INTERNAL EXTERNAL CONTROL, OIRECTIVENESS ANO STATUS OF TIIERAPISTS, ANO OUTCOME IN ‘ PSYCHOTHERAPY Thesis for the Degree Of M. A . MICHIGAN STATE UNIVERSIW ' OAVIO MICHAEL RUBIN 1975 . o . 4 - ABSTRACT INTERNAL-EXTERNAL CONTROL, DIRECTIVENESS AND STATUS OF THERAPISTS, AND OUTCOME IN PSYCHOTHERAPY BY David Michael Rubin The purpose of this study was to test the hypotheses that (1) clients beginning psychotherapy with an internal locus of control orientation would be more successful with nondirective therapists, while (2) those beginning with an ex- ternal locus of control orientation would be more successful with directive and (3) high status therapists. Client locus of control was measured by Rotter's I-E scale (1966). Therapist directiveness-nondirectiveness was measured by a method of rating written transcripts of counseling interviews developed by Aronson (1951). Outcome measures included: (1) Pre-post therapy differences in I-E; (2) Pre—post differences on the Number of Deviant Signs of the Tennessee Self Concept Scale; and (3) Therapist ratings of how helpful therapy was for the client. Twenty-two therapist-client diads were studied. The clients were undergraduate students at Michigan State Univer- sity being seen at the Counseling Center by either Ph.D. David Michael Rubin psychologists or by advanced psychology graduate student interns. Subjects were split at the median in order to desig- nate clients as either internal or external and therapists as either directive or nondirective. The results of the statistical analyses indicated the following: (1) Internals were not significantly more successful with nondirective than with directive therapists; (2) Externals were not significantly more successful with directive therapists than with nondirective therapists. Indeed, contrary to predictions, an inspection of the data indicates that externals were actually more successful with nondirective therapists than with directive therapists; and (3) Externals were not significantly more successful with senior staff than with junior staff. Various explanations of the above results were discussed. First of all, there was some question as to the validity of the Aronson scale for differen- tiating between directive and nondirective therapists and the method by which it was used in the present study. Further- more, the small sample size did not permit a strong test of the hypotheses and the examination of more discrete, less overlapping groups of subjects. Also, differences between the present and previous studies relating I-E to the effects of status on influence change may account for externals not having worked more successfully with supposedly high status therapists than with supposedly low status therapists. Impli- cations of these results in terms of past theories and David Michael Rubin research were discussed, as well as suggestions for future research. Lastly, this study suggests that, contrary to predictions that were based upon past I-E research, externals may actually be more successful with nondirective than with directive therapists. INTERNAL-EXTERNAL CONTROL, DIRECTIVENESS AND STATUS OF THERAPISTS, AND OUTCOME IN PSYCHOTHERAPY BY David Michael Rubin A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1975 AC KNOWLEDGMENTS Sincere appreciation is extended to my chairman, Gordon Williams for his help, direction, and support throughout the long process involved in develOping, executing, and writing this thesis. Thanks are also offered to Larry Messe, especially for his time and advice on the statistical analysis, and to Dozier Thornton for his concern and sugges- tions at various stages during the preparation of this study. In addition, I would like to acknowledge with gratitude Linda Giacomo and Thomas Negri for their help in coding the therapy transcripts. ii LIST OF TABLES . . . . . . . . . . . . . . . . . TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . . . . . . . . . Relationship to Pathology . . . . . . . . . . . . Reaction to Influence Attempts . . . . . . . . . Relationship to Psychotherapy . . . . . . . . . . The Present Study . . . . . . . . . . . . . . . . METHOD . . . . . . . . . . . . . . . . . . . . . . . Subjects . . . . . . . . . . . . . . . . . . . . Materials . . . . . . . . . . . . . . . . . . . . Procedure . . . . . . . . . . . . . . . . . . . . RESULTS 0 C O O O O O O O O O O O O O O O O O O O O C Hypotheses 1a and 2a . . . . . . . . . . . . . . Hypotheses lb and 2b . . . . . . . . . . . . . . Hypotheses 1c and 2c . . . . . . . . . . . . . . Hypotheses 3a and 4a . . . . . . . . . . . . . . Hypotheses 3b and 4b . . . . . . . . . . . . . . Hypotheses 3c and 4c . . . . . . . . . . . . . . DISCUSSION 0 O O O O O O O O O O O O O O O O O O O O 0 Sample Size . . . . . . . . . . . . . . . . . . . Limited Range of Therapeutic Styles . . . . . . Counselor Techniques in Therapy Scoring System Weaknesses in Experimental Procedure . . . . . Underlying Assumptions . . . . . . . . . . Lack of Significant Differences of Internals With Both Senior and Junior Staff . . . . . . . . Differences Between Ritchie and Phares' Study and the Present Study . . . . . . . . . . . . . Incidental Findings and Tentative Implications . Areas for Future Research . . . . . . . . . . . . iii Page [—1 NKDGN 19 20 21 23 23 25 27 29 32 34 38 38 39 40 42 . 42 45 45 49 50 iv Page APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . 52 A. Social Reaction Inventory . . . . . . . . . . . . 52 B. Therapist Rating Scale . . . . . . . . . . . . . 56 C. Definitions and Instructions for Counselor Coding Categories . . . . . . . . . . . . . 57 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . 64 Table l. 8. 9. 10. ll. 12. LIST OF TABLES Analysis of Variance of Decrease in I-E: Group (Initial I-E) by Treatment (Directive- Nondirective) . . . . . . . . . . . . . . . . Group Means . . . . . . . . . . . . . . . . . . . Analysis of Variance of Decrease in NDS: Group (Initial I-E) by Treatment (Directive-Non- directive) O O D 0 O O O O O O O O O C O O 0 Group Means . . . . . . . . . . . . . . . . . . . Analysis of Variance of Therapist Ratings: Group (Initial I-E) by Treatment (Directive-Non- directive) O O O O O C O O O O O O O O I O 0 Group Means . . . . . . . . . . . . . . . . . . . Analysis of Variance of Decrease in I-E: Group (Initial I-E) by Treatment (Staff Level) . . Group Means . . . . . . . . . . . . . . . . . . . Analysis of Variance of Decrease in NDS: Group (Initial I-E) by Treatment (Staff Level) . . Group Means . . . . . . . . . . . . . . . . . . . Analysis of Variance of Therapist Ratings: Group (Initial I-E) by Treatment (Staff Level) . . Group Means . . . . . . . . . . . . . . . . . . . Page 24 25 27 27 29 29 31 31 33 34 35 INTRODUCTION Within recent years, the construct known as Internal- External Locus of Control (I-E) has been researched by literally hundreds of investigators (Rotter, 1966; Lefcourt, 1966; Throop and MacDonald, 1971; MacDonald, 1972a, and 1972b). It is based largely upon Julian B. Rotter's social learning theory (1954). According to Rotter, a reinforcement of a particular event or behavior acts to strengthen the expec- tancy that the event or behavior will be similarly reinforced in the future. As a result of an individual's personal life history, it follows that he would have a certain degree of expectancy that his own actions will or will not lead to reinforcements. Rotter sees this expectancy as falling on a continuum from internal to external control of reinforcement. Individuals at the external end perceive reinforcements which follow their actions as not being necessarily contin- gent upon their actions, but rather resulting from fate, chance, luck, or the actions of powerful others. On the other hand, the more internal individual views his reinforce- ments as being contingent upon his own behaviors and actions. (The higher the I-E score is, the more external is the individual.) The concept of internal-external control has been related to an extremely wide range of phenomena. The three 1 which are of particular interest to the present study include the relationships between locus of control and (1) pathology, (2) reactions to influence attempts and (3) psychotherapy. Relationship to Pathology The first area to be reviewed, that of the relationship between internal—external locus of control and pathology, has been examined by many researchers. It has been consis- tently found, throughout the I-E literature, that neurotic or pathological behaviors are associated with a characteris- tically external control. The more externally oriented the individual is, the more likely it is that he is less well- adjusted, and is exhibiting deviant behaviors. Indeed, it follows logically that, in general, those individuals who feel that they have little control over what happens to them are those who are typically labelled as maladjusted. Lefcourt (1966a, p. 191) points out that "many forms of deviant behavior recognized as symptoms of psychOpathology may profitably be described as resulting from a disbelief that efforts to behave in socially constructive, approved ways would be successful." In general, internals are seen as having more ego strength, while externals are seen as exhibiting more pathology. For example, Burnes, Brown and Keating (1971) found, among 25 rescue squad workers between the ages of 17 and 30, that I-E correlated significantly with three MMPI scores (F, K and Hy). Externals were thus found to be low in ego strength and high in pathology. According to Burnes, Brown and Keating, "the I—E scale's correlations with MMPI scales suggest that a sense of control over external events is related to self control and competence in handling internal events" (p. 301). Similarly,in a study involving 225 male veterans, Palmer (1971) compared the I-E orienta- tion of patients who had been hospitalized for psychiatric reasons to those who had been hospitatlized for medical (non-psychiatric) reasons. He found the former group to be significantly more external than the latter group (Efl<°02)' Among the psychiatric patients, those who had demonstrated the lowest social competence expressed significantly greater externality than those who had demonstrated the highest social competence (p <.02). In a study involving 648 introductory psychology students, Watson (1967) found a significant positive correlation (p}<.01) between reports of anxiety, as measured by the Taylor Manifest Anxiety Scale and an external locus of control orientation. He also found a significant positive correlation between I-E and the Debilitating Anxiety Subscale of the Alpert-Haber Achievement Anxiety Test (Efi<.01). Like- wise, Ray and Katahn (1968) found among two large samples of introductory psychology students that locus of control was positively correlated to the Manifest Anxiety Scale and also to a Test Anxiety Scale (p<:.01). Furthermore, I-E has also been found to correlate significantly in a positive direction with anxiety as measured by the Pt scale of the MMPI (Hersch and Scheibe, 1967). Hersch and Scheibe also found I-E to correlate significantly and postively with maladjustment on the Incomplete Sentences Blank. Wareheime and Foulds (1971) found an internal locus of control to be associated with personal adjustment (p‘<.05) among introductory psychology students as measured by the Internal Support Scale of the Personal Orientation Inventory. However, the relationship was not significant when males were examined separately, although it was significant for females (pg<.01). Ducette and Wolk (1972) found a general indication among 173 freshmen at a girls suburban high school, that externals are typically more deviant or extreme in their behaviors. They concluded that externality not only implies a belief in the fact that one's behavior is under external control, but that it also implies that one prefers it that way and will work to attain such an end. According to Ducette and Wolk, externals thus place themselves in situations where they can have little information about how much control they can have over their fate. High self-esteem is generally associated with a more internal locus of control orientation. In one study, Fish and Karabenick (1971) examined 285 males enrolled in an intro- ductory psychology class. They found that, individuals who scored as being high in self-esteem on the Janis and Field Feelings of Inadequacy Scale, were more internally oriented (p}<.OOl). Similarly, Fitch (1970) studied the self-esteem of 135 undergraduate business students. He found that U1 individuals with low self-esteem, as measured by the Tennessee Self-Concept Scale, tended to score toward the external end of the I-E continuum (p;<.05). Lefcourt (1966) points out that externals lack con- fidence. In a related study, Harrow and Ferrante (1969) examined 128 consecutive admissions to a short-term psychia- tric inpatient service. The major types of psychiatric dis- orders generally necessitating hospitalization were repre- sented, including 40 schizophrenics, 41 depressives, 34 character disorders, 5 manics, and 8 of varied diagnoses. Using a self-confidence scale devised by Brim, Glass, and Lavin, Harrow and Ferrante found I-E to correlate significant- ly in a negative direction with self-confidence (pp<.02). They also found I-E to correlate significantly and positively (E<:.01) with a scale for measuring frustration over previous failures. Externals have been found to have a greater preference for immediate gratification (Bialer, 1961). Furthermore, in a study of 104 graduate students in education, Baker (1971) found highly external individuals to exhibit a signi- ficantly higher degree of escapism than highly internal individuals. In a related finding, Abramowitz (1969) found externals to be more depressive than internals. By examining 235 introductory psychology students, Williams and Nickels (1969) found externality to be directly related (E<:.01 for males, p}:.05 for females) to suicide potentiality, as measured by Farberow and Devries' MMPI Suicide Scale. Thus, an external locus of control orientation has been found to be associated with deviant or pathological behaviors low social competence, poor personal adjustment, high anxiety and low self-esteem. Internals on the other hand have been shown to be high in ego strength, social competence, personal adjustment and self-esteem. Reaction to Influence Attempts The second general area to be reviewed in the present study is that dealing with the relationship between I-E and reaction to influence attempts. Lefcourt (1972) points out that individuals, who feel themselves to be responsible for their own fates, are probably more cautious about accepting advice and influence from others. Those who do not see themselves as being in control of their lives would be less cautious. Odell (1959), for example, found a significant relationship between Rotter's I-E Scale and Barron's Inde- pendence of Judgement Scale. Those showing a greater likeli- hood of conformity were found to be more external. In a similar vein, Crowne and Liverant (1963) studied conformity among 110 introductory psychology students. They found ex- ternals to be less confident in their own judgements. They report that, in an Asch-type task, externals wagered less money than internals on the correctness of their judgements, when making independent rather than conforming judgements. Externals also conformed more to group judgements than did internals. The authors concluded that the external's low evaluation of himself and his fear of social rejection result in a strong disposition on the external's part to conform. Another related study was performed by Hjelle and Clouser (1970), who attempted to induce attitude change among 64 undergraduate students. The experimenters found, on a broad range of current campus relevant issues, that externally oriented individuals were more susceptible to attitude change than were internally oriented individuals. Biondo and MacDonald (1971) also examined response to influence attempts among 144 undergraduates. They found, under involving conditions (a proposed grading procedure), that externals conformed to both high and low levels of influence. Internals were not responsive to low levels of influence and actually reacted against high levels of influence. Furthermore, they point out that "It is crystal clear that internals did not manifest reactance to the low or subtle influence in the experiment" (p. 415), but rather only to the high influence attempts. Getter (1966) also found externals to be more sug- gestible and dependent on cues from other persons than were internals. In that study, subjects were given a contrived oral test of abstract ability in which responses were rein- forced during acquisition trials (Subtest I). No reinforce- ments were given to responses which were subsequently given (Subtest II). It was found that externals learned to give answers which would be reinforced and continued to do so, even after reinforcements were no longer present. Internals did not give the "correct" answer during the reinforcement period (Subtest.I)but did give the correct answer once rein- forcement had been discontinued (Subtest II). This suggests that the internals viewed reinforcement as a manipulation and thus chose not to give the correct answer until the reinforce- ment had been discontinued. Strickland (1962) studied reactions to influence attempts by utilizing a verbal conditioning paradigm. She discovered, on the basis of post-experimental inquiries, that internals who were more aware of the reinforcement contingency (the experimenter saying mmm-hmmm) failed to condition significantly less often than externals who were aware. Doctor (1971) found that externals, who were selectively reinforced in a sentence construction task, evidenced signi- ficantly greater performance gains than internals. It was found that internals were nonresponsive or resistive to influence. Similarly in still another verbal conditioning experiment, it was found that internals were more likely to deny influence by the experimenter and in some instances were more resistant to extinction than were externals (James and Rotter, 1958; Strickland, 1970). Ritchie and Phares (1969) studied attitude change among students enrolled in a general psychology class. One group received information from a supposedly "high-prestige" source. The other group received the same information from a supposedly "low prestige" source. The information (national budget expenditures) was considered not to be viewed as important by the subjects. Externals were found to exhibit significantly less change than internals in the low prestige condition and more changes than internals in the high prestige conditions (p:<.05). Externals changed morezhxthe high than inthe low prestige conditions (EW<°025)° The amount of Opinion change among internals did not vary according to the status of the influencer. Under high prestige conditions, externals exhibited significantly greater change than did internals (p< .05). Thus, it has consistently been found that individuals with an external locus of control are typically more suscep- tible to the influence of others than are individuals with an internal locus of control. Indeed, internals often react negativistically to influence attempts, changing their opinions and beliefs in the direction opposite to that advo— cated by an outside source. Relationship to Psychotherapy The third general area to be discussed in the present paper is that of the relationship between Internal-External control and psychotherapy. As Lefcourt (1966) points out, perceived control should have some importance as a goal for psychotherapy. A number of studies have shown clients to become more internal as a result of the psychotherapeutic experience. Gillis and Jessor (1970) studied 29 psychiatric patients, who could be contacted within one week of admission, and who had no more than one previous psychiatric lO hospitalization. Thirteen of these received some form of individual or group therapy for a minimum of 10 weeks. The remining 16 did not receive psychotherapy. Among those who received therapy, those who were judged as having shown some or marked improvement also had become significantly more internal than they had been before therapy (pg<.05). The therapy group also became significantly more internal than did the no therapy group (p;<.05). In another study, Dua (1970) looked at first year university females who had come to a counseling center and who had expressed concern about their ability to interact with others in interpersonal situations. Clients in the experimental groups, a "reeducation program" which focused on cognitive processes and verbal interactions, and a behaviorally oriented action program, were found to have become more internal at the end of therapy than had a control (non-therapy) group. It is interesting to note that the behaviorally oriented group had a significantly greater change in the internal direction than did the reeducation group (pg<.05). Furthermore, when analyzed separately, although the former group differed significantly from the controls (Ef‘°05)' the latter did not. During a summer camp experience, an attempt was made to affect locus of control scores of deprived inner city adolescents (Nowicki and Barnes, 1972). The camp was des- cribed as being highly structured with emphasis placed on contingent reinforcement for good and poor performance. 11 There was an overall change in I-E in the internal direction (2 <.002). Similarly, Lesyk (1969) found female schiZOphre- nics to become more internal after five weeks on a token economy, operant conditioning ward. Changes in locus of control have also been examined as a function of life crisis resolution. Smith (1970) studied 30 patients (mean age = 28 years) who appeared at a neuropsychiatric emergenCy room because of acute life crises. These patients received six weeks of intensive treatment which focused on helping the individual to learn and adopt more effective coping mechanisms so that he might gain more positive control over his current life situation. The change in I-E in the internal direction, between pre- and post-intervention measures, was significant for the crisis group (p <.01). A non-crisis control group showed no significant changes in I-E. Although it has been generally found that clients have become more internal at the end of therapy than they were at the beginning, Pierce, Schauble and Farkas (1970) did not find the change in I-E at the end of therapy to be a statistically significant one. Other interesting data were found by Farkas (1969). She did not find support for her prediction that clients who were initially internal would be more successful in therapy than clients who were initially external. There has been some evidence that internals perceive various forms of psychotherapy in a different manner than do externals. Under modified desensitization relaxation 12 techniques, internals view the experimenter as dominating, while externals view him as loving (Nowicki, Bonner, and Feather, 1972). Helweg (1971) studied the relationship between I-E and perceptions of directive and nondirective psychotherapeutic approaches among 77 college undergraduates and 77 hospitalized psychiatric patients. Subjects were presented with sound—film recordings depicting therapy inter- views typical of the directive (Ellis) and nondirective (Rogers) approach. It was found that both students and patients who prefer a directive therapeutic approach are more external than those who prefer a nondirective approach. Both students and patients who prefer a nondirective therapeutic approach were found to value independence, as a basis for relating to others, more than did those who prefer a directive approach. Jacobson (1971) in a study of 100 undergraduates found that subjects who chose behavior modification techniques were more external than were those who chose more analytically oriented therapy. Thus, it has been found that individuals tend to increase in the expectancy for internal control as a result of having participated in some form of psychotherapy or behavior change projects. Internals seem to prefer a more nondirective approach, while externals seem to prefer a more directive approach . 13 The Present Study The present study attempted to relate to one another, in a meaningful manner, the three areas discussed above. Since one's locus of control orientation does affect the ways in which one relates to life expriences, it may very well significantly affect the ways in which one deals with and behaves in therapy. Therefore, in a general sense, it seems that the client's I-E orientation would have some effect upon the therapeutic process and hence, upon success in therapy. It might be predicted that one's locus of control, in inter- action with the therapist and therapy process, would signifi- cantly affect his behaviors in and reactions to therapy. More specifically, it would be expected that an indivi- dual's typical reactions to external influence and suggestion would influence his actions and feelings in therapy. The research, described above, relating I-E to reactions to influence attempts may thus be quite relevant to the study of the therapeutic process. Since externals are generally more conforming, one would predict that an individual beginning therapy as an external would be more likely to go along with and trust the verbal statements and communications of his therapist. An internal, on the other hand, would react nega- tively towards any indication that the therapist was trying to influence him. It was predicted that clients who begin therapy as more internal will be more resistant to therapists who come across as more influencing and directive. Externals, 14 on the other hand, should do very well with just such types of therapists. Gains from therapy, or success of outcome was assessed by three measures. The first was computed by how much the client's I-E score changed in the course of therapy. As has been previously discussed, a more external locus of control has generally been shown to coincide with more pathological behaviors and personality characteristics. On the other hand, an internal locus of control has been associated with high ego strength and better social adjustment. Thus, an increase in Umeexpectancy for internal control can be used as one measure of successful therapeutic outcome. The second measure of success was taken from the Tennessee Self Concept Scale (TSCS), an instrument which consists of 100 self-descriptiVe statements which the sub- ject rates on a scale from "completely false" to "completely true" of himself. The TSCS was given to each client at the beginning and end of therapy. Degree of improvement or success in therapy was measured by the difference betweeen the pre- and post-therapy scores on the scale's index of psycho— logical disturbance, the Number of Deviant Signs (NDS). According to Fitts (1965), the N08 is the scale's best index of psychological disturbance, identifying deviant individuals with about 80% accuracy. The difference between pre- and post-terapy NDS scores has been successfully used as a measure of client improvement through psychotherapy (Ashcraft and Fitts, 1964). 15 The third measure was the therapist's rating on a seven-point scale of how helpful therapy was for the client (Appendix B). Each client was rated by his therapist at the end of therapy. In order to assess degree of non-directiveness of the therapist, Counselor Techniques in Therapy, a scale developed by Aronson (1951), was used as the basis for rating. The scale measures the type of verbal behavior used by therapists in therapy and specifically refers to the use of directive and non-directive methods. The responses which go into these categories are defined in Appendix C. A single score for degree of non—directiveness can be computed. Thus, the main hypotheses of the study were as follows: Hypothesis la - For internals with nondirective thera— pists, success in therapy as assessed by an increase in the expectancy for internal control is greater than for internals with directive therapists. Hypothesis 1b - For internals with nondirective thera- pists, success in therapy as assessed by a decrease in NDS is greater than for internals with directive therapists. Hypothesis lc - For internals with directive therapists, success in therapy as assessed by therapist ratings is greater than for internals with directive therapists. Hypothesis 2a - For externals with directive therapists, success in therapy as assessed by an increase in the expectancy for internal control is greater than for externals with non- directive therapists. 16 Hypothesis 2b - For externals with directive therapists, success in therapy as assessed by a decrease in NDS is greater than for externals with nondirective therapists. Hypothesis 2c - For externals with directive therapists, success in therapy as assessed by therapist ratings is greater than for externals with nondirective therapists. Additional hypotheses related locus of control, success in therapy, and therapist status to one another. As has been previously discussed, Ritchie and Phares (1969) found externals to change their attitudes significantly more when receiving information from a high prestige source than from a low prestige source. Internals did not differ significnatly in their reactions to high and low prestige sources in that study. One might predict, then, that externals would be more influenced by and more successful with therapists whom they perceived as being more prestigious than by those whom they perceived as being somewhat less prestigious. Internals would not be expected to differ according to the status of the therapist. It was assumed that senior staff members, who held the Ph.D. degree and were generally somewhat older, would be perceived as having a higher status than would graduate student interns. Hypothesis 3a - Internals do not differ significantly in the degree of success in therapy, as assessed by an increase in the expectancy for internal control, between those seeing senior staff members and those seeing junior staff members (interns). 17 Hypothesis 3b - Internals do not differ significantly in the degree of success in therapy, as assessed by a decrease in NDS, between those seeing senior staff and those seeing junior staff. Hypothesis 3c - Internals do not differ significantly in the degree of success in therapy, as assessed by therapist ratings, between those seeing senior staff and those seeing junior staff. Hypothesis 4a - Externals are significantly more success- ful in therapy, as assessed by an increase in the expectancy for internal control, with senior staff than with junior staff. Hypothesis 4b - Externals are significantly more success- ful in therapy, as assessed by a decrease in NDS, with senior staff than with junior staff. Hypothesis 4c - Externals are significantly more successful in therapy, as assessed by therapist ratings, with senior staff than with junior staff. These hypotheses are important for a number of reasons. Previous studies have shown internals and externals to differ as to the type of therapy which they prefer. However, none has examined the relationship between (1) success in therapy, and (2) the interaction of client I-E and degree of directive- ness of the therapist. If support were indeed found for the hypotheses of the present study, a relatively simple means of facilitating the "matching" of clients and therapists would 18 be available for agencies which offer psychotherapy. Also, therapists might modify their own behavior somewhat according to the client's I-E so as to be maximally effective with their clients. Another purpose in executing this study was to clarify some past research findings. Farkas (1969) and Pierce, Schauble and Farkas (1970) did not find a relationship between initial I-E, as measured by the Rotter Scale, and therapeutic outcome. By examining the possible effects of the interaction of clients I-E and directiveness of the therapist upon success in therapy, seeming inconsistencies between previous research studies might be clarified. METHOD Subjects The 22 cases to be examined in the study are part of a tape recording library which has been compiled at the Michigan State University Counseling Center. The tapes record the total therapeutic experience of 75 Michigan State University under- graduate students who came voluntarily to the Counseling Center, between 1967 and 1969, requesting help with personal and social problems. Much testing and evaluation of clients was performed before, during the course of, and at the end of therapy. The specific criteria for subjects to be used in the present study include: (1) participation in pre- and post-therapy testing and (2) therapy having been conducted by one therapist who worked with the client through termination. Client S's consisted of six males and 16 females, who were seen by 23 psychotherapists at the Counseling Center: i.e., no two students were seen by the same therapist. The therapist group consisted of 11 senior staff members who held the Ph.D. degree in either clinical of counseling psychology; five second year interns; and six first year interns (all interns were completing training for the Ph.D. in clinical or counseling psychology). Fifteen therapists were male and seven were female. 19 20 Materials The Internal-External Locus of Control Scale is a 29- item forced-choice test, including six filler items which are intended to make the purpose of the test somewhat more ambiguous (Appendix A). Extensive normative, reliability, and validity data have been obtained on various populations (Rotter, 1966). The Number of Deviant Signs (NDS) is a subscale of the Tennessee Self-Concept Scale (TSCS), which consists of 100 self-descriptive statements. The subject uses these to por- tray his own picture of himself by rating the item's degree of applicability on a five-point scale which ranges from "completely false" to "completely true." The items of the TSCS are organized into a number of subscales, such as the neurosis scale and the personality disorder scale. The NDS score is a count of deviant features on all subscale scores. Reliability and validity data are reported by Fitts (1965). The therapist's rating of success of therapy was obtained as part of the post-therapy questionnaires given to each counselor. It consists of a seven-point Likert type scale ranging from extremely harmful to the client to extreme- ly helpful to the client (Appendix B). Counselor Techniques in Therapy (Aronson, 1951) is a method of rating therapy interview transcripts, which assesses the type of verbal behavior used by the therapist in therapy and Specifically refers to the use of directive and non- directive methods. The responses which go into and define 21 those categories are presented in Appendix C. Aronson found reliability for the directive and non-directive cate- gories from a reliability sample in which seven coders independently classified the counselor statements in three sample interviews. The method of average intercorrelations was used. The average intercorrelation for directive tech- niques yielded a coefficient of .66 with an estimated reliability of .98. For non-directive techniques the coef- ficient was .88 with an estimated reliability of .98. The directive and non-directive scores can be combined to obtain a single corrected non-directive score. Procedure All of the client data had been obtained as part of the Counseling Center tape library. It included I-E scores (pre- and post-therapy), Tennessee Self Concept Scale scores (pre- and post-therapy), the therapist's rating of success in therapy, and tape recordings of therapy sessions. The initial locus of control for each client was assessed by his score on Rotter's I-E scale. Those above the median (11.5) were designated as being external, while those below the median were designated as being internal. In order to assess the degree of directiveness-non- directiveness of the therapist, the following procedure was used. For each client, written transcripts of four two- minute tape-recorded segments of the first, middle and last therapy sessions were transcribed (twelve segments per client). 22 In those cases in which the tape of a particular session was either missing or inaudible, the tape of the preceding or following session was used. The transcripts were rated according to the categories defined by Aronson (1951) in the Counselor Techniques in Therapy (Appendix C). Raters were two graduate students in clinical psychology. The various categories described by Aronson were given to the raters in written form and were also discussed with the raters by the experimenter. Raters were asked to code sample, practice transcripts over a two day period. Each rater then rated all twelve segments for each client-therapist diad. The scores for the segments were added together so that each client-therapist diad had two overall scores for directive- ness-nondirectiveness (one made by each rater). Inter- rater reliability was assessed by means of a Pearson product- moment correlation (r = +.78). The ratings made by each rater were then averaged in order to obtain one directiveness- nondirectiveness score for each client-therapist diad. These scores were then split at the median, resulting in each client being designated as having obtained either directive or nondirective therapy. RESULTS Hypotheses 1a and 2a Pre and post scores on Rotter's I-E scale were the dependent variables relevant to hypotheses 1a and 2a. Independent variables were the directiveness-nondirectiveness of the therapist and the initial I-E of the client. A 2 (directiveness-nondirectiveness) x 2 (initial I-E) x 2 (pre and post I-E) analysis of variance with repeated measures on the third variable was used to compare groups' pre and post treatment scores. Due to unequal group sizes, an unweighted means solution was used. Hypothesis 1a states that among clients who are ini- tially internal, success in therapy as assessed by an increase in the expectancy for internal control will be significantly greater for those seen by nondirective therapists than for those seen by directive therapists. Hypothesis 2a states that among clients who are initially external, success in therapy as assessed by an increase:h1the expectancy for internal control will be significantly less for those seen by non- directive therapists than for those seen by directive therapists. If these hypotheses are valid, results should yield both (a) a significant pre-post main effect (overall pre I-E scores > overall post I-E scores) and (b) a significant 23 24 interaction effect (decrease in I-E for internals with directive therapists < decrease hTI-E for internals with nondirective therapists and decrease in I-E for externals with directive therapists > decrease in I-E for externals with nondirective therapists). Table 1 shows that, although the pre-post main effect was in the predicted direction, it was only a trend (2 <.10). The interaction effect, however, did not even approach significance. Thus, neither hypothesis 1a nor 2a was supported. Table 1. Analysis of Variance of Decrease in I-E: Group (Initial I-E) by Treatment (Directive— Nondirective) Source SS df MS F A(Directive-Nondirective) 24.54 1 24.54 1.57 B(Initial I-E) 460.86 1 460.86 29.60** AB .25 l .25 .01 Subject within 280.30 18 15.57 C(Pre-post I-E) 19.88 1 19.88 3.49* AC .78 l .78 .13 BC 9.50 l 9.50 1.67 ABC 7.88 1 7.88 1.38 C x subject within 102.43 18 5.69 *p< .10 **B< .00004 25 Indeed, in contrast to hypothesis 1a internals increased their expectancy for internal control with directive therapists and actually decreased slightly in the expectancy for internal control after having seen nondirective therapists. Also, contrary to hypothesis 2a, externals experienced a greater decrease in I-E with nondirective than with directive therapists (Table 2). Table 2. Group Means Initial Pre I-E Post I-E Directive Internal 7.4 6.4 External 13.83 12.67 Nondirective Internal 8.17 8.34 External 16.6 13.2 Hypotheses lb and 2b Pre and post scores for each subject on the Number of Deviant Signs of the Tennessee Self Concept Scale were the dependent variables relevant to hypotheses 1b and 2b. Independent variables were the directiveness-nondirectiveness of the therapist and the initial I-E of the client. An analysis of variance similar to that performed for hypotheses 1a and 2a was performed with the exception that the third variable consisted of pre and post NDS scores. Hypothesis 1b states that among clients who are initially internal, success in therapy as assessed by a decrease in the 26 Number of Deviant Signs on the Tennessee Self Concept Scale will be significantly less for those seen by directive than by nondirective therapists. Hypothesis 2b states that among clients who are initially external, success in therapy as assessed by a decrease in the Number of Deviant Signs Scale will be significantly greater for those seen by direc- tive than nondirective therapists. If these hypotheses are valid, results should yield both (a) a significant pre-post main effect (overall pre NDS scores > overall post NDS scores) and (b) a significant interaction effect (decrease in NDS for internals with directive therapists < decrease in NDS for internals with nondirective therapists and decrease in NDS for externals with directive therapists > decrease in NDS for externals with nondirective therapists). Table 3 shows that the pre-post main effect was sig- nificant in the predicted direction (p <.05). The interaction effect, however did not even approach significance. Thus, neither hypothesis 1b nor 2b was supported. Internals with both types of therapists decreased almost identical amounts on the NDS. This result is somewhat confounded by the fact that internals with nondirective therapists were initially somewhat higher on the NDS. Contrary to hypothesis 2b, externals working with directive therapists again actually had less of a drop in NDS than did externals working with nondirective therapists (Table 4). Table 3. 27 Analysis of Variance of Decrease in NDS: Group (Initial I-E) by Treatment (Directive-Nondirective) Source SS df MS F A(Directive-Nondirective) .16 1 .16 .001 B(Initia1 I-E) 926.76 1 926.76 7.11* AB 46.61 46.61. .36 Subject within 2345.17 18 130.28 C (NDS) 791.34 1 791.34 6.88* AC 15.63 1 15.63 .14 BC 241.22 1 241.22 2.10 ABC 15.27 1 15.27 .13 C x subject within 2069.83 18 114.99 E<:.05 Table 4. Group Means Initial Pre NDS Post NDS Directive Internal 16.0 12.2 External 26.67 15.83 Nondirective Internal 13.83 10.0 External 31.0 15.4 Hypotheses 1c and 2c A rating of success in therapy made for each subject by his therapist was the dependent variable relevant to hypotheses 1c and 2c. Independent variables were the directiveness-nondirectiveness of the therapist and the 28 initial I-E of the client. A 2 (directiveness-nondirective- ness) x 2 (initial I-E) analysis of variance was used to compare groups' post-treatment scores. Due to unequal group sizes, an unweighted means solution was used. Hypothesis 1c states that among clients who are initially internal, success in therapy as assessed by therapist ratings will be significantly less for those seen by directive than nondirective therapists. Hypothesis 2c states that among clients who are initially external, success in therapy as assessed by therapist ratings will be significantly greater for those seen by directive than by nondirective therapists. If these hypotheses are valid, results should yield a significant interaction effect (success for internals seeing directive therapists < success for internals seeing nondirective therapists and success for externals seeing directive therapists > success for externals seeing non- directive therapists). Table 5 shows that the interaction effect did not even approach significance. Thus, neither hypothesis 1c nor 2c was supported. Indeed, contrary to hypothesis lc, internals were actually judged to be more successful with directive than nondirective therapists. On an absolute basis, externals were judged as more successful with directive than nondirective therapists, though the result did not even approach signi- ficance (Table 6). 29 Table 5. Analysis of Variance of Therapist Ratings: Group (Initial I-E) by Treatment (Directive-Nondirective) Source SS df MS F A (Directive-Nondirective) .001 1 .001 .001 B (Initial I-E) .19 1 .19 .22 AB .04 1 .04 .04 Subject within 15.23 18 .85 Table 6. Group Means Directive Nondirective Internal 5.6 5.33 External 5.6 5.4 Hypotheses 3a and 4a Pre and post scores on Rotter's I-E scale were the dependent variables relevant to hypotheses 3a and 4a. Inde- pendent variables were the staff level of the therapist (senior or junior) and the initial I-E of the client. A 2 (staff level) x 2 (initial I-E) x 2 (pre and post I-E) analysis of variance with repeated measures on the third variable was used to compare groups' pre and post treatment scores. Due to unequal group sizes an unweighted means solution was used. 30 Hypothesis 3a states that among clients who are initially internal, success in therapy as assessed by a decrease in I-E will not be significantly different for those seen by junior and senior staff. If this hypothesis is valid, results should yield both (a) a significant pre-post main effect (overall pre I-E scores > overall Post I-E scores) and (b) a non-significant interaction effect (decrease in I-E for internals with senior staff = decrease in I-E for internals with junior staff). Hypothesis 4a states that among clients who are ini- tially external, success in therapy as assessed by a decrease in I-E will be significantly greater for those seen by senior staff than for those seen by junior staff. If this hypothesis is valid, results should yield both (a) a signi- ficant pre-post main effect (overall pre I-E scores > overall post I-E scores) and (b) a significant interaction effect (decrease in I-E for externals with senior staff > decrease in I-E for externals with junior staff). Table 7 shows that, although the pre-post main effect was in the predicted direction, it was only a trend (p<=.10). The interaction effect did not even approach significance. Thus, neither hypothesis 3a nor 4a was supported. An examination of Table 8 shows that contrary to hypo- thesis 3a, internals did not change significantly in their expectancy for internal control with either senior or junior staff. Furthermore, contrary to hypothesis 4a, externals actually had a bigger decrease in I-E when working with junior than with senior staff members. Table 7. Analysis of Variance of Decrease in I-E: Group (Initial I-E) by Treatment (Staff Level) Source SS df MS F A (Staff Level) 2.64 2.64 .17 B (Initial I-E) 447.97 447.97 28.78* AB 21.63 21.63 1.39 Subject within 280.17 18 15.56 C (Pre-post I-E) 18.67 1 18.67 3.09** AC 1.40 1.40 .23 BC 9.66 9.66 1.20 ABC 1.68 1.68 .30 C x subject within 108.67 18 6.04 *p < .001 **pg<.10 Table 8. Group Means Initial Pre I-E Post I-E Senior Staff Internal 6.8 6.4 External 15.17 13.67 Junior Staff Internal 8.67 8.33 External 15.0 12.0 32 Hypotheses 3b and 4b Pre and post scores for each subject on the Number of Deviant Signs of the Tennessee Self Concept Scale were the de- pendent variables relevant to hypotheses 3b and 4b. Independent variables were the staff level of the therapist and the initial I-E of the client. An analysis of variance similar to that performed for hypotheses 3a and 4a was performed with the exception that the third variable consisted of pre and post NDS scores. Hypothesis 3b states that among clients who are ini- tially internal, success in therapy as assessed by a decrease in NDS will not be significantly different for those seen by junior and senior staff. If this hypothesis is valid, results should yield both (a) a significnat pre-post main effect (overall pre NDS scores > overall post NDS scores) and (b) a non-significant interaction effect (decrease in NDS for internals with senior staff = decrease in NDS for internals with junior staff). Hypothesis 4b states that among clients who are initially external, success in therapy as assessed by a decrease in NDS will be significantly greater for those seen by senior staff than junior staff. If this hypothesis is valid, results should yield both (a) a significant pre-post main effect (overall pre NDS scores > overall post NDS scores) and (b) a significant interaction effect (decrease in NDS for externals with senior staff > decrease in NDS for externals with junior staff). 33 Although the overall pre-post main effect was signifi- cant (Table 9), this was due to the change in NDS shown by externals. Examined separately, internals did not show a significant difference between pre and post NDS scores. Thus, since neither group of internals changed significantly as a result of therapy, hypothesis 3b was not supported. Also, contrary to hypothesis 4b, externals had a greater decrease in NDS with junior than with senior staff (Table 10). Table 9. Analysis of Variance of Decrease in NDS: Group (Initial I-E) by Treatment (Staff Level) Source SS df MS F A (Staff Level) .45 1 .45 .004 B (Initial I-E) 925.24 1 925.24 7.38* AB 121.71 1 121.71 1.00 Subject within 2269.62 18 121.09 C (NDS) 802.05 1 802.05 6.99* AC .36 1 .36 .003 BC 228.34 1 228.34 1.99 ABC 36.30 1 36.30 .32 C x subject within 2064.15 18 114.67 *p< .05 34 Table 10. Group Means Initial Pre NDS Post NDS Senior Staff Internal 14.2 8.2 External 29.5 18.0 Junior Staff Internal ‘ 15.33 13.33 External 27.6 12.8 Hypotheses 3c and 4c A rating of success in therapy made for each subject by his therapist was the dependent variable relevant to hypotheses 3c and 4c. Independent variables were the staff level of the therapist and the initial I-E of the client. A 2 (staff level) x 2 (initial I-E) analysis of variance was used to compare groups' post-treatment scores. Due to unequal group sizes, an unweighted means solution was used. Hypothesis 3c states that among clients who are initially internal, success in therapy as assessed by therapist ratings will not be significantly different for those seeing junior than for those seeing senior staff. If this hypothesis is valid, results should yield a non-significant interaction effect (success for internals with senior staff = success for internals with junior staff). Hypothesis 4c states that among clients who are initially externals, success in therapy as assessed by therapist ratings will be significantly greater for those seeing senior than 35 for those seeing junior staff. If this hypothesis is valid, results should yield a significant interaction effect (success for externals with senior staff > success for externals with junior staff). Table 11 shows that the interaction effect did not even approach significance. Thus, hypothesis 4c was not supported. Table 11. Analysis of Variance of Therapist Ratings: Group (Initial I-E) by Treatment (Staff Level) Source > SS df MS F A (Staff Level) 0.0 l 0.0 0 B (Initial I-E) 0.0 1 0.0 0 AB 0.01 1 0.01 0.01 Subject within 15.4 18 0.86 Indeed, there were no significant differences between any of the groups (Table 12). Table 12. Group Means Senior Staff Junior Staff Internal 5.6 External 5.5 5.6 36 The data for hypotheses 3c and 4c presented above brings into question the validity of the seven-point therapist rating scale. Since the range for all subjects was between four and seven, with a standard deviation of only .86, it would appear to be an inadequate means of differentiating between degrees of helpfulness of therapy. Furthermore, a Pearson product moment correlation between Number of Deviant Signs on the Tennessee Self Concept Scale, a well known and often used indication of success in therapy, and the therapist outcome rating yielded a non-significant E of .25. The therapist's rating scale therefore does not seem to be a reliable means of assessing success in therapy. Thus, although no differences were found between internals with senior or junior staff, hypothesis 3c was not supported. To summarize the results then, the hypotheses were not supported. This is partially due to the general lack of significant change made by the majority of clients as a result of therapy. Under none of the conditions did internals change significantly on either Rotter's I—E test or the Tennessee Self Concept Scale. Contrary to hypothesis 1, externals experienced greater change on the I-E and Tennessee scales with nondirective therapists than with directive therapists. Also, in opposition to hypothesis 4, externals changed a greater amount on the Tennessee scale with junior staff than with senior staff. Externals did not change 37 significantly on I-E with either senior or junior staff. Furthermore, the therapist rating scale has been shown to be a highly questionable outcome measure, thus making any conclusions drawn from that measure extremely tentative in nature. Incidental, post hoc findings will be examined in the discussion section. DISCUSSION A task of this discussion must be to answer why the four hypotheses were not supported and what these results meaanIterms of the hypotheses that internals would be more successful with nondirective therapists while externals would be more successful with directive therapists and with senior staff members. There are several possibilities as to why the hypotheses were not supported: 1. Small sample size. 2. Limited range of therapeutic styles. 3. Counselor Techniques in Therapy scoring system. 4. Weakness of experimental procedure. 5. Underlying assumptions. 6. Lack of significant differences for inernals with both senior and junior staff. 7. Differences between Ritchie and Phares' study and the present study. Sample Size The small number of subjects used in this study did not permit as strong a test of the hypotheses as one might nor— mally like. Because not all clients whose records were part of the Counseling Center tape library had participated in the testing necessary for the current experiment, only twenty- two counselor-client diads could be studied. This resulted 38 39 in an average of only 5.5 subjects per cell. A larger sample size would have permitted a stronger testing of the hypotheses. The small sample size also forced the experimenter to split client and therapist groups at their respective medians in order to designate clients as internal or external and therapists as directive or nondirective. As a result, subjects with fairly close raw scores had to be placed in different, Opposing groups. More discrete and less overlap- ping groups could have been examined had there been a larger and hence wider range of subjects. Limited Range of Therapeutic Styles The absence of support for the hypotheses related to the therapist's degree of directiveness-nondirectiveness may in part be a reflection of the general therapist population studied. Although there were some differences between counselors, the majority worked out of a client-centered, somewhat nondirective approach. More directive types of therapies such as behavior modification and rational emotive therapy were not widely represented among the population studied. As a result, it can be argued that even though certain of the therapists were more directive relative to the other therapists studied, they might, when compared to a wider range of possible therapeutic approaches and styles, be seen as more nondirective. 4O Counselor Techniques in Therapy Scoring System The validity of the "Counselor Techniques in Therapy" scoring system must be examined. In the original study in which the scale was used (Aronson, 1951), seven judges were asked to make independent ratings as to whether certain coding categories and their supportive descriptions (Appendix C) were directive, nondirective, or nonclassifiable in nature. [It should be noted that for purposes of discussion, the current author is making a distinction between Aronson's 21 coding categories (e.g., "Restatement of Content," "Accurate Clarification of Feelings," etc.) and the three scoring classifications (directive, nondirective, and nonclassifiable).] When five or more of the judges had agreed to the designation of a category of responses as either directive or nondirective, it was assigned to that classification. However, in the course of rating the transcripts in this study, it became apparent that certain minor, technical differences in the way in which a therapist made a statement could result in its being placed in a directive rather than a nondirective category and vice versa. Thus, although raters may have agreed reliably as to whether certain therapist responses should be classified as directive or nondirective, there may have been no major differences between some that were classified as directive and some that were classified as nondirective. 41 For example, Aronson distinguishes between two coding categories, "Accurate Clarification of Feelings" and "Inaccu- rate Clarification of Feelings." An accurate clarification would be one which "puts the client's feelings or affective tone in a clearer or more recognizable form; or any effort to show that the counselor is accurately recogninzing the feeling of the client's statement by understanding it" (p. 99). A response falling within this category would be designated as nonclassifiable. A problem arises, however, because the use of written transcripts makes it difficult to assess whether a counselor's statement is indeed accurate or inaccurate. For example, a statement such as "You seem kind of sad" could be categorized as either accurate or inaccurate clarification of feelings. The rater would have to make a decision as to whether the statement was indeed accurate or inaccurate. This decision might be difficult to make for a number of reasons. For example, although a client may verbally agree or disagree with the therapist's statement, the rater cannot really be sure whether the client is being truthful, is seeking approval, or is being resistant. Furthermore, the nonverbal cues and communications which may have contri- buted to the counselor's assessment of the client's affect are not available to the rater. Thus, a client may say, "No, I'm not sad“ and yet have a very sad expression on his face. Whereas the counselor would have access to the client's 42 sad expression, the rater would not. Thus, the rater might score a reasonably accurate therapist statement as being inaccurate. As a result of these types of rating difficulties the counselor's overall score for directiveness-nondirective- ness would be affected. Weaknesses in Experimental Procedure The procedure by which the Aronson scale was used in the current study may also have contributed to the lack of significant findings. The raters in the present study scored twelve, two-minute segments of the interviews for each counselor-client diad. On the basis of a two-minute segment the rater may not have had enough information on which to make an accurate decision between placing a response in two or more different coding categories. Since the raters did not know the content of the entire interviews, their ratings may have been adversely affected. For example, within the context of a two-minute segment, a statement might have appeared to be an "Interpretation" (directive), while in the context of the entire interview, it might have been more accurately described as a "Restatement of Content" (nondirective). Underlying Assumptions Another possible explanation for the lack of significant findings for hypotheses one and two is that the basic assump- tions underlying these hypotheses were in error. In other 43 words, although internals and externals have been found in the literature to react differently to influence attempts, these differences may not apply to the process of psycho- therapy. These hypotheses were based on the assumption that a directive therapist would be perceived by the client as trying to influence him more than would a nondirective therapist. The accuracy of this assumption must be called into question. It is quite possible that certain therapeutic styles which would be classified as directive in nature might not be viewed by the client as ones in which the therapist is trying to exert strong influence. This would be especially true in relation to the directiveness-non- directiveness scale used in the present study. For example, according to Aronson the statement "How do you feel about that?" would be classified as directive, even though the terapist might not be perceived as particularly trying to in- fluence the client. "Approval and Encouragement," another directive category which includes statements such as "That's fine" and "You bet," might not always be seen as the thera- pist's trying to sway or manipulate the client. It might be perceived as approval and encouragement for the client's having acted according to his own value system rather than simply having gone along with that of the therapist. Thus, there might not be a clear relationship between the directive- ness of a therapist's statement(s) and the perception on the part of the client of strong attempts to influence. 44 In a similar manner, one must consider whether there is really a clear relationship between a therapist's non- directive behavior and the perception of a lack of influence attempts. It can be argued that no matter how nondirective and non-value oriented a therapist might try to be, he is still, at some level, trying to influence the client. The fact that a therapist is working with a client implies that he is trying to affect some sort of change in that client. Thus in an extreme case, one can never really avoid the client's perceiving some attempt at influence on the part of his therapist. The relevant literature previously discussed may also help to clarify the lack of significant results for hypotheses one and two. Biondo and MacDonald (1971) for example, found that having received high influence messages, internals would move in a direction opposite to the one advocated by the persuader. However, in that same experiment, internals did p95 react significantly against low influence messages. In the current experiment, since internals may not have perceived directive therapists as exerting high levels of pressure or influence, they may not have reacted against the more subtle and low-keyed influence attempts which were made. Thus, they were "successful" with directive therapists. Furthermore, it might be argued logically that internals, who are generally psychologically well adjusted, would be able to integrate the ideas of their therapists when they were not viewed as out and out manipulations. 45 In still another study, Getter (1966) found that internals did not give "correct" answers during a reinforce- ment period, but did give correct answers once the reinforce- ment had been discontinued. This suggests that although internals might not want to Openly admit or show it, they might indeed be influenced by the statements and ideas of others, including their therapists. Hence, they might be successful with directive therapists even though they would be loathe to admit it. Lack of Significant Differences for Internals With Both Senior and Junior Staff It was predicted in hypothesis three that there would be no significant differences for internals between the degree of success with senior staff and degree of success with junior staff. The lack of significant change for internals on both I-E and the Tennessee scale, combined with the lack of vali- dity on the third outcome measure, made in impossible to adequately test this hypothesis. Thus, although internals working with senior staff did not differ significantly from those working with junior staff, any conclusions drawn from these findings must be made with extreme caution. Differences Between Ritchie and Phares' Study and the Present Study Hypothesis four stated that externals would be more successful with senior staff than with junior staff. This hypothesis was based primarily upon Ritchie and Phares' 46 study on attitude change (1969). As was discussed previously, it was found in that study that externals had greater attitude change as a result of communications from high pres- tige sources than from low prestige sources. It was pre- dicted in the current study that those results would similarly affect the process of psychotherapy. However, there are a number of important differences between the two studies which may help to explain the absence of support for hypothesis four. On the basis of the previous literature, Ritchie and Phares felt that more attitude,change would occur under low involvement conditions than under high involvement conditions. For that reason, they chose to present the tOpic of national budget expenditures to their subjects in order to maximize the likelihood that attitude change would take place. In the current study, however, the topics being discussed were extremely involving and important for the subjects. One would predict that psychological and social concerns would be far more involving for clients in the present study than were the arguments about national budget expenditures for the subjects of the Ritchie and Phares study. Thus, although on a personally unimportant sub- ject externals might be swayed more heavily by high than by low prestige sources, the same might not be true for the much more important and personally meaningful tOpics discussed in the course of psychotherapy. 47 Another difference between the two studies concerns the relative positions of the high and low prestige communi- cators. In the Ritchie and Phares study the high prestige source was described as having been associated with the federal government for a period of over ten years in key positions including Secretary of the Treasury. The low status communicator was described merely as a twenty year old college sophomore. Thus, considering the nature of the opinions to be changed, there was an extremely large discrepancy between the high and low prestige sources. On the other hand, in the present study the discrepancy between high and low prestige sources was not nearly as great. Members of the high prestige group held a doctorate in clinical or counseling psychology and were generally older in age. The low prestige group consisted of advanced gra- duate students in the same area and were generally younger in age. Thus, the difference in prestige might not have been perceived as being all that great by the client subjects (college undergraduates). Indeed, a client might very well have viewed an advanced graduate student in psychology as being a high prestige source of information concerning social and psychological problems. Clients therefore may not have viewed senior and junior staff members as differing all that much in status or knowledge. The two studies are also dissimilar in that whereas the status of the communicator in the Ritchie and Phares 48 study was fixed and constant, the perceived status of the counselors in the present study was not necessarily so. For example, even if a client had initially perceived a junior staff member as being somewhat less prestigious, he might have increased his view of that therapist's prestige over the course of therapy. Likewise, a senior staff member might have been preceived initially as prestigious and expert, yet during the course of therapy fallen in prestige in the eyes of the client. Such changes in perceived status were not possible in the Ritchie and Phares study. Ritchie and Phares discuss the possibility that need for approval affected their results. They speculate that externals may have changed towards the opinions of the high prestige communicators as a means of fulfilling a need for approval rather than as being strictly a result of perceived status. If such were indeed the case, need for approval could similarly have affected the results of the current study. The need for approval of one's therapist is quite often very strong regardless of the therapist's staff level or perceived status. One could argue that the need for approval of one's therapist by a client would be much greater than that for Ritchie and Phares' Secretary of the Treasury by their subjects. Therefore, if it were a factor in the Ritchie and Phares study, it would be an even greater factor in the current study. If, as they suggest, Ritchie and Phares' findings were possibly due to need for approval 49 rather than perceived status, that might explain the lack of differences between high and low prestige sources in the present study. To summarize then, as has been demonstrated, the experimental situation set up by Ritchie and Phares differs in many respects from the process of therapeutic interactions studied by the present author. Since that study provided the original basis for the hypothesis that externals would be more successful with senior staff members the absence of support for that prediction may be attributed to the differ- ences between those studies. Incidental Findings and Tentative Implications Post hoc analyses of the data show that, contrary to the hypotheses, externals actually exhibited significantly greater change with nondirective than with directive therapists. While these findings cannot be asserted as facts, they are worthy of some discussion. One possible explanation is externals' strong need to conform. It might be that clients are generally more aware of the values Opinions, and ideas of directive therapists than of non- I directive therapists. Consequently externals may give verbal agreement to their directive therapists, although such agreement may not actually be integrated and utilized by these clients outside of therapy. With nondirective therapists however, externals may not have as great an awareness of the beliefs and values of their therapists. 50 Such a situation may force the external to develop and act out of his own values and belief systems rather than those of the people around him. A consequence of this might be a better psychological adjustment, a more adaptive behavioral repertoire, and an increase in the expectancy for internal control. Thus an external might actually experience more change with a nondirective than with a directive therapist. The findings in the present study also suggest that the differential reactions to influence attempts found in previous studies may not apply to the process of psycho- therapy and other more naturalistic settings. Whereas internals and externals may show differences in their reac- tions to structured and somewhat artificial experimental situations, those differences may not be as dramatic in less controlled and more complex, ambiguous settings. Areas for Future Research The above explanations of this study's results suggest a number of possibilities for future research. First of all, a larger sample size would permit a stronger testing of the hypotheses and their underlying theoretical assumptions. Because of the small number of subjects in the present study, client and therapist groups had to be split at the median, in order to place subjects in a particular category. A greater sample size would permit one to look at individuals who differ more clearly as to their locus of control and counseling styles. 51 As was discussed, the method used in the current study for designating a therapist as directive or nondirective can be greatly improved upon, with future experimentation providing a more valid means of labeling counseling tech- niques and styles. For example, the rating of transcripts of entire interviews rather than of isolated segments would enable raters using the Aronson or similar scales to be more accurate in their scoring. Even greater oppor- tunities for accuracy would be available if one-way mirror observations or video tape recordingss of counseling inter— views were used hiplace of written transcripts. Furthermore, a broader range of therapeutic styles and approaches could provide a more adequate means of testing whether internals and externals do indeed achieve differential levels of success as a result of varied types of experiences in therapy. The effects upon internals and externals of more specific types of treatment modalities could be compared to one another. This would be especially useful in light of the post hoc findings that externals seem to be more successful with nondirective than with directive therapists. Such research might also provide a means of assessing the overall merits of different types of counseling techniques. APP END ICES APPENDIX A SOCIAL REAC TION INVEN TORY We are interested in the way different people look at things which happen in our society. We have listed below 29 pairs Of statements. You will probably agree more with one of the two statements than you will with the other one. Sometimes neither of the two statements will really say what you would like for it to say. If this happens, just choose the one which is closest to what you believe. ‘ There are no right or wrong answers. Just choose the one which is closest to what you really believe, and circle the appropriate letter. Go ahead and start. Remember to choose the one which is closest to what you really believe. ******** l. A. Children get into trouble because their parents punish them too much. B. The trouble with most children nowadays is that their parents are too easy with them. 2. A. Many of the unhappy things in people's lives are partly due to bad luck. B. People's misfortunes result from the mistakes they make. 3. A. One of the major reasons why we have wars is because peOple don't take enough interest in politics. B. There will always be wars, no matter how hard people try to prevent them. 4. A. In the long run people get the respect they deserve in this world. B. Unfortunately, an individual's worth often passes unrecognized no matter how hard he tries. 5. A. The idea that teachers are unfair to students is nonsense. B. Most students don't realize the extent to which their grades are influenced by accidental happenings. 52 10. ll. 12. 13. 14. 53 Without the right breaks one cannot be an effective leader. Capable people who fail to become leaders have not taken advantage of their opportunities. No matter how hard you try some people just don't like you. People who can't get others to like them don't under- stand how to get along with others. Heredity plays the major role in determining one's personality. It is one's experiences in life which determine what they're like. I have often found that what is going to happen will happen. Trusting to fate has never turned out as well for me as making a decision to take a definite course of action. In the case of the well prepared student there is rarely if ever such a thing as an unfair test. Many times exam questions tend to be so unrelated to course work that studying is really useless. Becoming a success is a matter of hard work, luck has little ornothing to do with it. Getting a good job depends mainly on being in the right place at the right time. The average citizen can have an influence in govern- ment decisions. This world is run by the few people in power, and there is not much the little guy can do about it. When I make plans, I am almost certain that I can make them work. It is not always wise to plan too far ahead because many things turn out to be a matter of good or bad fortune anyhow. There are certain people who are just no good. There is some good in everybody. 15. 16. 17. 18. 19. 20. 21. 22. 23. 54 In my case getting what I want has little or nothing to do with luck. Many times we might just as well decide what to do by flipping a coin. Who gets to be the boss often depends on who was lucky enough to be in the right place first. Getting people to do the right thing depends upon ability, luck has little or nothing to do with it. As far as world affairs are concerned, most Of us are victims of forces we can neither understand, nor control. By taking an active part in political and social affairs the peOple can control world events. Most peOple don't realize the extent to which their lives are controlled by accidental happenings. There really is no such thing as "luck." One should always be willing to admit mistakes. It is usually best to cover up one's mistakes. It is hard to know whether or not a person really likes you. How many friends you have depends upon how nice a person you are. In the long run the bad things that happen to us are balanced by the good ones. MOst misfortunes are the result of lack of ability, ignorance, laziness, or all three. With enough effort we can wipe out political corruption. It is difficult for peOple to have much control over the things politicians do in office. Sometimes I can't understand how teachers arrive at the grades they give. There is a direct connection between how hard I study and the grades I get. 24. 25. 26. 27. 28. 29. 55 A good leader expects people to decide for themselves what they should do. A good leader makes it clear to everybody what their jobs are. Many times I feel that I have little influence over the things that happen to me. It is impossible for me to believe that chance or luck plays an important role in my life. PeOple are lonely because they don't try to be friendly. There's not much use in trying too hard to please people, if they like you, they like you. There is too much emphasis on athletics in high school. Team sports are an excellent way to build character. What happens to me is my own doing. Sometimes I feel that I don't have enough control over the direction my life is taking. Most of the time I can't understand why politicians behave the way they do. In the long run the peOple are responsible for bad government on a national as well as on a local level. APPENDIX B THERAPIST RATING SCALE Place an (x) at the point on the scale that best des- cribes your feelings about whether counseling helped your client to solve his problems. Emuemebr flamed HBnle nxfiifenxm flamed Ikflpai Vaqrmudr- hanMifl the the neflimm the the waseadxeme- to the client client helped nor client client 1y helpful client quite a somewhat banned the somewhat quite a to the lot client lot climt 56 XRC XCFa XCFi APPENDIX C DEFINITIONS AND INSTRUCTIONS FOR COUNSELOR CODING CATEGORIES Restatement of Content A simple repeating of what the client has said without any effort to organize, clarify, or interpret it, or any effort to show that the counselor is appre- ciating the feeling of the client's statement by understanding it. The wording need not be identical with that of the client. Emphasis here is on statement of attitudes of others toward the client; statements of fact, statement of conditions of the environment. These statements usually reflect the intellectual rather than the affective aspects of the client's response. Accurate Clarification of Feeling A statement by the counselor which puts the client's feeling or affective tone in a clearer or more recog- nizable form, or any effort to show that the counselor is accurately recognizing the feeling of the client's statement by understanding it. Emphasis here is on the client's attitudes and feelings toward the topic being discussed. The clari— fication, or reflection of the counselor must be reason- ably accurate to be scored under this category. Inaccurate Clarification of Feeling A statement by the counselor which expresses atti- tudes and feelings of the client different from those he has expressed or implied. A mistake or an error has occurred in attempting to clarify the client's verbal- ized feelings or attitudes. These statements are characterized by: 1. Refelcting a minor feeling and ignoring a major feeling when both are present in the client's statement. 2. Gross understatement of the client's feeling. 57 XCFu XIT 58 3. Real errors or mistakes as a result of misunder- standing the client. Clarification of Unverbalized Feeling A statement by the counselor which expresses unverbalized attitudes or feelings of the client. A recognition or clarification of a feeling or an atti- tude which the client has not verbalized but which is clearly implied in the client's previous statements and is in context with these previous statements. The emphasis here is on recognition or clarifica- tions which go beyond what the client has verbalized but which are implied in his previous statements. "Shrewd guesses" of the client's attitudes which are Obtained from the counselor's knowledge of the total situation are coded in this category. Feeling must be clarified to use this category. Interpretation Any counselor statement which indicates, even vaguely, a causal relationship in the client's behavior; points out a characterization, explains, or informs the client as to his patterns or personality; provided he has not specifically mentioned it in presvious state- ments. These statements frequently represent the coun- selor's attempt to impose his "diagnostic" concepts. Scoring Notes l. Differentiating XRC from XIT: An XIT may be a nonfeeling statement and confused with an XRC. However, the presence of a causal inference in the statement would place it in the X13 category. Pointing out a characterization, explaining, or informing the client as to his patterns or personality goes beyond a restatement of content and would be an §£I° If the client had pointed out the characteriza- tion himself in the previous statement, the counselor response would be x39. 2. Differentiating XCFu from XIT. An XCFu and an XIT might both have elements of unexpressed feeling (see definition of XCFu), but if, in addition, the statement contains elements of causal inference it is classified as XIT. XCS XND XFT 59 3. If no feeling has been clarified it cannot be considered an XCFu. Structuring Statements which explain the counseling procedure; state the expected outcome of the treatment process in general (not in the client's specific case); the limi- tations of time; or the responsibilities of the coun- selor or client. These statements emphasize the process of counsel- ing itself. ‘ Nondirective Leads Counselor responses which are aimed at eliciting from the client a further statement of his problem. These responses are planned in such a manner as to avoid limiting the nature of the discussion to a narrow topic. EXAMPLES: "What would you like to talk about today?" "How have things been going?" "How are you today?" (If asked in a general sense.) Forcipg the Topic Attempts by the counselor to redirect to the client the responsibility for selecting a topic for discussion; emphasis upon discussing specific tOpic or suggestions that the client discuss or develop a specific topic. EXAMPLES: "How do you feel about that?" "Tell me how you felt then." "Can you tell me more?" Proposing Client Activity Any statement that implies that the client should take any kind of action. This does not imply a change of attitude. EXAMPLES: "You should work out in the gym sometimes." "Why don't you read Shaffer's book on psychology?” XPS XSA XRS 60 Direct Questions Questions asked by the counselor to obtain specific information from the client. Asking an outright ques- tion that requires the giving of a factual answer. It does not include counselor statements phrased in the form of a question that really only clarify or restate the previous statement of the client. EXAMPLES =. "How old are you now?" "Did you read that book I suggested?" Persuasion Any attempt to persuade the client to accept an alternate point of view; an implication that the client should change his attitude or frame of reference. EXAMPLE: "Don't you think it would be better that way, now?" Simple Acceptance Simple agreement; statements that indicate under- standing or assent, but do not imply approval or dis- approval. This category is used if the counselor statement is not an answer to a question. EXAMPLES: "Yes," "Hhmm," "I see." "That's right." (If not in answer to a question.) Reassurance Counselor statements which encourage the client which are intended to reassure the client's self-esteem or self-assurance; or which imply sympathy. Emphasis here is on items tending to alleviate anxiety by changing the client's evaluation of himself through a minimization of his problem. Approval and Encouragement Counselor statements which evaluate the client or his ideals in terms of the counselor's own attitudes in such a manner as to provide emotional support. 61 This is emphatic acceptance, an obvious reward given by the counselor for an activity of the client. EXAMPLES: "That's fine." "You bet." “You've covered a lot of ground today; that's good." XDC Disapproval and Criticism Any expression of disapproval or criticism of the client by the counselor. EXAMPLE: "You need to get hold of yourself." XFD Friendly Discussion Any statement of friendly discussion with the client, unrelated to his problems, which are designed to maintain a positive rapport with the client. XEC Ending of the Contact Any statement involving the ending of the contact, or making future appointments. XES Ending the Series of Interviews Any statement involved in ending the series of interviews which result from the client's discussing the ending of the series. XUNt Unclassifiable: Due to Transcription Difficulties Any statement not classifiable because parts of it are missing, it was not clear on the recording, or for any transcription difficulties. XIX Giving Information Statements supplying factual data to the client. XUN Unclassifiable Any statements not classifiable into one of the other categories. 62 GENERAL INSTRUCTIONS FOR CODING COUNSELOR RESPONSES: l. 2. Carefully read the client statement so you will know if the counselor is accurately clarifying it, etc. Read the counselor response. Decide which category it represents and place the number of the response on the work sheet and check the correct column for the category of the response. Place an "a," "i," or "u" in the XCF column if the coun- selor has reflected feeling, depending on the type of reflection or clarification of feeling made. If more than one type of category is represented in the counselor response indicate the end of each type of response and code as separate responses. Use subscripts of a, b, c. etc. under the number of the counselor state- ment. Put each coding on a separate line. 63 Case Nutber: Interview Nunber: Classifier: RESPONSE (DUNSELOR CATEIDRIES NUMBER XRC XITXIS )GJQHQDMEWSMXCAWS MIBDR REFERENCES REFERENCES Abramowitz, S. I. Locus of control and self-reported depression among college students. Psychological Reports, 1969, gg, 149-150. Aronson, M. A study of the relationships between certain counselor and client characteristics in client-centered therapy. Unpublished Ph.D. dissertation, Pennsylvania State College, 1951. Aronson, M. A study of the relationships between certain counselor and client characteristics in client-centered therapy. In W. V. Snyder, Group Report of Research in Psychotherapy. Pennsylvania State College, 1953. Aschcraft, C., and Fitts, W. H. Self-concept change in psychotherapy. Psychotherapy, Theory, Research and Practice, 1964, I, 115-118. Baker, S. R. Relations of locus of control to escapism. Psychological Reports, 1971, 39, 313-314. Bialer, I. Conceptualization of success and failure in mentally retarded and normal children. Journal of Personality, 1961, g9, 303-320. Biondo, J., and MacDonald, A. P. Internal-external locus of control and response to influence attempts. Journal of Burnes, K. B., Brown, W. A., and Keating, G. W. Dimensions of control. Correlations between MMPI and I-E scores. J. of Consulting and Clinical Psychology, 1971, fig, 301. Crowne, D. P., and Liverant, S. Conformity under varying conditions of personal commitment. J. of Abnormal and Social Psychology, 1963, gg, 547-555. Doctor, R. M. Locus of control of reinforcement and respon- siveness to social influence. J. of Personality, 1971, 39, 542-551. Dua, P. S. Comparison of the effects of behaviorally oriented action and psychotherapy reeducation on intraversion-extraversion, emotionality and internal- external control. J. of Counseling Psychology, 1970, 11, 567-572. 64 65 Ducette, J., and Wblk, S. Locus of control and extreme behavior. J. of Consultingyand Clinical Psychology, 1972, 33, 253-258. Farkas, A. The internal-external dimension of experience in relation to the process and outcome of psychotherapy. Unpublished Master's thesis, Michigan State University, 1969. Fish, 3., and Karabenick, S. A. Relationship between self- esteem and locus of control. Psychological Reports, 1971, 33, 784. Fitch, G. Effects of self-esteem, perceived performance and choice on causal attributions. J. of Personality and Social Psychology, 1970, 33, 311-315. Fitts, W. H. The experience of psychotherapy: What it's like for client and therapist. Princeton, N.J.: D. Van Nostrand Co., Inc., 1965. Getter, H. A personality determinant of verbal conditioning. J. of Personality, 1966, 34, 397-405. Gillis, J. S., and Jesson, R. Effects of brief psychotherapy on belief in internal control: An exploratory study. Psychotherapy: Theory, Research and Practice, 1970, 1, 135-137. Harrow, M., and Ferrante, A. Locus of control in psychiatric patients. J. of Consulting and Clinical Psychology, 1969, 33, 562-589. Helweg, G. C. The relationships between selected personality characteristics and perceptions of directive and non- directive psychotherapeutic approaches. Dissertation Abstracts International, 1971, 33, 2396. Hersch, P. D., and Scheibe, K. E. Reliability and validity of internal-external control as a personality dimension. J. of Consulting Psychology, 1967, 33, 609-613. Hjelle, L. A., and Clouser, R. Susceptibility to attitude change as a function of internal-external control. The Psychological Record, 1970, 33, 305-310. Jacobson, R. A. Personality correlates of choice of therapist. Dissertation Abstracts International, 1971, 33, 5626. James, W. H., and Rotter, J. B. Partial and 100% reinforce- ment under chance and skill conditions. J. of Experi- mental Psychology, 1958, 33, 397-403. 66 Lefcourt, H. M. Belief in personal control: Research and implications. Journal of Individual Psychology, 1966, 33, 185—195. (a) Lefcourt, H. M. Internal versus external control of rein- forcement: A review. Psychological Bulletin, 1966, 63, 206-220. (b) Lefcourt, H. M. Internal versus external control of rein- forcement revisited: Recent developments in B. A. Maher (ed.) Progress in experimental personality research. New York: Academic Press, 1972. Lefcourt, H. M., Lewis, L., and Silverman, I. W. Internal versus external control of reinforcement and attention in a decision making task. Journal of Personality, 1968, 36, 663-682. Lesyk, J. J. Effects of intensive Operant conditioning on belief in personal control in schizophrenic women. Dissertation Abstracts International, 1969, 33, 4849. MacDonald, A. P. Internal-external locus of control: A partial bibliography. (II) Selected Documents in Psychology, JSAS, 1972, Manuscript No. 156. (a) MacDonald, A. P. Internal-external locus of control: A partial bibliography. Unpublished manuscript, West Virginia University, 1972. (b) Nowicki, S. J., Bonner, J., and Feather, B. Effects of locus of control and differential analogue interview procedures on the perceived therapeutic relationship. J. of Consultingsand Clinical Psychology, 1972, 33, 434. Nowicki, S., and Barnes, J. Effects of a structured camp experience on locus of control orientation. J. of Genetic Psycholggy, 1973, 123. Odell, M. Personality correlates of independence and con- formity. Unpublished master's thesis. Ohio State University, 1959. Palmer, R. D. Parental perception and perceived locus of control in psychopathology. Journal of Personality, 1971, 420—431. Paul, G. L. Insight vs. Desensitization in Psychotherapy. Stanford University Press, 1966. 67 Pierce, R. M., Schauble, P. G., and Farkas, A. Teaching internalization behavior to clients. Psychotherapy: Theory, Research, and Practice, 1970, 7,,217-220. Ritchie, E., and Phares, E. J. Attitude change as a function of internal-external control and communicator status. Journal of Personality, 1969, 31, 429-443. Ray, W. J., and Katahn, M. Relation of anxiety to locus of control. Psychological Reports, 1968, 33, 1196. Rotter, J. B. Social Learning and Clinical Psychology, New York: Prentice Hall, 1954. Rotter, J. B. Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 1966, 30 (1, Whole No. 609). Smith, R. E. Changes in locus of control as a function of life crisis resolution. J. of Abnormal Psychology, 1970, 13, 328-332. Strickland, B. R. The relationships of awareness to verbal conditioning and extinction. Dissertation Abstracts International, 1963, 33, 2988. Strickland, B. R. Individual differences in verbal con- ditioning, extinction and awareness. J. of Personality, 1970, 33, 364-378. Sundland, D. M., and Barker, B. N. The orientations of psychotherapists. Journal of Consulting Psychology, 1962, 33, 201-212. Throop, W. F., and MacDonald, A. P. Internal-external locus of control: A bibliography. Psychological Reports, 1971, 13, 175-190. Warehime, R. G., and Foulds, M. L. Perceived locus of control and personal adjustment. J. of Consultingfiand Clinical Psychologyp 1971, 31, 250-252. Watson, D. Relationship between locus of control and anxiety. Journal of Personality and Social Psychology, 1967, g, 91-93. Williams, C. B., and Nickels, J. B. Internal-external control dimension as related to accident and suicide proneness. J. of Consulting and Clinical Psychology, 1969, 33, 485-494. MICHIGAN STATE UNIV. LIBRARIES lllllll Hill! N Ill ll I"! III Ill 9 312 30000 Illilillfllllfll III 138 74