_" :gJ; [03 Wm? “301371 %g{ ’ G}... C3 32%;:013 ABSTRACT THE RELATIONSHIP BETWEEN COUNSELOR-TRAINEES' PERCEPTIONS OF FILMED CLIENT-PERCEPTS AND THEIR COUNSELING INTERVIEW BEHAVIOR WITH SPECIFIC CLIENTS by Jack D. Thorsen The purpose of this investigation was to develop an instrument for measuring counselorbtrainee perceptions of filmed c1ient~percepts and to study the relationship between counselor-trainees? perceptions of the filmed clientepercepts and these counselor-trainees’ subsequent counseling interview behavior with particular clients by: (1) creating a set of filmed counseling segments, (2) developing an objective-meanS'for tabulating counselor- trainee perceptions of filmed client-percepts using a semantic differential format, (3) taping samples of counselor-trainee interview behavior, and (4) comparing statistically counselor-trainee perceptions of the filmed client-percepts with selected interview behavior. The subjects in this study were thirty graduate students enrolled in the 1965-66 academic year guidance and- counseling institute at Michigan State University (Group I) and thirty graduate students enrolled in the 1966 summer practicum at Michigan State University (Group II), For all Jack D. Thorsen analyses of the data Group I was considered the experimental group and Group II the replicating group. The Person Perception Test (PPT), developed specifically for this study, was used as the independent variable. The PPT consisted of five filmed counseling segments judged on a thirty scale graphic semantic differ— ential. Factor analysis of the thirty scales revealed four judgmental dimensions: (1) evaluative, (2) emotional, (3) potency, and (4) evaluative-activity. Each scale was assigned a factor and the scale positions were numbered 1 through 7. Test-retest reliability estimates ranging from .55 to .80 were computed by product-moment (3). Average absolute deviations of less than one scale unit for each of the four factors were found. The Counselor Response System (CRS) was used as the dependent variable. Three judges, working independently and from interview tapes, rated the counselor-trainees' reSponses along the following dichotomous dimensions (inter- judge reliability estimates in parentheses): (1) Content: Follow-Shift (.95), (2) Control: Expansive-Restrictive (.92), (3) Referent: Client-Other (.73), and (4) Reinforcing-Nonreinforcing (.94). Hypotheses tested for this study were grouped under two headings: (1) Person Perception Test hypotheses and (2) validity hypotheses. The first set of hypotheses tested Jack D. Thorsen the internal consistency of factor items with total factor scores and with client-percept factor scores; the differences among client-percept factor scores; and the theoretical notion of the perceiver‘s "generalized meaning."' The second set of hypotheses tested the validity of the PPT by analyzing its relationships with the.CRS dimensions. Major Findings The following were the major findings of the study: 1. The data collected for this study did not support the validity hypotheses.and, therefore, did not support the predicted relationship between counselor-trainee53-perceptions of filmed client- percepts and their subsequent interview behavior with a specific client.' Counselor-trainees.judge clients along several dimensions with an evaluative dimension being the most dominant, followed by emotional and potency dimensions. Counselor—trainees value more highly those clients who:are most.active as evidenced by the dependence of the activity factor on the evaluative factor for Group I and by its coalescence with the evaluative factor for Group II. The PPT is an objective means for tabulating counselor-trainee perceptions of filmed client- Jack D. Thorsen percepts as evidenced by the ease with which various scoring procedures can be applied. The PPT total factor scores represent counselor- trainees' generalized meaning for the client- percepts as evidenced by.the positive relation- ships found between client-percept factor scores and total factor scores. If a perceiver's comparative particular judgment of a person percept were known, the perceiver's total judgment of the person percept could be predicted. But his comparative total judgment of people in general could not. If a perceiver's comparative total judgment of a person percept were known, the perceiver's comparative total judgment of person percepts in general could be predicted. But his comparative total judgment of another person percept could not. The PPT needs.further refinement (before discarding it or the theory upon which it was based) in an attempt to bring out better possible relationships. THE RELATIONSHIP BETWEEN COUNSELOR-TRAINEES' PERCEPTIONS OF FILMED CLIENT-PERCEPTS AND THEIR COUNSELING INTERVIEW BEHAVIOR WITH SPECIFIC CLIENTS By Jack D. Thorsen A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOHPY College of Education Department of Counseling, Personnel Services, and Educational Psychology 1968 {a 1n: med ,- I V 2:: I L" ) 1' 1}}. I“; /’ /‘(‘ .-’ .5 / c? k i i." r ’2’ ACKNOWLEDGMENTS The author wishes to express his appreciation to the many people who generously gave their time, experience, and support. To Dr. Richard Rank, chairman, who gave direction in the planning and reporting of this study and who remained with the project to the end despite his personal move to Georgia. To Drs. William Farquhar, Randall Harrison, and Fred Vescolani for their interest and their willingness to serve as members of my guidance committee. To Dr. Robert Campbell and Nancy Abe for graciously being the unseen and unheard counselors; and to Peggy, Bill, Bob, Lynn, and Terry for being not only seen but heard. Without them there would have been no film. To John Bogner, Barry Stearns, and William Zuhl for serving as judges of interview material and as editors and critics of the text in its various stages of development. To Dr. Gwen Norrell who was a very special person in my doctoral program. To my wife Mary and to my childrenéeJack, Dana, Kent, Denise, and Dale-~who have more patience than I ever imagined and who are as much a part of this project as anyone. ii TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . LIST OF TABLES . . . . . . . . LIST OF FIGURES . . . . . . . . LIST OF APPENDICES . . . . . . . Chapter ‘ I. INTRODUCTION . . . . . Purpose . . . . General Relationship Hypothesis . . . Theory. . . . . . Person Perception Defined The Meaning of Perception Concept Formation and Attribute Seeking The Semantic Differential . . . The Semantic Differential Technique Analysis of Semantic Differential Data Osgood's Meaning of "Meaning" Theory and the Semantic Differential Definitions of Relevant Terms . . Overview . . . . . . . II. REVIEW OF RELATED LITERATURE . . . . . Person Perception . . . . . . . The Semantic Differential . . . Stability of Factorial Structures Scale- -Checking Style Behavioral Validity of the Semantic Differential A Film Test of Counselor Perceptions . . Summary . . . . . . . . . III. DESIGN. . . . . . . . . . The Subjects . . . . . . The Person Perception Test. . . . . iii Page ii vi xii xiii 0001 13 19 19 20 20 31 43 48 52 52 S4 Chapter Selection of Client-Percepts Filming of Client-Percepts Selection of Bipolar Adjective Scales Factor Analysis Scoring Procedures Reliability Counselor Response System . Hypotheses. Person Perception Test Hypotheses Validity Hypotheses Collection of Data. . . . The Independent Variable The Dependent Variable Analysis Of Data . . . . Person Perception Test Hypotheses Validity Hypotheses Exploratory Analyses Summary . . . . . . IV. ANALYSIS OF DATA: PERSON PERCEPTION TEST HYPOTHESES . . . . . . . Internal Consistency . Evaluative Factor Emotional Factor Potency Factor Evaluative-Activity Factor Client-Percept Meanings . . Evaluative Factor Emotional Factor Potency Factor Evaluative-Activity Factor Person Perception Theory . . . . Evaluative Factor Emotional Factor Potency Factor Evaluative-Activity Factor Discussion. . . . . . Summary . . . . V. ANALYSIS OF DATA: VALIDITY HYPOTHESES. . Approach-Avoidance: Evaluative Factor. . Approach-Avoidance: Interrelated Factors . Scale-Checking Style . . . . . . Stereotopy and Rigidity . . . . Exploratory Analyses . . . . . . Discussion. . . . . . . . . Summary . . . . . . . . . iv Page 79 82 91 93 96 99 100 111 123 135 138 143 144 145 149 150 154 158 164 Chapter Page VI. SUMMARY AND CONCLUSIONS . . . . . . 168 Summary . . . . . . . . . 168 Conclusions . . . . . . . . 177 Future PPT Research: A Suggested Procedure 180 BIBLIOGRAPHY. . . . . . . . . . . 183 APPENDICES . . . . . . . . . . . 192 LIST OF TABLES Table Page 3.1 Sample page of the Person Perception Test . . 65 3.2 Factor analysis of Person Perception Test scales: Group I . . . . . . . . 67 3.3 Factor analysis of Person Perception Test scales: Group II . . . . . . . . 68 3.4 Person Perception Test factor composition and factor loading for Group I and Group II counselor-trainees. . . . . . . . 71 3.5 Person Perception Test test- retest reliability estimates using product- moment (r) for each client- -percept factor score (all- scoring procedures) for Group II . . . . . . 75 3.6 Group II test-retest factor average absolute deviations by client-percepts and by total expressed in scale units . . . . . . 80 4.1 Items significantly related to total factor score ("1 -7" procedure) by both groups: Evaluative factor . . . . . . 102 4.2 Item/client- percept score ("1- 7" procedure) correlation coefficients (r) for both groups: Evaluative factor . . . . . . . . 103 4.3 Items significantly related to total factor score ("l-7” procedure) by both groups: Emotional factor . . . . . . . . 105 4.4 Item/client-percept score (”l-7" procedure) correlation coefficients (1) for both groups: Emotional factor . . . . . . . . 106 4.5 Item/client-percept score ("l-7" procedure) correlation coefficients (3) for both groups: Potency factor. . . . . . . . . 108 vi Table 4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13 4.16 4.17 Items significantly related to total factor score ("l-7" procedure) by both groups: Evaluative-activity factor. . Item/client- percept score ("1- 7" correlation coefficients (r) for Evaluative- activity factor. . procedure) both groups: Factor means by client-percepts for both groups. 0 O O O O 0 Average semantic meanings for the client- percepts by factor for both groups . . Differences among client-percept ("l-7" procedure) by analysis of both groups: Evaluative factor Differences among client-percept ("l-7" procedure) by analysis of both groups: Emotional factor. Differences among client-percept ("l-7" procedure) by analysis of both groups: Potency factor . Differences among client-percept ("l-7" procedure) by analysis of both groups: factor scores variance, for factor scores variance, for factor scores variance, for factor scores variance, for Evaluative—activity factor . C1ient-—percept factor score comparisons by the rank- difference method (rho), for both groups: Evaluative factor . . . . 0 O 9 Relationships between client-percept factor scores and total factor score using the rank- difference method (rho), for both groups: Evaluative factor . . . . C1ient-—percept factor score comparisons by the rank- difference method (rho), for both groups: Emotional factor . . . . Relationships between client-percept factor scores and total factor score using the rank- difference method (rho), for both groups: Emotional factor . . . . vii 0 Page 110 112 114 115 117 119 121 122 126 128 128 130 Table Page 4.18 C1ient-—percept factor score comparisons by the rank- difference method (rho), for both groups: Potency factor. . . . . . . . . 130 4.19 Relationships between client-percept factor scores and total factor score using the rank- difference method (rho), for both groups: Potency factor. . . . . . . . . 130 4.20 Client-percept factor score comparisons by the rank-difference method (rho), for both groups: Evaluative-activity factor. . . . . . 132 4.21 Relationships between client-percept factor scores and total factor score using the rank- difference method (rho), for both groups: Evaluative- activity"f3ctor. . . . . . 134 4.22 Summary of results of Person Perception Test hypotheses tested . . . . . . . . 139 5.1 Correlation coefficients for hypothesized relationships between Counselor Response System scores and Person Perception Test evaluative scores using the "1~7" procedure . 146 5.2 Multiple correlation coefficients between Counselor-Response System scores and Person Perception Test scores using the "1-7" procedure . . . . . . . . . . 148 5.3 Multiple correlation coefficients between Counselor Response System scores and Person Perception Test scores using the £2356 procedure . . . . . . . . . 151 5.4 Multiple correlation coefficients between Counselor Response System scores and Person Perception Test scores using the variance scoring procedure . . . . . . . . 153 5.5 Person Perception Test total factor scores (f1 7 procedure) compared with Counselor Response System dimensions by product- moment (a) and by multiple correlation (R) . . . 155 5.6 Person Perception Test factor variance scores compared with Counselor Response System dimensions by product-moment (r) . . 157 viii Table 5.7 5.8 B.8 B.9 Counselor Response System dimension means and standard deviations for the experimental, the replicating, and the DeRoo groups . . . Summary of results of validity hypotheses tested. . . . . . . . . . Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group I by subject: Evaluative factor . . . Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group II by subject: Evaluative factor . . . Raw scores for the Person Perception Test by client- -percept, total, and variance, using the "1-7" scoring procedure for Group I by subject: Emotional factor . . . Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group II by subject: Emotional factor . . . Raw scores for the Person Perception Test by client-percept, total, and variance, using the ”1-7" scoring procedure for Group I by subject: Potency factor . . . Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group II by subject: Potency factor. . . . Raw scores for the Person Perception Test by client-percept, total, and variance, using the ”1-7" scoring procedure for Group I by subject: Evaluative-activity factor . Raw scores for the Person Perception Test by client- -percept, total, and variance, using the "1- 7" scoring procedure for Group II by subject: Evaluative- -activity factor. Frequency scores for the Person Perception Test by client-percept and total for Group I: Evaluative factor . . . . . . ix Page 163 165 202 203 204 205 206 207 208 209 210 Table B.10 Frequency scores for the Person Perception Test by client-percept and total for Group II: Evaluative factor . . . . . . . Frequency scores for the Person Perception-' Test by client-percept and total for Group I: Emotional factor . . . . . . . Frequency scores for the Person Perception Test by client-percept and total for Group II: Emotional factor . . . . . . . Frequency scores for the Person Perception Test by client-percept and total for Group I: Potency factor. . . . . . . . Frequency scores for the Person Perception Test by client-percept and total for Group II: Potency factor. . . . . . . . Frequency scores for the Person Perception Test by client-percept and total for Group I: Evaluative-activity factor. . . . . Frequency scores for the Person Perception Test by client-percept and total for Group II: Evaluative-activity factor. . . . Person Perception Test: distribution of responses by item . . . . . . Person Perception Test-retest item reliability estimates for Group II by factor . . Frequency scores for the Counselor Response System dimensions for Group I . . . Frequency scores for the Counselor Response System dimensions for Group II. . . . Counselor Response System: distribution of responses for both groups . . . . Item/total factor score correlations using the ”1-7" procedure for both groups: Evaluative faCtoro O O O O 0 O O O Item/total factor score correlations using the "1-7" procedure for both groups: Emotional factor. . . . . . . . . . Page 211 212 213 214 215 216 217 219 230 247 248 250 252 253 Table Page I.3 Item/total factor score correlations using the "1-7" procedure for both groups: Potency factor. O O O O O O O O O O 253 1.4 Item/total factor score correlations using the "1-7" procedure for both groups: Evaluative- activity factor . . . . . . . . 254 J.l Person Perception Test total factor scores ("l-7"procedure) compared with Counselor Response System dimensions by product- moment (I)° . . . . . . . . . 256 K.l Person Perception Test total factor scores (£2356 procedure) compared with Counselor Response System dimensions by product- moment (I). . . . . . . . . . 258 L.1 Person Perception Test total factor scores (f procedure) compared with Counselor Regponse System dimensions by product-moment (1) and by multiple correlation (R) . . . 260 M.1 Person Perception Test item validity for both groups using the Counselor Response System dimensions as the criteria: Evaluative factor. . . . . . . . . . . 262 M.2 Person Perception Test item validity for both groups using the Counselor Response System dimensions as the criteria: Emotional factor. . . . . . . . . . . 265 M.3 Person Perception Test item validity for both groups using the Counselor Response System dimensions as the criteria: Potency factor . 267 M.4 Person Perception Test item validity for both groups using the Counselor Response System dimensions as the criteria:. Evaluative- activity factor . . . . . . . . 269 xi LIST OF FIGURES Figure 3.1 Floor plan of filming room. . . 3.2 Per cent of items yielding observed absolute deviations between test-retest for Group II expressed in scale units . xii Page 60 78 LIST OF APPENDICES Appendix Page A. Person Perception Test . . . . . . . 193 B. Person Perception Test: Raw scores. . . . 201 C. Person Perception Test: Distribution of reSponses by item . . . . . . . . 218 D. Person Perception Test: Test-retest reliability estimates . . . . . . . 229 E. Counselor Response System . . . . . . 231 F. Counselor Response System: Rating sheet . . 244 G. Counselor Response System: Frequencies by subjects 0 O O O O O O O O O 246 H. Counselor Response System: Dimension frequencies. . . . . . . . . . 249 1. Person Perception Test: Item/total factor score correlations ("l-7" procedure) . . . 251 J. Person Perception Test-Counselor ReSponse System: Comparisons ("l-7" procedure) . . . 255 K. Person Perception Test-Counselor ReSponse System: Comparisons (£2356 procedure) . . . 257 L. Person Perception Test-Counselor Response System: Comparisons (f4 procedure). . . . 259 M. Person Perception Test-Counselor Response System: Item validity . . . . . . . 261 xiii CHAPTER I INTRODUCTION This study was designed to explore the relationship laetween person perception and behavior in the counseling zictivity. More specifically, it focused upon the :relationship between the generalized "meaning" a counselor- ‘trainee attached to clientépercepts and the interview 'behavior he displayed with a particular client. The possibility of predicting counselorbtrainee‘interview behavior-from that relationship should be explored, since a measure of person perception might be one effective criterion for selecting prospective counselor—trainees. Counselor educators today recognize that present counselor selection procedures are inadequate.1 Selection~ methods currently practiced by graduate schools are based primarily upon the applicant's intelligence and his general ability to do didactic graduate work. In a questionnaire study Santavicca found that 85 per cent of 170 responding colleges and universities used an applicant's undergraduate record and a measure of 1J. D. Linden, S. C. Stone, and B. Shertzer, "The Development and Evaluation of an Inventorytfor Rating Counseling," Personnel and Guidance Journal, 44:267. scholastic aptitude as the main criteria for selecting counselor-trainees.2 Similarly, Hill found that eligibility and potential competence for graduate training was currently the chief counselor-trainee selection criterion.3 He noted that some schools gave a variety of non-academic tests but that there was little evidence that they were used for selection. Patterson, in his survey of the most commonly used counselor-trainee selection methods, found that two- thirds of the schools he studied used undergraduate grade point average as the most general requirement for entrance.4 Yet, even though a certain level of intelligence is needed to be an effective counselor, intelligence, per se, and counselor effectiveness do not appear to be linearly related. As Stoughton wrote, "Good scholarship is important but does not guarantee good counseling ability."5 Because academic ability alone is not a sufficient ingredient for counselor success, some other selection 2G. G. Santavicca, "Supervised Experience and Selection of Counselor Trainees," Personnel and Guidance Journal, 38:195-197. 3G. W. Hill, "The Selection of School Counselors," Personnel and Guidance Journal, 39:357-358. 4C. H. Patterson, "Selection of Rehabilitation Counseling Students," Personnel and Guidance Journal, 41:318-320. 5R. W. Stoughton, "The Preparation of Counselors and Personnel Workers," Review of Educational Research, 27:175. criterion should be used in addition to the applicant's academic ability. Recent literature supported the notion that the best predictor of counselor effectiveness is a measure of counselor-trainee perception of client communication. Davitz wrote that when he was a graduate student, he was encouraged to be "sensitive," to "empathize," to "understand" how the other person felt, to listen with his "third ear," to ”let himself go," to react "spontaneously and intuitively," and finally to undergo psychoanalysis.6 He wrote further that when he became a counseling supervisor, he found himself echoing these same phrases to his students. In his research on the "Ideal Therapeutic Relation- ship," Fiedler referred to the therapist's ability to communicate with and understand the client.7 He generalized that the effective "therapist is able to participate completely in the patient's communication." Rogers reported that clarity of communications and clarity of perceptions are functions of the personal characteristics of speaker and listener: The more that Y (the counselor) experiences the communication of X (the client) as a congruence of experience, awareness, and communication, the more the ensuing relationship will involve: a tendency toward more mutually accurate understanding of the 6J. R. Davitz (ed.), The Communication of Emotional Meaning (New York: McGraw-Hill, 1964), p. 2. 7F. Fiedler, "The Concept of Ideal Therapeutic Relationship," Journal of Consulting Psychology, 14:242. communications; improved psychological adjustment and functions in both parties; and mutual satisfaction in the relationship.8 Stating that the communication of ideas, feelings, and facts is of primary importance in the counseling process, Rank concluded that the perception of client communication 9 is directly related to effectiveness of counseling. In Counseling: Content and Process, Fullmer and Bernard cited a research project by Combs which indicated that good teachers could be differentiated from poor teachers by their perception of youngsters, even though they could not be distinguished by what they knew about teaching. "It (perceptual ability) may be the most helpful criterion we have ever had," concluded Fullmer and Bernard, "for selecting and training candidates for teaching and counseling."10 If the perceptual ability of the counselor is important, then an instrument which measures counselor- trainee perception could be a useful selection tool. 8C. R. Rogers, On Becoming a Person (Boston: Houghton-Mifflin Company, 1961)] pp. 344-345. 9R. C. Rank, "Counseling Competence and Perception," Personnel and Guidance Journal, 45:359-365. 10D. W. Fullmer and H. W. Bernard, Counseling: Content and Process (Chicago: Science Research Associates, 1964), p. 128 PURPOSE The purpose of this investigation was two-fold: (l) to develop an instrument for measuring counselor-trainee perceptions of filmed client-percepts and (2) to study the relationship between counselor-trainees’ perceptions of filmed client-percepts and these counselor-trainees' subse- quent counseling interview behavior with particular clients by: a. creating a set of filmed counseling segments, b. developing an objective means for tabulating counselor-trainee perceptions of filmed client-percepts using a semantic differential format, c. taping samples of counselor-trainee interview behavior, and d. comparing statistically counselor-trainee perceptions of the filmed client-percepts with selected interview behavior. Recently, Whiteley pleaded for shifts in research from the global studies of effective versus ineffective counselors to the studies that focused on "what a counselor dggg and how he is to behave."11 His prime concern was that no behavior distinguishing effective from ineffective counselors seemed ever to be specifically identified in global studies. Though Whiteley's remarks were made well -11John M. Whiteley, "The Selection and Evaluation of School Counselors," a paper presented at the American Person- nel and Guidance Association convention, April, 1968, pp. 1, 8. after this study was begun, they served to underscore the still present need for a behavioral criterion. Also counselor-trainee interview behavior proved a more stringent test for the developed instrument than would a global measure such as counselor effectiveness. GENERAL RELATIONSHIP HYPOTHESIS It was hypothesized that a significant relationship existed between the generalized meaning a counselor-trainee had for filmed client-percepts and subsequent selected counseling interview behavior he displayed with a particular client.1 THEORY In his ambitious review of perceptual theories, Allport has clearly shown that "perception" has many definitions.13 As is the case with many such constructs it falls to the investigator to select one which best meets the needs of his study. Since one of the two directions of person perception research has been the relationship between perception and 12This hypothesis will be expanded and restated in statistically testable form in Chapter III. 13F. H. Allport, Theories of Perception and the Concept of Structure (Neinork: John Wiley 8 Sons, Inc., 1955). action,14 the definition used by those studying person perception was used for this study. Person Perception Defined Tagiuri defined person perception as ”whenever the perceiver regards the object as having the potential of representation and intentionality."15 Specifically, Tagiuri referred to the perceiver's observation 3nd inferences about the object's "intentions, attitudes, emotions, ideas, abilities, purposes, traits," i.e. about things inside the person. It is as the existentialist describes "the one who looks back." So it is in the counseling activity: the per- ceived object is always another person "who looks back." But this definition has not been without its critic. Pastore objected, saying that perception, per se, is different from interpretations of and inferences made from perceptions.16 He claimed that interpretations and inferences change according to the perceiver's need even though the perceived object has not changed a bit. But Murphy and his associates, in reply to Pastore's criticism, 14R. Tagiuri and L. Petrullo (eds.), Person Perception and Interpersonal Behavior (Stanford: Stanford University Press,T1958), pp. xi-Xii. 15Ibid., p. x. 16N. Pastore, "Need as a Determinant of Perception," Journal of Psyghology, 28:457-475. said that such a distinction cannot be made operationally.17 There is no way to separate perceptions, they contend, from interpretations and meanings of the perceptions. And this inability to operationally separate physiological perception from the meaning of the perception appears to be partic- ularly true with person perception where the perceiver is not just looking at the object from the outside (as he would a stone) but is inferring inner states of intentions which cannot be physiologically perceived. The "Meaning" of Perception Berlo wrote: "Your meanings for things consist of the ways that you respond to them, internally, and the prediSpo- "18 sition which you have to respond to them, externally. Tolman's "map room" where stimuli are sorted out and arranged conceptually before a response ever occurs is similar.19 Bruner, Goodnow, and Austin state that these concepts are learned by the association of external stimuli with internal . . . . 20 . . mediating stimuli. So meanings are internal, conceptual, and learned. 17I. Chein, R. Levine, G. Murphy, H. Proshansky, and R. Schafer, "Need as a Determinant of Perception, a Reply to Pastore," Journal of Psychology, 31:129-136. 18D. K. Berlo, The Process of Communication (New York: Holt, Rinehart and Winston, Inc.,—1960), p. 184. 19J. S. Bruner, J. Goodnow, and G. Austin, A Study of Thinking (New York: John Wiley 6 Sons, Inc., 1965), p. vii. 201bid., p. 79. Tagiuri observed that how a person behaves toward another depends upon what that other person means to him. As the First Century A. D. stoic, Epictetus, wrote: Men are not disturbed by things (perceptions), but by the views they take of things (interpretations of perceptions). Thus death is nothing terrible . . the terror congists in our notion of death that it 15 terrible. It is not the perception, per se, which causes men to act but the meaning attached to the perception. Thus in counseling, clients will be perceived differently by different counselors, not because the client- percept has changed from counselor to counselor, but because the interpretation and the meaning given to the client- percept is different from counselor to counselor. The counselor sorts out the various stimuli coming to him and arranges them into a conceptual pattern which has meaning for him. The counselor responds to this arrangement of stimuli into a meaningful concept--not to the objective percept itself. Heider points out that this arrangement of perceptual stimuli into meaningful concepts helps the . . 22 perceiver in several ways: 21Epictetus, The Works of Epictetus, trans. E. Carter and T. W. Higginson (Boston: Little, Brown, 1865), p. 377. 22F. Heider, The Psychology of Interpersonal Relations (New York: John Wiley 8 Sons, Inc., 1958), pp. 151, 157. 10 1. "It gives us control over the environment." Conceptual structure allows the perceiver to predict. For example, if the counselor can have some immediate meaning of his client, he is better able to predict the dynamics of his client. 2. "It helps us evaluate." By categorizing stimuli the perceiver can have a meaningful, predictable way of saying this stimulus is good, that one bad. Thus the counselor is able to assess what he sees. 3. "It serves as a motivation for further action." It serves as a starting place for making hypotheses about the client. 4. "It gives us the ability to report." Since the perceiver cannot attend to all elements of the stimuli bombarding him, he selectively attends to those elements which have the most meaning and lets the rest go relatively unnoticed, giving him a shorthand for describing what he sees. Concept Formation and Attribute Seeking Arrangement of stimuli into meaningful concepts is an ongoing process. Perhaps, as Bruner, Goodnow, and Austin state, the perceiver learns to classify pe0p1e as "honest," "somewhat shifty," and "downright crooked" early under the tutilage of parents and peers.23 He learns quickly to find 23Brunner, Goodnow, Austin, pp. 69, 209. 11 selected attributes he can use as a basis for predicting a broader range of behaviors of the observed person. Moreover, he is constantly classifying people by their "type," into those who are "his kind" and "not his kind," into the class of those who are "reliable" and those who are not. Attribute seeking is so important that Heider wrote, "My entire relationship with the other person may hinge on attribution. Those pleasures attributed to him as a person reveal the kind of person I believe him to be and I am 24 But what a person accordingly drawn to him or repelled." sees, Heider pointed out, is dependent upon what he is tuned to see. And as Berlo said, "We are tuned to see what we believe; our beliefs determine what we see."25 Bruner and associates described this as a "thematic process" or as an ”effort after empirical verisimilitude."26 That is, the perceiver tends to interpret his perceptions congruently with his own attitudes and values in spite of the fact that he may be incorrect logically. And once an interpretation has been established and the person been identified by his attributes, the perceiver will tend to "rectify" or "normalize" the attributes which deviate from expectancy to fit his (the perceiver's) needs and wishes. Z4Heider, pp. 151, 157. ZSBerlo, pp. 230-231. 26Bruner, Goodnow, Austin, p. 104. 12 Soon a core of generally consistent categories is developed describing all people, a consistent core which Bruner and Tagiuri call the perceiver's "expressive style," which Heider labels "perceptual style,” and which Cronbach 27 And this calls the "generalized meaning for the other." learned generalized meaning for the other will determine how the perceiver will behave toward the other. So it is with the counselor as perceiver. "We act and choose on the basis of what we see, feel, and believe; mean- ings and values are part and parcel of our actions. When we are mistaken about things, we act in terms of our erroneous notions, not in terms of things as they are. To understand human action, it is therefore essential to understand the conscious mode in which things appear to us."28 But even though it is generally known that the perceiver Spontaneously categorizes peOple, Tagiuri and Petrullo noted that the relationship between person 29 perception and behavior is in great need of study. A goal 27J. S. Bruner and R. Tagiuri, "The Perception of People,” in G. Lindzey (ed.), Handbook of Social Psychologx, Vol. 11 (Cambridge: Addison-Wesley, 1954), p. 650; Heidér, p. 56; L. J. Cronbach, "Proposals Leading to Analytic Treatment of Social Perception Scores," in R. Tagiuri and L. Petrullo, Person Perception and Interpersonal Behavior (Stanford: StanfOrd University Press, 1958), p. 363. 288. E. Asch, Social Psycholo (Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 19%;), p. 65. 29Tagiuri and Petrullo, pp. xiv-xv. 13 of this thesis was to develop an instrument to study the relationship in the context of the counseling activity. THE SEMANTIC DIFFERENTIAL30 A major concern of this study was the choice of an objective technique for measuring person perceptions. Because of Osgood's background in social psychology and his attempts to tie the measurement of meaning with learning theory, the semantic differential technique is particularly applicable for any study focusing on person perception and meaning. The Semantic Differential Technique The semantic differential technique is Osgood's attempt to quantitatively and objectively measure meaning. The technique consists of‘a concept (in this study the name of the filmed client-percept) and a set of bipolar adjectival scales. Selection by the subject (in this study the counselor-trainees) among successive pairs of bipolar adjectives gradually isolates the "meaning" the concept has for the subject. Osgood visualized the process analogous to the game of "Twenty Questions': Peggy (concept)--Is she Open or closed? Is she strong or weak? Is she fast or slow? 30C. E. Osgood, G. J. Suci, and P. H. Tannenbaum, The Measurement of Meanigg (Urbana: University of Illinois Press, 1957). 14 To increase the sensitivity of the instrument, a seven-point scale is inserted between each pair of adjectives so that the subject can indicate both the direction and the intensity of each judgment as in the example below: Peggy open : : : : : : closed strong : weak fast slow Analysis of Semantic Differential Data To analyse the data generated by the semantic differ- ential technique, Osgood postulated a Euclidian semantic space defined by a set of orthogonal dimensions (identified by factor analysis of the bipolar adjectival scales). The meaning of a concept is that point in the semantic space specified by a series of differentiating judgments. Thus, in the example above, the meaning of the concept "Peggy" is defined by the subject's successive choices from among a set of given scaled semantic alternatives--essentially a combina- tion of controlled association and scaling procedures. And since no two people have identical experiences, and since meanings are derived from experiences, it follows in strict logic that no two counselors will perceive precisely the same meaning for any given filmed client-percept (concept). 15 Osgood's Meaning of "Meaning” Osgood's meaning of "meaning" is based upon a representational mediation process, a linguistic and communication approach to learning theory. According to the mediation process, certain stimulus patterns have a ”wired- in" connection with certain behavior patterns, and other stimulus patterns have acquired a connection with certain behavior patterns. In classical conditioning terminology the behaviors elicited through the "wired-in” connections are the unconditioned reflexes while the behaviors elicited through the acquired connections are the conditioned reflexes. To this point the representational mediation process and classical conditioning are alike. But there are some important differences between the two. The classical conditioning model is a three-step process: (1) a proximal stimulus elicits reflexive responses in the individual; (2) distal stimuli are paired with the proximal stimulus; and, finally, (3) the individual begins to respond to the distal stimulus in ways that he responded to the proximal stimulus. This three-step process (called single stage conditioning) assumes that the individual responds to the distal stimulus identically as he responded to the proximal stimulus. Tolman likens this approach to a "telephone switchboard connecting stimuli and responses." The representational mediation process does not make that assumption. Rather than responding totally to the distal stimuli in the same manner as he responded to the 16 proximal stimulus, Osgood stated that the individual detaches and internalizes some of his original responses to the proximal stimulus. This internalization of responses is what separates the representational mediation process from the classical conditioning model. The process is more like Tolman's "map room" referred to earlier ”where stimuli were sorted out and arranged before ever response occurred." The proximal stimulus, which Osgood called a significate (s), is defined as any stimulus which, in a given situation regularly and reliably produces a predict- able pattern of behavior (RT). Diagrammatically the model is: s ; RT where the arrow (—————9 ) is the "wired-in" connection between stimulus and response. The distal stimulus, which Osgood called a sign (5), is a stimulus which is not the significate but which elicits in the person: (1) part of the total behavior elicited by the significate and (2) responses which would not occur without the previous contiquity of non-significate and significate patterns of stimulation (rm). Diagrammatically the model is: s 3 RT — ’I” i '—> r‘x’n where rm is an internalization of some of the behavior (RT) elicited by the significate, é. These internal responses become relatively fixed over time and serve as a self- stimulus, sm, to the individual to make some sort of overt 17 response, RX' The complete representational mediation process model is: s ‘ R (“’a", T E ———-) r ----9sm -———) RX where ————+ rm———-esm«————§ is an acquired connection. The process is representational because the internal response, rm, is part of the same behavior, R produced '1‘, by the significate, é. It is mediational because the self- stimulation, 5 produced by making this short-circuited m, reaction can now be associated with a variety of acts, Rx, which "take account of" the significate. To the degree that the sign is similar to the significate, the internal response will be similar to the response made to the significate. Theory and the Semantic Differential To this point Osgood has presented two definitions of "meaning." In learning theory the meaning of a sign for a particular person is defined as the representational mediation process which it elicits. For the semantic differential the meaning of a sign is defined as that point in the semantic space specified by a series of differen- tiating judgments. One is a theoretical definition, the other an operational definition. The task now is to tie theory with the measurement technique itself. The bipolar adjectival scales used on the semantic differential have two properties: (1) direction and 18 (2) distance from the origin or midpoint. Likewise, the representational mediation process has two properties: (1) Quality of reaction (e.g. is the individual reacting positively or negatively) and (2) intensity of reaction. The direction of a point in the semantic space corresponds to the quality of the reaction elicited by the sign and the distance corresponds to the intensity of the reaction. Corresponding to each dimension in the semantic space is a pair of reciprocally antagonistic mediating reactions which Osgood has labeled r and Th. Each judgment m by an individual on the semantic differential corresponds to the acquired capacity of a sign to elicit either rm or rm, and the extremeness of judgments correSponds to intensity. The direction and distance of the judgment is assumed to be proportional to the quality and intensity with which the sign elicits rm or Th. Therefore, any concepts being judged on the semantic differential corresponds to a sign eliciting a distinctive set of rm's or Th's with differing intensities. Stated in another way, the filmed client-percepts being judged on the semantic differential for this study correspond to person perceptions eliciting a distinctive set of positive or negative reactions with differing intensities. It is these reactions, acquired via the previously mentioned film test, which this study prOposes to compare with counselor-trainee interview behaviors. 19 DEFINITIONS OF RELEVANT TERMS Throughout the course of this study four terms were used with specific meanings: Client-percept: The client viewed on the filmed counseling interview 229 the name of that client as it appeared at the top of the semantic differential page; usually referred to by Osgood as a "concept." Scale: Paired bipolar adjectives separated by a seven step interval. Item: A client-percept/scale pairing. Meaning: A judgment, or a combination of judgments, made by a counselor-trainee in response to an item. Generalized Meanigg: Meanings summed across client- percepts. OVERVIEW Chapter II, the review of literature, includes (1) a review of person perception research, (2) an examina- tion of several studies which tested the adequacy of the semantic differential, and (3) a detailed inspection of those studies from which this study directly springs. Described in Chapter III will be the development of the filmed client-percept protocols and the selection of the bipolar adjectival scales, as well as the statistical methods and procedures used in carrying out this study. The results of the study will be analysed and discussed in Chapter IV and in Chapter V. Chapter II will begin with a brief historical survey‘ of person perception research to give perspective to the entire study. CHAPTER II REVIEW OF RELATED LITERATURE This chapter begins with an historical review of per- son perception research to support the use of the semantic differential technique as the measurement device for this study. Next, a review of the semantic differential, with special emphasis on its measurement and stability proper- ties, will be discussed. Finally, the chapter will conclude with a review and critique of the study from which this study springs. PERSON PERCEPTION Person perception researchers focused almost entirely upon the face as a source of data and, with Allport, considered the face the most expressive region of the body. The face was where "most people locate the 'self‘" and where l "we give chief attention when we are observing others." Piderit developed a detailed set of line drawings of the 1G. W. Allport, Pattern and Growth in Personality (New York: Holt, Rinehart 8 Winston, 1961), p. 41. 20 21 human head that served as the ground work for much of the empirical work on facial expressions. More recently, but prior to World War II, Brunswik and Reiter used line-drawings of the head which, by systemat- ically moving the eyes, nose, and mouth, produced 189 drawings.3 Nine horizontal rows of these line-drawings were rank ordered by ten subjects according to the following seven qualities: (1) mood, (2) age, (3) beauty, (4) willens- starke (roughly translated as energy plus determination). Several facial "expressions" were consistently linked with certain qualities. The high mouth, for example, was associated with gaiety, youth, unintelligence, and lack of energy; the low mouth with sadness, age, bad character, intelligence, and energy. High and middle foreheads gave favorable impressions; low foreheads, unfavorable impressions. Samuels, replicating the Brunswik-Reiter study and using 247 American college students, found that an average of 88 per cent agreed with the Brunswik-Reiter results.4 But when photographs were selected to correspond to the 2Randall Harrison, "Pictic Analysis: Toward a Vocab- ulary and Syntax for the Pictorial Code, with Research on Facial Communication" (unpublished Doctoral thesis, Michigan State University, East Lansing, 1964), p. 44. 3E. Brunswik and Lotte Reiter, "Eindruckscharaktere- schematisierter Gesichter," 2. f. Psychologia, 142:62-134, in Harrison, Ibid., pp. 52-53. 4Myra R. Samuels, "Judgment of Faces," Character and Personality, 8:18-27. 22 line-drawn faces, Samuels found the percentage of "Correct" responses dropped to 63 with the additional noise of a complete photo. Dremensh, Winkler, Seiller, Kuhnel, and Harrison each continued research using line-drawn faces in the Brunswik-Reiter tradition.5 But line-drawings were not the only stimulus material used. For example, Charles Darwin, early interested in the recognition of emotion, typically showed photographs depicting various emotions to several subjects "without a word of explanation" and asked for judgment of the emotion expressed.6 Later investigators followed Darwin's procedure, sometimes adding a list of emotional words for the subject to choose from, other times leaving the subject free to choose his own words. Probably the first systematic series of phOtographed expressions were Rudolph's poses of a male actor, but the photographs themselves were touched-up and "idealized" making the poses appear artificia1:7"The-Feleky, Ruckmick, Frois-Wittman, and Lightfoot series of poses followed.8 5Harrison, pp. 53-54. 6Charles Darwin, The Egpression of the Emotions in Man and Animals (London, 1872), p. 14. 7R. S. Woodworth and H. Schlosberg, Experimental Psychology (New York: Henry Holt, 1954), pp. 116-123. 81bid. 23 "Unposed" photographs were used by Sherman who tried for spontaneous emotions by taking motion pictures of babies.9 Munn used "unposed" pictures taken from Lifg and ngk_magazines and found that showing the complete photo- graph elicited more sensitive judgments than did showing just the face.10 The additional information served as an aid rather than mere noise.’ Early person perception investigators studied primarily a subject's accuracy of perception, accuracy being defined by the similarity between how the subject described a facial-percept's emotions and a predetermined "correct" answer. Usually, little correlation was found and it soon became apparent that demanding pin-point accuracy of a subject was unrealistic. If, however, the subjects were given credit for "near misses,".the results were much more promising. Researchers then shifted to a more global approach in person perception research and turned to factor analysis as the basic statistical model.11 The first to use a global approach was Woodworth who, using the Feleky series, deve10ped a six-stage unidimensional 12 scale that he found quite reliable. His scale ran as 91bid. loIbid. 11C. E. 05good, "Dimensionality of the Semantic Space for Communication Via Facial Expressions" (Urbana: Univer- sity of Illinois, N. D.), pp. 1-6. 12Woodworth and Schlosberg, p. 118. 24 follows: (I) love-happiness-mirth, (2) surprise, (3) fear- suffering, (4) anger-determination, (5) disgust, and (6) contempt. But it was Schlosberg who was given credit for transforming the semantics of emotion from discrete labeling to continuous, multidimensional scaling.13 He discovered that sortings of the Frois-Wittman series:into the Woodworth six-dimensions produced a circular rather than a linear scale. Schlosberg started with a two-dimensional system and later, switching from the Frois-Wittman series to the Lightfoot series, showed that three dimensions could be used reliably: (1) pleasant-unpleasant, (2) tension-sleep, 14 Schlosberg used all 48 and (3) attention-rejection. Lightfoots and developed norms using 96 undergraduate subjects. Following Schlosberg's lead, several cross- cultural studies have been done which essentially replicate and corroborate Schlosberg's work.15 Osgood summarized the factor analytic research com- pleted by Schlosberg and many others and concluded that a' perceiver judges emotions of others primarily along three dimensions.16 According to Osgood, there seems to be 13Osgood, p. l. 14H. Schlosberg, "Three Dimensions of Emotion, Psychological Review, 61:81-88. 15Woodworth and Schlosberg, p. 132. 16Osgood, p. 39. 25 complete agreement that a "pleasant" dimension exists and almost complete agreement that an "activity" dimension exists. But although a third dimension is apparent, there is little agreement as to what it is. 'Schlosberg himself ”17 And was not happy with his term ”attention-rejection. Osgood concluded that even today naming the third dimension is still a problem. When Osgood counted "near misses" in hista e study, he also found reliable judgments along three dimensions: (1) intensity, (2) pleasantness, and.(3)..control.18 'And when later studies were conducted using the semantic differential technique, Osgood again found reliable judgments along three dimensions: (1) evaluative, (2) activity, and (3) potency.19 On theoretical grounds as presented in Chapter I, the semantic differential appeared to be an appropriate technique for measuring the generalized meaning a counselOr- trainee has for filmed client percepts. «It.is objective;* it employs a factor analytic model for arriving at the dimensions of percepts; and its versatility asa measuring instrument lends itself easily to person perception research even though the technique is not exclusively restricted to person -pe rcepts . 171bid., p. 37. 181bid., p. 12. 19C. E. Osgood, G. J. Suci, and P. H. Tannenbaum, The Measurement of Meaning (Urbana: University of Illinois Press, 1957), pp. 36- 8. 26 The versatility of the semantic differential can be attested to by the variety of subjects studied, by the variety of concepts and their presentation, and by the many uses to which the semantic differential has been put. Subjects have typically been college undergraduates, 20 But especially students in beginning psychology classes. elementary students (down to grade three), high school students, and college graduate students have also been studied.21 Similarly, nurses and physicians have been studied in the hospital setting as have various kinds of men- tal patients such as sexual psychopaths, schizophrenics, neu- rotics, functional psychotics, acute organic psychotics, and 20R. E. Brown, "A Use of the Semantic Differential to Study the Feminine Image of Girls Who Participate in Compe- titive Sports and Certain Other School-Related Activities," Dissertation Abstracts 26:4426; M. R. Goldfried, "The Connotative Meaning of Some Animal Symbols for College Stu- dents," Journal of Projective Technigues and Personality Assessment, 27:60-67;*M.’W2 Otten and R. L. van de CastIe, "A comparison of Set 'A' of the Holtzman Inkblots with the Ror- schach by Means of the Semantic Differential," Journal of Projective Techniques and Personality Assessment, 27:452-460. 21M. R. Goldfried and S. Kissel,."Age aS'a Variable in the Connotative Perceptions of Some Animal Symbols," Journal of Projective Techniques and Personality Assessment, 27:171- 180; A. Jack Hafner and Ephraim Rosen, "The.Meaning of Rorschach Inkblots, Responses and Determinants as Perceived by Children," Journal of Projective Techniques and Person- ality Assessment, 28:192-200; R1TS. LiIly, "A Deve10pmental Study of the Semantic Differential," Dissertation Abstracts, 26:4063-4064; 0. A. Rosenthal, "A Semantic Differential Investigation of Critical Factors Related to Achievement and Underachievement of High School Students," Dissertation Abstracts, 26:3156; W. A. Lewis and W. Wigel, "Interpersonal Understanding and Assumed Similarity," Personnel and Guidance Journal, 43:155-158. 27 character disorders.22 Moreover, hard of hearing subjects,23 24 25 have been industrial workers, and various ethnic groups studied. It seemed reasonable to conclude that counselor- trainees could be studied with the technique. Ordinarily, the semantic differential concept is a word placed at the top of the page followed by a set of bipolar adjectival scales. But concepts have not been limited to words. Inkblots,26 for example, have been used 22R. D. Singer, "A Note on the Use of the Semantic Differential as a Predictive Device in Milieu Therapy," Journal of Clinical Psychology, 17: 376- 378; George Stricker, "Stimulus Properties ofithe Rorschach to a Sample of Pedophiles," Journal of Prgjective Techniques and Person- alit , 28:241-244; Melvin ax, Robert H. Heiselle, and Athan arras, "Stimulus Characteristics of Rorschach-Inkblots as Perceived by a Schizophrenic Sample," Journal of Projective Techniques and Personality Assessment, 24: 439- 443. 23E. J. Hardich, "The Self— -Concept of Hard- of- -hearing Adults as Measured by the Semantic Differential, " Disser- tation Abstracts, 25: 6826. 24H. C. Triandis, ”Differential Perception of Certain Jobs and PeOple by Managers, Clerks, and Workers in Industry," Journal of Applied Psychology, 43:221-225. 25E. R. Oetting, "Cross- Cultural Communication and the Semantic Differential, " Journal of CounselingPsychology, 11:292-293. 26A. I. Rabin, "A Contribution to the 'Meaning' of Rorschach's Inkblots Via the Semantic Differential," Journal of Consulting Psychology, 23: 368- 372; R. H. Loiselle and einschmidt, "A Comparison of the Stimulus Value of Rorschach Inkblots and Their Percepts," Journal of Projective Technigues and Personality Assessment, 27: 191- 194; A. Jack Hafner and Ephraim Rosen, "The Meaning of Rorschach Inkblots, Responses and Determinants as Perceived by Children," Journal of Projective Techniques and Personality Assessment, 28:I92- 200; E. L. Witt, "The Connotative Meaning of selected 'Forced' Rorschach Percepts as Measured by the Semantic Differential Technique," Dissertation Abstracts, 26:4083- 4084. 28 28 29 as have pictures,27 colors, and sonar signals. Glatter and Reece used actual art objects placed in a black, square, 30 Levy used curtained box and "tactually examined." editorials from American, Barron's, and Ebgny_magazines and McNelly used mass media messages.31 Van de Geer, Levelt, and Plomp used musical tones; Hoar and Meed used subliminal messages presented during a twenty minute film; Barclay and Thumin used slides of students studying while using either 32 It seemed reasonable to conclude coffee, tea, or No-Doz. that filmed client-percepts could be used as semantic differential concepts. 27C. J. Friedman, C. A. Johnson, and K. Porde, "Subjects' Descriptions of Selected TAT Cards Via the Semantic Differential," Journal of Consulting Psychology, 28:317-325. 28John E. Williams, "Connotations of Color Names Among Negroes and Caucasians," Perceptual and Motor Skills, 18:721-731. 29L. N. Solomon, "Semantic Reactions to System- atically Varied Sounds," Journal of the Acoustical Society of America, 31:986-990. 30A. N. Glatter and M. M. Reece, "Tactility and Sexual Symbolism," Perceptual and Motor Skills, 14:302. 31Sheldon G. Levy, "Multidimensional Content Analysis: An Extension of the Semantic Differential," Dissertation Abstracts, 25:1321. 32J. P. van de Geer, J. M. Levelt, and R. Plomp, "The Connotation of Musical Consonance," Acta Psycholo ical’ ‘ Journal, 20:308-319; J. R. Hoar and'E. E. Meek, "fifie Semantic Differential as a Measure of Subliminal Message Effects," Journal of Psychology, 60:165-169; A. Barclay and F. J. Thumin,’"M6dified Semantic Differential Approach to Attitudinal Assessment," Journal of Clinical Psycholggy, 19:376-378. 29 The technique has been used to quantify such projec- tive techniques as the Rorschach33 and the Thematic Appercep- tion Test.34 Clayson used the semantic differential to evaluate an experimental program of treatment of delinquent boys.35 Stromberg developed a "Semantic differential of social behavior" to predict how a person perceives, assigns meaning to, and behaves in the immediate environment from knowledge of previous experiences.36 Hypothesizing that the semantic differential would predict group participation on the basis of ratings on the.concept."the.way.l.am," Singer~ concluded that the.semantic differential might be quite 37 There- useful as a predictive device in milieu therapy. fore, the semantic differential technique should be useful in studying the relationship between a counselor-trainee's generalized meaning for client-percepts and his counseling interview behavior. The study which most closely resembled the present one was conducted by Greenberg and Bowes who attempted to isolate 33George Stricker, "Stimulus Properties of the Rorschach to a Sample of Pedophiles," Journal of Projective Techniques and Personality Assessment, 28:241-244. 34Friedman, Johnson, and Forde, pp. 317-325. 35Merrill D. Clayson, "Therapeutic Progress in Terms of Semantic Variability," Dissertation Abstracts, 25:623. 36C. E. Stromberg, "Semantic Differentiation of Social Behavior of Valued Persons by Female College Groups," Dissertation Abstracts, 23:3787. 37Singer, pp. 376-378. 30 independent factors of empathic judgment by showing eleven counseling vignettes to 31 experienced clinicians and 58 graduate students (half "high" in their ability to assess client problems and half "low"). The subjects in this study judged the client-percepts in the filmed vignettes along a 26 scale semantic differential. Factor analysis revealed four factors: (1) dependency, (2) avoidance, (3) anger, and (4) apprehension.38 Although Greenberg and Bowes did~not test their data against counselor behavior with an actual client, they concluded: (1) that expert counselors and counselor-trainees did judge clients along the same dimensions, (2) that there is a common frame of reference counselors use with clients, and (3) that the dimensions used to define that frame of reference is finite in number. The semantic differential technique is just now beginning to find a place in assessing various aspects of the counseling activity. The next section summarizes studies that have eval- uated the semantic differential as a measuring instrument. 38Bradley S. Greenberg and John Bowes, "Dimensions of Empathic Judgment of Clients by Counselors," in Norman Kagan and-David R. Krathwohl, Studies in Human Interaction (East Lansing: Educational Publication Services, College of’ Education, Michigan State University, 1967), pp. 213-262. 31 THE SEMANTIC DIFFERENTIAL Osgood made three assumptions with regard to the semantic differential: (1) intervals within scales are equal, (2) intervals between scales are equal, and (3) scale origin is at the centroid.39 To test these scaling assumptions,.Messick used the' method of successive intervals applied to nine scales over twenty concepts judged by 100 college students and concluded that interval sizes were fairly consistent between scales.4O And although the origin was not exactly at the centroid for any scale, its placement was consistent between scales, that is, slightly negative. Osgood has consistently found the origin placed slightly negative, but Mehling, using nine- point instead of the usual seven-point scale, concluded his results gave added weight to the assumption that "the middle (number) interval in the scales represents the neutral point 41 I! But the assumption of equal intervals within scales was not satisfied by the analysis. For example, Messick" found that interval "5" was less than half the size of 39Osgood, Suci, and Tannenbaum, pp. 146-153. 408. J. Messick, "Metric PrOperties of the Semantic Differential,” Educational and Psychological Measurement, 17:200-206. 41Reuben Mehling, "A Simple Test for Measuring Intensity of Attitudes," Public Opinion Quarterly, 23:567- 568. 32' interval "2."42 One wonders with Osgood, then, how far wrong an investigator would be by assuming equal intervals within scales, even though empirical evidence failed to support the 43 Messick concluded, after estimating distortion assumption. by correlating assumed boundary positions with scale boundary positions, that little distortion would be introduced by accepting the assumption of equal intervals within scales and by using successive integers as interval midpoints.44 Osgood wrote: Considering this andtne other indications of the present study, i.e., an approximate equality.of.. intervals between scales and a similar placement of“ origins across scales, it seems reasonable to conclude that the scaling properties assumed with the semantic differential have some basis other than mere assump- tion- ' ' Still another assumption is scale linearity. Osgood assumed that bipolar adjectival scales are represented by a straight line passing through the semantic space such that‘ scale poles are in opposite directions and equidistant from" the origin. Osgood reported a study by Taylor and Kumata that gave evidence supporting the linearity of at least some bipolar scales.46 42Messick, pp. 200-206. 43Osgood, Suci, and Tannenbaum, p. 152. 44Messick, pp. 200-206.‘ 45Osgood, Suci, and Tannenbaum, p. 152. 46Ibid., p. 153. 33 Taylor and Kumata had four concepts judged against ten scales in two ways: (1) bipolarly in the usual graphic differential manner and (2) unipolarly by using three-point single member scales. Of the fortyjudgments made, in only four were significant differences found between the algebraic summation of unipolar judgments and their corresponding' bipolar judgments, lending support for the linearity assump- tion. Further evidence supporting the linearity assumption was supplied by Malmstrom and“French.47" Starting with the' assumption that if adjectivesare truly opposite, concepts of exactly Opposite meaning should show mirror image profiles when rated by the same scales, Malmstrom and'French concluded that concepts of opposite meanings do yield mirror image results for those judgments for which relevant information is available and that scale symmetry is possible with pairs of bipolar adjectives. Stability of Factorial Structures Factor structure, according to Osgood,‘can.be» appraised in three ways: (1) the number.of factors needed to account for judgments, (2) the relative.weights given to the same set of factors, and (3) the nature of the factors used.48 Osgood reported that cross-cultural work by Kumata 47Edward J. Malmstrom and Gilbert M. French, "Scale- Symmetry and the Semantic Differential," American Journal of Psychology, 76:446-451. 48Osgood, Suci, and Tannenbaum, p. 222. 34 and Schramm indicated Americans, Japanese, and Koreans used the same major factors in about.the.same.relative.weights.49 And Elliott and Tannenbaum found that the same meaningful dimensions underly a large portion of humanbehavior.so Osgood concluded that, regardless of other differences between groups of subjects, factor structures:are pretty stable.51 Suci's high versus low ethnocentricistudy'investi- gated factor structures of tw0'quite.different'typeS'of” people, hypothesizing that high ethnocentrics would show more variance on the evaluative factor than on other dimensions.52 However, Suci found no such difference. 'Bop ‘compared the factorial structures of normal and schizophrenic subjects and found no differences either in relative weightS“of factors or in the nature of‘the'factors, and therefore, concluded the semantic structure for schizophrenic's does not differfrom normals.53 49Ibid., p. 175. 50L. L. Elliott and P. H. Tannenbaum,."FaCtor- structure of Semantic Differential ReSponses to Visual Forms and Prediction of Factor-Scores from Structural Characteristics of the Stimulus-Shapes," American Journal of Psych010gy, 76:589-597. . 51Osgood, Suci, and Tannenbaum, pp. 222-226. 521bid. 53Joan Bopp, "A Quantitative Semantic Analysis of Word Association in Schizophrenia" (unpublished Doctoral thesis, University of Illinois, Urbana, 1955). 35 This general support for factor stability has led many investigators to take for'granted the evaluative, potency, and activity dimensions in meaning without conduct- ing an independent factor analysis todetermine the specific factors for the concepts under study. This might be a dubious procedure, for at least one author has cast doubt on the generality of Osgood's three-factor interpretation of 54 Even Osgood has said his three dominant semantic space. factors do not exhaust the semantic Space.SS As a further example, Greenberg and Bowes found more than three dimensions and, more importantly, found that counseling experiences were 56"It seems related to the number of dimensions found. desirable for each investigator to conduct his own factor analysis, especially if the concepts are unique. Scale-Checking Style Early in his research Osgood observed that better educated subjects checked scale positions 2,3,5,6 more frequently than positions 1,4,7, and Kerrick found that subjects of low I.Q. checked positions 1 and 7 most 54R. E. Lana and F. J. Pauling, "Opinion Change When the Semantic Differential is a Pretest," Psychological Reports, 17:730. ' SSOsgood, Suci, and Tannenbaum, p. 72. 56Greenberg and Bowes, p. 246. 36 frequently.S7 Grade school children were found to perform better with a five-step scale than a seven-step scale, indicating that age was a factor in scale-checking style.$8 Subject's emotionality toward the concepts 8150‘ affected which scale positions were checked. Members of the American Legion used scale positions'1,4,7 more than positions 2,3,5,6.59 In a study of belief systems, Wozniak found that subjects with closed belief systems-checked scale positions 1 and 7 more than subjects with open belief' systems, and subjects with open belief systems checked scale position 4 more often than subjects with closed’belief' systems.60 On scale positions 2,3,5,6 no differences in checking style were found between the two groups. Wohl conducted a study testing scale-checking styles with personality "constriction" and concluded that "constric- ted" subjects responded on a semantic differential toward the neutral 4 position, avoiding the extreme I and 7 positions.61 S7Jean Kerrick, "The Effects of Intelligence and Manifest Anxiety on Attitude Change Through Communication" (unpublished Doctoral thesis, University of Illinois, Urbana, 1954). 58Osgood, Suci, and Tannenbaum, p. 227. 59Ibid. 60Daniel F. Wozniak, "A Factor Analytic Study of‘ Semantic Structures of Closed, Open, and Medium.Be1ief- Disbelief Systems" (unpublished Doctoral thesis, Michigan State University, East Lansing, 1964), p. 71." 61Osgood, Suci, and Tannenbaum, p. 227. 37 And finally, when Bopp compared the checking style of schizophrenics and normals, she found that normal subjects checked scale positions 2,3,5,6 significantly more frequently than did schizophrenics.62 She concluded that‘schizophrenics were less discriminatory in theirruse of semantic scales. In an attempt to compare Osgood's mediation theory with a verbal learning approach to meaning, Cotter concluded, "Semantic differential scores, the deviations from the neutral point on a seven-point scale, are measuring nothing "63..(It311c57mine.)‘“One'method more than response style. employed to overcome Cotter's Conclusion has been research in latency of subject's differential judgments.' In the Lyons and Solomon latency studies, subjects, acting as quickly as possible, made differential judgments by operating a lever to the right Orleft.64 During the allotted five-second interval for item judgment a split- second timer recorded'thecitem,cthe'subject;”the"latency"of‘ judgment, and the direction of judgment.. Forty subjects judged 150 items in the manner described in addition to' judging the same items in the usual graphic differential manner . 62Bopp, et passim. 63R. A. Cotter, "Verbal Learning in the Evaluation of Two Theories ofMeaning," Dissertation Abstracts, 26:6835. 64Osgood, Suci, and Tannenbaum, pp. 155-159. 38 Lyons and Solomon then alligned the favorable poles of each bipolar scale and compared mean*laten¢y scores with graphic differential scale judgments. They found that mean latency scores were shortest in the direction of the favor- able pole; that is, latencies were shorter forpositive judg- ments than for negative judgments. However, latencies were shortest at the extremes, increasing as judgments moved-- toward the 4 position, which gave support to Osgood's claim that extremeness of judgment on the semantic differential was a valid measure of habit strength.6S Osgood felt that latency possibly reflected the degree of conflict the subject was experiencing in making judgments. 'That is, "the more nearly equal the reaction tendencies, the slower the judgment and the nearer to the center of the scale the check-mark." But the problem was not that simple. :Tannenbaum‘s attitude change study revealed that amountiof:attitudeichange was least at the extremes (scale positions l.and 7) and increased linearly as the original attitude.became less intense, as was expected.66 'But when the original attitude was least intense (neutral 4 position), the amount of" attitude change unexpectedly decreased; ‘Latency studies have' confirmed these findings. Judgments on scale positions 1 and 7 required the shortest mean latencies, positions 2,3,5,6 required the longest, and the neutral 4 position fell 651bid., pp. 158-159. 661bid., p. 228. 39 somewhere in between. Osgood explained that when a subject had a response conflict he resolved it by avoiding a decision and by relying on a "neither" or "don't know" choice. This led Osgood to suggest three judgmental levels: (1) an all- or-nothing choice (positionsl and 7) being the easiest to make, (2) a "neither" choice (scale position 4) being the next easiest to make, and (3) a discriminatory choice (scale positions 2,3,5,6) being the hardeSt to make. Behavioral Validity of the Semantic Differential Of the three major dimensions OSgood typically used, one, the evaluative dimension, has been used as a measure of attitude. Suci, using evaluative scales of a semantic differential, was able to distinguish between high and low ethnocentrics by their ratings of variousethnic‘concepts.67 Suci as well as Tannenbaum and Kerrick was able to discriminate between shades of political preference by evaluative scale ratings.68 That the evaluative dimension of the semantic differential can be used as a measure of attitude was also supported by Osgood's highly significant correlations between 67G. J. Suci, "A Multidimensional Analysis of Social Attitudes with Special Reference to Ethnocentrism" (unpublished Doctoral thesis, University of Illinois, Urbana, 1955). 68Osgood, Suci, and Tannenbaum, p. 193. 40 semantic differential evaluative scores and scores on both the Thurstone (p<.01) and the Guttman (p<;01) scales.69 In the Manis study five undergraduate "communicators" wrote two short passages on their views-toward college life and rated their passages on a nine-scale semantic differential.70 The rating results were then compared with ratings given to the passages by30 undergraduate "recipients," and Manis concluded that the evaluative scales can be profitably used in assessing attitudes. Walker constructed a laboratory analogue for social attitude learning and used it to assess the attitudinal' validity of an evaluative semantiCWdifferential‘S‘capacity 71 The behavioral validity of an‘ tO‘predict behavior. evaluative semantic differential was partially confirmed. 'In cross-cultural attempts to asseSS'attitude'Diab and Rosen each concluded that attitude can be measured by the evalua- tive factor.72 But Diab wrote that the evaluative factor alone was not enough and Doob agreed that "overt behavior can seldom be 691bid., pp. 193-194. 70M. Manis, "Assessing Communication with the Semantic Differential," American Journal of Psychology, 72:111-113. 71Lawrence Walker, "A Concept Formation Analogue of Attitude Development," Dissertation Abstracts, 22:2482-2483. 72L. N. Diab, "Studies in Social Attitudes:III:Atti- tude Assessment Through the Semantic Differential Technique," Journal of Social Psycholo , 67:303-314; E. Rosen, "A Cross- Cultural Study of Semantic rofiles and Attitude Differ- ences," Journal of Social Psychology, 49:137-144. 41 73 Within the predicted from knowledge of attitude alone." theoretical structure of the semantic differential, attitude was only one of the dimensions of meaning; therefore, atti- tude gave only part of the information needed to make pre- dictions. To improve predictability of overt behavior, Osgood suggested the addition of the other semantic dimensions.74 To support his suggestion, Osgood reported a pilot study by Tannenbaum in which 40"subject5'judged"three nationality concepts against a set*of.scales representing the three major factors--evaluative, activity, and potency.7S The subjects also rated these nationalities on’a modified Bogardus Social Distance Scale.‘ Tannenbaum compared each factor score with each other and with the Bogardus ratings, and, as was expected, the evaluative score correlated the highest with the Bogardus ratings. ‘But multiple correlation analysis revealed a significant increase in predictibility: Concept E-Score/Bogardus E-A-P Scores/Bogardus Germans .22 .78 Chinese .62 .80 Hindus .59 .72 73Diab, Ibid.; L. w. Doob, "The Behavior of Attitudes," Psychological Review, 54:135-156. 74Osgood, Suci, and Tannenbaum, p. 198. 75Ibid., p. 199. 42 In their "tribes of stickmen" study, Solley and Messick investigated the degree to which the profile of an "arbitrary" concept described by the semantic differential could be predicted from the experiences presented in develop- ing the concept's meaning.76 They concluded:' I. The semantic differential provides a very accurate index of the final ratios . . . of guesses by.the subjects; it has high validity in this sense. 2. The semantic differential scores also reflect with considerable accuracy the input characteristics of 'the stimuli making up the arbitrary concept.* In other words, when we experimentally produce a complex "meaning" for a concept . . . the instrument faithfully reflects the learning experiences. Osgood also discussed the limitations of "reversibility of the measurement”0perations"“as'atvalidity criterion. That is, given concept profiles produced by‘a subject, could the investigator discriminantly label the concept originally judged? He found that reversibility was successful when the concepts were both‘few‘infnumber and highly varied in meaning. But'when‘the‘concepts were many or when they were connotatively similar, then reversibility could not be accomplished with any degree of confidence. Furthermore, meaning is in the perceiver and before an investigator could predict a concept from a profile, he would have to know the meaning of the.concept for the subject who produced the profile. Osgood concluded, "Therefore, this is not a necessary validity criterion for 761bid., p. 164. 43 this type of measurement; its application would require that we reproduce the meaning of the concept from the profile, not the concept label."77 A FILM TEST OF COUNSELOR PERCEPTIONS Krathwohl wrote that no study‘startsde‘novo.78 This study was a direct outgrowth of one conducted by Rank at the University of Minnesota. ‘To examine the relationship between counselor perceptions and counselor effectiveness, Rank developed a Film Test of Counselor Perceptions (FTCP) which was administered to enrollees of three consecutive National Defense Education Act Guidance and Counseling'Institutes.79 The filmed portion of the FTCP consisted of ten 1 1/2 to 5 minute excerpts selected from filmed counseling interviews. The graphic portion of the FTCP consisted of 200 items generated by counselor educators from the Univer- sity of Minnesota who viewed the filmed excerpts and after’ each segment listed statements describing their observations of the client, the counselor, and the interaction of client 771bid., p. 166. 78David Krathwohl, How to Prepare-a Research Prgposal (Syracuse, New York: Syracuse University, I966), p. 26. 79R. C. Rank, "The Assessment of Counselor=trainee Perceptions of Interview Protocols Before and After an Intensive Practicum Experience" (unpublished Doctoral thesis, University of Minnesota, Minneapolis, 1964). 44 and counselor. A Likert scale was attached to each item as in the following examples: Barbara (the client) likes ( ) (') ( ) ( ) ( ) peOple. Barbara is anxious. ( ) ( ) (’) ( ) ( ) All students should be ( ) ( ) ( ) ( ) ( ) like Mary. The counselor is genuinely ( ) ( ) ( ) ( ) ( ) interested in Elizabeth. The client is happy with ( ) ( ) ( ) ( ) ( ) herself. Test-retest reliability coefficient of .69 (interval one week) was reported. Scores on the FTCP were correlated with the dependent variable "counselor effectiveness" defined by staff rankings and significant results were found: 1. y .4l--"counselor effectiveness” correlated with' post-test scored with pre-test key (p<.01). 2. r .54--"counselor effectiveness" correlated with - post-test scored with post-test"key (p<.005). 3 y .34--"counselor effectiveness" correlated with pre-test scored with post-test key (p<.025). Rank's study gave evidence that a relationship exists between "counselor-perceptions” and ”counselor effective- ness," and that a film of counseling interviews is an excellent vehicle for presenting stimulus material. Such stimulus material resembled a real counseling activity closely and at the same time allowed the investigator.to repeat the material exactly, a distinct'advantage for any instrument. Using live actors would be closer to reality, but then no two‘performances could be replicated in every detail as could a film. 45 But his was a pilot study and as such his definitions for the dependent and independent variables were"tentative. Improved definitions were sought for both ”counselor effec- tiveness” and "counselor perceptions." "Counselor effectiveness" defined by staff rankings did not satisfy Cronbach's criticism of global‘scores.80 Global scores, claimed Cronbach, compress many dimensions into a single index and thereby, significant relations are overlooked and results are interpreted as general without suff1c1ent‘evidence. To overcome the global index criticism and to attempt a more behavioral approach to the problem, Rank with DeRoo developed the Counselor Response System (CR8),81 a "method' for objectively analyzing the Verbal statements of counselors during counseling interviews."82 This multidimensional response system was composed of six dichotomous dimensions: (1) affective-cognitive, (2) affective-cognitive change, (3) content: follow-shift, (4) control: restrictive- 80L. J. Cronbach, "Proposals Leading to Analytic Treatment of Social Perception Scores,” in Renato Tagiuri and Luigi Petrullo, Person Percepgion and Interpersonal Behavior (Palo Alto: Stanford UniverSity Press, 1958), p. 355. \ 81R. G. Rank and W. M. DeRoo, "Counselor Response System of the Behavior Interaction Description System" (East Lansing: Michigan State University, 1965;’mimeographed). 82W.’M. DeRoo, "A Study'ofiRelationshipS“Between Counselor Personality and Counseling Behavior" (unpublished Doctoral thesis, Michigan State University, East Lansing, 1965). 46 expansive, (5) temporal: present-past/future, and (6) referrent: client-other. DeRoo used the CRS to study the relationship between counselor personality and counseling behavior.83 Counselor personality was defined by the Holtzman Inkblot Technique andbythe Rokeach Dogmatism Scale. Counseling behavior was defined by frequency of responses in each dimension of the CRS.’ DeRoo found a positive relationship (p<.025) between Human Movement scores on the Holthan and Client reference scores on the CRS. Significant, but'not predicted, relationships were found between Barrier and Integration scores on the Holtzman and Client reference“scores on~the CRS. DeRoo concluded that his resultS'"suggest’that the CRS is a potentially useful tool for research“inicounseling."84' In'a similar, but independent, study Riewald also used counselor interview responses ascthefidependent variable in his investigation of the relationship between tolerance of ambiguity and counselor behavior and found'that frequency 85 of certain counselor responseS‘was a"fruitful”criterion3 Riewald hypothesized a positive relationship between 831bid. 84Ibid., pp. 87-88. 85Arthur G. Riewald, "The Relationship of Counselor's Tolerance of Ambiguity to Counselor Behavior in the Counsel- ing Interview” (unpublished Doctoral thesis, Michigan State University, East Lansing, 1964). 47 intolerance of ambiguity, defined by Figure Recognition and Verbal Reasoning tests and: l. movement toward premature closure or resolution of the client's problem, defined by frequency of responses in which the counselor initiated,' interrupted, or changed the client's topic; and by frequency of counselor summary or closure statements; use of cognitive rather than affective statements; tendency toward value laden statements and conclusions. CNN 0 0 Analyzing the data generated by 23 beginning counselors at Wayne State University, Riewald found support for each of his hypotheses except movement toward premature closure when defined by frequency of summary or closure statements. Kagan and Krathwohl over the past few years also have been developing a method for quantifying counselor behavior.86 Their instrument, the Counselor Verbal Response Scale (CVRS), consisted of five forced choice dichotomous dimensions, much like Rank and DeRoo's system. Therefore an adequate solution to Rank's "global criterion" problem has been found by using the frequency of actual counselor interview responses to define the dependent variable. But to the time of this study no one had attempted to improve upon Rank's measurement of the independent variable. 86Norman Kagan and David R. Krathwohl, Studies in Human Interaction (East Lansing: Educational Pfiblication Services, College of Education, Michigan State University, 1967), p. 27. 48 Since the FTCP counseling segments included both the counselor and the client-percept, the viewer, because he was a counselor-trainee, might have concentrated upon the counselor and his technique rather than on the client. Filming only the client-percept should not only eliminate counselor technique but also should help create the illusion of a one-to-one relationship between filmed client-percept and viewer. No one had developed a film that concentrated solely on the client. Moreover, many of the FTCP items were concerned with counselor technique. No one had developed a graphic scale free from counselor technique to be used with a counseling film test. This study sought to improve the definition of Rank's independent variable by deveIOping an objective measuring instrument based upon person perception theory. SUMMARY In this chapter a review of the pertinent literature leading to the present study was presented. The chapter began with an historical review of facial expression studies in which line-drawn or photographed faces, both posed and unposed, were the stimulus material and the similarity between how a subject described a facial- percept's emotion and a predetermined "correct" answer was the criterion. But demanding pin-point accuracy of a subject proved unrealistic and a review of more global approaches 49 to the problem of recognizing emotions in others followed. Starting with Woodworth's six-stage unidimensional scale, Schlosberg transformed the semantics of emotion from discrete labeling to continuous, multidimensional scaling, eventually arriving at three reliable dimensions on which a perceiver judges emotions in others: (1) pleasant-unpleasant, (2) tension-sleep, (3) attention-rejection. 'Using the semantic differential technique, Osgood also found reliable judgments along three dimensions: (1) evaluative, (2) activity, (3) potency. A rationale for using the semantic differential technique in this study was then presented followed by a review of studies that examined the measurement character- istics of the technique. The major measurement assumptions made by Osgood were examined and the following conclusions were reached: (1) interval sizes between scales were fairly consistent, (2) the origin was consistent between scales, slightly negative of the centroid, (3) intervals within scales were not equal but little distortion would be introduced by accepting the assumption, and (4) scale linearity is possible. The factorial structure of the semantic differential was shown to be consistent from study to study and from group to group irrespective of other differences between studies or groups of subjects. But a note of caution was raised about blindly accepting Osgood's three major 50 dimensions without conducting an independent factor analysis, especially if the concepts under study were unique. Scale checking style was also examined and apparently three judgmental levels exist: (1) an all-or-nothing choice (positions 1 and 7) being the easiest to make, (2) a "neither" choice (scale position 4) being the next easiest to make, and (3) a discriminatory choice (scale positions 2,3,5,6) being the hardest to make. Differences in scale checking style were observed for various subjects: the less educated, less intelligent, more emotional, and more psychologically closed subjects checked the less discrimi- natory scale positions (1,4,7) more than did their respective counterparts. A review of the semantic differential as a measure of attitude followed and although the evaluative dimension was generally conceded to measure attitude, it was concluded that predictive validity could be improved by including the other dimensions and by conducting a multiple correlation analysis. The chapter concluded with a review and critique of Rank's Film Test of Counselor Perceptions (FTCP) which gave evidence not only that a relationship between counselor perceptions and counselor effectiveness exists but also that a film of counseling interviews is an excellent vehicle for presenting stimulus material. But Rank's definitions for "counselor perceptions" and counselor effectiveness" were questioned. Although it was pointed 51 out that Rank and DeRoo improved upon the criterion for "counselor effectiveness" by develOping the Counselor Response System (CR8), a more behavioral approach to the problem, to the time of this study no one had attempted to improve upon Rank's definition of "counselor perceptions." In the next chapter the design of this study will be presented. CHAPTER III DESIGN In the preceding chapter the literature concerned with person perception research, the semantic differential technique, and the Film Test of Counselor Perception was reviewed. This chapter will describe the design of this study. The chapter begins with a demographic description of the subjects included in the study. Development of the independent variable, the Person Perception Test, is then discussed followed by the dependent variable, tabulated counselor-trainee interview responses. The chapter concludes with hypotheses to be tested and procedures employed in collecting and analyzing the data. THE SUBJECTS The purpose of this investigation was to study the relationship between counselor-trainees' perceptions of filmed client-percepts and these counselor-trainees' sub- sequent counseling interview behavior with particular~ clients. Two groups of counselor-trainees were used in this study. 52 53 Group I included 30 graduate students attending the 1965-1966 National Defense Education Act Guidance and Counseling Institute at Michigan State University. The group was composed of 21 males and 9 females, and ranged in age from 23 to 38 with a mean age of 28.4. Twenty-three enrollees were married. Although the enrollees came to the institute from 14 states, Michigan contributed the most (13) and Illinois and Wisconsin contributed three each. Four other enrollees came from the Midwest, five from the East, and two from west of the Mississippi River. Only five enrollees had Master's degrees, none of which was in guidance or counseling; and only five enrollees had any counseling experience prior to attending the institute. Three enrollees had one year of counseling experience and one each had five years and seven years, respectively, of part-time counseling experience. Group II was composed of 30 graduate students enrolled in the counseling practicum for Master's degree students at Michigan State University in the summer of 1966. Since the practicum class was a requirement for completion of the Master's degree, none of the class members had a Master's degree in counseling and guidance prior to their enrolling in the practicum. The practicum class had 18 males and 12 females, and ranged in age from 22 to 58 with a mean age of 32.2. All but two of the practicum enrollees came from Michigan. 54 Data from each group were treated separately although the probability of obtaining significant results could have been increased by combining them. But with the great number of correlations computed for this study, the probability of finding significant relationships where none, in fact, existed was high. And if these significant relationships were found by using the groups combined, there would be no way of estimating how many of the relationships were significant merely by chance alone. With Group II as the replicating group, this problem was at least partially controlled. Separating the groups controlled for false positive results and also strengthened the conclusion that a relationship indeed did exist when it was confirmed by both groups. THE PERSON PERCEPTION TEST . The independent variable used in this study was the Person Perception Test (PPT), a specially developed film test based Upon person perception theory as described in Chapter I and upon Osgood's semantic differential technique. The test consisted of five filmed client-percepts and a 30 scale graphic semantic differential. DevelOpment of the PPT required several operations: (1) selection and (2) filming of client-percepts to be used as semantic differential concepts, (3) selection of bipolar adjectives to be used as semantic differential scales, (4) factor analysis of the 55 data generated, (5) development of scoring procedures, and (6) conducting reliability estimates. Selection of Client-Percepts Osgood listed three primary criteria for selecting semantic differential concepts: 1. The investigator should select concepts which are similar to the significate they represent. Osgood's representational mediation process suggested that the more a sign is similar to a significate, the more the mediating process will lead to overt behavior similar to the behavior elicited by the significate. Since the counselor-trainees to be tested were prospective secondary school counselors, filmed high school students were used as client-percepts. 2. The investigator should try to select concepts that vary in meaning one from the other. To obtain concept variability, both male and female percepts were filmed and one percept was Negro. Further concept variability was obtained by varying interview content. 3. The investigator should use "good judgment” when selecting concepts. The "good judgment" criteria for client- percept selection was (1) recommendation from the high school counselor, (2) verbal ability, and (3) a willingness to participate in the project. 1C. E. Osgood, G. J. Suci, and P. H. Tannenbaum, The Measurement of Meaning (Urbana: University of Illinois Press, 1957), pp. 77-78. 56 Using the above criteria, five students were selected from three area high schools. The following is a brief profile of each client-percept and the interview content assigned to each. Client-percept Peggyyand ClientLpercept Bill. Client- percept Peggy was a female, high school senior who, according to her counselor, was a good academic student who took part in many extra-curricular activities and planned to attend college after graduation. Client-percept Bill was a male, high school junior interested in science, particularly in computer programming and statistics. According to his high school counselor, he was a good academic student, partici- pated in some science-related extra-curricular activities, and planned to attend college after graduation. Interview content for both these client-percepts concerned a typical adolescent problem, educational- vocational planning. Harmon and Arnold found that 19 out of 20 counselors they studied dealt specifically with vocational and educational information.2 And 1152 out of 1282 counselors studied by Hitchcock assisted pupils with 3 educational-vocational plans. Schmidt concluded that 2Donald Harmon and Dwight L. Arnold, "High School Counselors Evaluate Their Formal Preparation," Personnel and Guidance Journal, 39:303. 3William Hitchcock, "Counselors Feel They Should," Personnel and Guidance Journal, 32:72-74. 57 secondary school counselors saw their ”ideal function” as including counseling students with vocational and educational problems.4 Client-percept Bob. Client-percept Bob was a male, Negro, high school senior on a cooperative work-study program attending school half days and working at a drive—in restaurant half days. No information was made available regarding his academic achievement. He was asked to focus upon his ambitions and goals and how he thought being a Negro affected them. With the passage of the various Civil Rights Acts, equal rights for minority groups has become a dominant social issue in public schools. The problems of Negro identity conflicts mentioned by Derbyshire, will undoubtedly be a problem brought to the secondary school counselor.5 Client-percept Lynn and Client-percept Terry. Client- percept Lynn was a female, high school junior with above average grades but not interested in attending college. Client-percept Terry was a male high school senior with average grades who participated in football and wrestling. His further educational goals were undecided. 4Lyle D. Schmidt, "Concepts of the Role of Secondary School Counselors," Personnel and Guidance Journal, 40:602. 5Robert L. Derbyshire, "United States Negro Identity Conflict,” Sociology and Social Research, 51:63-77. 58 In an attempt to capitalize on the principle that the perceiver invests ambiguous stimuli with those responses unique to his personality, the interview content for these two client-percepts was their reaction to three minutes of counselor silence. According to Bordin, counselor silence should produce an anxiety reaction not only in the client- percept but also in the counselor-trainee perceivers. With reference to ambiguity, Bordin wrote: "One possible consequence is the more direct expression of feelings toward the counselor. This can be very threatening to the counselor if he is not both personally and professionally secure."6 Filming of Client-Percepts Although films of counseling sessions frequently show both the counselor and the client, only the client was filmed for this study. Filming only the client had several advantages: 1. It eliminated counselor technique as a confounding stimulus, thereby allowing the counselor-trainees to concentrate solely on the client-percepts. 2. It helped create the illusion that the counselor- trainee was the interview counselor. 6Edward S. Bordin, P§ychological Counseligg (New York: Appleton-Century-Crofts, Inc., 1955), pp. 146-150. 59 3. The client-percepts thus filmed would more easily translate to the semantic differential as stimuli-concepts than if both counselor and client were seen. With the cooperation of the Michigan State University audio-visual service, the five client-percepts were filmed in a counseling practicum room. During filming, the camera and cameraman were present but behind a screen (as illustrated by Figure 3.1) to give both the client and the counselor the best possible opportunity to forget they were being filmed. The audio was picked up by a microphone sit- ting openly on a table between the counselor and the client. Because of financial limitations, filming time was restricted to 66 minutes, divided as follows: Client-percept Peggy (Educ.-Voc. Planning) 20 Min. Client-percept Bill (Educ.-Voc. Planning) 20 Min. Client-percept Bob (Negro Goals) 20 Min. Client-percept Lynn (Counselor Silence) 3 Min. Client-percept Terry (Counselor Silence) 3 Min. Total Filming Time 66 Min. So that the PPT could be administered in one fifty- minute class period, it was necessary to pare the film to a more reasonable length. Total film time was, therefore, reduced to twenty-one minutes by cutting each twenty-minute film segment to approximately five minutes. The two three- minute segments were left in tact. With the aid of a moviola the counselor's voice was eliminated and the film was spliced so that client verbal 60 Camera Filming Client Only Z/ // Counselor \& Micrgphone Q/ ‘/,z0ne-way Window Client "L l //{J F_T::i::P-Tape Recorder Timer Cl Monitor”’ Figure 3. 1--Floor plan of filming room. 61 content appeared continuous despite the absence of the counselor's comments. Specifically, the film footage was cut (1) when the counselor and the client were both talking at the same time and the counselor's voice could not be eliminated, (2) when the counselor asked a question (which the viewer ultimately would not be able to hear) to which the client was quite obviously just listening, (3) when the client answered a direct question (again ultimately not heard by the viewer) with a "yes" or "no" response, and (4) when client comments did not verbally make sense with immediately preceding uncut comments or when comments seemed to stray from the interview content assigned. Because the periodic jerking created by splicing would serve as a distracting element for the viewer, a fading technique was used to smooth the client's body movements while still maintaining continuity of verbal content. The dictates of the editing process brought the film time to twenty-one minutes, leaving twenty-nine minutes in the class period, which in retrospect worked out quite well. Fourteen minutes was estimated for camera loading and unloading, test direction reading, and miscellaneous classroom administrative duties. Fifteen minutes was thus allowed for counselor-trainee judgments of the five client-percepts. 62 Selection of Bipolar Adjective Scales Since the time available for making semantic judg- ments was limited to fifteen minutes, it was necessary to know what size differential the subjects could be expected to complete in the allotted time. Osgood's experience was that "even the slowest college student subjects can be expected to make judgments of at least 10 items per minute" (150 items in 15 minutes).7 The number of items on any form of a semantic differential is merely the number of scales times the number of concepts. 80 with five concepts to be judged on this 150 item differential, 30 scales, repeated in typical graphic differential form for each concept, could be easily completed in the available fifteen minutes. When selecting semantic differential scales, Osgood advised the investigator to follow three main criteria: (1) factor composition of the scales, (2) relevancy of the scales to the concepts being judged, and (3) semantic stabiligy of the scales for the concepts and subjects in a particular study.8 Because no similar use of the semantic differential had ever before been made, no adequate guide existed for choosing scales by their factor composition or semantic stability. Therefore relevancy was the most important scale selection criterion for this study. However, since Osgood usually arrived at three dominant 7Osgood, Suci, and Tannenbaum, p. 80. 8Ibid., pp. 78-80. 63 factors in his studies (evaluation, potency, and activity), a deliberate attempt was made to select scales which looked as if they might load on an evaluative, potency, or activity factor. To allow each subject equal use of the three major factors, balance of ten scales per factor was maintained as a fourth criterion for selecting differential scales. Relying on assumed factor composition and relevancy, the scales selected for this study were taken from two sources: (1) Osgood's Thesaurus study9 and (2) Harrison's thesis on pictic analysis.10 Six scales were selected as markers to help identify the factor composition of the remaining seales in the PPT. Harrison's open-closed and honest-dishonest seemed to load on an evaluative factor.and also seemed relevant for judging client-percepts. The activity factor was most purely defined by Osgood's active-passive and fast-slow; and the potency factor was best defined by Osgood's strong-weak and tenacious-yielding. Several scales selected seemed more applicable to the perceiver than to the person perceived. For example, because the focus of any counseling activity is on the client, it seemed more probable that the counselor-trainee 91bid., pp. 47-64. 10Randall Harrison, "Pictic Analysis: Toward a Vocabulary and Syntax for the Pictorial Code, with Research on Facial Communication" (unpublished Doctoral thesis, Michigan State University, East Lansing, 1964), p. 44. 64 would find himself agreeing or disagreeing with the client rather than perceiving the client as agreeing or disagreeing with the counselor. Nine scales focusing on the perceiver were separated slightly from the rest in the graphic presentation of the differential. Scales in the first part focused on the client-percept; scales in the second part focused on the perceiver, thus hopefully incorporating the essential dyadic nature of the counseling activity. The thirty selected scales were then randomly alternated by polarity direction and also randomly ordered with each part. Table 3.1 is a sample page of the PPT showing the name of the client-percept to be judged and listing the scales with a seven-step interval between each bipolar adjective. The complete test with directions modified from Osgood's typical semantic differential directions can be found in Appendix A. Factor Analysis One task in the development of the PPT was to deter- mine which factors emerged consistently from group to group and to discover which scales consistently defined which factors irrespective of group. The purpose of factor analysis of semantic differential data is to make possible the selection of a minimum number of scales which taken together can best define a judgmental dimension. As Osgood explained: 65 TABLE 3.l--Samp1e page of the Person Perception Test. Client #1: Peggy A. Please make your judgments on the basis of what this client has communicated to you about herself. The client is: closed :open transparent :opaque near :far tired :energetic active :passive fast :slow yielding :tenacious defensive :agressive excited :relaxed static :dynamic upset :calm unhappy :happy complex :simple strong :weak embarrassed :smug controlled__”___ :uncontrolled hopeful :fearful feminine :masculine constrained :free honest :dishonest profound :superficial B. Please make your judgments on the basis of how you feel toward the client. I feel: supportive :angry perplexed :understanding disagreeing :agreeing approving :disapproving surprised :bored unhurried :harrassed mild :intense patient :impatient satisfied :dissatisfied After viewing the next film sggment, turn the page and make your judgments of the next client. 66 l. The process of description or judgment can be conceived as the allocation of a concept to an experiential continuum defined by'a pair of 'bipolar adjectives. ‘ 2. Many different continua of judgments are essentially equivalent and hence may be repre- sented by a single dimension (factor). 3. A limited number of such continua, representative‘ ‘of the dimensionality of meaningful judgments, can be used to define a semantic space within whiph 'the meaning of any concept can-be specified. Theoretically, thirty factors could emerge from factor analysis of thirty scales, but in practice this never occurs. Osgood normally isolated three major factors. To determine scale factor composition,‘the PPT was administered to each group of subjects. 'The data were transferred to computer cards and submitted for'factor analysis using the Factor A: Principal Components and Orthogonal Rotations Program.12 This program called for a principal axis solution and a quartimax rotation; with the' Kiel-Wrigley criterion of three for terminating rotation. The Kiel-Wrigley criterion of three discontinueS'rotation when less than three scales have their highest loading on the factor‘under consideration. As can be seen from Tables 3.2 and 3.3 factor analysis of the counselor-trainee judgments computed for Group I isolated four factors accounting for 56 per cent of 11C. E. Osgood, "The Nature and Measurement of Meaning," Psychological Bulletin, 49:227. 12A. Williams, "Factor A: Principal Components and Orthogonal Rotations," Michigan State University Computer Institute for Social Science Research, Technical Report No. 31, October 21, 1966. 67 TABLE 3.2--Factor analysis of Person Perception Test scales: GrOUp I. Scales I 11 III IV h2 closed-open .62 .02 .30 -.50 .74 opaque-transparent 25 .21 .21 -.45 .36 far-near 53 .12 .15 -.62 .71 energetic-tired - 28 .ll .32 .70 .70 active-passive - 18 .08 .17 .81 .73 fast-slow - 28 .17 .32 .66 .66 yielding-tenacious 16 .26 .75 .00 .66 aggressive-defensive -.27 .25 .00 .50 .39 excited-relaxed .04 80 .11 .19 70 dynamic-static -.29 .13 .24 .72 .69 upset-calm .25 .77 .21 -.15 .73 unhappy-happy .34 .22 .18 -.62 .59 complex-simple -.15 .10 .55 .00 .34 strong-weak -.22 .50 .54 .04 .59 smug-embarrassed .23 .52 .20 .15 .39 controlled-uncontrolled -.17 .59 .29 -.16 .49 hopeful-fearful -.27 .64 .08 .32 .60 masculine-feminine -.02 .50 .04 -.18 .29 constrained-free .46 .03 .23 -.65 .69 dishonest-honest 75 .06 .15 -.16 .62 profound-superficial -.66 .28 .22 .00 .57 angry-supportive .84 .08 .01 -.Ol .72 perplexed-understanding .62 .06 .02 -.24 .45 disagreeing-agreeing .73 .05 .10 -.19 .58 disapproving-approving .79 .00 .05 -.22 .67 surprised-bored .07 .09 .54 .07 .31 harrassed-unhurried .57 .17 .08 -.15 .38 intense-mild .32 .06 .41 .13 .30 impatient-patient 80 .11 .07 -.07 .66 dissatisfied-satisfied 75 27 .01 -.24 73 Proportion of variance .22 ll .08 .15 56 Legend: I II III IV Evaluative factor Emotional factor Potency factor Activity factor TABLE 3.3--Factor analysis of Person Perception Test scales: GrOUp II. 68 Scales I II III h2 closed-open .86 .01 .00 .74 opaque-transparent .66 .00 .07 .45 far-near .84 .11 .00 .73 energetic-tired .64 .37 .38 .69 active-passive .70 .27 .41 74 fast-slow .65 .27 .47 .73 yielding-tenacious .58 .32 .30 .53 aggressive-defensive .56 .14 .36 .46 excited-relaxed .10 .84 .13 .73 dynamic-static .67 .14 .48 .71 upset-calm .23 .79 .02 .68 unhappy-happy .62 .12 ll .41 complex-simple .06 09 44 .20 strong-weak .33 .39 .59 .63 smug-embarrassed .34 .36 .27 .32 controlled-uncontrolled .21 .61 .22 .46 hopeful-fearful .45 .53 .32 .60 masculine-feminine .07 .55 .18 .35 constrained-free .72 .04 .11 .54 dishonest-honest .67 .12 .13 .49 profound-superficial 53 .35 .05 .40 angry-supportive .81 .02 13 .67 perplexed-understanding .76 .03 14 .61 disagreeing-agreeing .70 .06 .10 .50 disapproving-approving .80 .04 .01 .64 surprised-bored .13 .01 .35 .14 harrassed-unhurried .57 .31 .35 .55 intense-mild .21 .19 .68 .54 impatient-patient .66 .15 .21 .50 dissatisfied-satisfied .82 .00 .02 .68 Proportions of variance .34 ll .09 54 Legend: I = Evaluative factor II = Emotional factor III = Potency factor 69 the total variance and for Group II isolated three factors comprising 54 per cent of the total variance. Those scales which loaded highest on a given factor defined that factor's structure. A close inspection of the factor structures revealed striking similarities between the two groups indicating marked factcr stability. The first factor was named the evaluative factor. For Group I it was defined by ten scales: closed-open, dishonest-honest, superficia1:profound, angry-supportive, perplexed-understanding, disagreeing-agreeing, disapproving- approving, harrassed-unhurried, impatient;patient, dissatisfied-satisfied. These same ten scales also loaded highest on Factor I for Group II. The second factor, defined by the same six scales for both groups, appeared to be mainly an emotional factor: excited-relaxed, upset-calm, embarrassed-smug, uncontrolled- controlled, fearful-hgpeful, feminine-masculine. This factor not only was defined by the same six scales, but also accounted for 11 per cent of the total variance for each group. The third factor was named the potency factor and was defined by the same four scales for both groups: simple- complex, weak-strong, bored-surprised, mild-intense. Again the variance accounted for by this factor was similar for both groups: 8 per cent for Group I; 9 per cent for Group II. 70 Group I used a fourth factor, an activity factor, defined by nine scales: opaque-transparent, far-near, tired- energetic, passive-active, slow-fast, defensive-agressive, static-dynamic, unhappy-happy, constrained-free. Though Group II did not use a fourth factor, a close inspection of the scales revealed that each of the Group I activity scales had coalesced with Group II's evaluative factor, indicating a quasi-stability at least. For this study this factor, defined by the above nine scales, was treated as if it were stable and was named the evaluative-activity factor. Only the scale yielding:tenacious, which loaded high on the evaluative factor for Group II and high on the potency factor for Group I, was not assigned to one of the four factors discussed above. Each of the scale groupings have been summarized in Table 3.4 In summary, the factor analysis for each testing indicated that the counselor-trainees in the two groups judged the filmed client-percepts from several perspectives or dimensions: (1) an evaluative dimension, (2) an emotional dimension, (3) a potency dimension, and (4) an evaluative- activity dimension. These four dimensions constituted the framework by which the Person Perception Test was scored and results analyzed. ScoringProcedures After factor analysis the scales defining each factor were aligned within their respective factor dimensions. By 71 TABLE 3.4-~Person Perception Test factor composition and factor loading for Group I and Group II counselor-trainees. Scales Group I Group II I-Evaluative Factor closed-open .62 .86 dishonest-honest .75 .67 superficial-profound .66 .53 angry-supportive .84 .81 perplexed-understanding .62 .76 disagreeing-agreeing .73 .70 disapproving-approving .79 .80 harrassed-unhurried .57 .57 impatient-patient .80 .66 I-IV-Evaluative-Activity Factor opaque-transparent .45 .66 far-near .62 .84 tired-energetic .70 .64 slow-fast .66 .65 defensive-aggressive .50 .56 passive-active .81 .70 static-dynamic .72 .67 unhappy-happy .62 .62 constrained-free .65 .72 II-Emotional Factor excited-relaxed .80 .84 upset-calm .77 .79 embarrassed-smug .52 .36 uncontrolled-controlled .59 .61 fearful-hopeful .64 .53 feminine-masculine .50 .55 III-Potency Factor simple-complex .55 .44 weak-strong .54 .59 bored-surprised .54 .35 mild-intense .41 .68 Unnamed Factor _ yielding-tenacious III.75 1.58 72 convention the most valued, the most active, and the most potent pole of the scales were placed at the right of the seven-step semantic Space, and for this study the least emotional pole was also placed to the right. Numerical values of 1 through 7 were assigned to each scale position as in the example below: closed 1 : 2 : 3 : 4 : 5 6 7 open passive 1 : 2 3 4 : 5 6 7 active upset 1 : 2 : 3 : 4 5 6 7 calm weak 1 : 2 : 3 : 4 5 ° 6 7 strong Every counselor-trainee judgment was thus converted to a number for scoring purposes.13 Two scoring procedures have typically been used to analyze semantic differential data. One method, hereafter called the "1-7" procedure, was to sum by factor the numbers assigned to counselor-trainee judgments (l) by client- percepts to obtain a client-percept factor score and then (2) across client-percepts to obtain a total factor score. Thus, for each counselor-trainee an evaluative score was calculated for each client-percept and, by summing across client-percepts, for the total test. In a like manner emotional scores, potency scores, and evaluative-activity scores were computed. Each counselor-trainee was given 13Complete item distributions can be found in Appendix B. 73 twenty-four "1-7" procedure scores: 4 factors times 5 client-percepts plus 4 total factors. The second method, hereafter called the frequency or f-procedure, was to tabulate by factor the number of responses in scale positions 2,3,5,6 and to sum by client- percept and across client-percept for each factor. As was noted in Chapter II, Osgood suggested three scale-checking levels with the "discriminatory" level (scale positions 2,3,5,6) being the most difficult for subjects to make. Because of Wozniak's finding that subjects with open belief systems checked scale position 4 more than did subjects with closed belief systems, frequencies of responses in scale position 4 were also tabulated. To complete the f-procedure, scale positions 1 and 7 were similarly tabulated. With the f-procedure each counselor-trainee received 64 f-procedure scores: 3 scale-checking levels times 4 factors times 5 client-percepts plus 4 total factors. The two scoring procedures combined yielded 88 part and total scores. A third scoring system, using variance as a score, was explored. No precedent was found to justify using variance as scores with the semantic differential technique. But it was reasoned that a high variance score would represent more use of the available semantic space than would a low variance score, which in turn would represent a wider range of openness to stimuli on the part of the counselor-trainee (i.e. less rigidity and sterotopy). With 74 the "1-7" procedure as the base, variance scores were computed by using the five client-percept factor scores as the basic data for each counselor-trainee. One variance score was computed for each factor, yielding four additional scores 0 Reliability PPT test-retest data (interval one week) were collected from Group II. Reliability eStimates, using Pearson's product-moment correlations, weie computed for each total factor score as well as for each client-percept factor score. (See Table 3.5.) Reliability estimates for total factor scores were: evaluative factor .71; emotional factor .75; potency factor .80; and evaluative-activity factor .55. The low reliability estimate for the evaluative- activity factor possibly resulted because that factor was the least stable of the four and for Group II was completely artificial. Client-percept factor score reliability estimates were not quite so high, predominantly in the .60's and ranged from .35 for Client-percept Lynn emotional factor score to .78 for Client-percept Bob evaluative factor score. Although Osgood does not ordinarily estimate item reliability, PPT items were subjected to a test-retest reliability check. (See Appendix D.) Item reliability coefficients were not as high as those found for either the total factor scores or the client-percept factor scores. 75 TABLE 3.5--Person Perception Test test-retest reliability estimates using product-moment (r) for each client-percept factor score and for each total factor score for Group II. Client- Factors Percept Evaluative Emotional Potency Eval-Activity Peggy .60 .62 .62 .66 Bill .48 .67 .57 .72 Bob .78 .67 .60 .75 Lynn .76 .35 .61 .45 Terry .62 .66 .62 .51 TOTAL .71 .75 .80 .55 76 Less than half of the items (69 of 150) had estimates higher than .50 and almost a third (46 of 150) had estimates so low that the coefficients computed did not reach the .05 level of significance. However, for this study individual items had relevance only when they were combined to form various factor scores. Osgood argued against the product-moment as an appropriate procedure for semantic differential data on the grounds that 3 does not account for absolute differences between means.14 A perfect correlation could occur when absolute differences are great and no scores are reproduced. Or an indeterminate correlation could occur when each subject gave the same score on test-retest. As an alternative to the product-moment correlation, Osgood suggested computing an error of measurement based upon average absolute deviations. Osgood considered the average absolute deviation to be analoguous to the reproducibility criterion of reliability held for physical measurements. That is, perfect reliability is obtained only when measures on a second testing are identical with measures on a first testing. Deviations from zero represent a certain degree of unreliability. Following Osgood, then, test-retest average absolute deviations were computed by factor for client-percepts and for factor totals. The formula used was (1) to determine the 14Osgood, Suci, and Tannenbaum, pp. 126-140. 77 absolute difference from test to retest for each of the 150 items of the Person Perception Test, (2) to combine the items by factor and add the absolute differences obtained by client-percept and by total factor, and (3) to divide by the number of items included in client—percept factor score or the total factor score. For example, ten items comprise the evaluative factor for Client-percept Peggy. Therefore, after the ten absolute deviations were summed, that total was divided by ten and the resultant figure was the average absolute deviation from test to retest on the evaluative factor for Client-percept Peggy. Figure 3.2 shows the per cent of 150 items (30 scales by five client-percepts) yielding observed average absolute deviations between test-retest scores for Group II. The average absolute deviations by item ranged from as small as .30 scale units to as large as 1.87 scale units. Over half the deviations were less than one scale unit. Although the deviations were slightly greater than Osgood reported finding, the difference perhaps can be explained by the concepts being judged.15 Osgood's subjects were asked to rate generalized concepts (e.g. father, sin, lady), whereas the counselor-trainees in this study were asked to judge quite specific concepts (client-percepts). Deviations 15Osgood, Suci, and Tannenbaum, p. 129. Per Cent 78 15‘- H O 1‘ .//. .. .4 .6 .8 1.0 1:2 1T4 1:6 1.8 Scale Units Figure 3.2--Per cent of items yielding observed absolute deviations between test-retest for Group II expressed in scale units. 79 could be explained by real changes in judgments after having had the opportunity to view the client-percepts a second time. Table 3.6 shows the average absolute deviations, expressed in scale units, computed for each client-percept and summed across client-percepts for each of the four factors. Summing across client-percepts, average absolute deviations for each of the four factors were less than one scale unit with no apparent differences among the factors. As with item deviations, factor deviations were slightly greater than figures reported by Osgood.16 But "we can expect subjects, on the average, to be accurate within a single unit of the scale, which for practical purposes is satisfactory."17 COUNSELOR RESPONSE SYSTEM The depentent variable used in this study was the frequency of selected counselor-trainee interview responses tabulated according to four dimensions of the Counselor Response System (CRS) as rated independently by three judges from interview tapes. DeRoo described the CRS as a "method for objectively analyzing the verbal statements of 16Ibid., pp. 130-139. 17Ibid., p. 131. 80 TABLE 3.6--Group II test-retest factor average absolute deviations by client-percepts and by total expressed in scale units. Client- Evaluative- Percept Evaluative Activity Emotional Potency Peggy .83 .87 .99 .96 Bill .96 1.03 .91 .87 Bob .87 .93 .79 .85 Lynn .92 .95 1.15 1.18 Terry .96 1.17 .99 .91 TOTAL .91 .99 .97 .99 81 "18 counselors during counseling interviews. The dichotomous dimensions used were defined as follows: 1. Content: Follow-Shift. Rank and DeRoo define the Content: Follow-Shift dimension as concerning the general topic of discussion between the client's last state- ment and the counselor's response.19 A response is con- sidered "following" if it does not depart from the general topic. DeRoo reported interjudge reliability of .87 for this dimension. An interjudge reliability of .95 computed by analysis of variance was found for this study. 2. Control: Egpansive-Restrictive. This dimension concerns the extent to which the counselor allows the client to express himself freely. DeRoo reported an interjudge reliability coefficient of .89, while an interjudge reliability estimate of .92 was found for this study. 3. Referrent: Client-Other. This dimension concerns the extent to which the referrent of the counselor's response is the client. A reliability coefficient of .63 was reported 18William M. DeRoo, "A Study of Relationships Between Counselor Personality and COunselor Behavior" (unpublished Doctoral thesis, Michigan State University, East Lansing, 1965), p. 53. 19Richard C. Rank and William M. DeRoo, "Counselor Response System of the Behavior Interaction Description System" (Michigan State University, East Lansing, 1965, mimeographed). 82 by DeRoo and a coefficient of .73 was computed for this study. 4. Reinforcipg Statement. A reinforcing response is a supportive and/or encouraging or positive statement about the client's previous statement. DeRoo did not use this dimension in his research. But Rank had begun experimenting with it and it was decided to have the counselor-trainees in this study judged on the frequency of their reinforcing statements. At the time of this writing no comparative reliability estimates were available. An interjudge reliability coefficient of .94 was computed for this study. The CRS Temporal: Present-Past/Future dimension was also tabulated but interjudge reliability was so low the dimension was eliminated from further consideration in this study. The complete set of dependent variable rating scales, together with their respective definitions and judges' directions, can be found in Appendix E. HYPOTHESES Hypotheses tested for this study were grouped under two headings: (1) Person Perception Test hypotheses, and (2) validity hypotheses. Person Perception Test Hypotheses Person Perception Test hypotheses were concerned with relationships (1) between PPT items and total factor scores, (2) between client-percept factor scores and the semantic 83 differential criterion of concept variability, and (3) between the PPT and person perception theory. Internal Consistency. One mark of a test's validity is the degree its individual items relate with its total score. The PPT had four total scores, one for each factor. Although factor analysis by itself validated the 30 scales with their respective total factor scores, it did not validate the individual items (scale/client-percept pairings) with their respective total factor scores. So the following internal consistency hypotheses were tested: H101: A positive relationship will be found between evaluative items and the total evaluative score ("l-7" procedure). H1.2: A positive relationship will be found between emotional items and the total emotional score ("l-7" procedure). H1.3: A positive relationship will be found between potency items and the total potency score ("l-7" procedure). H1.4: A positive relationship will be found between evaluative-activity items and the total evaluative-activity score ("l-7" procedure). Client-percept Variability. As reported earlier in this chapter, one of the primary criteria for selecting semantic differential concepts was that the investigator 84 try to select concepts that vary in meaning one from the other. The concepts used in this study (client-percepts) were filmed counseling interviews of high school students. The following hypotheses were used to test the difference in meanings among the five client-percepts: ”2.1: A statistically significant difference will be found among the five client-percept evaluative factor scores ("l-7" procedure). H : A statistically significant difference will be found among the five client-percept emotional factor scores ("l-7" procedure). H2.3: A statistically significant difference will be found among the five client-percept potency factor scores ("l-7" procedure). “2.4: A statistically significant difference will be found among the five client-percept evaluative- activity factor scores ("l-7" procedure). Person Perception Theory. Under the theoretical discussion of person perception in Chapter I it was noted that the perceiver tends to interpret his perceptions congruently with his own attitudes and values in spite of the fact that he may be incorrect logically. This tendency was called the "perceiver's expressive style," by Brunner 85 and Taguiri or the perceiver's "generalized meaning for the other" by Cronbach.20 An attempt was made to test the theoretical notion of the perceiver's "generalized meaning" and to assess the degree total factor scores represent the concept of "gener- alized meaning." Theory suggests that, although he may not judge all five client-percepts the same, a counselor-trainee should, nevertheless, have style of judging that would be detectable when his judgments are compared with those judg- ments of other counselor-trainees. For example, if a counselor-trainee had a high potency score ("l-7" procedure) for Client-percept Peggy when compared with other counselor- trainees, he should also have a high potency score for Client-percept Bill, comparatively, despite possible differences in the actual degree of potency seen between the two client-percepts. Or if this same counselor-trainee had a comparatively high potency score for Client-percept Peggy, he should also have a comparatively high total potency score. To test the notion of perceiver's "generalized meaning," the following hypotheses were advanced:t 20For a more complete discussion please refer to Chapter 1, pages 6-12. “3.1‘ 3.2‘ 3.4‘ 86 A positive relationship will be found: a. when each client-percept evaluative factor score is compared with every other client- percept evaluative factor score; when each client-percept evaluative factor score is compared with the total evalua- tive score. A positive relationship will be found: a. when each client-percept emotional factor score is compared with every other client- percept emotional factor score; when each client-percept emotional factor score is compared with the total emotional score. A positive relationship will be found: a. when each client-percept potency factor score is compared with every other client- percept potency factor score; when each client-percept potency factor score is compared with the total potency SCOTS. A positive relationship will be found: a. when each client-percept evaluative- activity factor score is compared with every other client-percept evaluative- activity factor score; 87 b. when each client-percept evaluative- activity factor score is compared with the total evaluative-activity score. Validity Hypgtheses The validity hypotheses tested the relationships between various PPT scores and outside criteria of counselor- trainee interview behavior as measured by the Counselor Response System (CRS). Approach-Avoidance: Evaluative Factor. Osgood related the evaluative factor with a measure of "attitude." Attitudes are "predispositions to respond, but are distinguished from other such states of readiness in that they predispose toward an evaluative response," wrote 21 Osgood. He referred to attitudes as "tendencies to approach or avoidance." For this study it was hypothesized that a significant positive relationship would be found between the generalized evaluative meaning a counselor-trainee had for filmed client- percepts and selected counseling behavior displayed with a particular client. That is, the more positive the total evaluative score ("l-7" procedure) obtained by the counselor- trainee on the PPT, the more his counseling responses, as rated on the CRS, would tend to approach rather than avoid the client. 21Osgood, Suci, and Tannenbaum, p. 189. 88 The following specific hypotheses were advanced: H A positive relationship will be found between 4.1' counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Control: Expansive responses recorded on the CRS. H : A positive relationship will be found between counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Content: Follow responses recorded on the CRS. H : A positive relationship will be found between counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Referrent: Client responses recorded on the CRS. “4.4: A positive relationship will be found between counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Reinforcing responses recorded on the CRS. Approach-Avoidance: Interrelated Factors. In Chapter II it was reported that the evaluative factor alone was not enough for predicting behavior--that the addition of other semantic dimensions could improve predictability. Thus, the following four hypotheses were tested: H : A positive relationship will be found between 5.1 counselor-trainee interrelated PPT total factor S. 5.3‘ 5.4‘ 89 scores ("l-7" procedure) and the frequency of Content: Follow responses recorded on the CRS. A positive relationship will be found between counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Control: Expansive responses recorded on the CRS. A positive relationship will be found between counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Referrent: Client responses recorded on the CRS. A positive relationship will be found between counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Reinforcing responses recorded on the CRS. Scale-Checking Style. Also reported in Chapter II were several studies concerned with scale-checking style. The following hypotheses were based upon the findings of those studies: H 6.1: 6. 2: A positive relationship will be found between counselor-trainee interrelated PPT total factor scores (£2356 procedure) and the frequency of Content: Follow responses recorded on the CRS. A positive relationship will be found between counselor-trainee interrelated PPT total factor 6.3‘ 6.4: 90 scores (f2356 procedure) and the frequency of Control: Expansive responses recorded on the CRS. A A positive relationship will be found between counselor-trainee interrelated PPT total factor scores (£2356 procedure) and the frequency of Referrent: Client responses recorded on the CRS. A positive relationship will be found between counselor-trainee interrelated PPT total factor scores (£2356 procedure) and the frequency of Reinforcing responses recorded on the CRS. Rigidity and Sterotopy. The following hypotheses were tested using variance scores as indicators of counselor- trainees openness to stimuli: H 7.1‘ 7.2‘ 7.3‘ A positive relationship will be found between counselor-trainee interrelated PPT variance scores and the frequency of Content: Follow responses recorded on the CRS. A positive relationship will be found between counselor-trainee interrelated PPT variance scores and the frequency of Control: Expansive responses recorded on the CRS. A positive relationship will be found between counselor-trainee interrelated PPT variance 91 scores and the frequency of Referrent: Client responses recorded on the CRS. H7.4: A positive relationship will be found between counselor-trainee interrelated PPT variance scores and the frequency of Reinforcing responses recorded on the CRS. COLLECTION OF DATA Person Perception Test and Counselor Response System data were collected from both Group I and Group II subjects. The Independent Variable The PPT was administered to each counselor-trainee group studied. Group II received the test twice (one week interval) to estimate test-retest reliability. Procedures for administering the test were the same for each group. The counselor-trainees received a copy of the graphic differential and were instructed to read the directions silently while the examiner read them aloud. After all questions were answered, the first five minute film segment (Client-percept Peggy) was shown. At the conclusion of the segment the projector was turned off and the counselor- trainees were asked to rate, according to test directions, the client-percept they had just viewed. When all counselor- trainees had finished rating the first segment (no attempt was made to time them), the second client-percept segment 92 was shown and the procedure was repeated until all five client-percepts were judged. The raw data were transferred to computer cards, one card for each subject/client-percept pairing, and a separate card deck for each testing was then submitted to the computer for factor analysis, scoring, and reliability estimates. The Dependent Variable First-interview tapes were procured from counselor- trainees in both Group I and Group II. Twenty-nine tapes were turned in by Group I subjects, but only twenty-four were usable since five tapes were inaudible and no effective rating was possible for them. From Group II twenty-eight tapes were turned in and twenty-six were usable. For ease of judging, the first half hour of each usable tape was transferred to a fresh tape, allowing six half hour sessions to be placed on one continuous tape. Three judges, each with Master's or Specialist's degrees in counseling and guidance, were used to rate the tapes. They received their respective degrees in separate institutions: Michigan State University, Western Michigan University, and Florida State University.. The diversity of the judges' background precluded any systematic bias by theoretical training from contaminating the CRS dimensions. Before beginning to judge counselor-trainee interview responses on the CRS dimensions described earlier in this chapter, the three judges practiced on a tape not used for 93 study. After they were rating in a similar manner, the three judges independently rated the counselor-trainee tapes. Interjudge reliability was computed by analysis of variance. Average judges' ratings were used as the best estimate of counselor-trainee responses on each dimension. The following is a summary of interjudge reliability estimates for this study compared with those from DeRoo's study: CRS Dimension Present Study DeRoo Stugy Content: Follow .95 .87 P Control: Expansive .92 .89 Referrent: Client .73 .63 Reinforcing Statement .94 (*)22 £ ANALYSIS OF DATA Seven sets of hypotheses were tested in this study subsumed under two major categories: (1) Person Perception Test hypotheses and (2) validity hypotheses. Person Perception Test Hypotheses Internal consistency hypotheses were tested by the Pearson product-moment correlation coefficient method (1). A correlation large enough to reveal a significant relationship with 95 per cent confidence was accepted as evidence supporting the hypotheses. A 22DeRoo did not use this dimension in his study. 94 The four client-percept variability hypotheses were tested by a simple one-way analysis of variance computed across client-percepts. An F ratio large enough to be significant with 95 per cent confidence was accepted as evidence supporting the hypotheses that a difference in meaning existed among the client-percepts. Where the F ration was found to be significant, tests of differences by use of p'as described by Garrett were computed.23 The four theory hypotheses were tested by the rank- difference (£22) method to determine the degree of the relationships. The rank-difference method had the advantage of not assuming that the distribution of scores on the PPT were normal. A 222 large enough to be significant with 95 per cent confidence was accepted as evidence supporting the theoretical notion of perceiver's "generalized meaning." Validity Hypotheses The relationship between various PPT scores and selected counseling interview behavior as recorded on the CRS was also hypothesized. Gleser wrote:' The only valid basis for prediction is experimental verification of the relationship between potential predictors and actual performance Of a group of subjects who are representative of the popu- lation for which it is desired to make predictions.24 23Henry E. Garrett and R. S. Woodworth, Statistics (New York: Longmans, Green and Company, 1958), pp. 280-281. 246. C. Gleser, "Prediction," Encyclopedia of Educa- tional Research (3rd edition), C. W. Harris, editor (New YOrk: The Macmillan Company, 1960), p. 1039. 95 For the first set of validity hypotheses the rank- difference method was used since it had the advantage Of not assuming that the distribution of scores on the PPT nor on the CRS was normal. Moreover, the relative order of judgments were more important than the actual scores because of the experimental nature Of the instruments used which seemed to make the conservative rank-difference method an appropriate technique. A £22 large enough to indicate a relationship with 95 per cent confidence was accepted as evidence in support of the hypotheses. But for the remaining validity hypotheses, where the interrelated PPT factor scores were compared with the CRS criteria, the rank-difference method was contraindicated. Instead, a coefficient Of multiple correlation, R, was computed to test eaCh hypothesis. An R large enough to indicate a relationship with 95 per cen} confidence was accepted as evidence supporting the hypotheses. Exploratoyy Analyses In this day of computer analyses, it is just as easy to compute all possible relationships as it is to compute the hypothesized ones. Since, as Sproull noted, computers can "now provide fast and comprehensive analysis," many exploratory analyses of data were made:25 25Natalie Sproull, "Finding and Stating the Research Problem: A Suggested Approach with Examples from Higher Edu- cation," (mimeographed), School Of Advanced Studies, College of Education, Michigan State University, June 1967, p. 6. 96 1. In addition to the evaluative factor, each of the other PPT factor scores ("l-7" procedure)--emotional, evaluative-activity and potency--were compared with the CRS criteria by the product-moment, 1, method. 2. Using the f-procedure, each PPT factor score was compared with the CRS criteria by the product-moment method. 3. Using the variance procedure, each PPT factor score was compared with the CRS criteria by the product- moment method. 4. Each Of the 150 PPT items were compared with the CRS criteria by the product-moment method. SUMMARY The subjects in this study were graduate students enrolled in the 1965-66 academic year guidance and counseling institute at Michigan State University (Group I) or in the 1966 summer counseling practicum at Michigan State University (Group II). The PersonxPerception Test (PPT), developed especially for this study, consisted of five filmed counseling segments judged on a 30 scale graphic semantic differential. Factor analysis of the 30 scales revealed four judgmental dimen- sions: (l) evaluative, (2) emotional; (3) potency, and (4) evaluative-activity. Three scoring procedures were used to tabulate PPT data: (I) checked scale position numbers were summed by factor ("l-7" procedure), (2) frequency of responses in 97 various scale positions were summed by factor (f-procedure), and (3) variance scores were computed using the five client- percept factor scores as the basic data (V procedure). Test-retest reliability estimates ranging from .55 to .80 were reported using Pearson's product-moment. Item test-retest reliability was also reported using Pearson's product-moment. Average absolute deviations were compared favorably with Osgood's typical findings. The Counselor Response System (CRS), developed by Rank and DeRoo, was used as the dependent variable. Three judges, working independently and from interview tapes, rated the counselor-trainees' responses along the following dichotomous dimensions: CRS Dimensions InterjudggfiReliability Content: Follow-Shift .95 Control: Expansive-Restrictive .92 Referrent: Client-Other .73 Reinforcing-Nonreinforcing .94 Hypotheses tested for this study were grouped under two headings: (1) Person Perception Test hypotheses and (2) validity hypotheses. The first set of hypotheses tested the internal consistency of factor items with total factor scores, the differences among client-percept meanings, and the theoretical notion of the perceiver's "generalized meaning." The second set of hypotheses tested the validity 98 Of the PPT by analyzing the relationshipsfbetween various PPT scores and CRS dimensions. PPT and CRS data were collected from bothwGroup I and Group II subjects. Raw data were transferred to computer cards and submitted to the computer for factor analysis, scoring, and reliability and validity estimates. Internal consistency hypotheses were tested by the Pearson product-moment method. The differences among client- percept meanings were computed by one-way analyses of variance. The PPT-theory hypotheses were tested by the rank- difference method to determine the degree of the relation- ships. Validity hypotheses were tested by the Spearman rank- difference method or by the coefficient of multiple correlation. Many exploratory analyses of the data were made. In the next chapter a detailed analysis of the data resulting from the Person Perception Test hypotheses will be reported. CHAPTER IV ANALYSIS OF DATA: PERSON PERCEPTION TEST HYPOTHESES The preceding chapter described the design of this study. Particularly important was the discussion Of the Person Perception Test (PPT), an instrument especially created for this study. Development of the PPT was traced (1) from the selection and filming of the five client- percepts and the selection of the bipolar adjectives used as semantic differential scales, (2) through a factor analysis Of the data generated and the development of various scoring procedures, (3) to a presentation Of several reliability estimates. Three sets of hypotheses were advanced which were concerned with the PPT development. In this chapter results Of the PPT hypothesized relationships will be presented in the following order: (1) internal consistency--a study of the relationships between PPT items and total factor scores, (2) concept variability--a study of the differences among client- percept factor scores, and (3) perceiver's "generalized meaning"--a study of the relationship between individual client-percept factor scores and total factor scores and the relationships of client-percept factor scores with each 99 100 other. The chapter will conclude with a discussion of the results and a complete summary. Two separate groups of counselor-trainees were used to test the hypotheses: Group I was the experimental group; Group II, the replicating group. The .05 level of significance was accepted as the criterion for rejecting the null hypothesis. INTERNAL CONSISTENCY The hypotheses in this section tested the relation- ships between factor items and their respective total factor scores and between factor items and client-percept factor scores. Such relationships constituted one indication of the PPT's internal consistency. For a complete list Of item/total factor score correlations see Appendix 1. Because total factor scores were the Operational definitions of the perceiver's "generalized meanings" for person percepts, the hypotheses tested the relationships between a perceiver's particular judgments of a person percept and his total judgments Of person percepts generally. Likewise, the hypotheses tested the relationships between a perceiver's particular judgments of a person percept and his total judgments of that person percept by comparing items with client-percept factor scores. The Pearson product-moment was used to test all hy- potheses in this section, with the .05 level of significance used as the criterion for rejecting the null hypothesis. 101 Evaluative Factor H01 1: No relationship will be found between: ° a. evaluative factor items and the total evaluative factor score ("l-7" procedure); b. evaluative factor items and client-percept factor scores ("l-7" procedure). A positive relationship will be found between: a. evaluative factor items and the total evaluative factor score ("l-7" procedure); b. evaluative factor items and client-percept factor scores ("l-7" procedure). 1.1‘ The evaluative factor consisted of 50 items (10 scales times 5 client-percepts). With total factor score as the criterion, the experimental group found forty significant relationships from the possible fifty. And thirty of the fifty items were found to be significant by the replicating group. More importantly, however, 24 of the fifty items were found to be significantly related with the total factor score by both groups. (See Table 4.1.) When the items were correlated with their respective client-percept factor scores, all 50 possible relationships were found significant for the experimental group. (See Table 4.2.) Moreover, the replicating group confirmed 47 of the fifty significant relationships. Emotional Factor H01 2: No relationship will be found between: ' a. emotional factor items and the total emotional factor score ("l-7" procedure); b. emotional factor items and client-percept factor scores ("l-7" procedure). 102 TABLE 4.1--Items significantly related to total factor score (”l-7" procedure) by both groups: Evaluative factor. Evaluative Scales Client-Percept GrOUp I Group II closed-open Peggy .40 .38 Bob .59 .42 dishonest-honest Bob .65 .49 Terry .49 .40 superficial-profound Bob .52 .54 angry-supportive Bob .46 .61 Terry .58 .40 perplexed-understanding Bill .52 .50 Bob .67 .59 disagreeing-agreeing Bill .44 .47 Bob .56 .55 Lynn .62 .37 disapproving-approving Peggy .48 .49 Bill .37 .46 Bob .62 .54 Terry .49 .39 harrassed-unhurried Bob .59 .55 Lynn .45 .61 Terry .41 .50 impatient-patient Bob .53 .37 Lynn .57 .37 Terry .49 .37 dissatisfied-satisfied Peggy .64 .37 Bill .43 .60 103 TABLE 4.2-{tem/client-percept score ("l-7" procedure) correlation coefficients (1) for both groups: Evaluative factor. Client-Percepts Scales Group Peggy Bill Bob Lynn Terry closed-Open It .55 .83 .91 .64 .64 II .49 .60 .64 .15 .511 dishonest-honest I .74 .75 .93 .67 .68 II .43 .66 .75 .37 .55 superficial-profound I .65 .56 .87 .69 .71 II .68 .72 .82 .22 .37 angry-supportive I .73 .73 .87 .80 .76 II .45 .61 .82 .53 .87 perplexed-understanding I .61 .67 .63 .71 .69 II .37 .75 .77 .46 .47 disagreeing-agreeing I .69 .80 .89 .72 .77 II .61 .70 .69 .67 .64 disapproving-approving I .69 .73 .90 .82 .85 II .45 .74 .82 .63 .86 harrassed-unhurried I .59 .70 .52, .58 .74 II .36 .46 .78 .76 .39 impatient-patient I .72 .80 .88 .75 .81 II .75 .63 .52 .72 .77 dissatisfied-satisfied I .81' .78 .94 .79 .81 II .63 .83 .59 .49 .66 104 H1 2: A positive relationship will be found between: ° a. emotional factor items and the total emotional factor score ("l-7" procedure); b. emotional factor items and client-percept factor scores ("l-7" procedure). The emotional factor summed across the five client- percepts consisted of thirty items (6 scales times 5 client- percepts). Twenty-one of the thirty possible relationships between item and total factor score were found to be significant by the experimental group. The replicating group confirmed sixteen of the twenty-one significant relationships. Excited-relaxed and upset-calm related significantly with total factor score for all five of the client-percepts. (See Table 4.3.) Fearful:hopeful was found significantly related when paired with every client- percept except Client-percept Peggy. On the other hand, feminine-masculine did not relate significantly with total factor score for any Of the client-percepts. The items were also compared with client-percept factor scores. (See Table 4.4.) The scale feminine- masculine was generally found to be not significantly related with the criterion or if a relationship were found by the experimental group, it was not confirmed by the replicating group. The scales embarrassed-smug and uncontrolled- controlled when paired with Client-percept Bob and Client- percept Lynn respectively had unconfirmed relationships with the criterion. All other items had confirmed significant relationships with the criterion. In total the experimental 105 TABLE‘4.3~{tems significantly related tO total factor score ("l-7" procedure) by both groups: Emotional factor. i L Emotional Scales Client-Percept Group I Group II excited-relaxed Peggy .46 .44 Bill .43 .37 Bob .45 .64 Lynn .40 .42 Terry .55 .66 upset-calm Peggy .54 .65 . Bill .52 .59 Bob .48 .57 Lynn .39 .51 Terry .38 .68 embarrassed-smug Bill .38 .47 uncontrolled-controlled Lynn .42 .37 fearful-hopeful Bill .49 .45 Bob .50 .41 Lynn .53 .37 Terry .66 .63 feminine-masculine none —- — -_ 106 TABLE 4.4--Item/client-percept score ("l-7" procedure) correlation coefficients (3) for both groups: Emotional factor. Client-Percepts Scales Group Peggy Bill Bob Lynn Terry excited-relaxed I .77 .60 .66 .82 .82 II .70 .79 .88 .75 .83 upset-calm I .85 .79 .82 .83 .85 II .81 .85 .88 .87 .83 embarrassed-smug I .63 .73 .59 .58 .44 II .39 .43-.07 .58 .56 uncontrolled-controlled I .66 .54 .70 .30 .79 II .60 .46 .72 .67 .58 fearful-hopeful I .74 ‘ .70 .67 .64 .74 II .73 .83 .85 .75 .62 feminine-masculine I .28 .49 .63 .03 .36 II .05 .42 .17 .21 .33 107 group found twenty-eight of a possible thirty relationships to be significant, with twenty-six of the twenty-eight -significant.relationships.confirmed by the replicating group. PotengyAFactor H01 3: No relationship will be found between: ' a. potency factor items and the total potency factor score ("l-7" procedure); b. potency factor items and client-percept factor scores ("l-7" procedure). H1 3: A positive relationship will be found between: ' a. potency factor items and the total potency factor score ("l-7" procedure); b. potency factor items and client-percept factor scores ("l-7" procedure). Twenty items (4 scales times 5 client-percepts) comprised the potency factor. With the total factor score as the criterion, the experimental group found only three significant relationships from the possible twenty. And although the replicating group found seven significant relationships, none of them confirmed the experimental group's findings. The potency factor, alone Of the four factors, yielded no confirmed evidence for rejecting the null hypothesis when total factor score was the criterion. Potency items were then compared with client-percept factor scores. (See Table 4.5.) By contrast, seventeen of the possible twenty relationships were found significant for the experimental group, each of which was confirmed by replication. Only for bored-surprised when Client-percept Peggy factor score and Client-percept Lynn factor score were the criteria and mild-intense when Client-percept Lynn 108 TABLE 4.5--Item/client-percept score ("l-7" procedure) correlation coefficients (r) for both groups: Potency factor.‘ _ Client-Percepts Scales Group Peggy Bill Bob Lynn Terry simple-complex I .60 .64 .64 .69 .65 II .72 .61 .68 .62 .52 weak-strong I .57 .46 .71 .57 .42 II .79 .68 .60 .57 .57 bored-surprised I .24 .79 .57 .28 .60 II .51 .58 .44 .65 .46 mild-intense I .66 .49 .39 .24 .56 II .70 .55 .70 .72 .78 109 factor score was the criterion were no significant relationships found. Evaluative-Activity_Factor H01 4: NO relationship will be found between: ' a. evaluative-activity factor items and the total evaluative-activity score ("l-7" procedure); b. evaluative-activity factor items and client-percept factor scores ("l-7" procedure). H1 4: A positive relationship will be found between: ' a. evaluative-activity factor items and the total evaluative-activity score ("l-7" procedure); b. evaluative-activity factor items and client-percept factor scores ("l-7" procedure). The evaluative-activity factor consisted of forty- five items (9 scales times 5 client-percepts). Twenty-eight items were found by the experimental group to relate significantly with the total factor score. Eighteen significant relationships were found by the replicating group, but only nine were common to both groups. (See Table 4.6.) Moreover, five of the nine were contributed by Client-percept Bob, while Client-percepts Peggy and Lynn, the two female percepts, contributed no confirmed items. Three scales had no confirmed relationships with the total factor score irrespective of its client-percept pairing. One of these scales, slow-fast, was a marker scale for this factor. When the evaluative-activity factor items were correlated with their respective client-percept factor 110 TABLE 4.6--Items significantly related to total factor score ("l-7" procedure) by both groups: Evaluative-activity factor. . . . Client- Evaluative-ActiVity Scales Percept Group I Group II opaque-transparent none -- -- far-near Bob .60 .53 tired-energetic Bob .44 .48 passive-active Bill .41 .37 Bob .49 .51 slow-fast none -- -- defensive-aggressive Bob .39 .52 Terry .48 .45 static-dynamic none -- -- unhappy-happy Bob .56 .40 Terry .47 .37 constrained-free Terry .53 .43 111 scores, forty-two of the forty-five relationships were found to be significant by the experimental group. (See Table 4.7.) Furthermore, forty-one of the forty-two significant relationships were confirmed by the replicating group. The four non-significant Or unconfirmed items involved only two scales: gpaque-transparent and unhappy- 9222):. To summarize, briefly, when the total factor Scores were the criteria, the null hypothesis was rejected forty- eight of a possible 145 times. The evaluative factor contributed the most confirmed items (24 of a possible 50), the emotional factor contributed the most confirmed items by percentage (16 of a possible 30), while the potency factor contributed no items to the total of confirmed items. With the client-percept factor scores as the criteria a much higher percentage of confirmed relationships was found. Of the possible 145 relationships, 134 were found significant by the experimental group and confirmed by the replicating group. The evaluative factor had confirmed significant relationships for forty-nine of a possible fifty items; the evaluative-activity factor had forty-two of a possible forty-five; the emotional factor, twenty-six of thirty; and the potency factor, seventeen of twenty. CLIENT-PERCEPT MEANINGS One criterion for selecting semantic differential concepts was to select concepts that vary in meaning one 112 TABLE 4.7--Item/client-percept score ("1-7" procedure) correlation coefficients (3) for both groups: Evaluative- activity factor. ‘ i f Client-Percepts Scales Group Peggy Bill Bob Lynn Terry opaque-transparent I .ll .40 .20 .59 .66 ‘ II .17 .62 .62 .36 .60 far-near I .72 .86 .81 .65 .80 II .44 .78 .77 .44 .59 tired-energetic I .67 .58 .71 .65 .78 II .43 .68 .72 .54 .56 passive-active I .61 .84 .79 .51 .74 II .74 .71 .78 .59 .63 slow-fast I .60 .70 .62 .64 .77 II .81 .62 .79 .55 .77 defensive-aggressive I .47 .68 .50 .46 .65 II .48 .76 .57 .57 .62 static-dynamic I .66 .73 .64 .70 .53 II .55 .75 .73 .49 .79 unhappy-happy I .70 .42 .77 .24 .64 II .52 .57 .69 .41 .11 constrained-free I .58 .72 .59 .62 .65 II .61 .74 .66 .50 .77 113 from the other. To determine client-percept meanings, separate factor means ("l-7" procedure) were computed for each client-percept. Table 4.8 presents the means per factor for each group. By dividing each factor mean by the number of scales comprising each factor, an average semantic meaning was computed for each factor/client-percept combination. (See Table 4.9.) A Some general statements about each client-percept/ factor combination can be made: 1. Evaluative factor: Client-percepts Peggy, Bill, and Bob were valued similarly, and highly (Group 1, 5.0-5.4; Group II, 5.3-5.5) while Client-percepts Lynn and Terry were value similarly and somewhat lower (Group I, 4.0-4.6; Group II, 3.0-3.6). 2. Emotional factor: Again, both groups saw the client-percepts relatively similarly. Client-percept Bob was the most emotive (Group I, 5.0; Group II, 4.8), Client- percept Peggy the least (Group I, 2.9; Group II, 3.2). 3. Potency factor: Though both groups saw Client- percepts Bill as the least potent, the range over all client- percepts was small (one scale unit for Group 1; less than one scale unit for Group II). I 4. Evaluative-activity factor: Client-percept Peggy was judged by both groups as being the most active (Group I, 5.2; Group II, 5.3), while Client-percept Lynn was seen as the least active (Group I, 3.1; Group II, 2.5). Except for 114 4.8-~Factor means by client-percepts for both groups. W Client-Percept Group I Group II Evaluative Factor (10 scales) . Peggy 54.0 53.4 Bill 50.2 54.6 Bob 50.8 54.7 Lynn 40.2 29.8 Terry 46.2 36.2 Emotional Factor (6 scales) Peggy 17.6 19.1 Bill 25.7 28.8 Bob 29.8 32.0 Lynn 23.0 23.3 Terry 25.8 25.4 Potency Factor (4 scales) Peggy 16.8 17.4 Bill 15.5 15.3 Bob 19.4 19.2 Lynn 19.6 18.7 Terry 17.6 17.8 Evaluative-Activity Factor (9 scales) Peggy 46.9 48.1 Bill 30.6 34.4 Bob 43.2 42.2 Lynn 27.6 22.1 Terry 36.4 28.0 115 TABLE 4.9--Average semantic meanings for the client- percepts by factor for both groups. =— Client-Percept Group I Group II Peggy B111 Bob Lynn Terry Peggy Bill Bob Lynn Terry Peggy Bill Bob Lynn Terry Peggy Bill Bob Lynn Terry Evaluative Factor b-bU‘ILDU'I GOP-‘05 0404010101 C‘OMU‘M Emotional Factor bLNU‘I-D-N 040000450 bum->04 wamN Potency Factor hhbwb #LOCXDKON #bbub muooOO-h Evaluative-Activity Factor #Mbmm OHm-hN CNN-#0401 l—amoooooa 116 Client-percept Terry, both groups viewed each of the client- percepts relatively similarly. In summary, there were apparent differences in meaning among the five client-percepts on each factor as judged by the two studied groups. To determine if the apparent differences in meaning were statistically signifi- cant, a simple one-way analysis Of variance was computed acrOss client-percepts by factor. Where the F ratio was found to be significant, tests of differences by use of p_as described by Garrett were computed.1 Evaluative Factor H02 1: No statistical difference will be found among ' the five client-percepts when evaluative scores ("l-7" procedure) are used as the criterion. H2 1: A statistically significant difference will ' be found among the five client-percepts when evaluative scores ("l-7" procedure) are used as the criterion. A Significant difference was found among the five client-percepts on the evaluative dimension by both the experimental and replicating groups. (See Table 4.10.) Thus, the null hypothesis was rejected, the evidence supporting the alternative hypothesis. Specifically, the experimental group found significant differences between Client-percepts Peggy.and Lynn, Bill and 1Henry E. Garrett and R. S. Woodworth, Statistics (New York: Longmans, Green and Company, 1958), pp. 280-281. 117 TABLE 4-10--Differences among client-percept factor scores ("l-7" procedure) by analysis of variance, for both groups: Evaluative factor. Source df 55 ms F SD Group I--Experimenta1 Group Between 4 3241.31 810.33 8.77*- Within 140 17320.10 123.72 11.1 Total 144 20561.41 Group II--Replicating Group Between 4 16894.30 4223.58 62.30* Within 145 9830.07 67.79 8.2 Total 149 26724.37 *Significant at .01 level. Tests of differences by p: Client-Percept Group I Group II Peggy-Bill NS NS Peggy-Bob NS NS Peggy-Lynn .01 .01 Peggy-Terry .01 .01 Bill-Bob NS NS Bill-Lynn .01 .01 Bill-Terry NS .01 Bob-Lynn .01 .01 Bob-Terry NS .01 Lynn-Terry NS .01 Legend: NS=not significant; .01=significant at the .01 level. 118 Lynn, Bob and Lynn, and Peggy and Terry, each of which was confirmed by the replicating group. ’In addition, the replicating group found differences betweenvClient- percepts Bill and Terry, Bob and Terry, and Lynn and Terry. In summary, Client-percepts Peggy, Bill, and Bob were seen evaluatively the same as each other (i.e. no significant differences among them) and evaluatively different from Client-percepts Lynn and Terry. Emotional Factor H02 2: NO statistical difference will be found among ' the five client-percepts when emotional scores ("l-7" procedure) are used as the criterion. H2 2: A statistically significant difference will ' be found among the five client-percepts when emotional scores ("l-7" procedure) are used as the criterion. A significant difference was found among the five client-percepts on the emotional dimension by both the experimental and the replicating groups. (See Table 4.11.) Therefore, the null hypothesis was rejected and evidence was submitted hi support of the alternate hypothesis. The experimental group found significant differences among nine of the ten comparisons, eight of which were confirmed by the replicating group. Only the difference found between Client-percepts Lynn and Terry were not seen as significantly different by the replicating group. On the other hand, the replicating group also found significant differences in nine of the ten comparisons, 119 TABLE 4.11--Differences among client-percept factor scores ("l-7" procedure) by analysis of variance, for both groups: Emotional Factor. Source df 55 ms F SD Group I--Experimental Group Between 2391.64 2391.65 597.91 26.48* Within 140 3274.41 23.39 4.75 Total 144 5666.06 Group II--Replicating Group Between 4 2961.29 740.32 23.81* Within 145 4508.50 31.09 5.50 Total 149 7469.79 *Significant at .01 level. Tests of differences by p: Client-Percept Group I Group II Peggy-Bill .01 .01 Peggy-Bob .01 .01 Peggy-Lynn .01 .01 Peggy-Terry .01 .01 Bill-Bob .01 .05 Bill-Lynn .05 .01 Bill-Terry NS .05 Bob-Lynn .01 .01 Bob-Terry .01 .01 Lynn-Terry .05 NS Legend: NS=not significant; .01=not significant at the .01 level; .05=significant at the .05 level. 120 with the difference between Client-percepts Bill and Terry not confirmed. Potency Factor H02 3: No statistical difference will be found among ' the five client-percepts when potency scores ("l-7" procedure) are used as the criterion. H2 3: A statistically significant difference will be found among the five client-percepts when potency scores ("l-7" procedure) are used as the criterion. A significant difference was found among the five client-percepts on the potency dimension by both the experimental and replicating groups. (See Table 4.12.) Therefore, the null hypothesis was rejected. For the experimental group seven of the ten client- percept comparisons were found significant and each of the seven.‘was confirmed by the replicating group. Evaluative-Activity Factor H02 4: NO statistical difference will be found among ‘ the five client-percepts when evaluative- activity scores ("l-7" procedure) are used as the criterion. 2 4: A statistically significant difference will ' be found among the five client-percepts when evaluative-activity scores ("l-7" procedure) are used as the criterion. A significant difference was found among the five client-percepts on the evaluative-activity dimension by both the experimental and replicating groups. (See Table 4.13.) The null hypothesis was, therefore, rejected in favor Of the alternate hypothesis. 121 TABLE 4.12--Differences among client-percept factor scores (81-7" procedure) by analysis of variance, for both groups: Potency factor. Source ,df 55 ms F SD Group I--Experimenta1 Group Between 4 354.79 88.70 13.10* Within 140 947.24 6.77 ~ 2.60 Total 144 1302.03 Group II--Replicating Group Between 4 278.60 69.65 5.58* Within 145 1810.90 12.49 2.36 Total 149 2089.50 *Significant at .01 level. Tests of differences of p: Client-Percept Group I Group II Peggy-Bill NS .01 Peggy-Bob .Ol .01 Peggy-Lynn .01 .05 Peggy-Terry NS NS Bill-Bob .01_ .01 Bill-Lynn .01 .01 Bill-Terry .01 .01 Bob-Lynn NS .NS Bob-Terry .01 .05 Lynn-Terry .01 .01 Legend: NS=not significant; .01=significant at the .01 level; .05=significant at the .05 level. 122 TABLE 4.13--Differences among client-percept factor scores (1"-7" procedure) by analysis of variance, for both groups: Evaluative-activity factor. Source df 55 ms SD Group I--Experimental Group Between 4 7708.87 1927.22 30.50* . Within 140 4 8857.79 63.27 7.94 Total 145 16566.66 Group II--Replicating Group . Between 4 13162.46 3290.62 49.19* Within 145 9699.37 66.89 8.18 Total 149 22861.83 *Significant at .01 level. Tests of differences by p: Client-Percept Group I Group II Peggy-Bill .01 .01 Peggy-Bob NS .01 Peggy-Lynn .01 .01 Peggy-Terry .01 .01 Bill-Bob .01 .01 Bill-Lynn NS .01 Bill-Terry .01 .01 Bob-Lynn .01 .01 Bob-Terry .01 .01 Lynn-Terry .01 .01 Legend: NS=not significant; .01=significant at the .01 level. 123 With~the client-percept comparisons, thetexperi- mental group found significant.differences between eight Of the ten possibilities: "No-significantidifferences were foundzbetween Client-percepts Peggy and"BOb and: between'Client-percepts.Bill and Lynn. The“rep1icating group foundtsignificant.differences“for*all;ten:possible comparisons; thus confirming the eight found by the experimental“group. In summary, the F ratios computed to testithe'above" four hypotheses were all significant; 'Therefore, it was concluded that the semantic differential concepts variabilityicriterion‘was'met'for"thisfstudy:i"Not'only were all four F ratios significant, but*of"the'tota1 of forty pfs:the experimental group had.twenty-eight A significant £25 and the replicating group had thirty- four.' In addition, the two groups agreedron the: significance of twenty-seven of the differences in meaning. PERSONuPERCEPTION'THEORY-. An attempt was made to test the theoreticalznotion' of the perceiveris'"generalized“meaning”+and—torasse35"the' degree"thatttotal factor scores represented the concept Of’ "generalizedtmeaning." ‘Theory suggestSithat, although he” may not'judge all five clienttpercepts the same, a counselor- traineeshould nevertheless,'have a style of judging that would be detectable when his judgments are compared with 124 those judgments of other counselor-trainees; ~For example, if a counseloretrainee had a high petency score ("l-7" procedure) for Client-percept;Peggy7when:compared with ' other"counselorrtrainees,'he“should'alsoxhave“a'high potency score for'Client-percepthill,:comparatively, despite'possible‘differencesrinfthecactua12degree“of potency~seen'between the.twO'clientspercepts.:‘Or if thiS' same-counselorttrainee'had'a'comparativelythigh"potency score'for‘Client-percept Peggy,.he should also have a comparatively high total potency score; '"Generalized meaning"-was assessed in two ways:' (1)2by~comparing-client-percept-factortscoreS‘withieach other'andx(2).by.comparingrclientepercept:factor;scores with total factor scores.- The firstéwas a quitegrestrictive and-stringent view of perceptual style; :With this view, if the comparative meaning attachedito a particular person perceptiwere known, thenthecomparative:meaningiattached' to another‘particular person:percept could be predicted. The secondTwas a general view of perceptual style;. That is, if the'comparative meaning.attached:t0harparticular'person percept were“known, the comparative meaning attached to person percept in general could be predicted.: 7““Each set of client-percept factor scores and~each' total factor score was ranked separately.and comparisons' werermadeiby'the ranktdifference method (222) to determine the existence of a relationship. 125 Evaluative Factor H03 1: NO relationship will be found: ‘ a. when each client-percept evaluative factor. score is compared with every other client- percept evaluative factor score ("1- -7" procedure); . b. when each client- -percept evaluative factor score is compared with the total evalua- tive score'(Pl-7" procedure). H3 1: A positive relationship will be found: ' a. when each client-percept evaluative factor score.is compared with every other client- percept evaluative factor score ("l-7" procedure); b. when each client- -percept evaluative factor score is compared with the total evalua- tive score (”1- 7" procedure). Significant relationships were found when Client- percept Peggy was comparedlwith:Client-percept Lynn and when Client-percept Lynn was comparedqwith Client-percept Terry. (See Table 4.14.) None.of thelotherrclient-percept comparisons were significant-for the experimental group. Under replication, however, neither significant relationship was confirmed despite the fact thatithe replicating group had three significant relationships itself. There- fore, the null hypothesis was.net rejected for any of the ten possible comparisons. , Despite lack of confirmed client-percept relation- ships, however, nine Of the comparisons were in the predicted direction for the experimental group, and ten of the comparisons were in the predicted direction for the replicating group. 126 TABLE 4.14--Client-percept factor score comparisons by the rank- difference method (rho), for each group: Evaluative factor. Client-Percept Exgggigehtal Regiiggtiig Peggy-Bill .11 '21 Peggy-Bob ~32 '38* Peggy-Lynn .59** 9 -16 Peggy-Terry ~21 '00 Bill-Bob .21 -40* Bill-Lynn .00 I '40* Bill-Terry -.03+ -11 Bob-Lynn ~34 '09 Bob-Terry ~23 '07 Lynn-Terry .78** '02 Legend: ** relationship significant at .01 level; relationship significant at .05 level, direction Opposite that which was predicted. 127 When client-percept factor totals were compared with total factor score, the experimental group found all. relationships to be significant. '(See Table 4.15.) All significant relationships were-confirmed by the replicating group. Emotional Factor “03.2: No relationship will be found: . a. when each client-percept emotional factor score is compared with every other client- percept emotional factor score ("l-7" procedure); b. when each client-percept emotional factor score is compared with the total emotional score (”l-7" procedure). H3.2: A positive relationship will be found: a. when each client-percept emotional factor score is compared with every other client- percept emotional factor score ("l-7" procedure); b. when each client-percept emotional factor score is compared with the total emotional score ("l-7" procedure). NO significant relationships were found for any of the ten client-percept comparisons for the experimental group, though a significant relationship was found when Client-percept Bill was compared with Client-percept Bob for the replicating group. (See Table 4.16.) Therefore, the null hypothesis was not rejected. All ten client-percept comparisons were in the predicted direction for the experimental group. Eight of the ten were confirmed by replication. Five of five possible relationships were found to be significant by the experimental group and confirmed by the 128 TABLE 4.lS--Re1ationships between client-percept factor Scores and total factor score using the rank-difference ethod (rho), for each group: Evaluative factor. . _ Group I V Group 11 Client Percept Experimental Replicating Peggy .64** .54** Bill .37* .72** Bob .68** .69** Lynn .83** .52** Terry .69** .45* Legend: ** relationship significant at .01 level; 1: relationship significant at .05 level. TABLE 4.16--Client-percept factor score comparisons by the rank-difference method (rho), for each group: Emotional factor. ‘ . Grou I Grou II Client-Percept Experimgntal Replicgting Peggy-Bill .19 .19 Peggy-Bob .17 .20 Peggy-Lynn .16 .10 Peggy-Terry .13 .15 Bill-Bob .24 .39* Bill-Lynn .02 -.06+ Bill-Terry .17 .26 Bob-Lynn .13 -.10+ Bob-Terry .34 .25 Lynn-Terry .15 .25 Legend: * = relationship significant at .05 level; T = direction Opposite that which was predicted. 129 replicating group when client-percept factor scores were compared with their respective total factor score. (See Table 4.17.) Therefore, the null hypothesis was rejected. Potency Factor H03 3: No relationship will be found: ' a. when each client-percept potency factor score is compared with every other client-percept potency factor score ("l-7" procedure); b. when each client-percept potency factor score is compared with the total potency score ("l-7" procedure). H3 3: A positive relationship will be found: ' a. when each client-percept emotional factor score is compared with every other client- percept emotional factor score ("l-7" procedure); b. when each client-percept potency factor score is compared with the total potency score ("l-7" procedure). NO significant relationships were found for any of the ten client-percept comparisons for the experimental group.' The replicating group found two significant relation- ships, but, of course, none of them were substantiated by the experimental group. (See Table 4.18.) The null hypothesis, therefore, was not rejected. Seven of the ten comparisons for the experimental group, however, were in the predicted direction. But only two comparisons were confirmed by the replicating group. Three of five possible relationships were-confirmed significant by the two groups when clienttpercept factor scores were compared with totaleact r scores. :(See Table 4.19.) Both groups found four significant relationships, 130 TABLE 4.17--Re1ationships between client-percept factor score and total factor score using the rank-difference method (rho), for each group: Emotional factor. . Grou I Grou II Client-Percept Experimgntal _ Replicgting Peggy -56** '54** Bill .54** ~59** Bob .65** .56** Lynn ,50** .48** Terry ~65** ’69** Legend: ** = relationship significant at .01 level. TABLE 4.18--Client-percept factor score comparisons by the rank-difference method (rho), for both groups: Potency factor. _ Group I Grou II c11ent-Percept Experimental Replicgting Peggy-Bill -16 '48** Peggy-Bob -25 ‘51** Peggy-Lynn -.27+ "34+ Peggy-Terry -.12+ ‘14 Bill-Bob -14 -31 Bill-Lynn -05 -15 Bill-Terry .04 ‘-03* Bob-Lynn . e .08 --19+ Bob-Terry '-121 '09 Lynn-Terry -24 "07+ Legend: ** = relationship significant at .01 level; T = direction Opposite that which was predicted. TABLE 4.19--Relationships between client-percept factor scores and total factor score using the rank-difference method (rho), for each group: Potency factor. . Grou I Group II Client-Percept Experimgntal Replicating Peggy -47** '62** Bill -59** '69** B013 .31 .59** Lynn ~42** '33 Terry -38** ‘41* Legend: ** relationship significant at .01 level; it = relationship significant at .05 level. 131 but each group found one relationship not confirmed by the other. -The experimental group, for instance, found a significant relationship with Client-percept Lynn, but the replicating group did not. The Opposite occurred with- Client-percept Bob. Evaluative-Activity Factor “03.4: NO relationship-will be found: a. when each clienttpercept evaluative- activitytfactorpscoreeis.comparediwith every other clienttpercept evaluative- activity factor.score ("l-7" procedure); b. when each.elient-percept evaluative- activity factor score is compared with the total evaluative-activity score ("l-7" procedure). H3 4: A positive relationship will be found: ° a. when each client-percept evaluative- activity factor score is compared with every other client-percept evaluative- activity factor score ("l-7" procedure);. b. when each client-percept evaluative- activity factor score is compared with the total evaluative- -activity score ("1 -7" procedure). A significant relationship was found when.Client- percept Lynn was compared with Clienttpercept Terry for the‘ experimental group and the relationship was confirmed by the replicating group. (See Table 4.20.):“Each group had two other but unconfirmed significant relationships.‘ For the replicating group, Client-percepts Peggy and Bill were significantly related, but in a direction opposite that which was predicted. Because only.one of ten comparisons was significant‘and confirmed, the null hypothesis was not rejected. 132 TABLE 4.20--Client-percept factor score comparisons by the rank-difference method (rho), for each group: Evaluative-activity factor. Client-Percept Gr9UP I Group II Experimental Replicating Peggy-Bill -.05 -.47" Peggy-Bob .43* -.05+ Peggy-Lynn .32 -.05+ Peggy-Terry .42* -,06+ Bill—Bob .16 .52** Bill-Lynn .08 -.05+ Bill-Terry .00 -.07+ Bob-Lynn .36 -.17+ Bob-Terry .25 -.05+ Lynn-Terry .66** .42* relationship significant at .01 level; relationship significant at .05 level; relationship significant at .01 level but in a direction opposite that which was predicted; + = direction opposite that which was predicted. Legend: ** * 133 Nine of the ten.re1atienships were in the predicted direction for the experimental_group. But only two of these were confirmed by the replicating group. Client-percept factor scores-were compared with the total factor score and each of the five relationships were found to be significant for the experimental group. ‘(See Table 4.21.) Though four.relatienships were confirmed by the replicating group, the relationship between Client- percept Peggy factor score and the total factor score was not. 'In summary, when client-percept factor scores were compared*with each other, only five of forty relationships were found to be significant, and only one of the five was confirmed by the replicating.group. 'The replicating group found nine of forty relationships significant. But thirty-five of forty.relationships were in.the predicted direction for the experimental group with.twenty-three Of the thirty-five positive directions confirmed by the replicating group. ,.The relationships between client-percept factor scores and total factor scores were found to be significant seventeen of the possible twenty times (S'clientepercept items times 4 factors) by the experimental group and all seventeen were confirmed by the replicating group. 134 TABLE 4.21--Re1ationships between client-percept factor scores and total factor score using the rank-difference method (Egg), for each group: Evaluative-activity factor. Client-Percept Expgiimgnial - Regiiggtiig Peggy .64** .08 Bill .37* .52** Bob .67** .67** Lynn .77** .40* Terry .76** .53** relationship significant at .01 level; Legend: ** * relationship significant at .05 level. 13S DISCUSSION When PPT items were compared.with.their.respective total factor score, forty-nine of the 145 items correlated significantly for.both the experimental andrthe replicating groups. Two factors, the evaluative factor and the emotional factor, contributed forty Of the forty-nine itemsi(twentye fouriof fifty for the evaluative factor and sixteen of“ thirty for the emotional factor). .Though.the-item5' contributed by the evaluative-activity and the potency factors were disappointingly few, there may be some possible explanations for their lack of item correlations. To be considered significantly related to their total factor score, items had to be found significantly related by pppp validating groups. As discussed in Chapter III, the evaluative-activity factor was only quasi-stable at best and, in fact, was not validated by Group‘II at all.' There- fore; the low number of items contributed was not tOO' surprising since the factor was an artificial one forced upon the data in an attempt to account for all-the scales-- The artificiality of the factor could also explain why the scale slow-fast was not significantlyrelated with the total factor Score.irrespective of.client-percept*pairing. Although factor artificiality may be claimed for the- evaluative-activity factor, no such claimican be advanced for the potency factor. Both groups revealed a potency’ factor through factor analysis. .Yet no potency items 136 correlated significantly with the total factor score consistently for both the experimental and replicating groups. 'Perhaps, of all the judgmental.dimensions, potency is the least generalizable across.person percepts and, therefore, the most specialized. Markedly different results were Obtained,.however, when items were compared with their respective.clienta percept factor scores. Only ten of the 145 relationships studied were not significant for both groups.. Of course, there always exists the danger of having.spuriously high relationships when items are comparedei h scor 5 because” the_items make up part of the score;-and one-would expect higher correlations when items.are-compared with client- percept factor scores than.when items arecompared with totalifactor scores because each item made more of a contribution to the client-percept.factor score than it did to the total factor score. For example, an evaluative itemtwas only one of fifty items that.comprised the total' factor score but was one of ten items that comprised-the client-percept factor score. Nevertheless, the data d0“ indicate that when compared with clienttpercept factor scores the item judgments made by the counselor-trainees- were internally consistent despite.limited internal.consist- ency when the total factor score was the.criterion.' The dangers of spurious relationships, however, were even more present when client-percept factor scores instead 137 of items were compared with total factor scores since the client-percept factor score had a much larger influence on the total factor score than did items. Anastasi called the relationship between test part scores and total scores the degree of test homogeniety.2 But naive psychology suggests that a striking parallel exists (1) between a factor item/client-percept factor score comparison and the real world where a perceiver's total judgment of a person percept is composed of his particular judgments of that person percept and (2) between the client- percept factor score/total factor score comparisons and the real world where a perceiver's comparative generalized meaning for person percepts is composed of his comparative meanings for single person percepts. Perhaps what was indicated by the data was that if a perceiver's particular value judgment (for example) of a given person percept were known, the perceiver's total value judgment of that person percept could be predicted. But hiS‘ comparative total value judgment of people in general could not. On the other hand, the data that compared client- percept factor scores with client-percept factor scores and client-percept factor scores with total factor scores suggested that if a perceiver's comparative total value judgment of a person percept were known, the perceiver's 2Anne Anastasi, Psychological Testing (New York: The Macmillan Company, 1954), pp. 9 -101. 138 comparative-total value judgment Of person percepts in general could be predicted, even though his comparative total value judgment of another person percept could not. At this level, at least, the scores on the PPT ("l-7" procedure) have referents in the real world and their relationships as discussed in this chapter lend support for person perception theory. .The semantic differential criterion that client- percepts should vary in meaning one from the other was apparently met since significant differences were found among the five client-percepts in each of the four PPT Dimensions. Moreover, the experimental group and the replicating group agreed on the significance of twenty- seven of the forty differences between.specific.client- percepts. Only the results obtained for the potency factor were surprising in light of "eye-ball" inspection of the data. >Little_practica1 differences seemed to appear among the judgments made for each of the client-percepts (less than one scale unit separating most potent from least potent). Evidently the differences had little to do with mean differences but rather with counselor-trainee variability in judgments. SUMMARY, Table 4.22 which follows contains.a summary of the PersonPerception.hypotheses.including the statistics used and the results found. 139 TABLE 4.22--Summary of results of Person Perception Test hypotheses tested. Hypothesis Results “1.1: A positive relationship will be found between: a. Evaluative factor items and the total evaluative factor score ("l-7" procedure) b.. Evaluative factor items and client- percept factor scores ("l-7" procedure). A positive relationship will be found between: a. Emotional factor items and the total emotional factor score ("l-7" procedure) b. Emotional factor items and client-, percept factor scores ("l-7" procedure). A positive relationship will be found between: a. Potency factor items and the total.po- tency factor score ("l-7" procedure) b. Potency factor items and client-percept factor scores ("l-7" procedure). A positive relationship was found and confirmed for 24 Of 50 items using the product-moment (p). A positive relationship was found and.confirmed for 48 of 50 items using the product-moment (p). A positive relationship was found and confirmed for 16 of 30 items using the product-moment (p). A positive relationship: was found and confirmed for 26 of 30 items using the product-moment (p). A positive relationship was found and confirmed for none of the 20 items using the product-moment (5). A positive relationship was found and confirmed for 17 of 20 items using the product-moment (I): 140 TABLE 4.22--(continued) ”1.4: 2.3‘ 2.4‘ A positive relationship will be found between: a. Evaluative-activity factor items and the total evaluative- activity factor score ("l-7" procedure) b. Evaluative-activity factor items and client-percept factor scores ("l-7" procedure). A significant difference will be found among the five client-percepts when evaluative scores ("l-7" procedure) are used as the criterion. A significant difference will be found among the five client-percepts when emotional scores ("l-7" procedure) are used as the criterion.' A significant difference will be found among the five client-percepts when potency scores ("l-7" procedure) are used as the criterion. A significant difference will be found among the five client-percepts when evaluative-activit scores ("l-7" procedurei are used as the criterion. A positive relationship was found and confirmed for 9 of 45 items using the product-moment (p). A positive relationship was found and confirmed. for 41 of 45 items using the product-moment (p). A difference was found among the.c1ient-percepts at .01 by analysis of variance.- Differences were found. between 4 of 10 client- percept comparisons by _t_ test. . A difference was found among the client-percepts at .01 by analysis of variance.. Differences were found between 8 of 10 client- percept comparisons by .2 test. . A difference was found among the client-percepts at .01 by analysis of variance. Differences were found between 7 Of 10 client- percept comparisons by t test. A difference was found .among the client-percepts at .01 by analysis of variance. Differences- were found between 8 of 10 client-percept comparisons by 3 test. 141 TABLE 4.22--(continued) “3.1‘ 3.2‘ 3.3‘ A positive relationship will be found: 8. When each client- percept evaluative factor score is compared with every other client- percept evaluative factor score ("l-7" procedure) When each client- percept evaluative factor score is compared with the total evaluative score ("l-7" procedure). A positive relationship will be found: a. When each client- percept emotional factor score is compared with every other client-percept emotional factor score ("l-7" procedure) When each client- percept emotional factor score is compared with the total emotional score ("l-7" ‘procedure). A positive relationship will be found: a. When each client- percept potency factor score is compared with every other client- percept potency factor score ("l-7" procedure) None of the ten client- percept comparisons were found significant by rank- difference method. , 9 of 10 relationships were in the predicted direction. A positive relationship. was found and confirmed~ in'S of 5 comparisons by rank-difference method. None of the ten client-‘ percept.comparisonsrwere found significant by rank- difference method. '. 8 of 10 relationships were in the predicted direction. A positive relationship was found and confirmed in 5 of S comparisons by rank-difference method. None of the.ten client-‘ percept comparisons were found significant by rank- difference method. ' 2 Of 10 relationships were in the predicted direction. 142 TABLE 4.22--(continued) b. When each client- percept potency factor score is compared with the total potency score ("l-7" procedure). H : A positive relationship 3.4 a. will be found: When each client- percept evaluative- activity factor score is compared with every other client-percept evaluative- activity factor score ("l-7" procedure) When each client- percept evaluative- activity factor score is compared with the total evaluative- activity score ("l-7” procedure). A positive relationship was found and confirmed in 3 of 5 comparisons by rank-difference method. 1 of 10 client-percept comparisons was found to be significant by rank- difference-method. 2 of 10 relationships were in the predicted direction. A positive relationship was found and confirmed in 4 of 5 comparisons by rank-difference method. CHAPTER V ANALYSIS OF DATA: VALIDITY HYPOTHESES Chapter III hypothesizedrelationships were grouped under two headings; (1) Person Perception Test hypotheses‘ and (2) validity hypotheses. Person Perception Test data were analyzed in the preceding chapter and included, hypotheses for internal consistency,differences among client-percepts, and counselor-trainee "generalized meaning." This chapter will analyze the data generated by the validity hypotheses. Relationships between various PPT scores and counselor-trainee interview responses as recorded on the Counselor Response System will be reported. The chapter discusses: (l) approach-avoidance hypotheses-- evaluative factor; (2) approach-avoidance hypotheses-- interrelated factors; (3) scale-checking style hypotheses; (4) rigidity and stereotOpy hypotheses; (S) exploratory analyses; (6) discussion of results; and (7) summary. The two counselor-trainee groups-comprisedvthe experimental (Group I) and replicating (Group“II).groups. A relationship found to be significant at’ortbeyondithe .05 level for both groups was accepted as evidence for rejecting the null hypothesis. 143 144 APPROACH-AVOIDANCE: EVALUATIVE FACTOR Because Osgood linked the evaluative factor with a measure of attitude, it was hypothesized that.a.significant positive relationship would be found between the generalized evaluative meaning counselor-trainees had for the filmed- client-percepts (i.e. total factor score using the."1-7" procedure) and selected counseling behavior displayed with a particular client as recorded on the Counselor Response System. Specifically, the following hypotheses were tested: H 4.3: 04.1‘ NO relationship will be found'between counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Content: Follow responses recorded on the CRS. A positive relationship will be found between counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Content: Follow responses recorded on the CRS. * No relationship will be found between counselor-traineeitotal evaluative.scores on the PPT ("l-7" procedure) and the frequency of Expansive responses recorded on the CRS. A positive relationship will be found between counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and-the frequency of Expansive responses recorded on the CRS. No relationship will be found between counselor-trainee total evaluative scores on’ the PPT ("l-7" procedure) and the frequency of Referrent: Client responses recorded on the CRS. ‘ A positive relationship will be found between counselor-trainee total evaluative scores on"' the PPT ("l-7" procedure) and the frequency of Referrent: Client responses recorded on the CRS. 145 H04 4: NO relationship will be found between ° counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Reinforcing responses recorded on the CRS. H : A positive relationship will be found between 4'4 counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency,of Reinforcing responses recorded on the CRS. The Spearman rank-differenceprocedure was used and coefficients were computed to test the hypothesized relation- ships. The resultant correlation coefficients (rho), using the "1-7" procedure are presented in Table 5.1. No significant correlations were found for either the experimental or the replicating group for any of the four hypothesized relationships. Therefore, null hypotheses 04.1, 04.2, 04.3, 04.4 were not rejected. APPROACH-AVOIDANCE: INTERRELATED FACTORS In Chapter II it was reported that the evaluative factor alone was not enough for predicting behavior--that the addition of other semantic dimensions could improve predictability. Therefore, the following hypotheses were tested: H0501: No relationship will be found between counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Content: Follow responses recorded on the CRS. H : A positive relationship will be found between 5’1 counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Content: Follow responses recorded on the CRS. 146 TABLE 5.l--Correlation coefficients for hypothesized relationships between Counselor Response System Scores and Person Perception Test evaluative scores using the "1-7" procedure. — #— Counselor Response System Person Perception T35t Group I GrOUp 11 Content: Follow -.12 -.04 Control: Expansive -.02 .11 Referrent: Client -.17 .07 Reinforcing -.08 -.08 147 H05 2: NO relationship will be found between ' counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Control: Expansive responses recorded on the CRS. HS 2: A positive relationship will be found between ' counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Control: Expansive responses recorded on the CRS. H05 3: NO relationship will be found between ° counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency Of Referrent: Client responses recorded on the CRS; H5 3: A positive relationship will be found between ' counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Referrent: Client responses recorded on the CRS. H05 4: No relationship will be found between ‘ counselor-trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Reinforcing responses recorded on the CRS. HS 4: A positive relationship will be found between ° counselor-trainee:interrelated PPT total factor scores ("l-7" procedure) and the frequency of Reinforcing responses recorded on the CRS. A multiple correlation analysis (R) was computed to test each hypothesis. The resultant correlation coeffi- cients, using the "1-7" scoring procedure are presented in Table 5.2. Neither the experimental nor.the replicating group found significant relationships between the PPT and the four CRS dimensions. Therefore.null hypotheses 05.1, 05.2, 05.3, and 05.4 were not rejected. 148 TABLE 5.2--Mu1tiple correlation coefficients between Counselor Response System scores and Person Perception Test scores using the "1-7" scoring procedure.. Person Perception Test Counselor Response System Group I Group II Content: Follow .29 .40 Control: Expansive .30 .39 Referrent: Client .39 .41 Reinforcing .42 .49 149 SCALE-CHECKING STYLE In Chapter II it was reported that subjects' emotionality toward semantic differential concepts affected which scale positions were checked. ‘Osgood suggested three judgmental levels: (1) a dogmatic choice (positions 1 and 7), (2) an avoidance choice (scale position 4), and (3) a discriminatory choice-(scale positions 2,3,5,6). Based upon the findings of these studies the following scale-checking style hypotheses were tested using the frequency scoring procedure: ”06.1‘ “06.2‘ “6.2‘ “06.3: No relationship will be found between counselor-trainee interrelated PPT total factor scores (£2356 procedure) and the frequency of Content: Follow responses recorded on the CRS. A positive relationship will be found between. counselor-trainee interrelated PPT total factor scores (£2356 procedure) and the frequency of Content: Follow responses recorded on the CRS. No relationship will be found between counselor-trainee interrelated PPT total factor scores.(f235 procedure) and the frequency of Control: Expansive responses recorded on the CRS. A positive relationship will be found between counselor-trainee‘interrelated.PPT total factor scores(f235 procedure) and the frequency of Control: Expansive responses recorded on the CRS. No relationship will be found between counselor-trainee interrelated PPT total factor scores (£2356 procedure) and the frequency of Referrent: Client responses recorded on the CRS. 150 H6 3: A positive relationship will be found between ° counselor-trainee interrelated PPT total factor scores (£2555 procedure) and the frequency of Referrent: Client responses recorded on the CRS.‘ “06.4: No relationship will be found between counselor-trainee interrelated PPT total factor scores (£2355 procedure) and the frequency of Reinforcing responses recorded on the CRS. 6 4: A positive.relationship will be found between ' counselor-trainee interrelated.PPT total factor scores (£2356 procedure) and the frequency of Reinforcing responses recorded on the CRS. . A multiple correlation analysis (R) was computed to test each hypothesis. The resultant correlation coeffi- cients, using the £2356 scoring procedure are presented in Table 5.3.. No significant relationships were found by either the experimental or the replicating groups.‘ So the null hypotheses 06.1, 06.2, 06.3, and 06.4 were not rejected. STEREOTOPY AND RIGIDITY. In Chapter III the use of available semantic space was linked with the extent a counselor-trainee was open to stimuli.. The less use of the semantic space, the more rigid the respondent and the more stereotyped his reactions. Variance factor scores were computed to signify use of the semantic space and the following hypotheses were tested: H0701: No relationship will be found between . counselor-trainee interrelated PPT variance factor scores and the frequency of Content: Follow responses recorded on the CRS. 151 TABLE 5.3--Multiple correlation coefficients between Counselor Response System scores and Person Perception Test scores using the f2356 scoring procedure. Person Perception Test Counselor Response System GTOUp I Group II Content:, Follow .36 .28 Control: Expansive .30 .34 Referrent: Client .46 .24 Reinforcing .43 .27 152 H : A positive relationship will be found between 7.1 . . . . counselor-trainee interrelated PPT variance factor scores and the frequency of Content: Follow responses recorded on the CRS. H : NO relationship will be fOund between 0702 a o o counselor-trainee interrelated PPT variance factor scores and the frequency of Control: Expansive responses recorded on the CRS. H7 2: A positive relationship will be found between ° counselor-trainee interrelated PPT variance factor scores and the frequency of Control: Expansive responses recorded on the CRS. H : No relationship will be found between 07.3 . . . counselor-trainee interrelated PPT variance factor scores and the frequency of Referrent: Client responses recorded on the CRS. H7 3: A positive relationship will be found between ° counselor-trainee interrelated PPT variance factor scores and the frequency of Referrent: Client responses recorded on the CRS. H07 4 No relationship will be found between ' counselor-trainee interrelated PPT variance factor scores and the frequency of Rein- forcing responses recorded on the CRS. H7.4: A positive relationship will be found between counselor-trainee interrelated PPT variance factor scores and the frequency of Reinforcing responses recorded on the CRS. A multiple correlation analysis (R) was computed to test each hypothesis. The resultant correlation coeffi- cients, using the variance scoring procedure, are presented in Table 5.4. Again, no significant relationships were found by either the experimental or the replicating group. The null hypotheses 07.1, 07.2, 07.3, and 07.4 were not rejected. 153 TABLE 5.4--Multiple correlation coefficients between Counselor Response System scores and Person Perception Test scores using the variance scoring procedure. Person Perception Test Counselor Response System Group I Group II Content: Follow .19 .36 Control: Expansive .29 .49 Referrent: Client .27 .42 Reinforcing .17 .55 154 EXPLORATORY ANALYSES Several non-hypothesized relationships were explored and are presented briefly here: 1. Using the "1-7” procedure, the PPT evaluative- activity, emotional, and potency total factor scores were compared with each of the four CRS dimensions by the product- moment (3). No relationships were found significant for either the experimental or the replicating group. (See Appendix J for the complete data.) 2. Using the £2356 scoring procedure, each of the four PPT total factor scores was compared with each of the four CRS dimensions by the product-moment (E). No relation- ships were found significant for either the experimental or the replicating group. (See Appendix K for the complete presentation of data.) 3. Using the f1,7 scoring procedure, each of the four PPT total factor scores was compared with each CRS dimension by the product-moment (p). Also, interrelated total factor scores were compared with each of the four CRS dimensions by the multiple correlation (R). (See Table 5.5.) The experimental group found a significant negative relation- ship between the evaluative-activity factor score and the frequency of Control: Expansive responses. But the relationship was not confirmed by the replicating group. On the other hand, the replicating group found a significant positive relationship (not confirmed) between the potency 155 TABLE 5.5--Person Perception Test total factor scores (f 7 procedure) compared with Counselor Response System dimensions by product-moment (p) and by multiple correla- tion (R). Counselor Response System Content: Follow Control: Expansive Referrent: Client Reinforcing Content: Follow Control: Expansive Referrent: Client Reinforcing Content: Follow Control: Expansive Referrent: Client Reinforcing Content: Follow Control: Expansive Referrent: Client Reinforcing Content: Follow Control: Expansive Referrent: Client Reinforcing Person Perception Test Group I Group II Evaluative Factor -.17 -.03 -.19 -.06 -.19 -.08 .15 .10 Emotional Factor .06 .20 -.15 -.03 -.06 .10 .08 .20 Potency Factor -.09 .23 -.29 .23 -.26 .10 -.04 .39* Evaluative-Activity Factor -.16 .24 -.41* .01 -.26 .13 .04 .06 Interrelated Factors .39 .42 .53 .35 .43 .29 .25 .69** Legend:. ** is significant at .01 level; significant at .05 level. 156 factor score and the frequency of reinforcing responses and between the PPT interrelated factor scores and-the frequency of reinforcing responses. 4. Using the'f4 procedure, each of the four PPT total factor scores was compared with each of the four CRS dimensions by the product-moment (I)° Moreoever, interrelated total factor scores.were compared with each of the four CRS dimensions by.multiple correlation (R). (See Appendix L for complete.presentation of the data.) No significant relationships were found by the experimental group, although the replicating group found a significant negative relationship between the evaluative-activity dimension and the frequency of reinforcing responses. 5. Using the variance scoring procedure, each of the four PPT total factor scores was compared with each of the four CRS dimensions by the product-moment (3). Though the experimental group found no significant relationships, the replicating group found significant positive relationships between the frequency of reinforcing responses and both the evaluative and evaluative-activity total factor scores. (See Table 5.6.) 6. Each of the PPT items was compared with each Of the CRS dimensions by the product-moment (p). 'The experi- mental group found thirty-seven significant relationships of a possible 600 and the replicating group found thirtye two. Two significant relationships were confirmed: When 157 TABLE 5.6--Person Perception Test factor variance scores compared with Counselor Response Systems dimensions by product-moment (p). Counselor Response System Person Perception TBSP Group I Group II Evaluative Factor Content: Follow , .ll .25 Control: Expansive -.15 .27. Referrent: Client -.07. .25 Reinforcing -.08 .38* Emotional Factor Content: Follow -.13 .05_ Control: Expansive -.16 .03 Referrent: Client -.18 .15 Reinforcing .15 .08 Potency Factor Content: Follow -.05 9.30 Control: Expansive -.12 .28 Referrent: Client -.22 .35 Reinforcing .01 .17 Evaluative-Activity Factor Content: Follow .00 .04 Control: Expansive -.24 -.15 Referrent: Client . -.12, -.03 Reinforcing -.06 .50** Legend: ** significant at .01 level; it significant at .05 level. 158 paired with Client-percept.Terry,:scales simple-complex and weak-strong (both potency scales) were negatively related with the Control: Expansive dimension. (See Appendix M for a complete presentation of the data.) DISCUSSION No significant predicted relationships were found by either the experimental or the replicating group. Moreover, over half the coefficients computed, although non-significant, were in the Opposite direction from that which was predicted. In the exploratory phase Of this study a significant inverse relationship was foundby the experimental group between total evaluative-activity_scores‘(f1,7 procedure) and the frequency of Control:-'ExpanSive responses. Though it was in the predicted direction the relationship was not confirmed by the replicating group. The replicating group found significant relationships between the reinforcing dimension and: 1. interrelated factor_scores (f1,7 procedure); 2. total potency scores (f1,7 procedure--relation- ship positive); 3. total evaluative-activity (f4 procedure-- relationship negative); 4. evaluative variance scores (relationship positive); 159 5. evaluative-activity variance scores (relationship positive). But each of these relationships was unconfirmed. It could be possible that the results of the validity hypotheses represent an accurate picture of real-li£e-- that there is, in fact, no relationship between the generalized meaning counselor-trainees have.for clients and the behavior counselor-trainees display with specific clients. If this were true, then the theory upon which much of this study was based is inaccurate.. But it is-premature to conclude that the theory does nOt represent the real world. Or it may well be that the relationship between generalized meanings and behavior exists but that it is not linear as was assumed with the statiStics used in this study. In fact,.future studies of the relationship might find it to be circular in nature much-as Schlosberg did with his sortings of the Frois-Wittman photos. .Since the PPT was developed especially for this study and because the primary interest of this investigator is with the further-development of the instrument, the results of the validity hypotheses point to a need for‘ further refinement of the instrument (before discarding it or the theory upon which it was based) in an attempt to bring.out possible relationships. “Refinement of the PPT will be the major consideration in the discussion of future research in Chapter VI.. 160 Besides the need for PPT refinement, several other“ possibilities could explain the validity hypotheses results and these possibilities should be kept in mind and, controlled for during further.deve10pment of the PPT. Future administrations Of the PPT would be needed to offset the possibility that the two groups used in this study might have been uniquely homogeneous and;thattno "real" difference should have been expected.. However, there is no reason to believe that-either group was any more‘ homogeneous than any other group of counselor-trainees that have been or might be studied.. Or perhaps the-PPT was not administered in a standard manner to both groups. 'Besides the directions on the test booklet itself which were‘read by the test administrator, no additional standard testing procedures were written for the administrator to follow.x But there*is no evidence t0"‘ support the notion that the two groups-were administered the: PPT such that the test results were differentially affected. Another possibility is that perhaps the counselor- trainees responded on the PPT as they thought they should rather than as theyhonestly felt. In other words, the PPTsmight be perfectly valid if only the.suhjects would answer honestly. Of course, this argument.can be used on any test in which there are no wrong or right answers. The researcher must rely on faith that his subjects were honest. 161 It could also be.arguedrthatrthegstatistics.used to test1thexhypotheses wereiinappropriate for.thedata and that a different statistic mightrhaveTyielded-significant“r results.. However, this argument seems weak in lightiof' Garrett's suggestion that the Spearman rank-difference method be.used when the number of subjects is small.1 Garrett stated that pppfyieldscasradequatearesults'as Pearsonistproduct-momentpercept factor for 48 of 50 items using scores ("1-7" the product-moment (I). procedure). H1.2= 1.3‘ 172 A positive relationship will be found between; a. Emotional factor items and the total emotional factor score ("l-7" procedure) b. Emotional factor items and client- percept factor scores ("l-7" procedure). A positive relationship will be found between: a. Potency factor items and the total potency factor score ("l-7" procedure) b. Potency factor items and client-percept factor scores ("1~7" procedure). A positive relationship will be found between: a. Evaluative-activity factor items and the total evaluative- activity factor score ("1-7" procedure) b. Evaluativefiactivity factor items and client~percept factor scores ("l~7" procedure). A significant difference will be found among the five client~percepts when evaluative scores ("1~7" procedure) are used as the criterion. A positive relationship was found and confirmed for 16 of 30 items using the product-moment (r). A positive relationship was found and confirmed for 26 of 30 items using the product-moment (1). A positive relationship was found and confirmed for none of the 20 items using the product-moment (1). A positive relationship was found and confirmed for 17 of 20 items using the productamoment (r). A positive relationship was found and confirmed for 9 of 45 items using the product-moment (r). A positive relationship was found and confirmed for 41 of 45 items using the product-moment (5). A difference was found among the client-percepts at .01 by analysis of variance. Differences were found between 4 of 10 client-percept comparisons by t test. 2.3: 173 A significant difference will be found among the five c1ient~percepts when emotional scores ("1-7" procedure) are used as the criterion. A significant difference will be found among the five clientwpercepts when potency scores (”l-7" procedure) are used as the criterion. A significant difference will be found among the five clientwpercepts when evaluative-activity scores ("1~7" procedure) are used as the criterion. A positive relationship will be found: a. When each client» percept evaluative factor score is compared with every other clients percept evaluative factor score ("l~7" procedure) b. When each client» percept evaluative factor score is compared with the total evaluative score ("l~?" procedure). A difference was found among the client-percepts at .01 by analysis of variance. Differences were found between 8 of 10 client- percept comparisons by E test. A difference was found among the client-percepts at .01 by analysis of variance. Differences were found between 7 of 10 cliente percept comparisons by t test. A difference was found* among the client-percepts at .01 by analysis of variance. Differences were found between 8 of 10 client- percept comparisons by t test. None of the ten clienta percept comparisons were found significant by rank» difference method. 9 of 10 relationships were in the predicted direction. A positive relationship was found and confirmed in 5 of 5 comparisons by rank-difference method. 3.3‘ 174 A positive relationship will be found: a. When each client- percept emotional factor score is compared with every other client- percept emotional factor score ("1-7" procedure) b. When each client- percept emotional factor score is compared with the total emotional score ("l-7" procedure). A positive relationship will be found: a. When each client- percept potency factor score is compared with every other client- percept potency factor score ("l-7 procedure) b. When each client- percept potency factor score is compared with the total potency score ("1-7" procedure). A positive relationship will be found: a. When each client- percept evaluative- activity factor score is compared with every other client- percept evaluative- activity factor score ("l-7" procedure) None of the ten client- percept comparisons were found significant by rank- difference method. 8 of 10 relationships were in the predicted direction. A positive relationship was found and confirmed in 5 of S comparisons by rank-difference method. None of the ten client- percept comparisons were found significant by rank- difference method. 2 of 10 relationships were in the predicted direction. A positive relationship was found and confirmed in 3 of 5 comparisons by rank-difference method. 1 of 10 client-percept comparisons was found to be significant by rank- difference method. 2 of 10 relationships were in the predicted direction. 4.4‘ 175 b. When each client~ percept evaluativee activity factor score is compared with the total evaluative- activity score ("1-7" procedure). A positive relationship will be found between counselor~trainee total evaluative scores on the PPT (”l-7" procedure) and the frequency of Content: Follow responses recorded on the CRS. A positive relationship will be found between counselor-trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Control: Expansive responses recorded on the CRS. A positive relationship will be found between counselor'trainee total evaluative scores on the PPT ("l-7" procedure) and the frequency of Referrent: Client responses recorded on the CRS. A positive relationship will be found between counselor~trainee total evaluative scores on the PPT ("1=7” procedure) and the frequency of Reinforc= ing responses recorded on the CRS. A positive relationship will be found between coun~ selor-trainee interrelated PPT total factor scores ("l-7” procedure) and the frequency of Content: Follow responses recorded on the CRS. A positive relationship was found and confirmed in 4 of S comparisons by rankwdifference method. The null hypothesis was not rejected using the rank-difference method. The null hypothesis was not rejected using the rank-difference method. The null hypothesis was not rejected using the rank-difference method. The null hypothesis was not rejected using the rank-difference method. The null hypothesis was not rejected using multiple correlation (R). 5.2‘ 176 A positive relationship will be found between counselor- trainee interrelated PPT total factor scores ("l-7" procedure) and the frequency of Control: Expansive responses recorded on the CRS. A positive relationship will be found between counselor- trainee interrelated PPT‘ total factor scores ("l-7" procedure) and the frequency of Referrent: Client responses recorded on the CRS. A positive relationship will be found between counselor» trainee interrelated PPT' total factor scores ("1~7" procedure) and the frequency of Reinforcing responses recorded on the CRS. A positive relationship will be found between counselor- trainee interrelated PPT' total factor scores (f 6 procedure) and the frequency of Content: Follow responses recorded on the CRS. A positive relationship will be found between counselor- trainee interrelated PPT , total factor scores (£2356 procedure) and the frequency of Control: Expansive responses recorded on the CRS. A positive relationship will be found between counselor- trainee interrelated PPT‘ total factor scores (£2356 procedure) and the frequency of Referrent: Client responses recorded on the CRS. The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). 7.1‘ 7.2‘ 177 A positive relationship will be found between counselor» trainee interrelated PPT total factor scores (£2356 procedure) and the frequency of Reinforcing responses recorded on the CRS. A positive relationship will be found between counselor» trainee interrelated PPT variance scores and the frequency of Content: Follow responses recorded on the CRS. A positive relationship will be found between counselor~ trainee interrelated PPT variance scores and the frequency of Control: Expansive responses recorded on the CRS. A positive relationship will be found between counselor» trainee interrelated PPT variance scores and the frequency of Referrent: Client responses recorded on the CRS. A positive relationship will be found between counselor» trainee interrelated PPT variance scores and the frequency of Reinforcing responses recorded on the CRS. The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). The null hypothesis was not rejected using multiple correlation (R). Several exploratory analyses of the data were made but no consistent relationships were found. obtained during the development of the Person Perception Test CONCLUSIONS The first three conclusions are based upon information as described in Chapter III: 178 l. Counselor-trainees judge clients along several dimensions with an evaluative dimension being the most dominant, followed by emotional and potency dimensions. 2. Counselor-trainees value more highly those clients who are most active as evidenced by the dependence of the activity factor on the evaluative factor for Group I and by its coalescence with the evaluative factor for Group.II.. 3. The PPT is an objective means for tabulating counselor-trainee perceptions of filmed client- percepts as evidenced by the ease with which various scoring procedures can be applied. The PPT data indicate that item judgments made by the counselor-trainees were internally consistent when client-percept factor scores were the criterion but had limited internal consistency when total factor score was the criterion. Test homogeneity was also attested to by the consistently significant relationships found between client- percept factor scores and total factor scores. Naive psychology suggests a striking parallel between PPT scores and possible referrents in the "real world." If an item is considered a particular judgment made by a perceiver of a person percept, if a client-percept factor score is considered the total judgment along a particular dimension made by a perceiver of a person percept, and if the total 179 factor score is considered the judgments made by the fi perceiver of person percepts in general, then several conclusions about the data may be.expressed which seem to fit everyday experience: 4. The PPT total factor scores represent counselor- trainees' generalized meaning for the client- percepts as evidenced by the positive relation- ships found between client—percept factor scores and total factor scores. If a perceiver's comparative particular judgment of a person percept were known, the perceiver's total judgmentof the person percept could be predicted. But his comparative total judgment of people in general could not. If a perceiver's comparative-total judgment of a person percept were known, the perceiver's comparative total judgment of person percepts in general could be predicted. But his comparative total judgment of another person percept could not. The data collected for this study did not support the validity hypotheses and, therefore, did not.support the predicted relationship between counselor-trainees' generalized meaning for filmed client-percepts and their subsequent interview behavior with a specific client. Therefore: 180 7. The PPT needs further refinement (before discards ing it or the theory upon which it was based) in an attempt to bring out better possible relation- ships. PUTURE PPT RESEARCH: A SUGGESTED PROCEDURE The focus of this study was on the development of the Person Perception Test. The primary conclusions suggest that the PPT, in its present state of development, has promise as an instrument for measuring counselor-trainee perception. However, the PPT needs refinement as a predictor of counselor-trainee interview behavior. The following list represents a suggested procedure for continuing research with thelPPT: l. The least expensive and most practical next step in the development of the PPT would be to concentrate on its reliability. Obviously, before any test can be valid it must be reliable. The primary purpose of such studies would be to stabi- lize the factors, particularly the potency and evaluative-activity factors. What the PPT measures, at this point, would not be a concern. The investigator would have to experiment with many different scales in an attempt to obtain a clear definition of the factors present by administering the PPT to many counselor-trainee groups to determine which scales consistently 181 loaded high on which particular factor. And while experimenting with new scales, the investigator would determine which scales "hold up" best from test to test. Once a consistent instrument is developed, one that clearly defines all factors present, the question of validity can be tackled. The criteria used in this studx_was the frequency of certain counselor-trainee responses. as recorded on the Counselor Response System. 'With a more consistent PPT it seems reasonable to ‘suggest replication of this study with a special focus on increasing.CRS dimension variance (perhaps by using coached clients or even actors) and with a prime concern with the relationship's possible non-linear function. If the relation- ships are no better than were reported in this study, then it might be well to count more coun} selor responses or even to use complete interview- tapes; or counselor-trainee responses to several clients might be indicated; or additional coun- selor response dimensions might be explored. If a relationship between the PPT and some counselor-trainee interview behavior can be shown to exist, then it would seem profitable to explore the predictive validity of the PPT with 182 the goal of someday using the instrument as one tool for selecting counselor-trainees. 4. Although expensive, different filmed client- percepts might be tried. Perhaps, more extreme 'personality types might yield a more complete saturation of the semantic space. 5. When the PPT has been refined enough to predict counselor-trainee behavior, a series of causal- comparative studies comparing counselors trainees with other graduate students, with 'their clients, and with their supervisors would be in order. In closing, a goal of this thesis was to study the relationship between person perception and behavior in the context of the counseling activity. The validity results were disappointing, and indicated that the independent. variable needs more study and the dependent variable needs more variance. But the results did indicate that counselor- trainees judged clients along several dimensions, and if these dimensions can be made more stable, perhaps the PPT can yet demonstrate a relationship between person perception and behavior. BI BLIOGRAPHY 183 BIBLIOGRAPHY. Allport, F. H. Theories ofqurception and the Concept of, Structure. New York: John WiIey 6 Sons, Inc., 1955. Allport, G. W. Pattern and Growth in Personalityt. New York: Holt, Rinehart E Winston, 1961. ' ‘ Anastasi, Anne. PsychologiCal Testing. New York:_ The Macmillan Company, 1954. Asch, S. B. 'Social Psychology.' Englewood Cliffs, New Jersey: Prentice-Hal , Inc., 1952. ' ' Barclay, A., and Thumin, F. J. "Modified Semantic Differential Approach to Attitudinal Assessment," Journal of Clinical Psychology, 19:376-378 (1963). Berlo, D. K. The Process of Communication. New York: Holt, Rinehart E Winston, Inc., 1960. Bopp, Joan. "A Quantitative Semantic Analysis of Word Association in Schizophrenia." Unpublished Doctoral dissertation, University of Illinois, 1955. Bordin, Edward S. Psychological Counselin . New York; Appleton-Century-Cro ts, Inc., 195 . l. Borg, W. R. Educational Research.‘ New York: David McKay Company, Inc., 1963. ‘ ' Brown, R. E. "A Use of the Semantichifferential to Study the Feminine Image of Girlstho Participate in Competitive Sports and.Certain Other School-Related Activities," Dissertation Abstracts, 26:4426 (1966). Bruner, J. 5., Goodnow, J., and Austin, G. A Study of Thinking. New York:. John Wiley 8 Sons, Inc., 1965. , and Tagiuri, R. "The Perception of PeOple," in G. Lindzey (ed.) Handbook of Social Psychologx. Vol. II. Cambridge: AddisonsWesley, 1954. 184 185 Brunswik, E., and Reiter, Lotte.: "Eindruckscharakter- schematisierter Gesichter," 2. f. Psychologia, 142:62-134, in Harrison, ”Pictic.Ana y51s: . . ." Chein, I., Levine, R., Murphy, 6., Proshansky, H., and Schafer, R. "Need as a Determinant of Perception, and Reply to Pastore," Journal of Psychology, 31: 129- 136 (1951). Clayson, Merrill D. "Therapeutic Progress in Terms of Semantic Variability," Dissertation Abstracts, 25:623 (1964). Cotter, R. A. "Verbal Learning in the Evaluation of Two Theories of Meaning," Dissertation Abstracts, 26:6835 (1966). Cronbach, L. J. "Proposals Leading to Analytic Treatment of Social Perception Scores," in Tagiuri and Petrullo, Person Perception and Interpersonal Behavior. Stanford: ’Stanford University Press, 1958: Darwin, Charles. The Expression of the Emotions in Man and Animals. ’London, 1872. ‘ Davitz, J. R. (ed. ) The Communication of Emotional Meaning. New York: McGrawsHill, 1964. Derbyshire, Robert L. "United States Negro Identity Con- flict, " Sociology and Social Research, 51: 63- 77 (1967). DeRoo, William M. "A Study of Relationships Between Counselor Personality and Counselor Behavior." Unpublished Doctoral dissertation, Michigan State University, 1965. Diab, L. N. "Studies in Social Attitudes: III: Attitude Assessment Through the Semantic Differential Technique," Journal of Social Psychology, 67: 303- 314 (1965). Doob, L. W. "The Behavior of Attitudes,” Psychological Review, 54: 135-156 (1947). Elliott, L. L., and Tannenbaum, P. H. "Factor~structure of Semantic Differential Responses to Visual Forms and Prediction of FactorsScores from Structual Characteristics of the Stimulus-Shapes," American Journal of Psychology,76:589-597 (1963). 186 Epictetus, The Works of Epictetus. Translated by E. Carter and T. W. Higginson. Boston: Little, Brown, 1865. Fiedler, F. "The Concept of Ideal Therapeutic Relationship," Journal of ConsultingPsychology,.14:242 (1950). Friedman, D. J., Johnson, C. A., and Forde, K. "Subjects' Descriptions ofSelected TAT Cards Via the Semantic Differential," Journal of ConsultingPsychology, 28:317-325 (1964). Fullmer, D. W., and Bernard, H. W. Counseling: Content and Process. Chicago: Science Research Assoc1ates, 1964. Garrett, Henry E., and Woodworth, R. S. Statistics. New York: Longmans, Green and Company, 1958. Glatter, A. N., and Reece, M. M. "Tactility and Sexual Symbolism," Perceptual and Motor Skills, 14:302 M (1962). Gleser, G. C. "Prediction," Encyclopedia of Educational Research (3rd edition), C. W. Harris (ed.) New York: The Macmillan Company, 1960. Goldfried, M. R. "The Connotative Meaningof Some Animal ' Symbols for College Students," Journal of Projective Techniques and Personality Assessment, 27:60-67 (1963). , and Kissel, S. "Age as.a Variable in the Connotative Perceptions of Some Animal Symbols," Journal of Projective Techniques and Personality Aesessment, 27:171-180 (1963). Greenberg, Bradley 8., and Bowes, John. ,"Dimensions of Empathic Judgment of Clients by Counselors," in Kagan and Krathwohl, Studies in Human Interaction, 1967. Hafner, A. Jack, and Rosen, Ephraim. "The Meaning of Rorschach Inkblots, Responses and Determinants as Perceived by Children," Journal of Projective Techniques and Personality Assessment, 287192-200 (1964). Hardich, E. J. "The Self-Concept of Hard-of—hearing Adults as Measured by the.Semantic Differential," Dissertation Abstracts, 25:6826 (1964). 187 Harmon, Donald, and Arnold, Dwight L. "High School Counselors Evaluate Their Formal Preparation," Personnel and Guidance Journal, 39:303 (1959). Harrison, Randall. "Pictic Analysis: Toward a Vocabulary and Syntax for the Pictorial Code, with Research on Facial Communication.". Unpublished Doctoral dissertation, Michigan State University, 1964. Heider, F. The Psycholo of Interpersonal Relations. New York: John Wiiey E Sbns, 1958. Hill, C. W. "The Selection of School Counselors," Personnel and Guidance Journal, 39,357.353 (1959). Hitchcock, William. "Counselors Feel They Should," Personnel and Guidance Journal, 32:72-74 (1953). Hoar, J. R., and Meek, E. E. "The Semantic Differential as a Measure of Subliminal Message Effects," Journal of Psychology, 60:165-169 (1965). Kerrick, Jean. "The Effects of Intelligence and Manifest Anxiety on Attitude Change Through Communications." Unpublished Doctoral dissertation, University of Illinois, 1954. Krathwohl, David. How To Prepare a Research Proposal. Syracuse, New York: Syracuse Univer§1ty, 1966. Lana, R. E., and Pauling, F. J. "Opinion Change When the Semantic Differential is a Pretest," Psychological Reports, 17:730 (1965). Levy, Sheldon G. "Multidimensional Content Analysis: An Extension of the Semantic Differential," Dissertation Abstracts, 25:1321 (1964). Lewis, W. A., and Wigel, W. "Interpersonal Understanding and Assumed Similarity," Personnel and Guidance Journal, 43:155-158 (1964). Lilly, R. S. "A Developmental Study of the Semantic Differential," Dissertation Abstracts, 26:4063- 4064 (1966). ‘ Linden, J. D., Stone, S. C., and Shertzer, B. "The Development and Evaluation of an Inventory for Rating Counseling," Personnel and Guidance Journal, 44:267 (1965). 188 Loiselle, R. H., and Kleinschmidt, A. "A Comparison of the Stimulus Value of Rorschach Inkblots and Their Percepts," Journal of Pro'ective Techniques and Personality Assessment, 27:191-194 (1963). Malmstrom, Edward J., and French, Gilbert M. "Scale- Symmetry and the Semantic Differential," American Journal of Psychology, 76:446-451 (1963). Manis, M. ‘"Assessing Communication with the Semantic Differential," American Journal of Psychology, 72:111-113 (1959). Mehling, Reuben. "A Simple Test for Measuring Intensity of Attitudes," Public Opinion Quarterly, 23:567-568 (1959). Messick, S. J. "Metric Properties of the Semantic Differential," Educational and Psychological Measurement, 17:200-206 (1957). Getting, E. R. "Cross-Cultural Communication and the Semantic Differential," Journal of Counseling Psychology, 11:292-293 (1964). Osgood, C. E. "Dimensionality of the Semantic Space for Communication Via Facial Expressions." Urbana: University of Illinois, N. D. (Mimeographed). . "The Nature and Measurement of Meaning," Psychological Bulletin, 49:197-237 (1952). . Suci, G. J., and Tannenbaum, P. H. The Measurement of Meaning. Urbana: University of I1linois Press, 1957. Otten, M. W., and Van de Castle, R. L. "A Comparison of Set 'A' of the Holtzman Inkblots with the Rorschach by Means of the Semantic Differential," Journal of Projective Techniques and Personality Assessment, 27:452-460 (1963). ' ‘ Pastore, N. "Need as a Determinant of Perception," Journal of Psychology, 28:457—475 (1949). Patterson, C. H. "Selection of Rehabilitation Counseling Students," Personnel and Guidance Journal, 41:318- 320 (1961). llabin, A. I. "A Contribution to the-*Meaning‘of~ Rorschach's Inkblots Via the Semantic Differential," Journal of ConsultingPsychology, 23:368-372 (1959). 189 Rank, R. C. "The Assessment of Counselor-Trainee Perceptions of Interview Protocols Before and After an Intensive Practicum Experience." ~Unpublished Doctoral thesis, University of Minnesota, 1964. ‘ . "Counseling Competence and Perception," *Personnel and Guidance Journal, 45:359-365 (1966). , and DeRoo, William M. "Counselor Response System of the Behavior Interaction Description- System." Michigan State University, 1965 (Mimeographed). Rogers, C. R. On Becomin a Person. Boston: Houghton- Mifflin Company, 1 61. Rosen, E. "A Cross-Cultural Study of Semantic Profiles and Attitude Differences," Journal of Social Psychology, 49:137-144 (1959). Rosenthal, O. A. "A Semantic Differential Investigation of Critical Factors Related to Achievement and Under-achievement of High School Students," Dissertation Abstracts, 26:3156 (1966). Samuels, Myra R., "Judgment of Faces," Character and Personality, 8:18-27 (1940). Santavicca, G. G. "Supervised Experience and Selection of Counselor Trainees," Personnel and Guidance Journal, 38:195-197 (1958). Schlosberg, H. "Three Dimensions of Emotion," Psychological Review, 61:81-88 (1954). Schmidt, Lyle D. "Concepts of the Role of Secondary School Counselors," Personnel and Guidance Journal, 40:602 (1962). Singer, R. D. "A Note on the Use of the Semantic Differen- tial as a Predictive Device in Milieu Therapy," Journal of Clinical Psychology, 17:376-378 (1961). Solomon, L. N. "Semantic Reactions to Systematically Varied Sounds," Journal of the Acoustical Society of America, 31:986-990 (1946).‘ Sproull, Natalie. "Finding and Stating the Research‘" Problem: A Suggested Approach with Examples from Higher Education." Michigan State University, 1967 (Mimeographed). 190 Stoughton, R. W. "The Preparation of Counselors and Personnel Workers," Review of Educational Research, 27:175 (1957). Stricker, George. "Stimulus Properties of the Rorschach‘ to a Sample of PedOphiles," Journal of Projective Techniques and Personality Assessment, 28:241-244 (1964): .. , ”' ' Stromberg, C. E. "Semantic Differentiation-of Social Behavior of Valued Persons by Female College Groups," Dissertation Abstracts, 23:3787 (1962). Suci, G. J. "A Multidimensional Analysis of Social Attitudes with Special Reference to Ethnocentrism." Unpublished Doctoral dissertation, University of Illinois, 1955-. Tagiuri, R., and Petrullo, L. (eds.) Person Perception and Interpersonal Behavior. Stanford: Stanford University Press, 195 . *‘ ‘ ' Triandis, H. C. '"Differential.Perception of Certain JObS' and People by Managers, Clerks, and Workers in Industry," Journal of Applied Psychology, 43: 221- 225 (1959). Van de Geer, J. P., Levelt, J. M., and Plomp, R. "The ‘ Connotation of Musical Consonance," Acta Psychological ' Journal, 20: 308- 319 (1962). Walker, Lawrence. "A Concept Formation Analogue of Attitude Development,” Dissertation Abstracts, '22:2482-2483 (1962). Whiteley, John M. "The Selection andevaluation' f School Counselors," a paper presented at the American Personnel and Guidance Association convention, April, 1968. Williams, A. "Factor A:. Principal Components and Orthogonal Rotations," Michigan State University Computer Institute for Social Science Research, Technical Report No. 31, October 21, 1966. Williams, JohnE. "Connotations of Color Names Among Negroes and Caucasians," Perceptual and Motor Skills, 18:721-731 (1964). 191 Witt, E. L.: "The Connotative Meaning of Selected 'Forced' Rorschach Percepts as Measured by the Semantic Differential Technique,".Dissertation Abstracts, 26:4083-4084 (1966). ' Woodworth, R. 8., and Schlosberg, H. Ex erimental Psychology. New York: Henry Holt, 1954. Wozniak, Daniel F. "A Factor Analytic Study of Semantic Structures of Closed, Open, and‘Medium‘Beliefa Disbelief Systems." Unpublished Doctoral dissertation, Michigan State University, 1964. Zax, Melvin, Loiselle, Robert H., and Karras, Athan;' "Stimulus Characteristics of Rorschach Inkblots aS‘ Perceived by a Schizophrenic Sample," Journal of Projective Techniques and Personality Assessment, 3 9- (1960} o ' APPENDICES 192 APPENDIX A PERSON PERCEPTION TEST 193 194 ”l .3 PERSON PERCEPTION TEST This test attempts to measure (1) how you assess people, and (2) how people make you feel. You will see filmed segments of counseling sessions after which you can rate what you saw. Please answer on the basis of what the client on the screen has communicated to you about himself, and on the basis of how this made you feel. On each page of this test you will find the name of the client to be rated and beneath it two sets of scales. You are to rate the client on the first set and your feelings in the second set. Here is how you are to use these scales: If you feel that the client is very closely related to one end of the scale, you should place your check-mark as follows: - honest X ° : : : : : : dishonest . . ---4 ---- 93; -------- \\‘ honest : : : : : g; X~ : dishonest If you feel that the client is quite closely_related to one or theTBther end of the scale but not extremely, you should place your check-mark as follows: strong : X : : : : : : weak 0 s trong : : : "" : : X : fl : weak/ If the client seems only slightly related to one side as opposed to the other side (but not really neutral), then you should check as follows: 1‘\\ active : : : X : : : : passive 9.3. ‘ \ active : : : : X : : : passive The direction toward which you check shows which of the two ends of the scale seem most characteristic of the client you're judging. If you consider the client to be neutral on the scale, both sides of the scales egually associated with the client, or if the scale is\ om letel \; unrelated to the client, then you should place your check-mark 1n the middle space: safe : : : X : : : : dangdrpus i 195 Page 2 IMPORTANT: (1) Place your check-marks in the middle of spaces, not on the boundaries: . : : X : : X~ : THIS NOT THIS ‘ .1- r‘ (2) Be sure you check every scale.for every client--dg not omit any. . (3) Never put more than one check-mark on a single scale. Make each item a separate sad independent judgment. Work rapidly through this test. Do not worry or puzzle over individual,items. The best answer is your impression, the immediate "feeling" about the items. On the. other hand, please do not be careless. This study consists of five counseling segments. After viewing segment #1, turn the page and make your judgments of client #1. a».}\ f. 0 DO NOT TURN THE PAGE UNTIL.AFTER VIEWING SEGMENT #1 196 Client #1: Peggy A. Please make your judgments on the basis of what this client has communicated to you about herself. The client is: closed : : : : : : :open transparent : : : : : : :opaque near : : : : : : :far tired : : : : . : :energetic active : : : : : : :passive fast : : : : : : :slow yielding : : . : : : :tenacious defensive : : : ° : : :agressive excited : . : . fl : :relaxed static : : : : ‘: :dynamic ’ upset : . : : : : :calm unhappy : : : . : :happy complex_ : : : : : : :simple strong : . : : :weak - embarrassed : . ° : : : :smug —- controlled : ° : : : :uncontrolled hopeful : : . : : : :fearful feminine : : : : : : :masculine constrained . : : . : : :free honest : : : : : : :dishonest- profound : : : : : : :superficial- 8. Please make your judgments on the basis of how you feel toward the client. I feel: supportive : : . : : : : :angry perplexed“ : : : : : :understanding disagreeing_ : : ° : : :agreeing approving : _: : : : : :disapproving surprised : : : : : : :bored ' unhurried : : : : : : :harrassed mild : : : : : : :intense patient : : : : : : :impatient satisfied : : : : : : :dissatisfied After viewing the next film seggent, turn the page and make your_judgmentr of the . next client. 197 Client #2: Bill A. Please make your judgments on the basis of what this client has communicated to you about herself. The client is: .\ Open Opaque“ far energetic. Passive slow tenacious agressive relaxed. dynamic calm happy simple weak smug“‘ uncontrolled fearful ”masculine free dishonest \superficial U , closed transparent near tired active fast yielding .defensive excited static upset unhappy complex strong' embarrassed controlled hopeful feminine constrained honest profound l .0 .0 l0. l0. l0. .0 l0. .0 .0 O. O. l I l l O. .0 O. '0. l ’ l I .0 .0 O. O. O. O. .0 .0 I l. ..|.. |.. I. l B. Please make your judgmentsoon the basis of how'you feel toward\the client. I feel: \ supportive : : : : : : : angr;\ perplexed : : : : : : : understanding disagreeing : : : : : : : agreeing approving : : : : : : : disapproving surprised. : : : : : : : bored . unhurried : : : : : : : harrassed mild : : : : : : : intense patient : : : : : : g impatient. satisfied ° ' ° : : : . : dissatisfied- . . . , I / After viewin the next film se ment, turn the page and make your jhdgments of the next client. \ 198 Client # 3 A- A‘ A. Please make your judgments on the basis of what this client has communicated to you about herself. \ The client is : ‘\ \ closed : : : : : : : open>Vw transparent : : : : : : : opaque . near : : : : : :_____: far ‘ tired : : : : : : :;.energetic active : : : : : : fl» : passive fast : : : : : : ‘” : slow yielding : : : : : : : tenacious defensive : : : : .:. : : agressive excited : : : : : i: v :..relaxad.- . static : : : : :r - : : dynamic upset : : : : :' :- : calm unhappy : : : : : : 3 haPPY complex : : : : : : : simple strong : : : : : : : weak embarrassed : : : : : : : \ ug controlled : : : : : : : u controlled hopeful : : : : : : : fgarful feminine : . : : : : : masculine constrained : : : : : : 3. free\\ / honest : : : : : : : dishonest profound : : : : : ' : : superficial B. Pleaae‘make your judgments on the basis of how you feel toward the client; I feel: 'i v! I It . supportive : : : : : : : angry perplexed : : : : : : : understanding disagreeing : : : : : : : agreeing‘ approvin : : : : : : : disapproving surprised : z : : : : : bored unhurried : : : : : : fv_ : harrassed mild ______:______: : : : : : intense patient :______3______3 : 2 : : impatient satisfied :____: : ': : : : dissatisfied 3 .After viewing_the next film segment, turn the page and make your judgments of-‘ the next client. fl ‘ ' Client #4 199 : Lynn \ A. Please make your judgments on the basis of what this client has\ébmmunicated to you about herself. The client is: closed transparent near tired active fast yielding defensive excited static upset unhappy complex strong embarrassed controlled hopeful feminine constrained honest profound Ii. I O. '0. ‘0. |.. 0. l0. .0 0. I. O. O O. O. I. O. O. O. O. l. I. O. O. O. \ :open :opaque :far . :energetic :passive :slow :tenaCious :agressivem ° * :relaxéd :dynamic :calm l ____:happy :simple :weak :smug :uncontrolled :fearful :masculine / :free :dishonest :superficial t B. Please make your judgments on the basis of how you feel toward the client. I feel: supportive perplexed disagreeing m 'C! '0 H O < P F surprised unhurried mild patient satisfied | I ll. :angry :understanding :agreeing :disapprovifig :bored :harrassed :intense :impatient :dissatisfied ,//' Aftrg;;viewing the next film segment, turn the page and make your judgments of the next client . . 200 Client #5: Terry A. Please make your judgments on the basis of what this client has communicated to you about herself. ‘ The client is: \\ ‘\ closed : : : : : : : Open transParent : : : : : : : Opaque m m m * -—-—- —-——— near :t : : : : : : far . w w m w m # ans—— tired : : : : : : : energetic active : : : : : : : passive faSE : : : : : : : slow . ' M m m * w m o yielding : : : : : : : tenaCious defensive : : ; : : z : agressive eXCited : : : : : : : relaxed static : .: : : : : : dynamic —-——-— m fl m M‘s—— upset ; g g : : : : calm unhappy : : : : : :______: happy complex ; ; g : : : : simple Strong : : : : : : : weak ————- urn—‘- —-— .n—s-‘D ” # embarrassed ; g g ; : : : smug controlled : : : : : : : uncontrolled -_ w w W * ” hopeful : : : : : : : fearful feminine : g g g : : : masculine . *— m— m —_——- ” # constrained : : : : : : : free ————— cum—III— ¢———II- -——-—-— o—-——- a“ . . honest : : : ‘: : : : dishonest * * ——-—- w w w . profound : : : : : : : superficial B. Please make your judgments on the basis of how you feel toward the client, I feel: 8UPP°rtive : : : ' : : : : angry perplexed : : : : : : : understanding disagreeing ; ; ; g ; : ; agreeing' aPPIOVing 5 : : : : : : disaPPrOViDS surprised {————-{———-_; g : : : bored unhurried : : : : : : : harrassed~ mild : : : : : : ‘: intense patient : : : : : : : impatient satiSfied : : : : : : : dissatisfied lifter viewing the next film segment, turn the page and make your judgments of the next client. APPENDIX B PERSON PERCEPTION TEST: RAW SCORES 201 202 TABLE B.1--Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group I by subject: Evaluative factor. Client-Percepts Subject Peggy Bill Bob Lynn Terry Total Variance 1 63 61 26 44 44 238 181.8 2 52 43 58 57 47 257 33.0 3 36 67 48 28 45 224 172.6 4 47 61 57 31 61 257 130.2 5 57 57 56 48 48 266 18.2 6 69 70 68 47 41 295 154.0 7 46 44 44 41 50 225 8.8 8 61 38 60 60 64 283 88.6 9 46 53 57 43 49 248 24.6 10 54 44 58 36 46 238 59.8 11 60 52 24 51 61 248 180.2 12 64 42 56 54 43 259 69.0 14 47 55 60 26 28 216 192.6 15 57 55 56 45 51 264 19.4 16 51 44 40 37 33 205 38.0 17 53 58 55 46 54 266 15.8 18 56 52 52 33 39 232 77.8 19 44 39 45 36 47 211 16.6 20 60 51 38 24 15 188 275.4 21 47 57 42 33 37 216 69.8 22 54 47 40 28 40 209 74.6 23 58 52 63 47 54 274 29.4 24 54 48 52 42 45 241 19.4 25 43 41 25 23 45 177 88.6 26 65 64 68 67 67 331 2.2 27 54 49 57 18 32 210 218.8 28 47 28 30 36 48 189 69.8 29 54 47 68 27 36 232 201.8 30 67 38 69 59 69 302 138.0 Mean 54.0 50.2 50.8 40.2 46.2 241.4 92. l-‘ 203 TABLE B.2--Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group II by subject: Evaluative factor. Client-Percepts Subject Peggy Bill Bob Lynn Terry Total Variance 1 51 62 47 21 48 229 182.2 2 6O 59 67 39 28 253 214.6 3 47 45 42 29 32 195 51.6 4 54 65 65 34 33 251 202.2 5 45 51 57 27 45 225 100.8 6 60 50 55 23 29 217 215.4 7 59 63 41 31 33 227 175.0 8 63 63 67 38 33 264 204.2 9 57 40 41 28 30 196 106.2 10 67 68 67 40 45 287 150.6 11 51 52 35 24 40 202 109.0 12 64 38 61 20 50 233 260.6 13 60 56 62 27 30 235 232.8 14 46 48 61 19 31 205 211.6 15 45 67 60 24 25 221 309.4 16 52 54 54 32 31 223 115.0 17 59 66 53 43 36 257 116.2 18 44 59 45 26 35 209 121.4 19 55 57 54 35 31 232 122.2 20 48 57 60 35 47 247 77.0 21 55 47 47 27 23 199 156.2 22 55 59 53 34 42 243 85.0 23 52 53 51 28 49 233 88.2 24 55 65 64 33 38 255 174.8 25 48 51 58 22 53 232 159.4 26 48 35 4O 30 19 172 94.6 27 41 51 45 40 40 217 17.8 28 54 53 70 20 20 217 401.4 29 47 51 56 40 43 237 32.2 30 60 54 63 25 47 249 183.8 \1 Mean 53.4 54.6 54.7 29.8 36.2 228.7 155. 204 TABLE B-3--Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" socring procedure for Group I by subject: Emotional factor. Client-Percepts Subject Peggy Bill Bob Lynn Terry Total Variance 1 30 31 31 22 29 143 11.4 2 15 25 35 29 21 125 46.4 3 12 22 27 20 21 102 23.4 4 14 23 36 22 30 125 56.0 5 22 28 34 26 24 134 17.0 6 18 31 32 24 28 133 26.2 7 11 22 30 23 23 109 37.4 8 16 25 32 28 27 128 28.2 9 15 31 23 28 28 125 31.6 10 17 30 28 23 24 122 20.2 11 10 21 27 15 33 106 67.4 12 22 15 28 37 26 128 52.2 14 26 24 33 15 20 118 36.2 15 12 27 27 20 19 105 31.6 16 18 22 26 22 27 115 10.4 17 21 30 33 20 32 136 31.0 l8 17 26 30 24 21 118 19.4 19 19 24 29 25 21 118 11.8 20 21 24 21 19 17 102 5.4 21 18 27 22 23 26 116 10.2 22 19 30 30 21 21 121 23.0 23 15 19 33 19 25 121 42.6 24 18 33 28 28 36 143 37.4 25 19 26 27 25 30 127 13.0 26 12 27 38 23 30 130 73.2 27 17 32 33 24 20 126 40.6 28 12 16 20 13 20 81 11.4 29 21 19 31 28 29 128 22.2 30 23 26 39 20 39 147 65.0 Mean 17.6 25.7 29.8 23.0 25.8 121.8 31. F—l 205 TABLE B.4--Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group II by subject: Emotional factor. Client-Percepts Subject Peggy Bill Bob Lynn Terry Total Variance 1 12 36 20 27 23 118 62.6 2 27 32 37 16 30 142 49.0 3 20 28 28 31 34 141 21.8 4 15 32 37 26 25 135 54.8 S 11 32 31 20 36 130 84.4 6 19 28 35 32 35 149 35.8 7 20 22 25 23 20 110 3.6 8 29 35 39 23 25 151 36.2 9 23 24 27 10 26 110 38.0 10 18 26 31 25 19 119 23.0 11 22 24 25 29 23 123 5.8 12 17 23 30 30 25 125 23.6 13 21 25 22 25 15 108 13.4 14 17 22 35 33 30 137 46.6 15 8 37 35 10 23 113 146.6 16 13 25 31 24 18 111 38.2 l7 19 29 34 28 28 138 23.4 18 26 30 34 23 23 136 18.2 19 21 33 38 20 36 148 57.8 20 20 33 36 31 28 148 29.8 21 18 22 28 30 31 129 25.0 22 30 37 36 26 34 163 16.6 23 20 33 34 24 22 133 33.4 24 22 39 38 27 21 147 59.4 25 22 27 31 16 16 112 35.4 26 19 29 34 23 24 129 27.0 27 17 23 26 19 25 110 12.0 28 12 18 38 14 17 99 87.4 29 17 28 33 22 23 123 29.8 30 19 32 33 12 26 122 63.4 H ZMean 19.1 28.8 32.0 23.3 25.4 128.6 40. 206 TABLE B.5--Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group I by subject: Potency factor. Client-Percepts Subject Peggy Bill Bob Lynn Terry Total Variance 1 22 21 25 16 15 99 14.2 2 18 10 21 20 18 87 15.0 3 20 16 20 24 18 98 7.0 4 17 20 20 18 18 93 1.4 5 15 16 20 18 14 83 4.6 6 17 15 18 19 19 88 2.2 7 18 15 18 20 17 88 2.6 8 15 14 21 18 15 83 6.6 9 14 16 22 19 23 94 11.8 10 20 17 22 20 16 95 4.8 11 15 17 24 23 15 94 15.4 12 16 16 17 26 17 92 14.6 14 18 10 22 17 14 82 15.4 15 19 16 17 20 20 92 2.6 16 14 17 18 21 23 93 9.8 17 17 17 17 18 19 88 0.6 18 20 13 19 15 16 83 6.6 19 15 11 16 18 16 76 5.4 20 15 16 17 20 19 87 3.4 21 20 15 16 16 16 83 3.0 22 14 11 16 18 19 78 8.2 23 20 16 22 20 20 98 3.8 24 16 18 20 21 19 94 3.0 25 12 16 13 20 18 79 9.0 26 19 18 20 18 16 91 1.8 27 13 12 22 23 16 87 20.6 28 14 20 17 19 14 84 6.2 30 18 16 23 22 21 100 6.8 Mean. 16.8 15.5 19.4 19.6 17.6 88.8 7.4 207 TABLE B.6--Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedure for Group II by subject: Potency factor. Client-Percepts Subject Peggy Bill Bob Lynn Terry Total Variance 1 17 18 21 27 12 95 24.4 2 22 17 22 16 18 95 6.4 3 15 ll 13 18 20 77 10.6 4 21 20 21 16 15 93 6.6 5 14 15 14 21 18 82 7.4 6 21 14 22 19 20 96 7.8 7 18 17 16 14 15 80 2.0 8 19 20 22 12 15 88 13.0 9 20 19 20 24 17 100 5.2 10 18 15 21 18 16 88 4,2 11 14 13 17 26 20 90 22.0 12 23 13 21 10 23 90 29.6 13 21 14 22 15 14 86 12.6 14 16 16 22 25 21 100 12.4 15 9 15 16 20 20 80 16.4 16 21 17 19 18 16 91 3.0 17 13 12 22 14 17 78 13.0 l8 18 15 19 20 16 88 3.4 19 20 16 21 20 14 91 7.4 20 16 15 17 18 13 79 3.0 21 14 13 17 23 17 84 12.2 22 17 17 13 19 18 84 4.2 23 12 14 21 22 16 85 15.2 24 17 16 23 16 19 91 7.0 25 22 24 23 25 25 119 1.4 26 16 12 16 23 22 89 17.0 27 15 14 16 16 17 78 1.0 28 13 8 19 18 17 75 16.4 29 18 15 18 19 17 87 1.8 30 22 14 23 10 27 96 39.0 LO Mean 17.4 15.3 19.2 18.7 17.8 88.5 10. 208 TABLE B.7--Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7” scoring procedure for Group I by subject: Evaluative- activity factor. Client-Percepts Subject Peggy Bill Bob Lynn Terry Total Variance 1 55 23 51 37 42 208 127.0 2 49 27 48 46 37 207 69.8 3 38 43 39 25 35 180 36.8 4 54 28 47 27 54 210 146.8 5 41 44 50 19 34 188 113.0 6 54 48 52 30 30 214 113.0 7 38 25 40 29 37 169 33.4 8 48 27 57 28 37 197 134.6 9 45 36 50 23 32 186 91.0 10 51 27 50 26 44 198 120.2 11 37 31 40 34 45 187 23.4 12 53 21 37 13 35 159 191.4 14 33 22 45 14 12 126 152.6 15 51 34 43 33 40 201 43.0 16 42 32 35 24 30 163 35.0 17 40 33 44 35 44 196 20.6 18 52 27 41 25 28 173 107.4 19 49 29 39 26 33 176 66.6 20 51 29 33 16 30 159 126.2 21 43 44 37 27 29 180 48.8 22 44 25 32 17 19 137 96.2 23 54 37 42 29 45 207 69.0 24 41 34 45 29 36 185 30.8 25 39 26 26 30 45 166 57.4 26 60 38 51 45 52 246 54.2 27 48 28 45 18 28 167 128.6 28 44 19 29 29 39 160 76.0 29 47 30 47 28 37 189 65.4 30 58 21 59 38 48 224 199.8 Mean 46.9 30.6 43.2 27.6 36.4 184.8 88. L0 209 TABLE B.8-~Raw scores for the Person Perception Test by client-percept, total, and variance, using the "1-7" scoring procedures for Group II by subject: Evaluative- activity factor. Client-Percepts Subject Peggy Bill Bob Lynn Terry Total Variance 1 4O 47 40 31 30 188 40.2 2 S4 39 58 18 15 184 315.8 3 49 27 31 28 30 165 66.0 4 47 46 50 18 19 180 206.0 5 53 26 28 34 37 178 91.4 6 55 36 51 15 23 180 239.2 7 41 30 41 24 22 158 65.8 8 55 50 62 19 19 205 337.2 9 57 20 26 15 18 136 235.0 10 47 42 45 26 32 192 65.0 11 SO 34 28 13 31 156 140.6 12 53 26 47 27 47 200 126.4 13 59 28 45 21 24 177 208.2 14 50 25 43 14 32 164 163.0 15 28 56 50 10 23 167 294.2 16 45 33 41 23 36 178 56.6 17 55 35 40 23 14 167 199.4 18 48 41 51 3O 33 203 66.6 19 52 32 39 25 30 178 87.4 20 55 29 42 27 38 191 101.4 21 45 24 24 21 15 129 103.0 22 35 44 44 33 38 194 20.6 23 39 36 42 18 34 169 69.8 24 43 42 54 25 38 202 87.4 25 52 34 44 18 37 185 128.8 26 47 29 36 25 24 161 72.6 27 44 35 26 25 24 154 59.0 28 43 22 55 19 16 155 234.0 29 50 38 43 18 21 170 155.6 30 52 26 40 20 41 179 130.6 Mean 48.1 34.4 42.2 22.1 28.0 174.8 138. to 210 TABLE B.9--Frequency scores for the Person Perception Test by client-percept and total for Group I: Evaluative factor. - L Client-Percepts Subject Peggy Bill Bob Lynn Terry Total 1 4-5-1 7-3-0 9-0-1 6-4-0 9-1-0 35-13- 2 2 9-0-1 2-0-8 8-1-1 7-1-2 3—2-5 29- 4-17 3 6-0-4 3-7-0 9-0-1 3-3-4 6-1-3 27-11-12 4 7-2-1 8-2-0 8-2-0 6-3-1 9-1-0 38-10- 2 5 3-4-3 3-4-3 7-1-2 1-4-5 1-2-7 15-15-20 6 1-9-0 10-0-0 2-8-0 1-5—4 1—3-6 5-35-10 7 8-0-2 9-0-1 9-0-1 7-0-3 9-0-1 42- 0- 8 8 9-1-0 8-1-1 8-2-0 8-2-0 6-4-0 39-10- 1 9 7-1-2 9-0-1 8-1-1 7-1-2 7—1-2 38— 4- 8 10 9-0-1 6-0-4 10-0-0 7-0-3 6-0-4 38- 0-12 11 7-3-0 9-1-0 2-8-0 5-5-0 5-4-1 28-21- 1 12 3-7-0 10-0-0 10-0-0 4-4—2 8-0-2 35-11- 4 14 10-0-0 10-0-0 10-0-0 4-5-1 2-4-4 36- 9- 5 15 10-0-0 10-0-0 8-0-2 8-0-2 6-2-2 42- 2- 6 16 7-0-3 6-0-4 6-0-4 6-0-4 4-2-4 29- 2-19 17 10-0-0 8-2-0 10-0-0 6-0-4 8-0—2 42- 2- 6 18 9-0-1 8-0-2 9-0-1 6-0-4 6—0-4 38- 0-12 19 7-0-3 5-0-5 7-0-3 7-0-3 5-1-4 31- 1-18 20 9-1-0 8-0-2 10-0-0 6—2—2 4-6-0 37- 9- 4 21 5-2-3 3-5-2 5-0-5 5-1-4 4-1-5 22- 9-19 22 8-1-1 8-0-2 10-0-0 8-2-0 6-1-3 40- 4- 6 23 9-1-0 7-0-3 7-3-0 9-0-1 8-0-2 40- 4- 6 24 3-3-4 5-2-3 3-3-4 3-1-6 1-1-8 15-10—25 25 9-0-1 7-0-3 10-0-0 8-1-1 5-0-5 39- 1-10 26 1-9-0 2-8-0 1-9-0 0-9-1 0-9-1 4-44- 2 27 10-0-0 10-0-0 9-0-1 6-4-0 10-0-0 45- 4- 1 28 9-0-1 6-1-3 10-0-0 8-2-0 8-1-1 41- 4- 5 29 8-2-0 6-1-3 2-8-0 5-3-2 5-0-5 26-14-10 30 2-8-0 5-0-5 1-9-0 4-6-0 1-9-0 13-32- 5 Legend: First number is frequency of responses in scale positions 2, 3, 5, 6. Second‘ number is frequency of responses in scale positions 1, 7. Third number is frequency of reSponses in scale position 4. Evaluative factor. Lynn Terry Total Client-Percepts 211 Bob Bill Peggy TABLE B.10--Frequency scores for the Person Perception Test by client-percept and total for Group II: Subject 6978667457757738200510017.4585538 1 1 111 2 l 11 1 ..__._..—_.-____..-...-__..... 681689907684470293366810410668 1 1 2 2 ll 2 l l 211 2 1 -_..._..-_.__-..__..-_..__..__ 832665468751967099236126115914 323233323123232422233241323132 6374877.77836655853356175306654 243212322541221212340827253235 .._.._...._........._......._. 3207.342221454520412122120040525 ._..........._..__......_..._. 547664566424437857550007101.7350 004000101030000005203210032021 .._......._............_....._ 070501070000331.6010320107100025 .........._......._......._l.. 0365099392777940955877939700.64 l l l 0 004021100031100201213111052120 ._._........_....._...._...... 301520540911107272300207100501 ._............................ 70556946016880363759779295004009 l l . 013211010020110203320102024130 ._.............................. 060233154814502030322303000303 ..........._..........._._.... O37666946276498877468605086677 1 .1. 1 1234567009012345678901234567890 111111111122222222223 5’ 6. Second number is frequency of responses in scale positions 1, 7. Third number is frequency of responses in scale First number is frequency of responses in scale position 4. positions 2, 3, Legend: 212 TABLE B.11--Frequency scores for the Person Perception Test Emotional factor. by client-percept and total for Group I: Client-Percepts 1 a t 0 ml. V: r r e T n n V1 1.. b O B 1 1 .1 B V. g g e P t C e .11. b u S 64635364444555446462442646640 01...003331320002022305002000011 ....................._.._.... 10520100003240010030000011304 .__...___...._....._._....... 551143235343424634331664655351 10001111311100011101012000110 ......._.....__........._._._ 00030202001000010000000030004 ._....................._.__.. 5663535335456664556565463/0552 12002111110000012401111201002 ....._.........__.._..._..... 10010200002000010010001010200 .._...__..._.._......_...._.. 44654355554666644255554455464 10001000100110011111111101002 ........._..........._..._... 11200111103000011000011020111 ........._..........._....... 45465555463556644555544545553 12345678901245678901234567890 11111111122222222223 7. Third number is frequency of responses in scale Second number is frequency of responses in scale position 4. First number is frequency of responses in scale positions 2, 3, 5, 6. positions 1, Legend: Emotional factor. Lynn Terry Total Client-Percepts 213 Bob Bill Peggy TABLE B.12--Frequency scores for the Person Perception Test by client-percept and total for Group 11: Subject 460635272705122565343845171235 1 1 .............-......__._._..__ 660696296290373341910635720519 1 1 1 1 1 ._....__..___......_......_._. 080889642115615204857630219366 212111212211221222112121222121 111011020021010123001201020001 _._............._........_.... 000032101031211000110211400003 ............_......._........i. 5.135623545614445543555254246362 112411130522011111111151304.00111 .._......................._.__ 110102003022034.0101400113300403 .._........._..._............. 444153533122621545415140326252 012101110131000210111110001111 ._....._..........._._..._.... 030311051213131120420205010412 _..........._................. 624254505322535336135351655143 113111012110101111110011110101 ...._...._.__._........._..... 200020010011003000230205000100 ....................._...._... 353535644545562555326450556465 122001000021000010010010000011 ...... . . . . . _ . _ . . . . . _ . . . _ . . . . .. 320231131023004211110011010301 ...... _ . . . . . . . . . . _ . . . . . . _ . . . .. 224434535622662445546645656354 123456789012 34 11 67 90 1111112 22 23 24 25 26 29 3O 5 00 1 1 7. Third number is frequency of reSponses in scale Second number is frequency of responses in scale position 4. First number is frequency of responses in scale positions 1, positions 2, 3, 5, 6. Legend: Potency factor. Total Terry Client-Percepts Lynn 214 Bob Bill Peggy TABLE B.13--Frequency scores for the Person Perception Test by client-percept and total for Group I: Subject 30064098415540SSS/07800236531466 1 ..__....._..._._.._....._.__. 41330402004210000000001010214 ......._..................... 311307249514955543220073569430 11111 11111111111111111111111 13313331021100012204101210232 _._.._......_._..........._.. 00010101000000000000000000001 ........._............._..... 31121012423344432240343234211 00021321101001111212011010011 ...._...__................... 202.00100001210000000001000111 .........._...._._......._... 24223023342233333232432434322 01.112110011101211.1030001110121 .........................._.. 20110201002000000000000010002 ........._.................._ 23222133431343233414443324321 11102110011002111021011200001 ...._........_._............. 01000000001000000000000000100 ...................._._...... 32342334432442~333423433244343 1.5102112011201111321112001101 ............................. 00010000000000000000000000000 ...._......_._............... 31332332433243333123332443343 12345678901245678901234567890 11111111122222222223 Third number is frequency of responses in scale Second number is frequency of responses in scale position 4. First number is frequency of responses in scale positions 1, 7. positions 2, 3, 5, 6. Legend: 215 TABLE B.14--Frequency scores for the Person Perception Test Potency factor. by client-percept and total for Group II: Client-Percepts Subject Total Bill Bob Lynn Terry Peggy 3500263354616313841363525134355 1 1 ._...___.._...____.._.__...... 330113132234015010101200540916 ._........._...._..._._..._... 422734624260782259646385436849 11111111111111111 111 11111 1 112011120201100224110301021000 .... . _ . . . . . . . . . . . _ . . . . . . . . . _ .. 000002000001000000000000220103 ................_...._........ 332431324242344220334143203341 011111011002001202110307.1121le ....._.... . . . . . . . _ . . . . . . . . . . .. 310111022022012000001100320212 ._..._.... . . . . . . _ . . . . . . . . . . . .. 123223411420431242333042012120 112010111101011111010310000111 ....._....._........_._....... 010000010100001010100100000301 ..._.._...........__._.__._... 322434323243432323333034444032 000021001101101.503021301000112 ._................._._.._..._. 000000100100001000000000000200 ....._........................ 434423343243342141423143444132 113111111211100011111311001020 ._.............._............. 010000000011001000000000000100 ....._........................ 32133333322234343333313344.3324 123456789012345678901234 111111111122222 s, 6. 7. Third number is frequency of responses in scale Second number is frequency of responses in scale position 4. First number is frequency of responses in scale positions 1, positions 2, 3, Legend: Total Evaluative- Terry Client-Percepts Lynn 216 Bob Bill Peggy TABLE B.15--Frequency scores for the Person Perception Test by client-percept and total for Group I: activity factor. Subject 16315423734227023788644.1025sz 1 111 11 1 1 1 1 ..............._....___._.... 002125503109260104063642146714 11 1 11 1 2 ..._._......._.._......._.... 67125639722.1784300814379317328 32332133343323333232332433331 ....._._......_.............. 9360020400888738558271795979816 04003741010011222315104100020 ....._....................... 5056.5200002660100041410106407 .........................._.. 45431058987328677643485793572 01000010100002120403212001200 .._..............._.......... 20010303106100000010001050205 ..._..._............._....... 7009896006793897879586786948594 13213011312003121252122001200 ..._._........._..__......... 00000300001100000000001010205 _......_._.._..._............ 00678668868689687874787690000594 12002010201010220112212000111 ................_...._...._.. 02440500000440003000030060016 ..................._... . . . . .. oo55574009798549776887757939872 12345678901245678901234567890 11111111122222222223 s, 6. 7. 3. Third number is frequency of responses in scale Second number is frequency of responses in scale position 4. First number is frequency of responses in scale positions 1, positions 2, Legend: 217 TABLE B.16--Frequency scores for the Person Perception Test by client-percept and total for Group II: Evaluative- activity factor. Client-Percepts Subject Peggy B111 Bob Lynn Terry Total 1 7—2-0 8-0-1 8-0-1 4-3-2 8-0-1 35- 5- 5 2 5-4-0 8—0-1 5-4-0 4-4-1 3-5-1 25-17- 3 3 7-1-1 6-0-3 3-0-6 4-0-5 4-0-5 24- 1-20 4 8-0-1 9-0-0 9-0-0 0-6-3 1-5-3 27-11- 7 5 4-5-0 7-0-2 8-0-1 6-2-1 6-0-3 31- 7- 7 6 6-3-0 8-0-1 6-3-0 0-7-2 4—4-1 24-17- 4 7 8-0-1 8-0-1 6-1-2 5-3-1 6-3-0 33- 7- 5 8 6-3-0 6-2-1 1-8-0 2-5-2 1-5-3 16-23- 6 9 5-4-0 8-1-0 8-0-1 0-7—2 5-4-0 26-16- 3 10 7-2-0 5-1-3 7-1-1 8-0-1 6-0-3 33- 4- 8 11 5—2-2 8-0-1 7-0-2 1-7-1 3-2-4 24-11-10 12 3-5-1 7-1-1 8-0—1 2-4-3 6-1-2 26-11- 8 13 3-6-0 9-0-0 8-0-1 4-3-2 6-3-0 30-12- 3 14 7-1-1 8-0-1 9-0-0 5-4-0 3-0-6 32- 5- 8 15 6-3-0 6-3-0 8-1-0 1-8-0 6-3—0 27-18 -0 16 6-2-1 7-0-2 6-0-3 4-3-2 7-0-2 30- 5-10 17 5-4-0 7-1-1 9-0-0 6-2-1 4-5-0 31-12- 2 18 4-3-2 4-0-5 9—0-0 3-0-6 2-0-7 22- 3-20 19 5-4-0 8-1-0 6-1-2 5-2-2 2-1-6 26- 9-10 20 4-4-1 9-0-0 6-1-2 3-2-4 7-1-1 29— 8- 8 21 8-0-1 9-0-0 9-0-0 6-3-0 3-5—1 35- 8- 2 22 8-0-1 5-1-3 8-0-1 0-1-8 1-0-8 22- 2-21 23 8-1-0 8-0-1 7-0-2 4-4-1 7-0-2 34- 5- 6 24 5-3-1 6-1-2 4-5-0 1-4-4 5-0-4 21-13-11 25 6-3-0 9-0-0 9-0-0 0-6-3 5-2-2 29-11- 5 26 9-0-0 7-0-2 6-0-3 1-3-5 2-3-4 25- 6-14 27 7-0-2 9-0-0 9-0-0 8-0-1 9-0-0 42- 0- 3 28 6-2-1 5-2-2 5-4-0 3-5-1 1-6-2 20-19- 6 29 8-1-0 9-0-0 7-1-1 6-2-1 5—2-2 35- 6- 4 30 7-2-0 9-0-0 7-0-2 1-5-3 4-3-2 28-10- 7 Legend: First number is frequency of responses in scale positions 2, 3, 5, 6. Second number is frequency of responses in scale positions 1, 7. Third number is frequency of responses in scale position 4. APPENDIX C PERSON PERCEPTION TEST: DISTRIBUTION OF RESPONSES BY ITEM 218 TABLE C.-1--Person Perception Test: responses by item. 219 distribution of g: Scale Position Item 0 5 1 2 3 4 s 6 7 Mean SD Closed-Open Peggy I 0 0 1 O 5 17 6 5.93 .89 II 0 0 1 2 2 15 10 6.03 .98 Bill I 0 3 6 2 8 7 3 4.66 1.54 11 0 1 6 0 5 12 6 5.30 1.49 Bob I l 2 3 0 9 11 3 5.03 1.54 II 0 1 4 0 5 13 7 5.53 1.38 Lynn I 13 10 1 1 1 1 2 1.82 1.82 11 26 2 l 1 0 0 0 1.23 .66 Terry 1 7 4 1 4 3 6 4 3.90 2.22 11 13 3 2 8 3 1 0 2.60 1.62 Dishonest-Honest Peggy I 0 0 0 1 8 14 6 5.86 .77 II 0 0 0 2 7 9 12 6.03 .95 Bill I 1 0 1 3 7 12 5 5.45 1.29 11 0 0 1 2 1 13 13 6.17 1.01 Bob I 1 4 1 2 7 7 7 5.03 1.79 II 0 1 2 2 6 12 7 5.57 1.28 Lynn I 1 3 3 9 3 5 5 4.55 1.72 II 1 1 2 13 6 4 3 4.53 1.36 Terry 1 0 1 4 12 1 8 3 4.69 1.37 II 1 0 3 19 3 3 1 4.20 1.08 Superficial-Profound Peggy I 0 7 9 3 5 5 0 3.72 1.44 11 0 5 9 4 6 6 0 3.97 1.40 Bill I 0 5 9 6 4 3 2 3.90 1.47 11 1 10 0 6 l 2 0 3.07 1.25 Bob I 2 8 6 3 4 5 1 3.62 1.71 11 5 7 8 3 5 2 0 3.07 1.50 Lynn 1 1 4 1 15 2 5 1 4.10 1.40 11 0 1 2 21 2 1 3 4.30 1.10 Terry I 0 4 3 15 6 0 1 3.93 1.10 11 1 0 2 20 2 2 3 4.33 1.22 TABLE C.1--Continued 220 Scale Position 0 Q. :3 Item 8 1 2 3 4 5 6 7 Mean SD L9 Angry-Supportive Peggy I 0 0 0 1 5 12 11 6.14 .82 II 0 0 0 2 9 12 7 5.80 .87 Bill I 0 O 0 2 0 8 9 5.83 .95 II 0 0 l 2 6 12 9 5.87 1.02 Bob I 1 0 3 0 7 9 9 5.59 1.47 II 0 0 2 4 3 12 9 5.73 1.21 Lynn I 1 l 4 2 8 5 8 5.14 1.66 11 3 3 8 12 3 1 0 3.40 1.20 Terry I 1 0 2 4 8 9 5 5.24 1.38 II 1 3 5 8 7 4 2 4.23 1.48 Perplexed-Understanding Peggy I 0 0 2 0 5 14 8 5.90 1.03 II 0 0 1 0 S 14 10 6.07 1.01 Bill I 0 0 4 6 6 7 7 5.41 1.39 II 0 0 3 2 4 11 10 5.77 1.26 Bob I 0 1 3 1 4 14 6 5.55 1.36 II 0 1 2 0 2 16 9 5.90 1.22 Lynn I 2 6 4 1 2 8 6 4.48 2.10 11 11 6 8 2 0 3 0 2.43 1.52 Terry I 3 1 0 4 3 11 7 5.21 1.84 II 7 7 4 1 5 5 1 3.30 1.95 Disagreeing—Agreeing Peggy I 0 O O 9 7 9 4 5.28 1.05 II 0 1 1 8 9 8 3 5.13 1.17 Bill I 0 1 1 12 8 4 3 4.76 1.17 11 0 l 1 6 8 8 6 5.30 1.24 Bob I 0 1 5 5 3 11 4 5.03 1.45 II 1 1 7 5 1 9 6 4.83 1.73 Lynn I 0 0 6 11 6 2 4 4.55 1.27 II 1 5 6 15 3 0 0 3.47 .99 Terry I 0 1 0 12 2 11 3 5.07 1.23 II 1 2 5 18 2 2 1 4.03 1.17 TABLE C.1--Continued 221 ======== = T 9. Scale Position Item 3 1. (3 1 2 3 4 5 6 7 Mean SD Disapproving-Approving Peggy I 0 0 1 6 6 11 5 5.49 1.10 II 0 0 2 5 6 10 12 6.67 1.69 Bill I 0 0 2 10 8 6 3 4.72 1.35 11 1 0 1 6 9 8 5 5.20 1.33 Bob I 0 3 1 4 7 10 4 5.10 1.45 II 0 2 1 6 5 9 7 5.23 1.35 Lynn I 0 2 9 6 5 4 3 4.31 1.47 II 1 8 8 11 1 1 0 3.20 1.08 Terry I 1 1 1 11 4 8 3 4.79 1.42 11 2 6 4 8 4 5 1 3.83 1.61 Harrassed-Unhurried Peggy I 0 0 4 7 10 4 4 4.90 1.21 II 0 1 6 8 4 7 4 4.73 1.44 Bill I 0 0 1 7 6 10 5 5.38 1.08 11 0 0 0 5 '6 14 5 5.63 .95 Bob I 0 0 6 5 4 8 6 5.10 1.45 11 0 0 2 4 6 9 9 5.63 1.22 Lynn I 2 6 6 7 3 3 2 3.69 1.64 II 7 9 2 5 1 4 2 3.13 1.95 Terry I 1 2 5 7 6 4 4 4.48 1.59 II 2 7 9 4 3 2 3 3.57 1.71 Impatient-Patient Peggy I 0 1 3 0 7 9 9 5.61 1.37 11 0 3 7 1 4 6 9 5.00 1.93 Bill I 0 1 2 2 5 13 6 5.55 1.27 11 0 1 2 0 4 11 12 5.93 1.29 Bob I 1 1 3 1 9 8 6 5.21 1.54 II 0 1 4 2 3 10 10 5.57 1.50 Lynn I 3 4 4 5 4 5 4 4.31 1.93 11 6 12 3 3 1 5 0 2.86 1.71 Terry I 1 2 2 4 7 8 5 5.00 1.60 11 1 5 9 2 3 9 1 4.06 1.69 222 TABLE C.1--Continued g: Scale Position Item 2 o 1 2 3 4 5 6 7 Mean SD Dissatisfied-Satisfied Peggy I 0 4 4 4 7 4 6 4.72 1.67 II 0 2 1 4 10 9 4 5.17 1.29 Bill I 0 2 9 6 5 4 3 4.31 1.47 II 0 2 6 4 3 10 5 4.93 1.59 Bob I 1 4 2 5 3 11 3 4.72 1.72 II 0 1 4 3 7 12 3 5.13 1.31 Lynn I 7 6 5 4 1 3 3 3.23 2.01 II 16 7 4 2 1 0 0 1.83 1.10 Terry I 3 6 4 8 1 3 4 3.72 1.88 11 6 9 4 8 2 1 0 2.80 1.37 Opaque-Transparent Peggy I 0 1 5 0 l4 5 4 5.00 1.31 II 0 3 2 3 10 11 1 4.90 1.33 Bill I 1 5 4 2 3 3 1 3.48 1.43 II 0 4 8 1 10 6 1 4.30 1.46 Bob I l 5 3 5 5 8 2 4.38 1.69 II 0 6 5 3 7 7 2 4.20 1.62 Lynn I 6 11 1 4 2 3 2 3.07 1.91 II 20 4 1 4 0 0 1 1.80 1.42 Terry I 2 3 2 9 5 6 2 4.31 1.62 II 10 2 3 10 4 1 0 2.97 1.61 Far-Near Peggy I 0 2 2 3 10 10 2 5.03 1.05 II 0 3 3 2 5 13 4 5.13 1.53 Bill I 1 4 3 4 2 4 1 3.62 1.45 II 2 2 4 1 11 5 5 4.73 1.73 Bob I 2 3 3 4 5 9 3 4.59 1.77 II 0 4 2 3 2 14 5 5.17 1.63 Lynn I 12 11 1 2 l 1 1 2.17 1.56 II 23 5 0 2 0 0 0 1.37 .79 Terry I 6 4 3 7 3 5 1 3.55 1.85 II 12 5 2 8 2 l 0 2.53 1.56 TABLE C.1—-Continued 223 e. Scale Position Item 8 - 5 1 2 3 4 5 6 7 Mean SD Tired-Energetic Peggy I 0 2 0 0 3 13 11 6.00 1.01 11 0 0 0 0 0 12 18 6.60 .49 Bill I 0 2 13 6 7 1 0 3.72 1.01 II 0 4 8 8 7 3 0 3.90 1.19 Bob I 0 0 1 2 9 15 2 5.52 .85 II 0 1 3 7 6 9 4 5.03 1.38 Lynn I 0 2 3 11 6 5 2 4.52 1.28 II 1 4 7 16 1 1 0 2.83 1.20 1 2 3 6 7 10 0 4.59 1.40 Terry 11 3 1 4 12 6 4 0 3.97 1.38 Passive-Active Peggy I 0 1 0 0 3 15 10 6.10 1.26 11 0 2 1 .1 0 10 16 6.10 1.42 Bill I 0 7 11 1 7 3 0 3.59 1.35 II 0 0 11 1 5 3 0 3.33 1.35 Bob I 0 2 0 0 1 9 7 5.59 1.25 11 0 1 5 3 6 10 5 5.13 1.43 Lynn I 6 2 7 3 6 4 1 3.59 1.58 II 11 9 l 7 1 1 0 2.37 1.43 Terry I 2 1 1 3 9 l3 0 4.90 1.44 II 8 7 1 4 4 6 0 3.23 1.93 Slow-Fast Peggy I 0 0 0 4 5 11 9 5.86 1.01 II 1 0 1 2 2 15 9 5.83 1.32 Bill I 1 6 12 6 3 1 0 3.48 1.16 II 0 8 13 4 3 2 0 3.27 1.17 Bob I 1 1 7 2 8 7 3 4.66 1.56 II 0 2 8 5 9 4 2 4.37 1.36 Lynn I 1 8 4 5 5 5 1 3.83 1.64 II 7 7 S 9 0 2 0 2.80 1.42 Terry I 0 3 2 11 7 6 0 4.38 1.18 II 6 7 2 9 4 2 0 3.13 1.57 224 TABLE C.l--Continued 7' 9‘ Scale Position 3 Item 0 :3 1 2 3 4 5 6 7 Mean Defensive-Aggressive Peggy I 0 5 5 7 2 0 3.97 I 1 4 1 6 3 2 3.77 Bill I 0 1 3 6 6 0 4.69 II 0 2 8 5 6 0 4.30 Bob I 4 11 4 1 2 2 3.21 11 3 13 7 0 3 1 3.00 Lynn I 1 3 1 4 9 10 5.34 11 0 3 0 6 4 12 5.43 Terry 1 0 3 7 8 7 l 4.24 II 1 2 3 7 11 4 4.87 Static:Dynamic Peggy I 0 0 2 4 6 5 1 5.41 II 0 1 2 1 10 9 7 5.50 Bill I 0 9 9 7 2 2 0 3.28 II 1 3 3 6 4 3 0 3.60 Bob I 0 l 1 2 13 1 1 5.34 II 0 2 2 4 ll 9 2 4.97 Lynn I 5 3 9 4 5 2 1 3.38 11 9 7 4 8 1 1 0 2.60 Terry I 1 1 7 9 7 4 0 4.10 11 2 10 3 11 2 2 0 3.23 Unhapny-Happy Peggy I 1 1 4 0 8 3 4.97 II 0 1 2 8 1 4 5.20 Bill I 1 5 9 3 1 0 3.38 II 0 2 7 4 2 2 3.90 Bob I 0 1 6 8 3 1 4.17 II 0 4 7 7 5 5 2 4.20 Lynn I 7 6 6 9 l 0 0 2.62 11 4 8 4 11 3 0 0 3.03 Terry I 1 4 9 9 3 2 0 3.41 II 3 7 7 11 2 0 0 3.07 r—Ir—I Hl—l HH r—u—Ir—I r—ar-a t—lr-a t—H—I l—H—IH r—H—I l—‘H r—Ir—I r—w—I t—‘b—J .22 .50 .97 .24 .82 .63 .83 .62 .41 63 .99 .26 .17 .22 01 .28 .67 .40 .21 .33 .40 .33 .10 .30 23 .47 .23 .25 .17 .17 TABLE C.1--Continugd 225 g5 Scale Position Item 0 (S 1. 2 3 4 5 6 7 Mean SD Constrained—Free Peggy I 0 1 4 3 7 12 2 5.07 1.29 II 1 2 5 1 7 10 4 4.90 1.51 Bill I 4 6 12 3 2 1 1 3.00 1.36 11 2 7 9 1 3 7 1 3.70 1.76 Bob I 0 6 7 1 4 9 2 4.62 1.75 II 0 5 7 1 7 6 4 4.47 1.71 Lynn I 12 11 2 2 0 2 0 2.07 1.37 II 17 3 4 5 0 0 1 2.07 1.48 Terry I 2 8 8 7 2 0 2 3.24 1.46 11 12 5 6 3 0 3 ‘1 2.57 1.76 Excited-Relaxed Peggy I 8 15 5 0 1 0 0 2.00 .48 II 9 10 6 0 2 2 1 2.53 1.65 Bill I 0 3 4 7 9 5 1 4.41 1.30 II 0 2 3 4 7 11 3 5.03 1.38 Bob I 0 1 10 2 8 8 0 4.41 1.30 II 0 2 3 5 8 8 4 4.97 1.40 Lynn I 3 5 5 8 5 2 1 3.59 1.54 II 1 3 6 13 1 3 3 4.03 1.49 Terry I 0 3 12 7 4 1 2 3.79 1.33 II 2 8 10 2 1 4 3 3.53 1.80 Upset-Calm Peggy I 1 9 12 3 4 0 0 3.00 1.05 11 3 7 12 3 2 3 0 3.10 1.37 Bill I 1 3 11 0 8 6 0 4.00 1.49 11 0 3 7 2 6 7 5 4.73 1.65 Bob I 0 3 6 1 2 5 2 4.55 1.43 II 0 3 5 0 7 8 7 5.10 1.55 Lynn I 8 1 10 4 5 1 0 3.00 1.51 II 4 7 3 5 2 5 4 3.83 2.03 Terry I 0 7 14 1 4 1 2 3.45 1.43 11 4 6 11 0 3 5 1 3.37 1.74 TABLE C.1--Continued 226 3‘ Scale Position Item 0 (S 1 2 3 4 5 6 7 Mean SD Embarrassed-Smug Peggy I 0 0 2 7 15 5 0 4.79 .92 II 0 0 5 9 14 2 0 4.10 .91 Bill I 0 0 3 12 8 6 0 4.59 .93 II 0 0 2 20 4 4 0 4.33 .79 Bob I 0 2 7 11 7 2 0 4.00 .95 II 0 l 7 16 5 0 1 4.63 1.13 Lynn I 1 8 5 6 4 2 3 3.76 1.69 II 7 4 5 7 3 2 2 3.30 1.81 Terry I 0 2 6 10 8 2 l 4.17 1.15 II 2 7 6 5 2 4 4 3.87 1.88 Uncontrolled-controlle Peggy I 0 4 7 3 12 3 0 4.10 1.27 II 1 1 9 1 7 8 3 4.60 1.60 Bill I 0 0 2 2 9 12 4 5.48 1.04 II 0 l 2 0 8 17 2 5.23 1.53 Bob I 0 0 0 2 7 15 5 5.79 .92 II 0 0 2 l 5 13 9 5.87 1.42 Lynn 1 0 1 2 1 6 12 7 5.62 1.27 II 4 0 0 7 4 8 7 4.97 1.89 Terry I 0 1 8 2 6 9 3 4.79 1.46 11 0 2 4 3 8 11 2 4.94 1.36 Fearful-Hopeful Peggy I 0 2 3 2 8 13 1 5.03 1.28 11 1 6 8 1 6 4 4 4.10 1.66 Bill I 0 0 5 3 9 9 3 5.07 1.24 11 4 4 5 0 5 10 2 4.20 1.99 Bob I 3 2 6 3 1 4 0 2.97 1.45 11 1 9 3 2 2 3 0 2.47 1.65 Lynn I 0 0 2 12 5 5 5 4.97 1.25 11 0 0 2 13 6 5 4 4.87 1.18 Terry I 1 2 7 9 5 5 0 4.03 1.30 11 0 2 0 13 8 4 3 4.70 1.22 TABLE C.-1--Continued 227 9. Scale Position 3 Item 0 :3 1 2 3 4 5 6 7 Mean SD Feminine-Masculine Peggy I 9 18 0 1 0 1 0 1.90 1.00 II 16 13 1 0 0 0 0 1.50 .57 Bill I 0 0 2 1 13 10 3 5.38 .96 II 0 0 0 0 6 16 8 6.03 .68 Bob I 0 0 0 1 5 17 6 5.97 .71 II 0 0 0 0 1 12 17 6.53 .76 Lynn I 3 7 8 4 4 3 0 3.28 1.48 11 3 15 8 1 3 0 0 2.53 1.05 Terry I 0 0 1 2 6 12 8 5.83 1.02 11 0 0 0 1 6 13 10 6.03 .81 Simple-Complex Peggy I 0 1 6 2 5 14 1 4.97 1.35 II 0 2 4 1 14 9 0 4.80 1.20 Bill I 0 l 6 1 15 5 1 4.69 1.18 II 0 4 5 3 14 3 1 4.33 1.32 Bob I 0 1 2 1 11 11 3 5.31 1.06 11 0 0 2 3 9 l4 2 5.37 .98 Lynn I 0 1 2 10 4 9 3 4.93 1.28 II 0 1 3 ll 5 3 7 4.90 1.45 Terry I 0 0 6 10 8 5 0 4.41 .98 II 0 3 2 13 6 4 2 4.40 1.28 Weak—Strong Peggy I 0 5 9 7 6 2 0 3.69 1.18 II 1 0 8 3 8 8 2 4.63 1.45 Bill I 0 4 12 2 9 2 0 3.76 1.22 II 1 2 13 3 4 6 1 3.97 1.49 Bob I 0 2 2 0 10 9 6 5.38 1.37 11 0 1 2 3 5 14 5 5.47 1.26 Lynn I l 1 5 3 7 9 3 4.83 1.53 II 3 2 2 11 6 4 2 4.17 1.59 Terry I 0 0 2 7 12 6 2 ' 4.97 .97 II 0 1 6 5 8 8 2 4.73 1.30 228 TABLE C.1--Continued g. Scale Position Item 2 U 1 2 3 4 5 6 7 Mean SD Bored-Surprised Peggy I 0 0 2 22 5 0 0 4.10 .48 II 0 0 4 19 5 2 0 4.17 .73 Bill I 1 4 5 16 3 0 0 2.86 1.19 II 0 2 5 15 6 2 0 4.03 .95 Bob I 0 0 3 20 3 3 0 4.21 .76 II 0 0 4 17 5 3 1 4.33 .95 Lynn I 0 1 0 9 11 5 3 4.97 1.10 II 2 3 l 3 4 9 8 5.10 1.89 Terry I 0 1 2 16 7 2 1 3.66 1.18 11 0 4 4 7 5 5 5 4.60 1.62 Mild-Intense Peggy I 0 9 3 2 13 2 0 3.86 1.43 11 2 2 4 6 13 3 0 4.17 1.35 Bill I 1 6 12 3 4 3 0 3.41 1.07 11 2 12 9 4 1 2 0 2.87 1.23 Bob I 0 1 5 4 12 4 3 4.76 1.28 11 3 5 3 1 12 6 0 4.07 1.69 Lynn I 0 2 2 3 14 5 3 4.97 1.26 11 3 2 4 3 7 6 5 4.57 1.87 Terry I 1 3 8 8 5 4 0 3.86 1.30 11 0 4 11 4 3 6 2 4.07 1.55 Yielding-Tenacious Peggy I 1 1 3 3 11 9 1 4.83 1.40 II 0 2 1 2 9 13 3 5.30 1.19 Bill I 1 0 2 6 13 6 1 4.79 1.13 II 0 2 5 5 11 7 0 4.53 1.20 Bob I 3 11 9 1 3 2 0 2.86 1.36 II 3 12 8 1 2 3 1 3.00 1.58 Lynn I 9 12 3 4 0 1 0 2.21 1.21 II 12 7 5 6 0 0 0 2.17 1.16 Terry I 2 6 8 7 3 3 0 3.41 1.38 II 7 8 3 11 0 1 0 2.73 1.34 APPENDIX D PERSON PERCEPTION TEST TEST-RETEST ITEM RELIABILITY ESTIMATES 229 TABLE D.1--Person Perception Test test-retest item 230 reliability estimates for Group II by factor. Client-Percepts Scale Peggy Bill Bob Lynn Terry Evaluative Factor closed-open .24 .20 .32 .78 .54 dishonest-honest .69 -.07 .71 .54 .23 superficial profound .49 .28 .31 .52 .70 angry-supportive .34 .47 .62 .41 .72 perplexed-understanding .53 .40 .63 .55 .32 disagreeing-agreeing .29 .32 .42 .13 .57 disapproving-approving .42 .77 .64 .27 .68 harrassed-unhurried .13 -.07 .53 .53 .65 impatient-patient .65 .40 .54 .59 .44 dissatisfied-satisfied .51 .23 .60 41 .42 Emotional Factor excited-relaxed .26 .28 .44 .36 .41 upset-calm .61 .52 .79 .44 .60 embarrassed-smug .49 .57 -.02 .62 .48 uncontrolled-controlled .25 .44 .44 .19 .58 fearful-hOpeful .40 .37 .49 .20 24 feminine-masculine .35 .45 58 .38 57 Potency Factor simple-complex .27 .52 .28 .60 .33 weak-strong .46 .47 .58 .45 .58 bored-surprised .24 .64 .58 .16 .67 mild-intense .37 .59 .52 .77 .38 Evaluative-Activity Factor Opaque-transparent .37 .27 .61 .33 .22 far-near .53 .52 .67 .36 .40 tired-energetic .08 .56 .68 .10 .57 slow-fast .29 .52 .45 .34 .62 defensive-aggressive .55 .60 .50 .53 .37 passive—active .23 .63 .52 .54 .57 static-dynamic .68 .28 .53 .32 .17 unhappy-happy .72 .61 .77 .72 .52 constrained-free .47 .49 .75 -.01 .35 Unnamed Factor yield-tenacious -.02 .39 .54 .34 .41 APPENDIX E COUNSELOR RESPONSE SYSTEM 231 1’) 232 O COUNSELOR RESPONSE SYSTEM 0 of the EF‘VIOR INTERACTION DESCRIPTION SYSTEM by Richard C. Rank . and William M. De Roo MICHIGAN STATE UNIVERSITY Rev. 4/65 t‘“ 233 '.ha Counselor Res sponse Sysc en (CRA) is a method for analyzing verbal responses (or statements) of counselors duiizg counseling inte- viei' This - '__-..- ,1 ° -.-.....-'~..' '.. . h ace of use with a high degree of r o sensitivity to theoretically relevant aspcc s of counselor behavior. The A F? system is designed to describe, but not evaluate, counselor respon as. (0 Each counselor statement is rated on six dichotomous dimensio.s: l. Affective - Cognitive Contert 22 Affective Cognitive Change 3. Content - Follow - Shift Present vs. Past or Future -!"-‘ Q 0 Evaluation of each statement involves maP1n_ six dic Y notomous judgemer ts, (IQ one for each dimension*. With this system, a counselor response could have Sions at one time. Judges need only to be familiar with counseling practice The six dimensions do not pr6vide a complete description of all theo— retically relevant dimensions, but ratner are highly relevant to the coun- g process, and are amenable to objec2ive description. They have been derived from counseling theory, but not exclusively from any single theory. 0 attempt has been made to determine which response characteristics are ”good” or ”bad," "effective" or "inef fective." Theoretical and research “,ia-Jv a- v d d r... “31' ' ' n- .1” . .c.atule Have not as yet prov1 e adequate e-ines for judging' 'goou 6: “*ad" responses. *See attached ratinr ~ -—\ U sheets. I "fferent descriptive profiles. One person can adequately judge two dimen- 234 EEEEEIQION AND DESCRIPZICNS OF T“" DIJEXSIOZ\ S I r‘ a I h . b O I O -\ n v x '4‘ .- — . ' ‘ - CCClVC - Cognitixe Ce .eht Dlrsn;13n This dimension indicates thetl er or no: client expression of affect of .excrcnce to affect is present in a counselor response. The presence absence of affective content is denoted by the cognitive category The categories are more explicitly defined as follows: t I _.‘: a. : lective Respm ses An affective response is one in which e counselo deals a-:ect1y o: reflecti.g client expression of mood, feeling or emotion, or by O calling attention to or remarking about mood, feeling, or emotion on the part of the client or anyone els=. Note: Counselor ex- : emotion are considered to '(1 H (1 U) 0) ’1 O :1 (I) O f‘h {1‘ VJ: (O O s P wn mood, ieeling, be ““nctive r“ "‘n“es “s are stnt rents "bo‘t mood feeling or t. 6-1.:— c. -b-J Junb , a it. tact...t..lt. c. b In , t. “o L emotion on the part of any person as related by either the cli nt ve response must rare: to or incorporate an expres- h~ :1 (‘1 H] Fh O H- sion of effect. It is the presence of affective content that is of importance and not tne level or eelinn evidenced by the response. Verticular care should be used when judging responses contain- the verb "to feel". Some ecunselors indiscriminantly use tnis word in reference to opinions rather than true feeling. Only when fee- is used to refer to true feelinr mood, or emotion, should ' is meant the response be categorized as eling' strong feelings. Me re lines or dislikes are not strong feelings, and responses dealing watn tne. are not considered affective. 235 Encriles: r 1. "That seems to mate you angry." 2. 'You seem Jery nappy today.” ”7' :i ‘ ,- ' 0 _ -, 3. now do you let. w.en they ignore you?" ' ".T' -‘ H ' w 0 5A... q ' J . t- 4. it annoys me when you arrive late for your appointment." “ o u 1 ‘1 . v I\-. 5. ”ULC that mane your parents nappy?" \ \v »\ " n .. ' '7 \ s. cognitive nesponses response is any statement or question which does a cognitive not refer to or incorporate expressions of feeling, mood, or. motion on part of the client or anyone else. Cognitive responses often material or content, but may be found to follow client if the counselor does not'deal affect. n a general sense) H- un}. o *- ‘ 1 ‘- v‘ ' ‘A’ (.hL ltbbk—JLAULDG "How are you today?” "What do you think about your grades in'Mathematics?" V" :3 bob rr 0. ‘- -\ cab r3" 3. "You did quite well on "So you feel you should loo? rore seriously at teaching 1 . i 3 s ole career. H as a po m “A - 1 'Vv-s: : ”LA .rm -—'f\ eetiwe - COgnitlve change Dimens-on . This dimension deals with gross changes in feeling level between a counselor response and the precedin client statement. More specifically, if the client's statement was primarily coenitive, does the counselor follow with a response that is also largely at the cognitive level of feel- ing, or does he change to a more affective feeling level? And if the ' O client's statement was ;rimarily affective, does the counselor follow at to a more cognitive level? 236 This direnslon, while sone .at rel ted o the Affective-Cognitive Content dinenSion, does not deal so teen with e"pressions of, and refer- ences to affect, as it does with differ noes in the ge ene ral feeling level betw ween client and counselor Statements. For example, it is possible for the counselor to refer to client affect without really re5pondin3 at the cane feeling level; it is also possible to deal with strong client affect in a non-emotior al, objective ranner and still remain at the client's level of feeling. , General consistency in fe lin. level between cl'ent affect and'COunse- lor responses is denoted by the "followi ng" category, and gross dif- ferences in f ling level are dero ed by the "chang ing" ategory. Kore explicit efinition follow: A. Following Responses A following response is one in which the counselor respor ds at the same, or nearly tne sane, feeling level as that of the client's prev; 'ous st tement. A response at an affective level to an affective statement is a following response, as is a re- sponse at a cognitive level t a cognitive statement. Examples 1. Cl: ”Every time he says that, I could just sit down and bawlf" (arrecrive st:tement) Co: ”It really makes you eel worthl ess.‘ (An a ffe et- ive Stuteient: if ent counselor responds with the some level of feeling, this would be a following response) 2. Cl: "I just wondered i; you had any tests I could tabe to see if I next semester." your should try a tougher English course (Cognitive statement) eral tests which might help you, but in Freshcen Englis is probablyt he 237 r O 1 J. COUESC .n'w ebb q h t . I -- P‘ sod-V ‘J .0 'tive level a cogni ‘— 3L resionse '5 G. O A la _b T e m Cement. Sta C O n‘les H n La ~v- -« . ‘J ¢\h.‘ d t down an just Si ‘4 Lu u ... —‘ .a ‘- ‘. A‘L—v..-) e ith him?" to talk‘it over a» C {L g . "nees “An: .. u 3 C ‘I ‘0 1 LS V3-5. OCC' 0 103 3038 di -L‘ls h', C ollowlth C c- 0 f" - -— LL. to a d scussion 0 di .1 .1 Or a.» o. .l C. «5 {selor A L. Q tflC COD one in wlich is ’5 \. nt's prev , LC e cl' ‘G AL C 88 t opi 5. L. L ne 4 a —- A ‘~.‘ I deals :A counselor ma ‘5‘; Pf. ~v v~ A.» AAV n -Atb "foww dered to be _I ‘- CONS C 2 t *1. C :L 3. b.» 238 Exa..ples: .' . g'T fl-‘v1n. ‘5 *-.‘~-s I‘ ‘ 4‘ 1 I ., ‘.""’ 1. Cl. l aLWuys Seen LO co poorly on filst ory tests." AA, 39“.... in ,. v. w,‘ 1. '5 II oo. tuna: was your g-aoe on the last one9 1 ‘h' rfih1_ n -— 'A‘ a J 1 .-. n r _ o " 2. Cl: My lather says . s.oulu be an A.gineer. . A "V'A ‘1‘. . ‘— v ca 1:...”4“ O‘v‘m - 'A . 1‘ I Co: now does it mane you lee- when ne tries to tell you 7 to do some hir.g you don't want to do?” it Shifting Responses 'k) ;. To A topic shifting response is one in which the general topic 5 _u “l-l'-a!- . “A : ‘:::fm-l 5“ z - - or the counselor s response is dillelent llom that or the precedin client statement. Included in this category are counselor responses 0) n m rt (D m r? I'" in. r E D '— H O A rt. r34 0 U) U) I) P- H} H (u {‘4 r1 0 C.) m f" 'h 'I] 0 ’1 m 5 [’1' r? 0 U *4. O H. 3 ff :3‘ m M :5 rt (1) '4 m S H0 5 00 h 'I I '--'-4--.v\ ‘A~ .-,Af\~"'\ . ’nb " 1. cl: I ve been ,ettlng low Ilaees in hath. 0 Co: ”How are your in inelish?” (Note: this would be a ”following" response if there had been a discussion of grades ress in.Mathematics has A I . I ‘ m (D {J 0 H m *4 r1- 0 U u (‘1' :3‘ r- u 0) H~ (D I!) (O .V H. fl) FT Ho D 09 H m U) 'U 9 d O v ' to ner about your grades?" orget, I want to as“ you if I could take one P' 3] Cl: "Belore I of tnose interest tests.” Co: "You were saying you thought you might talk to Miss Jones DJ. ”1* Q ‘1“ V:r‘ m .- auOdt YOuL histo_y gr aees... v H‘ at, 1 3“. ' n A-' 7. l u lV. -ne Co tlol BluefiSlOn (aestsietive - ixpanSiJe) ent to which the counselor limi"s (D This dimension deals with the or permits freedom of expression by the client. It should be noted that the counse -lor can focus on specific 3 and still permit the client to express . "r‘ r) 5" .-.-.- 'uvgnbbe-V- 4 t7 4‘“.J o -4...‘ ‘. 7‘ . AL. '1 Ct ; A. set ., UCLC v-‘m . l. l r-v‘ '1‘. l:‘.*.' {IA-h... ‘- q eely. PT. I- n .L— W A - p< z. .A LII-SK: A - O .C t r e n 0 w L H a t Lu .1 Us 3 r A...» . .1 .. . I 3 3 So m :1 s w .t s e r .n r 1 S r a 1 L C C C .l S . O 3 .. v .8 3 O C I t O 3. a . S n .1 C t a u t C t m ? 8 h o. t f e - f t .1 r no 0 d S g m 0 ; .3 a n r u n D. 1 O U. C 9. O O n a e O y l d e n s v. u a l h e v l t e e . 3 S . O C C C S .l e l. t C n A e S y S f 3 t V U r a r .1 n n O n a C C u f I D. h C u C .1 .C nu .1 O . x t v 0 V e l n f y S C d e n C e r C C O f I h C I O a e t t .1 d t t .1 O a .C e 2 .I d 1 n :1 t :1 .n d S a ’ U 8 h .1 y t in C m e r O .1 C C r O S O S . V. e a C l .1 C O S C an O. n O O r l c .n o. .c c s o 1. so u W S 3 h a .1 e D. S 3.. a C . e d S r .n r s .1 1 n S i r h n r n i 8 w e .n c i t e t .1 O a .. 1 O I .G r .n o. C i n t e s .. a C d C d x a.“ .. .. 1 .n . 1 . e r 3. O n 3 e e 3 Q m C o S O r a .o t 1. a 3 O u . 1 .1 no .0 1... .2 .1 r . n a d 3 U. Q .. n K t 1m 0 . :1 C O C I 1.. C .l r C .1 a in .c C 5 M 1 V C e «.1 C C no t . 1 . C no a 7g 5 I r .. o .1 1 C O I a a... n O a V. O a o C ._ 1 C .L 11 _. .1 r .C I :1 D. 8 O 8 O t s w o. e a O o t s t r y C . C C K V t t E C .e .1 s I C a n s e e .3 c V. r n I R I 3 C ... C .1 C C . u 0,. t C 2w 0 I W S T .1 .. l r I u G ....c be S ”J. r FL nu ml. 50 :A C S 5‘ AU 0 VJ I.— .. C S 0 VJ O C .1 C b y r S C S 1 t y l S o. .C C C m S t LL. .1 c F. I .l R . 1 C S C .R .1 .1 I d t e U C O 8 a S a. «n U . t t u .1 . 1 0 Cu 3. 1. Q S 3; L1. 1.; 1. u vL A- U S .6. 5L 5L AG w wed n s n e l . r o i w r n S a O C .1 S C C 3. 7o «n a C O .. .3. no 0. R t c n a u .. .n . 1 S .1 O u n e S C C C a.“ O u .1 a n . 1 l . 1 O O r e 3 . l o. .n W” .4 B. .e o a c e v... 1 o. r 1 u 1 1 5 t .. .. 1 C . V n a r .1 _. .. .. ,r o a A“ S .0 m e _.. .. e t . e r n s n l e u n e u n d s I.“ H C i» «C a A. o . ._- vA .l C Au 1A. «L AC 0 C 7. .1 R t V y l 11 2 .1 E l O t O l 11 :1 r n .1 . 1 3. 3 C :1 «C a . C 0.. S 1 1 t e 3 C m a .. a C 1“ AC 5.» AD I; C AG D. e y 3. AU 4* LL 3 O S C l S a 1 x n .P. a. n A. . .m“ 0. v d m R C .1 l. 1 2 E t . C .L E l E :L t 3 Lu 3b . c o S S A B n 3 O a A). w. Cu ”J. your- 5V o—A OVC‘. ‘4. L D d d that just to .11] \JHAKA’ C t’1~‘:’ ‘A yQU. TCH11J q .a—poo bookb «Jory, as to is Pr e ‘- L. fi 1 , J. S W8 A a. 3.8 into one ca utur I: ence of the counselor's J- hich the most emphas .1. -A ‘9‘ 1r... w .- V 240 to 'ni .1 a. H; ”A q formed oy co ,- LS mph e primary a th 1 ‘lCJ .. ct V “SCS 13 W 830 b e 3 tion, or feeling. condi 5 a U. a at 2 live h a pas on you ‘1‘? Q o s. 5 1 10 focuses & q WUC CIGRCQ .C.‘ A. at you no longe th re (Note that alt‘ W1 1" L A ure event, he L. -g to college do “w t in fu u a .L v ‘n 1 F‘.’ 00 VII. J. :1 m (r ‘0 ' .5— 1; q 1 sf» "1 I-l-i Kr“. ‘— DOUC d..\—(.b y :‘1 ‘s’LlLC ere you w” fe . o ,. G q 0 1 0" I“ b ‘QL.\ L -A m A. a' l ‘1 c 5 lee \ c T d q W 01 IBSUORSCJ t do you J co you U a (no. -‘ fat. "v_./‘\,v -gU'U counselor b ”~71 ”(we ese event, condition, 0 . g A, ‘QL 'fi fit C. /~ on 241 VI. or with another 8- ient cl ient the.cl ht by noug J— L 0: ,— ' 1’- Ant ether or not t Q 3 W8 t1 5 W1 .- :A ,1 A7 ll‘1k/Il neg; me 3“ -1 -, -1 h L. ,— '\".. . y..\.~\.—.. ”I ‘.\ r7 "‘ 9 CO errin. spenses ref ences of the c f-reler sel ,7 flu 1a hts, oug fl AL h’ L. in any way focus upon the , -rr UGISOH. L OLL‘ 3-1. kebbh . w do you U 'IV:,., .. 9 It other 1 eacn en you c .5 l w' ”An ICC now do you tt~~ friends don't listed - I. 1 er you wnen you Q 1" 58 'A ‘- 40C to you." ~oa A ‘A' L- a bK—Atcc‘n L— .Q .L. .1. Sea or her persons as well to n~ ab .: V tne ermi +— I— de no the I C \8030 A. .1.‘ A .IA soLAuLC.;LL) 0 non“' ‘ .5. L. CICRCES 1.1 such reference ‘4: l1 category “.6— LA. .> -5 ’— .13-fl- ._, 0... other 1.. 8 :1 «1 C at S o . A S oes your .C 40W IF about you?" cglw 3.96; now does your 0 .‘V L. ' a-thou; ) ‘v4- 455-. _ OI (:0th ° out 1“ A 55*- b*~‘-b he chased t our dog ind . H“ ._ fit. LL . 4 I q 243 Vf q. §‘\ fin..- ,‘: r can be .- -‘- Us 0 V .v c». .N n 2:: m L‘J—H .AAN—oh-‘b A 3 A-yv‘j'“. (— LVV....J\..~ V r'flfl "T“ A. m (3:33), ‘7'”. Un— Sy”‘ ion script erac;- I ,- ‘T'.,. A— ioral int DCUCV t1"; Ci ‘»fl “‘5‘ ’ P \- w. 1 V; c . l .n T fl CCU” tn V DO” 7 APPENDIX F COUNSELOR RESPONSE SYSTEM: RATING SHEET 244 24S CRS RATING SHEET d O N N a 5m "LTII-II fini'lfl'l II".I'II.IIIIII rlgl'1-nll4 'lu' III-'A-"Ill'nl fi'l-nfl'l I‘ll-I Inlnll'nl'llllul. rIIIu x .\W p \I\ -11. S O H mm“ m all-l-" 1"..-Illllk-nl3"...vl-nllul'nl.fi"l..l" II... III" IN" .llp'an-l-lnl" u" .lI-IrII-II 1'- m e T. s an - 0 3 m N Iu-IIIrI-Ilu lll'I.III-.llllvll.rlulul nlllI-l'- Ilnll III!- '1'!" III” ”lull.lnlnln.nllllr'lu I'I-l .Inlllllll loll- S m .v... w m I 2 m 'lull" III-ll.I'I]".Ilulllillul.l.l..l.w|'l 'llulll-lnwlu'l. I.-I!ll.nll.'lll.ll-II u“ .lI-l-T'nl l'- n s E E V. w m 1 m nan}... .i!. I....1...!.....- 1133341...- 3.... Tn...1........i..:..!. I............:..!. stun-.. in S m. m E S w 1 2 3 4 S 6 7 8 9 0 1 2 3 4 5 6 7 8 9 O “a. 1 l l 1 1 1 1 1 1 1 2 136w APPENDIX G COUNSELOR RESPONSE SYSTEM: FREQUENCIES BY SUBJECTS 246 247 TABLE G.1--Frequency scores for the Counselor Response System dimensions for Group I. Counselor Response System Dimensions Subject Follow EXpansive Client/Focus Reinforcing 1 15 3 17 2 2 *1: *5: ** *9: 3 ** *9: *a'c ** 4 12 1 14 l 5 12 8 16 O 6 ** ink ** *1’: 7 19 6 18 2 8 12 5 18 O 9 11 3 7 3 10 9 3 16 1 11 15 5 12 l 12 fink ** *7: ** 14 12 3 14 1 15 12 3 13 l 16 17 1 17 l 17 10 6 12 0 18 12 6 16 1 19 19 6 20 1 20 17 3 15 1 21 7 l 8 1 22 *1: ink *1: *4: 23 6 1 9 1 24 11 6 10 0 25 8 1 12 O 26 4 3 6 3 27 ll 5 6 1 28 9 5 15 2 29 12 5 9 1 30 18 0 14 1 Mean 12.08 3.71 13.08 1.08 **No tape available for this subject. 248 TABLE G.2--Frequency scores for the Counselor Response System dimensions for GrOUp II. Counselor Response System Dimensions Subject Follow Expansive Client/Focus Reinforcing 1 12 8 18 1 2 10 3 9 2 3 11 3 15 O 4 *2? *9: *9: ** 5 2 O 2 O 6 15 4 10 2 7 16 5 17 O 8 7 3 9 O 9 19 3 19 2 10 *8: *3! *1: *3? 11 13 1 9 2 12 15 4 10 1 13 10 3 16 1 14 8 2 12 0 15 13 3 15 1 16 14 5 13 0 17 8 5 14 0 18 5 7 16 0 19 15 3 16 O 20 ** *5: *9: *9? 21 12 5 9 1 22 12 3 11 O 23 7 4 12 0 24 16 5 18 0 25 7 3 S 1 26 9 4 10 O 27 11 5 13 0 28 12 7 12 1 29 *2! *1: 22* ink 30 14 5 16 0 Mean 11.27 3.96 12.54 0.58 **No tape available for this subject. APPENDIX H COUNSELOR RESPONSE SYSTEM: DIMENSION FREQUENCIES 249 250 distribution of TABLE H.1--Counselor Response System: responses for both groups. Control: Referent: Expansive Content: Follow Reinforcing Client Group 000000000000000000000000000000000020344755 1 11. 100122213422311233011324101020010000010000 00000000000000000000000011025057047901511.11 002100200223020274321221221310011000000100 13.08 12 54 71 .96 11.67 11.27 Mean APPENDIX I ITEM/TOTAL FACTOR SCORE CORRELATIONS ("l-7" PROCEDURE) 251 252 TABLE I.1--Item/total factor score correlations using the "1-7" procedure for both groups: Evaluative factor. m Client-Percepts Scales 7 Group Peggy Bill Bob Lynn Terry closed-open I .40 .18 .59 .43 .44 II .38 .68 .42 -.17 .07 dishonest-honest I .57 .17 .65 .74 .49 II .33 .49 .49 .06 .40 superficial-profound I .38 -.10 .52 .50 .52 II .16 .48 .54 .09 .10 angry-supportive I .33 .18 .46 .69 .58 II .53 .34 .61 .20 .40 perplexed-understanding I .32 .52 .67 .63 .53 II .27 .50 .59 .28 .10 disagreeing-agreeing I .56 .44 .56 .62 .47 II .16 .47 .55 .37 .20 disapproving-approving I .48 .36 .51 .62 .49 - II .49 .46 .54 .29 .39 harrassed-unhurried I .35 .17 .59 .45 .41 II .02 .45 .55 .61 .50 impatient-patient I .36 .30 .53 .57 .49 II .34 .36 .36 .36 .37 dissatisfied-satisfied I .64 .43 .66 .70 .69 II .37 .60 .34 .24 .22 253 TABLE I.2--Item/total factor score correlations using the "1—7” procedure for both groups: Emotional factor. Client-Percepts Scales Group Peggy Bill Bob Lynn Terry excited-relaxed I .46 .43 .45 .40 .55 II .44 .37 .64 .42 .66 upset-calm I .54 .52 .48 .39 .38 II .65 .59 .57 .51 .68 embarrassed-smug I .34 .38 .44 .16 .23 II .26 .47 -.06 .07 .19 uncontrolled-controlled I .64 .11 .43 .42 .35 II .18 .24 .35 .36 .25 fearful—hopeful I .43 .49 .50 .53 .66 II .31 .45 .41 .37 .63 feminine-masculine I -.19 .04 .27 -.21 .49 II .00 .11 ~.15 .10 .23 TABLE I.3-~Item/total factor score correlations using the “1—7” procedure for both groups: Potency factor. Client-Percepts Scales Group Peggy Bill Bob Lynn Terry simple-complex I .42 .61 .21 .10 .06 II .27 .34 .25 .45 .43 weak-strong I .22 .43 .35 .26 .32 II .38 .29 .28 .24 .32 bored-surprised I .01 .33 .08 .17 .33 II .49 .35 .26 .02 .10 mild-intense I .27 .02 .04 .22 .17 II .62 .70 .55 .19 .16 254 TABLE 1.4--Item/tota1 factor score correlations using the "1-7” procedure for both groups: factor. Evaluative-activity Client-Percepts Scales Group Peggy Bill Bob Lynn Terry opaque-transparent I -.01 .10 .28 .27 .58 II .03 .50 .24 .23 .28 far-near I .36 .28 .60 .47 .59 II .07 .34 .53 .27 .35 tired-energetic I .49 .30 .44 .48 .56 II .12 .36 .48 .32 .29 slow-fast I .58 .41 .49 .31 .40 II .18 .36 .51 .11 .28 defensive-aggressive I .59 .13 .32 .60 .70 II .01 .25 .58 .07 .30 passive-active I .36 .30 .39 .51 .48 II .17 .44 .52 .17 .45 static-dynamic I .52 .00 .34 .59 .41 II -.08 .37 .61 -.09 .23 unhappy-happy I .20 .40 .56 .29 .47 II -.02 .19 .40 .44 .37 constrained-free I .18 .21 .32 .43 .53 II -.07 .41 .36 .34 .43 APPENDIX J PERSON PERCEPTION TEST-COUNSELOR RESPONSE SYSTEM: COMPARISONS ("l-7" PROCEDURE) 255 256 TABLE J.1--Person Perception Test total factor scores (“l-7” procedure) compared with Counselor Response System dimensions by product-moment (r). Counselor Response System Person Perception Test Group I Group II Emotional Factor Content: Follow -.02 -.16 Control: Expansive -.04 -.22 Referrent: Client -.10 -.14 Reinforcing -.20 -.23 Potency Factor Content: Follow .03 .10 Control: Expansive -.28 ~.06 Referrent: Client -,10 —,04 Reinforcing .24 .34 Evaluative-Activity Factor Content: Follow -.06 -.29 Control: EXpansive -.09 -.06 Referrent: Client —.04 —.17 Reinforcing .20 -.28 APPENDIX K PERSON PERCEPTION TEST-COUNSELOR RESPONSE SYSTEM: COMPARISONS (f PROCEDURE) 2356 257 258 TABLE K.1--Person Perception Test total factor scores (£2356 procedure) compared with Counselor Response System dimensions by product-moment (r). Counselor Response System Person Perception Test GrOUp I Group II Evaluative Factor Content: Follow .17 .04 Control: EXpansive .12 .10 Referrent: Client .25 .08 Reinforcing .04 .08 Emotional Factor Content: Follow -.12 -.14 Control: Expansive -.01 .12 Referrent: Client -.01 .04 Reinforcing -.O4 -.15 Potency Factor Content: Follow -.02 -.09 Control: Expansive .04 -.21 Referrent: Client .00 -.O3 Reinforcing .32 .04 Evaluative-Activity Factor Content: Follow .10 -.21 Control: Expansive .19 -.05 Referrent: Client .26 -.13 Reinforcing .18 -.18 APPENDIX L PERSON PERCEPTION TEST-COUNSELOR RESPONSE SYSTEM COMPARISONS (f4 PROCEDURE) 259 260 TABLE L.l—-Person Perception Test total factor scores (f4 procedure) compared with Counselor Response System ' dimensions by product-moment (r) and by multiple correla- tion (R). — Person Perception Test Counselor Response System Group I Group II Evaluative Factor Content: Follow -.02 -.01 Control: Expansive .ll -.05 Referrent: Client .01 .04 Reinforcing -.31 -.26 Emotional Factor Content: Follow .07 -.08 Control: Expansive .19 -.17 Referrent: Client .05 -.14 Reinforcing -.05 -.O6 Potency Factor Content: Follow .07 -.07 Control: Expansive .12 .06 Referrent: Client .09 .02 Reinforcing -.35 -.32 Evaluative-Activity Factor Content: Follow .06 -.04 Control: Expansive .23 .04 Referrent: Client -.02 .08 Reinforcing -.24 -.39* Interrelated Factors Content: Follow .15 .10 Control: Expansive .25 .28 Referrent: Client .19 .25 Reinforcing .42 .46 Legend: * = significant at .05 level. APPENDIX M PERSON PERCEPTION TEST4COUNSELOR RESPONSE SYSTEM ITEM VALIDITY 261 262 TABLE M.l--Person Perception Test item validity for both groups using the Counselor Response System dimensions as the criteria: Evaluative factor. Person Perception Counselor Response System Test Group Follow Expan. Client Reinf. Peggy closed-Open I -.16 -.44* -.20 .41* II .16 .04 .00 -.02 dishonest-honest I .02 -.21 -.16 .02 II .13 -.24 .02 .29 superficial-profound I .17 -.14 .23 -.43* II .37* .09 .31 -.13 angry-supportive I -.O6 .14 —.21 -.02 II -.06 -.ll -.18 -.O4 perplexed-understanding I .08 .00 .02 .20 II .05 -.31 .03 .25 disagreeing-agreeing I -.O7 -.28 -.15 .03 II .26 .18 .07 .37* disapproving-approving I .08 -.04 -.07 .36 II .14 -.05 -.22 .18 harassed-unhurried I -.29 .02 -.25 .05 II .30 .27 .07 .19 impatient-patient I .08 .25 .07 -.01 II .23 .05 .03 .10 dissatisfied-satisfied I .06 .14 -.03 -.10 II .38* .21 .29 .28 Bill Closed-open I -.35 -.O9 -.41* .20 II -.05 .20 .09 -.04 dishonest-honest I -.13 .05 -.34 -.29 II —.29 -.02 .00 -.38* superficial-profound I .02 -.08 -.17 -.05 II -.05 .03 .22 -.18 angry-supportive I -.40* .00 -.22 -.04 II -.21 .06 .04 -.05 perplexed-understanding I -.26 -.04 -.O4 .10 II -.13 .20 .09 -.13 disagreeing-agreeing I -.21 -.19 -.37* .14 II -.36 .18 -.12 -.14 263 TABLE M.l--Continued Person Perception Counselor Response System Test Group Follow Expan. Client Reinf. Bill (continued) disapproving-approving I -.32 -.O4 -.28 .23 II -.12 .02 .06 .00 harassed-unhurried I -.21 -.06 -.37* .30 II .53** .26 .53** .02 impatient-patient I -.05 -.OS -.41* .19 II .06 .13 .33 -.23 dissatisfied-satisfied I -.18 .04 -.10 .28 II .12 .34 .57** -.13 Bob Closed-open I -.24 -.O9 -.29 .01 II .01 -.Ol -.12 .16 dishonest-honest I -.17 -.06 -.23 -.05 II -.15 -.28 —.21 -.18 superficial-profound I -.13 .08 -.15 -.09 II -.24 .10 .01 -.14 angry-supportive I -.19 -.15 -.3O .03 II -.04 -.02 -.12 .07 perplexed-understanding I -.12 .29 -.18 .15 II -.23 .25 .05 -.38* disagreeing-agreeing I -.23 -.19 -.34 .11 II -.19 -.12 -.36 -.09 disapproving-approving I -.12 -.13 -.29 .10 II —.10 .04 -.29 .07 harassed-unhurried I -.16 .24 -.18 -.05 II .06 -.05 -.O3 .07 impatient-patient I -.29 -.05 -.33 -.10 II -.05 .01 -.06 .OO dissatisfied-satisfied I -.31 -.03 -.31 -.00 II .05 -.02 -.05 .33 Lynn closed—open I -.O8 .23 -.OS .21 II .03 -.O3 .13 -.24 dishonest-honest I .16 -.02 .06 .09 II -.34 -.09 -.20 -.20 264 TABLE M.l--Continued Person Perception Counselor Response System Test Group- Follow Expan. Client Reinf. Lynn (continued) superficial-profound I .30 .11 .15 -.10 II -.33 .02 -.10 -.15 angry-supportive I .03 -.07 .29 .06 II .12 .01 .08 -.21 perplexed-understanding I .02 .16 -.01 .37* II .04 '004 .14 015 disagreeing-agreeing I .13 -.08 .07 .32 II -.11 -.05 .03 -.31 disapproving-approving I -.16 -.08 —.O8 .36 II .32 .08 .41* -.26 harassed-unhurried I -.40* .01 -.19 .10 II -.06 -.10 -.02 .05 impatient-patient I -.16 .04 .03 .00 II -.09 .01 -.10 -.36 dissatisfied-satisfied I -.21 -.06 -.03 -.09 II .18 .01 .08 ~.39* Terry closed-open I .05 -.10 .15 -.06 II -.36 -.24 -.37* .01 dishonest-honest I -.17 -.23 -.18 -.10 II -.39* -.23 -.16 -.35 superficial-profound I -.09 -.23 .04 -.28 II -.28 -.06 .02 .52** angry-supportive I -.14 -.29 -.02 .23 II -.17 -.25 -.20 .02 perplexed-understanding I -.08 .25 .01 .27 II -.15 .04 -.05 -.12 disagreeing-agreeing I .05 .03 -.03 .38** II -.12 .12 -.23 .16 disapproving-approving I -.02 -.05 .12 .28 II -.25 -.18 -.26 -.14 harassed-unhurried I -.26 .10 -.09 .02 II .16 .02 .04 .14 impatient-patient I —.14 .12 .05 .00 II .21 .23 .14 -.OS dissatisfied-satisfied I -.37* -.14 -.26 .03 II .16 .14 .13 -.18 Legend: ** significant at .01 level; * significant at .05 level. TABLE M-2--Person Perception Test item validity for both groups using the Counselor Response System dimensions as the criteria: Emotional factor. Person Perception Counselor Response System Test Group Follow Expan. Client Reinf. Peggy excited-relaxed I .29 .23 .31 .24 II .09 .12 .15 .08 upset-calm I .24 .11 .06 .10 II .03 .25 .08 .03 embarrassed-smug I .06 .26 .03 .12 II .36 .11 .19 .15 uncontrolled-controlled I .08 .07 .11 .49** II .12 .10 .10 .28 fearful-hopeful I .28 .17 .22 .19 II .12 .33 .05 .06 feminine-masculine I .09 .13 .17 .16 II .28 .01 .04 .34 Bill excited-relaxed I .09 .17 .09 .15 II .25 .02 .15 .29 upset-calm I .16 .05 .27 .27 II .23 .04 .03 29 embarrassed-smug I .19 .03 .19 .05 II .02 .13 .08 .04 uncontrolled-controlled I .16 .01 .24 .02 II .10 .07 .19 .22 fearful-hopeful I .19 .02 .29 .01 II .00 .02 .09 .09 feminine-masculine I .29 .18 .15 .23 II .14 .17 .07 .05 Bob excited—relaxed I .03 .08 .12 .04 II .21 .14 .30 .22 upset-calm I .12 .12 .01 .07 II .12 .01 .18 .16 embarrassed-smug I .08 .14 26 .05' II .16 .23 .08 .08 uncontrolled-controlled I .12 .02 .03 .16 II .07 .10 .14 .33 266 TABLE M-2--Continued Person Perception Counselor Response System Test Group Follow Expan. Client Reinf. Bob (continued) fearful-hOpeful I -.O4 —.15 .04 -.09 II -.12 -.10 -.29 -.11 feminine—masculine I .00 .05 .15 -.20 II -.03 -.25 -.05 .05 Lynn excited-relaxed I -.30 .44* -.36 -.08 II .22 .22 -.O6 -.07 upset-calm I -.31 .39* -.30 -.20 II .08 -.Ol .04 -.11 embarrassed-smug I -.07 .07 .03 -.19 II -.21 .13 -.11 .10 uncontrolled-controlled I .31 -.04 .01 -.12 II -.25 -.01 -.07 -.27 fearful-hopeful I -.19 .12 -.17 .00 II -.14 -.21 -.22 -.25 feminine-masculine I .45* -.33 .42* .19 II -.03 .10 -.05 -.12 Terry excited-relaxed I .10 -.O4 -.19 -.41* II -.11 -.40* -.24 .01 upset-calm I -.12 -.15 -.26 -.18 II -.13 -.26 -.15 -.24 embarrassed-smug I -.ll .09 -.20 .24 II .28 .00 .10 -.O4 uncontrolled-controlled I .02 -.30 -.12 .07 II -.40* -.21 -.35 -.03 fearful-hopeful I .03 -.36 .04 -.04 II -.01 -.39* -.17 -.03 feminine-masculine I -.13 -.14 -.33 -.23 II .18 -.50** -.23 .27 Legend: ** = significant at .01 level; * = significant at .05 level. TABLE M.3--Person Perception Test item validity for both groups using the Counselor Response System dimensions as the criteria: Potency factor. Person Perception Counselor Response System Test Group Follow Expan. Client Reinf. Peggy simple-complex I .12 .23 .15 .12 II .SO** .42* .28 .25 weak-strong I .16 .05 .27 .23 II .14 .08 .14 .40 bored-surprised I .05 .14 .17 .09 II .19 .13 .01 .18 mild-intense I .29 .17 .21 .48* II .01 .15 .02 .04 Bill simple-complex I .09 .09 .05 .21 II .08 .20 .11 .18 weak-strong I .25 .14 .31 .16 II .07 .12 .06 .09 bored-surprised I .35 .19 .07 .22 II .15 .18 .ll .00 mild-intense I .02 .27 .33 .28 II .04 .19 .05 .06 Bob simple-complex I .16 .02 .14 .15 II .26 .33 .37* .29 weak-strong I .33 .08 .15 .01 II .16 .11 .00 .07 bored-surprised I .13 .00 .24 .16 II .04 .01 .02 .09 mild-intense I .26 .15 .42* .58** II .10 .11 .07 .20 268 TABLE M.3-~Continued Person Perception Counselor Response System Test Group Follow Expan. Client Reinf. Lynn simple-complex I .01 .06 -.02 -.20 II .16 -.O3 -.12 .62** weak-strong I .17 .03 -.10 -.23 II -.16 -.18 -.14 .22 bored-surprised I .03 .10 -.O8 -.12 II -.33 -.19 -.23 -.05 mild-intense I .11 -.32 -.20 .40* II -.03 .03 .08 .06 Terry simple-complex I .02 -.43* .04 -.19 II .17 -.40* .03 .28 weak-strong I .21 -.53** -.12 .05 II .08 -.6l** -.26 .21 bored-surprised I .04 -.19 -.O7 .15 II -.04 .26 -.24 -.O6 mild-intense I .09 -.11 -.33 .18 II —.12 -.ll -.27 -.17 Legend: *3': significant at .01 level; significant at .05 level. 269 TABLE M.4--Person Perception Test item validity for both groups using the Counselor Re5ponse System dimensions as the criteria: Evaluative-activity factor. Counselor Response System Person Perception Test Group Follow Expan. Client Reinf. Peggy opaque-transparent I -.02 -.44* -.05 .20 II -.42* -.12 -.16 .27 far-near I -.14 -.6l** -.29 .29 II .03 -.32 -.17 .26 tired-energetic I -.12 -.32 -.13 .22 II .20 .05 .02 .27 slow-fast I .08 .07 .02 .20 II .05 .01 -.03 .14 defensive-aggressive I -.02 -.14 .00 -.04 II -.07 -.09 -.12 .16 passive-active I -.27 -.12 .01 .20 II .17 -.14 .09 -.O7 static-dynamic I .00 -.05 .22 .11 II .05 -.19 -.22 .30 unhappy-happy I -.06 -.23 -.12 .16 II -.06 -.Ol .02 .14 constrained-free I -.04 -.Ol -.03 .47* II -.39* -.51** -.31 -.12 Bill Opaque-transparent I -.38* .09 -.21 .00 II -.02 -.O3 -.01 .04 far-near I -.36 .05 -.30 -.04 II -.10 .13 .11 -.22 tired-energetic I -.08 -.20 -.40* .24 . II -.17 .09 -.O6 -.21 slow-fast I -.26 -.O3 -.44* .11 II -.03 .10 .05 .05 defensive-aggressive I -.30 .16 -.26 -.25 II .08 .28 .42* -.14 passive-active I -.22 -.10 -.34 .04 II -.01 .03 -.01 .11 static-dynamic I -.10 .18 -.23 -.07 II -.05 .03 .04 .14 270 TABLE M.4-~Continued Counselor Response System Person Perception Test Group Follow Expan. Client Reinf. Bill (continued) unhappy-happy I -.22 .04 -.33 -.31 II -.26 -.10 -.08 -.08 constrained-free I -.30 .20 -.O8 -.16 II -.19 .23 .11 -.06 Bob opaque-transparent I -.18 -.02 .00 .11 II -.25 -.01 -.ll -.01 far-near I -.17 -.18 -.21 .17 II -.09 .20 .01 -.08 tired-energetic I -.01 .12 -.15 -.22 II .13 .36 .13 .32 slow-fast I .15 .05 -.O6 .14 II .18 .38* .27 .11 defensive-aggressive I .40* .13 .30 -.21 II .11 .13 .13 .06 passive-active I .10 .15 .10 -.14 II -.17 .23 -.04 -.O8 static-dynamic I .39* .31 .13 .13 II -.17 .08 -.O3 -.01 unhappy-happy I -.18 -.24 -.24 .10 II -.30 .07 -.25 .01 constrained-free I -.22 .06 -.21 .05 II .09 .07 .15 -.11 Lynn opaque-transparent I .08 -.17 .14 .17 II -.04 .16 .05 -.12 far-near I .14 -.14 -.09 -.12 II -.09 -.02 .08 -.24 tired-energetic I .18 .02 .04 .14 II -.03 .12 .01 -.29 slow-fast I -.44* -.20 -.20 .20 II -.15 .19 .17 -.37* defensive-aggressive I -.09 .37* .ll .18 II -.39* .09 .00 -.23 271 TABLE M.4--Continued Person Perception Counselor Response System Test Group Follow Expan. Client Reinf. Lynn (continued) passive-active I -.28 .12 -.34 .02 II -.18 .15 -.19 -.18 static-dynamic I .05 -.16 .01 .27 II -.27 .21 -.09 -.29 unhappy-happy I -.22 .08 -.12 -.10 II -.12 .10 -.O6 -.26 constrained-free I -.27 .18 -.28 .34 II .18 -.01 .09 -.43* Terry opaque-transparent I .01 -.13 .06 .21 II -.12 .ll -.20 -.02 far-near I .02 -.24 .06 -.08 II -.10 -.12 -.01 -.19 tired-energetic I -.25 -.24 -.26 .18 II .36 -.12 .05 .09 slow-fast I -.03 -.16 -.O4 .12 II .10 -.11 .05 -.20 defensive-aggressive I -.02 -.34 -.09 -.03 II -.03 -.O7 .02 -.33 passive-active I -.37* -.19 -.03 ~.12 II .05 -.22 -.O7 -.26 static-dynamic I -.49** -.10 -.31 .39* II -.11 -.ll -.24 -.36 unhappy-happy I .07 -.21 .01 -.12 II -.45* -.28 -.24 -.22 constrained-free I -.33 .01 -.10 .08 II -.07 .06 .Ol -.34 Legend: ** significant at .01 level; significant at .05 level.