Instrument Critique - Needed ASAP

Psychotherapy Volume 3 I/Fall 1994/Number 3

THE MYERS-BRIGGS TYPE INDICATOR PERSONALITY DIMENSIONS AND PERCEPTIONS OF QUALITY

OF THERAPY RELATIONSHIPS

BRUCE A. NELSON JAYNE E. STAKE University of Missouri—St. Louis

Associations between perceived quality of therapy relationships and therapist and client scores on the Myers-Briggs Type Indicator (MBTI; Myers & McCaulley, 1985) were explored. Participants were 35 experienced therapists and 53 adult outpatients in individual personal therapy. Therapy dyads had met for a median of45 sessions at the time of testing. Higher therapist MBTI Extroversion and Feeling scores were associated with client and therapist independent reports of more positive perceptions of the therapy relationship. Client perceptions of relationship quality were positively and linearly related to similarity of therapist-client a) overall MBTI profiles, b) Thinking/Feeling scores, and c) Judging/Perceiving scores. Clinical and methodological implications of the findings are discussed.

Social psychology research provides strong evidence that people who are more similar in self- descriptions and attitudes experience greater mu- tual attraction and liking (Griffitt, 1966; Hendrick & Page, 1970; Izard, 1960). The association be- tween similarity and attraction suggests that thera- pist-client dyads matched on personality vari-

Correspondence regarding this article should be addressed to Jayne E. Stake, Department of Psychology, University of Missouri—St. Louis, MO 63121.

ables may have more positive therapy relation- ships. This intriguing possibility has led research- ers to explore the value of therapist-client simi- larity with a variety of personality measures. Em- pirical support for the personality similarity hypothesis has been generally weak and inconsis- tent, however, when tested in the therapy setting (for reviews, see Atkinson & Schein, 1986; Beu- tler, Crago, & Arizmendi, 1986; Parloff, Waskow, & Wolfe, 1978; Ross, 1977). One rea- son for the lack of positive findings may be that researchers have failed to identify appropriate personality variables on which to measure thera- pist-client similarity (Carson & Llewellyn, 1966; Ross, 1977). As Ross (1977) cautioned, thera- pist-client similarity is unlikely to affect the qual- ity of therapy relationships unless the variables on which therapist and client are matched are directly relevant to the therapy setting.

Newman (1979) argued that the personality di- mensions measured by the Myers-Briggs Person- ality Inventory (MBTI; Myers & McCaulley, 1985) are relevant to the therapy relationship be- cause they pertain to characteristic ways of per- ceiving and interacting with oneself and the outer world. The MBTI assesses three personality di- mensions identified by Jung (1923/1971): a) Ex- troversion-Introversion: Involvement in the outer world of people and things versus interest in the inner, personal world of concepts and ideas; b) Sensing-Intuiting: a preference for direct data from the physical senses versus indirect, intuitive perception colored by unconscious associations; and c) Thinking-Feeling: a preference for judg- ments based on logical, impersonal processes ver- sus personal, subjective valuing (Myers & Myers, 1980). The MBTI also measures a fourth person- ality dimension, Judging-Perceiving, which is implicit in Jung's writings. Judging refers to a preference for order, closure, and decision-mak- ing, and Perceiving to a preference for informa-

449

B. A. Nelson & J. E. Stake

tion-gathering and assimilation of new ideas and information (Myers & Myers, 1980).

Newman (1979) suggested that amount of simi- larity between therapist and client MBTI profiles should influence the degree of correspondence between the type of help a therapist will offer and that which a client will perceive as helpful. For example, the therapist who has a stronger prefer- ence for judging may be more likely to approach problems in an orderly, systematic manner and to reach judgments quickly, which may be appre- ciated more by the client who has a stronger pref- erence for judgment. In contrast, a therapist with a greater preference for perceiving may spend more time gathering information and considering alternatives, an approach that may be accepted more by clients with a higher preference for perceiving.

Investigations of other kinds of relationships (e.g., marital, supervisor/worker) provides some support for the expectation that MBTI similarity is associated with more positive interactions and mutually satisfying relationships (Carlson & Will- iams, 1984; Yeakley, 1982). However, evidence for an association between therapist—client MBTI similarity and positive therapy interactions is equivocal. Mendelsohn & Geller (1965) reported that MBTI similarity was positively related to client ratings of comfort and rapport in a freshman sample, but this finding was not replicated in their sample of advanced students. Braun (1971) re- ported that similarity on the MBTI Think- ing-Feeling dimension was associated with an empathy measure—therapist ability to predict cli- ent descriptions of therapists—for female thera- pists but not male therapists. Other researchers found no associations between therapist-client similarity and measures of relationship quality (Colen, 1978; Mendelsohn & Geller, 1967). Re- tention in therapy and consistent attendance may be interpreted as indirect measures of relationship quality. MBTI similarity was positively related to length of treatment in two studies (Mendelsohn & Geller, 1963; Mendelsohn, 1966) but nega- tively related to consistency in attendance in an- other (Mendelsohn & Geller, 1967).

In addition to similarity of therapist-client MBTI profiles, the independent associations of therapist and client MBTI scores with therapy relationship variables have been investigated in a few studies. Client MBTI scores were not associ- ated with quality of therapy relationships or evalu- ation of treatment gains (Mendelsohn & Geller,

1965; Mendelsohn, 1966). There is no direct evi- dence that therapist MBTI scores are related to relationship quality either; however, there is some indirect support for this possibility. Therapist In- tuiting and Feeling scores related to higher coun- seling performance ratings in a small sample of counselor trainees (McNamara, 1967), and MBTI Extroversion scores (Newman, 1979) and Feeling scores (Emanuel, 1972) related to standardized measures of empathy. These findings suggest that there may be an association between therapist MBTI dimensions and quality of therapy relation- ships.

Given substantial evidence from social psy- chology research for an association between simi- larity and attraction (e.g., Hendrick & Page, 1970), and given that the MBTI assesses charac- teristic preferences for relating to self and others, why is there not stronger evidence for a relation between measures of relationship quality and therapist-client MBTI similarity or between mea- sures of relationship quality and therapist or client MBTI scores? Methodological limitations in the previous MBTI research may help to account for the lack of positive findings. First, therapy dyads in most studies met for an average of only two to three sessions, and some met for one session. Hence, assessments of relationships were often made after only a brief therapy experience, and there may not have been time in many of these relationships for the potential influences of the personality styles measured by the MBTI to be- come manifest. In addition, similarity has some- times been measured as a categorical (e.g., low, medium, high) rather than continuous measure (e.g., Jones, 1968; Mendelsohn & Geller, 1967), and client and therapist type have sometimes been treated as categorical variables (e.g., Extrovert vs. Introvert) rather than continuous scores (e.g., Mendelsohn, 1966; Mendelsohn & Geller, 1965). The categorical measures provide less discrimina- tion of individual differences along the MBTI di- mensions than the continuous measures, and they are less reliable (Carlson, 1985). The use of cate- gorical scores may therefore have obscured asso- ciations between the MBTI dimensions and ther- apy relationship variables.

In the present study, we investigated therapist and client MBTI dimensions in long-term, per- sonal therapy using continuous MBTI scores. The relation of client MBTI scores, therapist MBTI scores, and similarity of MBTI profiles to thera- pist and client perceptions of relationship quality

450

Therapist-Client MBTI Similarity

was tested. Based on the rationale and research reviewed here and the improved methodology of the present study, we expected that a) greater similarity of therapist and client MBTI profiles would be associated with higher ratings of rela- tionship quality; and b) therapist MBTI scores would be associated with ratings of relationship quality.

Method

Participants

Clients included 40 women and 13 men, who were primarily Caucasian (98%). The average age of clients was 36.3 years (range = 19-64). Cli- ents were seen in a private practice setting (70%), a community mental health clinic (13%), or a student counseling center (17%). Thirty-seven percent were single, 37% married, 21% divorced, and 2% widowed. College degrees were held by 53% of the clients, and all but 2% had completed high school. The median number of therapy ses- sions attended was 45 (range = 7-393).

Therapists included 22 women and 13 men, who were primarily Caucasian (97%). Therapists' highest educational degrees were: Ph.D. (46%), M.D. (3%), Psy.D. (3%), M.A. (26%), M.S.W. (9%), and other or unspecified (13%). Therapeu- tic modalities of therapists were: eclectic (43%), cognitive-behavioral (17%), humanistic/client- centered (15%), dynamic/psychoanalytic (15%), and other (10%). Therapists reported an average of 9.3 years of experience (range = 1-27). Eighteen therapists saw two clients each, and 17 therapists saw one client each.

Measures

Myers-Briggs Type Indicator (MBTI). Form G of the MBTI is a 126-item self-report question- naire that measures the four personality dimen- sions described earlier: Extroversion-Introver- sion (ET), Sensing-Intuiting (SN), Thinking- Feeling (TF), and Judging-Perceiving (JP). To obtain continuous scores for each dimension, the number of items endorsed for each side of the dimension is summed (with appropriate weight- ing as specified in the standard MBTI scoring instructions), the difference between the sums is calculated, and the difference score is converted to a standardized score {i.e., continuous score) according to the MBTI transformation table. Form G continuous scores for all dimensions are internally consistent and stable over time (Carl-

son, 1985). Extensive research provides support for the construct validity of the four MBTI scales (for reviews, see Carlyn, 1977; Carlson, 1985).

Therapist—client similarity. Similarity of profile type was calculated according to the proce- dure recommended by Cronbach & Gleser (1953). Differences between client and therapist continuous scores for each MBTI dimension were squared, and the squared values were then summed to yield a D2 similarity score. The range of D2 scores was 20-242.

Quality of the therapy relationship. Relation- ship quality was assessed from the perspective of the client and the therapist. Client perceptions of relationship quality were measured with 10 items adapted from the Patient Questionnaire (Strupp, Fox, & Lessler, 1969). Examples are, "I feel there is a good deal of warmth in the way my counselor talks to me," and "I feel my counselor is genuinely interested in helping me." Clients indicated the extent to which they agreed with each item on 7-point Likert-type scales ranging from strongly disagree (1) to strongly agree (7). The range of obtained scores was 4 2 - 7 0 . Higher scores represent more positive client perceptions. The internal consistency of the client relationship quality measure was high—coefficient alpha was .87 for the ten items.

Therapist perceptions of relationship quality were measured with five items adapted from the Therapist Questionnaire (Strupp et al., 1969). Ex- amples are, "How warmly do you feel toward this client?", and "How much do you enjoy work- ing with this client?" Therapists responded on 7- point Likert-type scales ranging from not at all (1) to a great deal (7). Therapist scores ranged from 13-35. Coefficient alpha for the five items was high, .85. The association between client and therapist ratings was significant, r(50) = + .35, p < .01.

Pretherapy adjustment. Clients rated their pretherapy adjustment on two 7-point Likert items. The first item, "Before entering the present counseling, how adequately do you feel you were dealing with your problems?", was presented with a scale ranging from very adequately (1) to very inadequately (7). The second item, "How much distress did you feel at the time you started the present counseling?", had a scale ranging from a tremendous amount (1) to none at all (7). Ratings on the former item were reversed so that higher scores reflected better adjustment on both items. Ratings on the two items were summed to obtain

451

B. A. Nelson & J. E. Stake

pretherapy adjustment scores. These scores ranged from 2 - 1 0 .

Procedure

The study was described to therapists as an exploration of basic personality preferences and the experience of therapy. The therapists who agreed to participate were asked to select one to two adult clients whom they were currently treat- ing in outpatient, individual psychotherapy. Ther- apists were instructed to select clients with whom they had had at least six therapy sessions. The minimum criterion of six sessions was chosen to ensure that in all dyads therapist and client styles of relating had had some opportunity to affect the therapy relationship. Participating clients were given the opportunity to receive feedback about their MBTI personality type. Informed written consent was obtained from clients and therapists prior to their participation.

Information and questionnaires for the study were given to therapists, who then provided the appropriate materials to their clients. Therapists completed a questionnaire for each of their partic- ipating clients. All participants were instructed to complete the forms privately and mail them directly to the first author.

Results

Client Perspective

Relations between the MBTI type variables and client ratings of relationship quality were tested in a hierarchical regression analysis. In Step 1 of the analysis, length of treatment (i.e., number of sessions) and pretherapy adjustment scores were entered as the first set of predictor variables. These variables did not account for a significant proportion of the variance in client ratings. In Step 2, the four therapist MBTI continuous scores were entered as a set. This set of variables resulted in a significant increase in the predicted variance of client ratings, F ( l , 41) = 2.88, p < .05. The size of this effect was in the medium to high range, r = .47 (Cohen, 1977; Rosnow & Rosen- thai, 1988). Because the effect of the set of thera- pist MBTI scores was significant, the effect of each therapist MBTI dimension was tested sepa- rately in the regression analysis (see Table 1). These analyses indicated that therapist El and TF scores were related to client ratings of relationship quality. Client ratings tended to be more positive when therapists rated themselves as relatively

TABLE 1. Relation of Therapist MBTI Continuous Scores to Perceptions of Relationship Quality

Client perceptions Therapist El Therapist SN Therapist TF Therapist JP

Therapist perceptions Therapist El Therapist SN Therapist TF Therapist JP

F

4.04" 1.02 6.12* < 1

5.29* 1.19 6.43* 3.97°

Effect Size (r)

.30

.36

.34

.37

.30

Note. Effects of therapist MBTI scores were tested after controlling for ratings of pretherapy adjustment and length of sessions

ap< .06. bp< .05.

more extroverted and feeling. Effect sizes were in the medium range (see Table 1). When the set of client MBTI scores was added in place of therapist MBTI scores at Step 2 of the regression analysis, the resulting increase in predicted vari- ance was not significant.

Before testing the effect of similarity of MBTI profile (D2 scores) on client relationship ratings, first length of treatment, pretherapy adjustment, client MBTI scores, and therapist MBTI scores were entered in the regression analysis. D2 simi- larity scores were then added as the final step in the regression analysis. D2 similarity scores significantly increased the predicted variance in client ratings, F ( l , 36) = 8.41, p < .01. The effect size was in the medium range (r = .36). Greater overall profile similarity was associated with higher client ratings of relationship quality. To test for a curvilinear relationship between sim- ilarity and client perceptions of relationship qual- ity, a quadratic similarity term was added to the regression model. This term did not significantly increase the predicted variance. In addition, the effect of similarity was tested separately for each dimension. Client ratings of relationship quality were positively related to similarity of TF scores, F ( l , 33) = 4.09, p < .06, and JP scores, F ( l , 33) = 4.46, p < .05. Effect sizes were in the medium range (.30 and .31, respectively).

Therapist Perspective Therapist ratings of relationship quality were

tested by the same procedure of hierarchical re-

452

Therapist-Client MBTI Similarity

gression analysis described for client ratings. Length of treatment and pretherapy adjustment scores did not account for a significant proportion of the variance of therapist ratings. Main effects of MBTI scores on therapist ratings of relation- ship quality were similar to those for client rat- ings. The set of therapist MBTI scores accounted for a significant increment in the predicted vari- ance of therapist ratings of relationship quality, F(4, 41) = 4.22, p < .01. This effect size was in the large range (t = .53). Separate tests of therapist MBTI dimensions indicated that, as in the case of client perceptions, therapist ratings of relationship quality were related to therapist El scores and TF scores {see Table 1). In addition, the relation between therapist JP scores and thera- pist ratings of relationship quality approached sig- nificance. Therapists who described themselves as relatively more extroverted, feeling, and per- ceiving rated their therapy relationships more pos- itively. Effect sizes were in the medium range. Thus, from the perspective of both therapist and client, therapist extroversion and feeling scores were associated with higher ratings of relation- ship quality.

Neither client MBTI scores nor similarity of client and therapist MBTI scores were signifi- cantly related to therapist perceptions of relation- ship quality.

Discussion A primary finding of our study was that similar-

ity of therapist and client MBTI profiles was asso- ciated with more positive client ratings of the quality of the therapy relationship. Before dis- cussing the implications of this finding, it should be noted that MBTI scores were obtained at the same time as therapy ratings, rather than prior to therapy. Hence, the causal link between MBTI similarity and client perceptions of relationship quality has not been established. It is possible to argue that MBTI similarity did not influence cli- ent perceptions of the therapy relationship but, instead, that clients who experienced a positive relationship with their therapist shifted their self- perceptions on the MBTI dimensions in the direc- tion of their therapist's MBTI scores. However, client MBTI scores probably did not change a great deal during therapy because continuous MBTI scores are highly reliable over time (Carl- son, 1985). Even if clients had changed their self- perceptions along the MBTI dimensions, it is likely that they were influenced by many factors

during the time of their therapy and that their therapist's MBTI type was not the only, or major factor in their change. Thus, it is reasonable to surmise that, although client experiences in the therapy relationship may have had some influence on therapist-client MBTI similarity, the effect was primarily in the other direction—that greater similarity in MBTI profiles led to more positive experiences for clients.

Client ratings of relationship quality were sig- nificantly related to therapist-client similarity on the Thinking-Feeling and Judging-Perceiving dimensions. One way to interpret these findings is to consider the possible effects of therapist-client dissimilarity in the context of Jung's (1923/1971) contracts of thinking—feeling and judging-per- ceiving. In regard to the thinking-feeling dimen- sion, recall that a feeling preference is associated with the use of personal, subjective values in forming opinions and making decisions, and a thinking preference is associated with the use of logical, impersonal bases for making judgments. A client who has a feeling preference may there- fore perceive a therapist who has a greater prefer- ence for thinking as insufficiently sympathetic, and a client who has a thinking preference may see the therapist with a greater preference for feeling as lacking in objectivity and overly influ- enced by subjective values. In regard to the Jud- ging-Perceiving dimension, as discussed earlier, individuals with a preference for judging focus more on order and decision making and perceivers more on gathering information and staying open to new information and ideas (Myers & McCaul- ley, 1985). Clients with a perceiving preference may therefore view a therapist who has a greater preference for judging as overly structured and eager to reach conclusions, and clients who have a judging preference may see a therapist who has a greater preference for perceiving as unorganized and indecisive. In this way, differences on the TF and JP dimensions would affect perceptions of the quality of therapeutic interactions.

Client MBTI scores were not related to therapy relationship ratings independently of therapist MBTI scores. As discussed earlier, previous re- search has also failed to find associations between client MBTI type and variables associated with quality of therapy relationships. In contrast, we found a substantial relation between therapist MBTI scores and ratings of relationship quality. Specifically, when therapists scored higher on Ex- troversion and Feeling, both they and their clients

453

B. A. Nelson & J. E. Stake

rated the therapy relationship more positively. To our knowledge, our findings are the first to show a direct association between therapist MBTI type and client and therapist ratings of therapy relation- ship quality. Our results are consistent with find- ings cited earlier that more extroverted and feel- ing therapists tended to score higher on stan- dardized measures of empathy (Emanuel, 1972; Newman, 1979), and counselor trainees with a feeling preference received higher ratings of their counseling performance (McNamara, 1967).

These findings may be interpreted within the context of Jung's (1923/1971) personality dimen- sions of extroversion-introversion and think- ing-feeling. As discussed earlier, extroverts have a stronger focus on people in their environment, whereas introverts focus more on their internal world of ideas. Extroverts may therefore express themselves and convey their understanding and caring to their clients more easily, which would lead to more positive therapy relationships. These findings suggest that more introverted therapists should make special efforts to ensure they are sufficiently expressive and interactive with their clients. In regard to the thinking-feeling dimen- sion, recall that individuals with a preference for feeling are thought to use subjective values of others and themselves in forming their opinions and judgments, and individuals with a thinking preference rely more on logic. Clients may there- fore experience the more feeling therapist as more attuned to their perspectives and, hence, as more empathic. These findings suggest that it may be important for therapists with a preference for thinking to expand their focus beyond the logical aspects of their clients' statements to attend to their subjective values and attitudes. However, because client perceptions of relationship quality were also related to therapist-client similarity on the Thinking-Feeling dimension, the attributes of the feeling therapist appear to be more im- portant to clients who have a feeling preference.

In conclusion, our results suggest that the MBTI may have value in helping to determine optimal therapist-client matches and in guiding therapists to understand better their impact on cli- ents. However, before using the MBTI to make therapy-relevant decisions, more research is needed to confirm our findings and to determine whether therapist MBTI scores are causally re- lated to perceptions of relationship quality. In ad- dition, therapy variables {e.g., treatment goals) that may mediate the association between therapist

MBTI dimensions and perceptions of relationship quality should be investigated. In researching these issues, careful consideration should be given to MBTI measurement and choice of therapy dyads. Continuous MBTI scores are more reliable and pro- vide better discrimination of individual differences in MBTI preferences than categorical indices of type, and potential influences of MBTI dimensions are more likely to be evident in well-established therapy dyads.

References ATKINSON, D. R. & SCHEIN, S. (1986). Similarity in counsel-

ing. Counseling Psychologist, 14, 319-354. BEUTLER, L. E., CRAGO, M., & ARIZMENDI, T. G. (1986).

Therapist variables in psychotherapy process and outcome. In S. L. Garfield and A. E. Bergin (Eds.), Handbook of psychotherapy and behavior change. Vol. 3 (pp. 257-310). New York: John Wiley.

BRAUN, J. A. (1971). The empathic ability of psychotherapists as related to therapist perceptual flexibility and professional experience, patient insight, and therapist-patient similarity. Dissertation Abstracts International, 32, 2391B-2392B. (University Microfilms No. 71-26,956)

CARLSON, J. G. (1985). Recent assessments of the Myers- Briggs Type Indicator. Journal of Personality Assessment, 49, 356-365.

CARLSON, R. & WILLIAMS, J. (1984). Studies of Jungian ty- pology: HI. Personality and marriage. Journal of Personal- ity Assessment, 48, 87-93.

CARLYN, M. (1977). An assessment of the Myers-Briggs Type Indicator. Journal of Personality Assessment, 41,461-473.

CARSON, R. C. & LLEWELLYN, C. E. (1966). Similarity in therapeutic dyads: A reevaluation. Journal of Counseling Psychology, 30, 458.

COHEN, J. (1977). Statistical power analysis for the behav- ioral sciences (2nd ed.). New York: Academic.

COLEN, R. S. (1978). The effect of therapist-client personal- ity similarity on the process and outcome of short-term psychotherapy. Dissertation Abstracts International, 39, 3502B. (University Microfilms No. 7900682)

CRONBACH, L. J. & GLESER, G. C. (1953). Assessing similar- ity between profiles. Psychological Bulletin, 50, 456-473.

EMANUEL, G. V. (1972). Affective sensitivity (empathy) as a function of therapist experience level, personality type and client cue source. Dissertation Abstracts International, 33, 2342B. (University Microfilms No. 72-28,535)

GRIFFITT, W. B. (1966). Interpersonal attraction as a function of self-concept and personality similarity-dissimilarity. Journal of Personality and Social Psychology, 4, 581-584.

HENDRICK, C. & PAGE, H. A. (1970). Self-esteem, attitude similarity, and attraction. Journal of Personality, 38, 588-601.

IZARD, C. E. (1960). Personality similarity and friendship. Journal of Abnormal and Social Psychology, 61, 47-51.

JONES, H. T. (1968). The relationship of counselor-client relationship similarity to counseling process and outcome. Dissertation Abstracts International, 29, 2962A. Univer- sity Microfilms No. 69-3390)

JUNG, C. G. (1923/1971). The collected works ofC. G. Jung: Vol 6, Psychological types (pp. 510-555). Princeton, NJ: Princeton University Press.

454

Therapist-Client MBTI Similarity

MARZIAU, E. (1984). Three viewpoints on the therapeutic alliance. Journal of Nervous and Mental Disease, YJl, 417-423.

MCNAMARA, T. C. (1967). A study of philosophical identities in a counseling practicum. Dissertation Abstracts Interna- tional, 28, 4878A. (University Microfilms No. 68-8164)

MENDELSOHN, G. A. (1966). Effects of client personality and client-counselor similarity on the duration of counseling: A replication and extension. Journal of Counseling Psy- chology, 2, 228-234.

MENDELSOHN, G. A. & GELLER, M. H. (1963). Effects of counselor-client similarity on the outcome of counseling. Journal of Counseling Psychology, 10, 71-77.

MENDELSOHN, G. A. & GELLER, M. H. (1965). Structure of client attitudes toward counseling and their relation to client-counselor similarity. Journal of Consulting Psychol- ogy, 29, 63-72.

MENDELSOHN, G. A. & GELLER, M. H. (1967). Similarity, missed sessions, and early termination. Journal of Counsel- ing Psychology, 14, 210-215.

MYERS, I. B. & MCCAULLEY, M. H. (1985). Manual: A guide to the development and use of the Myers-Briggs Type Inventory. Palo Alto, CA: Consulting Psychologists Press.

MYERS, I. B. & MYERS, P. B. (1980). Gifts differing (pp. 1-9). Palo Alto, CA: Consulting Psychologists Press.

NEWMAN, L. E. (1979). Personality types of therapist and client and their use in counseling. Research in Psychologi- cal Type, 2, 46-55.

ORUNSKY, D. E. & HOWARD, K. I. (1978). The relation of process to outcome in psychotherapy. In S. L. Garfield and A. E. Bergin (Eds.), Handbook of psychotherapy and behavior change, Vol. 2 (pp. 283-329). New York: John Wiley.

PARLOPF, M. B., WASKOW, I. E., & WOLFE, B. E. (1978). Research on therapist variables in relation to process and outcome. In S. L. Garfield and A. E. Bergin (Eds.), Hand- book of psychotherapy and behavior change, Vol. 2 (pp. 233-282). New York: John Wiley.

ROSNOW, R. L. & ROSENTHAL, R. (1988). Focused tests of significance and effect size estimation in counseling psy- chology. Journal of Counseling Psychology, 35, 203-208.

Ross, M. B. (1977). Discussion of similarity of client and therapist. Psychological Reports, 40, 699-704.

STRUPP, H. H., FOX, R. E , & LESSLER, K. (1969). Patients view their psychotherapy. Baltimore: Johns Hopkins Uni- versity Press.

YEAKLEY, F. R. (1982). Communication style preferences and adjustments as an approach for studying effects of similarity in psychological type. Research in Psychological Type, 5, 30-48.

455