EFFECTS OF SOCIAL MODELING AND COGNITIVE I ‘- STRUCTURING STRATEGIESVON AFFECTIVE SELF DISCLOSURE OF SINGLE UNDERGRADUATE COLLEGES MALES ' Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY V PAMELA s. HIGHLEN 1975 ~‘II— z": B IIIIIIII III ‘IIIIIIIIIIIIIIIIIIIIIII “ 3 1293 106841 nfiRARY “new This is to certify that the thesis entitled EFFECTS OF SOCIAL MODELING AND COGNITIVE STRUCTURING STRATEGIES 0N AFFECTIVE SELF-DISCLOSURE OF SINGLE UNDER- GRADUATE COLLEGE MALES presented by Pamela S. Highlen has been accepted towards fulfillment of the requirements for Ph.D degree in ML @6ng Major professor W on. W" Nov 16 zone ‘.. . '119 A 1m i3 8W “Mfr . cw.”.-‘.-- , "R _ “tact, In AWL; .. ‘- IVIQIIQI‘Ixic-ty. (A; 5‘... @ Ig1lfilosut?, 2 I it: ELL: :gal , cm_ ' 0 ed posts“: tr» sugg‘emm‘fld, kg... .1119 putwamt {AMA , fig trflfltfiyflt 56ftft$ 3.?) _.:7 ;, ABSTRACT EFFECTS OF SOCIAL MODELING AND COGNITIVE STRUCTURING STRATEGIES ON AFFECTIVE SELF-DISCLOSURE OF SINGLE UNDER- GRADUATE COLLEGE MALES BY Pamela S. Highlen This research involved an examination of the effects of social modeling and cognitive structuring multi-component treatment strategies on the affective self-disclosure of single Michigan State University undergraduate males. Specifically, the objectives of this investi- gation were: 1. to assess the impact of social modeling and cognitive structuring strategies on males regarding (a) amount of affect, (b) quality of affect, (c) con- comitant level of anxiety, (d) skill necessary for affective self-disclosure, and (e) attitudes toward disclosing feelings. 2. to examine treatment effects over time by administering a delayed posttest to subjects who did not-receive an intervening self—management procedure (Vbight, 1975). Pamela S. Highlen To meet these objectives, two videotaped presen- tations were developed with the intent of making sys— tematic outcome comparisons between the social modeling and cognitive structuring treatments. To ascertain the immediate treatment effect, social modeling and cognitive structuring treatment groups were contrasted with attention-placebo and no-treatment control conditions. The design employed for this study was an experimental posttest-only design recommended by Campbell and Stanley (1963). Subjects (Né48) were randomly assigned to the social modeling (SM), cognitive structuring (CS), attention-placebo (AP), and no-treatment control (NTC) groups. Social modeling, cognitive structuring, and attention-placebo treatments were administered indi- vidually and were of comparable length. Immediately following treatment, each subject took the performance test and two paper-and-pencil measures. In order to assess treatment effects over time, subjects who received no intervening treatment (Voight, 1975) were retested at a follow-up session three weeks later. The basic design for this phase was a 2 x 4 repeated measures design. Since only 13 of 16 possible subjects returned for the delayed posttest, unequal cell sizes were used. The dependent variables used in this study were created to assess amount and quality of affect, Pamela S. Highlen concomitant anxiety level, skill, and attitudes toward disclosing feelings. Both the skill and attitude measures were pretested. Prior to the actual experi- ment, a pilot study was conducted. Reliability coef- ficients for the skill and attitude measures were com- puted, as well as an item analysis of each measure, with only the most discriminating items retained for the actual study. Two raters were trained to evaluate amount and quality criteria. For the actual study, performance test audio— tapes were made into typescripts and were then rated by the two trained raters. Interrater reliability estimates were .99 for amount and quality, while the coefficients ranged from .99 to .97 for the quality subscales. Each subject's total score for the anxiety, skill, and attitude measures was computed. Reliability esti- mates for the skill and attitude measures were .87 and .84, respectively. For the immediate posttest, it was hypothesized that the effects of social modeling and cognitive structuring treatments would be greater than those for attention-placebo and no-treatment control conditions. A second hypothesis stated that the effects of cognitive structuring would be greater than those for social model- ing, while a third predicted that no difference would exist between attention—placebo and no—treatment control conditions. Pamela S. Highlen For the delayed posttest, the same directional hypotheses were formulated for the treatment effect. In addition, the researcher hypothesized that no measures or treatment-by-measures interaction effects would be found. Data for the immediate posttest were analyzed using a one-way multivariate analysis of variance pro- cedure. Data for the immediate and delayed posttest comparison were analyzed using a repeated measures multi- variate analysis of variance technique. A statistically significant (a=.05) effect was found for the contrast between the two experimental and the two control groups on the immediate posttest. Neither the planned contrast between experimental treatments nor the one between the two control groups was significant. Results for the immediate-delayed posttest comparison were not significant. Three supplemental analyses were conducted. The first, a repeated measures MANOVA, examined three factors-- type-of-role, type-of—feeling, and sex-of—best-friend-- in items on the performance and skill tests. For the performance test, a significant difference (a=.05) between the treatment versus no-treatment groups on the amount and quality measures was obtained. However, the treatment, not the measures, effect was responsible for these differences. Across subjects, however, the feeling Pamela S. Highlen effect was significant for amount and quality. Partici— pants were able to make more affective responses with higher quality to positive situations than negative ones. Only the multivariate measures effect was significant for the anxiety scale. Once again, the feeling effect pro- duced the only significant univariate result. This find- ing indicated that subjects were more anxious when making negative affective responses than when making positive ones. The cognitive structuring versus the social model- ing contrast for the skill test was also significant (a=.05). None of the other multivariate tests was sig- nificant. The second, a one-way ANOVA, used the mean quality score as the dependent measure. None of the planned contrasts proved significant. Third, descriptive data from the debriefing questionnaire showed that experi- mental treatment subjects generally felt that the video- tape presentations were valuable. Furthermore, they reported increased awareness of and improved skills in affective communication as the primary benefits from the training. EFFECTS OF SOCIAL MODELING AND COGNITIVE STRUCTURING STRATEGIES 0N AFFECTIVE SELF-DISCLOSURE OF SINGLE UNDER- GRADUATE COLLEGE MALES BY 0” . Pamela S. Highlen I A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Personnel Services and Educational Psychology 1975 ACKNOWLEDGMENTS Any_work of this magnitude cannot be completed without assistance from others. Therefore, the writer wishes to recognize and express her appreciation to the following people: To Dr. Norman R. Stewart, committee chairperson and major professor throughout my doctoral study, who has provided insight, encouragement, and support through- out my graduate work. His active recruiting and support of women in the doctoral program and his tireless dedi- cation as a counselor educator make me proud to be one of his doctoral candidates. To Dr. Herbert M. Burks, Jr., Committee member, who contributed valuable editorial comments to this research effort and consistently demonstrated an active interest in my graduate work. To Dr. William H. Schmidt, committee member, who offered expertise in research design and statistics. His high research standards were those I attempted to meet . ii To Dr. Dozier W. Thornton, committee member, whose knowledge of and interest in social learning theory contributed to my growth in this area. To Nancy L. Voight, close friend and colleague, whose knowledge, creativity, and sense of humor greatly contributed to this investigation. To Dennis O'Grady, Doug Mason, Margaret Baker, and Bill Brooks, who contributed their talents and count- less hours to make the videotaped presentations credible treatment strategies. To Bill Cruise, President of Cruise Communi— cations Company, who so generously gave of his time and talents in editing the videotapes. To the Office of Research Consultation staff, who provided valuable statistical advice and consul- tation. To the Instructional Media Center staff, who supplied the audiovisual equipment for this study, as well as technical production suggestions. Finally, to the many significant others who contributed to my background, growth, and motivation which culminated in this research effort. To Peg, Elouise, and Jim '7’. .WW” are “shaped and fashioned by xngntmse whom we love. Purpose Impm .a- . --Goethe 661191.) I. . . 915. Review ‘ an. 8e1:‘~;- ;.,'° goal : "n ' “ . 606...!- ‘{‘1 ' SCCIF . z‘ 5 . ‘ ‘ FaCLl. .' ‘.‘_ 7 iILLLL ‘ Perri: I 7: se‘.‘ . buaT"'.l LI 3([1'71 _ '. ... ._ , Trentn'c: I.” ~ .IuatuL-“f ... 3‘ jI'}-j.\i_vu_; . . sttech . ._..+ n;, mEffects 0‘; _l. . :"‘ .. 5:; géo‘quttwe s: ,---.: Wt‘w‘ ze3‘-tmtur.i:.m instrucrinn . . Wain” Self—modelzrze. !_\._0. v 0 o I. \ ‘ 5‘- 05:16:: AM. nmcrmar‘r' ‘2‘“? ' ' t of flea-urea. . . . ‘ltndy . . . . . 1‘. IN: :2. a. r TABLE OF CONTENTS Chapter I. INTRODUCTION AND REVIEW OF LITERATURE Introduction . . . . . . Purpose . . . . . . . . Importance . . . . . . . Generalizability . . . . . Review of Literature . . . . Self- Disclosure . . Social Learning and Self— Disclosure Social Learning Theory. . . Social Modeling . . . . . Facilitation of self- disclosure in interview settings. . Facilitation of social skills across settings . . . . . . Research Implications for Modeling Treatments . . . . . . Nature of modeling experience. Effects of reinforcement . Effects of overt rehearsal. Cognitive Structuring . . . Cognitive restructuring. . Self— —instruction . . . Cognitive self-modeling. . Summary . . . . . . . . II. EXPERIMENTAL DESIGN AND PROCEDURES Overview . . . . . . . . Pretest of Measures. . . . Pilot Study . . . . . . Adv. c Page mUIUJNH I—‘ 16 19 22 22 25 25 26 32 34 36 36 39 39 Chapter III. Hypotheses . . . . . . Definition of Terms . . . . Design over Time . . . Design over Variables. . . Sample . . . . . . . . Sample Selection Procedures. Sample Characteristics . . Subject Assignment. . . . Treatment . . . . . Social Modeling. . . Cognitive Structuring. Attention Placebo . . No-Treatment Control . Instrumentation and Data Collection. Performance Test . . . . Raters. . . . . . . . Reliability . . . . Paper-and— Pencil Measures . Data Collection. . . . . Data Analysis . . . . . RESULTS. . . . . . . . . Immediate Posttesting . . . General Findings . . . Hypotheses Test Results . . Delayed Posttesting . . . . General Findings . . . . Hypotheses Test Results . . Supplemental Findings . . . Repeated Measures Item Analyses Mean Quality Analysis of Variance. Debriefing . . . . . . Summary . . . . . . . . Page 41 42 45 47 74 74 77 84 84 86 96 97 118 120 125 Chapter IV. SUMMARY, DISCUSSION, AND IMPLICATIONS . . Summary . . . . . . . . . . . Limitations . . . . . . . . . . Discussion and Implications. . . . . Subject Variable. . . . . . . . Dependent Variables. . . . . . . Treatment Effects . . . . . . Supplemental Findings . . . . . . Implications for Counseling and Edu- cation . . . . . . . . . . Conclusion . . . . . . . . . . APPENDICES Appendix A. ADVERTISEMENTS. . . . . . . . . . B . FLYER. . O O I I C I I O O O I C. LETTER TO SUBJECTS . . . . . . . . D. SOCIAL MODELING VIDEOTAPE TYPESCRIPT . . E. COGNITIVE STRUCTURING VIDEOTAPE TYPESCRIPT F. ANXIETY SCALE . . . . . . . . . . G. AUDIO PERFORMANCE TEST . . . . . . . H. PERFORMANCE TEST MEASURES: AMOUNT AND QUALITY RATING CRITERIA . . . . . . I . SKILL TEST . I O C O I O O U U 0 J. ATTITUDE SURVEY . . . . . . . . . K. DEBRIEFING . . . . . . . . . . . L. INTERCORRELATIONS FOR DEPENDENT MEASURES . mFERENCES I O I O O I I I O O U I O Page 128 128 133 138 138 140 144 149 151 152 154 155 156 157 187 213 220 231 237 240 241 242 LIST OF TABLES Table Page 1. Hoyt Reliability Coefficients for the Total Rater Ratings for Amount and Quality of Affect and the Five Quality Subscales . . . 67 2. Cell Means and Standard Deviations for Social Modeling, Cognitive Structuring, Attention- Placebo, and No-Treatment Control Groups on Amount, Quality, Skill, Anxiety, and Attitude Measures. . . . . . . . . . 75 3. Multivariate and Univariate Analysis of Variance and Step-Down F Statistics for Quality, Amount, Skill, Attitude, and Anxiety Measures; Social Modeling and Cognitive Structuring (n = 24) vs. Attention-Placebo and No- Treatment Con- trol (n = 24) Comparison . . . . . . . 78 4. The Least Square Estimates, 95% Confidence Intervals, for the Quality, Amount, Skill, and Attitude Measures; Social Modeling and Cognitive Structuring (n = 24) vs. Attention-Placebo and No- Treatment Control (n = 24) Comparison . . . . . . . . . 80 5. Multivariate and Univariate Analysis of Variance and Step- Down F Statistics for Quality, Amount, Skill, —Attitude, and Anxiety Measures; Cognitive Structuring (n = 12) vs. Social Modeling (n = 12) Comparison . . . . . . . . . . . . 82 6. Multivariate and Univariate Analysis of Variance and Step- Down F Statistics for Quality, Amount, Skill, —Attitude, and Anxiety Measures; Attention-Placebo (n = 12) vs. No-Treatment Control (3 = 12) Comparison. 83 viii Table 7. 10. 11. 12. 13. Cell Means for Social Modeling, Cognitive Structuring, Attention-Placebo, and No— Treatment Control Groups on the Summation of and Difference between Amount, Quality, Skill, Anxiety, and Attitude Measures across Immediate and Delayed Posttesting . . . . . . . . . . . Multivariate and Univariate Analysis of Variance and Step-Down F Statistics for the Treatment-by-Measures Interaction Effect across Immediate and Delayed Posttests. . . . . . . . . . . . Repeated Measures Multivariate Analysis of Variance for the Planned Contrasts Testing the Treatment Effect across Immediate and Delayed Posttests . . . . . . . . . Repeated Measures Multivariate Analysis of Variance for the Planned Contrasts of the Measures Effect across Immediate and Delayed Posttesting . . . . . . . . Immediate Posttest Cell Means and Standard Deviations for the Four Treatment Con- ditions across the Five Dependent Variables. . . . . . . . . . . . Repeated Measures Multivariate and Uni- variate Analysis of Variance and Step—Down F Statistics for Amount and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by- Feeling, Role-by-Feeling, and Role-by-Sex- by-Feeling; Social Modeling and Cognitive Structuring (n = 24) vs. Attention-Placebo and No-Treatment Control (n = 24) Com- parison . . . . . . . .' . . . . Repeated Measures Multivariate and Uni- variate Analysis of Variance and Step—Down F Statistics for Amount and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by-Feeling, Role-by-Feeling, and Role-by-Sex—by-Feel— ing; Cognitive Structuring (n = 12) vs. Social Modeling (a = 12) Comparison . . . ix Page 85 86 87 88 90 98 99 Table Page 14. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down F Sta— tistics for Amount and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by-Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling; Attention-Placebo (g = 12) vs. No-Treatment Control (3 = 12) Comparison . . . . . . 100 15. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down 3 Sta- tistics for Amount and Constant, Role, Sex, Feeling, Role-by-Feeling, Sex-by-Feeling, Role-by-Feeling, and Role-by-Sex-by-Feeling across the Four Treatment Groups . . . . 101 16. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down F Sta- tistics for Quality and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by-Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling; Social Modeling and Cognitive Structuring (n = 24) vs. Attention—Placebo and No- Treatment Control (2 = 24) Comparison. . . 102 17. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down 3 Sta- tistics for Quality and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by-Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling; Cognitive Structuring (n = 12) vs. Social Modeling (5 = 12) Comparison. . . . . . 103 18. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down E Sta- tistics for Quality and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by-Feeling, Role- by-Feeling, and Role-by-Sex-by—Feeling; Attention-Placebo (n = 12) vs. No—Treatment Control (3 = 12) Comparison . . . . . . 104 19. Repeated Measures Multivariate and Univariate Analysis of Variance and Step—Down F Sta- tistics for Quality and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by-Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling across the Four Treatment Groups . . . . 105 Table Page 20. Repeated Measures Multivariate and Univariate Analysis of Variance and Step—Down E Sta- tistics for Anxiety and Constant, Role, Sex, Feeling, Role—by-Sex, Sex-by-Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling; Social Modeling and Cognitive Structuring (n = 24) vs. Attention-Placebo and No- Treatment Control (3 = 24) Comparison . . . 106 21. Repeated Measures Multivariate and Univariate Analysis of Variance and Step— Down F Sta- tistics for Anxiety and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by- Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling; Cognitive Structuring (n = 12) vs. Social Modeling (a = 12) Comparison . . . . . . 107 22. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down F Sta— tistics for Anxiety and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by—Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling; Attention—Placebo (n = 12) vs. No-Treatment Control (n = 12) Comparison . . . . . . 108 23. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down F Sta— tistics for Anxiety and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by- Feeling, Role— by-Feeling, and Role-by- Sex- by—Feeling across the Four Treatment Groups. . . . . 109 24. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down F Sta- tistics for Skill and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by- Feeling, Role- by-Feeling,m Md Role-by- Sex-by-Feeling; Social Modeling and Cognitive Structuring (n— = 24) vs. Attention-Placebo and No- Treatment Control (a = 24) Comparison . . . 110 25. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down F Sta- tistics for Skill and Constant, Role, Sex, Feeling, Role-by- Sex, Sex-by- Feeling, Role- by-Feeling, and Role-by- Sex-by- Feeling; Cognitive Structuring (n = 12) vs. Social Modeling (a = 12) Comparison . . . . . . 111 xi Table 26. 27. 28. 29. 30. 31. Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down F Sta— tistics for Skill and Constant, Role, Sex, Feeling, Role-by—Sex, Sex-by-Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling; Attention-Placebo (n = 12) vs. No-Treatment Control (3 = 12) Comparison . . . . . Repeated Measures Multivariate and Univariate Analysis of Variance and Step-Down F Sta- tistics for Skill and Constant, Role, Sex, Feeling, Role-by-Sex, Sex-by—Feeling, Role- by-Feeling, and Role-by-Sex-by-Feeling across the Four Treatment Groups . . . Analysis of Variance of Mean Quality Score for Three Planned Comparisons . . . . Cell Means and Standard Deviations for Social Modeling, Cognitive Structuring, Attention- Placebo, and No-Treatment Control on Mean Quality Measure . . . . . . . . . Major Results from Debriefing Questionnaire Immediate Posttest Correlation Matrix for Amount, Quality, Skill, Anxiety, Attitude, and the Supplemental Mean Quality Measure Page 112 113 119 119 121 241 Figure 1. 2. 10. 11. LIST OF FIGURES Bandura's Social Learning Paradigm . . . Posttest—only control group experimental design over time . . . . . . . One-way design. . . . . . . . . . One-way repeated measures design. . . . Parallel processes involved in the social modeling and cognitive structuring treatments . . . . . . . . . . Representation of the three factors equally presented in the affective self-disclosure performance test (EFZ) . . . . . . Comparison of mean scores for amount of affect between immediate and delayed posttests. . . . . . . . . . . Comparison of mean scores for quality of affect between immediate and delayed posttests O O I O I I I I O O I Comparison of mean scores for skill level between immediate and delayed posttests. Comparison of mean scores for anxiety level between immediate and delayed posttests. Comparison of mean scores for attitude sur" vey between immediate and delayed post— tests . . . . . . . . . . . . xiii Page 12 46 47 48 60 63 91 92 93 94 95 CHAPTER I INTRODUCTION AND REVIEW OF LITERATURE Introduction Males in this culture have been socialized in ways that make it difficult for them to express feelings (Lynn, 1964). Specifically, males tend to inhibit verbal affective self-disclosure both as initiators and respon— dents in dyadic interactions. Current research in the area of social learning has demonstrated that emotional responsiveness can be affected by several treatment combination packages, each utilizing Bandura's theoreti- cal paradigm (l97lb). The first, social modeling, focuses on learning through symbolic modeling and overt behavior rehearsal, while the second, cognitive structuring, emphasizes the cognitive mediating factors that influence learning. Typically, these cognitive treatment strategies focus on the individual's self-verbalizations about the inappropriate behavior, self-instructions on how to change it, and cognitive self-modeling of the desired response. Although most of this research has been conducted on inhibiting inappropriate emotional arousal, the same strategies should be similarly effective in facilitating socially sanctioned emotional responsiveness. Purpose Single, undergraduate males at Michigan State University volunteered to participate in a research/ training program on disclosing feelings. The overall purpose of this study was to apply social modeling and cognitive structuring treatment strategies to increase the affective self-disclosure of the participants. Therefore, this researcher investigated whether social modeling and cognitive structuring multi—component strategies are effective methods for increasing the affective self—disclosure of single undergraduate college males. Specifically this study had a two-fold purpose: 1. to assess the impact of social modeling and cognitive structuring strategies on males regarding (a) amount of affect expressed, (b) quality of affect, (c) concomitant level of anxiety, (d) skill necessary for affective self-disclosure, and (e) attitudes toward disclosing feelings. 2. to examine the potency of social modeling and cognitive structuring treatments over time. Three weeks following treatment, subjects who did not receive a 33.. v»:- . H“ u).- a ‘ o .4 ,. .II I .-.. 2‘ 1 ”I l .. . ‘. ' '0‘ .I_ H i-. . 1 4 Id, self-management procedure from Voight (1975) were re— evaluated to assess the maintenance of treatment effects. Importance Although self-disclosure has been extensively studied by social psychologists, a review of this literature indicated conflicting results due to poor methodology and a paucity of actual performance measures. In addition, self-disclosure has been typically defined as "any information about himself which Person A communi- cates verbally to a Person B" (Cozby, 1973, p. 73). This definition seems inadequate when considering that an individual is a respondent as well as an initiator of dyadic communication. Furthermore, this definition makes no distinction between cognitive and affective self-disclosures. Hence, few studies have examined such differential effects. While social scientists have employed poor research methodology and inadequate operational defi— nitions of self-disclosure, behavioral scientists have almost completely ignored this area in research. As Kanfer and Phillips (1970) noted: "The potential of deliberate application of vicarious learning methods for training patients in improving their effective social interactions has barely been tapped" (p. 238). Typically, behavioral scientists have focused their investigations on affective behavior that is easily quantifiable. Although phobics have served admirably in the method- ological evaluation of treatment procedures, behavioral scientists need to expand their choice of clinical problems for investigation. This researcher, therefore, attempted to expand the realm of clinical problems investi- gated by behavioral scientists. In addition, this investigator examined the efficacy of using multicomponent treatment packages instead of a single treatment. Too often researchers isolate and test a single treatment variable only to find it to be insignificant in effecting change. Using multicomponent strategies consisting of discrete facets enabled this researcher to examine the efficacy of whole treatment packages. Subsequent research, then, can isolate specific causes through analyses of the discrete com— ponents. Empirically examining multicomponent treatment packages is quite similar to actual therapy conditions where multiple strategies are typically employed. Both treatment packages tested fit Bandura's social learning paradigm. The social modeling package focused primarily on external stimuli to effect learning. In contrast, the cognitive structuring package emphasized cognitive mediating factors that influence learning. By demonstrating the relative efficacy of each, this “It I... u.- .... 'Ulv 'v. A... 3.. '~ 94 ‘v - \§ .3 study has added to the body of empirical knowledge regarding social learning principles. Finally, the inclusion of a treatment follow-up also increased this study's import. A common criticism of laboratory experiments is that only the immediate treatment impact is assessed. Thus, the question of treatment potency over time usually remains unanswered. By including a short-term follow-up, this investigator attempted to address this question. Generalizability The findings of this study have impact beyond the limits of this study itself. First, since the single undergraduate male volunteers were from a large Midwestern university, it seems reasonable to suggest that the find- ings may generalize to other college male volunteers want- ing to learn affective communication skills. Second, no research evidence suggests that inhibition of affective self-disclosure is a problem unique to college males. Therefore, social modeling and cognitive structuring treatments also may be appropriate for others wishing to increase affective self-disclosure. Finally, social modeling and cognitive structuring may be applicable to a variety of interpersonal situations. For instance, these strategies to facilitate client self-disclosure during therapy or as adjunct strategies for clients concerned with improving social skills such as dating. Review of Literature Research for this study was reviewed in six major areas: Self—Disclosure, Social Learning and Self-Disclosure, Social Learning Theory, Social Modeling, Research Implications for Modeling Treatments, and Cog—- nitive Structuring. Self—disclosure includes a review of studies dealing with the sharing of oneself with another. The processes regulating self—disclosure are next discussed within the context of a social learning paradigm. A brief discussion of social learning theory follows and provides a theoretical framework for the reviews on social modeling, implications for modeling treatments, and cognitive structuring. These last three sections, then, contain in—depth reviews of research related to self-disclosure and social learning principles. Self-Disclosure In the late 1950's the term "self—disclosure" first appeared in psychological literature. Sidney Jourard and his associates were instrumental in promoting interest in and conducting research on this topic. Their efforts were not surprising when one realizes the importance Jourard ascribed to the construct. According to Jourard (1959), self-disclosure is the norm of mental health and, thus, a key factor in personality change. In an attempt to more fully understand self—disclosure and its correlates, Jourard devised a self-disclosure questionnaire (Jourard & Lasakow, 1958). Other measures were devised (e.g., Rickers-Ovsiankina & Kusmin, 1958; Polansky & Brown, 1967), and numerous studies ensued. However, research has not demonstrated a relation— ship between the Jourard Self-Disclosure Questionnaire and actual disclosure in a situation (Ehrlich & Graeven, 1971; Himelstein & Kinbrough, 1963; Lubin & Harrison, 1964; Vondracek, 1969a, 1969b), or with ratings of actual disclosure made by peers (Himelstein & Lubin, 1965; Hurley & Hurley, 1969). Likewise, research on self— disclosure as a personality trait has been inconclusive (Cozby, 1973). Altman and Taylor (1973) now advocate exploring the relationship between personality and self- disclosure within the context of specific relationships and settings. Uncontrolled situational factors most likely have contributed to the equivocal findings on the relationship between self-disclosure and age, social background, race, and sex. For instance, numerous inves- tigators reported females as being more disclosing than males (e.g., Himelstein & Lubin, 1965; Jourard & Lasakow, 1958; Pederson & Breglio, 1968; Pederson & Higbee, 1969), while yet others reported no sex differences (e.g., Certner, 1973; Dimond & Hellkamp, 1969; Vondracek & Marshall, 1971). However, since no studies have reported greater disclosure by males, Cozby (1973) suggested that such findings may be indicative of sex differences. The socialization of males in this culture (Lynn, 1964) tends to support such a sex difference. Conflicting results in self-disclosure research have been due to several factors: (a) ambiguous, con- flicting, and statis definitions of the construct; (b) heavy reliance on subjective measures; (c) lack of behavioral measures; and (d) poor research methodology allowing competing explanations for results. Several highly controlled research studies on operant condition- ing of self-referenced affect, however, demonstrate that extraneous variables can be controlled. Auerswald (1972), Barnabei (1972), and Crowley (1970) have demonstrated that proportional emission of affect, self-referenced statements can be effectively regulated by the type of interviewer response used. Even if self-disclosure researchers had adhered to stringent experimental con- trol, problems would still exist. For example, even the research using performance measures has exclusively relied on frequency or proportional emission rates to assess self-disclosure (e.g., Axtell & Cole, 1971; Kaplan, 1967). Thus, this research has neglected to assess the quality of self-disclosure. The most thorough definition of self-disclosure was presented from a communications perspective by Pearce and Sharp (1973). According to them, self-disclosure occurs when an individual honestly tries to tell another person things about himself which the other is unlikely to know or find out from other people. Furthermore, self-disclosure is distinguished from nondisclosure, revealing, and confession. Because self-disclosure is defined as voluntary, it excludes confession. Revealing responses are classified as unintentional cues expressing information about the person. Honesty is characterized by a sincere attempt to make the speaker known to the listener and emphasizes present experiences dealing with subjectively oriented personal involvement (Moustakas, 1962). By limiting their review to studies that employed their definitional criteria, Pearce and Sharp (1973) were able to identify five characteristics of self-disclosing communication: (a) relatively few communication trans— actions involve high levels of disclosure; (b) self- disclosure usually occurs in dyads; (c) in a dyad, self- disclosure is usually symmetrical; (d) self-disclosure usually occurs within the context of positive social relationships; and (e) self-disclosure usually occurs gradually and covaries with the stability of the relationship. -‘fié' _ 10 SOcial Learning and Self- Discibsure If disclosure is assumed to be a learned response that is affected by complex interaction of external and internal processes governing each person, then empirical attention must be given to the processes underlying self- disclosing communication. ’ Within a social learning framework (Bandura, 1971b), three inter-related processes can be identified that regulate and maintain self-disclosure response patterns. The first is stimulus control. In order to function effectively in a dyadic interaction, both par- ticipants must be able to anticipate the probable conse— quences of their behavior and act accordingly. Since emotional responsiveness is typically mediated through thought-produced arousal without the presence of directly conditioned stimuli, the thought of self—disclosing may produce heightened emotional arousal, thus inhibiting self-disclosure. Self—disclosure is also controlled by antecedent stimuli that convey information regarding the probable response consequences for self-disclosing. Each participant encodes cues from the other and from the situation itself in order to evaluate probable conse- quences for self-disclosing. Another factor affecting stimulus control of self-disclosure involves modeling stimuli. If one person initiates self-disclosing behavior, the other may match this behavior. The second ll regulatory process is reinforcement control. All behavior, including self-disclosure, is extensively controlled by its consequences. Direct reinforcement of self-disclosing responses will tend to maintain and increase this behavior. Likewise, observing a model who is rewarded for self—disclosing responses can facilitate the same response in an observer. Either intrinsic or self-administered reinforcement for self-disclosure can facilitate this response even in the absence of external reinforcement. Finally, cognitive control affects the maintenance of a person's self-disclosing behavior. Since anticipated consequences are partly inferred from observed response consequences of others and a variety of other cues, actions may be guided by inaccurate anticipated consequences. When this occurs, behavior will be weakly controlled by its actual consequences until cumulative experiences provide more realistic outcomes. For instance, people may make inappropriate self-disclosures because they may have been told self- disclosure will make them popular. Until enough negative feedback for such inappropriate responding is perceived, these individuals will probably continue making this response. This psychological functioning involves a con- tinuous reciprocal interaction between self-disclosure and these three controlling processes. Bandura's social 1 12 learning model provides a conceptual framework for under— standing the process of self—disclosure. Social Learning Theory Since social modeling and cognitive structuring can be explained in terms of Bandura's social learning paradigm (1971b), this model will be considered briefly. Bandura's paradigm is schematically presented as Figure l. reinf. Anticipated S ._. Attention _§ S _. Symbolic 2:25.329 Cogn1t1ve Organiza— tion \Rehearsa Figure l. Bandura's social learning paradigm. According to Bandura (1969), vicarious learning is largely influenced by three sets of variables: (a) rein— forcement contingencies associated with behavior imitation in a specified setting, (b) model attributes, and (c) observer characteristics. The observer's anticipation of reinforcement for imitating a model's response will affect his level of attention to the model. The observer may anticipate direct reinforcement, such as social approval, or self— reinforcement, such as feeling good, for performing the matching response. Anticipation of positive reinforce- ment, therefore, can enhance the observer's learning by 13 increasing his attentiveness to the modeling stimuli. Furthermore, anticipated reinforcement can strengthen learning retention by motivating the person to code and rehearse modeled responses. Informing the observer in advance about the benefits for adopting modeled patterns of behavior, thus, should increase his attention to the modeling stimuli. Model characteristics can also influence the extent to which observers will imitate behaviors. Models perceived as possessing characteristics valued by the observer will be attended to more readily than those with negatively perceived attributes. A model's power to influence the responses of others can be mediated through competence, status, and interpersonal attractiveness. Observer characteristics, likewise, can influence vicarious learning. If an observed behavior is per— ceived as having appropriate or utilitarian value, an individual will more likely incorporate the modeled response into his repertoire. People who lack self- esteem (Gelfand, 1962), feel incompetent (Kanareff & Lanzetta, 1960), are highly dependent (Jakubczak & Walters, 1959), and have often been rewarded for imi- tation (Masters & Morris, 1971) are most likely to imitate model performance. Similarity between observer and model in age (Hicks, 1965; Kazdin, 1974b), and sex (Bandura, Ross, & Ross, 1963; Kazdin, 1974b), as well 14 as on other perceived similarities (Burnstein, Stotland, & Zander, 1961) has;also been demonstrated as facili- tating imitation. However, attention to the modeling stimuli is not sufficient for observational learning to occur. The observer must be able to properly encode the modeling stimuli in symbolic form so that he or she can produce the response at a later time. This cognitive retention process involves verbal and imaginal coding. Most cog— nitive processes regulating behavior are verbal (Bandura, 1971b). After modeled stimuli are transformed into images and verbal symbols, these cognitive memory codes become guides for the observer to subsequently reproduce the vicariously learned response. In addition, both mental (covert) and actual (overt) rehearsal of the modeled patterns of behavior will facilitate memory retention. In order to successfully implement a vicariously learned response, the observer must be able to combine a given set of responses according to the coded model patterns. The learner must possess the necessary sub- skills to effectively reproduce the modeled response and be physically capable of carrying out the process. A final component involves the observer's ability to refine rough approximations of the newly acquired 1 Me- 15 behavioral response. In most cases, learners will have to rely on feedback from the environment to make these refinements. Finally, for the imitative response to be overtly practiced, the learner will have to perceive some benefit for implementing it. If aversive consequences accrue from performing the modeled behavior, the person is less likely to continue performing it. Social Modeling According to Bandura (1971a, 1971b; 1969), vicarious learning can produce three differential effects in observers: (a) the modeling effect in which an indi- vidual acquires responses that are novel to him; (b) the inhibitory and disinhibitory effects in which responses already well established in the observer's repertoire are weakened or strengthened, and (c) the response facilitation effect in which previously learned responses are elicited through model observation. In response facilitation, no new responses are acquired, and dis- inhibitory processes are not involved, because the response is socially acceptable and, therefore, unencumbered by restraints. Expression of affective self-disclosure is classified under response facilitation effects of modeling since this response is socially sanctioned. 16 The review of literature on social modeling procedures, therefore, is focused primarily on studies dealing with social response facilitation of established behaviors. These studies are reviewed under two headings: (a) Facili- tation of self—disclosure in interview situations, and (b) Facilitation of social skills across settings. Facilitation of self-disclosure in interview settings. The majority of social facilitation modeling studies has dealt with increasing the self—referenced affect of subjects in individual or group counseling settings. Wilder (1968) compared the effects of verbal modeling and verbal reinforcement on the frequency of self-referred affect statements made by female college students. He found that verbal modeling increased the frequency of self-referred affect verbalizations in quasi-counseling interviews, while verbal reinforcement did not. In a closely related study, Brody (1968) found that greater and more enduring increases in self— referent verbalizations can be achieved by combining modeling procedures with social reinforcement of similar verbal responses. The efficacy of modeling procedures to facilitate self-disclosure can be augmented through vicarious rein- forcement procedures as well as through the direct rein- forcement method shown by Brody (1968). Marlatt, Jacobson, Johnson, and Morrice (1970) and Marlatt (1970a) 17 demonstrated that subjects revealed more personal problems after seeing a brief waiting-room conversation in which a model's personal disclosure was either neu— trally or positively reinforced than if the model's disclosure was negatively sanctioned or if subjects were not exposed to the model. In addition to social modeling and reinforcement, instructions can increase verbal self-disclosure. Merbaum and Lukens (1968) demonstrated that instructions are superior to reinforcement in eliciting affective verbal behavior. Furthermore, Green (1972a, 1972b) found that optimal versus minimal instructions were effective for increasing self-disclosure and that modeling, and modeling combined with either vicarious reinforcement or commentator guidelines were equally effective. However, the most facilitative effect was a combination of optimal instruc- tion and one of the modeling conditions. Further comment on Green's study is in order because of its relevance to this investigation. The dependent measure, talking into a tape recorder for 15 minutes, provided a very low-level simulation of dyadic interpersonal communication. This duration made the verbal task a lengthy private monologue that seldom, if ever, occurs in real life. Interjecting stimuli for the subject to react to would have improved the quality of this measure. Although the dependent variables con- sidered proportional amount of feeling statements emitted, 18 the investigator neglected to assess the quality of affective self-disclosure. Also, the time—sampling analysis procedure sacrificed raw data for the sake of investigator convenience. Although the reliability for this measure was .87, a higher coefficient most likely could have been obtained if typescripts had been rated instead of the audiotapes. Model and observer characteristics were examined by Sarason, Ganzer, and Singer (1972). They found that a nondefensive model facilitated negative self-disclosure made by low and high defensive subjects. In addition, results showed that only nondefensive models facilitated negative self-disclosures in highly defensive subjects. Modeling has proved effective in clarifying the client's role in counseling (Marlatt, 1970b). Clinical field studies have also been conducted in which modeling procedures have been used to teach psychotic patients and institutionalized delinquents what to expect from therapy (Truax, Wargo, Carkhuff, Kodman, & Moles, 1966). Finally, modeling procedures have been used in group settings to facilitate self-disclosure. Whalen (1969) found that neither instructions nor models alone could produce high levels Of self-disclosure. However, a combination of detailed, exhortative instructions plus modeling did produce the desired effect. Her results point to the importance of context and setting as factors influencing self—disclosure, since within the context of 19 an individual interview modeling has proved sufficient to elicit self-disclosure (Brody, 1968; Marlatt et al., 1970; Wilder, 1968). Schwartz and Hawkins (1965), using two patients as models of affective verbalizations and two who lacked affect and a therapist dispensing reinforcements, suc- ceeded in either increasing or decreasing adult schizo- phrenics' affect statements. These results demonstrate that significantly different group behavior can occur under the same reinforcement conditions, depending on the behavior of influential models. In social situations, therefore, behavior always remains partly under modeling stimulus control. Reinforcement procedures may prove relatively ineffective if influential models display behavior counterproductive to the aims of change agents. Facilitation of social skills across settingy. Based on the Schwartz and Hawkins' (1965) findings, one (can conclude that individuals may fail to exhibit responses within their capability because of inadequate reinforcement or lack of appropriate models. This conclusion is supported by a study comparing the effects of reinforced modeling alone on the popularity of chil- dren. Hansen, Niland, and Zani (1969) sociometrically evaluated changes in children's social status produced by three types of group counseling. They found that 20 children who received modeling and reinforcement achieved a significant rise in sociometric status that was main- tained at a two-month follow—up assessment. Guided practice as an adjunct to modeling has been proven effective. The efficacy of modeling with guided performance on increasing assertive behavior was demonstrated by Friedman (1968). He found that behavioral versus verbal modeling when combined with behavior rehearsal was the most powerful treatment. Bale (1972) examined the effect of discrimination training, guided practice, and role-playing in modeling procedures designed to teach socially withdrawn males how to make personal feeling questions. Both guided practice and role-playing when combined with discrimi- nation training proved to be highly effective for increasing frequency of personal feeling questions in socially withdrawn males. These findings, therefore, support those of Friedman (1968) by suggesting that modeling procedures are most powerful when supplemented with some form of behavior rehearsal. Model characteristics for eliciting personal feeling questions were examined by Stuhr (1972). He found that both mastery and coping model conditions were effective for increasing personal feeling questions in socially withdrawn males. These findings suggest that coping and mastery models may be essentially 21 equivalent when teaching conversational skills. As the investigator noted, whether increases in personal feeling questions affect quality and generalize to in vivo con— versations needs further study. Inclusion of a quality measure would have strengthened this study. The relative efficacy of the mastery and coping modeling conditions is interesting. Other studies (e.g., Meichenbaum, 1971; Kazdin, 1973a) have suggested that coping models are superior to mastery models in avoidance reduction of feared stimuli. Are the discriminations in conversa— tional skills too fine to be significant? Would con- trasting models demonstrating appropriate and inappro- priate social responses be more effective than coping or mastery models? Clearly, more empirical investigations are needed to clarify these issues. The social modeling studies reviewed clearly demonstrate that modeling can facilitate self-disclosure and other social behaviors. Likewise, modeling when combined with reinforcement (direct or vicarious) or behavior rehearsal served to increase verbal disclosure. The role of instruction in self-disclosure is not yet clear, but several studies suggested that it has some facilitative effect (Green, 1972a, 1972b; Merbaum & Lukens, 1968; Whalen, 1969). Behavioral measures have tended to rely on frequency or proportional emission data of self-disclosure. Self-referenced feeling statements 22 and personal feeling questions have been used as depen- dent measures. The quality dimension of affective self- disclosure continues to remain unassessed. Research Implications for Modeling Treatments A cursory examination of research across the three types of modeling effects was conducted to assess salient factors that facilitate vicarious learning. Results are briefly cited and are arranged under three headings: (a) Nature of modeling experience, (b) Effects of reinforcement, and (0) Effects of overt rehearsal. Nature of modeling experience. The circumstances surrounding vicarious learning are important in helping a person learn target responses. The following list summarizes relevant research findings on factors that facilitate the modeling process. 1. Live and symbolic modeling proCedures are likely to produce equivalent levels of observational learning (Bandura & Mischel, 1965; Bandura, Ross, & Ross, 1963). 2. The desired behavior should be repeatedly modeled, preferably by multiple models who demonstrate progressively more difficult performance (Bandura & Menlove, 1968; O'Connor, 1969). 23 3. Presentation of both inappropriate and appropriate modeling responses allows the observer to discriminate relevant behavioral cues associated with positive response consequences (Debus, 1970). 4. A coping model can be more effective than a mastery model in facilitating snake approach responses (Meichenbaum, 1971). 5. Mastery and coping models are equally effec- tive in eliciting personal feeling questions (Stuhr, 1972). 6. A mastery model is more effective than a fearful model (Geer & Turtletaub, 1967). 7. Behavioral modeling can be enhanced by the use of narratives (Bandura & Menlove, 1968). 8. Generalization of imitation increases when the modeling stimulus situation is similar to the original, real—life situation (Croner & Willis, 1961). 9. In a comparison of the strength of obser- vational learning manifested under different incentive conditions, increased imitation is expected under suc- cessively stronger incentive conditions. The results of Bandura (1965), Hicks (1965), and Mischel and Grusec (1966) showed more imitation manifested under maximal than under minimal or moderate incentive conditions. 24 10. Modeling procedures can induce behavioral, affective, and attitudinal changes in fearful subjects (Bandura, Blanchard, & Ritter, 1969). Effects of reinforcement. Direct, vicarious, and self-reinforcement procedures are powerful adjuncts to social learning. Research findings summarized below illustrate the effectiveness of reinforcement when coupled with modeling procedures. 1. Direct: Matching responses are performed at a high level when they produce rewarding outcomes, whereas modeled behavior is seldom reproduced when it results in punishment or goes unrewarded (Baer, Peterson, & Sherman, 1967; Kanareff & Lanzetta, 1958, 1960). 2. Vicarious: Observation of rewarding or punishing consequences to a model can substantially affect the extent to which observers willingly engage in imitative behavior (Bandura, 1965). The inhibiting effect of anticipated negative consequences can be over- come in observers through positive reinforcement of the model's responses (Walters & Parke, 1964). 3. Vicarious versus direct: Behavioral changes displayed by observers are generally of the same magni- tude as those achieved by reinforced performers (Kanfer & Marston, 1963), or under certain circumstances, may even exceed them (Berger, 1961). 25 4. Self: Individuals tend to adopt standards of self-reinforcement displayed by appropriate models, evaluate themselves in terms of that standard, and then serve as their own reinforcing agents (Bandura & Kupers, 1964). Desired responses can become partially independent Of external consequences through the development of self— regulatory functions (Bandura, 1969). Effects of overt rehearsal. Modeling procedures are most powerful when supplemented with guided perfor- mance in acquiring novel responses (e.g., Lovaas, Freitag, Nelson, & Whalen, 1967), eliminating inhibitory responses (e.g., Rimm & Mahoney, 1969), and facilitating socially sanctioned responses already within a person's repertoire (e.g., Bale, 1972). Cognitive Structuring As previously mentioned, Bandura's social learn- ing paradigm emphasizes the role of cognitive mediating factors in observational learning. Two studies have pro— vided evidence regarding the role of symbolic represen- tation in observational learning. Bandura, Grusec, and Menlove (1966) found that children who generated verbal equivalents of the modeled responses during presentation later reproduced more matching responses than children who attentively observed the modeled displays without verbalization. In turn, this latter group achieved a 26 higher level of observational learning than those engaged in symbolic activities designed to prevent implicit verbal coding of modeled behavior. Gerst (1971) compared three symbolic coding procedures-~summary labeling, imaginal coding, and verbal description—-in the acqui- sition of novel responses through observational learning. All three coding operations facilitated immediate repro- duction of modeled responses, with imaginal and summary label codes being the most effective. In a delayed test, the summary labeling condition proved superior to the other conditions, which did not vary from each other. Clearly, then, evidence suggests that cognitive mediating factors influence vicarious learning. The review of literature on cognitive structuring focuses on three areas dealing with cognitive mediating factors: (a) Cognitive restructuring, (b) Self- instruction, and (c) Cognitive self-modeling. The first two emphasize the verbal cognitive mediating component, while the latter focuses on the imaginal one. Cognitive restructuring. Central to the Rational Emotive Theory of psychopathology (Ellis, 1962) is the notion that thought must precede and accompany emotion and that irrational self-verbalizations maintain negative and self-defeating behavior. These ideas have been sum- marized into the A-B-C theory of emotional disturbances, 27 where A is the objective, experiential event; B the cog— nitive self-statements regarding event A; and C the emotional response generated by B. Rational—Emotive-Therapy (RET) focuses on B of this process, that is, the irrational self-statements a person makes about a specific event. RET, therefore, emphasizes the cognitive mediating factors identified in Bandura's social learning model. This emphasis on cog- nition is consistent with the social learning paradigm which views emotional responsiveness as typically being mediated through thought—produced arousal rather than being directly evoked by conditioned stimuli. There— fore, inattention to environmental stimuli, inaccurate coding of stimulus material, retentional deficiencies, and inferential errors such as inaccurate anticipation of consequences are possible explanations, couched in social learning terminology, for resultant irrational self-verbalizations. The irrational values and ideas Ellis (1970) considers the root of emotional disturbance can also be explained through social learning tenets. The behavioral standards people set for themselves represent these values and the anticipatory self" satisfaction and self-criticism for actions correspond- ing to or deviating from these adopted standards serve as the controlling influences. Take, for example, people who adopt the behavioral standard that they 28 should be thoroughly competent in everything they do. Their anticipatory self-criticism for actions that deviate from this standard will, therefore, serve as the controlling influence in determining internalized self-verbalizations which, in turn, will affect their behavior. The main thrust of cognitive restructuring therapy, then, is to logically show clients that certain self-adopted behavioral standards (values) are so unrealistically unattainable that self—criticism for failure to meet these standards is inevitable. Unfortunately, experimental research evaluating the effectiveness of cognitive restructuring has been minimal. Those studies reported typically employed poor research methodology; hence, data implications are limited. Two studies examined topics that might be classified under response facilitation. Baker (1966) evaluated the effects of cognitive restructuring on the frequency of positive self-references made by college students. Results showed that all subjects receiving the cognitive restructuring treatment significantly increased their positive self-references. Since this was an analogue study, the generalizability of the results to a clinical population should be made with caution. DiLoreto (1971) compared the efficacy of systematic desensitization, client-centered therapy, rational-emotive therapy. attention-placebo, and 29 no-contact control in reducing interpersonal anxiety. All three treatment groups were significantly more successful than the two control conditions, with systematic desen- sitization effecting greater anxiety reduction and rational-emotive therapy producing greater inter- personal activity. Since no direct performance measures were employed, this study's significance is somewhat diminished. Six of the studies reviewed dealt with inappro- priate emotional arousal: speech anxiety (Goldfried, Decenteceo, & Weinberg, 1974; Karst & Trexler, 1970; Trexler & Karst, 1972), text anxiety (Maes & Heimann in Rimm & Masters, 1974; Montgomery, 1971), and rat phobia (D'Zurilla, Wilson, & Nelson, 1973). The Goldfried et al., (1974) and Montgomery (1971) studies confined out- come assessment to paper-and—pencil measures, thus limit— ing the implications of the results. In the remaining four studies, which included both self-report and per- formance measures, three suggested that cognitive restructuring effects may be most apparent in the area of subjective distress rather than task performance (Karst & Trexler, 1970; Trexler & Karst, 1972; D'Zurilla et al., 1973). As reported by Rimm and Masters (1974), a 1970 study by Macs and Heimann demonstrated RET as being effective in reducing physiological arousal, but less impactful on reducing subjective distress. 30 Finally, in a laboratory analogue study, Burkhead (1970) found that both live and taped RET procedures effected significant anxiety reduction on physiological and subjective arousal measures. The research reviewed suggests that cognitive restructuring procedures seem to be most frequently employed with inappropriate emotional arousal and tends to rely heavily on subjective assessment of treatment effects. If anxiety is a contributing factor to the inhibition of affective self—disclosure, then RET principles may be appropriate in disinhibiting dis- closure. Self—instruction. Self-instructional training, like cognitive restructuring, emphasizes the role of private monologues in subjective distress and provides training in systematic observation leading to alteration of dysfunctional thought patterns. However, self- instructional training emphasizes the use of graduated tasks, cognitive modeling, directed mediational training, and self-reinforcement for improvement, whereas cog- nitive restructuring primarily relies on logical self- examination through client-therapist Socratic dialogue. The most extensive and impressive work in this area has been conducted by Donald Meichenbaum. Self- instructional packages have been used effectively in reducing impulsivity in children (Meichenbaum & Goodman, 31 1969, 1971), test anxiety (Meichenbaum, 1972; Sarason, 1973; Wine, 1971), speech anxiety (Meichenbaum, Gilmore, & Fedoravicius, 1971), and snake avoidance (Meichenbaum, 1971). Self-instructional training has been primarily used to reduce inappropriate anxiety and to increase task performance. Only one published report specifically dealt with dysfunctional self-verbalizations in social interactions. Meichenbaum and Cameron (1973) conducted two studies to assess the impact of self-instructional training on the attention, thought, and language patterns of institutionalized schizophrenics. They found in their first laboratory study that cognitive training improved schiZOphrenics' performance on two perceptual tasks. From results of the second study, which assessed the effects of extended training in self-instruction, they concluded that this cognitive training package effected significant decreases in "sick talk" during interviews and effected improvements on three of the other four dependent measures. More importantly, however, data collected in a three-week follow-up were interpreted to indicate that schiZOphrenics in the self-instruction group improved. As reported by Mahoney (1974), Meichenbaum (in press) has discussed implications for self-instructional training of the elderly who frequently make dysfunctional self-verbali- zations. 32 Like cognitive restructuring, self—instructional strategies focus on the internal verbalizations clients make to themselves. Altering these self-verbalizations may be one way of increasing affective self-disclosure. However, no research has been conducted on this topic. Cognitive self-modeling. According to Bandura (1970), the degree of observational learning can be enhanced through overt and covert behavior rehearsal. However, covert rehearsal may be of greater importance since an observer can readily engage in mental rehearsal at times when overt rehearsal is either impeded or impractical. Unfortunately, almost no empirical research has been conducted on covert selffmodeling, although as a component variable in systematic desensitization (Wilkins, 1971), it has been frequently employed. Several studies have been conducted on covert modeling, where subjects imagine a model other than themselves performing a fear avoidance task. Cautela, Flannery, and Hanley (1974) demonstrated that both overt and covert modeling pro— cedures were effective in reducing rat avoidance in female college students. However, the most impressive evidence of the efficacy of covert modeling has been demonstrated by Kazdin (1973a, 1974a, 1974b) in a series of highly controlled studies on snake avoidance. 33 Kazdin (1973a) found that a covert coping model was sig- nificantly more effective in improving snake approach behavior than a covert mastery model. In a second study, Kazdin (1974a) found that age and sex similarities between the "covert" model and observer were more important than the coping—mastery distinction. However, greatest improvement tended to occur in the coping, age-sex similarity condition. In a subsequent replication study (Kazdin, 1974b) no difference was found between covert selffmodeling and covert modeling treatments in effecting snake avoidance reduction behavior. Again, imagining a coping model proved superior to imagining a mastery model. As reported by Mahoney (1974), Kazdin examined the effectiveness of covert modeling in the development of assertiveness. Results showed that covert modeling, both alone and when combined with covert reinforcement, significantly increased assertive behavior. With only one report on the effects of cognitive self-modeling, it is premature to make definitive statements regarding the merits of this procedure. However, conceptually, cognitive self-modeling should provide a person with an opportunity to rehearse desired responses. Given that a person has the necessary information as well as the requisite skills to perform a desired response, self-modeling may facilitate the maintenance and generalization of response improvements. 34 Cognitive restructuring, self-instruction, and covert self-modeling procedures can be explained in terms of Bandura's social learning model. However, almost no research has been conducted on response facilitation behaviors using cognitive structuring techniques. Most of the research has dealt with anxiety reduction, although self-instructional training has been utilized in teaching some performance tasks. If cognitive mediating factors affect response facilitation behavior as suggested by Bandura's model, then the cognitive structuring pro- cedures may be viable methods for increasing affective self-disclosure. One purpose of this study was to empirically investigate these possibilities. Summary The rationale for this research is based on the following needs: (a) the absence of a specific oper- ational definition of affective self-disclosure, (b) the need for empirical research on affective self-disclosure using performance measures, (c) the absence of assessment procedures to measure the quality of self-disclosure, and (d) the need to examine the differential effects of overt and covert components of Bandura's social learning para— digm. Social learning research has demonstrated that instruction, modeling (overt and covert), and behavior 35 rehearsal independently and in combination are effective methods for teaching a variety of responses. The decision to use social modeling and cognitive structuring multi— component strategies in this study was based on the research reviewed in this chapter. Although most of this research was conducted on topics unrelated to self- disclosure, it was felt that the strategies should be equally effective for increasing the affective self- disclosure of undergraduate college males. The population of single undergraduate college males needs and uses affective communication skills daily. However, some research evidence has suggested that males in this culture tend to disclose their feelings less frequently than females. This population, there- fore, may especially benefit from affective communication skills training. CHAPTER II EXPERIMENTAL DESIGN AND PROCEDURES Overview The primary purpose of this study was to investi- gate whether social modeling and cognitive structuring multi-component strategies are effective methods for increasing affective self-disclosure of undergraduate, college males. In order to attain this objective, an operational definition of affective self-disclosure was formulated, and five criterion measures were develOped to assess it. The research was conducted with 48 undergraduate male volunteers at Michigan State University during the 1975 spring quarter. Subjects were randomly assigned to one of the four treatment conditions--social modeling (SM), cognitive structuring (CS), attention-placebo (AP), and no-treatment control (NTC). Subjects in the social modeling and cognitive structuring conditions viewed videotape presentations of comparable length. Attention-placebo subjects listened to a deep muscle relaxation tape of comparable duration. Subjects in the no-treatment control condition only received testing. 36 37 The testing phase consisted of three tasks. First, the subject listened and responded to 16 audio— taped stimulus situations which simulated dyadic inter— actions between the subject and his best male or female friend. Situations were constructed to equally require positive and negative feeling responses. In addition, the subject made responses as both a respondent to and initiator of feeling statements. Each response was then evaluated for amount and quality of affect. During this audiotape phase, the subject recorded the level of anxiety he felt in each situation. This constituted the second task. Finally, each subject completed two paper- and-pencil measures. The first, a multiple choice skill test, ascertained his ability to discriminate the most appropriate way to disclose feelings. The second, a Likert scale attitude instrument, measured his attitudes toward disclosing feelings. The three paper-and-pencil measures were then scored. Audiotaped responses were transcribed, and two raters were trained to rate them for amount and quality of affect. Hoyt reliability coefficients were then computed for the two phases of the training program and for the experimental data typescripts. Reliability estimates were made for the total amount and quality scores as well as for the five quality subscales. A 38 one-way multivariate analysis of variance using ortho- gonal planned comparisons was conducted to test for the treatment effect. Three weeks later, those subjects who did not receive a self-management treatment (Voight, 1975) were retested to ascertain the maintenance of the treatment effect. A repeated measures multivariate analysis of variance test was then conducted to determine potency of treatment over time. Three supplementary questions of interest were also explored. Since none of these questions were included in formal hypotheses, these results are reported as additional information only. The first dealt with factors influencing affective dyadic communication. Items from both the performance audiotape (amount, quality, and anxiety) and skill tests were analyzed to test for differences between type-of-role (initiator, respondent), type-of—feeling (positive, negative), and sex-of-best-friend (male, female). A repeated measures multivariate analysis of variance procedure was used to test this difference. The second question concerned the measure of quality used in the investigation. Would subject meag_ instead of total quality scores produce a differential effect among treatment conditions? To answer this 39 question, a one-way analysis of variance procedure was conducted using the mean quality score as the dependent measure. Third, the researcher was interested in subject assessment of the experiment. Immediately following de- layed posttesting, the 13 subjects who did not receive a self-management procedure (Voight, 1975) completed a debriefing questionnaire. Descriptive statistics on these data were then compiled. Pretest of Measures Both the skill and attitude measures were pre- tested on a sample population (NEZO) of undergraduate males enrolled in psychology courses at Olivet College. This pretest was conducted prior to the pilot study. None of these subjects received an experimental treatment. Thus, each only completed the two paper-and~pencil measures. Hoyt reliability coefficients (Guilford, 1954) were computed for both measures. Reliability estimates were: r = .31 for the skill test and r = .83 for the attitude survey. Pilot Study Prior to the actual study, the treatments and dependent measures were pilot tested. Subjects for this pilot study were 20 males who were either graduate students in counseling or Juvenile Court caseworkers. Each subject was randomly assigned to one of the four treatment 40 conditions. Results from this pilot study were used to improve and refine the dependent measures. In addition, reliability coefficients for the skill and attitude measures were obtained. Finally, results from the one- way multivariate analysis of variance were used to formu- late directional hypotheses for the actual study. For both the skill and attitude measures, indices of item discrimination and difficulty (Mehrens & Lehmann, 1973) were computed. Only the most discriminating paper- and-pencil items were retained for the actual study. Hoyt reliability coefficients (Guilford, 1954) for both paper- and-pencil measures were also computed. Reliability esti- mates were: 3 = .91 for the skill test and r = .86 for the attitude scale. Three planned contrasts were tested using a multi- variate analysis of variance procedure. The first contrast between the two experimental treatments and the two control groups was significant, F (5, 12) = 6.10, p < .005. The second comparison between cognitive structuring and social modeling was also significant, F (5, 12) = 4.52, p < .02. Finally, the third contrast between attention-placebo and no-treatment control was not significant, F (5, 12) = 1.35, p < .31. Based on these results, the directional hypothe- ses for the actual study were formulated. 41 Hypotheses Two questions of general interest were addressed in this study with relation to undergraduate college males. They were: (a) Will Treatment strategies employing social learning principles be effective for increasing affective self-disclosure of males? More specifically, will social modeling and cognitive structuring treatment packages be more effective than attention-placebo and no-treatment conditions? (b) How potent are social modeling and cognitive structuring treatments over time? The effects of social modeling, cognitive structuring, attention-placebo, and no-treatment control conditions on the affective self-disclosure of college males were tested by assessing: (a) amount of affect, (b) quality of affect, (c) concomitant anxiety level, (d) skill level, and (e) attitudes toward disclosing feelings. Based on results of a pilot test, the following hypotheses were formulated for the first question of interest. The first two are directional, the third non- directional. Hypothesis 1: The main effects of the social modeling and cogni- tive structuring groups will be greater than those of the attention-placebo and no-treatment control groups. 42 Hypothesis 2: The main effect of the cognitive structuring group will be greater than that of the social modeling group. . Hypothesis 3: There will be no difference between the main effect of the attention-placebo and no-treatment control groups. For the second question of interest, the follow- ing hypotheses were formulated. Hypothesis 1: There will be no treatment-by-measures interaction. Hypothesis 2: There will be a treatment effect. Hypothesis 3: There will be no measures effect. Definition of Terms The following key terms have been repeatedly used throughout the study. In order to provide a common basis for understanding, they are defined below. 1. affective self—disclosure--a speaker's volun— tary verbal statement made as an initiator or respondent in a dyadic interaction which expresses his emotions in feeling terms, is present-oriented, and self-referenced (using the word "I"). 43 2. amount of affect--the number of responses containing at least one verb, adverb, or adjective used as an affect word which expresses emotion in the context of the response given by the Speaker in a simulated dyadic interaction. 3. quality of affect--in a simulated dyadic interaction the content of the speaker's responses pre- viously classified as containing amount of affect is rated for quality on the following dimensions: (a) refer- ence, (b) time orientation, (c) apprOpriateness, (c) reason, and (e) specificity of reason. 4. concomitant anxiety level--the Speaker's self-reported anxiety level on a 1-7 scale in simulated dyadic interactions. 5. skill leve1--the subject's skill in dis- criminating affective self-disclosure statements as measured by a paper-and-pencil test. 6. attitudes toward disclosing feelings--the subject's evaluative responses regarding affective self- disclosure as measured by a paper-and-pencil instrument. 7. social modeling treatment--a videotaped multi-component treatment package emphasizing learning affective self-disclosure through overt vicarious con- ditioning procedures.‘ This package consisted of an introduction and four discrete learning components: (a) instruction, (b) behavioral modeling, (c) overt behavior rehearsal, and (d) a review of the procedures 44 covered in the learning package. Paper-and-pencil self-tests were given at the end of parts a and b, with provision for recycling the subject through the component if the minimum criterion of acceptable performance was not met. 8. cggnitive structuring treatment-~a videotaped multicomponent treatment package emphasizing learning affective self-disclosure through both verbal and imaginal covert cognitive mediating processes. This package consisted of an introduction and four discrete learning components: (a) instruction, (b) behavioral and cognitive modeling, (c) cognitive self-modeling, and (d) a review of the procedures covered in the learning package. Paper-and-pencil self-tests were given at the end of parts a and b, with provision for recycling the subject through the component if the minimum criterion of acceptable performance was not met. 9. attention-placebo treatment—-an audiotaped presentation of training in deep muscle relaxation which was included to control for any demand characteristics or cues which influenced a subject's perception of his role in the experiment. 10. no-treatment--a waiting list control group that only received posttesting. ll. immediate posttest--a performance test, measuring amount and quality of affect and the subject's concomitant anxiety, as well as two papereand-pencil 45 measures on skill and attitudes. These five measures were given immediately following treatment. 12. delayed posttest—-the same performance and paper-and-pencil instruments were readministered to those subjects not receiving a self-management follow-up pro- cedure (Voight, 1975) approximately three weeks after the immediate posttest. 13. follow-up treatment--any one of three treat- ment packages. These treatments, defined by Voight (1975), are: (a) self-reinforcement, (b) goal-directed practice, and (c) traditional follow-up. Design Over Time This study employed a posttest-only control group experimental design with one independent variable con- sisting of four levels, as shown in Figure 2. A delayed posttest was administered to those subjects who did not receive treatment in the self-management follow-up study by Voight (1975). Random assignment of Se to the four treatment groups permitted the absence of pretest measures . Data for the major question of interest regarding the treatment effect were analyzed separately utilizing a posttest-only control group design. .This design con— trolled for all sources of internal invalidity as well as the interaction of testing and treatment threat to 46 external validity (Campbell & Stanley, 1963). There were no other major concerns for the validity of this experi- mental phase. R X 0 O 1 1 1' R x2 02 02, R x3 03 03. R x4 04 04. Figure 2. Posttest-only control group experimental design over time. Legend: R = random assign— ment, X = treatment: X1 = social modeling package, X2 = cognitive structuring package, X3 = attention-placebo, X4 = no-treatment control. Ol--O4 = immediate posttesting; Ol.--O4u = delayed posttesting. The second question of interest regarding main— tenance of treatment effects over time involved analyzing repeated measures--the immediate and delayed posttest-- on those subjects receiving no treatment in the self- management follow-up study (Voight, 1975). Sources of internal invalidity for this immediate-delayed posttest follow-up were controlled (Campbell & Stanley, 1963). However, because of a possible testing-by-treatment interaction, results of the follow-up assessment can only be generalized to persons who are tested more than once after they receive treatment. There were no other major concerns for the external validity of the follow- up phase of this experiment. 47 Design over Variables The variable matrix for the first question of interest is presented as Figure 3. It takes the form of a one-way design with 12 subjects in each cell. The design variable was type of treatment and consisted of four levels: social modeling, cognitive structuring, attention-placebo, and no-treatment control. TYPE OF TREATMENT SOCIAL COGNITIVE ATTENTION- NO— MODELING STRUCTURING PLACEBO TREATMENT S1 S13 S25 S37 S12 S24 S36 S48 Figure 3. One-way design. Equal nfs per cell. The dependent variables were: (a) amount of affect, (b) quality of affect, (c) concomitant level of anxiety, (d) affective self-disclosure skill level, and (e) attitudes toward disclosing feelings. A multivariate one—way analysis of variance (MANOVA) procedure was used to analyze the data. The alpha level for this MANOVA procedure was .05 for a one- tailed test. The second question of interest concerned main— tenance of treatment effect between immediate and 48 delayed posttesting. The data matrix, presented as Figure 4, takes the form of a repeated measures design with four subjects each in the social modeling and cognitive structuring conditions and three and two respectively in the attention-placebo and no-treatment control conditions. The design variable was type of treatment. The repeated measures variable was time of posttest. The dependent variables were the same. A repeated measures multivariate analysis of variance procedure was used to analyze the data. The alpha level for this analysis was .05 for a one-tailed test. TIME OF POSTTEST IMMEDIATE DELAYED S SOCIAL 1 MODELING 2 54 E4 Z s E COGNITIVE 5 g STRUCTURING : m S m 8 [-4 ELI o ATTENTION- S9 m PLACEBO I a S11 54 NO- S12 TREATMENT 2 S13 Figure 4. One-way repeated measures design. Unequal 3's per cell. 49 Sample A sample of 52 volunteer §s was obtained from the undergraduate male population at Michigan State University during the 1975 Spring quarter. However, because of a tape recorder malfunction, data were incomplete for four sub- jects. Therefore, a total of 48 subjects were used for actual data analyses for the first question of interest. For the follow-up, only 13 subjects returned. Data from all 13 subjects were analyzed, thus creating an unbalanced design. If subjects had been randomly deleted to create equal cell size, statistical power would have been drasti- cally reduced. Hence, the decision was made to analyze all data. A thorough delineation of sample selection proce- dures, sample characteristics, and subject assignment follows in order to permit readers to judge how this popu- lation compares with populations to which they might wish. to generalize (Cornfield & Tukey, 1956). Sample Selection Procedures Initially volunteers who participated in this research project were located in response to a 3 l/2" x 3 l/4" advertisement placed in the State News, the Michigan State University student neWSpaper (see Appendix A). This advertisement ran two days. Concur- rently, flyers were distributed to Michigan State Uni- versity residence halls and posted in Erickson Hall, the College of Education building (see Appendix B). 50 When these efforts failed to produce a sufficient number of subjects, a classified advertisement was placed in the State News which appeared for five days. The text of this advertisement appears in Appendix A. In addi- tion, an undergraduate male recruiter was paid to make announcements in various classes and make personal con- tacts with undergraduate males in residence halls. In all efforts to secure subjects, the experiment was called a research/training project, focusing on the expression of feelings to others. The only criteria for participation were that volunteers be single, male under- graduate students at Michigan State University. Further- more, volunteers were informed of the number of sessions, the total time commitment, and that the training was free. Thus, all subjects knew the topic of the study and the total time commitment necessary for participation prior to their involvement in the experiment. The flyers and advertisements listed the investi- gator's phone numbers in Erickson Hall, as well as her home phone number. The majority of men who contacted the investigator stated they did so in response to the per- sonal contact made by the paid recruiter and to one of the two advertisements in the State News. In all cases, the investigator questioned the men in order to determine their suitability for the study. The relevant criteria included: single marital status, male sex, Michigan 51 State University undergraduate student status, knowledge of study's purpose, and agreement to study's time commit- ment. Those men who agreed to participate provided the investigator with their names and telephone numbers. At the time of this first phone contact, volunteers were scheduled for their first session. Each subject received a phone call to remind him of this appointment. At the end of the first session, a three-week follow-up appoint- ment was scheduled for each participant. Several days before the second session, each subject received a letter reminding him of the date and time of the follow-up session (see Appendix C). In order to insure a high return rate, subjects were offered a token reward for keeping this appointment. Finally, each participant was contacted by phone to remind him of his second appointment time. Sample Characteristics 1. Sex: 48 males 2. Age: range = 18-26 years; mean = 21.5 years; median = 21 years 3. Marital status: 48 single 4. Class standing at Michigan State University: 4 freshmen, 9 sophomores, 13 juniors, 16 seniors. Since five subjects did not complete the class standing portion of the debriefing questionnaire, these 52 sample characteristics are based on 43, instead of 48, subject-responses. Approximately 25% of the subjects were advisors in residence halls; hence, the subject sample contained more juniors and seniors than freshmen and sophomores. In addition, the volunteers who agreed to participate are probably older than the average age for their class rank. Most likely, these two factors inflated both the mean and median age of subjects. Subject Assignment Prior to initial phone contact with participants, a randomized treatment schedule was constructed. Thus, when a caller decided to participate, he was immediately assigned to one of the four treatment conditions. This initial random assignment procedure was also used for the follow-up study by Voight (1975). Hence, both investi- gators knew the two treatment conditions for each subject after the first phone contact. For the first phase of the experiment, all data for 48 subjects were collected. Since only 13 of 16 sub- jects who did not receive a follow-up treatment (Voight, 1975) returned for the second session, only data on these 13 subjects were collected and analyzed. Treatment Subjects received one of four treatments. The social modeling and cognitive structuring treatments were presented on videotape, the attention-placebo on audiotape. 53 These three treatment conditions were of comparable length, each lasting approximately 50 minutes. Treatment packages were presented to subjects individually. The same male narrator, as well as the same male and female models, were used across the two social learning con— ditions. Both the social modeling and cognitive structur- ing treatments contained four discrete components, with self-tests administered at the end of the first two sections. The no-treatment control subjects were only tested during the experiment. A description of each treatment condition follows. Social Modeling This 50-minute videotaped presentation followed the social learning model outlined by Bandura (1971b). However, the actual content emphasized external stimulus cues that affect vicarious learning of affective self- disclosure. Each component of this treatment package was presented on videotape. In the introduction, a male narrator provided the subject with preorganizers by outlining the components to be covered and by giving instructions for taking the self- tests at the end of parts one and two. In addition, he gave instructions regarding the videotape equipment. During part one, instruction, the narrator dis- cussed affective self-disclosure in terms of social learning principles. Using these principles, he 54 explained how males in this culture have learned to inhibit affective self-disclosure more than females. He emphasized that inhibition of affective self-disclosure is a learned response and, therefore, can be altered through a relearning process. Finally, benefits for increasing affective self-disclosure were discussed. At the conclusion of this section, the narrator instructed the subject to take the written self-test over the material presented. The narrator read each question, and the sub- ject had a c0py of the self-test in front of him. The narrator read the correct answers to the self-test, and the subject scored the test himself. Since the primary purpose of this self-test was to motivate the subject to attend to the instructional material, this honor system scoring procedure was considered apprOpriate. If the subject met the minimum criterion for acceptable perfor- mance, eight correct responses to 10 questions, he was instructed to continue with part two. If the minimum criterion was not met, he was requested to ask an assis- tant to rewind the tape so that he could view part one again. In part two, behavioral modeling was presented. Each modeling vignette was short and lasted less than one minute. A male model was shown as a respondent in dyadic interactions, using inappropriate and appropriate affec- tive self-disclosure. An equal number of positive and negative feeling responses were presented. The model 55 separately interacted with a male and a female, both designated as the model's best friends. Immediately following the model's response, his best male or female friend verbally reacted to his statement. If the model made an inappropriate feeling statement, the verbal conse- quence was negative. If the reSponse was appropriate, the consequence was positive. After each of the eight vignettes, the narrator provided feedback on the model's performance. This same procedure was followed with the male behavioral model as an initiator of appropriate and inappropriate feeling responses in dyadic social inter- actions. Following this second set of eight modeling presentations, the subject was instructed to take the written self-test on this section. If the subject met the minimum criterion for acceptable performance, four correct responses to five situations, he was told to con— tinue with part three. If not, he was instructed to ask an assistant to rewind the tape and then view part two again. The procedure of administering and scoring the self-test was the same as that used in part two. During part three, behavior rehearsal, the narrator described new stimulus situations. The subject then ver- bally practiced affective self-disclosure both as an ini- tiator and as a respondent in dyadic interactions. Situ- ations were constructed so that the subject received practice in responding to a male and female whom he regards as his best friends. A total of eight stimulus 56 situations was presented, equally representing positive and negative feeling conditions. The subject directly responded to the taped stimuli. Between stimulus situ- ations, the narrator provided the subject with guidelines for evaluating his performance. The subject's performance was not monitored during this phase. Therefore, the investigator did not know whether or not the subject actively participated in the behavior rehearsal. In part four, review, the narrator summarized the material covered in each section of the learning package. At the end of this summary, the subject was instructed to ask an assistant to stop the videotape. The assistant then gave him instructions for the next phase of the pro- gram, the posttest. The verbatim typescript of this videotaped presentation is located in Appendix D. Cognitive Structuring As with the social modeling treatment, the cog- nitive structuring videotaped package followed Bandura's social learning paradigm. However, the actual content focused on the verbal and imaginal cognitive mediating factors which affect vicarious learning. All parts of this treatment package were presented on videotape. In the introduction, a male narrator provided the subject with preorganizers by outlining the components to be covered and gave instructions for taking the 57 self-tests at the end of parts one and two. In addition, he gave instructions regarding the videotape equipment. During part one, instruction, the narrator dis- cussed affective self-disclosure in terms of cognitive structuring processes. Using this procedure, he explained how males in this culture have learned to inhibit affec- tive self-disclosure more than females. Following an approach similar to Ellis' (1970), the narrator empha- sized the role irrational self-verbalizations play in inhibiting affective self-disclosure. The anticipated consequences, as well as the self-evaluative statements regarding them, were stressed. Through a discussion of the cognitive restructuring process, benefits for increasing affective self-disclosure were highlighted. At the conclusion of this presentation, the narrator read the lO-item self-test. The subject also had a copy of the self-test in front of him. The narrator read the correct answers to the self-test, and the subject scored the test himself. Since the primary purpose of this self- test was to motivate the subject to attend to the instruc- tional material, this honor system scoring procedure was considered appropriate. If the subject met the minimum criterion for acceptable performance, eight correct responses to 10 questions, he was instructed to continue with part two. If the minimum criterion was not met, the subject was recycled through the instructional unit. 58 In part two, behavioral and cognitive modeling was presented. Each modeling vignette lasted less than one minute. The behavioral and cognitive model was shown first as a respondent in dyadic interactions using inappropriate and appropriate affective self- disclosure. An equal number of positive and negative feeling responses were presented. The model separately interacted with a male and female who were designated as the model's best friends. Eight vignettes were presented, one for each type-of-response (inappropriate, appropriate) by type-of-feeling (positive, negative) by sex-of-best-friend (male, female). Before the model's response, a voice overlay depicted his self- verbalization (i.e., internal thoughts) about his feelings in each situation. The model's response immediately followed. However, his best male or female friend did not verbally react to his feeling statements. Thus, no positive or negative consequences for affective self-disclosure were presented. This same procedure was followed with the male behavioral and cognitive model as an initiator of appropriate and inappropriate feeling responses in dyadic social interactions. Following the 16 modeling presentations, the subject was instructed to take the written self-test on this section. If the subject met the minimum criterion for acceptable per- formance, four correct responses to five situations, he was told to continue with part three. If not, he was 59 instructed to ask an assistant to rewind the tape and then view part two again. The procedure of administering and scoring the self-test was the same as that used in part two. During part three, cognitive self-modeling, the subject was given situations in which he covertly imagined himself practicing affective self-disclosure as an initi- ator and as a respondent in dyadic interactions with his best male or best female friend. After each self-modeling vignette, the narrator provided the subject with guide- lines for evaluating his performance. A total of eight situations was provided for the subject to cognitively rehearse. The subject's performance was not monitored during this phase. Therefore, the investigator did not know whether or not the subject actively participated in the cognitive self-modeling. In part four, review, the narrator summarized the material covered in each section of the learning package. At the end of this summary, the subject was instructed to ask an assistant to stop the videotape. The assistant then gave him instructions for the next phase of the program, the posttest. The cognitive structuring videotape typescript is located in Appendix E. Both the social modeling and cognitive structuring treatments are schematically depicted as Figure 5. SOCIAL MODELING TREATMENT INTRODUCE COMPONENTS l, 1.0 INSTRUCT IN AFFECTIVE SELF-DISCLOSURE WITH SOCIAL LEARNING MODEL ‘Tr l CHECK UNDERSTANDING EEITERION NOT MET 3.1+ pRITERION MET 3.4 _. l 3.0 J PRESENT BEHAVIORAL MODELING J. CHECK UNDERSTANDING pRITERION NOT MET 5.1t FRITERION MET 5-H l 5.0 .y J INITIATE BEHAVIOR REHEARSAL 6.0 J. REVIEW COMPONENTS A . TERMINATE TREATMENT 8.0 Figure 5. 60 COGNITIVE STRUCTURING TREATMENT INTRODUCE COMPONENTS J INSTRUCT IN AFFECTIVE SELF—DISCLOSURE WITH COGNITIVE SOCIAL LEARNING MODEL J. L CHECK UNDERSTANDING PFRITERION NOT MET 3. pRITERION MET 3 I 3. I U? Q AND COGNITIVE MODELING 4.0 PRESENT BEHAVIORAL 4.0 «- EL CHECK UNDERSTANDING 4CRITERION NOT MET S-H pRITERION MET 5.2 I 5.0 L INITIATE COGNITIVE SELF- MODELING 6.0 REVIEW COMPONENTS 7.0 TERMINATE TREATMENT 8.0 Parallel processes involved in the social modeling and cognitive structuring treatments. 61 Each has parallel, discrete components in order to make both of comparable length and potency. Attention Placebo According to Orne (1969), "demand characteristics" are those cues which affect a subject's perception of his role in an experiment. Subjects often evince substantial progress after receiving only suggestions of therapeutic improvement (Kazdin, 1973b). Therefore, an attention" placebo control group was included to afford an index of behavioral, anxiety, skill, and attitudinal changes due to factors of attention, exposure to training materials, and any demand characteristics inherent in the dependent measures. The attention-placebo condition consisted of two parts. Before listening to the deep muscle relaxation tape, the subject read the rationale and instructions for the training. The rationale stated that oftentimes peOple do not interact well socially because they are nervous. The technique of deep muscle relaxation was then introduced as a method for reducing anxiety in social situations. The subject was also instructed to practice the deep muscle relaxation exercises as they were being described on the audiotape. After the subject read these instructions, the audiotaped muscle relaxation exercise was presented. During the tape, the subject 62 was seated in a comfortable recliner chair with the room dimly lit. This presentation was of comparable length to the social modeling and cognitive structuring treatment conditions and lasted approximately 50 minutes. An assistant stopped the tape and gave the subject instruc- tions for the next phase of the program, the posttest. No-Treatment Control Normally, one control group is sufficient for any experimental design. However, since a self—management follow-up study (Voight, 1975) immediately followed this one, a no-treatment control condition was included so that the main effect of self-management procedures could be independently assessed in the second investigation. This no-treatment control group, therefore, only received immediate posttesting in this experiment. Instrumentation and Data Collection Two measurement procedures were employed. The first involved a performance test which yielded three dependent measures: (a) amount of affect, (b) quality of affect, and (c) self-reported concomitant anxiety. Paper-and-pencil testing comprised the second procedure and consisted of an affective self-disclosure skill test and an attitude measure on disclosing feelings. Each of these measures was devised for this study and the self-management follow—up (Voight, 1975). All measures were individually administered to the subjects- 63 Performance Test The performance test consisted of 16 audiotaped situations. Type-of-subject-role (initiator, respondent), type-of-feeling response (positive, negative), and sex-of- best-friend (male, female) were equally represented in the audiotape stimuli, as presented in Figure 6. TYPE OF ROLE Q TYPE OF 2 FEELING INITIATOR RESPONDENT E m m m H POSITIVE 5. § U) EL] 3 &* NEGATIVE [:4 O m POSITIVE x A m § m NEGATIVE Figure 6. Representation of the three factors equally presented in the affective self-disclosure performance test (Q?2). Subjects were instructed to consider the taped voices as their best male or female friend to control for intimacy level in the simulated dyadic interactions. The performance test was played on one tape recorder. After each stimulus, the subject was instructed to respond to the situation for 15 seconds into a second tape recorder. After recording his response, the sub- ject reported the anxiety he felt while responding to ' that stimulus. A 1-7 Likert-type paper-and-pencil scale was used for this self-reported anxiety measure (see 64 Appendix F). A warm-up or pretest stimulus situation was given in order to habituate the subject to the test. This stimulus situation was not evaluated. The complete typescript of the performance test is located in Appendix G. Unitized typescripts were prepared from the tape recorded performance tests. A modified version of the Auld and White (1956) rules for dividing continuous con- versations into sentences was employed. The investigator assumed major responsibility for preparing the unitized typescripts. The performance test typescripts were then rated for amount and quality of affect by two previously trained raters. Raters first evaluated each unit within a response for amount of affect. A list of feeling words developed by Crowley (1970) and expanded by Auerswald (1972) was given to the raters as examples of affect. They were instructed to rate a unit as containing affect if it met two criteria: (a) Crowley's rules (1970), and (b) defi- nition of affect. If one unit within a response met the criteria for affect, an amount score of one was given for the total response. On the other hand, if no unit within a response contained affect, the total response was given a zero. Raters were further instructed to only rate those units for quality which had been previously identified as containing affect. The affect quality score, therefore, was a subject's total score across the number of his 65 responses containing affect. The quality measure had five subscales, each with varying numerical values assigned: (a) reference, 0-12 points; (b) time orien- tation, 1-6 points; (c) apprOpriateness, 0-8 points; (d) reason, 0-5 points; and (e) specificity of reason, 0-8 points. Both amount and quality of affect criteria are provided in Appendix H. Scores on the amount and quality variables were totaled for each subject. The investigator also computed a total score for the anxiety measure. The performance test, therefore, yielded three numerical totals which corresponded to the three depen- dent variables of amount and quality of affect and self- reported anxiety. Raters Two raters were selected and trained by the researcher to make assessments of amount and quality of ‘affect on the unitized typescripts. Both raters were female, high school English teachers who had been close friends for six years. Each had a bachelor's degree and 30 graduate level course hours. Because of their exper- tise in English and grammar, both raters were considered competent by this investigator to rate the typescripts. The raters were trained to evaluate units within each subject's responses for amount and quality of affect. The first training session consisted of the following experiences. 66 l. introduction to amount and quality of affect 2. explanation and discussion of the rating criteria 3. observation of the investigator modeling the rating procedure using pilot test typescripts 4. practice rating of pilot test typescripts 5. discussion and further practice The second session was composed of the following tasks. 1. review of the criteria 2. rating two pilot test typescripts for reliability estimate 3. discussion of typescripts rated for relia— bility purposes 4. clarification and modification of rating criteria 5. rating two pilot test typescripts for a second reliability estimate The training continued until doubts and concerns on how to rate the typescripts were removed through continued discussion and practice. The researcher systematically presented a variety of pilot test typescripts during both sessions to insure rater competence in using the rating criteria. When evaluating typescripts for reliability purposes, each 67 rater was seated at a separate work station, and no com- munication between raters was allowed. This procedure was used to insure independence of ratings. Reliability Interrater reliability for Training Sessions 1 and 2 as well as for the actual study was computed using the analysis of variance technique developed by Hoyt (Guilford, 1954). The ANOVA was calculated on a Control Data 6500 computer using a program developed by Jennrich (1961). This procedure yielded estimates of reliability of the total raters' ratings on amount and quality of affect, as well as on the five quality subscales. Results are reported in Table 1. Table l Hoyt Reliability Coefficients for the Total Rater Ratings for Amount and Quality of Affect and the Five Quality Subscales Trainin Training Actual Affect Criteria SeSSlon Session 2 Study Data Data Data Amount 1.00 1.00 .99 Quality .09 .99 .99 Reference .51 1.00 .99 Time Orientation .00a 1.00 .98 Appropriateness .00 1.00 .97 Reason .84 .85 .98 Specificity of Reason 1.00 .99 .99 aThis coefficient was actually -80.00, which indi- cated that the Time Orientation rating for Training Session 1 had no reliability. 68 In estimating interrater reliability for each set of Training Session 1 and 2 data, two typescripts from the pilot test were used. The subject pool used in estimating the rater reliability for the actual study con- tained eight randomly selected typescripts, two from each treatment condition. As evident from the reliability coefficients reported for training session 1 in Table l, the raters were not consistent on the quality dimension. The time orientation and appropriateness reliability coefficients of .00 indicated a lack of consistency between raters and suggested much more interaction of variance between them than due to differences between subjects. After the first reliability rating, the researcher found that one rater was consistently making errors. The rating criteria were discussed and refined before the second interrater reliability measure was taken. As obvious from the coefficients reported for training session 2, the rating errors were sufficiently corrected to provide significantly higher reliability coefficients. Finally, the reliability data from the actual study are sufficiently high to conclude that raters were consistent in rating amount, quality, and the five quality subscales across the four treatments on exper- imental data. 69 Paper-and-Pencil Measures Affective self-disclosure skills and attitudes toward disclosing feelings were jointly developed by this investigator and Voight. The skill test was a multiple choice measure with correct answers keyed to the definitional criteria for affective self-disclosure. Some distractors were cognitively oriented, while others contained "other-than-self" referenced affect, past tense, and vagueness of reason. Stems contained positive and negative stimuli, initiator and respondent situations, and best male and female friends as inter- actants. A pool of skill items was developed; however, only the most discriminating items were retained after pilot testing. Scores on the skill test were totaled, with one point for correct answers and zero points for incorrect responses. Appendix I contains the 24-item skill test. The attitude survey was comprised of items selected from a pool of pilot-tested items which con- tained only the most discriminating items. Items on the attitude survey were marked on a Likert scale of 1-7, ranging from strongly agree to strongly disagree. The investigator totaled each subject's score following scale adjustment to account for positive-cued questions. Appendix J contains the 30-item attitude survey. 70 The reliability coefficients for the two paper- and-pencil instruments used in the actual experiment were: £.= .87 for the skill test and E = .84 for the attitude survey. Data Collection Subjects were seen twice. During the first session, all received a treatment and took posttest measures. After a three-week period, those subjects who did not receive an intervening self-management treatment (Voight, 1975) were retested. Upon completion of the testing, subjects evaluated the training by com- pleting a debriefing questionnaire. Appendix K contains this questionnaire. Data Analysis Data were keypunched and hand verified by the investigator. The various statistical analyses were calculated on a Control Data 6500 computer. An analysis of variance procedure developed by Finn (1968) formed the base for this data analysis. A multivariate analysis of variance test was conducted to ascertain the immediate treatment effect. Thus, the data collected were analyzed using analysis of variance techniques for a one-way design with five dependent variables. Based on pilot test results, directional hypotheses were formulated. Orthogonal 71 planned comparisons were used to test these hypotheses at the .05 alpha level, using one-tailed tests for sig- nificance. In order to ascertain the effects of treatment over time, the five dependent measures on subjects who received immediate and delayed posttesting with no inter- vening treatment were analyzed using a repeated measures multivariate analysis of variance procedure. Directional hypotheses were stated since specific results were expected. The alpha level for the three hypotheses of interest was set at .05. The ordering of the five depen- dent variables for both procedures was based on pilot test outcome data. Three supplemental questions of interest were also addressed. An item analysis across three factors of type-of-role (initiator, respondent), type-of-feeling (positive, negative), and sex-of—best-friend (male, female) was conducted for amount and quality of affect, as well as for anxiety level on the performance test. This same analysis was also conducted for the skill test. Thus, four repeated measures multivariate analyses of variance tests were performed. A one-way analysis of variance procedure was con- ducted to test for differences across treatments using the mean instead of the total quality score. This test was run to ascertain whether or not a measure of affect 72 quality not influenced by the amount score (i.e., the mean quality score) would produce significant differences across treatment groups. For each of the two supplemen- tary statistical tests, an alpha level of .05 was used. Finally, descriptive statistics for the debrief- ing questionnaire were computed. Thirteen subjects who received no self-management procedure completed this questionnaire after the delayed posttest. However, since only eight had received an experimental treatment, only their questionnaires were used. Since none of these three supplemental issues were formally included in the hypotheses for this investigation, the results are reported as additional information only. The outcomes of these data analysis procedures are reported in chapter III. CHAPTER III RESULTS The results of this investigation were based upon measures of amount and quality of affect expressed in simulated dyadic interactions, a self-report concomi- tant anxiety level, and two paper-and-pencil measures assessing skill level and attitudes toward disclosing feelings. Total scores for each of these measures were obtained. To test for the main effect immediately following treatment, a one-way multivariate analysis of variance utilizing orthogonal planned comparisons was employed. A repeated measures multivariate analysis of variance test was used to ascertain the maintenance of treatment effects over time. The same orthogonal planned comparisons were again tested. Supplemental research questions were also of interest. One dealt with the nature of affective dyadic communication. Three factors--type-ofarole (initiator, respondent), type-of-feeling (positive, negative), and sexbof-best-friend (male, female)--were assessed in repeated measures multivariate analyses of variance on 73 74 amount and quality of affect, as well as the concomitant self-reported anxiety level on the performance test. This same analysis procedure was applied to the items on the skill test. A second supplemental question was also explored. Will differences among the four treatment conditions exist using the mean quality score instead of the tgtal quality score? A one-way analysis of variance procedure was used to test for the main effect using this dependent measure. For all formal hypotheses and supplemental tests, the .05 probability of a Type I error (alpha) with appropriate degrees of freedom was used. Finally, the subjects' assessment of the two experimental treatments after the delayed posttest was of interest to this researcher. Therefore, results from the debriefing questionnaire for eight subjects are presented as descriptive statistics and frequencies. Immediate Posttesting General Findings The immediate treatment effect can be observed in terms of the result means for social modeling, cognitive structuring, attention-placebo, and no-treatment control. The means for each group on the five dependent measures are recorded in Table 2. Cell means for both social modeling and cognitive structuring groups were higher 75 hm.mH mm.NvH hH.NN oo.vma mh.MH mm.mmH vm.ma mm.HmH ccsuauua Nc.cH mc.ma mc.aH ~m.ca mv.mH ma.om mm.mH mm.aa muoaxca mm.m hm.m mo.m mo.m mo.m mm.ma no.m SH.m Hawxm hm.oma Ho.om~ mm.mm mm.mha om.maa mm.amm mm.moa vm.mmm muflamso um.v mm.m Hw.~ Nv.m mm.m mh.oa oo.m mo.HH ucsoad om cmmz am GMT: .mm cmmz mm cum: 5 u 3 12 u 3 12 u 3 5 u .1: 838: Houucoo onmomam mcHHSDODHum mcflampoz ucmpammmn ucmEummualoz Icowucmuum O>Huflcmou Hmwoom monummmz mcsufluud 6cm .humflxcd .Hflwxw .hufiamno .ucsoE< co masono Houucoo ucmfiummhetoz cam .onmomamucowucmuu< .mcflusuonnum O>Huflcmoo .mcwamwoz amaoom How chfluma>OQ pumpcmum paw mammz HHOO ~ mance 76 than those for attention-placebo and no-treatment control groups on the amount, quality, skill, and attitude measures. The lowest cell mean on the anxiety scale, which indicated the least anxiety experienced during the per- formance test, was scored by the attention-placebo group. Intercorrelations for the five dependent measures and the supplementary mean quality measure are reported in Appendix L. Of these measures, amount and quality were the most positively correlated (.99). This high cor- relation suggests that amount and quality were measuring similar components. The correlations between mean quality and quality (.51) and mean quality and amount (.49) sug- gest that mean quality measures some different components than amount or quality. Therefore, these correlations indicate that the quality measure reflects amount. The correlations of anxiety and attitude with the other measures were low, thus suggesting that these measures were not closely related to the other three. The .40 correlation between skill and amount, as well as with quality, suggests that being able to discriminate affec- tive self-disclosure on a paper-and-pencil test is not the same as verbalizing it in a performance situation. With the background of these general findings in mind, the results of hypotheses dealing with the immediate main effect of the four treatment conditions are reported. 77 Hypotheses Test Results Hypothesis 1 was planned to test the difference between social modeling and cognitive structuring versus attention-placebo and no-treatment control groups. A one-way multivariate analysis of variance procedure was used to test Hypothesis 1. The results of this analysis are shown in Table 3. This hypothesis was stated in a directional manner based on pilot test results. The .05 probability of a Type I error was used to test the multivariate sta- tistic (E = 5.97, d: = 5,40). The .05 level of error was partitioned for the five dependent measures. Thus, each measure was tested at the .01 level. Partitioning of the alpha level was considered appropriate to control for the overall Type I error. The multivariate sta- tistic (F = 5.97, d; = 5,40) for the five dependent measures was significant at the .0004 level. The uni- variate F results indicate that quality, amount, skill, and attitude contributed to the differences between the experimental and control groups, while anxiety did not. Furthermore, step-down F probabilities suggest that the quality measure was the most important factor contribut- ing to group differences. Based on this conditional test, the other measures seemed to be largely redundant and did not greatly contribute to the overall differences. Therefore, Hypothesis 1 was supported. It stated: 78 Table 3 Multivariate and Univariate Analysis of Variance and Step- Down F Statistics for Quality, Amount, Skill, Attitude, and Anxiety Measures Social Modeling and Cognitive Structuring (n = 24) vs. Attention-Placebo and No-Treatment Cofitrol (n_= 24) Comparison Multivariate F = 5.97 with 5, 40 d3, —P_ < .0004 Dependent 82:22. Univafiag; z B 3235' 2 Between ’ -—— F Quality 286597.52 19.94 < .0001 19.94 < .0001 Amount 275.52 22.38 < .0001 3.54 < .0669 Skill 280.33 13.33 < .0007 2.24 < .1417 Attitude 2366.02 7.66 < .0083 2.11 < .1543 Anxiety .45 < .5061 .09 < .7619 82.69 79 Hypothesis 1: The main effects of the social modeling and cogni- tive structuring groups will be greater than those of the attention-placebo and no-treatment control groups. To further examine the effects of the quality, amount, skill, and attitude variables, a 95% confidence interval was constructed for each (see Table 4). The least square estimates provide estimates of the population means for the four measures based on sample data. With infinite replications of this study, it is expected that the subsequent sample mean scores will fall within the reported confidence intervals 95% of the time. Standard deviations were also computed to measure the magnitude of differences between the two treatment and two control groups. The standard deviations for these measures were: 1. quality - 2.58 2. amount - 2.73 3. skill - 2.11 4. attitude - .58 The amount, quality, and skill standard deviations suggest that the two treatments were quite effective in producing desired changes. On the other hand, the atti- tude measure was less effective in producing a dif- ferential effect between the two treatment and two con- trol groups. 80 Table 4 The Least Square Estimates, 95% Confidence Intervals, for the Quality, Amount, Skill, and Attitude Measures Social Modeling and Cognitive Structuring (n = 24) vs. Attention-Placebo and No-Treatment Control (2.: 24) Comparison Dependent Least Square Confidence Measure Estimate Interval Quality 309.08 l70.64-447.53 Amount 9.58 5.53-13.64 Skill 9.67 4.37-l4.96 Attitude 28.08 7.25-48.38 81 Hypothesis 2 was structured to test for dif- ferential effects between cognitive structuring and social modeling groups. Again, a multivariate analysis of variance procedure was used to compare effects of the two experimental treatments. The results are presented in Table 5. The multivariate statistic F (5,40) = 2.33 failed to reach the established .05 significance level. Therefore, Hypothesis 2, which stated that the effects of cognitive structuring would be greater than those of social modeling, was not supported. Examining the uni- variate F statistics, the researcher found that none of the measures produced a significant result at the .01 level. Furthermore, step-down F statistics suggested that none of the measures, when tested conditionally, produced significant results. Hypothesis 3 was structured to compare the effects of the attention-placebo with the no-treatment control group. The same multivariate analysis of variance procedure was used to test this hypothesis. The results are located in Table 6. The multivariate statistic F (5,40) = .70 was not significant, thus indicating that the two control groups did not differ from each other. The step-down and univariate statistics also indicated that none of the measures significantly contributed to differences between the two control groups. Therefore, these results supported the third 82 Table 5 Multivariate and Univariate Analysis of Variance and Step- Down F Statistics for Quality, Amount, Skill, Attitude, and Anxiety Measures Cognitive Structuring (n = 12) vs. Social Modeling (3 = 12) Comparison Multivariate F 2.33 with 5, 40 df, p < .0603 Dependent MW 5232?. Univaggagg E 2 $2352" 2 Between ' —— F Quality 161.72 .01 < .9161 .01 < .9161 Amount .67 .05 < .8171 6.30 < .0160 Skill 100.04 4.76 < .0346 5.12 < .0289 Attitude 30.38 .10 < .7554 .00 < .9541 Anxiety 51.04 .28 < .6011 .13 < .7166 83 Table 6 Multivariate and Univariate Analysis of Variance and Step- Down E Statistics for Quality, Amount, Skill, Attitude, and Anxiety Measures Attention—Placebo (n = 12) vs. No-Treatment Control (3 = 12) Comparison Multivariate E = .70 with 5, 40 df, Dependent E < ’6954 523:2. Univafiagg : E 3235' B Between ’ —— E Quality 15713.28 1.09 < .3015 1.09 < .3015 Amount 12.04 .98 < .3281 .12 < .7303 Skill 2.04 .10 < .7569 .01 < .9311 Attitude 468.17 1.52 < .2249 1.43 < .2393 Anxiety 73.50 .40 < .5306 .88 < .3553 84 research hypothesis, which stated that there would be no difference between the attention-placebo and no- treatment control groups. Delayed Posttesting General Findings The treatment effect over time can be observed in terms of result means for the summation of and dif- ference between the five dependent variables taken at immediate and delayed posttesting sessions. These data are reported in Table 7. The social modeling and cog- nitive structuring means for the two test sessions were higher than those for the attention-placebo and no- treatment control groups on the amount, quality, and skill measures. The two experimental groups also had a lower mean anxiety level than the two control groups did. On the attitude survey, however, the no-treatment con- trol group mean score exceeded that of the social modeling group. The mean differences between immediate and delayed posttesting indicate that both control groups made higher scores on the delayed posttest for amount and quality measures. Furthermore, the no-treatment control group improved its score on the skill test as well, while both experimental groups' scores on amount, quality, skill, and attitude declined at posttesting. With this background information in mind, the hypotheses dealing with the effect of treatment over time are reported. 85 .ummuumom UmmmHmw ou mamacmEEH Eoum mmuoom cmmE cmm3umn mocmumMMHw msu .mummuumom pmwmamc 6cm muMfimeEfl mmouom monoom come no cowumfifidm may MHQ «a oo.m mm.m oo.m om. mvsufluud man. oo.mom nm.vmm co.>om om.mnm mmsuwuum 85m oo.m I mm.m mm.m 00.5 mumfixam man oo.mHH hm.voa mm.mm om.om mpowxcd Sum oo.m I mm. oo.m mm. HHflxm MHQ oo.m mm.ma om.mm mh.ha HHme 85m oo.hmI om.vm| mm.m~ mH.mm unflamso mac o~.m~m mm.emm mm.vnm m¢.m~m mundane sum om.m I hm.m I mu. om.m ucaofid ««Mfla om.mH mm.HH mh.ma oo.ma unnoE¢ «55m comm snow comm comm Am u so Am n no as u so A. u no musmmmz Houucou onmowam mcwusuosuum mafiamooz usaoammmn ucmfiummHBIoz IGOflucouud m>wuflcmoo Hmwoom mcwummuumom cmmmHmo 0cm mumwwofifiH mmouom mmusmmmz mosufluud wcm.mumflxcm .Hawxm .muwamso .ucsoEd cwmzumn wocmumMMflo cam mo coflumfifism mcu co masono Houucoo ucmfiummusloz 6cm .onmomamICOflucmuud .mcahfiuosnum m>w¢wcmou .mcflamcoz Hafioom How mammz HHGU a magma 86 Hypotheses Test Results Hypothesis 1 was formulated to test for a treat- ment-by-measures interaction effect. A repeated measures multivariate analysis of variance was conducted to test this hypothesis. These results are located in Table 8. As predicted, the treatment-by-measures interaction was not significant, 3 (15, 14.20) = .96, p < .53. However, the univariate E probabilities for both amount and quality were .07 and .08, reSpectively, thus suggesting some degree of interaction for these two variables across treatments and measures. Step-down E statistics also suggested that the amount measure contributed the most toward an interaction effect. Table 8 Multivariate and Univariate Analysis of Variance and Step- Down E Statistics for the Treatment-by-Measures Inter- action Effect across Immediate and Delayed Posttests Multivariate F = .96 with 15, 14.20 Dependent di, E < .5297 Measures Mean Univariate Step- Square §_ 2_ Down 2. Between 3, 9 Q: E Dif Amount 24.26 3.26 < .0734 3.26 < .0734 Dif Quality 25087.28 3.10 < .0822 1.12 < .3964 Dif Skill 11.27 .70 < .5767 1.38 < .3247 Dif Anxiety 74.45 .65 < .6014 .24 < .8690 Dif Attitude 38.68 .45 < .7266 .53 < .6839 87 Hypothesis 2 was planned to test the treatment effect over time. A repeated measures multivariate analysis of variance using the sum of each dependent measure across immediate and delayed posttests was con- \ ducted. The results of this test are reported in Table 9. The multivariate analysis of variance indicated no dif- ferences among treatments across immediate and delayed -posttesting, E (5,5) = .93, p < .53 for the contrast ,between the two experimental and the two control groups 3 (5,5) = .54, p < .75 for the contrast between experi- mental groups, and E (5,5) = 1.03, p < .49 for the con- ‘trast between the two control groups. Therefore, the results do not support the research hypothesis that a treatment effect would be found. 4 Table 9 Repeated Measures Multivariate Analysis of Variance for the Planned Contrasts Testing the Treatment Effect across Immediate and Delayed Posttests ‘ l Multivariate F Contrast 5'5 9:“ — 2 (SM + AP) - (AP + NTC) .93 < .5310 CS - SM .54 < .7471 AP - NTC 1.03 < .4894 1 SM = social modeling; CS = cognitive structuring; AP = attention-placebo; NTC = no-treatment control. In Tables 9, 10, 11, and 28 the following letters will be used to designate the four treatment conditions. 88 Hypothesis 3 stated that there would not be a measures effect. A multivariate analysis of variance procedure was used to test each of the three planned contrasts for a measures effect. These results are reported in Table 10. None of the contrasts were sig- nificant at the established (.05) level. Thus, these data suggest that the difference between the four treatment groups did not significantly differ across the two testing times. These results support the second research hypothesis. Table 10 Repeated Measures Multivariate Analysis of Variance for the Planned Contrasts of the Measures Effect across Immediate and Delayed Posttesting Multivariate F Contrast 5' 5 Q: -— p_ (SM + CS) - (AP + NTC) 1.46 < .3452 CS - SM .21 < .9424 For the immediate-delayed posttest repeated measures multivariate analysis of variance test, small and unequal cell sizes ranging from four to two subjects were used. Hence, statistical power may have been too weak to pick up treatment effects if any existed. In order to explore this possibility, the researcher 89 compared immediate posttest mean scores for all subjects who completed the immediate posttest (g = 48) with scores for subjects who did not receive a follow-up procedure (Voight, 1975) before the delayed posttest (E = 13). The cell means for both groups, as well as the standard deviations for the first, are reported in Table 11. By comparing the mean scores and taking into account the standard deviations, the researcher noted that the cell means of Group 2 were comparable to those of Group 1. Therefore, the investigator concluded that the two groups were equivalent at the time of immediate posttesting. Finally, mean scores for the 13 subjects used for the repeated measures analysis were computed for the delayed posttest. A comparison of immediate and delayed posttest means for each treatment group was then made. These results are graphically represented in Figures 7, 8, 9, 10, and 11. Interactions between the no-treatment control and social modeling groups occurred for both amount and quality measures. In both cases, the social modeling group's score declined, and the no-treatment control group's score increased, thus contributing to the negation of the significant immediate posttest results. These interactions may have been artifacts of chance. On the skill test, all groups except no- treatment control declined in delayed posttest scores. 90 Table 11 Immediate Posttest Cell Means and Standard Deviations for the Four Treatment Conditions across the Five Dependent Variables Group 1 (E = 48) Group 2 (E = 13) Dependent Variable Mean §2 Mean Amount SM 11.08 3.00 10.75 CS 10.75 3.33 10.25 AP 5.42 2.71 4.33 NTC 6.83 4.67 6.00 Quality SM 356.37 102.32 356.28 CS 361.56 113.20 348.83 AP 178.83 92.36 136.67 NTC 230.01 160.27 213.10 Skill SM 9.17 3.07 9.25 CS 13.25 6.06 12.75 AP 6.08 3.06 8.33 NTC 6.67 5.35 3.00 Anxiety SM 47.83 12.95 48.75 CS 50.75 15.48 45.75 AP 44.92 14.68 54.00 NTC 48.42 10.62 57.50 Attitude SM 151.33 16.94 140.00 CS 153.58 13.76 156.50 AP 134.00 22.17 134.00 NTC 142.83 16.37 157.00 Mean Amount of Affect Figure 7. 91 15-— g ‘1 ..+ 10-— ‘i /° 1’ . z' _ . / 1' / 5-— ’e/ / 0 ~— = Social Modeling _ 0.0-. = Cognitive Structuring .. - . = Attention-Placebo ‘ . ..... o= No-Treatment Control L 4- Immediate Delayed Posttest Posttest Comparison of mean scores for amount of affect between immediate and delayed posttests. 92 370-“ 325-— p o _ m In m _ fi M _ 0 B1 250—— -H g _ 0 g. _ / a ' / m m _ / S ./ -« / / 175 —- / /' ~ / / _ / O 125-— o-—-. = Social Modeling -—— o-ua = Cognitive Structuring o--m = Attention-Placebo = No-Treatment Control 0-..... 0 _I- I .L Immediate Delayed Posttest Posttest Figure 8. Comparison of mean scores for quality of affect between immediate and delayed posttests. 93 15“ Mean Skill Level L Social Modeling Cognitive Structuring Attention-Placebo Immediate Delayed Posttest Posttest Figure 9. Comparison of mean scores for skill level between immediate and delayed posttests. Mean Anxiety Level Figure 10. 94 40 - 30- 20“ *-- = Social Modeling o-«a = Cognitive Structuring _ --m = Attention-Placebo °«w~ = No-Treatment Control 10.— I 1 Immediate Delayed Posttest Posttest Comparison of mean scores for anxiety level between immediate and delayed posttests. 95 160 W 3=4m 4 " \ ..‘: ‘ .‘. ~. 150‘ .'u. .4 U) 140‘ T 4 m H o _ O ._ ~ _ m ----._____. a) 130') 'o 5 .‘J — -H 13 d 120 8 4 g °-* = Social Modeling 110.. o-o-o = Cognitive Structuring ---o = Attention-Placebo - -«-==No-Treatment Control L lOO‘L T I I Immediate Delayed Posttest Posttest Figure 11. Comparison of mean scores for attitude sur- vey between immediate and delayed posttests. 96 Likewise, anxiety level decreased for all groups except no-treatment control. Finally, all groups' mean scores on the attitude survey decreased. However, the no- treatment control group, which had the highest score on the immediate posttest, only had the second highest mean score on the delayed posttest, thus creating an interaction with the cognitive structuring group on this measure. As evident from Figures 7-11, then, experimental treatment scores sufficiently declined while no-treatment scores sufficiently increased to make the treatment effect over time nonsignificant. Supplemental Findings The supplemental findings which are reported in this section include: (a) repeated measures multivariate analyses of variance on amount, quality, anxiety, and skill measures for three factors: type-of-role, type-of- feeling, and sex-of-best-friend; (b) an analysis of variance to test for the treatment effect using the mean quality score; and (c) descriptive statistics and fre- quencies for the debriefing questionnaire. Since none of these statistical analyses were conducted to test formally stated hypotheses, these results are reported as additional information only. Each test was conducted after formal hypothesis tests were completed. 97 Repeated Measures Item Analyses On the performance and skill tests, items were constructed to systematically examine three factors involved in affective dyadic communication. The type- of-role (initiator, respondent), the type-of-feeling (positive, negative), and sex-of-best-friend (male, female) were combined across items in order to examine any differential effects treatments might have on sub- ject responses to these three factors. Four separate repeated measures multivariate analyses of variance were conducted with an alpha level of .05 set for each. The transformation of the input factors produced the follow- ing dependent variables: constant, role, sex, feeling, role-by-sex, sex-by-feeling, role-by-feeling, and role- by-sex-by-feeling. The results of these analyses are presented in Tables 12-27.2 Whenever a multivariate test was significant for one of the planned contrasts, both univariate and step- down E statistics were examined. The .05 alpha level was partitioned for the eight dependent measures. Thus, the .006 level was used to ascertain significance for the individual test results. 2In these tables, the following letters are used to denote the eight dependent measures: C = constant, R = role, 8 = sex, F = feeling, R x S = role-by-sex, S x F = sex-by-feeling, R x F = role-by-feeling, and R x S x F = role-by-sex-by-feeling. 98 coma. v ma.m mhha. v mm.a mo. m x m x m mnmm. v ma. mmmv. v mm. Ha. m x m mmoo. v Hm.m name. v ma.m mm. m x m mmwm. v mm. Hmav. v mm. Ha. m x m Hmvm. v nm.a mvma. v mv.m m>.m m mmmm. v em. nmam. v Ho. Ho. m mmva. v o~.~ ammo. v mm.~ m~.m m Hooo. v am.mm Hooo. v Hm.m~ mo.mmm o m m m I mm as .H cmmsumm azooImmum m mumwum>wca mumsqm cams whammmz usmocmmma mooo. v m .mc hm .m spas am.m n m mumflum>fluaaz acmwummfioo Avm ".mv Honucoo unmfiummHBIoz cam onmomamIGOflucmuud .m> Ava n mv msflusuosuum m>auwcmou can mcflamooz HMfloom madamGMImQIxomIanImHom cam .mcflammthnImHom .msfiammm ImnIxmm .xmmenImHom .mcwammh .xmm .mHom .ucmumcoo Ucm unsoa¢ How mowumwumum M szoalmmum can mUCMflHm> mo mwmhamad mumwnm>wss tam mumanm>aua52 mmusmmmz Umummmmm NH OHQMB 99 swam. v 00. mmmv. v we. mo. m x m x m meme. v mm. Nmo>. v ma. mo. m x m ovmm. v 00. hmmm. v om. mo. m x m coma. v mm.m mmmo. v ma. mo. m x m vomm. v ha. hmmm. v Ho. Ho. m mnmm. v on. mnmm. v HH.H Hm. m ommo. v mm.v vomo. v Ho.m Hm.m m mmmm. v mo. mmmm. v mo. mm. 0 m m m I mm as .H cmmzumm czoaImmum . m mumaum>fl25 mnmsvm cmmz musmmmz ucmocmmma hams. v m .mm mm .m an“; mo.a n.m mumflum>flpflsz comflummaou ANA n mv mcflamooz HMfloom .m> ANA n my mcflusuosuum o>wuwcmou mcflammhImnIxmmlmnImaom can .mcflaomhlwnImHom .mcflaomm Imnlxmm .xmmenImHom .mcwammm .xmm .maom .ucmumcou cam “£5084 How mOflumaumum M GBOQImmum cam moccanm> mo mfimhamsm mDMHHm>flGD tam mumanm>wua52 mondmmmz woummmmm ma manna 100 mamm. v ov. momm. v mo. oo. m x m x m ammo. v mo. oooo.a v oo. oo. m x m Amos. v Ha. oooo.a v oo. oo. m x m whom. v oo. oooo.a v oo. oo. m x m ooam. v om. aoom. v mm. mm. m homo. v Ho. ovum. v mo. oo. m «mom. v om. mmom. v om. mm. m comm. v mm. moon. v mm. oo.oa U m m m I m 3 . a 595me c3ooImmum m mumaum>flco mumsvm cmoz musmmmz I ucmccmmmo mmmm. v m .mm mm .m spas mm. u m mumuum>fluasz GOmHHmmEoo Ama n mo Honucou ucoaummuBIoz .m> ANA n my oncomamIGOflucmuud mcwamomhnIxmthnImHom can .mCHHmMMImnImHom .mcwammm Ianwam .xmmlhnIoaom .mcwammm .xmm .maom .ucmumcoo cam unsofid How unflumflumum m c3oaImmum can mosseum> mo mammamad mumaum>wcs 6cm mumaum>fluasz mmusmmmz Umummmom «A OHQMB 101 mmam. v mo. . Hmhv. v mm. mo. m x m x m mmmm. v av. mmao. v mm.m mm. m x m vmma. v ¢N.H ommo. v mm.v mm. m x m mmmN. v mm.H mvmh. v mo. Ho. m x m vomo. v mm.m Hooo. v mh.mm mN.Hv h momH. v mm.m hmam. v Ho. Ho. m vam. v we. mhom. v ho.H hH.H m Hooo. v ow.amm Hooo. v mm.HmN mH.mHmm U m czowmmoum m M wwmawm>mwn oHMMWWuNmmz . . whammmz II I so comm Hooo. v m .88 an .m nuns am.ov n m mumflum>fluasz u o a h mmoouo usmfiumoua Hsom on» mmouom mcflammmImnIxmmImnlmHom cam .mcaamwMImnImHom .mcwawwm IhnIxmm .msHHQMMImnImHom .mcaammm .xmm .maom .ucmumcou cam ucsoad mom moflumflumum CBOQImmum was moccaum> mo mflwhamaé onwanm>acs can mume~m>aua52 monsmmmz omummmmm ma OHQMB 102 mmmv. v mv. mama. v om.m mm.~ma m x m x m. mmmo. v no. ommm. v om. mm.mm m x m mvmo. v mo.v omoo. v vm.m mo.~m¢ m x m mmam. v mm.a momm. v mo. Hm.v m x m Homm. v no. mmmm. v vv.a mm.vvam m «mom. v mm.m omom. v mo. ov.va m hmma. v mo.a mmma. v mo.m mm.mmmm m Hooo. v No.om Hooo. v mh.om mm.moomom U m M m I mm vv .H cmm3umm csoalmmum m mummum>flca mumsvm cam: whammmz pcmvcmmmo 38. v m 7% s... .m fig Hm; I M 338353: acmflnmmfioo Ava!" mo Houucoo usmfiummHBIoz ocm onmomamIcoHucmuud .m> Avm n so maausuosuum m>HuHcmoo mam mcwamooz ammoom mcflammmenIxmmImnlmaom mam .mcflammMIwaImaom .mcflammm hnIxmm .xmmImnlmaom .mswammm .xmm .maom .uGMMmcou mam muwamso How moflumwumum M czooImmum mam mocmwnm> mo mammaqu mamanm>fica 6cm mummum>muasz meSmmmz mmummmmm mm magma 103 ooom. v mm. omom. v om. mh.ov h x m x m thm. v ma. mamm. v mm. mH.Hv h x m mmhm. v oo. mmmm. v mm. Ha.ov m x m mNHN. v Hm.H Noam. v mm. Av.H¢ m x m Hmmm. v mm. Hoam. v wo. mh.mmm h HmmH. v No.m mmHH. v mm.~ mm.mvva m mmho. v mm.m mono. v mm.m mm.mmH¢ m mmmm. v 00. mmmm. V CC. om.m U m m m mm as .a cmmsumm conImmum m mumwnm>flcs mumsvm cmmz musmmmz unmosmmma mama. v m .mm_sm .m spas oo.a u m mumflnm>fluasz 'I'I comfinmmaoo ANH n my mcflaoooz ammoom .m> ANH n my mcflusuosuum m>wumcmou mcHHmMMImnlxmmlmnIoHom can .mcfiammmImnImHom .mcwammm IhnIxmm .xmmImnImHom .mcflammm .xmm .maom .ucmumcoo ocm huwamso How momummumum M czoolmmum ocm monmwum> mo mammmmsd mumflnm>wca 6cm mumaum>wuasz monummmz omumommm ha manna 104 ovum. v mo. momm. v mo. mm.H m x m x m mama. v Ho. ommm. v mm. mm.ov m x m momm. v oo. mmmm. v Ho. 5H.H m x m ommo. v oo. mmmo. v no. oo.HH m x m momv. v mm. ammo. v mm. mm.mmm m mmmm. v Ho. namn. v om. mo.mm m Hmmm. v mm. mmmm. v mm. HH.omv m swam. v oo.H Hemm. v oo.M ma.~vmva o m m m I mm 3. . A $953 c3oaImmum m maneum>flcs mnmsvm com: whammmz unmocmmmo mmmm. v m .mm mm .m spfl3 mm. n_M mummnm>wuaoz comwnmmeoo ANA u do HOHucou ucmfiummHBIoz .m> ANH n my onmomamIGOHucmuud mafiammmImnIxmmImnImaom cam .mcflamoMIhnIoaom .mcwammm Imnlxmm .xmmImnImHom .mcflammm .xmm .maom .ucmumcou cam huflamso Mom moflummumum M c3ooImmpm 6cm mocmflnm> mo mammamcd mumwnm>mco mam mumwum>fiuasz mmHSmmmz Umummmmm ma magma 105 ammo. v oa. ommv. v mm. Hm.mm m x m x m namm. v mo. ammo. v om.m mm.mHHH m x m vmmm. v mo. mooo. v mo.o mm.HHHH m x m mmma. v mv.m ommo. v mo. oo.vH m x m mmmo. v mm.m Hooo. v om.Hv HH.HmmHm m memo. v mm.m ovmo. v NH. mv.mm m mmom. v mo. momH. v mm.H mm.momm m Hooo. v mH.Hmm Hooo. v mH.Hmm mo.a~amvmm U m M m I mm mm .H cmmzumm csooImmum m mumaum>flcs mnmsvm cmmz whammmz II. usmocmmmn Hooo. v m .mm om .m cums mo.mm n M mumanm>flumsz mmsouw usmfiummna usom map mmouom mafiammMImnIxmmlmnImHom mam .mcflammmImnlmHom .mcwammm IanIxmm .xmmImnImHom .mcflammh .xmm .mHom .ucmumcoo cam wuflamso How moflumflumum M G3ooImmum mam mocmwum> mo mammamcfi mumaum>flas mam mummum>fluasz mmusmmmz Umummmmm ma GHQMB 106 momm. v NH.H vvma. v mv.~ mv.H m x m x m Naom. v ma. mmmv. v mv. mm. m x m omno. v vm.m mmma. v vv.m ma.vm m x m momm. v om. momm. v mm.a oo.m m x m mvmv. v mm. oooo.H v oo. oo. m movm. v om. mmmm. v Ho. mo. m momm. v Hm. ovum. v om. mm.mh m Nmmm. v mv.a mmmm. v mv.a mo.moom U m M m I. mm vv .H cmmzumm conImmum m mummum>wcb mumsvm G802 whammmz I I a mvmv. v m .mo mm .m nufi3 mm. u m mumfium>fluasz unmocm ma GOmHHmmEoo Avmlu_mv Houucou ucmfiummHBIoz mom camomamlcoflucmpum .m> Avm n so mcHHsuosuum m>fluflcmoo mam msflmmmoz Hafioom mcflammmlmnIxmmenImaom mam .mcflammthnImHom .mcfiammm ImnIxmm .xmmImnImHom .mCHHmmm .xmm .maom .ucmumcou mam muomxsm How mowumfluMDm M c3oalmmum mam mocmwum> mo mwmhamsd mumanm>wco can mumanm>aua52 monummmz mmummmmm om magma 107 mmhm. v Hm. hmMB. V dd. mm. m x m x m mmHm. v mo. mono. v mv.m Hm.m m x m obvm. v hm. Hmom. v mm.H mm.mm m x m ommo. v mm.N vmvm. V Ho. Ho. m x m mmmo. v mm.m mmhv. v Hm. vo.m h mNNo. v mm.m mmvm. v Hm. vo.N m Hva. v Ho. thH. v mm.H mm.mmH m vaa. v mm.N vaa. v mm.N hH.ONNm U m M m I mm vv .H ammzumm czonlmmum m mUMHHm>HcD mumsvm com: mnsmmmz usmvcmmma mmmo. vom .mm km .m 38H; va.a I‘m mumflum>fluasz comHHmmEoo ANH u do mamamoos ammoom .m> “mm u do mcHHSDODHum o>fluflcmou mcflammmImnIxmthnImHom mam .maflammmImnImaom .mcflammm ImnIxmm .xmmlwnlmaom .mcflammm .xmm .maom .ucmumcoo mam mumflxc4 How moflumwumum M conImmum mam mUCMHHm> mo mflmhamcd mumwnm>flcs mam mamaum>fluasz mmusmmmz omummmmm AN manna 108 mmmm. v mo. mmvm. v Ho. oo. m x m x m mamm. v mv. mmao. v va. mm. m x m momv. v ov. momm. v mo. vo.H m x m mamm. v Ho.H mamm. v mm. mm. m x m ammo. v oH. mmmo. v no. om. m vmmh. v ma. mmmm. v mo. 5H. m momv. v mv. omam. v Ho. vo.m m hmmm. v vo. nmmm. v vo. om.om o m m m I mm vv .H cmmsumm GBOQImmum m mumflnm>flcb mumswm cmmz whammmz ucmmcmmoo mmmm. v m .mm an .m and; «m. um mumflum>fluasz GOmHHmmEoo ANA u my Honucoo ucmEummHBIoz .m> ANA n mo onmomamIcowusmuua msflammmImnIxomenImHom can .mcflammmlmnlmaom .mcwammm ImnIxmm .xmmenImHom .mCHHmmh .xmm .oaom .ucmumcou new wumflxcd How momumflumum M conImmum ocm mommanm> mo mflmhamsm mummum>wcs mam mumahm>flumsz monsmmmz mmummmmm mm magma 109 mmvm. v mm. ovmo. v av.m ma.m m x m x m mmam. v mm.a anon. v am.a om.~ m x m vmmo. v av.m mmmo. v va.m mo.vaa h x m mmom. v oo. mmom. v aa.a mm.m m x m mmao. v mm.m aooo. v ma.mm mo.vmm h mamm. v mo. mmao. v va. mm.a m aooo. v mm.voa vmmm. v vv.a mo.oma m aooo. v vv.om aooo. v vv.hm mo.o~mvma U m M m I MW vv .a cmm3umm GBOQImmum m mumaum>acs mumsmm cmmz whommmz I unmmnmmmo aooo. v m .mm am .m mugs os.mm u m mumaum>auasz masonw usmfiummus Hoom map mmonom mCaammMIhnIxmmlmnImaom cam .mcaammMImnImaom .mcaammh ImnIxmm .xmmImnlmaom .msaammm .xmm .maom .ucmumcoo mam aumaxcm How moaumaumum M csooImmum mam mocmaum> mo mamwams¢ mumanm>ass mam mumaum>auasz mmHSMmmz mmummmmm mm maQMB 110 mmmv. v hv. mmom. v vv. oo. h x m x m mama. v 00. momN. v om.a mo. m x m mmmm. v am. mmom. v mo. oo. m x m hwmb. v mo. wham. v mo. co. m x m mvbm. v mm.a mavh. v aa. mo. m mamh. v ma. vmam. v mm.a ma. m omvm. v av.a thm. v mm. vo. m 5000. v mv.ma 5000. v mv.ma va.am U m czoommwum m M mWMHWM>MMD mHMMMMuMMmE . . whammmz II I ucmocmmwo oamo. v m .mo am .m spas mm.a u m mumaum>auasz cowanmmfiou Avflln.mv aouucoo ucmsummHBIoz mam onmomamIcoaucmuu4 .m> Avm n so mcausuosubm m>auasmoo mam msaammoz amaoom mcaammMImnIxmmlmnImaom was .msaammMIwnImaom .mcaammm IhnIxmm .xmmenlmaom .mcaammm .xmm .maom .ucmumcoo cam aaaxm Mom moaumaumum M c3onlmoum cam mocmaum> mo mamwamad mumanm>aca mam mumaum>auasz mounmmmz Umummmmm vm manMB lll oomv. v mm. mmvm. v mm.a ao. m x m x m momm. v mm.a nmmm. v vm. ao. m x m Nmmo. v ma. mmmm. v mm. ao. m x m mamo. v mo.m mmoo. v mm.o mm. m x m mvmo. v mv.m mmoo. v mm.m vm. m maam. v ao. ammm. v oo. oo. m mavm. v ao. ommh. v ma. ao. m mmvo. v vm.v mmvo. v vm.v ao.oa U m M m I mm vv .a cmmsumm czoaImmpm m mumanm>aco mumsvm cam: whammmz ommo. v m .mm mm .m nuaz mm.m u M mumaum>auasz usmmcmmma cowaummaou Ana n mo mcaawooz amaoom .m> Ama u mo mcausuosnum m>apacmoo mcaamMMIhnIxmmenImaom cam .mqaamGMIhnlmaom .mcaammm IanIxmm .xmmImQImaom .mcaammm .xmm .maom .ucmumcoo mam aaaxm now moaumaumum M csooImmnm mam mocmaum> mo mamhamcd mumaum>acb mam mumaum>aua52 monsmmmz moummmmm mm maQMB 112 momm. v mo. mvom. v ma. oo. m x m x m vmmm. v ov. mvmm. v oo. oo. m x m vomv. v am. ammo. v oo. oo. m x m mmmv. v am. mamm. v mm. vo. w x m mmmv. v ao. whom. v ao. oo. m mmma. v mv.m mmoa. v mm.a ma. m mmmm. v ao. «mom. v mo. oo. m mans. v mo. mmoo. v mo. om. o m m m mm vv .a cmmsumm csoaImmum m mumaum>acn mumsvm cmmz whammmz II I m ommm. v m .mo an .m spas mv. u m mumaum>auasz unmonm ma comaummEou Ama n do aouucou unmfiummHBIoz .m> Ama n my onmomamIcoaucmuu¢ asaammMIanIxmmlmnlmaom mam .mcaammmImnImaom .msaammm ImnIxmm .xmthbImaom .maaammm .xmm .maom .ucmumcoo can aaaxm How moaumaumum m sBOQImmum mam mocmanm> mo mamxamsd mumanm>aco mam mumaum>auasz mmusmmmz mmummmmm mm maQMB 113 mmmm. v ma. ommm. v oo.a ao. m x m x m momo. v am.m mama. v mo.a mo. m x m mvmv. v mm. ommm. v ma. oo. m x m mmmv. v om. ammo. v mo. oo. m x m ammo. v vm.v mmoa. v oo.a mm. m mamm. v mo. omoo. v ma.oa mm. m vomo. v mo.m moma. v mm.m mm. m aooo. v mm.oma aooo. v mm.oma mm.>av U m M m I W 3 . a 525mm czooImmum m mumaum>aco mnmsom cmmz musmmmz I ucmmcmmma 88. v m .Mm mm .m 8:3 3.3 u m 3883632 mmsouw ucmfiummne udom ocu mmouom mcaamQMImnIxmmenImaom mam .mcaammhImnlmaom .mcaammm IMQIxmm .xmmImnImaom .mcaammh .xmm .maom .ucmumcoo mam aaaxm Mom moaumaumum M c3oalmmum mam mommaum> mo mammamad mumaum>acb mam mumanm>aua52 mmnsmmmz pmummmmm hm maQMB 114 For the contrast between the treatment and con- trol groups on amount, a significant multivariate result was obtained, E (8,37) = 5.61, p < .0002 (see Table 12). Examining step-down E and univariate probabilities, the investigator noted that the constant contributed most significantly to the differences between treatments and controls, F (1,44) = 23.61, p < .0001. This constant effect is actually the treatment effect, since it excludes the repeated measures. Therefore, the treatments were primarily responsible for the multivariate results. For the comparisons between the two treatments (see Table 13) and between the two controls (see Table 14), neither multivariate result was significant. The measures effect across treatments, however, produced a significant result, E (8,37) = 40.54, p < .0001. Since the constant effect for this test is the grand mean, this result was not relevant. However, the feel- ing effect did produce a significant result, E (1,44) = 36.78, p < .0001. The step-down result, though, was not significant, E = 5.82, p < .0204. These results indi- cate that if feeling is conditioned on role and sex, its effect is not significant. If taken as an indepen- dent measure, however, the feeling effect is significant across treatment conditions. Across treatment conditions, the amount mean for positive feelings was 5.21, while the amount mean for negative feelings was 3.63. 115 Therefore, subjects more frequently responded to positive feeling stimuli than they did to negative ones. Similar results were found for the quality measure. Only the contrast between the treatments and controls was significant, E (8,37) = 4.31, p < .001 (see Table 16). Only the constant univariate and step-down E probabilities were significant, E (1,44) = 20.72, p < .0001. Therefore, the difference between the treatment and control groups was due to the treatment, not measures, effect. Neither the treatment (see Table 17) nor the control (see Table 18) contrast produced a significant result. The quality measures effect across treatments, however, did produce significant results, E (8,37) = 38.76, p < .0001. The only significant univariate result besides the constant effect was that for feeling, E (1,44) = 41.60, p s .0001. However, the step-down test result was not significant, E = 6.25, p < .0166. These findings suggest that when the feeling effect was not conditioned on other measures, it produced significant differences across sub- jects, regardless of treatment. The quality mean for expressing positive feelings was 177.46. In contrast, the quality mean for expressing negative feelings was 107.15. As with the amount measure, these results suggest that subjects were able to express better quality responses in positive situations than they were in nega- tive ones. 116 Since none of the anxiety repeated measures planned contrasts produced significant multivariate results (see Tables 20-22), the measures effect was examined. This multivariate test was significant, E (8,37) = 65.70, p < .0001 (see Table 23). This result indicated that a measures effect occurred across treat- ment conditions. In examining univariate and step-down E probabilities, the researcher did not consider the constant effect, since it was the grand mean. However, two effects can be discussed. For the role (initiator, re5pondent) effect, the step-down E probability was sig- nificant, E = 104.38, p < .0001; however, the univariate result was not, E (1,44) = 1.44, p'< .2364. This incon- gruency is difficult to explain; however, the most plausible explanation is related to the nature of these two tests. The step-down E statistic took into consider- ation the arbitrariness of the anxiety scale, whereas the univariate test did not. Most likely, the significant step-down E probability indicated that the anxiety scale had a floor effect. In other words, the anxiety scale did not measure lower levels of anxiety. Therefore, subjects who felt calm and relaxed had scores clustered around the lowest rating on the 1-7 Likert-type scale. This arbitrariness of the scale affected the step-down probability, thus making it significant. On the other hand, the univariate E test result, which did not take 117 into consideration the arbitrary nature of the scale, did not reach significance. Therefore, this researcher concluded that the role effect was not significant across treatment groups. For the feeling (positive, negative) effect, the reverse was true; the step-down probability was not sig- nificant, E = 6.39, p < .0155, whereas the univariate probability was significant, E (1,44) = 96.13, p < .0001. These results indicate that when feeling was conditioned on the constant, role, and sex effects, the feeling effect was not significant. However, by itself, the feeling effect was significant. Across treatment con- ditions the anxiety mean for positive feelings was 19.52, whereas the anxiety mean for negative feelings was 28.23. This finding indicates the subjects were more anxious when reSponding with negative feelings than with positive ones. For the skill test, only the contrast between cognitive structuring and social modeling was significant, E (8,37) = 2.38, p < .0357 (see Table 25). This finding indicates that the treatment and measures effects con- tributed to these significant differences. By examining univariate and step-down probabilities, the investigator determined that none of the measures effects was sig- nificant. The measures effect result across treatments was significant, E (8,37) = 29.39, p < .0001. However, 118 the constant (grand mean) accounted for differences across subjects. None of the other seven variables had a significant result. These results were reported as supplemental find- ings since no formal hypotheses were formulated regarding the outcome. Mean Quality Analysis of Variance The one-way analysis of variance procedure used to test the first research question employed a subject's E933; quality score as one dependent measure. Since this score reflected the amount, as well as the quality, of a subject's response on the performance test, the researcher decided to compute mega quality scores for each subject. The mega quality score, therefore, produced an independent measure of quality unaffected by amount. A one-way analysis of variance for the three planned comparisons with appropriate degrees of freedom and a total alpha level of .05 was then computed. These results are reported in Table 28. None of these univariate test results was significant. However, the finding for the social modeling and cognitive structuring versus the attention-placebo and no-treatment control contrast was E (1,44) = 2.83, p < .10. The cell means and standard deviations for the four treatment conditions on the mean quality measure are presented in Table 29. Treatment 119 Table 28 Analysis of Variance of Mean Quality Score for Three Planned Comparisons Mean Square Univariate E Contrast Between 1, 44 g: E (SM + CS) - (AP + NTC) 200.08 2.83 < .0996 CS - SM 31.05 .44 < .5110 NTC - AP 20.35 .29 < .5943 Table 29 Cell Means and Standard Deviations for Social Modeling, Cognitive Structuring, Attention—Placebo, and No-Treatment Control on Mean Quality Measure Mean Quality Treatment Mean EE Social Modeling 31.99 1.87 Cognitive Structuring 34.27 2.00 Attention-Placebo 29.97 9.69 No-Treatment Control 28.13 13.47 120 cell means were ordered as the investigator predicted; however, the differences among them were slight. Notice, though, that the standard deviations of the two control groups were large, while those of the two treatments were small. These results suggest that the control groups' scores had a wide range of variance, and the treatment groups' scores were quite homogeneous. Again, these findings were reported as additional information only, since no research hypotheses were formulated. Debriefing The 13 subjects who did not receive an intervening treatment (Voight, 1975) prior to the delayed posttest completed a debriefing questionnaire after the testing. Of the 13 returning subjects, eight had received a video- tape treatment, four each in the social modeling and cog- nitive structuring conditions. The major results from this questionnaire are presented in Table 30. Since not all eight subjects answered each question, the response frequency varies from question to question. Subjects generally felt that the videotape or its specific com- ponents such as instruction and modeling were of greatest value in the training. For those who stated that they learned something new about themselves from the training, all responses involved some aspect of increased self- awareness regarding their role in affective communication. Their reported benefits from the experience focused on m n amuoe a cumuumm ”commum moonomcaom a momcommon kuumn mom 302 mmocmaummxm m coapmoaGSEEoo mm>oumaa onu Eonm mouammcmn N mmmcmumsd m>.som amom so» om 30m .m m u amuoe a human @80003 was maaaxm m mmmccmmo N mmcaammm l mmmumxm ou mnmn m.ua umcu mmmcmumsé mumcz R m mmw .Om ma mmammusom usonm m Maammu no: .02 30c mcacumcm cumma so» can .N o u amuoe m ommuooma> a mummocoo mmcumma >am50a>mnm mo 3ma>mm a mcaamooz a mcaEMaQ mo mmmbmca mcaamnma mmcacamnu a coauommmOHucH may ca mpam> umoE.Ho was a mmam anoma>IOamD¢ ammm so» om umns .aamnm>o .a mocmsvmum mmcommom coaymmso muamccoaummno mcammannma Eoum muasmmm om magma Hommz 122 m u Hmuos a ucm>maou mmnm pomhbsm m mmasm 0 mm own on maanumfiom a amcoammmm Ioum mnoE Emmm mocmaummxm mom: a achaumosmm pun msonoasm a moom wamfimuuxo mumuomumno m mm» m m0 ocflx «you was no was «damn o 02 moamfimxm mcaammoE map can .m mumppmn m Houmummc Hmuumm smog m>mc pa mason 30m o n amuoe 5 mm» mummao coapuom o 02 amcoauosnumca may mmz .m "mmmuoo©a> mcu mcamummmm m u amuoe a uomucoo aMGOmHmm 0H0: a ucoaummup cocumcma a coauosmougca mafimmamn mHoE coon ca HoumHHmc Mo huaamso m>oumEH w>mc ua masoo mhwz #853 CH .v mocmsomuh mmcommmm coaumoso concauCOUIIom magma 123 m u amuoa a aamB haamsqm mmmcaaomw m>aummms swap v mow mmmnmxm ou Hmammm o oz mmcaammm w>auamom mum .aa III MO m a p B mmnz m>oE umuaw 03» ome mmgu mmsmomn mmucmEmuwum a Ucommmm ocaawmm mumauaca ou mmcmanm mcofim coaummaabo cm ma Ho mucmfimumum mcaamom mmcaawow mcanmnm mmsmomb cua3 GOmHmm Hngocm op a mumauaca Ucommmn op umammm ua ma .oa I mxaacu v I amuoe 90% oo many mm3 max m aamB maamsom cuom mmmamemm 0» HO mmamfiw a mwamemm ou mmdaammw Hsom mmmumxm H moan: ou nmammm pa mean so» can .m mocmsomnm mmcommom GOaummso concaucooIIom magma 124 improved communication skills, as well as increased awareness of affective self-disclosure. Suggestions for improving the presentation included using a more pro- fessional narrator, lengthening the treatment, and using personal interactions instead of tape recorders. All respondents felt that the instructional portion of the videotape was clear, and three suggested that a better narrator would have improved this section. Six of eight respondents felt that the modeling examples were helpful. According to them, the examples served as guides for their performance, the actors were good, and the subject matter relevant, educational, and humorous. Finally, subjects were asked questions regarding type-of-role, type-of-feeling, and sex-of-best-friend in affective dyadic communications. For these questions, the results were equivocal. The two peOple answering the question on type-of-role split on whether being an initiator or respondent was the easier role to assume. The greatest consensus occurred regarding the type of feeling that is easier to express. Four stated that positive feelings are easier to express than negative ones, while one per- son stated he could do both equally well. Finally, two of four respondents said they could express feelings equally well to males and females. One said it was easier for him to express his feelings to women because 125 they are more empathic, while the other preferred males because he is self-conscious around females. Summary For the primary question of interest, the analysis of immediate posttest data yielded significant results at the .05 significance level. The one-way multivariate analysis of variance performed on the five dependent measures for the first planned comparison indicated that the combined effects of social modeling and Cogni- tive structuring treatments were greater than those for attention-placebo and no-treatment control conditions at the .0004 significance level. Contrary to the investi- gator's prediction, the second contrast, between cognitive structuring and social modeling, was not significant. The third contrast, between attention-placebo and no-treatment control, was also not significant, thus supporting the investigator's research hypothesis. Thus, the research hypOtheses contrasting the treatments versus the controls, as well as the contrast between the two controls, were supported. The question of secondary interest focused on the effects of treatment over time. The repeated measures multivariate analysis of variance test comparing immediate and delayed posttest data indicated no sig- nificant differences for the three planned contrasts for both the treatment and measures main effects. The 126 treatment-by-measures interaction was also not signifi- cant. Thus, the two research hypotheses for the measures and interaction effects were supported. Three supplemental questions of interest were also examined. The first dealt with four repeated measures multivariate analyses of variance procedures to test for type-of-role by type-of-feeling by sex-of- best-friend factors in items on the performance and skill tests. Amount, quality, and anxiety were the three measures assessed on the performance test. A significant difference between the treatment versus no- treatment groups on the amount and quality measures was obtained. However, the treatment, not the measures, effect was responsible for these differences. Across treatments, the feeling effect was significant, thus suggesting that subjects responded more frequently and with higher quality affect to positive stimuli. None of the planned contrasts for the anxiety measure was sig- nificant; however, the measures effect was. The feeling effect was significant across treatments, indicating that subjects were more anxious when responding with negative than positive feelings. For the skill test, the cogni- tive structuring versus social modeling contrast was sig- nificant. None of the other multivariate tests for the four measures was significant. 127 The second supplemental test conducted was an analysis of variance to ascertain the immediate posttest main effect using the mean quality score as the dependent measure. Results of this test were not significant for the three planned contrasts. Third, the results from a debriefing questionnaire given to eight subjects who received no intervening treatment (Voight, 1975) prior to delayed posttesting were reported as supplementary data. CHAPTER IV SUMMARY, DISCUSSION, AND IMPLICATIONS Summary The effects of social modeling and cognitive structuring multicomponent treatment strategies on the affective self-disclosure of single undergraduate” Michigan State University males was the subject of this investigation. Specifically, the objectives of this research were: 1. to assess the impact of social modeling and cognitive structuring strategies on males regarding (a) amount of affect, (b) quality of affect, (c) concomi- tant level of anxiety, (d) skill necessary for affective self-disclosure, and (e) attitudes toward disclosing feelings. 2. to examine treatment effects over time by administering a delayed posttest to subjects who did not receive an intervening self-management procedure (Voight, 1975). 128 129 To meet these objectives, two videotaped presen- tations were developed with the intent of making systematic outcome comparisons between the social modeling and cogni- tive structuring treatments. To ascertain the immediate treatment effect, social modeling and cognitive structur- ing treatment groups were contrasted with attention- placebo and no-treatment control conditions. The design employed for this study was an experimental posttest- only design recommended by Campbell and Stanley (1963). Subjects (E = 48) were randomly assigned to the social modeling (SM), cognitive structuring (CS), attention- placebo (AP), or no-treatment control (NTC) group. Social modeling, cognitive structuring, and attention-placebo treatments were administered individually and were of comparable length. Immediately following treatment, each subject took the performance test and two paper-and- pencil measures. In order to assess treatment effects over time, subjects who received no intervening treatment (Voight, 1975) were retested at a follow-up session three weeks later. The basic design for this phase was a 2 x 4 repeated measures design. In other words, the design contained the immediate and delayed posttest measures across the four treatment conditions. Since only 13 of 16 possible subjects returned for the delayed post- test, unequal cell sizes ranging from two to four were used. 130 The dependent variables used in this study were created to assess amount and quality of affect, concomi- tant anxiety level, skill, and attitudes toward dis- closing feelings. Both the skill and attitude measures were pretested and found to have .31 and .83 reliability coefficients, respectively. Prior to the actual exper- iment, a pilot study was conducted to refine the depen- dent measures and formulate research hypotheses. Relia- bility coefficients for the skill and attitude measures were .91 and .86, respectively. Discrimination and dif- ficulty indices were calculated for both measures, and only the most discriminating items were retained. Selected performance test tapes were used to train raters in using the amount and quality criteria. At the end of training, interrater reliability coefficients of 1.00 and .99 were obtained for the amount and quality measures. For the actual investigation, performance test audiotapes were made into typescripts and were subse- quently rated by two trained raters. For reliability ‘ purposes, the raters independently evaluated eight typescripts, two randomly chosen from each treatment condition. Reliability estimates were calculated for total typescript ratings on amount and quality of affect, as well as for the five quality subscales. Reliability estimates were .99 for amount and quality, while 131 the coefficients ranged from .99 to .97 for the quality subscales. The investigator calculated each subject's total score for the anxiety, skill, and attitude measures. Reliability coefficients were also computed for the skill and attitude measures. The reliability estimate for the skill test was .87, while the coef- ficient for the attitude survey was .84. For the immediate posttest, it was hypothesized that the effects of social modeling and cognitive structuring treatments would be greater than those for the attention-placebo and no-treatment control con- ditions. A second hypothesis stated that the effects of cognitive structuring would be greater than those for social modeling, while a third predicted that no difference would exist between attention-placebo and no-treatment control conditions. For the delayed posttest, the same directional hypotheses were formulated for the treatment effect. In addition, the researcher hypothesized that no measures or treatment-by-measures interaction effects would be found. The data for the immediate posttest were analyzed using a one-way multivariate analysis of variance pro- cedure. Data for the immediate and delayed posttest comparison were analyzed using a repeated measures multi- variate analysis of variance technique. 132 A statistically significant (a=.05) effect was found for the contrast between the two experimental and the two control groups on the immediate posttest. Neither the planned contrast between experimental treatments nor the one between the two control groups was significant. Results for the immediate-delayed posttest comparison were not significant. Three supplemental analyses were conducted. The first, a repeated measures MANOVA, examined three factors--type-of-role, type-of-feeling, and sex-of- best-friend--in items on the performance and skill tests. For the performance test, a significant difference between the treatment versus no-treatment groups on the amount and quality measures was obtained. However, the treatment, not the measures, effect was reSponsible for these differences. Across subjects, however, the feeling effect was significant for'amount and quality. Participants were able to make more affective responses with higher quality to positive situations than negative ones. Only the multivariate measures effect was sig- nificant for the anxiety scale. Once again, the feeling effect produced the only significant univariate result. This finding indicated that subjects were more anxious when making negative affective responses than when making positive ones. The cognitive structuring versus social modeling contrast for the skill test was also significant. None of the other multivariate tests was significant. 133 The second, a one-way ANOVA, used the mean quality score as the dependent measure. None of the planned contrasts proved significant. Third, descriptive data from the debriefing questionnaire showed that experi- mental treatment subjects generally felt that the video- tape presentations were valuable. Furthermore, they reported increased awareness of and improved skills in affective communication as the primary benefits from the training. Limitations Before presenting positive conclusions and implications of this investigation, limitations of the study will be discussed. By specifying limitations, inferences regarding this study will be properly qualified. Furthermore, they will suggest possible refinements and extensions for future research. The posttest-only experimental design employed for the first research question controls well for threats to internal validity (Campbell & Stanley, 1963). However, a possible threat to external validity was the reactive effect. The research was conducted in an experimental setting, and all subjects knew they were participating in an experiment. Therefore, treatment effects cannot be legitimately generalized to indi— viduals receiving training in nonexperimental settings or under nonexperimental conditions. Another threat to 134 external validity involves the interaction of the popu- lation selected and treatment effects. The possibility exists that treatment results are valid for only the unique pOpulation used for the study. Since volunteers had to be actively recruited, the selection-by-treatment interaction poses a more viable threat to external validity. The immediate-delayed posttest design also controls well for threats to internal validity. However, since subjects were tested twice, results of this phase of the experiment can only be generalized to subjects receiving treatment under similar testing conditions. For both videotaped experimental treatments, the male narrator was a junior at Michigan State University. Although the narrator was chosen with no intentional selection bias, several subjects stated that they knew him. Therefore, this acquaintance with the narrator limits external validity of the experiment's results. An additional factor limiting the validity of the experimental treatments might have been the length of the presentation and the amount of material presented. Both the social modeling and cognitive structuring videotaped packages were approximately 50 minutes long. During each presentation, however, instructions, model examples, and guided practice were introduced. There- fore, the amount of material covered within a limited period of time needs to be considered when reviewing the results of this study. Strictly Speaking, the results 135 of these two experimental treatments cannot be generalized beyond the Specific videotaped materials developed for this investigation. Since the investigator could not secure a random sample of single male undergraduate students at Michigan State University, the results cannot be generalized beyond the sample in this study. Subject parameters as well as the sampling procedure were outlined in chapter II. However, by applying the Cornfield-Tukey (1956) bridge argument, the reader may generalize these findings to other populations with similar characteristics. The use of volunteers needs further comment. This investi- gation was advertised as a research/training program to increase affective communication skills. Therefore, sub- jects may have been more amenable to the experimental treatments than nonvolunteers. Furthermore, the volun- teers may have been less inhibited in disclosing feelings than nonvolunteers would have been. Another sample limi- tation was the number of subjects used in the experiment. For the first research question of interest, the inves- tigator attempted to obtain 60 subjects; however, only 48 volunteered to participate. A more serious problem arose for the second question of interest. Given the nature of the design and the follow-up study conducted by Voight (1975), a maximum of 16 subjects could be used for the immediate-delayed posttest comparison. 136 Thus, the statistical power of this repeated measures multivariate test was reduced. Furthermore, since only 13 subjects returned for posttesting, data were analyzed using unequal cell sizes. As is true for many prelimi- nary investigations such as this, a larger number of subjects would have been preferable. The dependent measures used in this study also had specific limitations. Although a performance measure was used to assess amount and quality of affec- tive disclosure, it was only a simulation of actual dyadic interactions. Having situations presented via audiotape with the subject making his responses into a tape recorder was artificial. However, this evalu- ation procedure was chosen in order to hold constant the type of role, type of feeling, and sex of best friend for dyadic situations presented, as well as for the length of re3ponse time across subjects. A performance test involving the subject and his best male and female friends would have been ideal. However, such a test was unfeasible, if not impossible, to construct. Another limitation of the performance test was the criteria it measured. Only the verbal content of subjects' responses was rated. Therefore, the investi- gator neglected to assess voice qualities, such as tone and inflection, as well as nonverbal behavior. However, adequate criteria for the assessment of these variables 137 do not exist. Under these circumstances, the researcher felt that developing original criteria to measure quality of affect content would add to the body of knowledge on self-disclosure research. The rating material and criteria used appeared to be reliable. Both the skill and attitude instrument reliability estimates ranged from .87 to .84, while both the amount and quality interrater reliability coefficients were .99. The performance test raters were described so that the reader can assess their comparability to other populations of raters. Inherent limitations in the use of human raters, however, warrants some discussion. According to Guilford (1954), using peOple as raters assumes that the observer is capable of some degree of objectivity and precision. While varying degrees of confidence can be placed in quantitative human evaluations, weaknesses resulting from personal bias in those judgments must be acknowledged. Rater biases such as errors of leniency, central tendency, and proximity, as well as the halo effect can, at least to some extent, be controlled through the careful selection and training of raters. As discussed in chapter III, qualified people were chosen as raters, and their training sessions planned in detail. A systematic program was used to provide instruction, modeling, practice, and discussion of the rating criteria. Although reliability ratings were high 138 for the second training session and for the actual study, the problems of rating bias and consistent judgmental errors remain as possible limitations of this investi- gation. Each of the dependent measures was constructed for this experiment. Although reliability data were obtained, no attempts were made to validate the instru- ments. Therefore, the absence of validity data is a serious limitation of this study. Finally, a limitation inherent in most educational research applies to the current study. Although the videotaped treatments were effective, what is their cost- efficiency ratio of effectiveness? Are videotape pro- cedures more effective than in vivo presentation methods? If so, does the cost warrant their use? Discussion and Implications Both the discussion and implications of this research must be interpreted within the parameters of this study. The population, procedures, materials, treatments, and dependent measures used in this investi- gation, therefore, define the degree of generalizations the reader can make. Subject Variable Single, undergraduate college males were chosen for this study because previous research suggested that 139 single males disclose feelings less frequently than females or married males. Based on subjective impres- sions of the investigator, the volunteers differed widely in their interpersonal communication skills. Some were advisors in residence halls who had partici- pated in several communication skills workshops. Others had received no formal training, but wanted to improve their social relationships with others. Through random assignment to treatment groups, these subject differ- ences should have been equally distributed across treat- ment conditions. For future research, therefore, sub- jects' social communication skills might be assessed prior to treatment. The length and potency of treatment needed to improve affective communication skills then could be determined. Other subject parameters should also be assessed. Subjects' marital status, sex, age, and mentality are variables worth examining in subsequent research. Also, clinical investigations using patients in therapy are suggested in order to assess the treatment effects of this analogue study. Furthermore, subject learning styles should also be assessed. For instance, it would be helpful to know which subjects learn best from video- tape presentations and which learn best from guided practice. Subject profiles delineating learning styles, 140 modes of processing information, and interpersonal compe- tencies, as well as personality characteristics, would be helpful in determining learning and training sequences for individuals and homogeneous subject groups. Each of these individual and group qualities suggests potential ideas for subsequent research. Dependent Variables Since the amount, quality, skill, and attitude instruments were Specifically developed for this investi- gation and Voight's follow-up (1975), pilot testing of these measures was deemed essential. The refined paper- and-pencil measures used in the actual investigation had respectable reliability coefficients, thus suggesting that both measures were consistent. The reliability of the raters' scoring (chapter III) suggests that trained judges can use amount and quality criteria to make con- sistent ratings from audiotape typescript data. In subsequent research, however, the validity of each instrument needs to be assessed. The present investigation incorporated both amount and quality ratings based on the assumption that each type of data contributed necessary information relevant to affective self-disclosure skills. However, the intercorrelations of the dependent measures (see Appendix L) indicate that the quality and amount measures are largely redundant. In other words, 141 using both of them did not contribute greatly to dif- ferences among treatments. Therefore, future research might pursue the use of either the quality or amount measure in assessing affective self-disclosure skills. Subsequent research might also examine the efficacy of using the mean quality score as a dependent measure. In this investigator's opinion, however, the quality criterion is preferable, since it takes into account the frequency of affective disclosure. Perhaps the amount criterion could be deleted, and the quality measure used exclusively in the performance test. In addition, factors other than content need to be assessed so that more complete ratings of affective self-disclosure can be obtained. Subsequent research, therefore, might focus on developing voice inflection and nonverbal cri- teria to further assess the quality of affective com- munication. Content, verbal, and nonverbal components should be assessed separately by independent raters, since one rater may not be able to attend to three stimuli simultaneously. Each criterion should be rated and then a global rating computed to give an overall assessment of affective self-disclosure skills. Future research might then explore the relative importance of content, verbal, and nonverbal components in affecting the quality of affective self-disclosure. 142 In the performance and skill tests, three situ- ational factors--type-of-role, type-of-feeling, and sex- of-best-friend--were controlled. Future research should be focused upon relationships other than with best friends. For instance, how would subjects respond to acquaintances and strangers in affective dyadic communication? Another factor held constant in the performance test was the length of time (15 seconds) subjects were given to respond to each stimulus situation. Is length of reSponse time a factor influencing amount and quality of affect? To assess this dimension, response time should be varied from less than 15 seconds to an unlimited amount. As previously mentioned, the audiotaped perfor- mance test was a low-level simulation of dyadic inter- actions. Future investigators, therefore, should attempt to use actual people instead of a tape recorder for these interactions. Perhaps standard situations could be role played to hold type-of-role and type-of—feeling constant for all subjects. The confederate roleplayer could then assume different identities such as best friend, acquaintance, and stranger. Both male and female role- players could be used to control for the sex variable. The anxiety self-report measure used with the performance test did not reflect differences among the four treatment conditions. Is anxiety level a factor 143 unrelated to affective self-disclosure? To test this question, future affective self-disclosure research might employ physiological instead of self-report measures to assess concomitant anxiety. If anxiety level continues to produce nonsignificant differences, then the measure probably should be deleted from the instruments used for research on this topic. Although the skill test presented eight types of situations across type-of-role, type-of-feeling, and sex-of-best-friend, it did not contain an equal number of items for each situation. Therefore, mean instead of total scores were computed. Subsequent research should alter the skill test to include an equal number of items for each type of situation. By doing so, total scores for each type of situation could be computed. Using total scores would give more com— plete data than mean scores. Finally, the attitude instrument apparently did not contribute significantly to differences found across the treatment conditions. Was this due to factors associated with the treatments, subject characteristics, or to weaknesses within the instrument itself? Clearly, then, subsequent research should examine the attitude survey in relation to its validity for measuring affec- tive self-disclosure. 144 Treatment Effects The data for the first research question of interest indicate that single male undergraduates at Michigan State University can be trained through social modeling and cognitive structuring videotape presen- tations to increase their affective self-disclosure. Reasons for the absence of significant differences between the cognitive structuring and social modeling treatments may be due to the effects of the anxiety and attitude measures. An analysis of this contrast using only the quality, amount, and skill measures is suggested in order to ascertain whether significant differences exist between the two treatments or whether the treat- ments are, in fact, equivalent. Replications of the current study are needed in order to answer this question. However, results of this study support the effec- tiveness of instruction, modeling, and guided rehearsal for increasing the affective self-disclosure of single undergraduate college males. These three components in each of the treatment packages need to be indepen- dently assessed in order to ascertain the differential effects of each. Next, a comparison between discrete components of the social modeling and cognitive structur- ing packages should be conducted to determine if one approach is superior to the other. In addition, dif- ferent methods for presenting the instruction and 145 modeling sections might be explored. For instance, would written instructions be as effective as those presented on videotape? Also, different types of model- ing procedures might be compared. Contrasting coping, mastery, and appropriate-inappropriate modeling pro- cedures would determine whether or not a particular procedure was more effective. To facilitate learning during guided rehearsal, situations identical to modeling examples could be practiced first. Following this practice, new situations could be introduced. This procedure may help subjects retain affective self- disclosure skills. Immediate posttest results also demonstrated that the attention-placebo and no-treatment control groups did not significantly differ from each other. These results suggest that attention-placebo subjects' expec- tations for gain did not make their performance superior to that of no-treatment control subjects. Therefore, replications of this experiment would not have to include two control groups. The second question of interest dealt with the effects of treatment over time. No significant results were obtained for the treatment, measure, and inter- action effects. The absence of a treatment effect might be explained in several ways. First, since the cell size ranged from two to four, statistical power was 146 greatly reduced and may not have been sufficient to detect group differences. However, examining group means for the immediate and delayed posttests (see chapter III), the investigator noticed that experimental treatment means declined whereas control group means increased on the delayed posttest. Therefore, lack of statistical power does not appear to be a viable expla- nation. A more plausible explanation for treatment groups' losses involves the curve of deterioration. Treatment subjects may have forgotten enough of the experimental training to lower their delayed posttest scores. Quite plausibly, the two 50-minute videotaped presentations may not have been potent enough to main- tain treatment effects over three weeks. On the debrief- ing questionnaire, several subjects stated that too much material was presented within the 50-minute presentation. Therefore, perhaps the nature of the presentation made retention of skills too difficult over the three-week period. The sharp gains made by control group subjects are more difficult to logically explain. Typically, control group scores remain relatively stable over time. What, then, caused their scores to increase so dramati- cally on the delayed posttest? These increases may be due to information gained from the immediate posttest experience. Perhaps subjects sufficiently learned from 147 the testing experience alone to improve their affective communication skills. Another competing explanation is that some learning occurred prior to the delayed posttest. The possibility exists that control subjects either learned through self-practice based on the immediate posttest experience or communicated with experimental subjects. If this is true, then control subjects may have learned affective self-disclosure skills before delayed posttesting. In an attempt to control for this confounding effect, the investigator asked subjects not to discuss the experiment with others participating in the training. However, the investigator had no way of enforcing this request. Although the two preceding explanations are plausible, a third seems more viable. The extremely small cell size for each control group (attention-placebo gy= 3; no-treatment control g_= 2) lacked stability. Therefore, the dramatic increase for the no-treatment control group and, to a lesser extent, for the attention-placebo condition may simply be due to measurement error. Therefore, the most plausible expla- nation for no-treatment effect is that treatment sub; jects' retention of learning sufficiently deteriorated. The gain by control subjects is more difficult to under- stand. Since maintenance of desired change is an impor- tant criterion for any treatment, future research efforts 148 should include a delayed posttest follow-up. However, based on the results of this experiment, the potency of the two experimental treatments should be increased. Potency could be increased in several ways. First, each videotaped component could be lengthened to make the presentation more thorough. Second, the number of training sessions could be increased. For instance, an hour session might be devoted to instruction, a second to modeling, and a third to guided practice. Ideally, these sessions would be spread over several weeks. Of course, a larger sample size should be used for any immediate-delayed posttest comparison to reduce chances for measurement error to affect the results. The results from the follow-up study by Voight (1975) indicate that subjects who had one of the videotape treatments and one of her self-management procedures did significantly better than those who had only one of the treatments or had neither. She also found no differences between the effects of videotape, self-management, and no-treatment conditions. Her findings suggest that the effects of the videotape treatments can be significantly increased by adding a three-week self-management follow- up procedure. Therefore, replications to test this combination of procedures are recommended. 149 Supplemental Findings The feeling effect was the only repeated measure that produced significant results. This effect, however, was only significant across treatment conditions. In other words, a differential feeling effect was obtained for all subjects, regardless of treatment. This feeling effect was significant for the amount, quality, and anxiety measures. These results suggest that subjects were able to respond affectively more often, with better quality, and with less self-reported anxiety in positive, rather than negative, situations. Therefore, investi- gators might emphasize the development of affective self- disclosure skills in negative stimulus situations. Instruction, modeling, and guided practice could be constructed to emphasize the expression of negative feelings, since the results of this investigation indi- cate that negative feelings are more difficult to express than positive ones. In addition, this investigator recommends that sex-of-best-friend, as well as type-of- feeling, be the subject of subsequent research. Although the sex-of-best-friend effect was not significant in any of the repeated measures analyses, the reduced alpha level of .006 for each univariate test contributed to this lack of differences. By using only the feeling and sex variables, along with the constant, the overall alpha level would remain .05, but the alpha level for 150 each univariate test would be changed to .01, rather than the extremely rigorous .006 level used for this study. This larger univariate alpha level would most likely detect sex differences, if, in fact, they exist. The marginal results for the sex and feeling-by-sex effects across groups suggest that sex is a situational factor influencing the expression of feelings. In light of these findings, future research on these situational variables is recommended. To ascertain whether or not sex of subject is a factor, future inves- tigations should include females in the sample. As reported in chapter III, the ANOVA using the mean quality score as the dependent measure pro- duced no significant results for any of the three planned contrasts. These findings suggest that a pure measure of quality not influenced by amount does not differentiate among the four treatment conditions. Therefore, subse- quent research on affective self-disclosure might best use the total instead of the mean quality score as one dependent measure. Finally, the debriefing questionnaire results provided the investigator with subjective feedback from the subjects. The results from this questionnaire suggest the following implications for future research. First, both videotaped presentations should be redone, using a more professional narrator. Second, segments, 151 especially the one on guided practice, should be expanded. Third, actual people should be used instead of the tape recorder for the performance test. Implications for Counseling and Education The results of this study have provided prelimi- nary evidence suggesting that treatments based on social learning principles are effective methods for increasing affective self-disclosure. Although replications and additional research are needed to substantiate these treatment effects, some practical implications for coun- seling and education can be tentatively postulated. Counselors working with clients deficient in social communication skills could employ the treatments used in this investigation to help improve their expres- sion of feelings to others. In both individual and group settings, the counselor could serve as an instructor and model of self-disclosure skills. Furthermore, the counselor could help clients practice these skills through guided practice procedures. In addition, both social learning treatments might be used as precounsel- ing strategies. In other words, clients could view a social learning videotape prior to their involvement with counselors. By doing so, clients would learn that disclosing feelings is a responsibility they have in 152 the counseling relationship. Therefore, these video- taped presentations may help clients to better under- stand their role with counselors. Educational implications for these social learning treatments are also exciting. In a broad sense, these strategies might be incorporated into preventive mental health programs for children. These programs should help children express feelings more openly and more appropriately. Therefore, children may become better adjusted through exposure to these affective communication skills procedures. The dif- ferential effectiveness of these treatments for dif- ferent age groups needs to be assessed. However, research in this area offers exciting possibilities for viable contributions to be made in the educational domain. Conclusion In the discussion and implications of this investigation, the major foci have been: (a) the training of single undergraduate college males using social modeling and cognitive structuring techniques for increasing affective self—disclosure immediately following treatment, and (b) the comparative assessment over time of these affective communication skills developed in training. The role of situational 153 factors in affective dyadic communication was also explored as a supplementary question of interest. In the past, behavioral scientists have been criticized for not conducting rigorous experimental research in the affective domain. This researcher, therefore, attempted to expand knowledge of affective communication by developing dependent measures to quan- titatively assess affective self-disclosure. In addition, two treatment strategies based on social learning prin- ciples were devised to train single undergraduate males at Michigan State University to improve their affective communication skills. Results suggest that both methods were effective in accomplishing this objective. Further research in these social learning procedures would increase effectiveness in the training of verbal and nonverbal affective communication skills with a variety of populations. The criterion measures may also be applied to future descriptive studies to examine various situational factors operating in affective dyadic com~ munication. This research has attempted to demonstrate that the humanistic goal of improving affective communication skills can be attained through behavioral methods. The investigator believes that these skills are vital for effective interpersonal communication and that their acquisition can be greatly facilitated through the use of behavioral techniques. APPENDICES APPENDIX A ADVERTISEMENTS APPENDIX A ' IMPROVE YDIIR .COMMIIIICAI’IOI SKILLS . A short term research and training 1 program is being offered during the remainder of this term. The focus of the training is on the expression of feelings to others. If you are a single undergraduate male you I qualify for the program. There will be two sessions and total time commitment will be approximately 4 hours. Sessions will be arranged according to free time in your class schedule. All materials will be provided, and there is no charge for this training. I I If you In Interested, call 353-3190. This advertisement appeared in the State News, Michigan State University's student newspaper, on Aprl 24-25, 1975. COMMUNICATIONS SKILLS TRAINING. Learn to express personal feelings more effec- tively. Free 4-hour training. Call 353-3738 or 355-1755. Volunteers. This advertisement appeared in the classified section of the State News, Michigan State University's student news- paper, on April 28-May 2, 1975. 154 APPENDIX B FLYER APPENDIX B COMMUNICATION SKILLS A slam-4 - 'l-erm resan’Clfi/‘h’famins program W3” be conducl-ed during #he remainder O‘F’ ‘H'Ie. ieI-meI-m program will focus owl-he. expressim 0‘ 4::9, ““95 ‘lo other-s, IF you are a singh, male. undergraduai-e, ou 03M: 410v- “M's ram. There will Be ZYSeSSions,Yand +o+al +73 commi-I-menl will be Approximately Ll hour-s, Ses- sions will be arranged according +0 ‘Free, +ime 'In our class Schedule. All ma‘l'cv-Ials wIll be Pro~ vided, and Hus-e, Is no charge {or 4555 hralnins. IF YOU ARE I NTERESTED, CALL: ESE-HS 5 on. 353-3798 on 3'42- 896 ANYTIME. 155 APPENDIX C LETTER TO SUBJECTS APPENDIX C Dear Your next appointment for Communication Skills Training is scheduled for at in room 250 Erickson Hall. If you have been working on a self-management project, please bring your cards in with you. Otherwise, all materials will be provided. It is VERY IMPORTANT to us that you come in for this appointment. In fact, we have planned to present some tangible rewards for your attendance. You have a choice of money, a bottle of wine, or a plant. If you need to change the time of your appointment or have any other problem, please call one of us soon. Otherwise we'll see you at your appointment time. Thank you for your c00perationl and I-k—bkflek MW Phone numbers: 355-1755 353-3798 372-8913 156 APPENDIX D SOCIAL MODELING VIDEOTAPE TYPESCRIPT APPENDIX D SOCIAL MODELING VIDEOTAPE TYPESCRIPT Introduction NARRATOR: Interpersonal communication--or talking with peOple-- is something each of us does every day. Because this process is such a basic part of our lives, we often don't think much about it. However, what we say and how we say it are important factors in determining the quality of relationships we have with other people. Basically, we communicate in two ways. We tell others what we think and what we Eggi. Our thinking communication conveys objective or factual information such as: "I have a test today" or "My friend is upset." We call this cognitive communication. In contrast, communicating feelings lets pe0p1e know our emotional or affective state. "I'm happy today" and "My feelings are hurt" are two examples of affective communi- cation. It's easier for us to communicate our thoughts rather than our feelings. However, heavy or exclusive reliance on c0gnitive communication tends to make our relationships with others superficial. Clearly communicating our feelings makes these relationships more meaningful. The purpose of this presentation, then, is to help you become a better communi- cator of feelings. 157 158 This presentation is divided into four parts. The first is an instructional unit. In this unit, we will talk about reasons why it's difficult to express feelings. We'll also talk about environmental factors that can increase or decrease the likelihood of peOple sharing their feelings. Advantages for sharing feelings will also be presented. This section will be followed by a brief self-test so that you can check your understanding of this material. Modeling is the title of the next section. Here appropriate and inappropriate examples of expressing feelings will be presented along with a brief discussion of each example. At the end of this unit, a second self-test will be given. The third phase is called Behavior Rehearsal. During this phase, you will practice making apprOpriate feeling responses to specific situations. Finally, the fourth section will briefly review the material covered in the other sections. At that time, I'll give you instructions for beginning the next phase of the program. Please do not adjust the videotape equipment during this presentation. If a technical problem occurs, go to the door and ask an assistant for help. Instructional Unit Generally speaking, males in this culture are raised to believe that it's "unmanly" to disclose their feelings to others. Females, on the other hand, are encouraged to com- municate feelings. This socialization process is learned by both males and females. Throughout our lives, we have 159 learned how and when to communicate feeling largely by observing others. While we were growing up, our fathers and other important men in our lives had a big impact on how we learned to communicate feelings. We also learned some "should's" regarding the expression of feelings. For example, "men should keep their feelings to themselves." "Expressing emotion is a sign of weakness." and "Only women should cry." With all of these socialization influences, it's a wonder that men express any feelings at all! But what has been learned can be unlearned, and new approaches to communicating can be sub- stituted for old ones. So, then, what are some of the factors that can help us ggy_in expressing our feelings more openly? First, the influence of others. If friends are inter- ested in knowing how we feel, we will be more likely to share our feelings with them. Likewise, if friends tell us how they feel, we will probably disclose feelings more often to them. Second, the consequences or outcomes for expressing feelings. If friends are receptive--that is, if they react positively to us when we disclose feelings, we will more likely continue sharing our emotions with them. Also, seeing others who disclose their feelings can also help--especially if the person they're talking to reacts in a positive way. In other words, watching how others share their feelings can help increase our own expression of emotions. What happens, though, if we want to express our feelings more often, but don't have friends who are interested in listening, or don't have friends 160 who usually share their feelings with others? Well, we can just go right ahead and share our feelings anyway-~and reward ourselves for doing so. For example, if I want to tell my best friend that I'm angry because he didn't show up for a meeting, I can do so even though he might not "thank" me for telling him my feelings. I can reinforce myself by saying-- "Gee, I feel a lot better now" or "I'm really proud that I could tell him my feelings!" Therefore, we really don't need acknowledgement from others in order to share our feelings with them. We can "pat ourselves on the back" instead. This gelff reward can keep us going as we try to be more open in dis- closing feelings. Third, the influence that our thoughts can have on whether or not we share feelings. If we inaccurately perceive the consequences for communicating our feelings, we probably won't express them. For example, if I make the assumption that I'll be rejected by a friend for telling him I'm angry, I'll probably keep that feeling to myself. However, it's yegy unlikely that I'll lose his friendship by just telling him my feelings. Therefore, inaccurate perceptions such as these can really get in the way of telling people how we feel. In summary, then, it's important to be aware of three factors that influence disclosure of feelings. First, other people who are receptive to us sharing feelings and who, in turn, share emotions with us will help increase our expression of feelings. 161 Second, positive consequences for sharing feelings with others can help, too. These positive consequences can come from others, from ourselves, and from observing others disclose feelings. And finally, our thoughts about the consequences for revealing feelings can help. However, these thoughts must be positive in order for us to be more likely to disclose emotion. Now that we know some of the factors that can help us express feelings, let's look at the ways we can communicate them. In communicating anger, notice the difference between these two statements: (a) "You make me angry" and (b) "I'm angry at you." In the first, the speaker is blaming the other person. In the second, the speaker is taking responsibility for them. In other words, he is owning his feelings. Blaming statements tend to put the listener on the defensive and most likely will cause that person to respond negatively. In order to reduce ambiguity or uncertainty, it's best to state a reason for the feeling expressed. In the preceding example, "I'm angry at you because you didn't keep our appointment" is better than simply saying: "I'm angry at you." There are three general advantages for disclosing feelings appropriately. First, if we share feelings, others will be more likely to communicate their feelings to us. Thus, disclosing feelings can bring us closer to others. 162 Second, in order to communicate feelings, we must first recog- nize and label them. Therefore, revealing emotions helps us get to know ourselves better. Third, sharing feelings simply makes us feel better! Verbalizing emotions can reduce tension and frustration caused by keeping feelings to ourselves. So, working on communicating our feelings more freely can benefit us all. Instructional Unit Self-Test NARRATOR: On the table in front of you is a PEEK packet titled: Instructional Unit Self-Test. Please refer to it now. As I read each question, please mark the appropriate response. 1. The ability to disclose feelings is a learned reSponse. (a) True (b) False 2. Through the socialization process, men have learned to disclose feelings more often than women. (a) True (b) False 3. A person will be more likely to express his feelings if: (a) Others reward him (b) He sees others praised for disclosing feelings (c) He rewards himself (d) All of the above (e) Only a & c 4. If friends share their feelings with us, we'll probably share our feelings with them. (a) True (b) False 5. Which of the following is an example of a blaming state- ment? (a) You're going to the store. (b) You're happy that he called. (c) It's your fault I lost my temper. (d) None of the above 10. [.4 O O 163 People don't learn much about communicating feelings by observing others. (a) True (b) False Which of the following is the best way to disclose feelings of depression? (a) You made me feel depressed because you called me a bad name. (b) I'm really feeling depressed. (c) You made me feel depressed. (d) I'm depressed because you called me a bad name. Disclosing feelings can help people get to know them- selves better. (a) True (b) False Sharing feelings can make people feel better. (a) True (b) False Expressing feelings does not facilitate the communication process. (a) True (b) False (PAUSE) The answers are: True False D True C False D True True False \DmflmUI-waI—J 0 If you got eight or more right, you're ready to move on to the second unit. Seven or fewer correct responses suggests that a review of this unit is in order. If you answered seven or fewer items correctly, go to the door and ask the 164 assistant to rewind the tape. Eight or more right, sit tight; unit two is coming up. Modeling Unit NARRATOR: In the communication process, a person assumes two roles: the initiator and the respondent. We initiate feeling statements, and we respond with feelings to statements made by others. In the last unit, we mentioned that the KEY we express feelings is just as important as the expression itself. In other words, some ways of expressing emotions are better than others. To make these distinctions clear, you will see sixteen vignettes with Denny interacting with his best male and best female friend. These examples will illustrate both positive and negative feeling expressions, with Denny acting as both as an initiator and respondent in the communication process. A five-item self-test will follow the model examples. (FADE OUT NARRATOR/FADE IN DENNY & DIANNE) DENNY: Hey Dianne, would you bring me my Pepsi? DIANNE: Where did you leave it? DENNY: I think I left it on the floor by your big plant, out there. Hey, I really like this chair. Did you carve it your- self? DIANNE: Out of an elephant tusk. Really. DENNY: Did you capture the elephant yourself? DIANNE: Yeah . DENNY: Well, it's really neat. I really like it. It's comfortable--sort of a mini chair. 165 DIANNE: Denny, I like you. I feel comfortable with you, and I think you're a fine person. DENNY: Well that's good. I think it's nice that people feel comfortable with each other. DIANNE: But, how do you feel? DENNY: About what? About people feeling comfortable with each other? DIANNE: No, no, about me. Are you comfortable? (FADE OUT/CUT SCENE BACK TO NARRATOR) NARRATOR: In this situation, Denny did not acknowledge Dianne's compliment. Instead, he made a general, evasive comment which added nothing to their verbal exchange. Notice also that Denny's reSponse did not satisfy Dianne. She picked up on the fact that he avoided communicating his own feelings. Let's view the same situation again. This time with Denny clearly communicating his feelings to Dianne . . . (FADE OUT NARRATOR/BACK TO DENNY & DIANNE) DENNY: Yeah, that looks really good. Could you pick up my Pepsi, too? Looks like nice snacks. Are we having a party this early in the afternoon? DIANNE: They aren't very good for you, but they taste good. (Drinks Denny's Pepsi) I'm sorry. That's yours, isn't it? I left mine in the other room. DENNY: It's ours. DIANNE: That's nice to share our Pepsi. DENNY: With a friend. 166 DIANNE: A good friend. Denny, I feel comfortable with you and real relaxed. I think you're a fine person. DENNY: That's really nice, for me. I feel comfortable with you, too, and relaxed. And that just makes me really happy that we can share that with each other and not feel threatened. I just think that's really neat. DIANNE: That's nice to hear. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: This time, Denny's clear communication of his feel- ings affected Dianne in a much more positive way. He took responsibility for his feelings. In addition, he told her HEY he felt good. By clearly stating his feelings and the reasons for them, Denny dealt directly with Dianne's statement. Notice, too, that Dianne appreciated Denny's Openness in sharing his positive feelings. But people communicate negative, as well as positive feelings. So let's look at another situation between Denny and Dianne. (FADE OUT/BACK TO DENNY & DIANNE) DENNY: Hey Dianne. Hey baby, nice to see you. DIANNE: Hey, hot shot. Nice to see you about 45 minutes late. Better late than never, I suppose. DENNY: Damn right. I'm here. DIANNE: Where were you? You're not going to get off that easy. Where were you? DENNY: I'm not trying to get off that easy. I was down at the bar having a drink with a couple friends of mine, Doug and Dan. 167 DIANNE: That's real nice. I've been sitting here for forty- five minutes picking my nose. DENNY: Get into that, do you? You look pretty serious. DIANNE: I am serious. Why don't you get serious? You can never be serious. Like at the party the other night . . . DENNY: Real nice party, wasn't it? DIANNE: Nice for you with all your friends. I didn't know anyone, and you took me there and didn't even bother to intro- duce me. DENNY: I swear you're so damn dependent, you can't even walk around the block. DIANNE: I don't need to depend on you; let me tell you that. DENNY: You sure don't look like it. What am I getting all this stuff for? You are just too damn clinging. DIANNE: I think I'll show you the door, boy. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: In this case, Denny didn't take responsibility for his feelings of anger. His response was an angry one that blamed Dianne for his feelings. Furthermore, his reply did nothing to facilitate their communication. Dianne became angry, and they're now involved in a heavy argument. A much better response would have been for Denny to tell Dianne he felt angry because of her comment. Let's look at the same situation again-~this time with Denny handling the situation more appropriately with much dif- ferent results. (FADE OUT NARRATOR/BACK TO DENNY & DIANNE) 168 DENNY: Chubby Checker doing his twist! Boy, did I have a nice time at the bar! DIANNE: Where the hell have you been? DENNY: I don't know if I can handle this. Maybe these bubbles are going to my mind. Maybe you're not really here. DIANNE: You're forty-five minutes late. I hardly believe you're here. DENNY: I kind of question it myself. DIANNE: I'm sick of this, Denny, I really am. I can't depend on you for anything. You were suppose to call me last night at eight, and you didn't. Today you show up forty-five minutes late, and last night, when we went to the party, you abandoned me the whole night--went off with your friends. I didn't know a soul there. I don't have to take this shit. DENNY: Yeah, I really agree with you. You don't have to take this shit, as you call it. I guess I've really come down about all this pretty quickly, but I'm kind of puzzled too. I didn't know that these things bothered you so much. That my not being here on time concerned you so much 'cause I was talking about you so much to my friends. DIANNE: You could be dead in a ditch for all I know. DENNY: So you're more concerned about me than . . . DIANNE: No, I'm more mad, I think, than concerned. DENNY: I caught a little bit of concern there 'cause like I'm concerned with you, and that's why I'd like to know more about this. But like I said, I'm pretty puzzled. It's not my inten- tion to be in some way showing you up or anything by not being here on time. 169 DIANNE: Well, maybe I jumped to some conclusions. I'm glad you're willing to talk about it, 'cuz I'm pretty upset. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Wasn't that an improvement? By recognizing and clearly stating his feelings and the reasons for them, Denny opened the way for a discussion of Dianne's hurt feelings. In contrast, his first response led right into an argument. It's important to clearly communicate our feelings to 3E; people. Let's now look at several situations where Denny is talking with Bill, his best male friend. (FADE OUT NARRATOR/FADE IN DENNY & BILL) DENNY: I'm still losing my voice from all that hooting and hollering at the bar last night. BILL: I'm still drinking. As usual, Lizard's and you go hand in hand. I had a hell of a good time, and not only that. You know, Denny, we've been friends for over 10 years now, and your friendship means a hell of a lot to me, Denny. DENNY: Same here. BILL: Well, you say same here, but it means more to me than that. It means more to me in the sense that I guess I love you like a brother. DENNY: It's uh good to have friends. BILL: Whew. It was hard for me to say something of that nature, and I guess you obviously don't feel the same way that I do. (FADE OUT SCENE/BACK TO NARRATOR) 170 NARRATOR: Denny has once again run away from feelings that his best friend has shared with him. By making a general statement that has nothing to do with his feelings at the moment, Denny has hurt Bill. Most likely, Bill will be reluctant to share his warm feelings for Denny as a result of this conversation. Let's look at a more appropriate way to handle this same situation. BILL: Lizard's was great last night, wasn't it? DENNY: Sure was. I tried to get a lot of sleep but . . . BILL: You see what I'm still doing here. DENNY: Sleeping? BILL: Drinking. I had a great time. DENNY: It's really great that we have such good times down at the bar and other places too, you know. BILL: I'm glad you brought that up because I was thinking after I went back to my place, "Man I'm really lucky to have such a good friend as Denny." DENNY: Same here. I feel the same about you. BILL: It's more than luck, and it's more than just friendship, you know. I guess I really love you a lot. DENNY: I'm really embarrassed by that Bill, but I'm deeply touched. In a way, I don't know how to handle that from you. We have been friends for 10 years, but still that doesn't make it any easier. But I'm deeply touched. It's just hard for me to say "I love you" to another man. It just seems that in this society, and, I guess I'll have to put the responsility on me, it's just hard for me, but I care a hell of a lot for you, too. 171 BILL: I guess what you're saying is that it's hard for you to accept this--as hard as it is for me to say that I love you, Denny. I'm real glad that we're getting this chance to talk about it. DENNY: Right, that just makes our friendship more and more and deeper and deeper. BILL: You bet. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Oftentimes, it's very difficult for us to share such intimate feelings with others--especially other men. Denny's second response really identified the feelings he was exper- iencing--his embarrassment, his difficulty in saying "I love you" and then sharing these feelings with Bill. Even though it was hard for Denny to respond with his real feelings, his efforts paid off. The two men feel much closer as a result. Now let's see how Denny handles another situation with Bill. This time, Bill is giving Denny some negative feedback. (FADE OUT NARRATOR/FADE IN DENNY & BILL) DENNY: Say Bill, how's it going? BILL: You took your sweet ass time answering the door, but I noticed that you didn't take your time moving in on my woman last night. I couldn't believe it--my favorite woman, Nancy. At the party last night, I take the time to introduce her to my best friend, and you just move right in; push old Bill to the side and start talking about your Goddamn karate lessons, and I felt lucky that I could drive Nancy home last night. Man, I didn't get a chance to say a damn thing to her. I'm really fucking hot. What have you got to say about that? 172 DENNY: Well first of all, I'd like to say that I'm feeling really puzzled about the party and about what you're saying right now. Nancy asked me a question about karate, and I guess I got carried away. It wasn't my intention, though, to push you out of the way. I'm just really puzzled. BILL: Well, I guess that makes me want to cool my jets. You know how jealous I get. I just didn't realize. I guess I was over imagining something. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Good for you, Denny! This time Denny identified the emotion he felt in response to Bill's criticism and stated it along with the reason for his feelings. Notice, too, that Bill did not take responsibility for his feelings toward Denny. Fortunately, though, Denny was able to own his feelings, thus facilitating their communication. Finally, because of the 33y Denny handled his feelings, Bill cooled down, and peace and friendship were restored. Let's quickly run through the same situation again. This time with Denny making an inappropriate response. (FADE OUT NARRATOR/BACK TO DENNY & BILL) (LOUD KNOCKING AT THE DOOR) DENNY: Come in. Come in. BILL: All day long I could knock on that door, and you wouldn't even move out of your damn chair. DENNY: You don't have to knock, Bill. We're good friends. 173 BILL: I wanted to get into something a little bit more than just knocking, because what I wanted to say was, you're slow moving here, you son-of-a-bitch, but you sure weren't slow moving in on my woman last night. I bring her to a party and introduce my favorite woman, Nancy, to my best friend, Denny, and, boy, I should have known it, you moved right in on the conversation and pushed me right out of the way, talking about karate and this and that, and you just pushed me right out of the way. What have you got to say about that? DENNY: I got to say that you should be big enough not to be pushed out of the way. You're just talking like a little boy. This is just totally ridiculous, you know. And if you can't keep your own women under control who's gonna do it for you? Can I help it if she's attracted to me? BILL: You S.O.B. I should have known it. Everytime I have a girl and I introduce her to you, you try to snake her. With friends like you, I don't need enemies! (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: What a blow-up! Denny got sucked into an argument by venting his feelings in a blaming way. The point, then, should be clear by now. When responding to positive or nega- tive statements made by others, it's best to: (a) Identify your feelings at the moment and reasons for these feelings; (b) Then think about the 33y to communicate these feelings. Will my response facilitate communication? g£_Will it cause a communication breakdown? (G) Then respond by clearly com- municating your feeling and reasons for them to the other person. 174 Interpersonal communication involves two roles. One as a respondent to communication; the other as an initiator of communication. The preceding examples showed Denny as a respondent to his best friends, Dianne and Bill. Although the role is slightly different as an initiator, the same basic rules apply. Now let's look at Denny as an initiator of feelings--first with Dianne, then with Bill. In the first situation, Denny and Dianne are sitting on the couch in her apartment. They just came from an exciting rock concert. Denny really likes Dianne and he's debating whether or not to tell her . . . (FADE OUT NARRATOR/FADE IN DENNY & DIANNE) DENNY: All I got to say is that was a nice show. DIANNE: Can I get you a beer? DENNY: No. Not right now. DIANNE: Rock yourself down here. That was fun, Denny. I liked it a lot. DENNY: Yeah. You know, this was the third time that we went out together, and the concert was really nice, but how do you feel about me? DIANNE: What made you say that? I don't know what to say! (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Looks like Denny chickened out. He sure knew how he felt about Dianne, but was just too uptight about her rejecting him to tell her. Instead, he put Dianne on the Spot by asking how gEg_felt about him. Dianne didn't know what to say, and said so. Thus, Denny made the situation uncomfortable 175 for both of them. Let's give Denny another chance using the same situation. (FADE OUT NARRATOR/BACK TO DENNY & DIANNE) DENNY: Hey, Dianne. What a nice concert, hey? DIANNE: Would you like anything from the kitchen? DENNY: No. No thanks. This is the third time we've gone out. We've had such a nice time. I feel comfortable with you. I feel good with you--a great time when we go out. You know, I like you a lot. It's really neat. DIANNE: I like you a lot, and that's really nice to hear. (DIANNE KISSES DENNY) (FADE OUT SCENE/ BACK TO NARRATOR) NARRATOR: Well, it sure looks like Denny can handle the situation from here! Sharing strong feelings usually makes us feel better--regardless of the outcome. And telling some- one how you feel helps clarify your relationship with that person. Keeping positive or negative feelings to ourselves can result in lots of frustration. Now let's look at another situation. This time Denny is jealous because Dianne went out with another man. He's gone over to her apartment to talk about it. (FADE OUT NARRATOR/ BACK TO DENNY & DIANNE) DIANNE: Well, Denny, what a nice surprise! DENNYk "What a nice surprise Denny. What a nice surprise. Sit down, Denny." A little Redbook huh? From what I hear you should have a little black book, baby. 176 DIANNE: What's wrong? DENNY: I'll tell you what's wrong. What the hell are you doing going out with that other guy? I heard about this. Aren't I good enough for you? DIANNE: Don't give me that jive, turkey. I don't have to take that from anyone. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Denny's sure venting his emotions, but not in a way that's going to help his relationship with Dianne. He is not owning his feelings of jealousy. Rather, he is putting Dianne on the defensive by blaming her. Let's look at a better way to handle the situation. (FADE OUT NARRATOR/BACK TO DENNY & DIANNE) DIANNE: Denny, what a nice surprise! DENNY: I imagine it kind of is, and it kind of isn't. Listen, I haven't been able to study all day. Like I heard this thing about--well, basically, I'm just jealous, "cause I care for you, and I heard that you're going out with this other guy, and you didn't even say anything about it. I just haven't been able to keep my mind on my books, and I just wanted to come over and talk about it and see if you do care for me, if things are going all right--if we can figure something out here. It's just really bothering me. DIANNE: Gee, Denny, I didn't know you felt that way. I'm really glad you came over. (FADE OUT SCENE/BACK TO NARRATOR) 177 NARRATOR: By communicating his feelings along with reasons for them, I'm sure Denny feels a lot better. He's also opened the way for a good discussion with Dianne. In contrast, his first response started one hell of an argument. Now let's go back to situations between Denny and Bill. In the first one, Denny has just returned from talking with Dianne. He's really excited about his relationship with her and wants to tell Bill about it. (FADE OUT NARRATOR/FADE IN DENNY & BILL) BILL: Denny, how you doing? DENNY: Nice to see you, Bill. I just came from Dianne's. We had a talk, and she just feels like we're getting along really well. She thinks we have a good relationship. BILL: Well, Denny how do ygg_feel about the situation? (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Although there's nothing really wrong with what Denny just said, he didn't communicate his egg feelings of excitement to Bill. Bill was more interested in Denny's feelings. Let's look at the same situation again. This time with Denny communicating his own feelings. (FADE OUT NARRATOR/BACK TO DENNY & BILL) BILL: Hey, Denny, how's it going? DENNY: Fantastic! I'm just feeling really happy. I just came from Dianne's, and we just had the nicest talk. She was saying that she feels really good about our relationship, and I do, too. I just feel really happy and lighthearted and like a bubble floating across the water on a spring day. 178 BILL: Denny, I feel happy that you feel happy. Tell me more about your relationship with Dianne. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Denny felt happy and excited because of what Dianne said. Generally speaking, it's better to communicate our feel- ings at a given moment as well as the reasons for these feel- ings. Now let's look at a final example of communicating our feelings to others. In this case, Denny has just finished taking a test which he thinks he's failed. He's really depressed and wants to talk with Bill about it. (FADE OUT NARRATOR/BACK TO DENNY & BILL) DENNY: Hey, Bill. BILL: I got a minute or two. Exam week, you know. I under- stand you had an exam this morning, man. DENNY: Yeah. A big one. I'm feeling pretty down and depressed about it. Like I studied these books so hard, and the material was so important for me, and it was a pretty important exam just in itself for my grade, you know? And I think I flunked it. I'm really down about it. It's good to see you studying, but, gees, I worked so hard, and I think I flunked it. BILL: Maybe you did a little better than what you're thinking. DENNY: I don't know. I appreciate your support. I knew that if I could come and talk to you you'd give me a hand with it. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Denny's statement was an appropriate example of clearly communicating negative feelings. Oftentimes in 179 situations like these, people have a tendency to emote feelings, but the feelings are used in a blaming way. Here's an example of "blaming feelings," using the same situation. (FADE OUT NARRATOR/FADE IN DENNY & BILL) BILL: What's happening, Denny? DENNY: What a paradox. BILL: Well, it's exam week. DENNY: It sure is exam week. Don't bother. Take my advice. BILL: I'm studying my ass off. What happened with your exam this morning? DENNY: Studying your ass off. You're just about as dumb as I am. One good friend to another. What happened in TX exam? Those asses. Those professors. They're so dumb, I run out of words for them. They're so ridiculous. The questions on the exam were so ridiculous. They don't know a thing. I don't know how they got their degrees. I think they just paid for it, and anyone who could persist long enough in swallowing all that bullshit. The test was just nothing. Give it up, man. BILL: No, no, no, Denny. I heard that from John about twenty minutes ago, and if you got nothing better to do than complain, let other people study. Please. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Denny vented his feelings all right, but he did so in a blaming way. Notice that Bill was not receptive to Denny's bitching, and the resulting argument certainly didn't do anything to help their friendship. 180 We've just seen eight different situations in which feelings were appropriately and inappropriately expressed. In order to effectively communicate feelings, remember the following guidelines. First, identify the feeling you have at a given moment in a Specific situation. Second, in com- municating it, be sure to own that feeling by using "I" statements that are present-oriented. For example, "I'm happy,‘ not "I was happy"--followed by a Specific reason for that feeling. By doing so, you will be able to more effectively communicate your feelings to others, which will improve your interpersonal relationships. Modeling Unit Self-Test In order to see if you understand the concepts pre- sented in this section, you will be presented with five situations similar to the ones previously seen. After each vignette, decide whether Denny's responses appropriately communicated his feelings to his best friend. On the table in front of you is a green sheet titled: Modeling Self-Test. Please record your answers to each situation presented on EEEE sheet. (FADE OUT NARRATOR/FADE IN SCENE) BILL: Hey, Denny. Looks like you're studying hard there. How's it going? DENNY: It's not going good. I'm feeling really uptight. I've got an exam in two hours, and I just don't think I'm ready for it. It's going to be a toughie. 181 (FADE OUT/15 SECONDS BLANK) DIANNE: Hey, I want to know what's going on here. You just told me the other night how much you like me, and now, all of a sudden, you're seeing someone else. DENNY: I don't see what business it is of yours. You're just like a clinging vine or something hanging on me. I think you're too dependent. Why don't you just lay off me. It's none of your business. (FADE OUT/15 SECONDS BLANK) DENNY: Hey, Bill. I'm really angry, man, because you had the study notes for our study group meeting, and you didn't show up. You said you were going to be there. I'm angry. Where were you anyhow? (FADE OUT/15 SECONDS BLANK) DENNY: Hi Dianne. Say, you look really nice today. It's a nice Spring day. DIANNE: Are you ready for Art History? DENNY: No. I don't think so. I don't get much out of that course and that instructor he . . . No, I don't think I'm going. Maybe we could do something else. (FADE OUT/15 SECONDS BLANK) BILL: Hi Denny. I stopped over to see how that job inter- view went this morning. DENNY: I'm feeling pretty positive about that. I got some positive feedback from her, and she thinks I'm a good candi- date, so I'm feeling pretty good. I think I might get the job. (FADE OUT/15 SECONDS BLANK/BACK TO NARRATOR) 182 NARRATOR: The correct answers to the five situations are: appropriate for #1; inappropriate for #2, since Denny made bliuning statements; appropriate for #3; inappropriate for #4, txecause Denny's response was cognitive and did not deal wiifli his feelings; and apprOpriate for #5. If you labeled four or five of the responses cor- ixurtly, you are ready to continue with the next section. If ytnixnissed two or more, then view this section again. Go to time door and ask an assistant to rewind the videotape. Please do not.rewind the tape yourself. Behavior-Rehearsal Unit ZNARRATOR; Now you're ready to practice expressing your feelings in appropriate ways. You will be presented with eight dif- ferent situations. Four will involve interacting with your own best EEAE friend, and four with your best female friend. After each situation is described, you will see either a male or female on the screen. Please pretend that this person is your own best friend and respond accordingly by expressing your feelings. Please make your reSponse out loud. For each situation, you will have 15 seconds in which to make your response before the next one is presented. After making your response out loud, silently reinforce yourself for expressing your feelings. You might say something like: "Hey! I really did a good job." 2E "I feel lots better having just said that!" If you have any questions regarding these instructions, please go to the door now and ask the assistant. (10 SECONDS BLANK) 183 NARRATOR: You're sitting in your room talking to your best male friend. You're involved in a heavy discussion that's lasted several hours. It's late, and you have an early class tomorrow. You're feeling exhausted and want to go to bed. You turn to him and say . . . (15 SECONDS ON BEST MALE FRIEND) (FADE OUT/BACK TO NARRATOR) NARRATOR: If you said something like: "Gee, I'm really tired. I want to hit the sack now since I've got an early class tomorrow. Let's continue our discussion tomorrow,‘ you've got the right idea. Remember to first identify and then com- municate your feelings and reasons for them to the other person. Now for the next situation. You and your best male friend are sitting in a bar drinking. You're having a great time and want to tell him how close you feel to him. You turn to him and say . . . (15 SECONDS ON BEST MALE FRIEND) (FADE OUT/BACK TO NARRATOR) NARRATOR: Were you able to tell him you care for him and why? (PAUSE) Now, put yourself in this situation. You're in a restaurant eating lunch with your best male friend. He casually mentions that he has a date tonight with the woman you've been dating. After he tells you this, you turn to him and say . . . (15 SECONDS ON BEST MALE FRIEND) (FADE OUT/BACK TO NARRATOR) 184 NARRATOR: Did you identify and then communicate your feelings in a nonblaming way and then give the reason for them? (PAUSE) Good! Now try this one. You're in your room and have just finished helping your best male friend write up a project for one of his classes. He turns to you and says . . . (FADE OUT NARRATOR/FADE IN BEST MALE FRIEND) BEST MALE FRIEND: I really appreciate us being such good friends. I'm glad we're friends. (FADE OUT/BACK TO NARRATOR) NARRATOR: You respond to him by saying . . . (15 SECONDS ON BEST MALE FRIEND) (FADE OUT/BACK TO NARRATOR) NARRATOR: Now imagine this scene. Your best female friend has just finished typing a term paper for you. You turn to her and say . . . (15 SECONDS ON BEST FEMALE FRIEND) (FADE OUT/BACK TO NARRATOR) NARRATOR: An example of an appropriate response is: "Thanks! I really appreciate the typing you did for me." Now put yourself in this situation. Your best female friend was supposed to meet you for lunch today, but didn't show up. You're angry with her for not keeping the date. Later on you run into her. You go up to her and say . . . (15 SECONDS ON BEST FEMALE FRIEND) (FADE OUT/BACK TO NARRATOR) 185 NARRATOR: One appropriate response would be: "I'm really angry that you didn't show up for lunch. What happened?" Now consider this situation. You and your best female friend are talking. In the conversation, she cuts down one of your good friends. You respond to her by saying . . . (15 SECONDS ON BEST FEMALE FRIEND) (FADE OUT/BACK TO NARRATOR) NARRATOR: You might have said something like this: "I'm upset by your negative comments about my friend. If you have some- thing to say about him, why don't you tell him directly?" Now please respond to one final situation. You have just finished changing a flat tire for your best female friend. She turns to you and says . . . (15 SECONDS ON BEST FEMALE FRIEND) BEST FEMALE FRIEND: Thanks for changing that tire! You're the greatest! (CAMERA STILL ON FEMALE FRIEND) NARRATOR: And you reSpond by saying . . . (15 SECONDS ON BEST FEMALE FRIEND) (FADE OUT/BACK TO NARRATOR) NARRATOR: A typical response might be: "Hey! I feel really good because you said that. Thanks a lot!" Well, how'd you do? In order to express emotions, we need to consciously practice every day. This behavior rehearsal is just a beginning. Hopefully, you'll transfer this learning into your every day interpersonal communications. (FADE OUT) 186 Review NARRATOR: In summary, then, clearly expressing feelings involves the following: First, own the feeling by using an "1" statement with the emotion; second, state the feeling, using the present tense; third, give a reason for the feeling. The appropriate expression of feelings tends to facilitate interpersonal communication. The modeling examples contrasted appropriate and inappropriate methods of disclosing feelings. The behavior rehearsal section provided Opportunities to practice the skills learned in preceding sections. However, as we just stressed before, only by transferring the skills learned here to your every day life will make this learning experience a meaningful one. You have now completed this phase of the program. Please Open the door and tell the assistant you are finished. The assistant will stop the videotape machine and then give you specific instructions regarding the next phase. (VOICE IN BACKGROUND: Hey, what about Denny & Dianne?) NARRATOR: Oh, yes. Well, they still have their up's and down's, but they're both still expressing their feelings in meaningful ways--both verbally--(CAMERA ON DENNY & DIANNE MAKING OUT) and nonverbally! APPENDIX E COGNITIVE STRUCTURING VIDEOTAPE TYPESCRIPT APPENDIX E COGNITIVE STRUCTURING VIDEOTAPE TYPESCRIPT Introduction NARRATOR: Interpersonal communication--or talking with people--is something each of us does every day. Because this process is such a basic part of our lives, we often don't think much about it. However, what we say and how we say it are important factors in determining the quality of relationships we have with other people. Basically, we communicate in two ways. We tell others what we E2325 and what we £22l° Our thinking communication conveys objective or factual information such as: "I have a test today" or "My friend is upset." We call this cognitive communication. In contrast, communicating feelings lets people know our emotional or affective state. "I'm happy today" and "My feelings are hurt" are two examples of affective communication. It's easier for us to communicate our thoughts rather than our feelings. However, heavy or exclusive reliance on cognitive communication tends to make our relationships with others superficial. Clearly communicating our feelings makes these relationships more meaningful. The purpose of this presentation, then, is to help you become a better communicator of feelings. The presentation is divided into four parts. The first is an instructional unit. In this unit, we will talk about reasons why it's difficult to express feelings. Thought processes that affect the communication of feelings will be discussed, and 187 188 advantages for sharing feelings will be presented. This section will be followed by a brief self-test, so that you can check your understanding of this material. Modeling is the title of the next section. Here appropriate and inapprOpriate examples of expressing feelings will be presented along with a brief discussion of each example. At the end of this unit, a second self-test will be given. The third phase is called self-modeling. During this phase, you will practice making appropriate feeling responses to specific situations. Finally, the fourth section will briefly review the material covered in the other sections. At that time, I'll give you instructions for beginning the next phase of the program. Please do not adjust the videotape equipment during this presentation. If a technical problem occurs, go to the door and ask an assistant for help. Instructional Unit NARRATOR: Generally speaking, males in this culture are raised to believe that it's "unmanly" to disclose their feelings to others. Females, on the other hand, are encouraged to communicate feelings. This socialization process is learned by both males and females. Thus, it follows that if men learn to suppress or inhibit their feelings, they can also learn to express them more freely. Our minds play an important role in the communication pro- cess. Before speaking, the brain assesses the situation and makes some quick decisions regarding what we actually say. These 189 mental processes largely determine whether we will express feel- ings, and, if we do, the way in which these feelings are stated. Two factors involved in this thinking process are: (a) anticipated consequences, and (b) irrational self-statements. Anticipated con- sequences are the results or outcomes we think will occur from disclosing feelings. Irrational self-statements are subjective, internal dialogues we hold with ourselves regarding these antici- pated consequences. To better understand these two concepts, consider this example. A man loves a woman. What may prevent him from sharing these feelings with her? First, he may anticipate that she will not feel the same way about him. If she doesn't, he will feel rejected. This is an example of an anticipated conse- quence that can hinder disclosure of feelings. While thinking of this anticipated consequence, he may make irrational self-statements regarding the outcome. For example: "Wouldn't it be horrible and terrible if She rejects me! And if she rejects me, that means I'm no good, and no other woman will want me." Even if the woman rejects his overtures of love, it certainly does not follow that this man is no good, nor that other women will find him unattrac- tive. Both anticipated consequences and irrational self-statements can hinder the disclosure of positive feelings. Likewise, anticipated consequences and irrational self- statements can hinder the expression of negative emotions. Often- times, people are reluctant to express negative feelings because they think others will view such expressions as personal attacks on them. A typical anticipated consequence for telling a best friend that you're angry is that he will be hurt, angry, or upset 190 with you. An irrational self-statement like this might follow: "My best friend won't like me anymore--how awful! I'll be rejected, and no one else will like me either!" These c0gnitive processes usually occur instantaneously. Oftentimes, we aren't even aware of them. But they g9 occur and can seriously inhibit the disclosure of feelings. So, an impor- tant first step in communicating feelings is to become aware of these cognitive processes. However, this awareness is not enough to make us effec- tive communicators of feelings. The E3y_in which feeling is expressed is as important as expressing the emotion. In com- municating anger, notice the difference between these two state- ments: (a) "You make me angry!" and (b) "I'm angry at you!" In the first, the speaker is blaming the other person for his feelings. In the second, the speaker is taking responsibility for them. In other words, he is owning his feelings. Blaming statements tend to put the listener on the defensive and most likely will cause that person to reSpond negatively. In order to reduce ambiguity or uncertainty, it's best to state a reason for the feeling expressed. In the preceding example, "I'm angry at you because you didn't keep our appoint- ment" is better than simply saying "I'm angry at you." There are advantages for disclosing both positive and negative feelings. First, if we share feelings, others will be more likely to communicate their feelings to us. Disclosing feelings, then, can bring us closer to others. Second, in order to communicate feelings, we must first recognize and 191 label them. Therefore, revealing emotions helps us get to know ourselves better. Third, sharing feelings can simply make us feel better. Verbalizing emotions can reduce tension and frus- tration caused by keeping feelings to ourselves. So, working on communicating our feelings more freely can benefit us all. Instructional Unit Self-Test NARRATOR: On the table in front of you is a Elge_packet titled: Instructional Unit Self-test. Please refer to it now. As I read each question, please mark the appropriate response. 1. The ability to disclose feelings is a learned response. a. True b. False 2. Through the socialization process, men have learned to disclose feelings more often than women. a. True b. False 3. "He will reject me if I tell him I love him" is an example of: a. Affective Self-Disclosure b. Anticipated Consequence c. Irrational Self—Statement d. Blaming Statement 4. "Wouldn't it be horrible if he rejects me; that will mean I'm no good" is an example of: a. Blaming Statement b. Irrational Self-Statement c. Anticipated Consequence d. Affective Self-Disclosure 5. "It's your fault I lost my temper" is an example of: a. Irrational Self-Statement b. Appropriate Affective Self-Disclosure c. Anticipated Consequence d. Blaming Statement 192 6. Learning to identify feelings is the only skill necessary for appropriately disclosing emotion. a. True b. False 7. Which of the following is the best way to disclose feelings of depression? a. "You made me feel depressed because you called me a bad name." b. "I'm really feeling depressed." c. "You made me feel depressed." d. "I'm depressed because you called me a bad name." 8. Disclosing feelings can help people get to know them- selves better. a. True b. False 9. Sharing feelings can make people feel better. a. True b. False 10. Affective self-disclosure does not facilitate the communication process. a. True b. False (PAUSE) The answers are: 1. True 2. False 3. B 4. B 5. D 6. False 7. D 8. True 9. True 10. False If you got eight or more right, you're ready to move on to the second unit. Seven or fewer correct responses suggest that a review of this unit is in order. If you answered seven or fewer items correctly, go to the door and 193 ask the assistant to rewind the tape. Eight or more right, sit tight; unit two is coming up. (10 SECONDS BLANK TAPE) Modeling Unit NARRATOR: In the communication process, a person assumes two roles: the initiator and the respondent. We initiate feeling statements, and we reSpond with feelings to statements made by others. In the last unit, we mentioned that the E§y_we express feelings is just as important as the expression itself. In other words, some ways of expressing emotions are better than others. To make these distinctions clear, you will see 16 vignettes with Denny interacting with his best male and best female friend. These examples will illustrate both positive and negative feeling expressions with Denny acting as both an initiator and respondent in the communication process. A five- item self-test will follow the model examples. (FADE OUT NARRATOR/FADE IN DENNY & DIANNE) DENNY: Dianne, would you mind bringing me my beer? I left it out there. DIANNE: Sure. DENNY: Thank you. It's nice around your apartment. DIANNE: Denny, I feel real at ease with you. I'm real com- fortable. I think you're a fine person. DENNY, VOICE OVERLAY: (Why did she say that? What does she want from me? I'm really uncomfortable.) 194 DENNY: Well Dianne, it's nice when people feel comfortable. (FADE OUT/CUT SCENE BACK TO NARRATOR) NARRATOR: In this situation, Denny felt uncomfortable because of Dianne's compliment. He chose not to acknowledge the compli- ment 2£ his embarrassment by making a general, evasive comment which added nothing to their verbal exchange. Notice also that his internal thoughts before his response negatively affected the type of reSponse he made. Let's View the same situation again--this time with different self-thoughts that lead to a more appropriate response. (FADE OUT NARRATOR/BACK TO DENNY & DIANNE) DENNY: Say, Dianne, could you please bring me my beer? I left it on your desk. I'd appreciate it. DIANNE: Here. And a flower for your hair. All of it. Denny I feel real relaxed with you and real comfortable, and I think you're a fine person. DENNY, VOICE OVERLAY: (Wow I really feel good! That's really nice of Dianne to tell me that.) DENNY: I think that's nice Dianne. I really appreciate that. I feel good about that. I'm comfortable with you, too. We have a really nice time, and that's just really good. (FASE OUT SCENE/BACK TO NARRATOR) NARRATOR: This time, Denny's internal processing of Dianne's compliment abknowledged the good feelings he felt. In his response to her, he owned his feelings by saying "I feel really good." In addition, he told Dianne why he felt good: "because 195 you feel so comfortable around me." Finally, he shared his feelings about her with her. Therefore, by accurately identify- ing our feelings (as represented by internal self-dialogue), we are in a better position to own them when communicating with others. Stating the reason for our feelings adds clarity to our response, thus facilitating interpersonal communication. But people communicate negative, as well as positive, feel- ings to us all the time. So, let's look at another situation between Denny and Dianne. (FADE OUT/BACK TO DENNY & DIANNE) DENNY: Hey ho, Dianne! Nice day, huh? DIANNE: Hey ho, big shot. Just trot yourself in here about 45 minutes late, why don't you? Where were you? DENNY: Out drinking. Having a really nice time. DIANNE: Yeah. Out drinking. Just like at the party the other night. You dropped me off at the door; we were supposed to be together, and that's the last I see of you. I always have to play second fiddle, don't I? DENNY, VOICE OVERLAY: (Gees, what a nagger! She can't even go to a party without hanging on me every minute.) DENNY: Listen, Dianne. I swear. Who needs you anyhow? Here I come over here in a good mood. You're just acting ridiculous. Stupid, in fact. And further more, what really hugs the shit out of me is that you're just so dependent. I'm sick of it. Just too dependent. (FADE OUT SCENE/BACK TO NARRATOR) 196 NARRATOR: In this case, Denny failed to recognize and take responsibility for his feelings of anger. His response was an angry one blaming Dianne. Furthermore, his response did nothing to facilitate their communication. A far better response would have been for Denny to tell Dianne he felt angry because of her comment. Let's look at the same situation again. This time with a different internal dialogue that leads to a better feeling response. (FADE OUT NARRATOR/BACK TO DIANNE & DENNY) DENNY: Hey ho, Dianne! What are you doing? I just been out drinking. . . . DIANNE: Sit down. But not too close. DENNY: Yes Ma'am. DIANNE: Forty-five minutes late. I ought to make you sit on the floor. Where have you been? DENNY: Out drinking. DIANNE: This is what I get for all the attention I pay to you and all the time I spend. I'm at your beck and call, and you can't even show up on time. I'm sick of it. DENNY, VOICE OVERLAY: (I didn't realize Dianne felt that way. That's upsetting.) DENNY: I'm really disappointed. I didn't realize that I neglected you in that way and that you took it in that way. That just upsets me. Like I came over here in a good mood today, and I wanted that to carry over, and I just didn't realize that I was doing that to you. Maybe we can figure something out. 197 (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Wasn't that an improvement? Denny was able to label his feelings and the reasons for them. He then communi- cated them to Dianne. His second approach to the situation has opened the way for a discussion of Dianne's hurt feelings; where the first paved the way for an argument. It's important to communicate our feelings to all people. Let's now look at several situations where Denny is talking with Bill, his best male friend. (FADE OUT NARRATOR/DENNY & BILL IN) BILL: We had a really good time at Lizard's last night, didn't we? DENNY: Yeah. BILL: Well, Denny, we've had some really good times over the years. I've always appreciated your friendship, you know, and I guess I want to take this time to tell you that your friend- ship means a lot more to me than just friendship. It's something that's kind of hard to say. I guess I kind of love you. DENNY, VOICE OVERLAY: God, I'm embarrassed. I'm stuck for something to say. I really care for Bill, but it's really strange to hear him say that he loves me. DENNY: Well, you know, hell. Friendship's really great, man. (FADE OUT SCENE/FADE IN NARRATOR) NARRATOR: Well, that time, Denny chose to evade Bill's feelings, as well as his own, by making a very general comment that says nothing about his feelings at the moment. A much better reSponse would be . . . 198 (FADE OUT/FADE IN DENNY) DENNY: Bill, I'm really embarrassed by that, but I'm deeply touched by your sharing those feelings with me. I really care a lot for you, too. It's just so hard for me to say that to another man. It just seems in this society that's wrong, and I have trouble dealing with that. I really appreciate you sharing that with me, you know. (FADE OUT/FADE IN NARRATOR) NARRATOR: Oftentimes it's very difficult for us to share such intimate feelings with others--especially other men. Denny's second response really identified the feelings he was exper- iencing--his embarrassment, his difficulty in saying, "I love you," and then sharing these feelings with Bill. Let's now see how Denny handles another situation with Bill. This time, Bill is giving Denny some negative feedback. (FADE OUT/FADE IN DENNY & BILL) DENNY: Come on in! BILL: Take your time answering the door. Have I got some things to say to you! DENNY: What's up? BILL: Last night is what's up. I can't believe it. My best friend. We go to a party together, and I was with Nancy, right? I bring her in and introduce her to you. You start impressing her; you make me look like an ass, because you didn't include me in the conversation, and it really bummed me out, man. I felt like a jackass. What have you got to say about that? 199 DENNY, VOICE OVERLAY: (What? How did he get that impression? Is that the way I really came across?) DENNY: Gees, Bill. I'm really confused, really perplexed about this. I really like you; I really like Nancy, and it was nice getting to know her, you know, because we're such good friends. Gee, she asked me a question about Sports. I guess that's where it maybe all started, and I did talk an awful long time about my karate, but really, one friend to another, it wasn't my intention at all to leave you out of the conversation. Really, I mean that. (FADE OUT/FADE IN NARRATOR) NARRATOR: Good for you, Denny! This time Denny identified the emotions he felt in response to Bill's criticism and stated it along with the reason for his feelings. Notice, too, that Bill did not take reSponsibility for his feelings toward Denny. Instead of saying "I really feel bummed out," Bill placed blame on Denny by saying "ygg really bummed me out." Fortunately, though, Denny was able to own his feelings, and by doing so avoided a possible argument. Let's quickly run through the same situation again--this time with an inappropriate response. (FADE OUT/FADE IN DENNY & BILL) DENNY: Say Bill, how's it going? BILL: Say Bill, say Denny, take your time answering the door. I'm really pissed off about last night, man. I had a real good time with my best friend and my best girlfriend with the way you acted. 200 DENNY: Good. BILL: Don't give me that, man. You acted like a super jock, tried to EEEEE.NanCY away from me. Tried to take over the con- versation. Man, that's it. That's it with our damn friendship. DENNY: Well, you know, that's probably a pretty good point you brought up. Who are you? Who the hell are you anyway? Mr. Dude? I mean, when it comes down to it, big man, I didn't put any button on your lip. You could have spoken up anytime you wanted to. Nancy's a nice lady. No one stopped you from talking. You're such a damn moron; sometimes you just have nothing vital to say. (FADE OUT/FADE IN NARRATOR) NARRATOR: The point, then, should be clear by now. When responding to positive and negative statements made by others, it is best to: (Cognitive Process) Identify (1) Your feelings at the moment (2) Reasons for these feelings, (ReSponse) And then communicate them to the other person. Interpersonal communication involves two roles. One as a :respondent to communication, the other as initiator of communi- Ication. The preceding examples showed Denny as a respondent to 1113 best friends, Dianne and Bill. Although the role is slightly different as an initiator, the same basic rules apply. Now let's look at Denny as an initiator of feelings, first with Dianne, then with Bill. In the first situation, Denny and Dianne are sitting on the couch in her apartment. They just came from an exciting rock concert. . . . 201 (FADE OUT/IN ON DENNY AND DIANNE) DENNY, VOICE OVERLAY: (I really like Dianne, but I don't know how she feels about me. Should I tell her I really like her? What if she rejects me?) DENNY: Gees, I just had such a nice time at that rock concert. The group was so good, you know. This is the third time we've gone out, you know. I think that's neat. Three times. How do you feel about me? (FADE OUT/ IN ON NARRATOR) NARRATOR: Looks like Denny chickened out. He sure knew how he felt about Dianne, but was too uptight about being rejected to tell her. Instead, he put Egg on the spot by asking how she felt about him. Generally speaking, it's usually best to share our feelings with others. Even if Dianne doesn't feel the same way about Denny, a rebuff from her won't be the end of the world for him. Let's give Denny another chance, shall we? (FADE OUT/IN ON DENNY & DIANNE) DENNY: You know, this is like the third time we've gone out. I've just had so much fun, you know, being with you. It's just been so nice. I just really care for you a lot. I just feel so comfortable. Like, I can laugh or I think I could cry with you or do anything. I just care for you a lot. It's really neat. (FADE OUT/IN ON NARRATOR) NARRATOR: Regardless of Dianne's response, I'm sure Denny feels better because he shared those feelings with her. Keeping positive or negative feelings to ourselves can result in lots 202 of frustration and uncertainty. Now let's look at another situation. This time, Denny is jealous because Dianne went out with another man. He's gone over to her apartment to talk about it. (FADE OUT/IN TO DENNY & DIANNE SITTING ON COUCH) DIANNE: Denny, what a nice surprise! I didn't expect to see you until tonight. DENNY: Yeah, I'll bet that you didn't expect to see me. DENNY, VOICE OVERLAY: (Yeah, baby, I bet you thought it was that other guy. I'm really jealous.) DENNY: Listen, I'll just lay it on out. I hear you've been going out with some other guy, as if I'm not good enough for you, as if whatever I have isn't enough. I think that's a bunch of bullshit. You really think you're tough shit, don't ya? (FADE OUT/IN TO NARRATOR) NARRATOR: Denny's sure venting his emotions, but not in a way that is going to help his relationship with Dianne. He is not owning his feelings of jealousy. Rather, he is putting Dianne on the defensive by blaming her. Let's look at a better way to handle the situation. (FADE OUT/IN TO DENNY & DIANNE) DIANNE: Come in, Denny. What a nice surprise! DENNY: Yeah, I imagine it kind of is and kind of isn't. Listen, I haven't been able to study all day. Basically, I'm just jealous, and I heard that you're going out with this other guy. You didn't even tell me, and I just haven't been able to keep my mind on my books, and I just wanted to come 203 over and talk about it. I wanted to see if you do care about me and if things were going all right. It's just really bother- ing me. (FADE OUT/ IN TO NARRATOR) NARRATOR: By communicating his feelings along with reasons for them, I'm sure that Denny feels a lot better. He's also opened the way for a good discussion with Dianne. In contrast, his first reSponse most likely would have started one hell of an argument! Now let's go back to situations between Denny and Bill. In the first one, Denny has just returned from talking with Dianne. He's really excited about his relationship with her and wants to tell Bill about it. (FADE OUT/FADE IN DENNY & BILL) DENNY: I just came from Dianne's. I was just talking to her. BILL: What's up? DENNY, VOICE OVERLAY: (My relationship with Dianne is really great!) DENNY: Well, I had some things on my mind as I came from there. She's just really happy about our relationship. She just thinks we're getting along fine. She thinks we've got it together. (FADE OUT/ FADE IN NARRATOR) NARRATOR: Although there's nothing really wrong about what Denny just said, he didn't communicate his gwn_feelings of excitement to Bill. In his internal self-dialogue, Denny identified his feelings and the reason for them. However, in talking with Bill, he chose to tell him Dianne's feelings. 204 Let's look at the same situation again, this time with Denny communicating his own feelings. (FADE OUT/FADE IN DENNY) DENNY: I just came from Dianne's. Such a nice talk, such a nice time. She's really happy about our relationship, and I'm really happy about how I'm becoming involved in the relation- ship, and things are just going really good. We just talk like you and I talk. We're really becoming friends, and more than friends. I'm just really happy. (FADE OUT SCENE/BACK TO NARRATOR) NARRATOR: Denny felt happy because of what Dianne said. Generally speaking, it's better to communicate our feelings at a given moment, as well as the reason for these feelings. Now, let's look at a final example of communicating our feelings to others. In this case, Denny has just finished taking a test which he thinks he's failed. He's really depressed and wants to talk with Bill about it. (FADE OUT/FADE IN SCENE) DENNY: Hi, Bill. DENNY, VOICE OVERLAY: (I'm so down. That test was just miserable. I wonder what I should tell Bill?) DENNY: Well, I'm just really depressed about that exam. I studied really hard for that. I went through these books back, forward, every which way, and I just don't think I passed this exam. It meant a lot to me. I'm interested in the class, and I really got into it, and I think I flunked it. 205 (FADE OUT/FADE IN NARRATOR) NARRATOR: Denny's statement was an appropriate example of clearly communicating negative feelings. Oftentimes in situ- ations like these, people have a tendency to emote feelings, but the feelings are used in a blaming way. Here's an example of "blaming feelings" using the same situation. (FADE OUT NARRATOR/FADE IN DENNY & BILL) DENNY: Those professors are such assholes! I swear; they don't even know how to blow their own noses! That test was ridiculous! Look at these books! Those professors and these books go hand in hand. No knowledge in either one. (FADE OUT DENNY AND BILL/FADE IN NARRATOR) NARRATOR: Denny was emoting, all right, but in a blaming way. He vented his anger, frustration, and depression by blaming his prof. A more appropriate way to vent feelings is to own them ourselves. "I'm angry/frustrated/depressed because I just failed a test" is a much better way to communicate negative feelings. We've just seen eight situations in which feelings were appropriately and inappropriately expressed. In order to effectively communicate feelings, remember the following guidelines: Our internal dialogues, or the statements we make to ourselves immediately before initiating or responding with feelings, are extremely important in determining what we actually say. Thus, it's important to first be aware that these self-statements occur even though they're often instantaneous. Next, focus on altering 206 self-statements so that you accurately identify and label your feelings at a given moment in a given situ- ation. In addition, identify the reasons for these feelings. Then respond by owning your feelings ("I feel" statements), using the present tense (e.g., "I'm happy" rather than "I was happy"), followed by the reasons for your feelings. By doing so, you will first become more aware of your feelings, and second, will be able to communicate them effectively to others, thus improving your inter- personal relationships. Modelinngnit Self-Test In order to see if you understand the concepts presented in this section, you will be presented with five situations similar to the ones previously seen. After each vignette, decide whether Denny's response appropriately communicated his feelings to his best friend. On the table in front of you is a yellow sheet titled: Modeling Self-Test. Please record your answers to each situation presented on this sheet. (FADE OUT NARRATOR/FADE IN DENNY & BILL) DENNY: I'm studying really hard. Bill, I'm really anxious. I got a really big midterm in a couple of hours that's worth half my grade, and I'm really not sure how I'm going to do on it. (FADE OUT/15 SECONDS BLANK) DIANNEH I really don't understand this, Denny. You told me the other day that you really like me, and now you're seeing another girl. 207 DENNY: Demand, demand, demand. Why don't you get off my back! We aren't married or anything. What the hell do you make yourself to be? (FADE OUT/15 SECONDS BLANK) BILL: Hi, Denny. What's up? DENNY: Well, a few things are up. You know we had that study meeting the other night, and I'm really angry because you had the notes, and we were counting on the notes. We couldn't do any studying! BILL: You mean you were waiting around for me? DENNY: Yeah, and I'm really pissed. (FADE OUT/15 SECONDS BLANK) DENNY: Busy, busy, busy. DIANNE: Yeah, are you ready to go to class? DENNY: Naw. I'm not going to that class. I don't think it's got much to offer me, and that professor isn't much anyway. So, I don't think I'm going. Maybe we could do something else. (FADE OUT/15 SECONDS BLANK) BILL: How'd that job interview go today? DENNY: It was for a part-time job. I really don't know how it went. I'm kind of uncertain. I'm really hopin' that she'll hire me, but I'm feeling up-in-the-air right now. (FADE OUT/15 SECONDS BLANK/FADE IN NARRATOR) NARRATOR: The correct answers to the five situations are: appropriate for #1: inappropriate for #2, since Denny made blaming statements; appropriate for #3; inappropriate for #4, 208 because Denny's response was cognitive and did not deal with his feelings; and appropriate for #5. If you labeled four or five of the responses correctly, you are ready to continue with the next section. If you missed two or more, then view this section again. Go to the door and ask one of the assistants to rewind the videotape. Please do not rewind the tape yourself. (15 SECONDS BLANK) Cognitive Self-Modeling Unit NARRATOR: Now you're ready to practice some situations your- self. The method we'd like you to follow involves imagining situations to yourself. I will describe eight different situ- ations. As I describe each one, I want you to close your eyes and put yourself into the situation. Really get into each one. Then imagine yourself making feeling responses to each situation. Say your response silently to yourself. You will have 15 seconds in which to practice your response before the next one is pre- sented. If you have any questions regarding these instructions, please go to the door and ask the assistant. (FADE OUT/15 SECONDS BLANK) NARRATOR: Relax by taking a deep breath, and get your body in a comfortable sitting position. Now close your eyes, and vividly imagine yourself in the following situation. (Slight pause) You are sitting in your room talking to your best male friend. You're involved in a heavy discussion that's lasted several 209 hours. It's late, and you have an early class tomorrow. You're feeling exhausted and want to go to bed. You turn to him and say . . . (BLANK TAPE FOR 15 SECONDS) NARRATOR: If you said something like: "Gee, I'm really tired. I want to hit the sack now since I've got an early class tomorrow. Let's continue our discussion tomorrow" you've got the idea. Remember to first identify and then communicate your feelings and reasons for them to the other person. Now close your eyes again, and imagine this next situ- ation. You and your best male friend are sitting in a bar drinking. You're having a great time and want to tell him how close you feel to him. You turn to him and say . . . (BLANK TAPE FOR 15 SECONDS) NARRATOR: Were you able to tell him you care for him and why? Close your eyes again, and put yourself in this situation. You're in a restaurant eating lunch with your best male friend. He casually mentions that he has a date tonight with the woman you've been dating. After he tells you this, you turn to him and say . . . (BLANK TAPE FOR 15 SECONDS) NARRATOR: Did you identify and then communicate your feelings in a nonblaming way and then give the reason for them? (Pause) Good! Now try this one. Again, close your eyes and imagine this situation. You're in your room and have just finished helping your best male friend write up a big project for one II III II‘ I [cl lull 210 of his classes. He turns to you and says: "I really appreciate all the help you've given me. You're the best friend I could ever have!" You respond to him by saying . . . (BLANK TAPE FOR 15 SECONDS) NARRATOR: Now imagine this scene. Your best female friend has just finished typing a term paper for you. You turn to her and say . . . (BLANK TAPE FOR 15 SECONDS) NARRATOR: An example of an appropriate response is: "Thanks! I really appreciate the typing you did for me." Now close your eyes again and imagine this situation. Your best female friend was supposed to meet you for lunch today, but didn't show up. You're angry with her for not keeping the date. Later on, you run into her on the street. You go up to her and say . . . (BLANK TAPE FOR 15 SECONDS) NARRATOR: One appropriate response would be: "I'm really angry that you didn't Show up for lunch. What happened?" Now imagine yourself in this situation. You and your best female friend are talking. In the conversation, she cuts down one of your good friends. You respond to her by saying . . . (BLANK TAPE FOR 15 SECONDS) NARRATOR: You might have said something like this: "I'm upset by your negative comments about my friend. If you have something to say about him, why don't you tell him directly? Now, please imagine one final situation.) You have just changed a flat tire for your best female friend. She thanks .‘1II|‘| III-lit! I] l. 211 you and says you're the greatest guy in the world. You respond to her by saying . . . (BLANK TAPE FOR 15 SECONDS) NARRATOR: A typical response might be: "Hey, I feel really good cuz you said that. Thanks!" Well, how'd you do? In order to express emotions, we need to consciously practice daily. This cognitive self-modeling practice is just a beginning. Hopefully, you'll transfer this mental practice into actual behavior in your own life. (FADE OUT) Review NARRATOR: In summary, then, clearly expressing feelings involves the following: first, own the feeling by using an "I" statement with the emotion; second, state the feeling using the present tense; third, give a reason for the feeling. An essential com- ponent in this process is being able to identify your feelings when they occur. The appropriate expression of feelings tends to facilitate interpersonal communication. The modeling examples contrasted appropriate and inappropriate methods of disclosing feelings. The cognitive self-modeling unit provided opportunities to practice the skills learned in the preceding sections. However, only by transferring the skills learned here to your everyday life will make this learning experience meaningful. (FADE OUT/FADE BACK IN ON THE NARRATOR) 212 NARRATOR: You have now completed this phase of the program. Please open the door and tell the assistant you are finished. The assistant will stop the videotape machine and then give you specific instructions regarding the next phase. VOICE IN BACKGROUND: "What about Denny and Dianne?" NARRATOR: Oh yes. Well, they still have their up's and down's, but both are expressing their feelings to one another in mean- ingful ways, both verbally (CAMERA PANS IN ON DENNY & DIANNE KISSING) NARRATOR: --and nonverbally! APPENDIX F ANXIETY SCALE INSTRUCTIONS: anxiety scale actually feel APPENDIX F For each situation, best male or female friends. EXANPLE SITUATION l 2 very calm calm and relaxed and relaxed SITUATION ONE 1 2 very calm calm and relaxed and relaxed SITUATION TWO 1 very calm 2 calm and relaxed and relaxed SITUATION THREE 1 very calm 2 calm and relaxed and relaxed SITUATION FOUR 1 very calm 2 calm and relaxed and relaxed 3 somewhat calm 3 somewhat calm 3 somewhat calm 3 somewhat calm 3 somewhat calm ANXIETY SCALE A somewhat nervous somewhat nervous h somewhat nervous h somewhat nervous somewhat nervous 213 Fl 5 nervous fl nervous 5 DGPVOUS K nervous C I nervous 6 very nervous 6 very nervous 6 very nervous 6 very nervous 6 very nervous circle the number on the Wthh best describes how nervous you would if you were saying your response to one of your 7 terrified 7 terrified terrified terrified 7 terrified 1" II I 1‘1. SITUATION FIVE 1 2 very calm calm and relaxed and relaxed SITUATION SIX 1 2 very calm calm and relaxed and relaxed SITUATION SEVEN 1 2 very calm calm and relaxed and relaxed SITUATION EIGHT 1 2 very calm calm and relaxed and relaxed SITUATION NINE 1 2 very calm calm and relaxed and relaxed SITUATION TEN 1 2 very calm calm and relaxed and relaxed SITUATION ELEVEN 1 2 very calm calm and relaxed and relaxed 3 somewhat calm 3 somewhat calm 3 somewhat calm somewhat calm 3 somewhat calm 3 somewhat calm 3 somewhat calm 214 A somewhat nervous somewhat nervous A somewhat nervous 4 somewhat nervous somewhat nervous somewhat nervous somewhat nervous q nervous S nervous S nervous S nervous g nervous S nervous S nervous 6 very nervous 6 very nervous 6 very nervous 6 very nervous 6 very nervous 6 very nervous 6 very nervous terrified terrified terrified terrified terrified terrified terrified 215 SITUATION TWELVE 1 very calm calm somewhat somewhat and relaxed and calm nervous relaxed SITUATION THIRTEEN 2 very calm calm somewhat somewhat and relaxed and calm nervous relaxed SITUATION FOURTEEN 2 3 very calm calm somewhat somewhat and relaxed and calm nervous relaxed SITUATION FIFTEEN 1 3 very calm calm somewhat somewhat and relaxed and calm nervous relaxed SITUATION SIXTEEN 1 2 very calm calm somgwhat somgwhat and relaxed and calm nervous relaxed S nervous S nervous S DGPVOIIS g nervous nerVOus 6 very nervous 6 very nervous 6 very nervous 6 very nervous 6 very nervous terrified terrified terrified terrified terrified APPENDIX G AUDIO PERFORMANCE TEST '217 Remember: 1. Listen to the description of the situation given on recorder #1. 2. Record your response on Recorder #2. 3. Rate your nervousness on the anxiety scale paper. 4. Make sure you respond to all sixteen situations and record your anxiety level for each. If you have questions, call an assistant in now. There will be a pause on the tape for you to do thlS. (lo-second pause) Please do not adjust either tape recorder during this presentation. If a technical problem occurs, go to the door and ask an assistant for help. (pause) Now let's try an example situation. Remember to make your response immediately following the example and then rate your anxiety level on the scale in front of you. Example Situation: Your best male friend has borrowed your notes for an important class. He told you he would return them two days ago, but he didn't. You have a test in this class tomorrow and desperately need your notes back to study for the exam. Finally, you find him in his room. You go up to him and say . . . (15-second pause) If you have any questions, please ask the assistant now, The first situation will be presented in 15 seconds. (15-second pause) Situation 1: You're giving a party, and the behavior of one of your best female friends has become increasingly objectionable and obnoxious over the course of the evening. You're upset with her behavior, so you go over to her and say . . . (15-second pause) Situation 2: A female friend is taking a carpentry course and has really struggled with it. Secretly, you think it's kind of a crazy thing for her to do, but for your birthday she brings you a really sharp coffee table that she made. After she gives you the gift,you say . . . (15-second pause) Situation 3: You are shopping with your best female friend. A clerk accidentally short changes you and runs half way across the mall to return your 20¢. Since this has been a rotten day for you, the clerk's concern almost restores your faith in people. You turn to your friend and say . . . (15-second pause) 218 Situation 4: Once again, you're having dinner with your best female friend. Oftentimes when you're talking, she taps her water glass with her fingers. This habit of hers is annoying, and you're even wondering if she's listening to what you're saying. You turn to her and say . . . (15—second pause) Situation 5: Your best male friend has finally convinced you to go to a play with him. Much to your surprise as the play ends, you find that you've enjoyed it immensely. You turn to him and say . . . (15-second pause) Situation 6: Your best friend works next to you at your summer assembly line job. He has been very lax the last few days, and both of your production quotas have fallen behind because of it. The supervisor stops you after work and warns you that you are replaceable. You then go to your friend and say . . . (15-second pause) Situation 7: You have a big evening planned at your apartment, and your roommate is going away. After a long day, you come home to find the apartment spotlessly clean, and your grocery shopping done. You go up to your room- mate and say . . . (15-second pause) Situation 8: A tornado watch has been announced, and you and your best friends are crowded together in the basement of your hall. They are all laughing and joking, but you have a real phobia about storms. Your closest male friend is sitting next to you. You turn to him and say . . . (15-second pause) For the next eight situations you will be responding to some- thing said by another person. I shall describe a situ- ation, and then a male or female voice will talk. You are to imagine that the voices are those of your best male or female friends and that you are actually in the situation and talking to them. Situation 9: You have worked your ass off preparing a really exotic dinner for a special date. Your date calls and says: "Gee, I'm really sorry, but I can't come over for dinner tonight cuz I have a really bad cold." Crushed, you respond to her by saying . . . (15-second pause) Situation 10: You are a volunteer in a program for retarded children. Your best female friend comes to the center one night for an Open House and sees how well you work with the children. As you are driving home together she says, "I'm really impressed by your concern and caring for the children. It's really neat that you're so sensi- tive to their needs." You say to her . . . (15-second pause) 219 Situation 11: You took your best female friend with you to vISit some old friends from home whom she didn't know. You thought the afternoon was pleasant and that everyone got along well. On the way home, however, your girlfriend is extraordinarily quiet and refuses to respond to your attempts at conversation. Finally, she says: "I wish you hadn't treated me like that back there." Unsure of what she means, you say . . . (15-second pause) Situation 12: Your best female friend's parents have just filed for dIVorce. Your friend is really upset and feels torn between them. You and she talk several hours, and you assure her that she can still love both of them. She's visibly relieved and says, "Thank you so much. I really needed to talk to someone, and you've helped a lot." You reply . . . (15-second pause) Situation 13: You're camping with your best male friend. It's night, and you're sitting together by the fire watch- ing the stars. He turns to you and says . . . "Hey! Isn't this fantastic?! Thanks for asking me to come. Listen, I think you're one hell of a guy—-I really care for you a lot." You respond by saying . . . (15-second pause) Situation 14: Your summer job involves working in teams of two. Your male partner, who has been on the job several years, says to you one morning: "I used to hate this job, but you've made it not only bearable but fun. I'm really happy you're my partner." You respond by saying . . . (15-second pause) Situation 15: Your roommate repeatedly borrows your razor without asking. This weekend, he took it along when he left campus. He returns Sunday evening and blithely says: "Thanks for letting me use your razor." Stroking the stubble on your face, you reply . . . (15-second pause) Situation 16: It's the worst day of your life. When you get home even your dog rejects you. Looking bright and cheerful, your roommate comes in and says: "Hey man, how's it going?" and you say . . . (15-second pause) You have now completed this phase of the program. Please go to the door and ask the assistant to stop both tape recorders and give you instructions for the next phase. APPENDIX H PERFORMANCE TEST MEASURES: AMOUNT AND QUALITY RATING CRITERIA APPENDIX H PERFORMANCE TEST MEASURES: AMOUNT AND QUALITY RATING CRITERIA INSTRUCTIONS FOR RATING AMOUNT OF AFFECT Amount of Affect. l--Affect Present. II A subject's verbal response to a stimulus situation is either classified as affect being present or absent. In order for a unit within a response to be classified as containing affect, the following criteria must be met: a. An affect word given in the Guide List of Affect Words (Crowley, 1970; Nye, 1971) or suggested by defi- nitions of affect (English & English, 1958; Goldenson, 1970) must be present, and b. The affect word must be used as a verb, adverb, or adjective. Any one unit of the subject's total response to a stimu- lus situation which meets these criteria is sufficient for rating the entire statement as having affect present. Examples: I feel sad. You're being obnoxious. She's lovely. 0--Affect Absent. Examples: No one unit of the subject's total response to a stimulus statement meets the criteria specified for affect present. - I'm an optimist. (Affect word used as noun.) I feel like getting drunk. (Cognitive statements I flunked that exam. which express thought or fact rather than feelings and emotions.) 220 221 INSTRUCTIONS FOR RATING QUALITY OF AFFECT A response to a stimulus situation is composed of one or more units. Each unit which has an affec- tive component will be scored for: I. Reference II. Appropriateness of Affect III. Time Orientation IV. Reason V. Specificity of Reason _ Scores for each affective unit within a response will be averaged to create one score for that response. A total of these reSponse scores will be computed for each subject. I. REFERENCE. The person or situation identified as being responsible for the affective com- ponent of the statement. 12 pts.--Self—Active. The nominative or possessive case self-reference singular personal pro- noun used as subject or modifier of subject (I, my) must explicitly occur in the state- ment. Examples: I feel angry. My feelings are hurt. 4 pts.--Se1f—Passive. The objective or possessive case self-reference singular personal pro- noun used as recipient of affect or modifier of the recipient (me, my) must explicitly occur in the statement. The subject of the statement can only refer to the general situation and not to a particular person. It must also be clear that the affect refers to the self and not to the situation. Examples: This confusion upsets mg. Getting a good grade makes my day happier. 222 3 pts.--Implicit Self-Reference. The recipient of the affect is implied but not explicitly stated. Reference is understood from the context of the statement. The subject can only be the general situation, not a specified person. Examples: It feels good (to me) when we talk like this. This really hurts (me). When the affect definitely refers to the situation and not to the person, score zero (0) for Reference. Example: That was really excellent. 2 pts.--Plural Active-Passive. The nominative, objective, or possessive case self—reference plural personal pronoun used as subject recipient, or modifier of either (we, us, our) must explicitly occur in the statement or be evident from a compound usage (e.g., You and I). When used as recipient or recipient modifier, the subject of the state— ment can only refer to the general situation and not to a particular person. Examples: We are really happy together. Arguing makes BE angry. Drinking makes 93; life miserable. 925 frustration increases daily. She and I are happy. (or You and I, He and I) This hurt you and me. (or her and me, him and me) 223 1 pt. --General Active-Passive. General terms which may include the speaker are used to define reference, and the individual participates only by implication. The term can be the subject. If used as recipient, the subject of the statement can only refer to the general situation and not to a particular person. Examples: People (including me) are sad at times like this. One (me also) cannot always be happy. Rejection hurts anyone (me too). O pts.--Other Reference. Another person is made responsIBle for the affect. A self-referenced pronoun may be used to indicate the recipient, but the subject of the affective component of the statement is identified as another person or another person's actions. Examples: You make me happy. Professor Jones excites me when I attend his classes. Your calling me a dummy really hurts. This/That's a good table. II. TIME ORIENTATION. The verb tense of the affective component of the statement. 6 pts.--Present. The action occurs in the present time and is expressed by the present tense of the verb. Present tense linking verbs (is, am, are, become) are included. Examples: I am happy. He makes me mad. I feel good. I am feeling sad. 3 pts.--Present Perfect. The action or linkage occurred in the past but continues to the present. A past tense verb or participle form is used with a present tense helping verb (have, has) III. 1 pt. 1 pt. 224 Examples: I have been happy. I have been feeling bad. He has made me mad. I have felt great lately. Past and Past Perfect. The action or linkage in the past and is completed. The helping verb hag_may be included with the past par- ticiple or past tense verb. Examples: I was happy. He made me mad. I felt bad. I had felt good lately. Future and Future Perfect. The action or linkage wilI occur in the future. The helping verb have may be included with will/ shall and as participle or past tense form of the verb. Examples: I shall be happy. He will feel good. He will make me mad. I'll havefibeen happy a long time by then. APPROPRIATENESS OF AFFECT. The emotion expressed is 8 pts. 0 pts. consistent with the affect presented in the stimulus situation. YES--According to the Guide List of Affect Words (Crowley, 1970; Nye, 1971) or words suggested by definitions of affect (English & English, 1958; Goldenson, 1970) the affect expressed falls into the same general cate- gory (love, fear, anger) as the stimulus situations. Example-stimulus is classified as presenting Love and affect is classified as representing Love. NO--According to the Guide List of Affect Words (Crowley, 1970; Nye, 1971) or words suggested by the definitions of affect (English & English, 1958; Goldenson, 1970) the affect expressed does not fall into the same general category (love, fear, anger) as the stimulus situations. Example-stimulus is classified as presenting a Negative feeling, and affect is classified as representing a Positive emotion. IV. REASON. 5 pts. 0 pts. 225 a rationale for the affect. YES-A cognitive statement, demonstrative pronoun (this, that, these, those) or other reference to the stimulus situation which notes a reason for the affect expressed is explicitly stated as either a part of the unit which includes the affective component or as a separate statement in the unit which immediately precedes or follows the affective component. Note: .In some cases a second affect unit may be the reason for the first affect unit. In these cases, score the second affect unit as the reason. Then score the second affect unit for the five quality measures. Examples: Cognitive statement within a unit. My feelings are hurt because you don't want to be with me right now. Cognitive statement in preceding/following unit. My feelings are hurt/I don't think you want to be with me right now. Demonstrative pronoun Within unit. That hurts my feelings/ I'm really pleased with you because of that. Preceding/following unit That does it/I'm frustrated with you W Other stimulus reference Your saying what you did hurts my feel- ings7 Our discussion exhilarates me/ I'm anxious about Professor Jones' class/ Since our fight, ITve been doing some thinking/I'm really still confused, however/ NO-No rationale for the affective component is explicitly stated within the unit,and the rationale is not in the unit which immedi- ately precedes or follows the affect unit. Examples: /I'm still really confused/ /My feelings are hurt/ ( REASON) V. 226 /You hurt my feelings/ /I'm excited today/ /I'm upset/I'm depressed/You stood me up/ affect 1 affect 2 reason (here there is no reason for affect 1) SPECIFICITY OF REASON. The degree to which the spe- 8 pts. 5 pts. cific component(s) of the stimulus situation which evoked the affect is explicitly ver- balized in the response. Specific Component-One or more of the spe- cific components of the stimulus situation are identified as a reason for the affect. Example Stimuli situations: a. You have just flunked a midterm exam in a required course. It's worth 50% of your final grade and you're sure you're going to fail the class. Your best friend comes up and asks how you're doing and you say . . . b. You are saying goodnight to a friend after a pleasant evening at the movies and drinking at a bar. You say . . . Examples using the stimulus situations: a. I'm depressed because I_just failed an. exam. b. Talking with you was really fun/ I like getting drunk with you. General Class-The entire stimulus situation or the class of situations of which it is a part is identified as the reason for the affect. Examples using the stimulus situations: a. I'm feeling shitty about school. a. Taking tests gets me down. b. I like being with you. b. Drinking is fun to me. (SPECIFICITY) 2 pts. 1 pt. 0 pts. 227 Implied Class. The entire stimulus situation or the class of situations of which it is a part is identified by a demonstrative or indefinite pronoun (this, that, these, those, it) or a word which strongly denotes such a pronoun as the reason for the affect. Examples using the stimulus situations: a. lt_was really shitty. b. This was fun, wasn't it? Generalized Additional Class. The reason for the affect is identified as a different and much broader class of stimulus situations than that to which the stimu- lus situation belongs. Examples using the stimulus situations: a. Life is hell. b. I just love being with people. Absent Reason. No rationale for the affective component is explicitly stated within the unit, and the rationale is not in the unit which immediately precedes or follows the affect unit. Examples using the stimulus situations: a. I feel depressed. b. This was a good evening/ It's late/ I'm feeling pretty happy. . 228 Supplemental Criteria for Ratinngape Transcripts 5/30/75 1. DOES THE AFFECT WORD REFER TO THE SPEAKER'S FEELINGS OR DOES IT DESCRIBE, COGNITIVELY, ANOTHER PERSON, SITU- ATION, OR THING? a. Criteria-~Look at the affect word, ask yourself: 1) Whg feels this way? 2) Is this just a statement of fact, or does it express the person's feeling state at the time? Examples: 1) This is a nice table. NO. Affect word refers to table; this is a statement of fact 2) That's really great. NO. What is great? The person? No. What- ever that is is what's great. Another statement of fact. 3) That made me really scared. YES. Who is scared? The person; there- fore, rate this as an affect unit. 4) This was really shitty. NO. Shitty refers to whatever thi§_was. These criteria supersede contradictory criteria stated under implicit self-reference on the Quality criteria. In order to have implicit self-reference, the affect must still apply definitely to the person. REFERENCE--QUALITY a. Note criteria for a 4 rating-~if ME or MY appears explicitly in the statement. Examples: Drinking makes me happy. This hurts me. My feelings are hurt. Special case of Situation 16. Often people respond with an adjective (and often a swear word). Shitty rates as affect. Understood subject rates 0. Shit is a noun; therefore, it is not considered as affect. Do not rate. 229 3. SPECIFICITY OF REASON--QUALITY a. AMOUNT O‘U‘lnwaH O QUALITY Look carefully at the situation to which it refers; to rate a 5 it would refer to the situation in gen- eral. To Fate an S, it must pick out a particular part of the situation. To rate a 2, it will use a pronoun to refer to reason. In all cases on specificity as well as reason-yes/no, remember to check the unit immediately predecing the affect unit and the unit immediately following it. Remember that one affect unit can serve as a reason for another affect unit if it makes sense to put because in front of the affect unit scored as reason. Scoring Check List Check for affect word in unit- Check affect unit for expression of feeling. Check if affect is verb, adverb, or adjective. Underline affect unit. Rate other units in response. Score response l and record as amount on tran" script if any unit receives ye§_on points 1, 2, and 3. Use only responses which have received a l for Amount. Score each unit in the response independently. Refer to Criteria for scoring, rating each under- lined affect unit for: Reference Time Orientation Appropriateness of Affect Reason Given Specificity of Reason Record scores on Rating Sheet for each unit. REFERENCES Crowley, T. J. The conditionability of positive and neg- ative self-reference emotional affect statements in a counseling-type interview. Unpublished doc- toral dissertation, University of Massachusetts, 1970. 230 English, H. B., & English, A. C. A comprehensive dic- tionary of psychological andgpsychoanalytical terms. London: Davids McKay, 1958. Goldenson, R. M. The engyclopedia of human behavior. New York: Doubleday, 1970. Nye, L. S. The development and implementation of two Ere-practicum training approaches and an evalu- atiOn of their effects upon counseling perfor- mance. Unpublished doctoral dissertation, Purdue University, 1971. APPENDIX I SKILL TEST APPENDIX I INSTRUCTIONS: Please read each item carefully, then choose the 1 -0 response you would make if you were in the situation. A special girl friend comes hack into town and calls you to go out. When you meet at the theatre, you want to tell her how you feel. You say ... A. l'm really excited that you called. B. Your call made my day. C. I was really excited that you called. D. l'm looking forward to seeing this movie with you. After you watch your best friend slave over a term paper all day, he asks you to read it. It's really well written and creative. He shyly asks, "What do you think?“ and you respond by saying ... A. Good paper! You're really creative! B. You certainly deserve a h.0 on this one! C. What a fantastic job! I'm really impressed! D. That's a really good paper! You have been working on your bicycle for three hours and finally have it repaired and ready for summer. A close female friend stops by and says, 'You really did a great job!" You say ... A. Thanks. I feel really good about it myself. B. Thanks. It was a real chore, but I think it's finally ready to go. C. Thanks. Let's go for a ride together soon. D. Thanks. I'm really pleased that you noticed my hard work. You are recovering from bronchial pneumonia and have been in bed for three weeks. Your best friend brings you a plant and visits a while. You say to him ... A. It's really good of you to come. B. Thank you. The plant is really nice. C. I'm happy you came. D. You're a good friend for coming to see me. 231 232 After studying three days and nights for the one exam that could make or break your whole college career, your professor comes up to you before the exam and says: "There has been a terrible mistake. You have a u.0 after all, and you don't have to take the exam." You say ... A. I'm relieved that I don't have to take the exam. B. I really studied hard for this exam, but I'll take a h.O any day! C. I wish you'd told me sooner. D. Thanks. I really needed the h.O! You have just been rejected for an important job you really wanted. After telling a good friend, she replies: "Friend, it's their loss. You're so talented I know something better will come." You say ... A. I feel better now that I've talked to you. B. You always make me feel better. C. Thanks. You're really a thoughtful person. D. Thanks. I can always count on you for cheering words. You're out with your favorite date. Your home team won a close game that you attended. You say to your date ... A. Our team really played a great game! B. We both got pretty excited, didn't we! C. I'm really excited that we won! D. Hey! Wasn't that great! You just got the results back from a blood test for mono, and found out that you don't have it. You run into a close male friend on your way home from the health center and say ... A. Hey! I'm feeling really good! B. I just got my blood test results back, and I don't have mono! ' C. I was real worried, but am feeling better now cuz I don't have mono. D. I'm ready to get back into action cuz I don't have mono. As a token of friendship, you give your best female friend a bottle of her favorite cologne. As you give her the present, you say ... A. I think good friends are hard to find, and I got you this to let you know that you're the greatest. B. I just want you to know how much I like you. C. You're very special to me. D. I thought you might like this. 10. ll. 12. 13. lb. 233 You're talking with a close male friend. He tells you that he just got a good paying part-time job. You respond to him by saying ... A. That's great! That extra money will sure come in handy, won't it? B. I'm really happy that the job came through for you! C. Hey! You must be really excited! D. Tell me all about it. You're walking through the woods with your best female friend. She turns to you and says: "You're the neatest person I've ever known. I hOpe we'll be friends forever." You reSpond by saying ... A. Thanks for Sharing your feelings with me. I hope we'll always be friends, too. B. I like you a lot, too. O. You're super! D. Having a close friend like you is the greatest gift anyone could have. You have just come from a class where the prof praised your term project in front of the entire class. When you run into your best male friend, you say ... A. I just got a u.O on my term project! B. [Nothing, since it's better to keep accomplishments to yourself.] C. I'm really proud of my term project! D. This is one of the best days of my life! You spent several hours picking out a birthday gift for a friend. He opens it and says, "Is this the only color it comes in?" You say ... A. You're disappointed that it's not the color you'd like. B. Don't you like the color? C. I'll take it back, then. D. I'm disappointed that you don't like the color. You best female friend enjoys cooking, but really lacks the knack. After a very stressful day, you once again find dinner burned. You say ... A. You know, cooking just doesn't seem to be your thing. B. You're no Julia Childs, are ya? C. I'm irritated that dinner's burned again. D. This has been a hard day, and I'm not really hungry. 15. 16. 17. 18. 19. 234 A good friend says to you, "You just don't have the brains to go to med school." You say to him ... A. I appreciate your honesty. B. It's not your opinion that counts. C. Bug off. D. I'm hurt that you don't think I can make it. During the course of a tense evening at your girlfriend's apartment, you spill an entire cup of coffee on her new couch. She leaps up and screams hysterically, "You stupid ass! You're hopelessly clumsy!" You say ... A. I'm upset that you're angry with me. B. It was really stupid of me to do that. C. I've been tense all evening and didn't mean to spill it. D. Here, I'll clean it up and pay for the cleaning bill. While waiting for an elevator during rush hour, a good female friend comes up to you, smiling, and says, "Your fly is open!" You reply ... A. Thanks for telling me. We all have our embarrassing moments, don't we? B. How stupid of me! I don't believe I did that! C. How embarrassing! Thanks for telling me. D. I'm really embarrassed. Thanks for telling me. As a big surprise, your roommate has painted the living room. You return from a day of classes and encounter four walls of a color you particularly dislike. You turn to your roommate and say ... A. [Nothingg maybe you'll get used to the color.] B. I'm upset that you painted the apartment without talking to me first. C. You really shouldn't have painted the walls without first talking to me. D. I know you've worked hard, but you really shouldn't have painted the walls without asking me first. You have borrowed a good pair of slacks from a friend to wear to a Special occasion. You take good care of them and return them promptly. Your friend notices a ragged hole in the cuff and says sarcastically, "Some friend you are! I can't even trust you with a pair of slacks!” You say ... A. Itxfirncyou're being unfair with me because the hole was there before I borrowed them. 20. 21. 22. 23. 235 B. I'm upset because the hole was there before I borrowed them. C. I really don't think I made the hole, but I'll see that the slacks are fixed. D. I think that hole was there before I borrowed them. What can I do to make things right? A state cop has pulled you over for a faulty brake light. You expect to be warned, but instead, he gives you a twenty dollar ticket. After he leaves, you say to a female friend who's with you ... A. I'm mad that I got that ticket. B. This just isn't my day! C. I just never seem to get a break! D. I'm so angry I could scream. Your best friend has borrowed your tape recorder and returned it with a note saying he broke it. He didn't offer to replace or fix it. You go to him and say ... A. You know, if you weren't my best friend, I'd be really hot that you busted my tape recorder! B. I thought I was a klutz, but you take first prize this time! C. Hey ! I'm really angry that you broke my tape recorder and didn't offer to pay for repairs! D. I don't believe you broke my tape recorder, and you didn't even offer to fix it! Your best female friend says, "You're a lousy driver! I never want to ride with you atain!" You say ... A. So who's asking you? You didn't have to come, you know. - B. I'm mad because you're knocking my driving! C. You're making me mad with this talk about my driving! D. I'm getting mad at you because of this! You've just finished typing a term paper for a close male friend. As he's proofreading it, he says, "God! You really made a lot of typos, didn't you?‘ You reSpond by saying ... A. I did the best I could. I'm upset that you're being so critical. B. Nothing; it's not worth the hassle. C. I'm not perfect, you know. I did the best I could. D. You really make me angry. 236 2h. You and a close female friend are walking by a river. Suddenly, a snake glides across your path. Startled, you turn to her and say ... A. That snake scared you, didn't it? B. I get cold chills whenever a snake startles me. C. I think snakes are ugly. D. Snakes give me the creeps. APPENDIX J ATTITUDE SURVEY APPENDIX J Tell how you feel about these statements by marking an opinion from STRONGLY AGREE to STRONGLY DISAGREE for each statement. It is important to be truthful when answering the statements. Please answer every question. Circle the correct letter. KEY a--strongly agree b--agree c--agree somewhat d--neutral e--disagree somewhat f--disagree g--disagree strongly a b c d e f g 1. When I tell someone how competent I feel, it's really like bragging. a b c d e f g 2. My good friends know how I feel without my telling them. a b c d e f g 3. People will think less of me if I tell them I've failed at a new task. a b c d e f g h. Nervous people make things worse by not telling others how they feel. a b c d e f g 5. If I'm afraid, I'll feel better by sharing my feelings with others. a b c d e f g 6. When I'm angry with others, it's best to tell them so. a b c d e f g 7. No one cares how I feel, when you get right f g 8. f 8 9. f g 10. f g 11. down to it. If I feel jealous, it will get me down unless I tell the person. When I'm discontented, just keeping busy will make the feeling go away. When I'm upset, it's better to work out my feelings alone. Although they pretend not to be, friends are leery of me after I talk about feeling worth- less. 237 17o 18. 19. 20. 21. 22. 23. 2h. 25. 26. 27. 28. 29. 238 It's Often best not to tell people I really like them. Proud feelings really shouldn't be shared. It's good to deal with feelings of failure myself. Expressing anger will get me in trouble. When I'm depressed, it's a good idea to handle it myself. When I love someone, I tell them. When I am attracted to a person, I should let them know how I feel. Others will think better of me if I will tell them when I feel helpless. It's best to express intense emotions rather then keep them inside. Expressing feelings of inadequacy makes others think less of me. Even if I think I'm really handsome, I shouldn't bore my friends by telling them 300 If I'm attracted to someone I just met, I'll keep these feelings to myself. Sharing feelings is one of the best ways to get close to people. I should be able to deal with most of my feelings myself. Feelings should be shared with counselors instead of friends. I'm more comfortable keeping my feelings to myself. PeOple are more comfortable when I don't tell them how I feel. When I like a person, it's good to tell him even if I don't know how he feels about me. 239 a b c d e f g 30. It's a good idea for me to express feelings verbally. APPENDIX K DEBRIEFING APPENDIX K DEBRIEFING Please answer the following questions to help us in plan— ning future training sessions. 1. Overall, what do you feel was of most value in the training? 2. Did you learn anything new about yourself? If so, what? 3. How do you feel you've benefited from the experience? 4. In what ways could it have been more helpful? REGARDING THE VIDEOTAPE--Please leave blank if you haven't had the videotape yet. 5. Was the instructional portion clear? How could it have been better? 6. Did the modeling examples help? Why or why not? REGARDING THE SELF-MANAGEMENT PROGRAM--Please leave blank if you haven't had self-management yet. 7. Did you actually d2 the project? 8. Did you regularly meet your goals? 9. Did you find it easier to express your feelings to males or to females? Why was this do you think? 10. Is it easier to respond to another person with feeling statements or to initiate feeling statements? Why? 11. Are positive feelings easier to express than negative feelings? 12. Please give any additional comments here. 13. I am ___ years old. 14. I am a freshman sophomore _ junior senior 240 APPENDIX L INTERCORRELATIONS FOR DEPENDENT MEASURES suaamso scmz use .mpspfluud .mumflxcm .aaflxm .muflamso luc308< wow whoaumamunooweucH oo.a ma. eo.u NH. am. as. Susanna cams G oo.a mm.u am. so. so. wannabes m oo.a mm. mm. em. sumaxqa e oo.a oe. oe. Haaxm m oo.a mm. suaamso m oo.a Dances a spmwwmo mesumuua sumaxca Hawxm suawmso pawoes Q s mashed: muflamso use: Hmucmfimammsm may use .mpsufluu¢ .mueflxcd .HHme .wwflamso .uc5054 How xfiwumz GOHDMHOHHOU umeuumom sunflpOeEH Hm magma mMMDmdmz Bzmazmmmo mom mZOHfidqmmmOUmMBZH A xHazmmmd 241 REFERENCES REFERENCES Altman, I., & Taylor, D. A. Social penetration: The development of interpersonal reIationships. New York: Holt, Rifiehart, & Winston, 1973. Auerswald, M. The differential reinforcing power of two counseling technigues, restatement and interpre- tation, on client productiEn of affect. Unpub- lished doctoral dissertatiOn, Purdue University, 1972. Auld, F., & White, A. M. Rules for dividing interviews into sentences. The Journal of Psychology, 1956, 52, 273-281. Axtell, B., & Cole, C. W. Repression-sensitization response mode and verbal avoidance. Journal of Personality and Social Psychology, 1971, lg, 133-137. Baer, D. M., Peterson, R. F., & Sherman, J. A. The development of imitation by reinforcing behavioral similarity to a model. Journal of the Experimental Analysis of Behavior, 1967, l9, 465-416. Baker, J. N. Reason versus reinforcement in behavior modification. (Doctoral dissertation, Uni- versity of Illinois, 1966). Dissertation Abstracts, 1966, 21, 2509-2510B. (University Microfilms No. 66-12, 286) Bale, Jr., R. N. The effect of discrimination training, guided practice, and role-playing in modeling treatments designed to increase bi-directional verbal behavior in socially withdrawn males. UnpubliShed doctoral dissertation, Stanford University, 1972. Bandura, A. Influence of models' reinforcement contin— gencies on the acquisition of imitative responses. Journal of Personality and Social Psychology, 1965, l, 589-595. 242 243 Bandura, A. Principles of behavior modification. New York: Holt, Rinehart, & Winston, 1969. Bandura, A. Modeling theory. In W. S. Sahakian (Ed.), Psychology of learnigg. Chicago: Markham, 1970. Bandura, A. Psychotherapy based upon modeling principles. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change. New York: Wiley, 1971. (aifi Bandura, A. Social learning theory. Morristown, N.J.: General Learning Press, 1971. (b) Bandura, A., Blanchard, E. B., & Ritter, B. Relative efficacy of desensitization and modeling approaches for inducing behavioral, affective, and attitudinal changes. Journal of Personality and Social Psychology, 1969, 13, 173-199. Bandura, A., Grusec, J. E., & Menlove, F. L. Obser- vational learning as a function of symbolization and incentive set. Child Develgpment, 1966, 31, 499-506. Bandura, A., & Kupers, C. J. The transmission of patterns of self-reinforcement through modeling. Journal of Abnormal and Social Psychology, 1964, 52, 1-9. Bandura, A., & Menlove, F. L. Factors determining vicarious extinction of avoidance behavior through symbolic modeling. Journal of Personality and Social Psychology, 1968, 8, 99-108. Bandura, A., & Mischel, W. Modification of self-imposed delay of reward through exposure to live and sym- bolic models. Journal of Personality and Social Psycholggy, 19657f2, 698-705. Bandura, A., Ross, D., & Ross, S. A. Imitation of film- mediated aggressive models. Journal of Abnormal and Social Psychology, 1963, fig, 3-11. Barnabei, F. Determining the effects of three counselor verbal responses on client verbal behavior. Paper presented at the Annual Convention of the North Central Association for Counselor Edu- cation and Supervision, Chicago, 1972. 244 Berger, S. M. Incidental learning through vicarious reinforcement. Psychological Reports, 1961, 2, 477-491. Brody, H. A. The effect of three modeling procedures on the frequency of emission of self-referent affect statements. (Doctoral dissertation, Columbia Uni- versity, 1968). Dissertation Abstracts, 1968, 22, 767B. (University Microfilms No. 68-11, 706) Burkhead, D. E. The reduction of negative affect in human subjects: A laboratory investigation of rational- emotive psychotherapy. (Doctoral dissertation, Western Michigan University, 1970). Dissertation Abstracts International, 1970, 31, 2097-20983. (University Microfilms No. 70-2IT 858) Burnstein, E., Stotland, E., & Zander, A. Similarity to a model and self-evaluation. Journal of Abnormal Psychology, 1961, 22, 257—264. Campbell, D. T., & Stanley, J. C. Experimental and quasi-experimental designs for research. Chicago: Rand McNally, 1963. Cautela, J. R., Flannery, R. B., & Hanley, S. Covert modeling: An experimental test. Behavior TherapY: 1974, 1, 494-502. Certner, B. C. Exchange of self-disclosures in same- sexed groups of strangers. Journal of Consult- ing and Clinical Psyghology, 1973, 52, 292-297. Cornfield, J., & Tukey, J. W. Average values of mean squares in factorials. Annals of Mathematical Statistics, 1956, 21, 907-949? Cozby, P. C. Self-disclosure: A literature review. Psychological Bulletin, 1973, 12, 73-91. Croner, M. D., & Willis, R. H. Perceived differences in task competence and assymetry of dyadic influence. Journal of Abnormal and Social Psychology, 1961, 22, 705-708. Crowley, T. J. The conditionability of positive and negr ative self-reference emotiOnal affect statements in a counseling-type interview. Unpublished doctoral diSSertation, University of Massachu- setts, 1970. Debus, R. L. Effects of brief observation of model behavior on conceptual tempo of impulsive chil- dren. Developmental Psychology, 1970, 2, 22-32. 245 DiLoreto, A. 0. Comparative psychotherapy: An experi- mental analySiS. Chicago: Aldine-Atherton, 1971. Dimond, R. E., & Hellkamp, D. T. Race, sex, ordinal position of birth and self~disclosure in high school students. Psychological Reports, 1969, 22, 235—238. D'Zurilla, T. J., Wilson, G. T., & Nelson, R. A prelimi- nary study of the effectiveness of graduated prolonged exposure in the treatment of irrational fear. Behavior Therapy, 1973, 1, 672-685. Ehrlich, H. J., & Graeven, D. B. Reciprocal self- disclosure in a dyad. Journal of Experimental Social Psychology, 1971, 1, 389-400. Ellis, A. Reason and emotion in psychotherapy. New York: Lyle-Stuart, 1962. Ellis, A. The essence of rational psychotherapy: A comprehensive approach to treatment. New York: Institute for Rational Living, 1970. Finn, J. D. Multivariance-univariate and multivariate analysis of variance and covariance: A fortran IV program, 1968. Friedman, P. H. The effects of modeling and role playing on assertive behavior. (Doctoral dissertation, The University of Wisconsin, 1968). Dissertation Abstracts, 1968, 22, 4844-4845B. (University Microfilms No. 69-912) Geer, J. H., & Turtletaub, A. Fear reduction following observation of a model. Journal of Personality and Social Psychology, 1967, 2, 327-331. Gelfand, D. M. The influence of self-esteem on rate of verbal conditioning and social matching behavior. Journal of Abnormal and Social Psychology, 1962, 221 259—265. Gerst, M. S. Symbolic coding processes in observational learning. Journal of Personality and Social Psychology, 1971, 11, 7-17. Goldfried, M. R., Decenteceo, E. T., & Weinberg, L. Systematic rational restructuring as a self- control technique. Behavior Therapy, 1974, 2, 247-254. 246 Green, A. H. The effect of various forms of modeling and instructional procedures upon the modification of self-disclosing verbal behavior. (Doctoral dissertation, The University of Wisconsin, 1972). Dissertation Abstracts International, 1972, 22, 3938B. (University Microfilms No. 72-31, 532) (a) Green, A. H., & Marlatt, G. A. Effects of instructions and modeling upon affective and descriptive ver- balization. Journal of Abnormal Psychology, 1972, 29, 189-196. (b) Guilford, J. P. Psychometric methods. New York: McGraw Hill, 1954. Hansen, J. C., Niland, T. M., & Zani, L. P. Model reinforcement in group counseling with elemen- tary school children. Personnel and Guidance Journal, 1969, 11, 741-744. Hicks, D. J. Imitation and retention of film-mediated aggressive peer and adult models. Journal of Personality and Social Psychology, 1965, 2, 97-100. Himelstein, P., & Kimbrough, Jr., W. W. A study of self- disclosure in the classroom. Journal of Psy- chology, 1963, 11, 437-440. Himelstein, P., & Lubin, B. Attempted validation of the Self-Disclosure Inventory by the peer nomination technique. Journal of Psychology, 1965, 61, 13-16. Hurley, J. R., & Hurley, S. J. Toward authenticity in measuring self-disclosure. Journal of Counseling Jakubczak, L. F., & Walters, R. H. Suggestibility as dependency behavior. Journal of Abnormal and Social Psychology, 1959, §g, 102-107. Jennrich, R. I. Wisconsin 1604 ANOVA (COOP ID, 64/Wisc/ ANOVA). Unpublished manuscript, University of Wisconsin, 1961. Jourard, S. M. I-thou relationship versus manipulation in counseling and psychotherapy. Journal of Individual Psychology, 1959, 12, 174-179. 247 Jourard, S. M., & Lasakow, P. Some factors in self— disclosure. Journal of Abnormal and Social Psychology, 1958, 22, 91-98. Kanareff, V. T., & Lanzetta, J. T. The acquisition of imitative and opposition responses under two con- ditions of instruction-induced set. Journal of Experimental PsycholoQY: 1958, 22, 516-528. Kanareff, V. T., & Lanzetta, J. T. Effects of task defi- nition and probability of reinforcement upon the acquisition and extinction of imitative responses. Journal of Experimental Psychology, 1960, 22, 340-348. Kanfer, F. H., & Marston, A. R. Human reinforcement: Vicarious and direct. Journal of Experimental Psychology, 1963, 22, 292-296. Kanfer, F. H., & Phillips, J. S. Learning foundations of behavior therapy. New York: Wiley, 1970. Kaplan, M. F. Interview interaction of repressors and sensitizers. Journal of Consulting Psychology, 1967, 21, 513-516. Karst, T. O., & Trexler, L. D. Initial study using fixed-role and rational-emotive therapy in treat- ing public-speaking anxiety. Journal of Consult- ipg and Clinical Psychology, 1970, 2, 360-366. Kazdin, A. E. Covert modeling and the reduction of avoidance behavior. Journal of Abnormal Psy- chology, 1973, 21, 87-95. Ta) Kazdin, A. E. The effect of suggestion and pretesting on avoidance reduction in fearful subjects. Journal of Behavior Therapy and Experimental Psychiatry, 1973, 1, 213-221. (b) Kazdin, A. E. Covert modeling, model similarity, and reduction of avoidance behavior. Behavior Therapy, 1974, 2, 325-340. (a) Kazdin, A. E. The effect of model identity and fear- . relevant similarity on covert modeling. Behav10r Therapy, 1974, 2, 624-635. (b) 248 Lovaas, O. I., Freitag, L., Nelson, K., & Whalen, C. The establishment of imitation and its use for the development of complex behavior in schizo- phrenic children. Behavior Research and Therapy, 1967, 2, 171-181. Lubin, B., and Harrison, R. L. Predicting small group behavior with the Self-Disclosure Inventory. Psychological Reports, 1964, 12, 77-78. Lynn, D. B. Divergent feedback and sex—role identifi- cation in boys and men. Merrill-Palmer Quarterly, 1964, 12, 17-23. Mahoney, M. J. Cognition and behavior modification. Cambridge, Ma.: Ballinger, 1974. Marlatt, G. A. Vicarious and direct reinforcement control of verbal behavior in an interview setting. (Doctoral dissertation, Indiana University, 1968). Dissertation Abstracts, 1968, 22, 1845-18468. (University Microfilms No. 68-15, 451) Marlatt, G. A. A comparison of vicarious and direct reinforcement control of verbal behavior in an interview setting. Journal of Personality and Social Psychology, 1970, 12, 695-703. (a) Marlatt, G. A. Exposure to a model and task ambiguity as determinants of verbal behavior in an inter- view. Journal of Consulting and Clinical Psy- chology, 1270, 21, 268-276. (b) Marlatt, G. A., Jacobson, E. A., Johnson, D. L., & Morrice, D. J. Effects of exposure to a model receiving evaluative feedback upon subsequent behavior in an interview. Journal of Consulting and Clinical Psychology, 1970, 22, 104-112. Masters, J. C., & Morris, R. I. Effects of contingent and noncontingent reinforcement upon generalized imitation. Child Development, 1971, 22, 385-397. Mehrens, W. A., & Lehmann, I. J. Measurement and evalu- ation in education and psychology. New York:y Holt, Rinehart, & Winston, 1973. Meichenbaum, D. H. Examination of model characteristics in reducing avoidance behavior. Journal of Personality and Social Psychology, 1971, 11, 298-307. 249 Meichenbaum, D. Cognitive modification of test anxious college students. Journal of Consulting and Clinical Psychology, 1972, 22, 370-380. Meichenbaum, D., & Cameron, R. Training schizophrenics to talk to themselves: A means of developing attentional controls. Behavior Therapy, 1973, 2, 515-534. Meichenbaum, D. H., Gilmore, J. B., & Fedoravicius, A. Group insight versus group desensitization in treating speech anxiety. Journal of Consulting and Clinical Psychology, 1971, 227 410-421. Meichenbaum, D., & Goodman, J. Reflection-impulsivity and verbal control of motor behavior. Child Development, 1969, 22, 785-797. Meichenbaum, D., & Goodman, J. Training impulsive chil- dren to talk to themselves: A means of developing self-control. Journal of Abnormal Psychology, 1971, 11, 115-126. Merbaum, M., & Lukens, Jr., H. C. Effects of instruc- tions, elicitations, and reinforcments in the manipulation of affective verbal behavior. Journal of Abnormal Psychology, 1968, 73, 376-380. Mischel, W., & Grusec, J. Determinants of the rehearsal and transmission of neutral aversive behaviors. Journal of Personality and Social Psychology, 1966, 2, 197-205. Montgomery, A. G. Comparison of the effectiveness of systematic desensitization, rational-emotive therapY: implosive therapy, and no therapy in reducing test anxiety in college students. (Doc- toral dissertation, Washington University, 1971). Dissertation Abstracts International, 1971, 32, 1861A. (University Microfilms No. 71-27, 337T Moustakas, C. Honesty, idiocy and manipulation. Journal of Humanistic Psychology, 1962, 2, 1-15. O'Connor, R. D. Modification of social withdrawal through symbolic modeling. Journal of Applied Behavior Analysis, 1969, 2, 15-22. Orne, M. T. Demand characteristics and the concept of quasi-controls. In R. Rosenthal & R. L. Rosnow (Eds. ), Artifacts in behavioral research. New York: Academic Press, 1969. 250 Pearce, W. B., & Sharp, S. M. Self—disclosing communi- cation. The Journal of Communication, 1973, 22, 409-425. Pederson, D. M., & Breglio, V. J. Personality correlates of actual self-disclosure. Psychological Reports, 1968, 22, 495-501. Pederson, D. M., & Higbee, K. L. Personality correlates of self-disclosure. Journal of Social Psychology, 1969, 12, 81-89. Polansky, N. A., & Brown, S. Q. Verbal accessibility and fusion fantasy in a mountain county. American Journal of Orthopsychiatry, 1967, 21, 651-660. Rickers-Ovsiankina, M. A., & Kusmin, A. A. Individual differences in social accessibility. Psycho- logical Reports, 1958, 2, 391-406. Rimm, D. C., & Mahoney, M. J. The application of rein- forcement and participant modeling procedures in the treatment of snake-phobic behavior. Behavior Research and Therngr 1969, 1, 369-376. Rimm, D. C., & Masters, J. C. Behavior therapy: Tech- niques and empirical findings. New York: Academic Press, 1974. Sarason, I. G. Test anxiety and cognitive modeling. Journal of Personality and Social Psychology, 1973, E, 58-61. Sarason, I. G., Ganzer, V. J., & Singer, M. Effects of modeled self-disclosure on the verbal behavior of persons differing in defensiveness. Journal of Consulting and Clinical Psychology, 1972, 22, 483-490. Schwartz, A. N., & Hawkins, H. L. Patient models and affect statements in group therapy. In 229- ceedings of the 73rd Annual Convention of the American Psychological Association. Washington, D.C.: American PsychologiCal Association, 1965. Stuhr, D. E. The effects of social model characteristics in elicitingipersonal feeling questions. Unpub- lished dOctoral dissertation, Stanford Uni- versity, 1972. 251 Trexler, L. D., & Karst, T. O. Rational-Emotive therapy, placebo, and no-treatment effects on public- speaking anxiety. Journal of Abnormal Psy- chology, 1972, 12, 60-67. Truax, C. B., Wargo, D. G., Carkhuff, R. R., Kodman, F., & Moles, E. A. Changes in self-concepts during group psychotherapy as a function of alternate sessions and vicarious therapy pretraining in institutionalized mental patients and juvenile delinquents. Journal of Consulting and Clinical Psychology, 1966, 22, 309-314. Voight, N. L. Effects of self-management strategies as initial training and as secondary training for affective self-disclosure of undergraduate males. Unpublished doctoral dissertatiBn, Michigan State University, 1975. Vondracek, F. W. Behavioral measurement of self- disclosure. Psycholggical Reports, 1969, 22, 914. (a) Vondracek, F. W. The study of self-disclosure in experi- mental interviews. Journal of Psychology, 1969, E, 55"590 (b) Vondracek, F. W., & Marshall, M. J. Self-disclosure and interpersonal trust: An exploratory study. Psychological Reports, 1971, 22, 235-240. Walters, R. H., & Parke, R. D. Influence of response consequences to a social model on resistance to deviation. Journal of Experimental Child Psy- chology, l964,1,269-280. Whalen, C. Effects of a model and instructions on group verbal behaviors. Journal of Consulting and Clinical Psychology, 1969, 22, 509-521. Wilder, S. N. The effect of verbal modeling and verbal reinforcement on the frequency of self-referred affect statements. (Doctoral dissertation, Columbia University, 1967). Dissertation Abstracts, 1968, 22, 4304-4305B. (UniVersity Microfilms No. 68-5632) 252 Wilkens, W. Desensitization: Social and cognitive factors underlying the effectiveness of Wolpe's procedure. Psychological Bulletin, 1971, 12, 311-317. Wine, J. Test anxiety and direction of attention. ngchological Bulletin, 1971, 12, 92-104. "I(ll'l'll'llllll'lll'ls