population consisted of 10 therapists who submitted a total of 15 therapy cases of schizophrenic patients. A combined outcome criteria score was obtained from the MMPI, Rorschach, T.A.T., self-concept Q Sort, WAIS, and a 246 item anxiety scale. All patients also completed a therapist relationship questionnaire. Analyses showed that approximately as many cases had positive as negative slopes over time in therapy. Case outcome was highly significant to therapist empathy, slightly related to congruence and not related to positive regard. It should be noted, as in the previous study, that 15 significantly more therapists of successful cases obtained high levels of congruence and empathy, yet case outcome was not related to patient's perception of therapist. Combs and Soper (1963) reported significant corre lations between supervisors' judgments of effectiveness of counselor and aspects of the counselor's attitudinal ori entation suggesting convergence of the facilitative dimen sions of empathy and warmth. They point out that effec tive counselors tended to assume the internal rather than the external frame of reference with others to be more people oriented. In a study examining analytically oriented thera pist's warmth, Strupp, Wallach, Wogan and Jenkins (1963) found substantial correlations between how therapist rated patients' outcome (improved or not) and the patients' own ratings of specific, attitudinal variables. They found significant relationships, between the therapist rating of patient's outcome and patient's perceived warmth of thera pist and liking of the therapist. In a similar study which examined psychiatrists and analysts and their patients, using an attitudinal questionnaire survey, Strupp, Wallach and wogan (1964), report the following: The emergence of a general "warmth" factor deserves emphasis. There was additional evidence to suggest that overshadowing this attitudinal emotional factor is the patient's conviction that he has the therapist's respect. This faith in the integrity of the therapist 16 as a person may be called the capstone of a successful therapeutic relationship subsuming other character istics. Technical skill on the part of the therapist may go a long way to capitalize on such a relation ship, although the present data do not specifically inform us how such a relationship comes into being, is deepened, and turned to a maximum therapeutic ad vantage. However, there is little doubt that a re lationship having these qualities represents the most basic ingredients of beneficial therapeutic influence, irrespective of the formal aspects of the setting Ipp. 36-37]. Strupp, et al. (1964) Concluded that it was defi nitely important to examine attitudes of both parties (therapist and client) with respect to outcome, and that the process needed more in-depth analysis. They reported that warmth and genuineness were both necessary for positive therapeutic outcome. Using 523 male neurotic, psychotic and other V.A. patients who were asked to fill out questionnaires, Love and McNair (1966) found that their factor of therapist understanding (accurate empathy) correlated .31 with pa tient rating of self-improvement, .19 with therapist rating of the patient's improvement, and .30 with patient's judged satisfaction with therapy. For therapist acceptance (warmth) the correlations were .24, .16, and .24 respec tively (all p<.01). For factors concerning low empathy and low warmth they obtained negative correlations with the judgments of therapeutic outcome. Thus, this study supports the contention that empathy and warmth are defin able and effect therapeutic outcome. Also, it appears 17 that therapists having low levels of empathy and low levels of warmth may even be therapeutically harmful to their patients. A major study reported by Truax, Wargo, Frank, Imber, Battle, Hoehn-Saric, Nash and Stone (1966) indicated greater improvement for patients who had therapists who offered high levels of accurate empathy, nonpossessive warmth, and genuineness, than patients receiving relatively lower levels of the combined conditions. Their data fur ther suggested that therapy can be better or worse depend ing on the level of therapeutic conditions offered by the therapist. When accurate empathy, nonpossessive warmth, and genuineness were analyzed separately, the data indi cated identical findings for empathy and genuineness, but a reversed tendency for nonpossessive warmth. They explain this in terms of the interrelationships between the three conditions themselves; i.e., the ranking of therapists on empathy and genuineness was identical (r=1.00), while non possessive warmth was negatively related to other condi tions (r=-.40). On the basis of this data Truax (1967) has suggested that when one of the three conditions is negatively related to the other two in any given sample of therapists, then patient outcome is best predicted by whichever two conditions are most closely related to each other. This would suggest the hypothesis that when two out of the three therapeutic conditions are sufficiently high, 18 positive patient or client change will occur. Ah attempt at measuring the generality of the facilitative conditions was made by Truax, Carkhuff and Kodman (1965) using 40 hospitalized patients (considered to be chronic patients) in group psychotherapy. The group met twice weekly for three months. The MMPI was admin istered at the beginning and end, this being used as the outcome measure. Patients were divided into groups using splits of high and low accurate empathy, genuineness, and nonpossessive warmth. There was a strong positive corre lation between empathy and warmth (r=.91) while genuineness was negatively correlated to both of these conditions. This would strengthen the sentiments of the previously men tioned study which indicated that if two of the facilita tive conditions are highly related, but the third is nega tively related, the predictions of outcome should be based on the two that are most highly related. In both studies the two negatively correlated conditions were not the most significant variable for either of the groups, while ac curate empathy was the most related to outcome. This find ing again suggests that when one of the therapeutic condi tions is low, but not significantly, it may not interfere with the process of the other two conditions. Another related unpublished study was conducted by Truax, Wargo, and Carkhuff (cited in Truax & Mitchell, 1971) in which the investigators attempted to replicate one 19 of their earlier unpublished studies which had involved 80 juveniles were seen on an outpatient basis rather than in patient. They found that patients receiving high levels of all conditions demonstrated improvement on 21 outcome variables and no improvement on two of the outcome meas ures. They found opposite results (using the same 23 outcome variables) from the patients who received relative ly low levels of the therapeutic conditions. Again, in this study accurate empathy, nonposses sive warmth, and genuineness were analyzed separately. The data suggested that nonpossessive warmth (p <.001), gen uineness (p< .001) and accurate empathy (p < .05) were important in predicting outcome for outpatients. This study, unlike other studies, suggested that there was very little deteriorative effects of low therapeutic conditions from outpatients. The authors suggest that since spon taneous improvement is relatively frequent with outpatients, low conditions might be expected to eliminate such spon taneous improvement and thus not result in absolute de terioration . A few control studies were reported by Truax, Wargo and Silber (1966) , Dickenson and Truax (1966) , and Carkhoff and Truax (1965) which basically showed that those patients who were in the groups with therapists who were judged to be high in facilitative conditions had greater improvement or more positive outcome in therapy, suggesting 20 again that therapists exhibiting these conditions are more effective. In the most recent studies it appears that re searchers have tended to accept the three facilitative conditions as predictors of positive outcome, Attempts have been made to show that type of therapist and type of patient may be relevant and important to the prediction process, however, outcome depends on the theoretical ori entation of the person designing the study (Beutler, Johnson, Neville, and Workman, 1972; Calder, 1971). There have been two studies reported by Truax and Mitchell (1971) which show an extension of the experimental design to student counselors (Wyrick and Mitchell, 1969) and college instructors (Wagner and Mitchell, 1969). The Wyrick and Mitchell study involved forty undergraduate dormitory counselors. One male and one female student were trained to act as a resident with a serious problem. The counselors were instructed to be as helpful as possible within the limits of their usual responsibilities. Again, ratings of tape recordings were made by clinical psycholo gist with five years of psychotherapy experience and half of the tapes were rated by education administration ad vanced students who had no prior experience in therapy or counseling. Inter-rater reliability was established at r*=.58 for empathy and r=.67 for positive regard and only r=.14 for genuineness. 21 Even after discussion on genuineness among the students and the psychologist the reliability on genuine ness was not raised. There was a significant correlation (p <.05) between students' perception of counselor effec tiveness and. counselor positive regard while the relation ship between empathy and perceived counselor effectiveness was significant at the .01 level. One might suggest that since accurate empathy gen erally has the lowest correlation of the three major facilitative conditions to outcome that it is less crucial. However, a closer examination of the conglomeration of findings show that, although accurate empathy is the least correlated with some outcome measures (p<.05) its positive findings remain constant across studies and measures while warmth or genuineness tend to fluctuate and to be either positively or negatively correlated to the remaining two conditions. From the literature thus far, it is apparent that there are at least three of the five Rogers' facilitative conditions which effect change in human behavior— accurate empathy, positive regard (warmth), and genuineness. Truax and Carkhuff (1967) have demonstrated that these variables are intercorrelated and are necessary and sufficient condi tions for therapeutic outcome. This review has suggested that the most reliable variable is that of empathy. The literature further indicates that more research must be 22 conducted using the measure of accurate empathy as pre dictor of academic success, at the college level. Empathy; Relationship to Student Achievement Re miners (1959) has suggested that teachers are evaluated by the university administration for the changes which they bring about in students and/or for those as pects of themselves (knowledge, skills, and attitudes) that are assumed to be related to their effectiveness in bring ing about desirable changes in the students. He suggests that teaching and learning represent a highly complex com munications system which must be explored and evaluated, rather than simply using teacher knowledge as the sole evaluative criterion. The previous findings which have' been applicable to psychotherapy have been generalized to research on stu dent academic achievement. These findings suggest that a teacher's ability to sense, recognize, understand, and communicate the understanding of problems, feelings and thoughts to students is important in facilitating students' learning (Pierce & Orasgow, 1969). It is these factors which appear to be most positively related to achievement. This process of being able to sense, recognize, understand and to communicate the understanding of problems, feelings, and thoughts to students is called empathy (Truax & Mitchell, 1971). The following literature review will 23 present evidence which supports the contention that erapa- they is an important factor when examining the educative process with outcome being student learning as defined by a variety of independent and nonindependent measures. An early study (Bendig, 1953) using Introductory Psychology students' achievement and course evaluation hy pothesized that there were two dominant factors, instructor empathy and instructor competence, which influenced the student's performance as measured by grades at the end of the term. He used the Purdue Rating Scale for instruction ratings regarding instructors' attitudes toward students to make up his factor-instructor empathy. He found that the students' mean judgments of instructional competence are highly and negatively related to the course achievement of students and that there was no relationship between student achievement and ratings of an instructor's erapa- thetic attitude toward his students. The findings from this study suggest that when empathy is defined as per ceived attitude of the teacher, this attitude is not corre lated to achievement, as measured by course grades. Aspy (1971) tried to systematically explore the relationship of empathy, positive regard, and genuineness to teaching effectiveness in terms of assessing the levels of these conditions that teachers were communicating and to determine the differential effects of these conditions upon cognitive growth. In this study six third-grade 24 teachers were randomly selected and out of each of their classes were chosen 5 boys and 5 girls with the highest I.Q. and 5 boys and 5 girls with the lowest I.Q. in the class. The subjects were administered 5 subtests of the Stanford Achievement Test. The teachers recorded 2 hours of their class interactions (reading lessons) and 8 4-minute segments were selected randomly from each teach er's performance. Trained raters (those who had been trained to rate counseling and psychotherapy sessions) rated the tapes for levels of empathy, positive regard, and congruence provided by the teacher. Empathy was as sessed according to the Truax-Carkhuff (1967) 9-point scale. Aspy (1971) found that the levels of empathy, posi tive regard, and congruence provided by the teachers in their actual classroom procedure related positively to the cognitive growth of their students, as measured by the Stanford Achievement Test. The Stanford Achievement Test was a group test administered by a group examiner, usually the teacher. That is, students who had teachers who scored higher in empathy, obtained higher achievement scores on the independent assessment instrument. They also reported, that these effects were "washed out"— did not occur— with students at an older age. Stoffer (1970) also reported a study involving emo tionally disturbed elementary school children (mean age 9.3 25 years with range 7.1 to 12.0) and their relationship to female volunteer teachers. The evaluation was made with respect to how the teacher-student relationship affected the results on intelligence tests (Stanford-Binet, Form L-M or Wechsler Intelligence Scales for Children), gains in achievement (as measured by the Wide Range Achievement Test and other achievement tests), and teacher grades. Subjective ratings concerning reduction in behavior prob lems and gains in motivation were made by the teacher. Three advanced graduate students in psychology rated the "helpers" on the three Truax scales— genuineness, positive regard, and accurate empathy as judged by videotapes. The children also rated the helpers and the helpers rated them selves. The results of this study showed that positive regard and accurate empathy were correlated at .70 as rated by the graduate student raters and as self-rated by the teachers. The volunteer helpers and the children were in agreement as to the global level of therapeutic conditions as shown by the children's ratings of the helpers as well as the helpers' ratings of themselves (r«°.62) , although the helpers rated themselves more favorably than did the chil dren. The correlation between positive regard as rated by the graduate students and the child's total score on the relationship rating scale was significant at the .05 level using a one-tailed test. This shows that raters rate helpers differently from the way helpers rate themselves 26 and that children also rated differently from the helpers. These results clearly indicate the necessity for outside evaluative criteria of helper's level of facilitative conditions. Stoffer found that the high levels of positive re gard as rated by the graduate student were significantly related to reduction of behavior problems, and gains in total outcome. Total outcome was a general index of be havior change which was based on all indices of behavior change. High levels of accurate empathy were significantly related to achievement and gains show in the total outcome index. The helpers' self-evaluation or overall perceptions of the relationship approached the .05 level of signifi cance of the combined total outcome only. This would rein force the notion that the helpers1 perception of the rela tionship is not necessarily accurately related to achieve ment, motivation, or change in behavior. The author also pointed out that as a group, the community helpers were providing relatively low levels of the rated conditions as compared to those studies on community helpers reported by Rogers (1957). Stoffer interpreted this as an indication that the elements of the therapeutic relationship are im portant for school children experiencing academic and be havior problems. Most pertinent to the present investigation is the aspect of a study cited by Truax and Mitchell (1971) , which 27 explored the relationship of instructors' and students 1 perception of the instructors' level of accurate empathy, positive regard, and genuineness and the students' final examination scores (Wagner & Mitchell, 1969) . The sub- jects included 29 instructors and 316 students at the University of Arkansas who taught and took five sections of Introductory Business; five sections of English Grammar and Sentence Completion; five sections of English Composi tion, and 14 sections of College Algebra. The College Algebra sections were divided into two parts: those class es composed of only freshman students and those classes with at least 24% upper-classmen. Each student in each section as well as each instructor completed the Relation ship Questionnaire as a measure of the facilitative dimen sions, empathy, positive regard, or genuineness for the instructor. It was found that there is a relationship between these dimensions and final examinations scores. In addition, the results indicate that students' perception of accurate empathy was more important than positive regard or genuineness. However, the instructor's perception of his own genuineness was more strongly related to final examination scores than his self-rating of either empathy or positive regard. Thus for students, the im portant variable is their perception of teacher empathy, but for the instructor, his perception of his genuine desire to be in this situation was a more crucial variable. 28 In an effort to reconcile these results with those obtained by Aspy (1971), Wagner and Mitchell suggested that there is a curvilinear relationship among the facilitative dimensions, grades and age. Apparently facilitative dimen sions may be negatively related to grades for adolescents. With the exception of a couple of studies cited, most of the previous research has been accomplished via video or audio tape with the use of outside judges. The dimensions are mostly judged according to how a trained rater sees the situation rather than how the subjects them selves perceive or view their experience. However, when subjects are used as the raters, positive relationships between facilitative conditions are obtained also as in the Wagner and Mitchell study reported above. A major criticism of the research reported here is that the criteria (grades) are usually determined by the teachers. Thus in the study reported above, the instructor may have been influenced in the final examination situation by his relationship to and knowledge of the students, which may account for the significant relationships obtained. Examining the intercorrelations of these facilita tive conditions as predictors of achievement has been a topic of investigation in recent years. Aspy and Roebuck (1972) present a detailed study of elementary school chil dren in which they examined thirteen variables in assessing the relationship between teacher's classroom behavior and 29 student's level of cognitive functioning. They videotaped one hour of class teaching. The three variables of empathy, genuineness and positive regard were chosen to replicate Aspy's earlier findings that these conditions facilitated cognitive gain (1971). They decided to change the measure ment of cognitive gain from performance on achievement tests to classroom assessment of cognitive gain as assessed by six factors of the Bloom's Taxonomy of Educational Ob jectives (knowledge, comprehension, application, analysis, synthesis, and evaluation). The only significant finding was that teacher's level of positive regard was judged sig nificantly higher for level 2-6 teachers whose students reflected cognitive growth (r=.82, p<.001). Accurate empathy and genuineness (r=.32 and .26 respectively), were not significant at conventional levels of statistical ap plication. The authors suggest that when cognitive gain was measured as an outcome or "product variable such as in the Aspy (1971) study using achievement tests, the inter personal dimensions are more significant than when measur ing learning process such as those reflected in Bloom's Taxonomy. They further suggest that the three facilitative conditions are necessary for knowledge (memory and recog nition as in achievement tests) but that once the student is past that level, only positive regard is important. One criticism of this research might be that only one hour of teaching was used, hardly enough time to establish any of 30 the facilitative dimensions. Since more recent research shows that positive regard or nonpossessive warmth is highly related to liking (Mullen, 1969) it is possible that the measure of cognitive gain for that hour is re lated more to liking than to any of the facilitative conditions. Chase (1971) conducted a study involving students and teachers in secondary school where he examined the re lationship between empathy, warmth, and genuineness with respect to achievement as measured by occupational choice and job satisfaction. He found that there was a signifi cant correlation between amount of facilitative conditions as judged by students and by raters and the amount of satisfaction one felt in his job, the number of years in volved, and hours of school completed beyond the Bachelor's degree. Truax and Carkhuff (1967) have reviewed several articles which would support a hypothesis that lower func tioning students or clients are more affected by the facil itative conditions than their counterparts who are less vulnerable. A partial explanation may be contained in the study by Anderson (1968) in which he noted that the high functioning therapists seemed to confront them with their limitations (weaknesses). Consequently, low functioning clients may have the most need for the particular behaviors exhibited by high functioning therapists. 31 In more recent years, some questions have been raised about the generalizability of the Truax scale to nonclient centered environments. These studies report no results or outcome that are unrelated to the facilitative conditions (Kratochvril, Aspy, & Carkhuff, 1967; Bergin & Jasper, 1969). The results from all studies would suggest that the facilitative conditions are important, but not the only variable needed to determine process and/or outcome in school learning. Carkhuff and Berenson (1967) propose that they account for only 20% of the variance, but suggest that they are necessary but not sufficient, or if they are sufficient, that there may be other variables which might heighten the effect. Empathy: Relationship with Liking An examination of the preceding definitions of ac curate empathy would readily suggest that a theoretical point of view is being reviewed, in that empathy, rather than a unidimensional concept, appears to be a multi dimensional construct. The studies reported on liking and interpersonal attraction at present consist primarily of a series of empirical findings that are not yet adequately bound into an integrated construct. Bersheid and Walster (1969) have suggested that positive regard is evoked in the liking process. They point out that it is one's qualities which attract and are 32 are attracted which gives predictive validity to interper sonal attraction studies. These studies are investigations in which one tries to define one aspect or another of like and dislike for other persons. Major studies on interper sonal attraction have examined empirical type questions which attempt to answer the question, "Why do people like some people and dislike others?” This could easily be translated into the question, "What are the attractive qualities between persons?" In conducting studies on interpersonal attraction the hypothetical proposition that "one's 'attraction' to another is assumed to be a more or less stable character istic which can be detected from the various behaviors in which he engages 1 must be accepted as a giverf' (Berstein & Walster, 1969, p. 3). There have been various methods used to measure interpersonal attraction. These methods include attitude scales (i.e., how strongly does one feel about another), tests of opinions disguised as facts, amount of eye contact (Argyle, 1967) , pupil size (Hess & Polt, 1960), distance one stands from another (Kiesler & Goldberg, 1968), subject's willingness to do favors for another (Bramel, 1969), and through sociometric choices. In all of the above research paradigms, the investigator has felt that he was measuring amount of interpersonal at traction, or "liking," Berschied and Walster (1969) have reviewed articles 33 which clearly demonstrate that "liking" is actually not a unidimensional variable as previous investigators have sug gested. They point out that probably the most frequently used psychological principle to predict interpersonal at traction is reinforcement. People like people who reward them, especially in terms of reducing stress, anxiety, loneliness, or insecurity. Proximity or geographic close ness of one person to another has been another well re searched area. They suggest that "proximity allows one to obtain an increased amount of information about the other person and to experience rewards and punishments from the other"[p. 49J." Also, proximity apart from information giving and apart from the rewards or punishments which the other may administer, may facilitate attraction as a by product of an individual's desire for cognitive consistency (Heider, 1958) . Another dimension which has been examined with re spect to interpersonal attraction has been what Walster and Berscheid consider the "Reciprocity-Liking Rule." In other words, one comes to like a person because he hears or in some way finds out that the other person likes him. When examining this rule, researchers have found that the way in which a person perceives how he is liked (or the reason why he is liked) has a tremendous effect on the amount of liking which is generated by the subject. For example, if the subject is liked by another for something 34 that raises his esteem, he will like the liker more than if the liking is expressed in terms of patronization. Interpersonal attraction or liking studies have been conducted considering another variable, similarity- dissimilarity. Walster and Berscheid (1969) provide evi dence that perceived similarity produces liking and that liking produces perceived similarity. Byrne and his col leagues (1966a, 1966b, 1966c) have demonstrated that inter personal attraction is a positive linear function of the proportion of attitude statements attributed to an individ ual which are in agreement with the attitudes of the sub ject. One of the reasons given for attitudinal similarity yielding "liking" is cognitive consistency, which is de pendent upon the individual liking himself. Festinger's theory of social comparison processes (1954) tends to pro vide some of the insights for this theory as does the research reported by Byrne (1961) when he made the follow ing summary statement: . . . any time that another person offers us vali dation by indicating that his percepts and concepts are congruent with ours, it constitutes a rewarding inter action and hence, one element in forming a positive relationship. Any time that another person indicates dissimilarity between two notions, it constitutes a punishing interaction and thus one element in forming a negative relationship. Disagreement raises the un pleasant possibility that we are to some degree stupid, uninformed, immoral or insane [p. 713]. Another reason why attitudinal similarity produces interpersonal attraction is that the attitude often pre dicts another's behavior. Expression of similar attitudes 35 toward something would imply that there are rewards for the personal interaction, while having different opinions about something would suggest that there will be more frustration in the relationship. The research cited above demonstrates that simi larity and liking- sure, reciprocal functions, i.e., as one increases so does the other. These findings are consistent with Newcomb's (1961) which demonstrated that "attraction breeds attraction." There are, of course, other similarities besides attitudinal similarities which seem to produce liking. These include physical characteristics and personality traits. The research literature abounds with studies which select a single characteristic in one person which can be termed similar in another person and find, as would be expected, liking between the two persons. Thus there are many factors which influence the liking variable, and any study reporting relationships be tween liking and achievement scores (grades, etc.) must specify which of many factors producing liking produces the relationship being examined. This point can be made an other way. If you don't know what variable determined the liking score, you don't know what "liking" means, because there is not yet a definition (parameters) for this term. From this brief review of the literature concerning liking and/or interpersonal attraction studies, it can be 36 seen that there are many parameters found through the empirical studies of liking that coincide with parts of the definition of accurate empathy. Therefore, it would be expected that there would be a certain amount of liking associated with empathy. The question which has not been answered in previous research is "how much of the liking variable is associated with the accurate empathy variable?" Student Personality: Self-Esteem It should be recognized that students1 personality characteristics also contribute to a major part of their learning process. The amount of students' self-esteem would appear to be one of the most influential personality characteristics on achievement criteria. Ziller, Kagey, Smith, and Lang (1969) after review ing the self-esteem literature concluded that a major dif ficulty with earlier self-esteem research has been the reliance on direct, subjective, self-report measures whose items are blatantly obvious. Berger (1971) points out that "it is likely that many tests of self-esteem measure to a great degree the individual's ability to assess himself ac curately and his motivation to report this assessment, rather than a statement of the individual as he genuinely is, or even as he actually sees himself [p. 8]." This statement suggests that many of the tests currently being used to measure self-esteem, may be measuring aspects of defensiveness. 37 The congruence between expectations and aspirations according to Coopersmith (1967) , differentiates between high and low self esteem in people. He found that the greater the discrepancy between expectation and aspiration, the lower the person's self-esteem, and the greater the congruence between these two variables (expectation and aspiration), the higher the self-esteem. In fact, Crowne and Stephens (1961) referred to this congruence as self acceptance. Thus the congruence between level of expecta tion and level of aspiration has been defined as synonymous with self-esteem or at least as an important component of "the self-esteem construct. Other investigators have pointed out that high self-esteem scores may in some cases reflect a need for social approval and defensiveness (Cohen, 1959; Silverman, 1964). Nisbett and Gordon (1967) reported that they found that the individual with high self-esteem tends to avoid unfavorable information about himself, as did Silverman (1964 a, b). Silverman's investigation showed that sub jects with high self-esteem respond to failure with avoid ance. Jacobson, Berger, and Millham (1970) , have de veloped a measure of self-esteem which seems to have avoided the problem of defensiveness while maintaining the congruence of aspiration and expectation. They conducted a study examining the relationship of the need for social 38 approval and self-esteem to failure avoidant behavior. In this study, subjects were able to avoid failure through cheating during a temptation period. Their sample con sisted of male and female undergraduate students who were administered the Marlowe-Crowne Scale. After they com pleted the questionnaires, they were presented with a modi fied version of the Digit Symbol subtest of the Wechsler Adult Intelligence Scale. The subjects were informed that the average performance for the American college students was 271 squares. At this point, the subject's level of aspiration and expectancy of success on the task were ob tained. The subjects were asked, "How many squares do you expect to complete?" and, "How many squares would you like to complete?" The discrepancy between the numerical re sponses to these two questions provided the self-esteem score. This measure of self-esteem is referred to as the Berger Self-Esteem Scale in other investigations (Chang, 1973; Chang, Birren & Wine, 1973). A small discrepancy in the score was interpreted as high self-esteem and a large discrepancy was interpreted as low self-esteem, as suggested by Coopersmith (1967) and Strickland and Crowne (1963). This measure is unique in that, although a self-report measure was used, a culturally sanctioned response to the items was not clear to the sub ject. Therefore, self-esteem was not determined on the basis of a score derived from a self-report inventory in 39 which socially desirable responses are clear. Using this type of measure would suggest that the self-esteem measure ment would be independent from need for social approval. In a study examining the self-deceptive personality, Berger (1971) examined the relationship between the Berger Self- Esteem Scale. He found that there was no correlation be tween his measure of self-esteem and those other measures. However, the Coopersmith measure was highly correlated with defensiveness (r=.44) and negatively related to trait anxiety (r=-.59). Therefore, this suggests that the Berger measure of self-esteem is a pure measure and is not confounded by social desirability and defensive denial, whereas a standard scale, such as Coopersmith's, is highly confounded. Berger’s research (Berger, 1971, Jacobson, Berger & MiIlham, 1970) has indicated that a combination of defensiveness and self-esteem is related to failure- avoidant behavior. Student Personality: Defensiveness and Social Desirability In constructing a social desirability measure which is unconfounded by psychopathology, Crowne and Marlowe (1960) developed a need for social approval scale. They developed this scale in an effort to clarify the high cor relation found in the Edwards Social Desirability Scale (ESD) with those scales on the Minnesota Multiphasic Per sonality Inventory (MMPI) which were related to psychotic 40 responding (Edwards, 1957; Edwards, 1970; Edwards & Walker, 1961). Therefore, the Marlowe Crowne Social Desirability Scale was an attempt to separate social desirability from psychopathology. From their studies, Crowne and Marlowe concluded that persons high in need for social approval were more sensitive and attuned to the norms of society. They sug gested that this awareness and sensitivity to social norms enabled the persons high in need for social approval to avoid disapproval through conformity. On later studies, Marlowe and Crowne (1964) reconceptualized their theory and reported that results from their investigations showed that the conformity of high need for social approval scorers represent "avoidant, self-protective behavior in anticipa tion of social rejection and threat to self-esteem [p. 165)." High need for social approval subjects were viewed as re pressors and the need for social approval variable was re conceptualized into a defensive style which attempted to protect self-esteem. Jacobson and Ford (1966) questioned this later conceptualization of Marlowe and Crowne (1964). They pointed out that the sensitization construct implies ap proach motivation, but the defensive construct implies avoidance motivation. They questioned whether it made theoretical sense to view the same person as both a re pressor and as a sensitizor simultaneously. 41 Jacobson and Ford predicted that high need for social approval scorers would be more sensitive to cultural norms than low need for social approval on the basis of the sensitizor hypothesis. On the basis of the repressor hy pothesis , they predicted that low need for social approval scorers would be more sensitive than high need for social approval scorers, since the former are less repressive and avoidant in viewing the world. In their investigation, Jacobson and Ford placed high and low need for social approval scorers in a situa tion in which the appropriate social responses were not clear. They found that low need for social approval scorers demonstrated greater sensitivity in social evalua tion than the high need for approval scorers. They, therefore, proposed that the social compliance of high need for social approval Ss was not merely defensive be havior but part of a defensive cognitive style aimed at reducing a strong fear of social disapproval and failure. Another study has provided additional evidence that high need for social approval scorers was less intropuni- tive than low need for social approval scorers (Ford & Hersen, 1967). This result further suggests that high need for social approval scorers deny the responsibility for their own failure. On the basis of these studies, it might be sug gested that in a real academic setting, unlike the 42 laboratory setting, that high need for approval scorers would perform better than low need for social approval scorers because of their need to conform to the socially approved tasks of academia; however, because of their de fensive nature the high need approval scorers would have more difficulty with the highly empathic teaching assistant (i.e., they would not know how to handle an outside au thority figure who tended to recognize, sense, and under stand their feelings, sense they themselves have an extreme need to deny them and only to behave in a socially desir able manner, thus avoiding disapproval through conformity). Support for this proposal can be found in data provided by Strickland and Crowne (1963). They found that high scorers on the Marlowe-Crowne scale tended to drop out of therapy and were rated by their therapists as defensive. Student Course Evaluations In 1960, McKeachie published his fourth edition of Teaching Tips which he had revised extensively from his previous editions. The significant portion of his work for this study is his extensive research and evaluation of teachings. He pointed out the validity or reliability, or both, of using students to evaluate courses (or instruc tors) by citing evidence that demonstrated that students' evaluation of courses did not significantly differ whether they were told it was for the personal use of the 43 instructor, or merely for an overall research evaluation (i.e., no matter which set of instructions he gave to the students, the evaluations maintained the same quality of assessment). Isaacson, McKeachie, Milholland, Lin, Hofeller, Baerwaldt, and Zinn (1964) reported on a more sophisticated type or evaluative questionnaire than previously used by students. The subject pool consisted of Introductory Psy chology students at the University of Michigan. The au thors were interested in more than an evaluation of course content and characteristics of teachers. They wanted to know the dimensions of teacher behavior as perceived by the students. Through factor analysis of 46 items (taken from an earlier factor analysis of 145 items) they obtained six factor groupings which they labeled skill, overload, struc ture, feedback, group interaction, and student-teacher rapport. Prior to this research, there had been very little attention given to student's perception of facilita- tive process in the instructor in relationship to course evaluations. Meredith (1969) reported his findings from a study conducted at the University of Hawaii among undergraduate Speech Communication students. There were 27 instructors (activity group instructors) comprising 60 sections with a total of 1,097 students. The students used a Course Evalu ation Questionnaire and a Student Rating Scale of the 44 Instructor to complete the evaluation. The first question naire consisted of questions pertaining to the mechanics of the course and teaching, while the second scale per tained to perceived attributes of the instructor. They point out that few faulty course evaluations consider the attributes of the learner (student). They found that 64% of the variance was accounted for by two variables, in structor impact (technique) and instructional impact (per sonal attributes of the instructor). It was found that teachers who were perceived as having higher empathy by their students also received more favorable course evalua tions from the students. McKeachie, Lin, and Mann (1971) further expanded on ratings of teacher effectiveness by measuring achievement using the Introductory Psychology students and the Intro ductory Psychology Criteria Test. This test was supposed to put more emphasis on applicability, analysis and various aspects of critical thinking rather than rote knowledge. Effects of intelligence scores were partialed out in the analysis. They found that there was a significant correla tion between skill, feedback, interaction and rapport (positive regard) and the achievement scores. The authors did further analyses on previously ob tained data (McKeachie et al., 1971). However, to their surprise and dismay they found a negative correlation with skill. As in the study reported above, feedback and 45 rapport (positive regard) were positively correlated with effectiveness. Thus, the research reported by Truax and Carkhuff and the research reported by McKeachie indicate that teachers who are rated high in empathy by students produce better learning by students and receive more favor able course evaluation from the students. Apparently stu dents really do respond favorably to learning. Prior to the development of the Psychological Cri teria Test used in the McKeachie, Lin, and Mann (1971) study cited above, these same authors had also conducted a previous experiment using a multiple choice test of knowl edge and an essay examination. They reexamined this data and found that the teachers who tended to be effective in terms of essay criteria were not e-ffective in terms of multiple choice criteria. In addition, positive regard was positively correlated with essay, but not with multiple choice criteria. Thus when the outcome measure requires more personal participation, the facilitative conditions become most relevant. McKeachie, Lin, and Mann (1971) also reported on a study which Mann had just completed using Introductory Economics students. The class consisted of 286 students taught once a week in large sections and three times a week by 18 advanced graduate students (similar to the proposed research design). An attitudinal scale was added to meas ure sophistication in economics. The grade or cognitive 46 achievement score was based on a series of course examina tions stressed in "thinking rather than knowledge." His results showed that rated teacher skill was positively cor related with both attitudinal and cognitive measures of achievement, and significantly so for females. Teachers who were rated as being high on change of beliefs (another factor added for this study, probably measuring flexi bility) were more effective in changing students' attitudes over the course. This series of studies consisted mainly of post hoc analyses and requires extension through a theoretical framework postulated prior to the examination of results. They demonstrate the importance and connection between empathy, students' attitudes, grades, and course evaluation. The course grades and criterion measures were essentially the same in these studies and were always determined by the subject matter of the given course. The third factor of this scale merely points to the personal relevance of the course as perceived by the student. This set of items has been reduced to 35 items through factor analysis and now constitutes the Course Comments Questionnaire (CCQ) used regularly at the University of Wisconsin, Green Bay. Hartley and Hogan (1972) report that students think that different courses have differential effects on their self development, and they suggest that if the students' judg ment of the instructor is important than the judgments of 47 their own self-development should be equally important. Teaching Assistants At the American Psychological Association Annual Meetings in 1969, a group of undergraduate students and their sponsors presented a symposium on the relevance of teaching assistants to the Introductory Psychology classes. This pilot program was at Cornell University and has been utilized since 1966. They pointed out the relevance of innovative ideas and motivation which is brought to the sections by the peer teaching assistant. They point out that little research has been done with undergraduate teaching assistants. Thus, the relationship of teacher empathy to achievement has not been demonstrated with teaching assistants. In addition, teaching assistants do not typically receive any type of evaluative structure on which to base future teaching other than casual observa tions made by a faculty member. They consider this to be a training situation without feedback, which needs to be remedied. In their subjective evaluations, they found that the effectiveness of the teacher on student achieve ment had little to do with the knowledge of the teaching assistant and a lot to do with the charisma and innovative ability. Thus, their results suggest that the relationship of empathy to achievement would be obtained with teaching assistants also in an empirical investigation. 48 The Present Study The present study was a field study designed to examine the relationship (as measured by grades and course evaluations) of teaching assistants1 empathy (both subjec tively and objectively assessed), students' self-esteem and need for social approval to teaching assistant effec tiveness. The students were enrolled in Introductory Psy chology at the University of Southern California, Fall, 1972. Structure of the Class The course was taught by one professor assisted by twelve teaching assistants who were graduate students in psychology. The class met twice per week. During one session, the whole class met in a large lecture hall where they were given a formal lecture by the professor. For their second class period, the students met with one of the teaching assistants in groups composed of approximately 25-40 students per group. Group Sampling At the first class meeting the students were asked to sign up for a discussion group on one of 15 sign-up sheets which listed the day, time, and place of section meetings. The name of the teaching assistant who was to be involved in the group was not listed with the sheet so that students were essentially randomly enrolled (within 49 the confines of their class schedules) to discussion sec tions meeting on a variety of days at varying times. The teaching assistants were randomly assigned to these dis cussion sections within the confines of their class schedules. Overall Grading System The grading of the students in this course was somewhat unique in that the final grades were determined by three separate factors. First was the students' per formance on conventional type examinations which included multiple choice, short answer and essay questions. The questions for the examinations came only from the major lecture each week and from assigned readings in the text book (i.e., teaching assistants could have or could not have covered the same material for which the student was responsible on the examinations). Second, teaching as sistants assigned a grade to each of their students accord ing to their own criteria, and third, the student was al lowed to decide a certain percentage of his own grade. In addition, the student was also allowed to decide the per centage of grade they wished to be assigned by the profes sor, and the percentage to be assigned by their teaching assistant. The combinations from which they could choose were 30-50, 40-40, or 50-30. They had to make this commit raent prior to knowing the results from the first examina- 50 tion, but after taking it. This allowed the student to have some information about his performance, but not enough to make him respond totally to the examination situ ation. Again, this was done to attenuate anxiety. The student was allowed to give himself the other 20% of his grade, using his own criteria for justifying his grade. Making a composite grade such as this, where the student would have some control over his success or failure, was planned so that the student would not feel threatened by his examination performance and would hopefully perform and possibly learn more by decreasing anxiety. Obviously, with at least 30% of the grade determination coming from the examinations, they would play a significant role in the course, but not an overwhelming one for students who hap pened not to do so well on a specific examination. In this study, performance on all three types of examination questions and grades based on those scores were used as the assessment criteria of academic achievement. Thus, the present study is an attempt to examine influences of both the teaching assistant and the student upon aca demic achievement in the actual environment in which these variables operate— the classroom. This study focused on a number of specific teacher and student variables in order to identify highly specific factors operating in course outcome (academic achievement) and course evaluations. An other way to state this is: What happens to students in a 51 classroom, how is it done to them, what about them lets it happen, and how do they feel about it. Teaching Assistants1 Empathy Empathy of teaching assistants was assessed both objectively and subjectively. The subjective measure con sisted of the students' responses on the Relationship Questionnaire (Truax and Carkhuff, 1967) on the empathy factor. The objective measure consisted of three different independent assessments. First, at the first of the semester an outside rater who was familiar with both the teaching assistants and the Truax criteria for finding stages of empathy for therapists, ranked the teaching as sistants from most to least emphati c. He also placed them in high and low empathy categories. Second, prior to the end of the semester, teaching assistants taped their classes to make tape segments available for outside raters to make a rating according to the nine stages of emphatic interaction. Third, the teaching assistants were asked to sort the training examples of the nine stages from least to most emphatic. These three outside or independent measures of empathy were used to measure objective empathy of the teaching assistant. Hypotheses Teachers with accurate empathy, as objectively rated, theoretically would facilitate learning in their 52 students. The first hypothesis therefore assumed that the teaching assistants who were objectively rated high in accurate exapathy would have students who would perform better than students of those teaching assistants who were objectively rated low in accurate empathy. Hypothesis 1. Students of teaching assistants who are objectively rated as being high in empathy will: A. do better on essay type questions than students of teaching assistants who are objectively rated as being low in empathy. (McKeachie, 1971). B. will rate the T.A. higher on the Course Comments Questionnaire when referring to responsiveness or interaction (Factor 2) and relevance (Factor 7) than students with low empathy T.A.'s (Hartley and Hogan, 1972). C. will rate themselves higher in general cognitive development (Factor 5) than stu dents with low empathy T.A.'s (Hartley and . Hogan, 1972). McKeachie and his associates (1971) have suggested that lack of enthusiasm and empathy affect the student's performance on multiple choice type questions less than on other types of written examinations. The third hypothesis 53 is designed to provide objective evidence to evaluate their assumption. Hypothesis 3. The tendency of students who per ceive the teaching assistants as being high in empathy to do better will be more pro nounced on essay type questions than on multiple choice type questions. The literature review has related studies which demonstrated that there is not a perfect correlation be tween the way in which a student or client perceives em pathy in his teacher (therapist) and the actual accurate empathy as rated objectively (Hartley and Hogan, 1972; Isaacson et al., 1964; Meredith, 1969; McKeachie et al., 1971; and Bendig, 1953). The fourth, fifth, sixth, and seventh hypotheses examine the relationships among subjec tive and objective accurate empathy ratings of T.A.'s as they relate to the dependent variables. Hypothesis 4. Students whose T.A.'s are objective ly rated higher and who rate their T.A.'s higher in empathy will: A. do better on essay type questions than students who perceived their T.A. as low in accurate empathy. B. will rate the T.A. higher on the Course Comments Questionnaire when referring to responsiveness (Factor 2) and relevance 54 (Factor 7) than students who perceive their T.A. as being lower in accurate empathy. C. will rate themselves higher in general cognitive development (Factor 5) than stu dents who perceive their T.A. as being lower in accurate empathy. Hypothesis 5. Students who have T.A.'s who are ob jectively rated higher in empathy and who rate the T.A.'s as low in empathy will obtain lower grades than students whose T.A.'s are objec tively rated high and who also perceive their T.A. as high in empathy. Hypothesis 6. Students whose T.A.'s are objective ly rated low on empathy and perceive their T.A. as high on empathy will do better than those students of T.A.'s objectively rated low on empathy and perceived as low on empathy by the student. Hypothesis 7. Students of T.A.'s who are rated low on empathy by objective methods as well as stu dents' perception will have the lowest achieve ment as measured by grades and will yield the lowest course evaluation ratings. CHAPTER II METHODS Subjects The subjects were 447 Introductory Psychology stu dents (249 males and 198 females) at the University of Southern California. Their ages ranged from 16 to 28 . years with a mean of 18 years, 3 months. The majority of the students (61%) were freshmen (N=300). The remainder were: sophomore (N=81), junior (N=35) , senior (N=7) , graduate students (Ns>2) , and information not given (N=22) made up the remaining 39% of the sample. The teaching assistants (TA) were seven male and five female volunteer graduate students in psychology. Five of these (three men and two women) were paid teaching assistants. Each teaching assistant was responsible for one or two discussion sections with one exception— one class was taught by two teaching assistants. Two TA's taught two sections each, eight TA's taught one section, and two TA's taught one section of their own and team- taught an additional section, making a total of 15 sections The students were randomly assigned to their discussion groups within limits of their class schedules. When the teaching assistants were asked to take the 55 56 task of leading a discussion group, it was explained to them that they would be having an experience in teaching for which they would be supervised. It was explained to them, as well as to the students, that these were smaller classes designed to give the students a more intimate con tact with psychology. It was the teaching assistant's re sponsibility to help provide learning and exposure through any means he/she wished. The TA's used several different methods which included formal lecture presentations, stu dent presentations, and/or student paper writing. Some TA's spent time discussing personal problems in a group therapy type situation, while others divided their section into smaller groups who were responsible for group pre sentations which ranged across the span of psychological literature from experimental to clinical research. Thus TA's method of teaching and the TA are intricately inter twined. In addition, if the students requested elaboration or clarification of lecture materials, it was up to the teaching assistant to provide for that need. Questions for the examinations were not taken from the discussion groups unless, by chance, the material covered in the discussion groups happened to overlap that covered in the lecture or in the textbook. Using this method, graduate students were allowed to plan and teach their own classes, with supervision as part of their gradu ate training. In return for their teaching experience, 57 they were asked to be responsible for grading examinations, giving students a part of their grade (T.A. grade) and passing out the necessary questionnaires for the present study. In all, there were 15 groups or discussion sections that were taught by the TA's. The one team taught discus sion section was not included in the data analysis (N=19 students). One other discussion section was dropped from this design because of the inability of the experimenter to get the data necessary for this research on time (N=39 students). Therefore, the total number of discussion groups used for this study was 13. Procedures The Shipley-Hartford C.Q. Scale Abstraction Test, Marlowe-Crowne Scale, and preobjective rating of TA empathy were administered during the first week of the semester. The Relationship Questionnaire (students' subjective rating of TA empathy), Course Comments Questionnaire, Berger Self- Esteem Scale, and the two postobjective measures of teach ing assistant empathy were made during the final weeks of the semester, prior to the final examination. The course examinations were given at regular intervals throughout the semester with the final exam being administered on the scheduled exam date. In all, there were four examinations whose scores were pooled to make up the achievement measure for this research study. 58 The examination room was a large auditorium which seated over 1,500 persons. The students sat in alternate seats so that there was no one sitting on either side of them. They were given an examination booklet containing all the questions and an IBM answer sheet. They placed multiple choice answers on the answer sheet and used the extra sheets of paper provided for the short answer and essay exams. The examinations were monitored by the in structor, experimenter and TA's who could attend at the given time. The third exam which was administered just prior to the Christmas vacation period was given in two forms, on Thursday and on the following Tuesday. The date on which the student took the exam was completely optional to the student. This format was used to accommodate those students who wished to leave early for the holidays. The exams were completely independent of each other, although they covered the same lectures and textbook assignments (i.e., no items appeared on both forms of the test). Coding of the data was also conducted by research assistants and was also checked by another person who had not done the original coding. Measuring Instruments Objective Measurement of TA Empathy Three objective measures of the teaching assis tant's empathy were made. At the first of the semester 59 TA's were rank ordered from high to low in empathy by an outside rater who was familiar with the Truax-Carkhuff (1967) training scales for judging empathy in therapists, as well as familiar with the TA's. This rater obtained a correlation of 1.0 between his ratings of criterion samples from the empathy training tapes (Truax and Carkhuff, 1967) and the actual ratings listed for those samples. Near the end of the semester, the second objective measure was ob tained. Teaching assistants taped their classes and randomly selected segments were sampled for two outside raters to judge using the nine stage scale for empathy, ordering TA's from least to most empathic. These outside ratings made up the objective measure of TA empathy. A description of the nine levels of the empathy scale which is listed in Truax and Carkhuff (1967) follows. Stage 1: Therapist seems completely unaware of even the most conspicuous of the client's feelings; his re sponses are not appropriate to the mood and content of the client's statements. There is no determinable quality of empathy, and hence no accuracy whatsoever. The therapist may be bored and disinterested or active ly offering advice, but he is not communicating an awareness of the client's current feelings [p. 47]. Stage 2: Therapist shows an almost negligible degree of accuracy in his responses, and that only toward the client's most obvious feelings. Any emotions which are not clearly defined he tends to ignore altogether. He may be correctly sensitive to obvious feelings and yet misunderstand much of what the client is really trying to say. By his response he may block off or may mis direct the patient. Stage 2 is distinguishable from Stage 3 in that the therapist ignores feelings rather than displaying an inability to understand them tp. 48]. 60 Stage 3; The therapist often responds accurately to client's more exposed feelings. He also displays con cern for the deeper, more hidden feelings, which he seems to sense must be present, though he does not understand their nature or sense their meaning to the patient [48]. Stage 4; Therapist usually responds accurately to the client1s more obvious feelings and occasionally recog nizes some that are less apparent. In the process of this tentative probing, however, he may misinterpret some present feelings and anticipate some which are not current. Sensitivity and awareness do exist in the therapist, but he is not entirely "with" the pa tient in the current situation or experience. The desire and effort to understand are both present, but his accuracy is low [p. 50]. Stage 5: Therapist accurately responds to all of the clientTs more readily discernible feelings. He also shows awareness of many less evident feelings and ex periences but he tends to be somewhat inaccurate in his understanding of these. However, when he does not understand completely, this lack of complete under standing is communicated without an anticipatory or jarring note. . . . This stage is the mid-point of the continuum of accurate empathy [p. 51], Stage 6: Therapist recognizes most of the client's present feelings, including those which are not readily apparent. Although he understands their content, he tends to sometimes misjudge the intensity of these veiled feelings, so that his responses are not always exactly suited to the exact mood of the client. . . . Although sensing the feelings he often is unable to communicate meaning to them [pp. 52-53]. Stage 7; Therapist responds accurately to most of the clxent's present feelings and shows awareness of the precise intensity of the underlying emotions. However, his responses move only slightly beyond the client's own awareness, so that feelings may be present which neither the client nor the therapist recognize. Stage 7 is distinguishable from Stage 6 in that often the therapist's response is a kind of precise pointing of the finger toward emotionally significant material [p. 54]. Stage 8: Therapist accurately interprets all the client's present, acknowledged feelings. He also un covers the most deeply shrouded of the client's 61 feelings, voicing meanings in the client's experience of which the client is scarcely aware. Since the therapist must necessarily utilize a method of trial and error in the new uncharted areas, there are minor flaws in the accuracy of his understanding, but these inaccuracies are held tentatively. With sensitivity and accuracy he moves into feelings and experiences that the client has only hinted at. . . . The content that comes to life may be new, but it is not alien. Although the therapist in Stage 8 makes mistakes, these mistakes are not jarring, because they are covered by the tentative character of the response. Also, this therapist is sensitive to his mistakes and quickly changes midstream, indicating that he has recognized what is being talked about and what the client is seeking in his own explorations. The thera pist reflects a togetherness with the patient in tenta tive seriousness and depth of his emphatic grasp [p. 55]. Stage 9: The therapist unnervingly responds to the client's full range of feelings in their exact in tensity. Without hesitation, he recognizes each emo tional nuance and communicates an understanding of every deepest feeling. He is completely attuned to the client's shifting emotional content; he senses each of the client's feelings and reflects them in his words and voice. With sensitive accuracy, he expands the client's hints into a full-scale (though tentative) elaboration of feeling or experience. He shows precision in both understanding and in communication of this understand ing, and expresses and experiences them without hesitancy [p. 56]. The third objective measure of teaching assistant's accurate empathy was assessed through the use of the Truax- Carkhuff (1967) empathy training examples of interactions between therapist and client (Appendix A). TA's were given the following instructions: On the following cards are typed several different con versations. Please read each conversation and then rank them from having least empathy to most empathy using the code on the back of the card in the lower left corner. Your first choice would be the one which you would consider to demonstrate the least empathy, 62 while your ninth choice would be the conversation which you consider to demonstrate the most empathy. Please put your name on your rating sheet and seal it in the envelope provided. Thank you very much. The rank order correlation between the preobjective measure and the third objective rating was .83. The pre objective ratings were divided into a high/low split for purposes of data analyses. This split correlated 1.0 with a high/low split based on both the second and third objec tive ratings. Thus, the original objective rating corre lated extremely highly with two subsequent and different methods of objectively ascertaining empathy and was there fore used as the basis for the independent variable of ob jective empathy rating of TA. The inter-rater reliability for the outside raters was .89, with Rater 1 having a 1.0 correlation with the criterion measure and Rater 2 correlating .89 to Rater 1 and the criterion measure. The preobjective empathy rating therefore, constituted an experimental treatment with sub jects being randomly assigned prior to the course. Students' Perception of TA Empathy The Truax-Carkhuff Relationship Questionnaire was used to measure amount of empathy as perceived by the stu dent. This scale measures the central therapeutic in gredients in order to determine therapeutic levels of interaction as perceived by the subject (Truax and Carkhuff 1967). The scale consists of 141 items which are forced 63 choice answer— true or false— of which 46 pertain to em pathy. Only the 46 empathy relevant items were used. These items were answered on the IBM answer sheet and machine scored. The subjects were divided into high and low categories, using a median split on the basis of their perception of accurate empathy in their TA's. The stu dents answered this questionnaire in the end of the semes ter, prior to final exam time. Students1 Course Evaluation Course evaluation and student's perception of self development were measured through the use of the Course Comments Questionnaire (Appendix A) which was modified to be applicable to the individual sections and to evaluate teaching assistants rather than instructor. The specific factors are discussed in the previous chapter. This ques tionnaire was administered at the end of the semester and the answers were placed on an IBM answer sheet for machine scoring. They were scored according to the seven factors, with five items per factor. Independent Assessment of Achievement Measures of achievement were assessed through the use of multiple choice (MC), short answer (SA), and essay questions (E) provided by the experimenter and the in structor. The students were given four examinations during the course. All except the final examination contained 64 all three types of assessment. The final exam was only multiple choice, and was cumulative over the entire course. Multiple choice questions accounted for the majority of points possible on each test. The specific number of items and point values were: Exam 1 multiple choice (MC) 50 items 75 points short answer (SA) 3 items 13 points essay (E) 1 item 12 points Exam 2 MC 50 items 75 points SA 3 items 12 points E 1 item 13 points Exam 3 MC 20 items 60 points SA 1 item 20 points E 2 items 20 points Final Exam MC 100 items = 100 points All the examination questions were written by the professor of the class and the experimenter. The multiple choice questions were answered on IBM answer sheets and machine scored, while the short answer and essay questions were graded, according to a set criteria, by the teaching assistants. The set criteria for grading the written 65 portions of the examinations were determined by the ex perimenter and the professor teaching the class. Exams 1 and 3 were graded by the student's teaching assistant and Exam 2 was graded by another teaching assistant. An outside grader (the instructor) was given a random sampling of essays and short answers from each of the teaching as sistant's groups to grade according to the same set cri teria. The correlation between the outside grader's assessment of the written work and that of the teaching assistants was .99. In fact, the scoring criteria were so clear that there were hardly any differences in total written scores between the TA and the instructor. This showed that teaching assistants were making an honest as sessment of the written work and that one TA was not grad ing more leniently or stringently than another. This is a critical point because any differences to occur on written exams cannot be attributed to differential scoring by T.A.fe. Grades for the exams were determined on the basis of total score on all four examinations. The criteria were A=294 or more, B=192-251, C=182-191, D=163-181 and F=162 and lower. For the purposes of this study, this was the student's grade. The test scores were divided for analysis accord ing to total score received for: (1) multiple choice ques tions (MC), (2) total score received for short answer questions (SA) , (3) total score received for essay 66 questions (E), (4) total written score which was SA+E (WRIT), and (5) the sum total score for the whole exam which was MC + WRIT (TOT). The sum total score as men tioned above was used to assess a letter-grade for each student. Independent Assessment of Intellectual Functioning An initial measure of intellectual functioning was assessed at the first of the semester, through the use of the Shipley-Hartford Verbal and Abstraction Scales which were to be used as a covariate to statistically control the effects of I.Q. score on achievement. The Verbal Scale was machine scored, while the Abstraction Scale was hand scored, by two graders with an intergrader reliability of .98. Personality Instruments Student Self-Esteem The Berger Self-Esteem Scale (1970) was used to assess the amount of self-esteem the student reported. Since the Berger Self-Esteem Scale was originally designed to be used in individual assessments of student's self esteem, the directions had to be modified to be used in group administration. Students were presented a booklet (Appendix B) which contained a coding task similar to that found in the Wechsler Intelligence Scales. The task was 67 called Proficiency Measurement Inventory and the instruc tions read as follows: This is an experiment in proficiency measurement. Notice above that each box has a number in the upper part and a mark in the lower part. Every number has a different mark. (Referring to the samples) In the lower row each upper part has a number, the lower part (Square) is blank. THE AVERAGE AMERICAN COLLEGE STUDENT CAN COMPLETE 271 OF THESE SQUARES IN FIVE MINUTES. Please thumb through the next five pages and give yourself an idea of the task involved. Then turn back and answer the questions immediately preceding the task. DO NOT DO the Proficiency Measurement Inventory task now. Thank you. At this point, the subject's level of aspiration and expectancy of success on the task was obtained. Among the questions on the questionnaire, the subjects were asked, "How many squares do you expect to complete?" and "How many squares would you like to complete?" The dis crepancy between the numerical responses to these two ques tions provided the self-esteem score. Subjects were divided into high and low scorers on the Berger self-esteem measure. Self-esteem was defined as the congruence between level of aspiration and expectancy of success. It should be noted that the smaller the dif ference between an individual's aspirations and expecta tions, the higher the self-esteem. Similarly, the greater the discrepancy between aspirations and expectations, the lower the self-esteem. Numerical values were assessed by two graders who had an interscorer reliability of .98. 68 Student Defensiveness or Need for Social Approvaf The Marlowe Crowne Social Desirability Scale was used to ascertain the student's need for social approval. This is a 33-item questionnaire to which the students answered true or false to statements as they pertained to themselves. The items were answered on an IBM sheet and were machine scored. The subjects were divided into high and low scorers, according to a median split, on the Marlowe-Crowne Scale. Like Scale In addition, students were asked to rate the amount which they liked their teaching assistant on a scale with one being the least anf five being the most. This scale was administered as part of the battery of questionnaires administered in the last weeks of classes. CHAPTER III RESULTS The data was analyzed using a multivariate analysis of variance computer program (Clyde, 1969). Analysis of variance was performed for the independent variables of objective rating of TA empathy (0), student's perceived empathy of TA (P), defensiveness (D), and student's self esteem, (E), and the dependent variables of achievement, course evaluation, liking, and intellectual functioning. In addition, analysis of covariance was performed. Academic Achievement The student's performance on exams was the cri terion used to measure academic achievement. Six cate gories of performance were examined in the analysis. These included the three types of exam questions, multiple choice (MC), short answer (SA) and essay (E). Also included were the three different composite scores: total written (WRIT) which included both types of written questions; the total exam score (TOT), which consisted of the MC and WRIT scores; and the grade assigned for performance on the examinations (GRADE). Table 1 presents the results of the analysis of the effects of objective empathy, perceived empathy, 69 Table 1 Analysis of Variance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness, and Self-Esteem on Academic Achievement as Measured by Performance on Course Examinations Source of Variance Academic Achievement Multiple Choice Short Answer Essay Written Total Grade F P F P F P F P F P F P Objective Empathy (0) 2.22 .14 4.28 .04 17.37 .001 11.24 .001 3.77 .05 6.07 .02 Student Per ceived Empathy (P) 4.67 .03 6.70 .01 1.54 .22 4.51 .04 4.20 .04 5.77 .02 Defensive ness (D) 3.60 .06 .29 .59 2.13 .15 .36 .55 5.41 .02 1.41 .24 Self-Esteem (E) 5.73 .02 .60 .44 .27 .60 .17 .68 5.57 .02 3.78 .05 0 x P 2.56 .11 .99 .32 .04 .85 .77 .38 .56 .46 1.39 .24 0 X D 1.68 .20 1.87 .17 .04 .84 .95 .33 1.89 .17 .44 .51 0 x E .39 .53 .24 .63 .53 .47 .66 .42 .001 .97 .03 .85 P x D 2.84 .09 .16 .69 4.52 .03 2.19 .14 2.43 .12 2.26 .13 P x E .94 .33 .35 .56 .002 .96 .27 .60 1.26 .26 .76 .39 D x E .04 .83 4.10 .04 1.05 .31 2.54 .11 1.65 .20 .06 .80 0 x P x D 2.79 .10 3.76 .05 2.16 .14 2.49 .12 2.13 .15 1.27 .26 0 x P x E 6.16 .01 2.53 .11 8.46 .004 7.74 .006 9.76 .002 7.40 .007 0 X D X E 1.66 .20 1.70 .19 .97 .33 1.68 .20 2.01 .16 .27 .61 P X D X E 1.31 .25 .26 .61 .52 .47 .72 .40 1.13 .29 .35 .55 0 X P X D X E .26 .61 .75 .39 .11 .74 .39 .53 .20 .66 .001 .98 df=l,251 •«j o 71 defensiveness, and self-esteem on the achievement perfor mance scores. Students of TA's who were objectively rated higher in empathy performed significantly better on short answer (p<.04) and essay (p < .001) questions than those students of TA's objectively rated lower in empathy. The means and standard deviations, for the significant effects of objec tive empathy can be seen in Table 2. Thus for the total written task (WRIT) students of high objectively rated TA's performed significantly better them those Ss of the low objectively rated TA-Js (p < .001). The same trend was true for total score (p=.05) and exam grade (p< .01). The means and standard deviations for WRIT, TOT and GRADE are pre sented in Table 2 also. Students who perceived their TAs as high in empathy also performed significantly better on the academic achieve ment measures than those who perceived their TA's as low in empathy. They performed better on five out of the six cri teria: MC (p < .03) , SA (pc.01), WRIT (p < .04) and GRADE (p<.02). The means and standard deviations for these effects are presented in Table 3. The students who were low in defensiveness (D) performed significantly better on the ‘ total exam variable (TOT) , them .those who were high in defensiveness (p < .02) . The means and standard deviations for this effect are presented in Table 4. Table 2 Means and Standard Deviations of Objective Empathy with Short Answer, Essay, Total Written Score, Total Exam Score, and Exam Grade Objective (0) Empathy Academic Achievement Short Answer Essay Written Total Grade M SD M SD M SD M SD M SD Low 29.75 7.32 28.83 7.80 58.74 13.00 256.12 34.05 4.74 .84 High 31.32 5.97 32.57 7.14 63.57 11.32 264.28 40.20 4.97 .80 -o to Table 3 Means and Standard Deviations of Student Perceived Empathy with Multiple Choice, Short Answer, Total Written Score, Total Exam Score and Exam Grade Perceived (P) Means , and Academic Achievement Empathy Multiple Choice Short Answer Written Total Grade M SD M SD M SD M SD M SD High 196.66 27.41 29.37 6.31 59.38 10.71 255.20 34.44 4.72 .79 Low 203.24 27.21 31.47 6.86 62.68 13.20 264.28 39.74 4.96 .84 u> 74 Table 4 Means and Standard Deviations of Student's Defensiveness with Total Exam Score Academic Achievement De fens ivenes s Total Low High M 263.35 257.40 SD 32.72 41.75 Using the Berger self-esteem measure (E), students with high self-esteem performed better than students with low self-esteem on MC questions (p4.02). Since multiple choice questions accounted for approximately two-thirds of the examinations, it would be expected that students who demonstrated high self-esteem would show better performance as measured by TOT and GRADE, than students who demon strated low self-esteem. This in fact, was the case (p < .02 and p=.05 for TOT and GRADE, respectively). The means and standard deviations for these effects are listed in Table 5. A significant interaction effect was found between perceived empathy (P) and defensiveness (D) on performance on essay questions (p <.03). In Table 6, which presents the means and standard deviations of the interaction 75 Table 5 Means and Standard Deviations of Student's Self-Esteem with Multiple Choice, Total Exam Score, and Exam Grade Self- Esteem Academic Achievement Multiple Choice Total Grade M SD M SD M SD Low 196.65 26.61 255.69 38.38 4.77 .79 High 204.43 28.03 265.82 36.05 4.96 .85 Table 6 Means and Standard Deviations of the Interaction Effects of Student Perceived Empathy and Defensiveness with Essay Score Perceived Empathy Defensiveness Academic Achievement Essay M SD Low Low 31.31 7.88 Low High 28.60 7.36 High Low 31.29 7.94 High High 31.40 7.99 76 effects of perceived empathy and defensiveness on academic achievement, it can be seen that those students who per ceived their TA's low in empathy and who were high in de fensiveness performed more poorly than the other three groups on essay type questions. An interaction between defensiveness and self esteem on short answer questions (p < .04) indicates that students high in defensiveness and high in self-esteem perform better on short answer questions than any other group and all other groups appear to perform equally well on short answer questions. The means and standard devia tions for this significant interaction are presented in Table 7. A triple interaction between objective rating of empathy, perceived empathy, and defensiveness was reflected in short answer questions (p=.05) . The means and standard deviations are presented in Table 8. These means are pre sented in graphic form in Figure 1, which indicates that the students whose TA's were objectively rated low in em pathy, who also perceived their TA as low in empathy, and who were low in defensiveness, performed significantly lower than the other groups. In contrast, when these students were students with TA's who were objectively rated high in empathy, their performance was greatly facilitated. Among the high objectively rated TA's those students who perceived the TA as low in empathy and who were themselves 77 Table 7 Means and Standard Deviations of the Interaction Effects of Student's Defensiveness and Student's Self-Esteem with Short Answer Scores Academic Achievement Defensiveness Self-Esteem -------------------------- Short Answer M SD LOW Low 30.53 57566 LOW High 29.432 6.651 High Low 29.961 7.020 High High 32.262 7.297 Means and Table Standard Deviations 8 of the Triple Interaction Effects of Objective Empathy, Perceived Empathy, and Defensiveness with Short Answer Scores Objective Perceived Defensiveness Short Answer M SD Low LOW Low 26.82 6.45 Low LOW High 29.67 7.03 Low High Low 31.04 7.72 Low High High 31.18 7.95 High Low Low 32.19 4.82 High Low High 29.11 6.81 High High Low 31.54 5.71 High High High 32.37 6.53 78 Figure 1. Graphic Presentation of the Three Way Interaction Effect of Objective Empathy, Perceived Em pathy, and Defensiveness on Short Answer Questions. Short Answer 32- 31- 30- 29- 28- 27- 26- High P/High D Low P/Low D High P/Low D Low P/High D -r- L —T“ H Objective Empathy 79 high in defensiveness obtained the lowest scores. A dominant finding emerged using academic achieve ment measures as criteria in the three-way interaction of objective rating of TA empathy, perceived empathy of TA and student's self-esteem. This interaction predicts all the dependent achievement variables except SA which was reflected in the previously discussed interaction (0 x P x D). The objective empathy, perceived empathy and self-esteem interaction is significant on MC performance (p <.01) , E performance (p< .002) , WRIT (p< .006) , TOT (p < .002) and GRADE (p<.007). This interaction is most dramatically and clearly illustrated by the differences in means as noted in Table 9 under the TOT scores. These means are also presented graphically in Figure 2. The students who performed best on this overall exam score were those who simultaneously had objectively rated high empathy TA's, who were per ceived as high in empathy by students who were high in self-esteem. On the other hand, those students who had objectively rated high empathy TA's, but perceived the TA's as low in empathy, yet still had high self-esteem, per formed the lowest on TOT scores. The difference appears to be a product of the Ss' self-esteem in relationship to his perception of TA's empathy as high or low in a TA objectively rated as high. Although there were no significant interactions 80 between objective empathy and perceived empathy using the achievement criteria, all of those students whose TA's were rated as high in empathy performed better on all academic criteria, than those whose TA's were objectively rated high and perceived as low in empathy. These relationships are presented graphically in Figure 3. Planned comparisons were performed as specified in the fifth and sixth hypothesis. A comparison on grades was made between students of TA's objectively rated as high in empathy and who were rated high in empathy and those stu dents who perceived their TA's as low in empathy, but who were objectively rated as high empathy TA's. There was no significant effect between these two variables on grades (p = .24) . The comparison of students of TA's who were objectively rated low in empathy and who perceived their TA's as high in empathy with those students who perceived TA as low in empathy and who also had objectively rated low empathy TA's on academic achievement shows that the low, low group performed on a lower level than the low, high, group. These comparisons were significant for SA (t=2.96, p< .001) and WRIT (t=2.54, p< .01) and were not significant for the other academic performance criteria MC (t=.16; p< .00) , E (t=1.02, df=126; p< .10) , Tot (t=.678; df=126; p <.10). Planned comparison t=tests were performed for the jo performance criteria as specified in the fourth hypothesis. Table 9 Means and Standard Deviations of the Triple Interaction Effects of Objective Empathy, Perceived Empathy, and Self-Esteem with Short Answer Scores Objec tive Per ceived Esteem Multiple Choice Essay Written Total Grade Low Low Low M 189.63 SD 22.44 M 26.94 SD 5.60 M 54.13 SD 11.81 M 243.31 SD 29.72 M 4.47 SD .80 Low Low High 205.93 25.46 29.55 6.24 58.62 10.44 264.52 30.96 4.90 .74 Low High Low 195.36 27.62 30.67 9.00 63.33 14.32 256.00 38.13 4.76 .88 Low High High 200.79 28.40 27.97 2.16 58.21 15.00 259.24 36.82 4.79 .92 High Low Low 198.30 32.69 32.90 7.42 63.33 10.91 261.47 39.58 4.87 .83 High Low High 189.33 29.79 30.63 5.50 61.25 9.33 248.08 36.25 4.58 .80 High High Low 201.14 24.70 31.58 6.78 62.51 10.76 258.65 44.18 4.91 .69 High High High 215.52 28.50 34.64 8.21 66.57 13.33 282.83 38.82 5.36 .91 o o H Figure 2. Graphic Presentation of the Three Way Interaction Effect of Objective Empathy, Perceived Empathy, and Self-Esteem on Total Exam Scores. Total Exam Scores 285- 280- 275- 2 1 0 - 265- 260- 255- 250- High P/High E Low P/Low E High P/Low E Low P/High E L H Objective Empathy 83 Figure 3. Graphic Representations of the Rela tionship of Objective Empathy and Perceived Empathy with Academic Achievement Variables. Total Exam Score Multiple Choice 275- High P 270 • 265 - 260- 255- - • Low P L H 210 - High P 205- 200- Low P 195* H L Objective Empathy Objective Empathy Total Written Short Answer Essay High 32" 65 High P 'Ow P 30- lOW P 55' L H H L High P 32- 30- Low P 28" H L Objective Empathy Objective Empathy Objective Empathy 84 A comparison on academic performance criteria was made be tween students of high objectively rated TA's who perceived their TA's as low in empathy and high objectively rated TA's who perceived their TA's as high in empathy. The stu dents of objectively rated high empathy TA's and who per ceived their TA's as high in empathy performed significant ly better than students of objectively rated high TA's who perceived their. TA as low in empathy on short answer (t=3.0 p < .002) essay (t=2.63, p< .005), and total performance (t=13.71, p <.001) with df=137. In addition, a comparison of course evaluation cri teria also indicated that students of objectively rated high empathy TA's who also perceived their TA’s as high in empathy rated their TA's significantly higher on interper sonal relationships (t=6.59, p^.001) global learning (t=5.95, p< .001) and on course relevance (t=6.74, p< .001) with df*=137, than students of objectively rated high em pathy TA's who perceive their TA's as low empathy. Student Course Evaluations The students' responses to the Course Comments Questionnaire were the criteria used to measure the stu dents' evaluation of the course. These factors included teacher's skill as reflected by his knowledge (K), an as sessment of student-teacher interpersonal relationship (I), difficulty of the class (D), skill as reflected in teachers' organization (0), student's perception of global 85 Table 10 Analysis of Variance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness, and Self-Esteem on Student Course Evaluations as Measured by Factors from the Course Comments Questionnaire Student Course Evaluations Source of ______________________________________ Variance Knowl- Interperson- Course edge al Relations Difficulty F P F P F P Objective Empathy (0) 1.74 .19 7.52 .007 .79 .38 Student (P) Perceived Empathy 109.64 .001 97.58 .001 15.29 .001 Defensiveness (D) 1.35 .25 .79 .38 .45 .50 Berger (E) Self-Esteem .50 .48 1.82 .18 .26 .61 0 X P .09 .77 1.71 .19 .11 .74 0 x D 2.28 .13 6.05 .02 .00 .98 0 x E .47 .49 .33 .56 1.27 .26 P x D .54 .46 2.74 .10 1.07 .30 P x E .01 .98 .11 .74 .23 .64 D x E 2.24 .14 .65 .42 .28 .60 0 x P x D 1.04 .31 5.76 .02 .37 .54 0 x P x E .06 .81 .44 .51 .81 .37 0 x D x E .53 .47 .16 .69 5.85 .02 P x D x E .030 .86 3.13 .08 .99 .32 0 x P x D x E .18 .67 .10 .74 1.37 .24 df - 1,251 86 Table 10— Continued Source of Student Course Evaluations Variance Skill Organi zation - Global Learning Specific Learning Course Relevance F P F P F P F P Objective Empathy (0) 6.53 .01 .10 .75 .79 .38 1.53 .22 Student (P) Perceived Empathy 118.37 .001 i 61.09 .001 37.04 .001 61.27 .001 Defensive ness (D) .50 .48 5.25 .02* 1.80 .181 .06 .81 Berger (E) Self- Esteem .22 .64 .00 .98 5.53 .02 .24 .62 0 x P .19 .66 1.34 .25 .10 .75 .00 .98 0 x D 4.95 .03 .01 .94 .09 .77 .32 .57 0 x E .21 .66 2.14 .15 .25 .62 .32 .58 P x D .62 .43 1.45 .23 6.77 .01 .90 .35 P x E .08 .78 0.00 1.00 1.36 .24 1.00 .32 D X E .39 .53 .01 .92 .09 .76 .54 .46 0 X P X D 1.07 .30 .02 .90 .56 .45 .02 .88 0 X P X E 3.63 .06 .08 .77 1.02 .31 .08 .78 0 X D X E .64 .43 0.00 .97 .99 .32 .25 .62 P X D X E 1.40 .24 2.45 .12 1.07 .30 1.02 .31 0 X P X D X E .06 .81 .55 .46 1.26 .26 .00 .96 87 learning (G), student's perception of specific learning (S) and his judgment of course relevance (R). Table 10 presents the results of the analysis of objectively rated empathy, perceived empathy, defensiveness, and self-esteem on the course evaluation factors: K, I, D, O, G, S, and R as dependent variables. Students of TA's who were objectively rated as higher in empathy, rated their student-teacher interper sonal relationships (I) and TA skill as reflected in course organization (O) significantly higher than students whose TA's were objectively rated as low in empathy (p<.007, p {.01, respectively). The means and standard deviations for these significant effects are listed in Table 11. Students who perceived their TA's as high in empa thy, evaluated all seven factors significantly higher than those students who perceived their TA's as low in empathy (p<.001 for all variables) as indicated by their means and standard deviations (Table 12). This shows that students' perceived empathy of TA's is directly related to the way in which students evaluate the course. Students who were high in defensiveness evaluated themselves as gaining significantly more global learning (G), than those students who were low in defensiveness (p < 02) . These means and standard deviations are pre sented in Table 13. 88 Table 11 Means and Standard Deviations of Objective Empathy with Skill as Reflected in TA Knowledge Student Course Evaluations Objective (0) ---------------------------------------- Interpersonal Skill/ Relationship Organization M SD M SD Low 3.930 1.072 3.156 1.252 High 4.273 .810 3.554 1.228 89 Table 12 Means and Standard Deviations of Perceived Empathy with Skill/Knowledge, Interpersonal Relationship, Course Difficulty, Skill/Organization, Global Learning, Specific Learning, and Course Relevance Perceived (P) Student Course Evaluations Empathy skill/Knowledoe Interpersonal Course SKxxx/nnowxeage Relationship Difficulty M SD M SD M SD LOW 2.009 1.868 3.383 1.215 2.409 1.275 High 4.112 1-.331 4*658 .609 2.961 1.127 Student Course Evaluations Skill/Organ ization Global Learning Specific Learning Course Relevance M SD M SD M SD M SD Low 2.374 1.424 1.730 1.585 2.374 1.538 2.678 1.425 High 4.112 1.100 3.263 1.560 3.513 1.432 3.961 1.103 Table 13 Means and Standard Deviations of Defensiveness with Students' Perception of Global Learning Defensiveness (D) Student Course Evaluations Global Learning M SD Low 2.346 1.572 High 2.858 1.569 91 In the previous section, it was reported that those persons with high self-esteem performed significantly better on MC, TOT, and GRADE. Corresponding to this per formance, the high self-esteem students also rated them selves as gaining more specific knowledge (S) than those students who were low in self-esteem (p <.02), as shown by the means and standard deviations in Table 14. Two significant interaction effects were found be- tween objective rating and defensiveness in relationship to interpersonal relations (p< .01) and (p < .03) . The means and standard deviations for these significant inter actions are presented in Table 15. The student-teacher interpersonal relationship finding indicated that those people who had TA's who were objectively rated as low in empathy and who were high in defensiveness judged the student-teacher relationship as being significantly worse than those in the high objective, low defensive (t»2.44, p<.05) and the high objective, high defensive (t=2.62, p< .01) groups. The skill as reflected by TA organization finding indicated that the low objective, low defensive group evaluates the TA as having less skill in organization than the other groups. This finding is significant when comparing the low objective, low defensive group to the high objective, low defensive group (t=*2.39, p < .05) and especially when comparing them with the high objective, high defensive group (t=3.05, p<.01). This suggests that Table 14 Means and Standard Deviations of Self-Esteem with Students' Perception of Specific Learning Self-Esteem (E) Student Course Evaluations Specific Learning M SD Low 2.790 1.480 High 3.271 1.475 Table 15 Means and Standard Deviations of Objective Empathy and Defensiveness with Interpersonal Relationships and Skill as Reflected in TA Organization Objective/Defensive Student Course: Evaluations 0 x D Interpersonal Relationship Skill/ Organization M SD M SD Low Low 4.018 1.142 3.088 1.348 Low High 3.859 1.044 3.211 1.175 High Low 4.250 .775 3.434 1.339 High High 4.302 .853 3.698 1.094 93 high defensive scorers, give a stronger socially approved response (i.e.r teacher has more organizational skill). An examination of the perceived empathy and de fensiveness interaction (P x D) shows a significant inter action effect with S (p< .01). Means and standard devia tions are presented in Table 16, which indicates that those students who had low perceived empathy and high defensive scores, rated their specific learning as being significants ly less than those with high perceived empathy and high defensiveness (t=7.47, p< .01). A triple interaction ef fect between objectively rated TA empathy, student per ceived empathy of TA and student’s (O x P x D) can be seen for the course evaluation factor I (p < .02). This sug gests that there is a relationship between objective empa thy, perceived empathy, defensiveness and student's perception of the student-teacher interpersonal relation ship. The main difference, as can be seen in Table 17 is reflected by those students who have objectively rated low TA's, perceive their TA's as low and who are themselves high defensive. On the other hand, those students who perceive their TA as high in empathy and have high de fensiveness , themselves report their TA as high in inter personal skills, irrespective of whether or not the T.A. is objectively rated as high or low in empathy. Another significant triple interaction emerged when examining objective and perceived empathy with self- 94 Table 16 Means and Standard Deviations of Perceived Empathy and Defensiveness with Specific Learning Perceived P Defensiveness Student Course Evaluations x D Specific Learning M SD Low Low 2.517 1.481 Low High 2.218 1.600 High Low 3.178 1.599 High High 3.823 1.314 Table 17 Means and Standard Deviations of Objective Empathy, Perceived Empathy, and Defensiveness with Student- Teacher Interpersonal Relationships Objective Perceived Defensiveness Student Course Evaluations Interpersonal Relations M SD Low Low Low 3.647 1.435 Low Low High 2.630 1.500 Low High Low 4.565 .622 Low High High 4.614 .765 High Low Low 3.462 1.130 High Low High 3.714 1.382 High High Low 4.660 .590 High High High 4.771 .431 95 esteem (Table 18)/ on the course difficulty factor (p <.02). The main difference in evaluation of course difficulty ap pears to be between those groups who are both high defen sive and high self-esteem scorers and whose TA's are ob jectively rated high or low empathy. The students who have objectively rated lower empathy TA's rate the class as far more difficult than students who have TA's who are objec tively rated as high in empathy. Liking Effects The student's response to the question/ How much do you like your TA? was the criterion used for this ef fect. They rated amount of liking on a Likert Scale with 1 being low and 5 being high. Table 19 presents the results of the analysis of variance of objectively rated TA empathy, student perceived empathy, defensiveness, and self-esteem on the like scale. A significant effect (Table 19) was demonstrated with stu dent perceived empathy on the like scale (p < .001). As seen in Table 20, students who perceived their TA as high in empathy, also liked their TA more. Liking was the only dependent variable with a quadruple interaction among all the independent variables. Table 21 presents the means and standard deviations for the quadruple interaction between objective empathy, subjective empathy, defensiveness, and self-esteem. The groups are Table 18 Means and Standard Deviations of Objective Empathy, Defensiveness, and Self-Esteem with Students' Perception of Course Difficulty Objective Defensiveness Esteem Student Course Evaluations Difficulty M SD Low Low Low 2.714 1.165 Low Low High 2.621 1.134 Low High Low 2.622 1.174 Low High High 3.176 .998 High Low Low 2.512 1.260 High Low High 2.818 .920 High High Low 2.933 1.770 High High High 2.455 1.138 Table 19 Analysis of Variance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness, and Self-Esteem on the Like Scale and Intellectual Functioning Source of Variance LIKE SHIQ F P F P Objective (0) Empathy .80 .37 .40 .53 Student (P) Perceived Empathy 114.03 .001 .74 .39 Defensiveness (D) .003 .96 12.11 .001 Berger (E) Self-Esteem .10 .76 .12 .73 0 X P .00 .99 1.18 .28 0 x D .40 .53 .06 .81 0 x E .03 .88 1.78 .18 P X D .31 .58 .02 .90 P X E .14 .71 .09 .76 D X E .03 .87 .05 .82 0 X P X D .24 .63 2.41 .12 0 X P X E 1.72 .19 2.15 .14 0 X D X E .25 .14 .25 .62 P X D X E 1.14 .29 4.58 .03 0 X P X D X E 6.65 .01 .21 .65 Table 20 Means and Standard Deviations of Empathy with Liking Perceived Perceived (P) Empathy Like M SD Low 2.991 1.159 High 4.329 .835 Table 21 Means and Standard Deviation of the Quadruple Interaction Effects of Objective Empathy, Perceived Empathy, Defensiveness and Self-Esteem on Liking (The Means Are Rank Ordered from Low to High) Objective/Perceived/Defensiveness/Self-Esteem Like High Low High LOW M 2.F3 SD lTTl High Low Low High 2.55 1.13 Low Low High High 2.75 .75 Low Low Low Low 2.94 1.39 High Low Low Low 3.07 1.39 Low Low LOW High 3.12 1.05 Low Low High Low 3.40 .91 High Low High High 3.46 1.13 Low High High Low 4.18 .10 Low High High High 4.23 .97 High High High High 4.25 1.02 High High Low High 4.32 .84 High High Low Low 4.36 .62 Low High Low Low 4.36 .92 Low High Low High 4.42 .67 High High High Low 4.67 .49 f 9 9 ! rank ordered according to their mean from lowest to highest j i in liking. Most obvious in this ranking is that the high ! perceived empathy students report liking TA's more than the low perceived empathy students regardless of whether the TA has been rated objectively as high or low in empathy. The low perceived empathy low defensiveness and low self- esteem students reported less liking of TA's independently of how the TA's were rated than the high perceived empathy, high defensiveness, and high esteem students. Intellectual Functioning The student's performance on the Shipley-Hartford C.Q. abstraction test interpreted into I.Q. equivalents was used as the criterion for the student's intellectual functioning. The results of the analysis of variance of objec tive empathy, perceived empathy, defensiveness, and self esteem on intellectual functioning are presented in Table 19. There was a significant interaction between defensive-: ness and intellectual functioning (p< .001). Table 22 pre sents the means and standard deviations for this effect, which shows that those students low in defensiveness performed significantly better than those who were high in defensiveness. A significant triple interaction occurred on the SHIQ variable with perceived empathy, defensiveness, and self-esteem (p<.03). The means and standard deviation for this interaction are presented in Table 23. The groups 100 Table 22 Means and Standard Deviations of Defensiveness with Intellectual Functioning Defensiveness (D) SHIQ M SD Low 122.609 8.179 High 118.679 9.604 Table 23 Means and Standard Deviations of Perceived Empathy, Defensiveness, and Self-Esteem with Intellectual Functioning Perceived/Defensiveness/Self-Esteem P x D x E M SD Low Low Low 122.937 8.271 Low Low High 120.964 11.168 Low High Low 116.033 11.660 LOW High High 120.120 6.854 High Low Low 122.359 9.400 High Low High 123.941 6.164 High High Low 120.027 9.912 High High High 118.524 9.619 101 jwho performed lowest on this measure were those low objec tive , high defensive, and low self-esteem groups and those high objective, high defensiveness and high self-esteem groups. Those groups who performed significantly higher I were all groups who had low defensiveness scorers. These included the low objective, low defensiveness, low self esteem groups and the high objective, low defensiveness, low self-esteem groups. Correlations Among the Variables Tables 24 and 25 present intercorrelations among selected variables for all subjects. Objectively rated empathy was more highly correlated to academic achievement them to course evaluations, while perceived empathy was more highly related to course evaluations than to academic achievement variables. Intellectual functioning (SHIQ) is moderately correlated with academic achievement measures, but not to course evaluation. On the other hand, the amount of student's reported liking of teaching assistants is moderately related to course evaluations and not to aca demic achievemant. As is presented in Table 24, specifically objec tive empathy was more highly correlated with written type exam performance which includes both essay and short answer questions. Perceived empathy was more positively corre lated with multiple choice type questions than objective Table 24 Summary of Selected Correlations of Objective Empathy, Perceived Empathy, Intellectual Functioning and Liking with Academic Achievement Multiple Choice Short Answer Essay Written Total Grade Objective Empathy* -.200 .120 .605 .669 .355 .223 Perceived Empathy .152 .216 .101 .188 .156 .169 Intellectual Functioning .405 .299 .232 .303 .355 .300 Liking .106 .236 .070 .154 .093 .141 ♦Rank order correlation coefficient. 1 102 Table 25 Summary of Selected Correlations of Objective Empathy, Perceived Empathy, Intellectual Functioning and Liking with Course Evaluations Skill/ Knowledge Inter personal Relations Course Difficulty Skill/ Organization Global Learning Specific Learning Course i Relevance Objective Empathy* .064 .560 -.263 .573 .010 .464 .437 | Perceived Empathy .494 .552 .202 .500 .360 .331 .424 Intellectual Functioning .031 .040 -.019 .032 -.056 -.042 i -.013 Liking .467 .487 .181 .442 .291 .340 .488 *Rank order correlation coefficient. 104 empathy. Intellectual functioning was moderately related to all academic achievement measures and most highly re lated to multiple choice type performance. Liking of the TA was only slightly correlated with academic achievement measures and was most related to short answer type ques tions . A summary of mean correlations of objective empa thy, perceived empathy, liking and intellectual function ing with the seven course evaluation factors is presented in Table 25. Objective empathy is highly correlated with student-teacher interpersonal relationships and skill as reflected in organization of the course, in addition, objective empathy is moderately related to the student's perception of specific learning, and course relevance. While objective empathy is negatively correlated to course difficulty, TA's perceived empathy was related positively to course difficulty. Perceived empathy was highly re lated to interpersonal relationship and skill as reflected in organization, as was objective empathy. In addition, perceived empathy was moderately related to skill as re flected in teacher knowledge. Perceived empathy was also moderately related to course relevance, global learning and specific learning. While intellectual functioning was not significantly correlated to the course evaluation fac tors , liking was moderately correlated to skill as re flected in both knowledge and organization, interpersonal 105 relationship and course relevance. In addition liking was slightly related to the person's perception of both spe cific and global learning. Thus, academic achievement is more highly related to objective rating of TA empathy and intellectual func tioning , while course evaluations are more highly related to student's perceived empathy of TA and amount of re ported liking of TA. These correlational findings support the analysis of variance results which were reported in the first part of this chapter, although they do not re flect the more complex interactions reported in that section. Analysis of Covariance An analysis of covariance was performed for the independent variables of objective rating of TA empathy (0), student's perceived empathy of TA (P), defensiveness (D) , and self-esteem (E) using intellectual functioning as the covariate. This analysis was performed on all de pendent variables (i.e., both academic achievement and course evaluation variables, as well as liking). A sum mary of the analysis of covariance with the independent variables and academic achievement variables is presented in Table 26. The only differences between the primary analysis and the covariance analysis for main effects occurred on the TOT scale. These changes occurred for the main Table 26 Summary of Analysis of Covariance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness, and Self-Esteem with Academic Achievement as Measured by Performance on Exams Academic Achievement Source of Variance Multiple Choice Short Answer Essay Written Total Grade F P F P F P F P F P F P Objective Empathy (0) 1.81 .18 3.87 .05 16.91 .001 10.90 .001 3.36 .07 5.69 .02 Student (P) Perceived Empathy 3.91 .049 5.96 .02 1.19 .28 3.81 .05 3.48 .06 4.99 .03 Defensiveness (D) .28 .60 2.04 .16 .61 .44 .16 .69 1.50 .22 0.00 .99 Berger (E) Self-Esteem 6.04 .02 .50 .48 .21 .65 .11 .74 5.62 .02 3.76 .05 0 x p 0 x D x X X X X X X X X o p p D 0 0 o p 0 E D E E P P D D P D E X E X E x D X E 3.02 2.31 .01 3.18 1.41 .02 6.22 4.22 2.63 4.71 .13 .08 .13 .93 .08 .24 .90 .02 .04 .11 .03 .72 1.06 2.17 .71 .14 .28 4.11 5.71 1.58 2.17 .00 1.01 .30 .14 .40 .71 .60 .04 .02 .21 .14 .98 .32 .04 .03 1.00 4.57 .01 .99 3.21 7.06 1.21 .09 .19 .84 .88 .32 .03 .92 .32 .07 .01 .27 .76 .67 .84 1.19 1.53 2.26 .21 2.53 4.39 6.00 2.24 .06 .62 .36 .28 .22 .13 .65 .11 .04 .02 ,14 .81 .43 .61 2.36 .18 2.57 1.65 1.62 4.28 7.73 2.78 3.41 .39 .44 .13 .68 .11 .20 .20 .04 .01 ,10 .07 .53 1.54 .62 .08 2.40 1.06 .03 3.04 5.51 .53 1.92 .02 .22 .43 .78 .12 .30 .85 .08 .02 ,47 .17 .89 106 107 j effects of objective empathy, perceived empathy and de fensiveness. Table 27 contains the means for each of these main effects and the adjusted means. In general, it can be seen that the adjusted means are essentially the same as unadjusted means. Thus, the primary difference indi cated in the covariance analysis is slight changes in the probability levels, rather than in the absolute magnitude or direction of these effects. In addition, a similar oc currence was obtained in the covariance for the defensive ness variable on the course evaluation measure. The sum mary table for this analysis is in Table 28. This same relationship also occurred on the global learning measure for the interaction of objective empathy by self-esteem. Differences in interaction effects between the primary analysis and the covariance analysis occurred for two effects only. In the primary analysis the perceived empathy by defensiveness by self-esteem interaction was not significant for Multiple Choice scores, however, in the covariance this effect was significant. The means and adjusted means are presented in Table 29. This difference occurred because the group that obtained the lowest scores on Multiple Choice also obtained the lowest I.Q. scores. In addition to the effect on short answer questions present in the primary analysis (p <.02) , the covariance indicated significant effects on MC (p < .02), WRIT (p <.04), TOT (p <.04). The means and adjusted means for 108 Table 27 Summary of Means and Adjusted Means of the Effects of Objective Empathy, Perceived Empathy, and Defensiveness on Total Exam Scores Means Total Adjusted Means Objective Empathy (0) Low 255.62 256.12 High 264.74 264.28 Perceived Empathy (P) Low 254.39 255.20 High 264.89 264.28 Defensiveness (D) Low 266.03 263.35 High 254.75 257.40 Table 28 Summary of Analysis of Covariance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness and Self-Esteem with Course Evaluation Course Evaluation Source of Variance Skill/Knowledge Interpersonal Relations Course Difficulty M SD M SD M SD “““ ■ 11 ■ ■— "■ )bjective (0) Empathy 1.95 .16 7.74 .006 .738 .39 Student (P) Perceived Empathy 112.51 .001 98.45 .001 15.50 .001 Defensiveness (D) .55 .46 1.19 .28 .26 .61 Berger (E) Self-Esteem in « .50 1.89 .17 .28 © VO • 0 X P .09 .76 1.71 .19 .11 .74 0 x D 2.39 .12 6.13 .01 .00 .99 0 X E .71 .40 .24 .63 1.39 .24 P x D .57 .45 .77 .10 1.08 .30 P x E .01 .91 .10 .75 .21 .65 D x E 2.34 .12 .67 .41 .27 .61 0 x P x D 1.44 .23 6.20 .01 .45 .50 0 X P X E .01 .95 .32 .57 .70 .41 0 X D X E .63 .43 .19 .67 5.93 .02 P X D x E 1.92 .17 .18 .67 3.58 .06 0 X P X D X E .23 .63 .08 .78 1.32 .25 df - 1,251 Table 28— Continued Course Evaluation Source of Variance Skill/ Organization Global Learning Specific Learning Course Relevance M SD M SD M SD M SD 1 " • — ■ ■ ■ Objective (0) Empathy 6.75 .01 .84 .66 1.60 .36 .78 .21 Student (P) Perceived Empathy 119.48 .001 64.79 .001 37.41 .001 61.62 .001 Defensiveness (D) .21 .65 2.84 .09 1.29 .26 .16 .69 Berger (E) Self-Esteem .24 .63 .01 .93 5.60 .02 .26 .61 0 x P .18 .67 1.37 .24 .10 .75 .00 .98 0 x D 5.03 .03 .01 .91 .08 .78 .33 .57 0 x E .31 .58 2.90 .09 .32 .57 .25 .62 P x D .64 .43 1.54 .22 6.79 .01 .91 .34 P x E .09 .76 0.00 .99 1.40 .24 1.02 .31 D x-E .41 .52 0.00 .95 .10 .75 .55 .46 0 X P X D 1.29 .26 .02 .89 .68 .41 .01 .94 0 X P X E 3.22 .07 .00 .96 .87 .35 .05 .83 0 x D x E .58 .45 .01 .93 .94 .33 .27 .60 P x D x E .80 .37 1.75 .19 3.73 .06 1.28 .26 0 x P x D x E .07 .79 .68 .41 1.30 .25 .00 .95 H H Table 29 Summary of Means and Adjusted Means for the Interaction Effect of Perceived Empathy, Defensiveness and Self-Esteem on Multiple Choice Multiple Choice Perceived/Defensiveness/Self-Esteem P x D x E Means Adjusted Means Low Low Low 198.00 195.209 Low Low High 205.93 205.53 Low High Low 189.37 194.95 Low High High 190.99 190.63 High Low Low 202.95 200.86 High Low High 208.06 204.05 High High Low 194.08 194.82 High High High 209.64 212.206 jthese last three variables are presented in Table 30. i Figure 4 is a graphic presentation of this effect on Mul- itiple Choice Scores and is representative of all the vari ables. For each of these variables, the students with high perceived empathy and high self-esteem gave the higher per formance on exams. The students who were high in defen siveness, continued to do more poorly across criteria than those students who were low in defensiveness. Students who had objectively rated high empathy TA's, who perceived their TA's as low in empathy, and who were high in defen siveness, performed at a lower level on MC than any other group, including those students who had objectively rated low empathy TA's. The analysis of covariance using intellectual functioning as a covariate showed no significant differ ences from the primary analysis when examining the liking variable. The summary of this analysis is presented in Table 31. Thus, there were essentially no differences in significant effects and interaction effects indicated, except in the interaction of objective rating of TA empa thy, student perceived empathy, and defensiveness. Those few instances where the primary analysis and the covariance analysis differed were generally related to the defensive ness variable. This occurred because those subjects who scored high in defensiveness scored significantly lower on 113 IQ, and IQ had a low, moderate correlation to academic performance in general. Table 30 Summary of Means and Adjusted Means for the Interaction Effects of Objective Empathy, Perceived Empathy and Defensiveness with Academic Achievement Variables, Multiple Choice, Written and Total Scores Ob je ctive/Perce ived/De fens ivenes s Multiple Choice Written Total 0 x P x D M ADJ M ADJ M ADJ LOW Low Low 197.91 195.15 55.97 55.10 255.27 252.18 LOW Low High 196.70 201.66 56.63 58.20 251.04 256.58 Low High Low 199.17 197.68 60.61 60.14 260.39 258.72 Low High High 197.57 198.75 60.80 61.17 256.20 257.53 High Low Low 206.65 206.40 66.70 66.61 273.27 272.98 High Low High 182.16 184.63 58.43 58.99 239.04 241.02 High High Low 208.16 204.49 64.00 62.84 272.18 268.07 High High High 208.37 210.74 65.26 66.01 268.34 270.99 115 Figure 4. Graphic Presentation of the Effect of Objective Empathy, Perceived Empathy, and Defensiveness on Academic Achievement as Represented by Multiple Choice Performance. m •* Amov a A- — — -A CoVftftXRNGE Multiple Choice I P/HXfcHO WWWD Low P/Love ^A LowI/IjOW© "'"’ ^ A HXM P/LowtJ 180' OOTECTtVE EMPATHY 116 Table 31 Analysis of Covariance of the Effects of Objective Empathy, Perceived Empathy, Defensiveness and Self-Esteem on Liking F P Objective Empathy (0) .78 .38 Student Perceived Empathy (p) 113.04 .001 Defensiveness (D) .01 .93 Self-Esteem (E) .10 .75 0 x P 1.18 .28 0 x D .39 .53 0 x E .03 .86 P x D .31 .58 P X E .14 .71 D X E .03 .87 0 X P X D .25 .62 0 X P X E 1.75 .19 0 X D X E 2.26 .13 P X D X E 1.06 .30 0 X P X D X E 6.59 .01 CHAPTER IV DISCUSSION The present study offered a systematically designed evaluation of student achievement as related to classroom environment and student personality characteristics. This study consisted of an exploration of the effects of objec tively assessed empathy of teaching assistant, students' perceived empathy of teaching assistant, students' defen siveness , and students' self-esteem on academic achieve ment, course evaluations, and liking. In order to more specifically examine the effects of objective empathy and perceived empathy on achievement, course examinations were divided into three kinds of questions: multiple choice, short answer, and essay. This study not only makes an assessment of academic achievement, but also examines the relationship of objective empathy, perceived empathy, de fensiveness (need for social approval), and self-esteem to specific types of course evaluation variables which may answer questions about the function of course evaluations in the university system. In addition, the function of students' liking for the TA was examined in relationship to his academic performance and course evaluations. 117 118 The results of this study on academic achievement may be summarized as: 1. Students of objectively rated high empathy TA's in general, perform at a higher level than students of objectively rated low empathy TA's, especially on written questions. 2. Students who perceive their TA as high in empathy performed better academically on all measures ex cept essay scores than students who perceived their TA as low in empathy. 3. Students who were low in defensiveness performed better than those students who were high in de fensiveness when examining the total performance scores. 4. Students who scored high in self-esteem obtained higher scores on multiple choice questions and consequently earned more total points and received higher grades. 5. Students who perceived their teaching assistants as low in empathy and who are high in defensiveness perform more poorly on essay type questions than any other group. 6. Students high in defensiveness and high in self esteem perform better on short answer questions, than all other groups which appear to perform equally. 119 7. Objectively rated high empathy TA's facilitate learning of students who perceive the TA as low in empathy and who are low in defensiveness. These TA's were not facilitative for students who were highly defensive and perceived the TA as low in empathy. 8. The relationship of student's self-esteem with ob jective empathy and perceived empathy is the most powerful and important interaction for predicting academic achievement as measured by multiple choice, essay, total exam scores, and grades. Thus, students who were high in self-esteem and who had TA's objectively rated as high empathy and who perceived their TA as high in empathy per formed better according to achievement criteria than all other groups. 9. Grades were not related to liking. 10. These results were independent of student's I.Q. scores. The results of this study on course evaluations may be summarized as: 1. Students of TA's who were objectively rated higher in empathy rated student-teacher interpersonal re lationships as better than those students of TA's who were objectively rated lower in empathy. They also rated the TA higher in skill as reflected by organization. Students of objectively rated low empathy TA's evaluate the course as more difficult than stu dents of those TA's objectively rated high in empathy. TA's perceived as high empathy received more positive ratings on all course evaluation criteria than TA's perceived as low empathy. High self-esteem students report more specific learning than low self-esteem students. Objectively rated low empathy TA's receive less positive evaluation on interpersonal relations from high defensive students than other combina tions of TA empathy and student defensiveness. Students who are low in defensiveness evaluate their objectively rated low empathy TA as having less organizational skill than other groups. The course evaluations were positively correlated to how much the student likes his TA. Students high in defensiveness who perceived their TA as low empathy gave objectively rated low em pathy TA's lower scores on interpersonal relations than any other group. Students high in defensiveness and self-esteem rated the course as more difficult than any other 121 group when assigned to an objectively rated low empathy TA. They rated the course as less diffi cult than any other group when assigned to an objectively rated high empathy TA. Evaluation of Hypotheses Hypothesis 1A. Students of teaching assistants who are objectively rated as high in empathy will do better on essay type questions than students of teaching assis tants who are objectively rated as low in empathy. Students of TA's who were objectively rated high in empathy performed better on short answer questions (p< .04) and essay questions (p<.001), than students of objectively rated low empathy TA's. For total written examination scores, students of objectively rated high empathy TA's did better than students of low objectively rated TA's (p < .001). Thus, the first part of Hypothesis 1 was supported. Hypothe sis IB. Students of TA's who are objective ly rated as high in empathy will rate TA higher on the Course Comments Questionnaire when referring to teacher re sponsiveness or interaction (factor 2) and relevance (fac tor 7), than students with objectively rated low empathy TA's. Students of objectively rated high empathy TA's rated their TA's as high in interaction (p< .007). How ever, when evaluating relevance there was not a significant effect with students of objectively rated high empathy TA's (p<.22). Thus the interpersonal interaction part of the 122 hypothesis was supported, which suggests that personal relevance would have been supported also. However, an examination of the items specific to the variable of rele vance reveals that all the items are not personally oriented. Hypothesis 1C. Students of TA's who are objec tively rated as high in empathy will rate themselves higher in general cognitive development (factor 5) than students with objectively rated low empathy TA's. Students of objectively rated high empathy TA's did not signifi cantly rate themselves higher on global learning (p=.75). Thus, Hypothesis 1C was not supported. The findings for the three parts of Hypothesis 1 support Bendig's (1953) findings in that there was no re lationship between achievement as measured by global learning and objectively rated empathy. In addition this hypothesis partially supports Aspy's (1971) findings with young students that empathy is related to positive academic growth in the classroom as measured by achievement tests. Aspy and Roebuck (1972) suggested that when meas uring achievement as outcome there was a positive rela tionship with empathy, but when measuring process which is equivalent to the global learning, there was no significant relationship. Hypotheses 1A and 1C supported this dual finding for achievement as measured by outcome and process criteria. 123 Hypothesis 2A. Students who perceived their TA's as high in empathy will do better on essay type questions than students who perceive their TA to be low in empathy. On short answer questions (p *.01) and on total written score (p <.04) students who perceived their TA's as high in empathy performed at a significantly higher level than students who perceived their TA's as low in empathy. On essay performance the same trend appeared, but not at a significant level (p <.22). Thus, this hypothesis was supported. Hypothesis 2B. Students who perceived their teaching assistants as high in empathy will rate the TA higher on the Course Comments Questionnaire when referring to responsiveness of interaction (factor 2) and relevance (factor 7), than students who perceive their TAs as low in empathy. Students who perceived their TA's as high in empathy rated them higher in interpersonal interaction (p< .001) and in relevance (p <.001) than students who per ceived TA's as low in empathy. Thus, the second part of the second hypothesis was supported. Hypothesis 2C. Students who perceived their teaching assistants as high in empathy will rate themselves higher in general cognitive development (factor 5) than students who perceive TA's as low in empathy. Students who perceived TA's as high in empathy rated themselves signifi cantly higher in global learning or general cognitive 124 development (p< .001) than students who rated TA's as low empathy. Thus, Hypothesis 2C was supported. McKeachie, Lin, and Mann (1971) suggested on the basis of their data that the personally facilitative condi tions were more relevant on outcome measures which require personal participation, as in written type examination questions. In addition, Meredith (1969) and Hartley and Hogan (1972) have reported extensively that teachers who are perceived as high in empathy will have more favorable course evaluations. The findings of this study were per fectly consistent with those proposals. Hypothesis 3. The tendency of students who per ceive the teaching assistants as high in empathy to do better will be more pronounced on essay type questions than on multiple choice type questions. Students who perceived TA's as high in empathy did perform better when considering total exam score (p< .04) than students who perceived T.A.'s as low in empathy. The high perceived empathy group performed better on short answer questions (p<.01) giving them a better total performance score (p< .04) than the low perceived empathy group. However, in multiple choice per formance the high perceived empathy group performed equally as well as they did on written type questions and conse quently out-performed the low perceived empathy group on multiple choice also (p < .03) . This hypothesis, then, was not supported in that the high perceived empathy group 125 performed at a higher level on the achievement measures both in terms of short answer type questions and multiple choice questions. McKeachie and his associate (1971) had suggested that lack of enthusiasm and empathy affect the students' written performance more than his multiple choice perfor mance. This proposal is supported where empathy is ob jectively ascertained, but not when student's perception is the criterion. Hypothesis 4A. Students whose TA's are objectively rated higher and who rate their TA's as higher in empathy will do better on essay type questions than those students who perceive TA's as low in empathy. Students of TA's ob jectively rated as high in empathy and who perceived the TA as high in empathy performed better on short answer (t=3.00; df" 137; p<.002), essay (t=2.63; df=137; p < .005) and total written (t=13.71; df=137; p< .001) than students of objec tively rated high empathy TA's who were perceived as lower in empathy. Thus, Hypothesis 4A was supported. Hypothesis 4B. Students whose TA's are objectively rated higher and who rate their TA's higher in empathy will rate the TA higher on the Course Comments Questionnaire when referring to responsiveness (factor 2) and relevance (factor 7) than students who perceive their TA as lower in empathy. Students whose TA's were objectively rated higher in empathy and who perceived the TA as high empathy did 126 rate TA's higher in responsiveness or interpersonal inter action (t=6.59; df=137, p<.001) and in relevance (t=6.74, df=137; p< .001) than students of objectively rated high empathy TA's who were perceived as low empathy. Thus, this part of Hypothesis 4 was supported in the present study. Hypothesis 4C. Students whose TA's are objectively rated higher and who rate their TA's higher in empathy will rate themselves higher in general cognitive development (factor 5) than students who perceive their TA as lower in empathy. Students of TA's objectively rated as high in empathy and who perceived their TA as high in empathy did rate themselves as gaining more global learning (t=5.95; df=137; p<.001) them those students of TA's objectively rated as high in empathy and who perceived the TA as low in empathy. Thus, Hypothesis 4C was also supported. McKeachie, Lin, and Mann (1971) found that there was a significant correlation between skill, feedback, interaction and achievement on an initial analysis. How ever, when they reanalyzed earlier data, they found that teacher skill was negatively correlated to achievement. The research reported by Truax and Carkhuff (1967) and the McKeachie et al. research suggested that teachers who are perceived as high in empathy have students who perform better on academic measures, especially with respect to essay type questions. These findings were supported by this study. 127 Meridith (1969) suggested that the student's percep tion of learning was not necessarily related to actual per formance or to traditional course evaluations. Hypothesis 4C does not support this contention that students do not perceive themselves as learning differentially. The TA's facilitative conditions provided for higher performance in essay type questions for students who perceived the TA as high in empathy. This group did in reality perform better and rated themselves as learning. Another way of express ing this would be that students who perceived their objec tively rated high empathy TA's as low empathy and performed lower reported perceiving themselves as learning more or as much as the students of those TA's high in empathy and who perceived the TA as low. Hypothesis 5. Students who have TA's who are ob jectively rated higher in empathy and who rate the TA's as low in empathy will obtain lower grades than students whose TA's are objectively rated high and who also perceived their TA as high in empathy. Students who had TA's who were objectively rated higher in empathy and who perceived TA's as lower in empathy did not perform at a significantly different level than students of TA's who were objectively rated as high empathy and who were perceived as high empathy by their students, as measured by exam grade (p=.24). Thus, this hypothesis was not supported. 128 Hypothesis 6. Students whose TA's are objectively rated low on empathy and perceive their TA as high on em pathy will do better than those students of TA's objective ly rated low on empathy and perceived as low in empathy by the student. Students whose TA's were objectively rated low in empathy and perceived their TA as high in empathy performed significantly better on short answer (p<.001) and written (p <.01) criteria of academic achievement than students of TA's who were objectively rated low in empathy and who were perceived as low in empathy. The trend re mained the same for other academic achievement criteria, but was not significant at conventional levels of statis tical acceptance. Thus, the sixth hypothesis was supported in the present study, especially with written performance. These findings would support the contention of Kratochvril, Aspy, and Carkhuff (1967) and Bergin and Jasper (1969), that academic achievement outcome are re lated to the facilitative conditions and that empathy is necessary, but not the only variable needed to determine process and/or outcome in school learning. Hypothesis 7. Students of TA's who are rated low on empathy by objective methods as well as student's per ception will have the lowest achievement as measured by grades and will yield the lowest course evaluation ratings. Students of TA's who were objectively rated as low in empathy, and who perceived their TA's as low in empathy had 129 I the lowest mean grade when compared to the other groups. This was also the case for all course evaluation variables except difficulty. Therefore, Hypothesis 7 was supported from the findings of this study. Relationship of Empathy to Academic Achievement and Course Evaluation Students of objectively rated high empathy TA's performed at a higher level on all aspects of written per formance than students of objectively rated low empathy TA's. However, students who perceived high empathy in the TA's, performed better on multiple choice and short answer questions but did not perform better on essay questions. This suggests that short answer and multiple choice type performance is more related to perceived empathy whereas the most abstract level of performance used in this study (essay) is more related to the objective rating of TA em pathy. This conclusion is consistent with the proposal by McKeachie, Lin, and Mann (1971). These findings are not simply a general empathy result, but are independent of each other as demonstrated in the significant planned com parisons of Hypothesis 4 indicating that objective and perceived empathy interact in their effects. The differ ences in these findings between objective rating and per ceived rating suggest that there is a complex relationship between the student and the TA. This is further considered in the next major section. 130 On the course evaluation variables, high objective empathy of TA was most related to interpersonal relations or more specifically student-teacher interaction and skill as reflected in the TA's organization of the class. Stu dents who perceived high empathy in their TA's evaluated them higher on all course evaluation variables as was sug gested by Meredith (1969). Again, as in academic achieve ment, the effects of objective and perceived empathy are independent of each other. High objectively rated empathy leads to higher abstract learning and to more specific out come on course evaluations. The high relationship between perceived empathy and evaluations is unclear because of the identical relationships between liking and course evalua tions. These findings have serious implications which will be explored in the later section. These findings further suggest that students who do well on multiple choice type questions may reflect their performance in the way they evaluate the course (i.e., if they get higher exam scores, they give their TA's higher evaluation on the evaluative measures). On the other hand, a TA who is objectively rated high in empathy will facili tate the process of abstract thinking and will be judged as high in interpersonal interaction or responsiveness. This can be stated more specifically as: when the class is organized around the student's need for interaction and re sponsiveness , abstract thinking is facilitated and to a 131 greater extent than in other alternative classroom environ ments. Of course, one would expect that students who could do well on essays could at least do moderately well on direct memory type learning as would be evidenced by the multiple choice criteria. This was the case in the present study since students of objectively rated high empathy TA's did not perform lower on multiple choice questions than students of TA's objectively rated as low empathy. Relationship of Student Personality to Academic Achievement and Course Evaluation The two personality variables examined for this study were self-esteem and defensiveness. The level of students' defensiveness or need for social approval seemed to be related to the student's total exam score on the academic achievement criteria. This finding suggested that students low in defensiveness or need for approval performed significantly better on overall performance. The overall performance was significantly influenced by multiple choice scores since approximately two-thirds of the score came from choice items. The low defensive stu dents (i.e., students who had low need for social approval) performed better also on essay and multiple choice than high defensive students, although these differences did not reach conventional, statistically significant levels. The high defensiveness group, however, claimed that ! 132 jthey learned significantly more than the low defensiveness group on the course evaluation criterion (global learning). This finding suggests that defensiveness or need for ap proval in students is one variable determining students• ratings on course evaluations. In addition, this finding suggests that the high defensiveness group needs to make socially desirable evaluations, and they, therefore, report themselves as gaining more general knowledge— which is socially desirable in an academic setting. As mentioned in the previous chapter, the student's self-esteem was a dominant factor in evaluating academic achievement criteria when it is combined with perceived and objectively rated empathy. Students' self-esteem, by itself, is related most to academic achievement variables of multiple choice and total exam score as was the defen siveness variable. In addition, exam grade was also sig nificantly related to the student's self-esteem. In all cases, students with high self-esteem performed at a higher level than students with low self-esteem. Students with high self-esteem reported more spe cific learning on the course evaluation criteria. This suggests that the person with high self-esteem is able to perform better on conventional type exams (i.e. , applica tion of rote learning), to such an extent, that the mul tiple choice performance alone determines the whole grade. In addition, these findings indicate that students with 133 high self-esteem make an accurate assessment of their learning and report this assessment accurately since mul tiple choice usually reflects specific learning. While the interaction of self-esteem and defensive ness or need for approval does not appear to be related to course evaluation criteria, the interaction is related to academic achievement on the short answer variable. This finding suggests that both the high defensiveness, low self-esteem arid the low defensiveness, high self-esteem groups perform at a lower level than the high, high or low, low groups. This finding suggests that for the high de fensiveness, high self-esteem group, their high defensive ness with their high self-esteem generates the best per formance because the high self-esteem attenuates the re sistant aspects of their defensiveness and allows them to satisfy their need for social approval through academic achievement. The low defensiveness, low self-esteem group also performs well, because the lack of defensiveness or need for approval readily compensates for the lower self-esteem and makes these students compliant to the demands of their teachers and they do not defend against social pressures. Thus, it might be suggested that the interaction between defensiveness and self-esteem, when one is high and the other is low, might be detrimental or facilitative of performance on short answer and possibly written type tasks 134 depending on if the combination leaves the student resis tant, nondefensive, compliant, or accessible. Another way of describing this effect would be to say that the high-low combination of defensiveness and self-esteem interferes more with the students' abstract thinking processes than with his more concrete thinking processes. Interaction of Student and Teacher The discussion thus far has reviewed the findings of the present study, related these results to the initial hypotheses, examined the relationship of teacher empathy to student academic achievement and course evaluation, and also examined the relationship of student's personality to academic achievement and course evaluation. In order for a study to have continuing meaningfulness, it is important to reflect and examine the process which led to the results making some suggestions and conjectures as to what might be occurring between students and teachers in the univer sity classroom, given the present data. This section, "Interaction of Student and Teacher," will attempt to give some meaningful interpretations of the findings as they apply to the ongoing classroom learning situation in the university. As can be seen from the previous discussion, objec tive rating of TA empathy, student perceived TA empathy, student's defensiveness (need for approval) and self-esteem 135 |all play a vital role in the students' classroom learning experience. While the objective empathy rating of the TA definitely reflects facilitation of written answers and perceived empathy reflects more rote processes, the per sonality of the student also is important. Most apparent is that students who have objectively rated low empathy TA's, who perceived low empathy in TA's, and who are them selves either high in defensiveness or low in self-esteem perform poorest in the classroom. In addition, if the TA is perceived by the student as being low in empathy, the student will give the TA lower course evaluation, with the exception of course difficulty, where these TA's are rated as teaching a more difficult course. The two course evaluation criteria which were sig nificant for objective rating of TA with student's defen siveness were interpersonal relationship and TA's skill as reflected in organization. In both cases the objec tively rated low empathy TA with high defensive students were rated as lower in responsiveness than any other group. On the other hand, the high defensiveness student with the objectively rated high empathy TA rated his TA as having the highest responsiveness. This finding suggests that objectively rated low empathy TA's are perceived for what they are, noninteractive, and especially so by highly defensive students (i.e., students who needed approval from their TA's). 136 The students of low objective empathy TA's who were low in defensiveness rated the TA's skill as reflected in organization as low, while objectively rated high empathy TA's with high defensiveness students were rated as high in this category; but low defensiveness students gave low ob jectively rated TA's the lowest rating in organization. This finding, like the previous one suggests that objective rating of TA empathy affects course evaluations. This finding further suggests that objectively rated high empa thy TA's are recognized as offering something positive by even the high defensive person as reflected in course evaluations. The three-way interaction effect of objective empathy rating, student perceived empathy of TA, and stu dent defensiveness was presented in Table 17. The group with objectively rated low empathy TA's who perceived TA's as low in empathy and who had a high level of defensive ness, rated their TA's as lowest on the interpersonal re lationship variable. The previous paragraph presented the same result on this variable for the two-way interaction of objective rating and defensiveness. The three-way interaction indicates that this effect is particularly potent when the TA is also perceived as low in empathy. This finding suggests that the highly defensive person sees the lack of communication and TA responsiveness as the TA's responsibility because these students are 137 individuals who deny such negative characteristics in themselves. The students from this group, therefore, blame the TA and withdraw from the situation (do not participate academically) in order to protect their perceived vul nerability. In addition, there were interactions between em pathy and both student personality variables. The group that perceived TA's as having the most difficult class were those who were low in objective empathy, had high defensiveness and high self-esteem. This finding suggests that high defensiveness is not compensated for enough by the high self-esteem when the TA is low in empathy. The students therefore, regard the course as very difficult. On the other hand, the students who had objectively rated high empathy TA's, who were high in defensiveness, and who had high self-esteem rated the course as not being diffi cult. This finding suggests that students high in defen siveness and self-esteem perceive the course as most difficult if they have an objectively rated low empathy TA and as least difficult if they have an objectively rated high empathy TA. In addition, it appears that high defen siveness students with low objective empathy TA's make the socially appropriate response of evaluating the course as difficult to compensate for the lack of TA responsiveness, while high defensive students with high objectively rated TA's have no need to make this compensation. 138 The students who performed lowest on short answer type questions were those whose TA's were objectively rated as low in empathy, perceived the TA as low in empa thy and were themselves low in defensiveness (Figure 1). This suggests that if the TA is low in empathy, and the student perceives him as low in empathy and the student's defensiveness is low, then the student is vulnerable— without internal resources to resist a low empathy TA— and performs poorly. It appears, in this case, that the vulnerable student cannot get whatever he needs from the teacher in order to facilitate his performance. However, when these students are with a high ob jective empathy TA, they perform equally as well as other students who perceive the TA as high in empathy. Thus, if the TA is objectively high in empathy but is perceived as low in empathy, the student's performance will be facili tated if the student is low in defensiveness and thus open to the TA even though he does not perceive the TA as em phatic toward him. Apparently, if the TA is perceived as low in empathy, and the student is high in defensiveness, the high objective empathy TA's of this study were not able to facilitate their performance. The most valuable contribution to classroom learn ing appears to come from a combination of objective empathy, perceived empathy, and student's self-esteem (Table 1, Table 9, and Figure 2). Three conclusions can be drawn 139 from these results. First, with one exception, students perform equally well or better with an objectively rated high empathy TA than they do with an objectively rated low empathy TA. Second, students who are most facilitated by an objectively rated high TA are those students who per ceive that TA as high in empathy and themselves are high in self-esteem. Third, those students whose performance is not facilitated by objectively rated high empathy TA's are those students who perceive the TA as low in empathy and are themselves high in self-esteem. Therefore, the findings from this study suggest that self-esteem has a major role in academic performance. In addition, student's self-esteem has more affect upon all academic performance criteria than does defensiveness, except on short answer questions which is related more to defensiveness. This finding suggests that for concrete learning processes as well as abstract organizational thinking process, self-esteem is the major contributor to high academic performance when combined with both objective empathy and student perception of empathy. Similar effects seem to occur when objectively rated high empathy TA's are perceived as low empathy by either high defensive or high self-esteem students. Spe cifically, what occurs is that the high objectively rated TA's do not facilitate the performance of these students, but rather seem to hinder their performance. This idea is 140 !supported by the fact that the poorest performance occurs for those students who combine high defensiveness with high self-esteem when they perceive an objectively rated high empathy TA as low in empathy. This contrasts with the fact that when these high defensive, high self-esteem students perceived the objectively rated high empathy TA as high in empathy, they obtain better performance scores than any other groups. Learning in the Classroom This section attempts to go one step further than the last in examining some of the behaviors and responses which students make in the classroom which offer support and at the same time examines the generalizability of the student-teacher relationship from this study to other similar university settings. Some additional questions were added to the Course Comments Questionnaire in order to obtain some indications of other behaviors and classroom situations. Although these were not analyzed in this study, they do shed some light on the teacher-student process. Students were also asked if they felt that their TA had empathy for them as an individual student and if the TA demonstrated empathy toward the whole class. These questions produced inter esting findings in that high objective empathy TA's stu dents saw them as having empathy toward them as individuals 141 as well as toward the class, while low empathy TA's were seen as having more empathy toward the class, but not necessarily toward the student. When students were asked where they sat in the class, high perceived empathy stu dents reported sitting in front more than in the back. In addition to asking students how they liked their TA's, in a similar manner they were also asked how they liked their peers. There was a significant three-way interaction among objective rating of TA empathy, perceived empathy and self-esteem with peer attitudes (F=4.38; df=l,229 ; p< .03) . For those students who performed the best in the course and who seemed to have the best relationships with their TA's, they had the least positive peer relations al though these relations were on the positive direction of the scale. In contrast, those students who had the lowest performance as a result of having low objectively rated or low perceived empathy TA's reported the most positive re lationships with their peers if they themselves were high in self-esteem. This suggests that if students had re sponsive TA's and if they perceived the TA's as respon sive , they participated in that interaction which was sufficient to facilitate their performance. However, if the TA's were not high in empathy, then the students re sorted to their peers for interpersonal relationships, especially for those students who have the internal 142 processes (high self-esteem) which enable them to partici pate more in such peer relationships. This study provides evidence that the teacher's ability to sense, recognize, understand and to communicate understanding to the students is an important aspect of the academic learning situation. While it is true that objective rating and perceived empathy of TA's differenti ally affect academic performance, the student's personality is an integral determiner of his reaction to TA and thus of his performance. In summary, then, the student's reaction to the classroom environment and his personality, in combination with the TA's empathy, determine academic performance. Objectively rated high empathy TA's seemed to facilitate learning and especially if the students could accurately recognize that quality in their TA and if the students were able to be less vulnerable or defensive toward their environment. In addition, high self-esteem in students seemed to greatly enhance the probability of the students utilizing the classroom positively as a learning environ ment. From this study, the model situation for learning would consist of a high empathy teacher on both perceived and objective measures and a student with a combination of high self-esteem and low defensiveness. Low perceived empathy of TA, whether or not a real istic perception by the student, was especially deleterious 143 to learning. This was most true for students who were highly defensive and who had low self-esteem. The course evaluations for both groups reflect the students' reac tions to perceived empathy and their response to their perception more than to what may actually have happened. It is apparent, however, that students who have low empathy TA's and are high in defensiveness attempt to make socially acceptable responses on course evaluations where those re sponses are well defined (course organization), but are less able to evaluate TA's in a socially desirable manner when the social norms are less well defined (interpersonal relationships). Limitations of This Study and Implxcations for Future Research The findings from this research range from the very specific to general implications. On a more general level, this study suggests that the college and university system needs to decide what it wishes to evaluate. Under the as sumption that the educative system is established to en courage and facilitate learning, it would be appropriate to assess not only the academic performance of the student but also the facilitative qualities of teachers that affect learning. This study presented evidence that conventional course evaluation results are independent of academic per formance, which support findings of Hartley and Hogan 144 (1972) and which support similar suggestions made by McKeachie, et al. (1971) that conventional course evalua tions do not necessarily tap the needed modalities for determining teacher effectiveness. In addition, this in vestigation provides data which indicates what aspect of the student-teacher interaction accounts for course evalua tion results as traditionally obtained— student empathy of teacher and the students' self-reports on the amount of liking for the teacher. Liking, as well as student per ceived empathy of teacher has positive effects on all course evaluation variables. Thus, outcome on course eval uations may be the result of student liking for the profes sor or the result of teacher qualities such as empathy as perceived by the student; but to accept such evaluations on their face value— the present practice— is a dangerous procedure which may in fact be anti-intellectual because of the autistic interpretations which may be applied. Possibly, one of the limitations to the evaluation of perceived empathy is that the questionnaire administered was designed more for evaluating psychotherapeutic situa tions than for evaluating academic situations. In an attempt to remedy this dilemma, the author designed a questionnaire which contained fourteen items which were relevant to how she thought empathy in academic settings would be reflected. The results of the analysis of the Chang Academic Empathy Scale suggests that the questions 145 were relevant only to course evaluations when considering skill as reflected in: Knowledge (F=19.55, df=l,233; p<.001), interpersonal relationships (F=5.30, df=l,233? p <.02) , skill as reflected in organization (F=10.15; df=l,233; p<.002), specific learning (F=9.31; df=l,233; p<.003), and course relevance (F=3.70; df=l,233; p<.05). Thus, this scale appears less confounded since course dif ficulty and global learning were unrelated to it and on an a priori basis they should not be. However, it was not related to any achievement criteria reflecting the autistic relationship between performance and course evaluations. This scale was also significantly related to liking (F= 19.17; df=l,233; p<.001). These findings indicate the difficulty in composing a self-report scale which would not be confounded by liking, but would be related to student performance on conventional academic criteria, and thus reflecting the professor's facilitation of that perfor mance . One of the questions raised by the previous conclu sions is the extent to which perceived empathy is reflected in the course grade and how much of the course grade is reflected in the student's reported perception of the teacher. The way this problem was approached in this study was that a pre-empathy measure was administered prior to feedback concerning performance on the first exam. As was suggested in formulating the methodology, pre-empathy does 146 i not have any effects reflected in course evaluations or performance, while post-empathy measures are reflected in course evaluations, and in this study, in every variable. An analysis of the differences between pre- and post perceived empathy indicated that the variance they had in common was the same as what would be accounted for by a general liking variable. Again, however, one cannot be sure how much liking facilitates performance through facil itating a perception of empathy and how much the perfor mance itself causes the student to like or dislike his TA. That is, does performance cause liking, or does liking cause performance through empathy. Students' post perceived empathy of TA's, however, also has an effect on exam performance which suggests that there is more to stu dent perception of empathy of TA than a mere confounding with amount of student liking of the teacher. This indi cates that liking is not related to grades but post-empathy is related to grades. A question may be raised about the confounding of a post-empathy measure and its relationship to performance. That is, is higher performance causing higher ratings by students of perceived empathy, or is perceived empathy, as a result of adequate interactions with the TA, causing greater performance? The use of a liking scale does not help in determining this issue. However, the Chang Scale provides strong evidence that the causative variable is 147 perceived empathy. Although the Chang Scale was highly correlated to the perceived empathy scale (r=.7117? p £.001) and the liking scale (r=.4936, p< .001) , the Chang Scale was unrelated to academic achievement. However, the perceived empathy scale was related to academic achievement directly and in interactions as reported earlier. Another implication from these research findings is that it is not the teacher alone who must be analyzed. In addition to teacher variables, the personality of the student needs to be considered. This study has examined only two pertinent personality variables— defensiveness and self-esteem— yet has indicated many interactions which could greatly facilitate or hinder the learning process. These findings suggest that not only do student personality variables need to be more strongly considered, but also the specific relationship of teacher personality to student per sonality needs to be investigated also. A field research study such as this one, which examines personality variables of both students and teach ers as these variables affect learning is very important if not crucial to the educational process and to the very con cept of academic freedom. Although the inclusion of the student evaluation of the professor is laudatory, it is clear that the evaluation process is an intricate interac tion of what actually happens to the student, what the student perceives as happening, and how he feels about it. 148 The grading system in this course was also an ex periment in lowering tension and anxiety to facilitate academic performance. The way in which this study was de signed, it was impossible to measure the effects of per formance on the letter grades students ultimately received. It is apparent that more research needs to be conducted in this area before specific relationships can be proposed. It should be recalled that the final course grade was a composite of the assessments of the instructor, the TA, and the student himself. Theoretically, however, the type of learning millieu used in this study should facilitate rather than hinder the learning processes of the student. In addition, the specific types of results found for each of the academic performance variables suggests that each variable is measuring a different type of learn ing and thinking process. In addition the relationship and interaction of student personality and teacher does not appear to be a simple and direct one to these different types of measures. More research needs to be designed and conducted that actively considers the issues raised by differences between multiple choice, short answer, and essay questions in order to determine the explicit and im plicit meanings of the relationships of those to each other and to student and teacher characteristics. Most vital to this particular investigation are the implications for training graduate students to become facilitative teachers. The results overwhelmingly indicate that the objectively rated high empathy TA's seem to facil itate academic performance at a higher level than objec tively rated low empathy TA's. As Truax and Mitchell (1971) and Carkhuff and Berenson (1967) have suggested, training of helpers (in psychotherapy) can be accomplished in 100 hours or less. This appears to be a relatively minimal investment for training future teachers in methods which would greatly enhance their communication with stu dents. This in turn would help students to perform on a higher level and allow students to develop their abstract level of reasoning better than with the typical, present academic environment. Since the development of higher mental functioning and the encouragement of creative think ing is presumably one of the goals of the university and college system, the training of teaching assistants in an active manner which definitely facilitates such processes would seem to be a high order responsibility. A last point may be made about objective ratings of empathy. Although the variable was highly related to academic performance, it is clear that students do not all perceive these TA's as high in empathy. With some students these teachers will still be able to facilitate their performance. However, this facilitation may not be re flected in the subject course evaluation by the student. Finally, the study found that there were some students jwhose performance was hindered by these TA's. At some point, a responsive university system must assess these relationships in greater detail and develop responsible 'programs as a result of the empirical findings. APPENDIX A TRUAX-CARKHUFF EMPATHY TRAINING EXAMPLES 151 Stage 1 C: I wonder if it's my educational background or if it's me. T: Mhm. C: You know what I mean. T: Yeah. C: [Pause] I guess if I could solve that I'd know just about where to hit, huh? T: Mhm, mhm. Now that you know, a way, if you knew for sure, that your, your lack, if that's what it is.... I can't be sure yet. [C: No.] [T continuing]...is really so, that it, it might even feel as though it's something that you just couldn't receive, that it, if that would be it? C: Well— I— I didn't uh, I don't quite follow you— clearly. T: Well [pause], I guess, I was, I was thinking that— that you perhaps thought that, that if you could be sure that, the, uh, that there were tools that, that you didn't have, that, perhaps that could mean that these— uh— tools that you had lacked— way back there in, urn, high school.... [C: yah] [T continu ing]...and perhaps just couldn't perceive now and, ah.... C: Eh, yes, or might put it this way, um [pause]. If I knew that it was, um, let's just take it this way. If I knew that it was my educational background, there would be a possibility of going back. T: Oh, so, I missed that now, I mean now, and, uh.... C: ....and really getting myself equipped. Stage 2 C: You've got to explain so she can understand.... T: Mhm, mhm [in bored tone] C: Without— uh— giving her the impression that she can get away with it, too [excitedly]. T: Well, you've got a job satisfying all the things that— seem important, for instance being consistent, and yet keeping her— somewhat disciplined and tell ing her it's good for her [conversationally], C: There's where the practical application of what we have just mentioned comes into being [laughs]. T: Mhm, mhm [sounding bored]. C: And when it's a theoretical plan— T: Mhm. C: It's beautiful [shrilly]. T: Mhm— mhm. 152 153 C: But.... T: [Interrupting] Something else about it that I feel really dubious about [banteringly]— what you can really do on the practical level [inquiringly]— I sometimes say that's what— we're most encouraged about, too [mumbling]. C: [Chiming in loudly] Yes— uh— there are many— uh problems in our lives in the practical applications of— trying to be consistent [informatively]. Stage 3 C: I' here, an' uh— I guess that maybe I'll go through with it and [nervous laugh]— I'll have to— there's no use— T: [Interrupting] You mean you're here— you mean you're right here— I wasn't sure when you said that • • • • C: Well.... T: ... Whether you meant you were— I guess you mean you were in— this is your situation. [stumbling] C: [Interjecting] I'm in— I'm in— I'm in the stage of suffering— well, yes, I'm here too because of that. An'— uh— [sighs audibly] but, I can see where— uh — .... [T murmurs Mhm after every other word or so. ] T: [Filling in] You feel it's— you feel it's a pretty rough situation to be in? [Inquiringly] C: Sometimes I do, sometimes I don't [casually]. Stage 4 C: If— if— they kicked me out, I— I don't know what I'd do— because.... T: Mhm. C: I— I— I am really dependent on it [stammering]. T: Even though you hate this part- — you— say, "my God, I— I don't think I could possibly exist without it either." [C: Mhm.] And that's even the— that's the worst part of it [gently]. C: [Following lengthy pause] Seems that [catches breath] sometimes I— uh— the only thing I want out of the hospital— 's tuh have everyone agree with IY16 • • • • T: Mhm, hm. C: ....that's— I— I— I guess that if [catches breath] everybody agreed with me— that everybody'd be in the same shape I was [seriously, but ending with nervous laughter]. 154 T: Mhm, well, this is sort of like— uh— feeling about the friend who— didn't want to do what I wanted to do what I wanted to do; that— even here— if you agreed with me— this is what I want because if you don't agree with me, it means you don11 like me or something [reflectively]. C: Mmmmmm [thoughtfully]— it means that I'm wrong [emphatically, quick breathless laugh]. Stage 5 C: I fcjave her— her opportunity.... T: Mhm. C: ...and she kicked it over [heatedly]. T: Mm first time you ever gave her that chance, and— she didn't take it? [inquiring gently] C; No. She came back and stayed less than two weeks— a little more than a week— and went right straight back to it [excitedly]. [T answers "mhm" after each sentence.] T: Mhm, mhm...it feels like it's sort of thrown— right up in your face [gently]. C: ...If I didn't demand some kind of assurances— that, that things was over with [firmly]. T: Mhm, mhm, it would be— pretty stupid to— put your self in that— same position where it could be sort of— done to you all over again [warmly]. C: Well, it could be— yes. I would be very stupid [shrilly]. T: Mhm. C: ...because if it's not him— it might be someone else [emphatically]. Stage 6 T: You're sort of— comparing things you do, do, things you have done— with what it would take to be a priest— is that sort of— the feeling? [Very gently] C: [Following long pause] I don't know. [Meekly, then a lengthy pause] T: Suppose we mean right now feeling real guilty? [Softly] C: [Sighs audibly] Real small. [Very softly— pro tracted silence]— I can't see how I could feel any different— other than— feeling small or bad.... T: Mhm.... C: ....guilty [softly]. 155 T: Things you've done just— so totally wrong to you— totally bad— you can't help sort of— hating your self for it? [Assuming client's tone]— is that the sort of quality? [Very gently, almost inaudibly]. C: [Following pause] And yet right now I feel as though I want to laugh— be gay. T: Mhm. C: I don't feel anything else [monotonously]. T: [Speaking with client] Right at this— at this moment? C: Mhm. T: So— it's too much to really— feel— very miserable and show it? [Inquiringly] C: Yeah, yeah [urgently], I— I don't want to show it anyway [haltingly]. Stage 7 C: Th— the last— several years— it's been the other way around— I mean he'll say, "Well let's go— go do this or that," and— and I— sometimes I actually wanted to, but I'd never go because— I feel like I'm getting my little bit of revenge or something. T: He owed it to you, and— if he didn't come through, you'll just punish him now.... C: Yah. T: ....now it's too late or— something [very softly]. C: [Laughingly] Yah— that's uh— that's just the way I— uh— now it's too late— it's your turn to take your medicine now. [Assuming the therapist's tone] T: Mhm— I'm gonna treat you like— you've treated me [pause]— uh.... C: Mhm....it's pretty— that's a— pretty childish way to think, but I know uh— if I went tomorrow, I'd do it tomorrow— if I had the chance [defiantly]. If.... T: [Interrupting and overtaking the client] One part of you could say, "Well, this is stupid and child ish 'cause I— I want to be with him," and yet— an other part says, "no, you gotta make him pay for it— you want him dangling there now" [gently]. Stage 8 T: The way she wanted me and I was always terribly afraid that she wouldn't put up with me, or would put me out, out. [C: Yeah.] I guess I can get something else there, too, how I was always afraid that she didn't really care. E<U E h U 156 C: I still think that thought. [T: mhm.] Cause I don't know for sure. T: Mhm. And don't really know for sure whether she cares or not. C: [Pause] She's got so many other, uh, littler kids to think about. T: Mhm. C: That's why.... T: Maybe she likes them better or.... C: No it's not that, I think she likes us all. [Pause] I think seein' that I'm the black sheep but, uh, the only one that served time and, that— 'n got in the most trouble. Seein' that I hurt her so much, that's why I think she's starting ta— she just don't care for me anymore. [T interjects "mhm" after most completed thoughts.] T: You believe, maybe, "because I have hurt her so much, maybe she's fed up with me, maybe she's gotten to the point where she just doesn't care." [Long pause] Stage 9 C: ...uh— I've always been— so afraid— uh— show just how I— how I felt [T: Mhm] and I— and I— I think • * • • T: [Interrupting] Showing feelings is weak or— some thing. [Gently fading to near inaudibility]. C: Yeah— that's how it seems to me [lengthy pause] I know I— I've been in the T.V. room— and all of a sudden— I had the feeling that— I was going to start crying [almost tearfully]. : Mhm. : ....and— uh— I knew then I'd have to leave and go somewhere.... : Mhm. : ...where nobody was, so in case I did start crying that nobody's see me [bashfully], T: Mhm— it'd be just be— terrible to stand if you— if you ever did show this much feeling. C: The thing is— that— I'm— I'm afraid of— well, I'd be embarrassed afterwards. T: Mhm— this would be— just terrible— uh— a mein wouldn't cry, a grown-up wouldn't cry. C: Yeah. T: ....or at least.... C: [Filling in for T] At least without an apparent reason. T: Mhm. 157 [Long pause] An'— uh— an'— I— I don't have— an apparent reason. ....It wouldn't only be weak, but— be crazy or something. [Chiming in] Yeah. [Very positively] APPENDIX B QUESTIONNAIRE 158 Name St.# Sex: M F The questions below refer to your performance on the task which is on the following pages (Proficiency Measurement). 1. How well do you expect to do in comparison to the average American college student? Much below average _____ Below average _____ Average _____ Better than average _____ Much better than average _____ 2. How many squares do you expect to complete? (Fill in the number) _____ 3. How well would you like to do in comparison to the average American college student? Much below average _____ Below average _____ Average _____ Better than average _____ Much better than average _____ 4. How many squares would you like to complete? _____ 5. Do you think that this experiment will be a valuable experience for you? No personal value _____ Little personal value _____ Some personal value _____ Much personal value _____ 6. Do you think that it will be easy or difficult for you to work proficiently at this task? Easy _____ Difficult _____ 7. Are your feelings toward this task positive or negative? Positive _______ Negative____ 159 160 Directions: This is an experiment in proficiency measure ment. Notice above that each box has a number in the upper part and a mark in the lower part. Every number has a different mark. In the lower row each upper part has a number, the lower part is blank., THE AVERAGE AMERICAN COLLEGE STUDENT CAN COMPLETE 271 OF THESE SQUARES IN FIVE MINUTES. Please thumb through the next five pages to give yourself am idea of the task involved. Then turn back and answer the questions immediately preceding this task. DO NOT do the Proficiency Measurement Inventory task now Thank you. PROFICIENCY MEASUREMENT INVENTORY / 2 3 H - 5 6 7 8 10 I I !Z — x 3 L L i O A X — I I a ( / s 5 3 11 7 9 10 1 £ II / 10 APPENDIX C COURSE COMMENTS QUESTIONNAIRE (CCQ) 161 Directions; For the following statements please answer yes or no in answering how the course was conducted or how the course effects you. Please use the separate answer sheet, marking column A for "YES" and Column B for "NO." Be sure there is one blacken column for each line. 1. I developed my ability to identify main points or central issues. 2. Did the teaching assistant put the material across in an interesting way? 3. I can recognize when people are using faulty arguments in this field. 4. Did the teaching assistant make the students develop intellectual discipline? 5. I developed increased interest in the field. 6. Did the teaching assistant use enough examples to clarify the material? 7. Was the teaching assistant actively helpful when students had difficulty? 8. It is now easier for me to identify factors related to conclusion. 9. Would you take another section from this teaching assistant, if you didn't have to? 10. In conversation I can recall important information in this field. 11. Did students have to work hard to meet section re quirements? 12. I had discussions of related topics out of class. 13. Did the teaching assistant present material in a well organized fashion? 14. Did the teaching assistant appear sensitive to the students' feelings? 162 163 15. I developed my ability to combine arguments and draw conclusions. 16. Considering everything, I would rate the teaching in this course highly. 17. I can now understand relatively advanced presenta tions on the subject. 18. Were students required to conform to high intellectual standards? 19. I became aware of ways the subject is involved in my own life. 20. Were the methods.used for evaluating (tests, grading, projects, etc.) reasonable? 21. Was the teaching assistant fair in his dealings with students? 22. I developed significant skills in the field. 23. Would you recommend this course to a friend who didn't have to take it? 24. I can identify values that enter into making judg ments in this field. 25. Was the amount of work required appropriate for the credit received? 26. I increased my concern for community projects related to the course. 27. Were the objectives of the course clear? 28. Were students free to ask questions, disagree, express their ideas, etc.? 29. I developed my ability to function creatively. 30. Considering everything I would rate this course highly. 31. I becaime able to analyze new and complicated material in the field. 32. Were the assigned readings and/or problems at an appropriate level? 164 34. Were the standards for grading clearly communicated to you? 35. Did the teaching assistant tell the students when they had done particularly well? / * ■ 1 REFERENCES 165 Anderson, S. C. Effects of confrontation by high and low functioning therapists. Journal of Counseling Psychology, 1968, 15, 411-416. Argyle, M. The psychology of interpersonal behavior. Baltimore: Penguin, 1967. Aspy, David. Helping and intellectual functioning. In R. R. Carkhuff, The development of human resources, education, psychology, ana social change. New York: Holt, Rinehart & Winston, Inc.,1971. Aspy, David N. and Roebuck, Flora N. An investigation of the relationship between student levels of cogni tive functioning and the teacher's classroom be havior. Journal of Educational Research, 1972, 65 (8), 366-366. Bendig, A. W. Student achievement in introductory psy chology and student ratings of competence and empathy of the instructor. Journal of Psychology, 1953, 36, 427-433. Berger, S. E. The self-deceptive personality. Unpublished Doctoral dissertation, University of Miami, Coral Gables, Florida, 1971. Berger, Stephen E. Therapist variables in psychotherapy. Unpublished paper, University of Miami, Coral Gables, Florida, 1969. Bergin, A. E. and Jasper, L. G. Correlates of empathy in psychotherapy: A replication. Journal of Abnormal Psychology, 1969, 74, 447-481. Berscheid, Ellen and Walster, Elaine H. Interpersonal attraction. Menlo Park, California: Addison- Wesley Publishing Co., 1969. Betler, Larry E., Johnson, Dale T., Neville, Charles W. and Workman, Samuel N. Accurate empathy and the A-B dichotomy. Journal of Consulting and Clinical Psychology, 1972, 3jS (3) , 372-375. Bom, David G., Gledhill, Stephen M. and Davis, Michael L. Examination performance in lecture discussion and personalized instruction courses. Journal Applied Behavior Analysis, 1972, 5 (1) 33-43"! 166 167 Bramel, D. Interpersonal attraction, hostility, and per ception. In Judson Mills (Ed.), Experimental social psychology. New York: Macmillan, ' 1968. Brummon, D. L. and John, Eve S. Changes over client- centered therapy evaluated on psychoanalytically based thematic apperception test scales. In C. R. Rogers and R. F. Dymond (Eds.), Psychotherapy and personality change, Chapter 11. Chicago: University of Direngo Press, 1954. Byrne, D. Interpersonal attraction and attitude simi larity. Journal of Abnormal Social Psychology, 1961, 62_, 713-715. Byrne, D. and Clore, G. L. Predicting interpersonal at traction toward strangers in three different stimulus modes. Psychonomic Science, 1966, 4 239-240. (a) Byrne, D. and Griffith, W. A developmental investigation of the law of attraction. Journal of Personality and Social Psychology, 1966, 4 _, 699-‘ 7ol. (FJ Byrne, D. and Griffith, W. Similarity Versus liking: a clarification. Psychonomic Science, 1966, 6, 295-296. (c) Calder, Peter. The effects of a system of matching clients to counselors on the level of counselor's empathy, warmth, and genuineness. (Doctoral dissertation, Indiana University) Dissertation Abstracts Inter national , 1972, 32 ^ (11—B) , 6631. Carkhuff, R. R. The development of human resources educa- tion, psychology, and social change. New York: Holt, Rinehart & Winston, Inc., 19 71. Carkhuff, R. R. The counselor's contribution to facilita- tive processes. Carkhuff, Robert R. Helping and human relations: a primer for lay and professional helpers. Vol. II. Practice and research. New York: Holt, Rinehart & Winston, Inc., 1969. Carkhuff, Robert R. and Berenson, Bernard G. Beyond counseling and psychotherapy. New York: Holt, Rinehart & Winston, 1967^ 168 Carkhuff, R. R. and Truax, C. B. Journal Consulting Psy chology , 1965, 29 , 333-336. Cartwright, Rosalind and Lerner, Barbara. Empathy: Need to change and improvement with psychotherapy. Journal Consulting Psychology, 1963, 2^7, 138-144. Carr, A. C. Evaluation of nine psychotherapy cases by the Rorschach. Journal of Consulting Psychology, 1949, 13 (3), 196-265. Chang, A. F. Some psychological and sociological corre lates of student activists and non-activists with emphasis on minority groups: An exploratory pilot study. California State Psychological Association Meetings, 1973. Chang, A. F. , Birren, J. E. and Wine, D. W. Psychologist as consultant: The measurement of perceived empa- they of therapist and change in self-esteem among patients in group psychotherapy using ward person nel. California State Psychological Association, Oakland, California, 1973. Clyde, D.S. Multivariate Analysis of variance on large computers. Coral Gables: Clyde Computing System, 1969. Cohen, A. R. Some implications of self-esteem for social influence. In C. J. Houland and L. I. Janis (Eds.), Personality and permisibility. New Haven: Yale University Press, 1957. Coombs, A. W. and Soper, D. W. The perceptual organization of effective counselors, journal of Counseling Psychology, 1963, 10^, 222-226. Coopersmith, S. The antecedents of self-esteem. San Fran cisco: Freeman, 1967. Crowne, D. and Marlowe, D. The approval motive. New York: Wiley, 1964. Crowne, D. and Marlowe, D. A new scale of social desir- ability independent of psychopathology. Journal of Consulting Psychology, 1960, 2 A_, 349-354. Crowne, D. and Stephens, M. W. Self-acceptance and self- evaluative behavior: A critique of methodology. Psychological Bulletin, 1961, 5£, 104-121. 169 Dickenson, W. A. and Truax, C. B. Group counseling with college underachievers: Comparisons with a control group and relationship to empathy, warmth, and genuineness. Personnel Guidance Journal, 1966, 45, 243-247. Ford, L. H. Jr. and Hersen, M. Need approval, defensive denial and direction of aggression in a failure- frustration situation. Journal of Personality and Social Psychology, 1967, 6, 228-232. Edwards, A. L. The measurement of personality traits. New York: Holt, Rinehart & Winston, Inc., 1970. Edwards, A. L. The social desirability variable in per sonality assessment and research. New York: Dryden Press, 1957. Edwards, A. L. and Walker, J. N. A short form of the MMPI: The SD scale. Psychological Reports, 1961, 8, 485-486. Festinger, L. A theory of social comparison processes. Human Relations, 1954, 1 _ , 117-140. Hartley, Eugene L. and Hogan, Thomas P. Some additional factors in student evaluation of courses. American Educational Research Journal, 1972, 9 _ (z) , 241-250. Heider, F. The psychology of interpersonal relations. Wiley, 1958. Hess, E. W. and Polt, J. M. Pupil size as related to interest value of visual stimuli. Science, 1968, 132, 349-350. Ingwell, Richard H. The relationship between perceived empathy, genuiness, and warmth conditions of charisma and external counseling conditions. Unpublished doctoral dissertation. University of Missouri, 19 71. Johnson, Relda J. An exploration of relationships between and among empathy, trust, and ego stage development in the adult learner. Unpublished doctoral disser tation, Michigan State University, 1971. Isaacson, Robert I, McKeachie, Wilbert J., Milholland, John E., Lin, Yi-Guang, Hofeller, Margaret, Baerwaldt, James W. and Zinn, Karl L. Dimensions of student evaluations of teaching. Journal of Educational Psychology, 1964, 55_, (6) 344-351. 170 Jacobson, L. I., Berger, S. E. and Millham, J. Individual differences in cheating during a temptation period when confronting failure. Journal of Personality and Social Psychology, 1970, 15, 48-5F7 Jacobson, L. I. and Ford, L. H. Need for approval, defen sive denial and sensitivity to cultural stereo types. Journal of Personality, 1966, 3 4 _ , 596-609. Kratochvril, D., Aspy, D. and Carkhuff, R. R. The differ ential effects of absolute level and direction of growth in counselor functioning upon client level of functioning. Journal of Clinical Psychology, 1967, 23, 216-217. Kiesler, C. A. and Goldberg, G. N. Multidimensional ap proach to the experimental study of interpersonal attraction: Effect of a blunder on the attractive ness of a competent other. Psychological Reports, 1968, 22^ 693-705. Lawson, Dene R. Indications of teacher ability to relate to students. Unpublished doctoral dissertation, Stanford University, 19 71. Lorr, M. and McNair, D. M. Methods relating to evaluation of therapeutic outcome. In L. A. Gottachalk and A. H. Auerbach (Eds.), Methods of research in psychotherapy. New YorJcl Appleton-Century Crofts, 196(5. McKeachie, Wilbert James. Teaching tips. Ann Arbor, Michigan: George Wahr, 1960. McKeachie, W. J., Lin, Yi Guang, Mann, William. Student ratings of teacher effectiveness: Validity studies. American Educational Research Journal, 1971, 8 (3), 435-446. Meredith, Gerald M. Dimensions of faculty-course evalua tion. Journal of Psychology, 1969 , 73^, 27-32. Mullen, J. A. An investigation of the variable of liking in therapy: Its relation to variables of outcome, empathy and therapist experience. Unpublished doc toral dissertation. Newcomb, T. M. The acquaintance process. New York: Holt, Rinehart & Winston, 1961. 171 Nisbett, R. E. and Gordon/. A. Self-esteem and suscepti bility to social influence. Journal of Personality and Social Psychology, 1967, 5> 268-2^6. Pierce, Richard and Drasgow, James. Teaching facilitative interpersonal functioning to psychiatric inpa tients. Journal of Counseling Psychology, 1969, 16, 295-2!T?r! Remitters, H. H. On students' perception of teacher effec tiveness . In W. J. McKeachie, The appraisal of teaching in larcje universities. Ann Arbor: University of Michigan Press, 1959. Rogers, C. R. The necessary and sufficient conditions of therapeutic personality change. Journal of Con sulting Psychology, 1957, 2JL, 95-123” Rogers, Carl R. A theory of therapy, personality, and interpersonal relationships as developed in the client-centered framework. In Sigmond Koch (Ed.), Psychology; A study of a science. Vol. Ill, Formulations of the personal ancTthe social context. New York: McGraw-Hill, 1959. Silverman, I. Differential effects of ego threat upon persuasibility for high and low self-esteem sub jects. Journal of Abnormal and Social Psychology, 1964, 6 9 _ , 567-572. fa) Silverman, I. Self-esteem and differential responsiveness to success and failure. Journal of Abnormal and Social Psychology, 1964, 69, ll5-ll$. (El Stoffer, Dean L. Investigation of positive behavioral changes or a function of genuineness, nonpossessive warmth, and emphatic understanding. Journal of Educational Research, 1970, 6J3 (5), 225-228. Strickland, B. R. and Crowne, D. P. Need for approval and the premature termination of psychotherapy. Journal of Consulting Psychology, 1963, 2J7, 95-101. Strupp, H. H., Wallach, M. S., Wogan, M. and Jenkins, Joan W. Psychotherapists' assessments of former patients. Journal of Nervous Mental Disorders, 1963, 137, 2T2-230. 172 Strupp, H. S., Wallach, M. S. and Wogan, M. Psychotherapy experience in retrospect: Questionnaire survey of former patients and their therapists. Psycho logical Monograph, 1964, 7j[(ll) (Whole No. 588). Truax, C. B. and Carkhuff, R. R. Toward effect counseling and psychotherapy. Chicago: Aldine Co., 1967. Truax, C. B., Carkhuff, R. R. and Kodman, F., Jr. Rela tionship between therapist-offered conditions and patient change in group psychotherapy. Journal of Clinical Psychology, 1965, 21, 327-329. Truax, Charles B. and Mitchell, Kevin M. Research on cer tain therapist interpersonal skills in relation to process and outcome. In Allen E. Bergin and Sol L. Garfield (Eds.), Handbook of psychotherapy and behavioral change. New York: John Wiley, 1971. Truax, C. B., Wargo, D. G., Frank, J. D., Imber, S. D., Battle, Carolyn C., Hoehn-Saric, R., Nash, E. H., and Stone, A. R. Therapist empathy, genuineness, and warmth and therapeutic outcome. Journal of Consulting Psychology, 1966, 22^, 331-334. Truax, C. B., Wargo, D. G., and Silber, L. D. Effects of high accurate empathy and non-possessive warmth during group psychotherapy upon female institution alized delinquents. Journal of Abnormal Psychology, 1966, 71, 267-274. Van der Veen, F. Basic elements in the process of psycho therapy: A research study. Journal of Consulting Psychology, 1967, 31_, 295-303. Vargas, M. Changes in self-awareness during client- centered therapy. In C. R. Rogers and R. F. Dymond (Eds.), Psychotherapy and personality change. Chi cago : University of Chicago Press, 1954. Wagner, H. M. and Mitchell, K. M. Relationship between’ perceived instructor's accurate empathy, warmth and genuineness and college achievement. Discus sion papers. University of Arkansas: Arkansas Rehabilitation Research and Training Center, 1969. Wyrick, T. J. and Mitchell, K. M. Relationship between accurate empathy, warmth, and genuineness and per ceived resident assistant effectiveness. Discus sion papers. University of Arkansas: Arkansas Rehabilitation Research and Training Center, 1969. 173 Ziller, R. L., Kagey, J. , Smith, M. D. C. and Lang, B. Self-esteem: A self-social construct. Journal of Consulting and Clinical Psychology, 1969, 33 84-95.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A Critical Test Of The Dominant Motive Of High Need-Approval Persons

PDF

The Effects Of Einstellung On Visual Perception Of Random Shapes: A Multidimensional Scaling Analysis

PDF

Relationship Of Personality And Task Demands To Self-Disclosing Behavior And Psychological Health

PDF

Mental Imagery As A Function Of Muscular Tension And Suggestion

PDF

The Effects Of Self Vs. Ideological Advocacy On The Self-Esteem And Endorsement Of Black Power Ideology Of Black College Students

PDF

Disponibilite In The Therapeutic Relationship

PDF

Relationship Between Perceptual Style, Achievement, And Childrearing Practices In Elementary-School Boys And Girls

PDF

Factors Of Adaptation And Rehabilitation In Home Hemodialysis

PDF

The Development And Evaluation Of Three Therapeutic Group Interventions For Widows

PDF

The Differential Effectiveness Of External Versus Self-Reinforcement On The Acquisition And Performance Of Assertive Responses

PDF

Self-Sacrifice, Cooperation, And Aggression In Women Of Varying Sex-Role Orientations

PDF

A Multiple Investigation Of Child-Rearing Attitudes

PDF

An Examination Of Positive And Negative Reinforcement In Classical And Operant Conditioning Paradigms In The Primary Psychopath

PDF

Weight Reduction As A Function Of The Timing Of Reinforcement In A Covertaversive Conditioning Paradigm

PDF

The Effects Of Anxiety And Threat On Self-Disclosure

PDF

The Effect Of Nonspecific Factors On The Modification Of Smoking Behaviorin Treatment Follow-Up

PDF

Modification Of Low Self-Confidence In Elementary-School Children By Reinforcement And Modeling

PDF

Attention, retention, and incentive processes in observational learning

PDF

Sound Blending: The Relationship Of Inter-Letter Interval, Consonant Category, And Stimulus Word Frequency To Word Identification

PDF

The Effect Of Homogeneity And Heterogeneity, In Terms Of Group Experience, On Success In Group Among Counseling Students

Asset Metadata

Creator Chang, Alice Frances (author)

Core Title The Relationship Of Teacher Empathy And Student Personality To Academic Achievement And Course Evaluation

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Education, general,OAI-PMH Harvest,Psychology, clinical

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Berger, Stephen E. (committee chair), Marston, Albert R. (committee member), Smith, Robert A. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-827367

Unique identifier UC11364283

Identifier 7409059.pdf (filename),usctheses-c18-827367 (legacy record id)

Legacy Identifier 7409059

Dmrecord 827367

Document Type Dissertation

Rights Chang, Alice Frances

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA