ACTION AND VERBAL t RELATIONSHIP INTERVENTIONS WITH COLLEGE MEN FEARFUL 0F DATING Basis for the Degree of PILD. MICHIGAN STATE UNIVERSITY NANCY. M. HAY 1972 LIBRARY Michigan State University [IF-1E5”- IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIJllIllllIlI 3 1293 10414 8378 This is to certify that the thesis entitled I ACTION AND VERBAIr-REIATIONSHBP INTERVENTIONS WITH COLLEGE MEN FEARFUL'OF DATING presented by Nancy Marden Hay g" has been accepted towards fulfillment of the requirements for Ph.D. 4168136 in Counseling Date W 0-7 639 ABSTRACT ACTION AND VERBAL-RELATIONSHIP INTERVENTIONS WITH COLLEGE MEN FEARFUL OF DATING By Nancy Marden Hay The purpose of this investigation was to assess the effectiveness of three different short—term treatment inter- ventions with college males characterized by little or no dating behavior and fears about dating. The treatments were designed to produce evidence concerning the effectiveness of lay versus professional action interventions, action versus talking—relationship treatments using professional helpers and male versus female helpers in working with college males who desired to increase their dating frequency. Specifically, the four treatments included: (1) professional counseling with a talking—relationship focus, (2) arranged social inter- action with a variety of peers, (3) professional counseling with an action focus and a variety of counselors, and (u) a wait-control group. Male and female helpers were used in a controlled fashion in the three experimental treatments. Four criterion measures were used to test effects of the experimental treatments: (1) individual differences in general anxiety were measured by post-test scores on the Willoughby Personality Schedule; (2) individual differences in dating anxiety were measured by scores on the Specific Nancy Marden Hay Fear Index; (3) individual changes in the frequency of dating behavior were measured by a pre— and post—study change score; and (H) client satisfaction was indicated and a rating scale administered following treatment. The selection of the helpers and the volunteer sample, pre- and post-testing random assignment of subjects to treat— ment and sex of helper and the experimental interventions took place between December, 1970, and July, 1971, in the St. Louis, Missouri area. Six male and six female profession- als were paid to serve as the counselors, and five male and five female undergraduates comprised the paid peer helpers in the study. Two design paradigms were formulated which permitted: (A) testing for effects for sex, treatments, and interaction; and (B) testing for differences between the experimental treatments and the wait—control group. The test statistics for the two anxiety criteria according to paradigms A and B were generated by the use of multivariate analysis of covari— ance using the pre—test scores on the two anxiety measures as the covariables. The data from the client satisfaction mea— sure were analyzed using analysis of variance. The dating frequency data, according to paradigm A, were analyzed using a two—way analysis of variance. A univariate analysis of gain scores was used on the dating frequency data to test for differences between the experimental and the wait—control groups. Nancy Marden Hay The only results of the analyses which reached signif- icance at the .05 level were those from the MANCOVA analyses of the general anxiety data. The four conditions differed on general anxiety following treatment when adjustment was made for initial differences in anxiety. Further analysis revealed that the professional action treatment made this difference. The data also revealed a trend toward a greater de- crease in specific anxiety in experimental groups, compared to the wait-control (p < .06), a greater reduction of general anxiety in those subjects who saw male helpers (p < .07) and a tendency toward the greater effectiveness of the professional as compared to the peer treatment(s). ACTION AND VERBAL—RELATIONSHIP INTERVENTIONS WITH COLLEGE MEN FEARFUL OF DATING By Nancy Marden Hay A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Educational Psychology, and Personnel Services 1972 s .43 L7 ACKNOWLEDGMENTS Like most dissertations, this one owes much to the help and encouragement of others. I particularly wish to ex- press my appreciation to the following people: To my committee, Professors Norman R. Stewart, William Mehrens, Gwen Andrew and William Farquhar, whose encouragement and helpful suggestions have been indispensable. My special thanks to my major professor, Dr. Stewart, for endless sup- port, patience, humor, and prompt and relevant comments and editing. To Drs. Mary Ellen McSweeney and Howard Teitlebaum for statistical and computer evaluation and consultation. To Dr. Richard C. Nickeson and the rest of the staff at the St. Louis University Counseling Center for their cheer- ful and generous cooperation before, during, and after the investigation. To Dr. F. L. Erlandson who helped me to grow suffi- ciently to attempt this hurdle. To the professionals, the peers, and most of all, to the participants without whom there would have been no inves- tigation at all. And, finally, to my family who all gave so much. Especially to Tom, without whose nurturance, patience and ii intellectual stimulation, this dissertation would not have been possible. iii Chapter I. II. III. IV. TABLE OF CONTENTS INTRODUCTION AND REVIEW OF THE LITERATURE General Purpose of the Investigation . . . . Emphasis on Methodology . . . The Dating Phenomenon . . . . . . . . . Lay and Professional Helpers . . . . . . . Action and Verbal Relationship Approaches. Summary . . . . . . . . . . . . . . . . . . EXPERIMENTAL DESIGN AND PROCEDURES Objectives and Overview of the Study Experimental Hypotheses . . . Experimental Design . . . . . . . . Schedule of Experimental Procedures . . . Selection of the Sample . . . . . . . Description of the Sample . . . . . . . . . Description of the Counselors . . . . . . . Description of the Peer Helpers . Presentation of Experimental Treatments Criterion Measures . . . . . . . . . Collection of Outcome Data . . Description of Outcome Data Analysis Procedures . . . . . . . . . . . . . . Summary RESULTS . . . . . . . . . . . . . . . . . . . Analysis of Outcome Data . General Findings . . . . . . . . . . . . Supplemental Findings . .‘. . . . . . . Summary . . . . . . . . . SUMMARY, DISCUSSION AND IMPLICATIONS . . . . Summary . . . . . . . . . . . Limitations of the Study . . . . . . . . . . Discussion . . . . . Implications for Eurther Research Implications for Counseling Practice Conclusion iv 3” 38 H0 H1 H3 H7 50 56 57 62 71 73 75 76 76 78 87 97 99 99 10% 111 129 13H 136 BIBLIOGRAPHY APPENDICES N302: 3t47 EXPLANATION AND DATING QUESTIONNAIRE NOTICE OF DATING PROGRAM . . . . LETTER TO TREATMENT I SUBJECTS . DESCRIPTION OF TREATMENT I FOR PROFESSIONAL HELPERS . . . . . . . . LETTER TO TREATMENT II SUBJECTS . . . DESCRIPTION OF TREATMENT II FOR PEER HELPERS LETTER TO TREATMENT III SUBJECTS DESCRIPTION OF TREATMENT III FOR PROFESSIONAL HELPERS . . LETTER TO TREATMENT IV SUBJECTS WILLOUGHBY PERSONALITY SCHEDULE SPECIFIC FEAR INDEX . . . . POST- TEST NOTE TO SUBJECTS . . . MISCELLANEOUS EXPLANATIONS TO PEERS AND PROFESSIONALS . . . . . . . . . CLIENT SATISFACTION SCALE CONTRACT WITH SUBJECTS WEEKLY NOTICES TO SUBJECTS IN TREATMENTS II AND III . . . . . . . . . . . . . . Page 137 1&6 1&9 150 151 152 153 155 156 158 159 161 162 163 167 168 169 TABLE 2.1 3.10 3.11 LIST OF TABLES Sample age, Anxiety, and Dating Behavior Prior to Treatment . . . . . . Therapist Orientation Sheet Multiple Correlations Predicting Post— treatment from Pre-treatment anxieties MANCOVA, Univariate and Step—down E Statistics for the Simultaneous Testing of the Four Experimental Conditions on the Anxiety Measures . . . . . . . . . . . . . . . . . . MANCOVA, Univariate and Step-down F Statistics for the Simultaneous testing of Sex, Treat- ment, and Sex by Treatment Interaction Effects on the Anxiety Measures ANOVA of Post—treatment Dating Gains . . . . ANOVA of Post-treatment Dating Gains for the Four Experimental Conditions . . . . . . . ANOVA of Post—treatment Client Satisfaction Cell Means of Post- treatment Dating Gains Reported by Sex of Helper and Treatment Conditions . . . . . . . . . . . . . Cell Means of Post- treatment Client Satis- faction Reported by Sex of Helper and Treatment Condition . . . . . . . . Pre— and Post- treatment Cell Means for General and Specific Anxiety Reported by Treatment and Sex of Helper . . . . . General Anxiety Change by Sex of Helper and Treatment . . . . . . . . . . . . . Adjusted Estimates of Treatment--Control Group Differences in General Anxiety vi PAGE us 53 77 81 81 82 82 83 83 84 85 86 87 TABLE ANOVA of Post—treatment Gains on Item 2 for Sex of Helper and Treatment Conditions ANOVA of Post-treatment Gains on Item 2 for Experimental Conditions . . . . . . Extra-treatment Behaviors Directed toward Dating Goals . . . . . . . . . . . . Helper Behaviors in the Action Treatments . Experimental Treatments: Sessions, DrOpouts, Goals, and Referrals vii PAGE 89 89 90 92 96 CHAPTER I INTRODUCTION AND REVIEW OF THE LITERATURE General_purpose of the investigation The purpose of this investigation is to assess the effectiveness of three different short-term treatment inter- ventions with college males who have dating problems and are characterized by fear of dating. The treatment results will permit the comparison of: (1) professional counseling with a talking—relationship focus, (2) arranged social interaction with a variety of peers, (3) professional counseling with an action-role playing focus and a variety of counselors, and (u) a wait control group. Male and female helpers will be utilized in a controlled fashion in the three experimental conditions. The resultant data should provide evidence con- cerning the relative effectiveness of lay versus professional interventions, action versus verbal—relationship treatments, and male versus female helpers in working with male clients. Concern with this project grows out of four current issues: (1) the general trend in the field of counseling and psychotherapy toward attempting to bring methodological order out of "chaos" (Rogers, 1963; Colby, 1964) in order to meet the rising demand for service, (2) a conviction that "dating" presents a sufficient problem to many college males to merit systematic attention from the mental health professions, (3) an interest in the use of "lay" helpers and the evalua- tion of their effectiveness compared to that of "profession— als" and (H) the controversy concerning relative effective- ness of action (or behavioral) interventions and talking— relationship interventions. Emphasis on methodology An acute demand for a service which unfortunately utilizes vague techniques which are effective only part of the time with some of the people has resulted in an expres— sion of concern for improved methodology in research and in practice (Kiesler, 1966; Paul, 1966b; Rogers, 1963, 1965; Stollak, Guerney and Rothberg, 1966; Whiteley, 1967), Edwards and Cronbach presented a positive case for formal design in psychotherapy research, analyzed the issues, and discussed the limitations of design in their 1952 paper. Many other professionals have expressed similar beliefs in the impor- tance of good research design in investigations of counsel— ing, psychotherapy, and education (Campbell and Stanley, 1963, 1966; Krumboltz, 1966b; Paul, 1966; Farquhar, 1967; Krumboltz and Thoresen, 1969). Kiesler (1966) stated that in the area of psycho- therapy (and it seems certain that he would not omit counsel— ing), "Research can no longer ignore the necessity for facto- rial designs [p. 33]." Other writers emphasize the importance of operational definitions of relevant variables (Patterson, 1966; Stollak, Guerney and Rothberg, 1966; Wellman, 1967), the control of rival hypotheses (Campbell and Stanley, 1963), of objective, individualized behavioral outcome criteria (Krumboltz, 1967; Paul, 1966; Truax, 1967; Zax and Klein, 1963, 1966), or the desirability of research directed toward meeting the "test of relevance." Krumboltz (1967) defined "relevance" to mean that the results of the research will have direct implications for the nature of counseling prac- tice. Methodological concern is also focused upon the behav— iors of the counselor and counselee. Krumboltz and Thoresen (1969) described counseling as a specific process in which client and counselor work together upon feasible, explicit, and mutually agreed upon goals. The counselor experiments systematically with different ethical techniques in order to find those which will be most effective with a particular client. The authors cited above emphasize the importance of research designs which provide data which are relevant to the hypotheses under test, the increased use of factorial designs, explicit operational definitions of relevant variables, clear outcome criteria and of research which will have direct impli— cations for practice. The pressure caused by a demand for services in counseling and psychotherapy has encouraged the investigation of the effectiveness of many procedures. This same pressure, a slowly growing scientific sophistication and a body of literature composed to an unfortunate extent of con- flicting and equivocal results, has led to an increasing focus on the importance of scientific methodology in investigations in the counseling field. ‘f ffi?wThe dating phenomenon The second source of interest in this investigation is the dating phenomenon. Authorities in several disciplines have stated that dating_and_courtship behaviors are important in the developmental process (Blocher, 1966; Gottlieb, 1964; V Levine, 1963; Lowrie, 19u8; Muus, 1962; Smith, 1962). Havig— hurst (19H8) considered the development of new relations with age~mateggofgbgth sexes‘:o be an important task of adoles— cense, "successful~aghievement of which leads to his happi- V nes§_and\tg_§uggg§§ with later tasks, while failure leads to M unhappiness in the individual, disapproval by the society, A . and difficulty withglate§_ta§ks Ep. 6]." Schoeppe and Havig- f— hurst (1960) supported this earlier statement with a longitu- dinal study of high school students. Dating seems to be a W- major way of establishing these new and different relation- “. .-.__-—---—-—"‘ ships with the opposite sex and to be important in learning . *_v#####, 1.11 an appropriate sex role. There is pressure to date‘from .1_11_~__._11___ ‘ '--~.-. ' *7 F‘ur-‘— _-.. ..-—~——..— peers, from parents, and from the schools. The school pro- - .,__1.'— vides "a multitude of non-academic programs [which] . . frequently call for the participation of male and female partners [Gottlieb and Ramsey, 1964, p. 171]." Some schools value popularity highly. In schools which emphasize dating . -*-______, 4__"#_________ activity, "the persons who achieve in this activity feel ‘ fl best about themselves and less often want to be someone dif— W _ K.— ferent . . . the effects-of these social systems on boys' or girls' self—evaluation are extremely powerful [Coleman, 1962, p. 236]." The individual who does not date may become "frus- A It # .'~ trated and turn to more introverted forms of reaction [Lowrie, 1948, p. 90]." Ngn-daters consider themselves as less valu— able than their peers and admit to feelings of "social fail— ure," They have to deal with a social stigma, a feeling of personal rejection and isolation (Levine, 1963), and possible pessimism about their eventual ability to find a marriage partner (Gottlieb and Ramsey, 1964). ~_',,.1 The hard data to support these assertions about the traumatic effects of non—dating are not conclusive. Garrison (1951), Meissner (1961), and Morgan (1969) provided data about the problems of high school students indicating that boy-girl concerns are a source of fear and worry to a size— able number of them. Early studies of college students did not suggest a great concern with personal adjustment or heterosexual rela- tionships. Congden (1943) felt, however, that this might be a reluctance to admit such concerns on the Mooney Problem Checklist. Several studies strongly support the position that college students are concerned about interpersonal rela- tionships--particularly those between men and women. Sister Jane Becker (1971) dealt with the presenting problems and goals of St. Louis University students. Her sample was small (24 Counseling Center clients and 20 control subjects), but it is particularly relevant since it is from the same university which provided 32% of the sample for the present study during the following year. Becker's subjects ranked 37 problems as to importance. A ranking of "one" was "very important" and "nine" was "least important" or "not at all important." Eleven of the 24 clients ranked "better relations with the opposite sex" among their four most impor— tant concerns. Seventeen of the 20 control subjects listed improved relations with the opposite sex as important goals for them. Certainly, Heath and Gregory's Harvard sophomores (1946), Gordon's (1950) Ohio State undergraduate women, Koile and Bird's (1956) East Texas students, Ottoson's (1967) lib- eral arts students, Martinson and Zerface's (1970) Indiana University students, Rust's (1960) Yale undergraduates, and the St. Louis University students indicate that they have concerns about social and recreational activity, courtship, and dating. A study done in 1962 at the University of Cali- fornia at Davis, which was quoted by Bolton and Kammeyer (1967), stated that juniors and seniors ranked "dating" a close second in importance to "living with peers" in the formation of their "self-definition." These students, appar- ently, agree with the authorities on the importance of dating in their developmental process. Misinformation and lack of familiarity with what is expected on a date contribute to concern over dating. Men ”—~ tend to be more romantic than women about datigg (Knox and H. ______._..« b—‘fi -rfiu— Sporakowski, 1968) and to_assume th§t_wgmen*§re_mor§_ggn- -— mun—1... (Balswick and Anderson, 1969; Blood, 1956; Hewitt, 1958) than the really are. Misconceptiogg such as these make dating appear more difficult than it needs to be and COEEEEREEE_FO the communication problems between men and women. W Correlational studies suggest that there may be an association between non—dating in college and social isolation and possible failure (Slocum, 1956; Rust, 1960; Adinolfi, 1970; and Martinson and Zerface, 1970). "Rates of emotional disturbance tend to rise as one moves from a condition of low friendship solidarity and academic dissatisfaction [Segal, 1967, p. 242]." With the average ability student at Duke grades vary inversely with anxiety level (Spielberger and Katzenmeyer, 1959; Spielberger, 1962). Grace (1957) found responsibility and low anxiety correlated with staying in college. Colleges characterized by a high frequency of informal dating are also characterized by high dropout rates. "Almost one—half of the students who left their first college indicated they were dissatisfied with their environment [Astin and Panos, 1969, Panos and Astin, 196813' No direct information is given to indicate the particular sources of dissatisfaction for these students. Students who indicated they had a number of close friends tended, however, to per- sist in college (Panos and Astin, 1968). One can only sus- pect that it may have been isolated non-daters who drOpped out of those college environments where informal dating was emphasized. \I' The data at least suggest that dating problems may have a detrimental influence on the college students' careers ._1 4"_____1 and on their self-esteem. The need is apparent for syste— . ’_,______ matic study of the effect of dating problems on academic achievement, self-esteem, interpersonal relationships, and psychosexual development. It seems equally clear that stu- dents are aware of these problems and would like some assis- tance with them (Martinson and Zerface, 1970; Becker, 1969). The entire institution of dating is a neglected research area (Martinson and Zerface, 1970; Lowrie, 1948; Gottlieb, 1964; Schoeppe and Havighurst, 1952, 1960). It is not surprising then that the literature indicates very few studies of programs designed to help students improve their dating skills. Martinson and Zerface (1970) and Morgan (1970) describe two such programs. Morgan (1970) used four short- term experimental conditions with "seldom—dating" Indiana University students: role models, model exposure plus re— hearsal, rehearsal only, and focused counseling. Behavioral rehearsal resulted in a decrease in anxiety on a self-report measure. Adding the model to behavioral rehearsal had a detrimental effect. The men exposed to this approach made fewer date requests following treatment. There were no statistically significant differences on the other criteria used in Morgan's study. In the study described by Martinson and Zerface (1970), 24 University of Indiana students characterized by fear of dating were assigned to professional, individual counseling, a semistructured social program with coeds who were interested in improving their own social skills or a delayed-treatment control group. Subjects were compared on change in general anxiety, change in specific fear of dating, on the number who were dating at post—treatment follow-ups, and on client satisfaction. The social program seemed to be the most effective in decreasing the specific fear of dating and in increasing dating behavior. The professional counsel- ing was most effective in decreasing general anxiety. It is difficult to interpret the results, however, because sex, status, and age of helped were confounded with treatment in the design. The present study partially parallels that of 10 Martinson and Zerface and attempts to control some of the methodological problems which blurred their results. ¥1College is sometimes seen by the students and the professionals as a place where the developmental problems associated with non-dating may be corrected (Blocher, 1966; Coons, 1970; Heath and Gregory, 1946, 1960). The available data strongly suggest that it is time that we take advantage of this Opportunity. Lay and professional helpers The demand for service has contributed to the use of lay personnel by mental health professionals in various disci- plines. This practice has continued to develop despite con- siderable reluctance to employ non-professionals, and anxiety about the possible consequences of their employment. The literature contains much support for the usefulness of lay mental health workers and much information about various pro- grams for training them. Lay personnel have been used effec— tively with hospitalized psychotics (Carkhuff and Truax, 1965; Poser, 1966), in community mental health and crisis centers (Farberow, 1969; Lynch, Gardner, and Felzer, 1968), with rehabilitation in-patients (Truax and Lister, 1970), with junior high blacks (Carkhuff and Griffin, 1970), with children in play therapy (Stollak, 1968), and with anxious college students (Neuman, 1968; Martinson and Zerface, 1970). Lay personnel have worked with groups and with individuals. 11 Training has ranged from sending the lay personnel a "struc- tured letter" (Martinson and Zerface, 1970), to the two-year program for college graduates conducted by Rioch (Magoon and Golann, 1966). Some studies compared the effectiveness of profes- sionals with that of non-professionals indirectly by evalua— ting the levels of the various process variables (empathy, genuineness, positive regard, etc.) offered by helpers having different status. [In] general, following training, on both identical and converted indexes [gig], lay trainees function at levels essentially as high or higher (never significantly lower) and engage clients in counseling process movement at levels as high or higher, than professional trainees [Carkhuff, 1969, p. 5]. While the number of comparative studies is limited, with both outpatients and inpatients, lay persons effect change on the indexes [sic] assessed at least as great or, all too frequently greater (never significantly less) than professionals [Carkhuff, 1969, p. 7]. Some of the studies which Carkhuff cites are not directly relevant to a college student pOpulation or to problems with interpersonal relations such as dating. Two of the six stud- ies to which he refers are not actually comparable in any experimental sense (Harvey, 1964; Magoon and Golann, 1966). Harvey (1964) described the use of trained and super- vised non—professional volunteers as marriage counselors in Australia. Harvey said, The results cannot be described in experimental language, but . . . the increase in numbers of professional peOple l2 willing to associate with such a counseling activity indicates that controlled and supervised services of selected and trained non-professional workers can pro— vide a useful service [p. 351]. No attempt was made by Harvey to compare the work of the non- professionals with that of professional counselors. The mental health workers with whom Rioch (1966) and Magoon and Golann (1966) were concerned were indeed evaluated as effec— tive by their supervisors and co—workers, but that study did not directly compare their effectiveness with that of more conventionally trained counselors or therapists. Anker and Walsh (1961) attempted to compare the results of group psychotherapy, group structure and a special activities program with hospitalized schizophrenic patients in a veterans hospital. The design was complex, the explana- tion poorly presented, the group psychotherapy all done by the senior author and the drama activity all led by one recreation worker, which confused the interpretation of the results considerably. According to the authors, the only treatment showing statistically significant improvement on a paper and pencil measure of behavioral adjustment was the activity group led by the "non-professional." Poser's study (1966) with schizophrenic in-patients also has several confounding variables (time and phenothiazine being two of them). In this study, the members of the lay— treated groups showed more improvement than those treated by professionals on the verbal and psychomotor pencil and paper 13 criteria. There was no significant difference in discharge rate. Mendell and Rapapport (1963) worked with schiZOphrenic out—patients in a study covering a 51—month period. Psychi— atrists, psychologists, social workers and psychiatric aides all had a once-monthly, 30—minute treatment interview with their schiz0phrenic out-patients. The patients were not ran— domly assigned, but were picked by their helpers, sometimes on the basis of continuing therapy begun when they were in- patients. Some of the patients were receiving medications, but no information was given about which patients in which groups were on what medications. The recidivism rates were 20 percent for those patients seen by psychologists, 23 per— cent for those seen by social workers, 34 percent for those seen by psychiatrists, and 36 percent for those seen by psy— chiatric aides. Who can say which of many fluctuating vari— ables may have caused this difference? In a study with superior design to those cited by Carkhuff and discussed above, Truax and Lister (1970) compared the effectiveness of counselors, counselor-plus-aides, and supervised aides working with rehabilitation clients. On all criteria, "the best results were obtained by the aides work- ing alone under the daily supervision of professional counsel— ors [p. 333]." Although sex of helper and age are uncon— trolled, the results were clear that the secretary helpers in this study were more helpful to clients than master's level counselors. 14 Studies dealing with college student populations are particularly relevant to this discussion. Wolff (1969) con- ducted a study with male freshmen at the University of Rochester. The experimental treatments consisted of small group discussions once weekly for a total of ten sessions. Four groups were led by advanced psychology doctoral students with considerable experience and four groups were led by undergraduate resident advisers who were supervised by the advanced doctoral students. The experimental subjects were compared on four criteria to two sets of control subjects (volunteers for the project who did not participate and non- volunteers). The outcome criteria included number of socio- metric choices by peers, grade point average, visits to the counseling service and number and kinds of activities as indicated on a form. "In general the results of comparing all the controls to all the experimental subjects suggest that group experiences can favorably affect the interpersonal relationships of freshmen [p. 301]." The experimental groups improved more than the control subjects on the sociometric criterion. "There was a trend suggesting that group E (grad- uate student led) increased more than all other groups on the percent of favorable other choices by the outgroup [p. 300]." The differences on the other three criterion measures were not significant, but it is clear that both lay- and profes- sionally (or at least semi-professionally) led group 15 discussions were helpful in the adjustment of freshmen col- lege males to same-sexed peers. In two other comparative studies with college stu- dents, Neuman (1968) and Zunker and Brown (1966) concluded that interventions which did not involve a professional coun— selor were equal to or more effective than those which used professionals. Both of these studies involved group inter— ventions. Neuman compared short-term group desensitization and group insight procedures used by professionals and sub— professionals and two control groups with test-anxious col- lege students. He used multiple outcome criteria, including various anxiety measures and client and therapist ratings. The sub-professionals proved more effective on one inventory for anxiety and the professionals on another. There were no significant differences between the two kinds of helpers on other criteria. Both experimental groups improved. Zunker and Brown (1966) implemented an activity se— quence for academic adjustment with freshmen. The prOgram consisted of four meetings, the first three of which were group meetings. Two male and two female professionals and four male and four female upperclassmen composed the helpers. Both groups of helpers were trained in a 50-hour training program. There was no significant difference according to status of helper on the effective study test criterion; but the subjects who had student counselors retained more 16 information on the Counseling Comprehension Test and rated the program higher than did those subjects with professional counselors. The student—counseled girls made better use of the study skills knowledge than other subjects and the stu- dent—counseled subjects as a whole were significantly higher on grade point averages! The professionals in this study were state certified and experienced school counselors who were candidates for the master's degree at Southwest Texas State. It could be argued that the study compares non- professionals and semi-professionals. All freshmen partici- pated as experimental subjects. Martinson and Zerface (1970) had both peers and pro- fessionals working with college males with dating problems, but the treatments utilized by each group were very differ- ent. The study does indicate that girls who have dating problems and utilize a semistructured program can be more effective in increasing dating behavior of college males than male counselors using traditional short-term counseling. The studies which compare the relative effectiveness of lay and professional counselors are surprisingly few and methodological problems are unfortunately frequent. The data certainly suggest, however, that interventions by non- professionals can be helpful with a variety of people with a variety of problems. In the comparative studies with stu- dents, the lay counselors certainly proved to be effective l7 and, in the Zunker and Brown (1966) study, to have better results on several criterion measures than did the master's level school counselors who comprised the professional helpers. There is insufficient evidence to indicate that professionals of various levels should turn direct service over to lay counselors, but the data do support the use of sub—professionals and the need for a great deal more compara— tive research. Action and verbal—relationship approaches The controversy between the "action-oriented" profes- sionals (behaviorists, socio-behaviorists, etc.) and the talking-relationship-oriented professionals (whom London [1965] refers to as "insight therapists") has resulted in a mammoth body of vituperative literature. Breger and McGaugh (1965), for example, criticized the learning theory explana- tions used by behavior therapists as simplistic, their meth— odology as weak and misleading, their conceptualizations of neurosis as inadequate and their success claims as inflated. Rachman and Eysenck (1966) attempted a rebuttal of this critique and also of the reformulation of behavior therapy suggested by Breger and McGaugh (1965). Essentially they accused Breger and McGaugh of making distorted, inaccurate and misleading criticisms and of presenting a vague and use- less reformulation. They reasserted that behavior therapy, 18 in spite of falling short of perfection, was making a greater contribution than other therapies because of its behavioral laws regarding the relationship of CS and UCS and condition— ing, extinction and different response rates. From these laws, they stated, effective techniques producing excellent results have developed. They repeated their assertions about the superiority of behavioral research. Breger and McGaugh (1966) replied with a denial of most of these charges, a reiteration of their original attack and a study by study analysis of the research referred to by Eysenck and Rachman to document the successes of behavior therapy and its research excellence. Breger and McGaugh contended that these studies were either irrelevant to the argument or uncontrolled case studies. There is no purpose here in describing the contro— versy between the "behaviorists" and the talking—relationship oriented therapists in greater detail. Neither side can boast of an explicit, integrated and generally accepted theory of personality behaviors from which a theory of personality change has been derived. In this discussion I will present a very brief comparison of the two points of view and an ex— ample of each. Albert Bandura (1969) represents an example of the action professionals and Carl Rogers of the talking- relationship viewpoint. 19 London (1965) stated that the most important commonality of the theories of the Action therapists is that all of them come under the heading of theories of learning . . . they are all agreed that only a very few principles of learning are needed to understand even the most complex kinds of behavior to which psychotherapy is applicable [p. 81]. In counseling, these therapists tend to focus on producing specific changes in overt behavior, to be very active and directive in treatment and to take more responsibility for process than do the talking—relationship group. Techniques include: systematic desensitization, conditioned avoidance procedures, role-playing, implosive therapy (a la Stampfl), assignments to perform various specified behaviors and the use of shaping and modeling procedures. Albert Bandura can serve as an example of the action- oriented professionals although his conceptualization of human learning is more broad than that of most of the learn- ing theorists to whom London referred. He concerns himself with human learning of three general types in his articula- tion of his social learning theory: (1) those behaviors which are a reaction to environmental stimulation and associ- ation (emotional behavior resulting from autonomic nervous system or behaviors resulting from physiological tensions such as hunger or fatigue, for example), (2) those behaviors resulting from feedback to previous behaviors (positive or negative reinforcement, for example), and (3) those behaviors which are the result of cognition (coding and synthesizing 20 data and hypothesis testing, for example). Behaviors are acquired and maintained on the basis of what Bandura refers to as "three distinct regulatory systems" which include the types of behavior discussed above and to which Bandura refers as (1) external stimulus control, (2) response feedback processes, and (3) central mediational processes. Bandura sees man as "neither an internally impelled system nor a pas— sive reactor to external stimulation. Rather, psychological functioning involves a reciprocal interaction between behavior and its controlling environment [Bandura, 1969, p. 63]." Bandura focuses on manipulating the social environ— ment, on an active planning counselor, on clear explication of the goals and sub—goals of treatment and on the use of models, reinforcement and withdrawal of rewards to induce behavior change. He emphasizes the importance of client behavior in the client's milieu rather than office behaviors. It follows from principles of generalization that the optimal conditions for effecting behavior changes from the standpoint of maximizing transfer effects, would require people to perform the desired patterns of behav— ior successfully in the diverse social situations in which the behavior is most appropriate. On the other hand, when treatment is primarily centered around verbal responses expressed in an invariant atypical context one cannot assume that induced changes will necessarily gen— eralize to real life performance to any great extent [p. 105]. Bandura does not see the relationship as essential in social learning and implies it is not essential in "change programs." His use of learning theory principles is broader than that of 21 many of the action people due to his consideration of hypothe- sis testing, coding and organizing in his discussion of the "central mediational processes." The talking-relationship oriented therapists share an emphasis on talking within a relationship which roughly par- allels the emphasis on "doing" among the Action therapists. The talking-relationship people tend to emphasize inter- and intra—personal relationships and internal more than external behavior. Emotions and the understanding of feelings are considered important areas of concentration as these counse- lors work to help their clients to change internal behaviors and (often) external behaviors as well. In general, practi- tioners with this point of view are less concerned with explicating goals for treatment, leave more responsibility for the conduct of the sessions up to the client and are less apt to involve themselves with environmental manipulation than are the action-oriented therapists. They tend to empha- size the internal.dynamics or experiences of the individual as manifested in social relationships (past or present) and to pay little attention to the details of the learning process in which these behaviors are learned or changed. The goals of therapy are usually phrased in terms of global, inter- or intra-personal behaviors or characteristics such as Openness, congruence, maturity, or self-acceptance rather than in terms of rather specific actions or groups of actions such as: g.‘ t . I I ’x\ r 22 increase assertive behaviors, eliminate a phobia, reduce test anxiety or increase dating behavior. "Behavior," according to Rogers (1951) "is basically the goal-directed attempt of the organism to satisfy its needs as experienced, in the field as perceived [p. 491]." A basic goal or need of the organism is the tendency to actu- alize, maintain and enhance itself. Rogers includes physio- logical tensions as behavior modifiers, but he does not deal with these at length. Perception by the individual is partly conscious and partly unconscious, but each person exists at the center of his own private world due to the uniqueness of his perceptions. His behavior is a total organized response to this private perceptual field and much of it is dependent upon that part of the total which the individual defines as self. Psychological maladjustment exists when the organism denies to awareness significant sensory and visceral experiences which consequently are not symbolized or organized into the gestalt of the self-structure. When this situation exists, there is a basic or potential psychological tension [p. 510]. The client, then, is the "only one who has the potentiality of knowing fully the dynamics of his perceptions and behav- ior [p. 221]." The activities of the counselor are directed toward helping the client to explore his conflicts and behaviors and toward helping him in "experiencing the inadequacies in old ways of perceiving, the experiencing of new and more accurate 23 perceptions, and the recognition of significant relationships between perceptions [pp. 222-23]." Rogers clearly sees this as a learning or re-learning process. He does not address himself to analysis of the learning process, however, but to the relationship within which the relearning can take place and to the qualities of that relationship. Earlier in his career, Rogers concentrated on empathy, positive regard and congruence from the therapist to the client in a relationship in which the therapist functioned as a mirror. Recently, Rogers has emphasized a more mutual relationship between client and counselor. He has continued to investigate em- pathy, positive regard and congruence offered by the thera- pist as important variables in effective therapy, but has changed his emphasis somewhat. "Therefore it would seem that this element of genuineness, or congruence is the most basic of the three conditions [Rogers, 1967, p. 100]." He assumes client vulnerability to anxiety (incongruence), psychological contact between therapist and client to the extent that "each makes a perceived or subceived difference in the experiential field of the other [p. 99]" and client perception of the therapists offered conditions to be necessary factors in con— structive client change. The therapist who is of Rogerian persuasion will con— centrate on "tuning in" to his client's world and upon the quality of their relationship. The main techniques are verbal 24 and the helper will be less active and directive than an action helper would be as he attempts to aid the client in learning about himself in order that he may change his be— havior. [The] changes in behavior keep pace with the changes in organization of self, and this behavior change is, sur- prisingly enough, neither as painful nor as difficult as the changes in self-structure. Behavior continues to be consistent with the concept of self and alters as it alters [Rogers, 1951, p. 195]. Rogers himself appears to be moving slightly toward the position that "much significant learning is acquired through doing [Rogers, 1969, p. 162]." In the classroom at any rate he has emphasized the importance of active partici- pation by the learner. He saw a free atmOSphere as a neces- sary condition for permitting the "facilitation of change and learning [p. 104]," which were his stated goals for educa- tion. The active learner will, Rogers indicated, feel better about himself and advance toward the goal of becoming a fully functioning person as the result of participation in and taking responsibility for the learning process. The implica- tion seems clear that the action results in a change in the person's self-concept. In this context, he indicated agree- ment with B. F. Skinner that one must engage in behavior to acquire behavior (Rogers, 1969, p. 140). Rogers does not seem, as yet, to have integrated the above point of view into his conceptualizations of psycho- therapy. The client "finds his behavior changing in 25 constructive fashion in accordance with his newly experienced self [Rogers, 1969, p. 281]." He was, at this writing, in the somewhat paradoxical position of holding his old view that overt behavior change follows changes in the perception of self and the conviction that, at least in some instances, overt behavior may lead to changes in self-concept (as Bandura contended). Both the Action and the Talking—relationship factions have done some self-examination in attempts to isolate rele— vant variables in client change. The data suggest that the positive results of desensitization are not due to extinction or placebo effects, but that relaxation training is an impor- tant ingredient for change, that the quality of the relation— ship makes a difference when combined with an action treat— ment for obese clients and that high levels of empathy, respect and genuineness offered by therapists are associated with positive changes in clients, while low levels are asso- ciated with negative changes. Laxer and Walker (1970) attempted to isolate vari- ables relevant to change in test—anxious high school students. They found that "treatment was only effective in those condi— tions where relaxation training was involved [p. 434]." They found no support for the hypothesis that systematic desensi- tization could be explained by simple extinction or placebo effects. 26 Stollak's (1967) study also gives some information about relevant variables. His results suggest that a com- bination of action (diary-keeping) and relationship "did result in significant weight loss. But when another variable such as shock, specific or non-specific, enters into this relationship the positive effect of the relationship dis- appears [p. 63]." Carkhuff (1969) has been involved in many studies which, he says, support the position that "helpees of high- level functioning helpers demonstrate constructive change on a variety of indexes [gig] while those of low-level function- ing helpers do not change or even deteriorate [p. 24]." He went on to refer to many studies which indicated that the helper and his levels of empathy, respect and genuineness have a direct effect on the process and outcome of the therapy. Truax (1963) attempted to relate measures of thera- pist empathy, unconditional positive regard and congruence to personality change in psychotherapy with schizophrenics. Those patients who improved had therapists who were rated higher on empathy, positive regard and self—congruence. The data suggested that the patients who were treated by thera— pists who were rated low on the characteristics showed nega- tive personality change on a battery of psychological tests. These tests included the Rorschach, the MMPI, the TAT, the 27 WAIS, Wittenborn Psychiatric Rating Scales and several anxiety measures. In a parallel study, also discussed by Truax (1963), those patients whose therapists were rated low on the therapeutic conditions showed increased anxiety on an Anxiety Reaction Scale. We have only begun to isolate the variables which are relevant to client change. Relationship variables (such as empathy, congruence and acceptance) have been shown to have a clear association with the amount and direction of client change. Reward and punishment, modeling behaviors, relaxa- tion training, role playing, and central mediational processes also have association with external and internal behavioral change. All of these variables and techniques may be useful in some cases, some may be useful in all cases, but we are not certain that all are useful in all cases. Case studies have long been available describing in— sight approaches and are easily accessible for action treat- ments as well (Krasner and Ullman, 1968; Krumboltz and Thoresen, 1969). Certainly these data suggest that peOple with anxiety and fears about Speaking up or asserting them- selves in performance, evaluation, or interpersonal situa- tions respond favorably to role-playing (Hosford, 1969; Varenhorst, 1969) or desensitization (D'Zurrilla, 1969; Emory, 1969; Paul, 1966, 1968). The lack of control of rival hypotheses in the one-shot case study, however, leaves one 28 stimulated, but unarmed with ammunition to answer the ques— tion of "which procedures and techniques, when used to accom- plish which kinds of behavior change, are most effective with what kind of counselors [Krumboltz, 1966, p. 22]." Comparative experiments, which have begun to appear more frequently in the literature in the past ten years, pro— vide some progress in our attempts to answer Krumboltz' ques- tion. Design problems prevent some of these from being clearly interpretable (Anker and Walsh, 1961; Martinson and Zerface, 1970). Martinson and Zerface, for example, had so many systematic confoundings of variables within their experi— mental treatments that one can only guess which of these may be relevant to the results which they found. An example of a recent comparative study is that of Marks, Sonoda and Schalock (1968). Chronic schizophrenics were assigned to either a behavior modification reinforcement program or a relationship treatment. The investigators found both effective. There was no consistent difference in over—all effective- ness between the two. Moreover, reinforcement showed no consistent differences from the relationship functions which it facilitated [p. 401]. Likewise, Hogan and Kircher (1968) investigated the relative effectiveness of implosive therapy, eclectic verbal therapy and bibliotherapy with coeds who were snake phobics.~ They reported no difference in the success rate between implosive and the eclectic verbal approaches, but bibliotherapy was 29 significantly less effective than other treatments. Compara- tive investigation of insight and action approaches was con- ducted by Hedquist (1968). In addition to the action group counseling and didactic insight group counseling experimental groups, he also included a control group and attempted to measure the effectiveness of each experience as he worked with college students on academic warning. He reported no differences in effectiveness between groups. Paul's elegantly designed experiment and the subse— quent follow-ups and amplifications (1966a, 1966b, 1966c, 1967, 1968) can be quite clearly interpreted as support for the superior efficacy of individual systematic desensitiza- tion to insight-oriented therapy with college students expe- riencing performance anxiety. Paul's work also supports the value of systematic desensitization procedures added to a group treatment with students experiencing performance anxi- ety. He found no symptom substitution problem in the follow— up studies. There were no relapses except among the control group students. Even the attention-placebo group was stable in improvement at the two-year follow—up. He suggests that this may have been the result of the relationship which was part of the placebo. The results of studies reviewed do not conclusively prove either action or talking—relationship approaches to be superior. They do suggest possible value in a 30 rapprochement. One sees the beginnings of this rapprochement in a cross-fertilization of techniques and methods. Some behaviorists are acknowledging that the warmth, empathy, and positive regard of the counselor must be communicated to the client (Krumboltz, 1966; Krumboltz and Thoresen, 1969; Bergin, 1966). The verbal-relationship oriented counselors are using various behavioral techniques and explicating the goals of treatment with their clients. The need is clear "for further experimentation in many areas that we may better pinpoint what treatment, by whom, is most effective for thi§_individual with that specific problem and under what set of circum— stances [Paul, 1966, p. 7]." At best the kind of question posed by Paul and Krumboltz can only be answered in terms of probabilities. However, it seems apparent that knowing what procedures are most likely to be helpful, utilized by X kind of helper, with Y kind of individual having Z problems and goals with which he wants help would greatly simplify coun— seling practice and training! One can speculate that both action and talking-relationship approaches will be a part of the answer to the above question, but this will have to be settled by continued experimentation. Summary Practitioners in counseling and psychotherapy are genuinely concerned with the methodological muddle in which they find themselves. The literature is monumental, but the 31 data relevant to improving the effectiveness of practice are not impressive. More issues exist than well-supported answers and more dogma than explanation. This has produced two alternative responses. Some researchers have reacted by an increased focus on improved research design, increased and more rigorous self—examination by members of the profession and a demand for relevance in questions the researchers have posed. The other reaction to the "state of chaos" (Rogers, 1 9 66, p. 11) is to turn away from research as useless or im- PO ssible. A sense of futility about doing research and of hopelessness about attempting to identify (much less to (11—1 antify) the relevant variables for client change is often expressed by grumbling graduate students and harassed practi— t:l.‘caners. This response is rarely specifically articulated in the literature, however, and is not a course of action likely to integrate chaos or to improve practice. This point of v iew has not contributed to the frame of reference of the QKlnrrent investigation. I kThe dating phenomenon appears to be relevant to the <1evelOpmental process. College students share problems in this area with students still in high school. The data sug— gest that inability to date, or intense anxiety about dating, may have a detrimental influence on the school eand personal development of students. Little research has been flu“ 1%---- -111, 7... conducted in this area in spite of the importance of the 32 jlssssue and the availability of college subjects for research F>zé<>jects. Perhaps the area has been neglected partly because 1:1113 demand for service is not as evident as it is with other E>crsciblems. Those males who hesitate to assert themselves in ciaai1:ing behavior are not likely to assert themselves in loud cries for help . The increasing demand for various kinds of mental 1163:511th services and an undersupply of trained professionals supply these services has had diverse effects. One result 't<:> IlaaLsss been the increased use of non-professionals as helpers. rI'I'1. literature indicates that these lay helpers can be very Lissmeszful with a variety of patients and clients. Here again, IICDVCAJever, the data are not sufficient to settle the questions C’f?‘ whether there is a difference in effectiveness between 131:‘<::fessionals and sub-professionals, in what areas the dif— feE=ZIPence may lie, if it exists, and in what circumstances lay C>3=5' professional helpers might be more appropriate. The controversy between the action and the talking— Ifi‘Eiilationship therapists presents us with a similar dilemma. ‘OIEE do not have clear theoretical positions to compare and 1:thiare are few comparative studies to answer questions about ‘tlle.relative effectiveness of the two approaches with various k:inds of problems and people. The purpose of this investigation was to compare the effectiveness of three short—term treatments and a wait-control 33 group. The treatment results provided evidence concerning the relative effectiveness of lay versus professional inter— ventions when both kinds of helper used an action approach, of action versus verbal-relationship treatments by profes- Sionals and of male versus female helpers (of either status) in working with college males'whose fear of dating was limiting their dating behavior severely. o o. . .. _ I vuvv- n-‘ ‘1 .- h- 'I ‘-~. 'v'. u I... 1‘: ll 1 f l" (I) CHAPTER II EXPERIMENTAL DESIGN AND PROCEDURES Ezlpjjectives and overview ggjf' the study The purpose of this study was to assess the effec— 1Zinveness of three short—term treatment interventions with Cnbllege males who seldom dated and who were characterized by ifear of dating. The three eXperimental treatments were com— Ipared with a wait—control treatment and the experimental groups were tested for differences between the treatments, sex of helper and the possible interaction of treatment and sex of helper. The current study is a partial replication of an earlier study done at the University of Indiana and described by Martinson and Zerface (1970). This study is discussed briefly in Chapter I of this paper. The present study deals with the same general problem pOpulation as do Martinson and Zerface. The subjects in the earlier study were fearful non- daters, those in the current study are fearful seldom—daters. The subjects in both studies were compared along the dimen— sions of general and specific anxiety change, dating frequency, and client satisfaction. Two of the criterion variables have been modified for the current design. These modifications 34 35 VJlel.be described in detail later in this chapter. The design of the current study has sex of helper and 1:11f3 three experimental treatments as completely crossed inde- E>€3I1dent variables. The three experimental conditions include EDIPCDfessionals in a talking-relationship mode and peers and l>lrc>fessionals each in an action-oriented mode of treatment. 7P11e3.fourth condition is a wait-control group. This differs ifxmom the Martinson and Zerface study which had only two ex- Exerimental treatments and a delayed-control group. The male I>rofessionals in one of their experimental conditions coun- Seled in an eclectic fashion and the female peers in the other experimental treatment participated in an arranged social interaction. Sex and status of helper and mode of treatment were thus confounded, which made interpretation of the results most difficult. The results shed little light on the questions underlying their research concerning the rela— tive effectiveness of peers and professionals and of the action and "traditional counseling" approaches to dating behavior change. Both the Martinson and Zerface and the cur- rent study involve the same duration and frequency of treat- ments. The statistical analyses of the dependent variables differ considerably in the current study from those utilized by Martinson and Zerface. This will be discussed more fully later in this chapter. 36 Four general questions of interest were addressed in 1:1163 study with relation to college males under 30 with dating cizigfficulties. They were: (1) Is there a difference in the €3:E2Eectiveness of male or female helpers? (2) Is there a dif- jfeaznence in the effectiveness of professionals in a talking- IreaiLationship-oriented treatment versus an action-oriented Eigxproach? (3) Is there a difference in the effectiveness of Iplcofessionals in an action-oriented approach versus peers in Gun action-oriented approach? (4) Is there a difference in (Dutcome between the experimental groups and the wait—control group? The three experimental treatments all involved having the subjects meet with a helper for one hour a week during a total of five weeks. One half of the subjects in each treat— ment saw all same-sexed helpers and the other half saw all opposite sexed—helpers. Prior to the beginning of treatment, all the subjects received letters explaining their particular treatment and giving them some idea of what their responsi- bilities would be and who their helper would be. In the case of those subjects in Treatments II and III, they were given the name and telephone number of their first helper (see Appendix). The treatments were as follows: (1) Professional counseling with a verbal—relationship orientation and the same counselor each week. Behavior modification techniques 37 vveure ruled out of this treatment. (2) Arranged interaction fi;r1 which the subjects saw a different peer helper each week. 171163 subjects were responsible for making arrangements to meet 1:11621r assigned helper each week at a convenient place. The earrqphasis was on performing behaviors similar to those in- \7c>1ved in dating. (3) Professional counseling with an action ifc>cus and a different counselor each week. The participants Vveare responsible for making arrangements to meet their as— ESigned helper each week at a convenient place. Any ethical IDehavior modification techniques were considered appropriate for this treatment. The emphasis was on the performance of behaviors relevant to dating. The three experimental treat- ments will be described in greater detail later in this chapter. Four criterion measures were used to test the effect of the various treatments: (1) individual differences in general anxiety as measured by post—test scores on the Willoughby Personality Schedule (1932); (2) individual dif— ferences in dating anxiety as measured by scores on the Specific Fear Index; (3) differences in the amount of dating behavior as derived by subtracting the number of dates which each subject had had in the six weeks prior to treatment from the total number of dates he had in the six weeks following treatment (A date was defined as any interaction with the op— posite sex in which the subject took initiative in the 38 airfranging for the interaction. It included taking a walk or $3't11dy or coke dates as well as more formal situations.); C 14-) client satisfaction as indicated on a rating scale by the I>aazrticipants following treatment. Based on the discussion and rationale presented in (Zlneapter I, the following research questions were formulated ifc>r~testing: Do the experimental treatments make a differ— €3rice on the anxiety criterion measures when adjustment is Inade for initial differences in the anxiety of the subjects? [kre there post-treatment differences on the anxiety criterion Ineasures between those subjects who have had same—sexed 'helpers and those who have had opposite—sexed helpers when adjustment is made for initial differences in the anxiety of the subjects? Does a combination of a particular treatment and sex of helper make a difference on the anxiety criterion measures when adjustment is made for initial differences in the anxiety of the subjects? Do the subjects in the experi— mental conditions differ from the wait-control group in the frequency of their dating behavior following treatment? Is there a difference in dating behavior or in client satisfac— tion between those participants who have female helpers and those who have male helpers? Is there an interaction between sex of helper and the treatment groups? The hypotheses were stated in the null rather than in the directional form because no body of data suggested reason 39 1:c> believe that: (1) same or Opposite-sexed helpers would t>ea :more effective, (2) the experimental treatments would nneaJ.N mm.m mm.NH Hm. mm. mm.am ms.“ :m.om sm.s: x Hayes so. so. mm.m H:.H mm.m mm.NH mm is so. so. NH.NN om.m mm.am oo.m: x om. ms. om.m mm.a mm.s mm.HH mm ms oz. om. om.HN OH.~ om.ma oa.m: x mm. Hm. oo.m sm.a ms.m mm.ma mm NH em. as. om.HN om.~ om.am o:.Hm x as. Hm. mm.m oo.H :s.: :m.NH mm as so. 0H. os.mm om.~ o:.mH oH.Hm x mxmmz Hum mxmmz :-o ucmapmmsa om< m EmpH ow hofihm mmpmm OOZ >poflxc< oamaoomm D >pmflxc< owmaommm hpmflxc< Hmsmcmw H.N mqmp< copmflq pGoEDMmQH H w H Ape wmammw omsmonmm . pmmmmdm xamH cospo< mHZMZBuoc oou uoa unmoH um uon .hmmoas usn hmwoso Huo> "Ammocs mHUsmuuxm m. m N m m c m N H .wmoem m>OH OH mHmMOHH mom ozHHH<3 mzHH UZOH < zH oanzo woo .uo: EUHa uoo ow ou .xUHm ou Gunshow waonoH EUHaowo o: .uaNUHmos uo: MHomha ouuow HH.H 8.H was HHuu HH.H ...oan Hoow .m=o>umc oou you E.H uon .zmwmcs usn .xmooaa auo> Hummus: xHoawuuxU m w . N e m e n N H .mmommm mm: :HHS mH¢m Dow Ham .m>HHOm mmfi mom mHum: oou uoc E.H won .hmwmas Una .xmmocs >um> H "mamas: mHuawuuxo a w n o n e m N H .HMHO mHmH :HHS MH mZHH HmmHm mmH mH mHMH .mHmwoc OOu uoc u.:03 asp .hmaoao HH.H uon .hmmwcs zuo> Hammad: hHUaouuxo m m N w m c n N H .mH QHDO3 DOW 20x3 024 msz 20:3 HmHo UZHHmm Hmuos ecu uoa uoc use .zmmwcs HH.H use .xmwmco muo> Hammond hHmaouuxo m m N c m e m N H .mmm IHHZ onHzou < Hm mzHOH mm zomz HMHO < 02¢ «(m xoH> mm mcoHum:UHm HmUHumsuoo%z wcH3OHHow msu UNHHm=mH> Cam mmho woo» omoHo mq