INTERNATIONAL JOURNAL OF GROUP PSYCHOTHERAPY, 58 (2) 2008 DIERICK AND LIETAER THERAPEUTIC FACTORS

Client Perception of Therapeutic Factors in Group Psychotherapy and Growth Groups: An Empirically-Based Hierarchical Model PAUL DIERICK, PH.D. GERMAIN LIETAER, PH.D. ABSTRACT To assess group participants’ perceptions of therapeutic factors, we developed an extensive questionnaire of 155 items that was administered to 489 members of 78 psychotherapy and growth groups of client–centered/experiential, psychoanalytic, behavioral, Gestalt and drama– and bodily oriented orientations. Using multivariate analyses we found a model that reveals the structure and connections of therapeutic factors as they are differentiated in the experience of the group members. Our model encompasses three hierarchical levels of abstraction: 28 Basic scales that appeared to be structured into seven main scales (Group Cohesion, Interactional Confirmation, Cathartic Self–Revelation, Self–Insight and Progress, Observational Experiences, Getting Directives, and Interactional Confrontation) and two dimensions (Relational Climate and Psychological Work). Validity for these therapeutic factors was found in their grounded content, statistically analyzed constructs, importance ratings, and correlations to intermediate outcome measures. Paul Dierick is a clinical psychologist/psychotherapist with specialization in client centered/experiential psychotherapy. He works in the gerontopsychiatry sector in the Psychiatric Center Sint–Norbertushuis at Duffel, Belgium. He is also connected to the Research Unit Psychotherapy and Depth Psychology of the Catholic University of Leuven. Germain Lietaer is Emeritus Professor at Catholic University of Leuven, Belgium and connected to the Research Unit Psychotherapy and Depth Psychology. This research project has been sponsored by the Belgian Foundation of Scientific Research and the Catholic University of Leuven.

203

204

DIERICK AND LIETAER

What are the clients’ core experiences of the group processes which they find important or helpful? This question has a long history. Corsini and Rosenberg (1955) were among the first to attempt an integrative theoretical framework of therapeutic factors. Subsequently, others like Yalom (1970, 1995) and Bloch and Crouch (1985) have worked out integrative visions on therapeutic factors in group psychotherapy and growth groups. Yalom has had the largest influence in this work because he embedded his framework of therapeutic factors into his well-known book of group therapy and because of his development of a questionnaire (Yalom, Tinklenberg, & Gilula, 1968), with several variations, that has generated a good deal of research. For example, Lese and MacNair–Semands (2000) recently developed the Therapeutic Factors Inventory (TFI) to assess the presence of the 11 therapeutic factors described by Yalom (1995), each represented by 9 items. They report good internal consistency and test–retest reliability for their scales, but they do not address the fundamental issue of construct validity. Despite the popularity of these questionnaires about therapeutic factors, several criticisms and questions can be raised (cf. Burlingame, MacKenzie, & Strauss, 2004). First, there are variations in the number and concepts of therapeutic factors that are assessed. The 12 scales that are measured with the original 60–item questionnaire (Yalom et al., 1968) differ from the 11 theoretically described therapeutic factors (Yalom, 1970, 1995). Yalom’s second operationalization of a 14–item questionnaire (Lieberman, Yalom, & Miles, 1973) shows a partly different list of therapeutic factors than the original 12 scales. Self–Disclosure was added and Cognitive Learning was divided into Genetic Insight and Self–Understanding. We notice that the term Cognitive Learning here refers to Self–Understanding in the 60–item operationalization and that Self–Understanding there already was a split from the theoretically described therapeutic factor Interpersonal Learning. The other key figures in this field of research, Bloch and Crouch (1985), formulated a different taxonomy of 10 therapeutic factors,

THERAPEUTIC FACTORS

205

which differs from Yalom’s on important aspects. Existential Factors and The Corrective Recapitulation of The Primary Family Situation are not incorporated in their model, whereas Self–Disclosure is added. Bloch and Crouch (1985) reversed the hierarchical relationship between Interpersonal Learning and Insight, compared to Yalom’s; they formulated a comprehensive cognitive factor of Self–Understanding that encompasses several forms of self–insight including interpersonal self–insight and also motivational and genetic self–insight, although these different forms of self–insight are not differentiated in their category system. On the other hand, they formulated Learning from Interpersonal Actions as a behavioral factor, in line with Yalom’s factor Interpersonal Learning Output. Another fundamental criticism concerning concept formation and model formulation is that the therapeutic factors in both Yalom’s and Bloch and Crouch’s lists differ very much in level of abstraction. For example, Interpersonal Learning is a comprehensive or wide–ranging factor, while Altruism is much more narrow. Furthermore, factor analytic research has not consistently provided the empirical support for Yalom’s scale formulations (Dierick, 2000, 2001; Eckert, Bierman–Ratjen, Tönnies, & Wagner, 1981; Fuhriman, Drescher, Hanson, Henrie, & Rybicki, 1986; Rohrbaugh & Bartels, 1975; Snijders, Van Der Spek, Colijn, Hoencamp, & Duivenvoorden, 1991; Stone, Lewis, & Beck, 1994). Finally, we note that research in this domain started from the (implicit) idea that the therapeutic factors that are to be formulated and differentiated are already known. In a comprehensive literature study (Dierick, 2000), we found that only a few studies have been done (Berzon, Pious, & Farson, 1963; Lieberman et al., 1973) in which group members could formulate their experiences on the basis of open–ended questions so that what they experienced as helpful could be explored without predetermined categories. Only in this manner can we discover therapeutic factors in the client perception that are not yet identified. Bednar and Kaul (1994) summarized the state of the art in the

206

DIERICK AND LIETAER

research on therapeutic factors in group therapy and the direction this research should take, as follows: In fact, each of the reports is seriously flawed in one way or another as are most reports in this difficult field. [With a view to further research they stress] the value of working diligently to develop a variety of conceptual schemes for identifying, describing, classifying, and understanding curative process from a variety of conceptual perspectives. Yalom has made an incisive step in this direction both substantively and methodologically by initially emphasizing careful observation and description of central group phenomena. His observations are outstanding candidates for the conceptual and measurement refinement needed to develop the conceptual tools and technology required for the next developmental step in group research and clinical practice. We simply must establish more consistency and depth in multidimensional conceptions and measurement of central curative factors before their effects can be tested and compared. (pp. 643-644; emphasis added).

Recently, Greene (2003) made similar comments: Yalom’s (1975) original contribution was his brilliant, clinically rich depiction of curative processes, but for the most part he stopped short of developing theories or models of the interrelationships among these processes. He himself doubted the capability of research for developing such models. To a significant degree, then, the research that followed has largely remained at a descriptive level, failing to advance understanding of dynamic pathways. (p. 132; emphasis added).

In line with Bednar and Kaul’s conclusions and recommendations we sought to devise a research project from which we can give a thoroughly empirically grounded answer to the questions: What are the therapeutic factors as seen from the perspective of the group therapy clients themselves and how are they differentiated and inter–related to each other? METHOD

Methodologically, three steps were taken: (1) the construction of a questionnaire that covered as wide a range as possible of helpful

THERAPEUTIC FACTORS

207

processes in all sorts of psychotherapy and growth groups; (2) the administration of this questionnaire to a very extensive and diverse sample of clients in group psychotherapy and growth groups; and (3) the use of a carefully chosen series of multivariate analyses to define the boundaries between the therapeutic factors and to clarify their degrees of interconnectedness. Construction of the Questionnaire

The first step was to create an extensive list of items describing helpful processes based on (1) our own exploratory research with open–ended questions administered to group members and therapists (Dierick & Lietaer, 1990); (2) existing questionnaires (Leszcz, Yalom, & Norden, 1985; Lieberman et al., 1973; Kahn, Webster, & Storck, 1986; Yalom et al., 1968) and category systems (Bloch & Crouch, 1985; Bloch, Crouch, & Reibstein, 1981; Bloch, Reibstein, Crouch, Holroyd, & Themen, 1979; Cabral, Best, & Patton, 1975; Corsini & Rosenberg, 1955; Drob & Bernard, 1986; Drob, Bernard, Lifshutz, & Nierenberg, 1986; Farell, 1962; Hill, 1957, 1975; Kelman, 1963; Lieberman et al., 1973; Lonergan, 1985; Maxmen, 1978; Papanek, 1958; Poey, 1985; Slavson, 1979; Yalom, 1970, 1995); (3) the available research studies on this topic (Berzon et al., 1963; Bloch & Reibstein, 1980; Butler & Fuhriman, 1983; Dickoff & Lakin, 1963; Eckert, Bierman–Ratjen, Tönnies, & Wagner, 1981; Freedman & Hurley, 1979, 1980; Fuhriman et al., 1986; MacKenzie, 1987; Rohrbaugh & Bartels, 1975); and (4) a number of publications representing different therapeutic orientations (Berk, 1986; Glass & Shea, 1986; Perls, 1973; Fagan, & Shepherd, 1970; Rogers, 1970; Yalom, 1970, 1983, 1995). We wrote all items from the perspective of the group member’s experience. Thus, interventions by the therapist or actions of other group members were described in terms of their effect on the group member. For example, the therapist’s empathy was written in terms of how this was experienced by the member. We constructed 155 sentences each expressing a possible therapeutic

208

DIERICK AND LIETAER

experience that a member in a therapy or growth group might have. For each item, the participant was asked to rate whether the description was applicable to his or her group experience on a 3–point scale ranging from not applicable to clearly applicable. If the item was rated somewhat or clearly applicable, the group member then used a second 3–point rating scale to indicate its importance (i.e., that the participant found the experience useful or helpful or that something was learned from it). We opted for rating a meaningful series of group sessions (e.g., a weekend of eight sessions, or “from the last vacation break until now” in a weekly group) instead of either only one session or the whole group therapy. In this way the range of experiences on which one reflects can be sufficiently large, but also sufficiently vivid in memory. This therapeutic factor questionnaire made up the central part of a more extensive battery of questionnaires including an inventory on disturbing or harmful factors, intermediate outcome rating scales, and other possibly intervening individual variables (such as motivation and engagement). The group therapists also filled out questionnaires. In this article we report only on the therapeutic factor model and its relationship to intermediate outcome ratings. Description of the Sample

To be able to conduct multivariate analyses we needed an extensive and heterogeneous sample. We sought a sample that would be representative of the complete domain of psychotherapy and growth groups in the Flemish area. Table 1 shows the distribution of 489 group participants who completed all questionnaires across 78 different groups, categorized by type of group and therapeutic orientation. The following five types were encompassed: inpatient psychotherapy groups, outpatient psychotherapy groups, personal growth groups, psychotherapeutically-oriented learning groups for psychotherapists in training, and “practicum group psychotherapy” groups for students in clinical psychology focus-

Therapeutic Orientation Client–centered Gestalt Psychoanalytic Behavioral Drama & Body Other N % Groups N % 29 37%

16 21%

31 4 3 1 53 11%

Outpatient 57 6 0 37 44 4 148 30%

Inpatient 14

9 12%

14 9 77 16%

9 6 87 18% 11 14%

Training 27 11 16

Clients

Growth 27 42

Type

13 17%

124 25%

Student 35 18 35 36

TABLE 1. Numbers of Clients and Groups by Type and Therapeutic Orientation

N 160 77 82 77 70 23 489

100%

% 33% 16% 17% 16% 14% 5%

78

N 26 9 14 15 10 4

Groups

100%

% 33% 12% 18% 19% 13% 5%

THERAPEUTIC FACTORS 209

210

DIERICK AND LIETAER

ing on personal development. Therapeutic orientations included: client–centered/experiential, Gestalt, psychoanalytic, behavioral, and drama– and body–oriented orientation. The last category encompasses therapeutic approaches that have in common working with drama–and/or bodily-oriented techniques and being experiential in the broad sense; they include Psychodrama, Pesso– (psychomotor) psychotherapy, bonding psychotherapy, bio–energetic Reichian therapy, and “Bodily expression and creativity with Gestalt.” RESULTS The Cluster– and Factor–Analytically Grounded Model of Therapeutic Factors

A matrix of gamma coefficients on the applicability ratings for the 155 items was calculated, followed by a hierarchical complete linkage cluster analysis (Aldenderfer & Blashfield, 1984; Delbeke, 1986; Nunnally, 1979). The complete linkage cluster analysis method makes high demands upon the homogeneity of the clusters that subsequently form the new entity for the next step in the agglomerative hierarchically clustering procedure, the results of which can be shown in a tree diagram. On a set of alternative clusters, selected from this cluster analysis, several item analytic procedures were done: replication of the cluster analysis on both randomly designed halves of the sample to study the stability of the clusters; calculating a goodness of fit index for a series of cluster solutions; determining the internal consistency (alpha coefficient) and iterative item analyses of preliminary cluster scales; and determining pure homogeneity by indices of inter–item association. These procedures resulted in the definition of 28 basic scales of therapeutic factors that accounted for 133 of the original 155 questionnaire items. Next, an exploratory principal factor analysis with orthogonal varimax rotation was conducted on these 28 basic scales, yielding a two–factor solution (Table 2). The first factor in the factor analysis

211

THERAPEUTIC FACTORS

TABLE 2. Factor Loadings of the 28 Basic Scales (N = 489) Basic Scale 1: Acceptance 2: Cohesive working group 3: Empathy & support 4: Expressing mutually positive feelings 5: Experiencing positive feelings encountering others 6: Finding self–confidence in helping others 7: Authentic self–expression 8: Intimate self–disclosure in a receptive climate 9: Getting things off one’s chest 10: Confidence in the group therapist 11: Understanding others 12: Being wrapped up in helping others 13: Making progress in relating to others 14: Self–understanding 15: Insight in and corrective re–experiencing of problems from primary family relationships 16: Becoming conscious of existential responsibility 17: Eliciting awareness by an event in the group 18: Discovering similarity with others 19: Hope by progress in others 20: Discovering the universality of problems 21: Getting suggestions from members 22: Learning a method to master a problem 23: Guidance from group therapist 24: Getting interpersonal feedback 25: Expressing negative feelings toward others 26: Modeling: imitation/trying out 27: Recognizing one’s transferences in the group 28: Expressing negative feelings toward the therapist Total communality Explained variance Relative Percentage

Psychological Work 0.018 0.079 0.288 0.370 0.438 0.437 0.238 0.393 0.491 0.167 0.311 0.195 0.663 0.700

Relational Climate 0.838 0.752 0.720 0.619 0.637 0.595 0.617 0.569 0.511 0.547 0.555 0.359 0.434 0.371

0.691 0.647 0.462 0.460 0.537 0.565 0.579 0.626 0.449 0.492 0.379 0.555 0.566 0.260

0.231 0.182 0.396 0.328 0.398 0.283 0.254 0.256 0.233 0.279 0.127 0.256 0.060 0.012

6.091 21.8% 50.8%

5.893 21.0% 49.2%

before rotation accounted for 76% of explained variance. This global factor demonstrates the overall high level of intercorrelations among the basic scales. The two factors we found after rotation were interpreted as Relational Climate and Psychological Work, and each accounted for about 50% of the explained variation. For each of these two dimensions we composed scales consisting of the 10 basic scales that loaded at least .49 on its own

212

DIERICK AND LIETAER

dimension and had a significantly lower loading (at least differing |.15|) on the other dimension. Between the basic scales that load high on the same factor we see further similarities and differences in meaning and patterns in the correlation matrix which are not revealed by the factor analysis. To further investigate relationships among the 28 basic scales, we conducted a second order cluster analysis on the correlation matrix of these scales. The hierarchical cluster solution yielded seven clusters of second order that map out 22 of the 28 first–order basic scales in a very meaningful way. We named these main clusters: (1) Group Cohesion, (2) Interactional Confirmation, (3) Cathartic Self–revelation, (4) Self–Insight and Progress, (5) Observational Experiences, (6) Getting Directives, and (7) Interactional Confrontation. With the two-factor analytically derived dimensions the latent variables or underlying constructs are revealed. The second order cluster analysis groups the manifest variables—the 28 basic scales—in a direct way and creates an intermediate level of abstraction between the more concrete basic scales and the more abstract two dimensions. The two dimension scales and the seven main scales are made up of all items of the basic scales by which they are composed. Besides the seven main scales we distinguish six separate basic scales, three within each dimension, that are not covered by a main scale. Although their minimum correlations with other main scales were under the cut–off line (i.e., less than minimum correlation 0.45), these separate basic scales each represent a meaningful therapeutic factor that has a meaningful loading on the two-factor analytic dimensions. In Table 3 we provide an overview of our model with a brief description of the 28 basic scales; in these descriptions we give the most weight to the central items that have the highest correlation with the basic scale. Interconnection and Differentiation of the Main Scales

In Table 4 some important aspects of the dimension and main scales are presented. The seven main scales can be considered as

THERAPEUTIC FACTORS

213

the core aspects of client perception of therapeutic factors. What strikes us here are the high intercorrelations among the scales, the pattern of which appears to be very meaningful. Within this pattern three main scales—Interactional Confirmation, Cathartic Self–Revelation, and Self–Insight and Progress—show very high intercorrelations, correlate also relatively high with the other main scales (from both dimensions), and show the highest correlations with the dimension scale of the other dimension (where they do not form part of). Therefore, these three scales appear to be central therapeutic factors, strongly associated with all therapeutic process components. The other four main scales show very varying correlations. The Therapeutic Proof of the Therapeutic Factors

Can we account for the epithet “therapeutic” in “therapeutic factors”? Are these factors experienced as therapeutic? And do these factors really make a difference between group members who make therapeutic progress and those who do not? We have at our disposal two sorts of data that address these questions. When group members rated an item as somewhat or clearly applicable, they were asked to indicate to what degree they found this important (helpful or instructive). As shown in Table 4, the mean degree of importance is rated very high and varies only to a small extent across the scales, obviously less than the degree of applicability. Our findings from the inquiry of the degree of importance confirm that the items we formulated for our questionnaire indeed are experienced by the group members as therapeutic. As to future research we conclude that rating the degree of importance of these therapeutic factors should not be necessary; the question can be restricted to the rating scale of the applicability. In line with Greene’s (2003) recommendations to link process data to outcome data, we made use of a number of ratings of intermediate outcome, ratings that were taken at the same time as the therapeutic factors questionnaire, after a series of group sessions. We present the results of the following three measures: (1) Client

214

DIERICK AND LIETAER

TABLE 3. Overview of Dimensions, Main Scales (Ms) and Basic Scales (Bs) (n = Number of Items) Dimension 1: Relational Climate Ms1 Group Cohesion (n = 16) consists of the following three basic scales: Bs1 Acceptance (n = 7): feeling accepted as I am, appreciated as a person, belonging. Bs2 Cohesive working group (n = 6): experiencing an atmosphere of team–spirit in the group. Bs3 Empathy and support (n = 3): experiencing empathy, support, involvement and warmth. Ms2 Interactional confirmation (n = 15) consists of the following three basic scales: Bs4 Expressing mutually positive feelings (n = 4): able to express positive feelings toward others; others telling me what they found attractive in me. Bs5 Experiencing positive feelings in encountering others (n = 5): experiencing that a genuine encounter with an other human being is possible; in touch with positive feelings in myself. Bs6 Finding self–confidence in helping others (n = 5): by my contribution in the group, increased belief that I had something to offer to others. Ms3 Cathartic self–revelation (n = 15) consists of the following three basic scales: Bs7 Authentic self–expression (n = 4): being able to show how I really felt instead of hiding my feelings, to express in a good way the things I felt. Bs8 Intimate self–disclosure in a receptive climate (n = 5): being able to express intimate and difficult personal things which I usually do not tell others and the group reacting openly and receptively. Bs9 Getting things off one’s chest (n = 6): being able to ventilate fee−lings repressed for so long; getting in better touch with what I feel inside. Separate basic scales with highest load on dimension 1 Relational Climate: Bs10 Confidence in the group therapist (n = 7): being optimistically disposed by seeing the group therapist having trust in him– or herself and in the helping capacities of the group; feeling really understood by the group therapist. Bs11 Understanding others (n = 5): learning to understand and respect other people in their being different; better understanding my own reactions to another group member by what he or she told about him– or herself. Bs12 Being wrapped up in helping others (n = 2): being able to put other’s needs above mine; forgetting myself and my problems. Dimension 2: Psychological Work Ms4 Self–insight and progress (n = 26) consists of the following four basic scales: Bs13 Making progress in relating to others (n = 7): learning to relate in a more satisfying way; trying out new ways of reacting; and experiencing progress. Bs14 Self–understanding (n = 12): becoming better aware of the heart of my problem; getting insight into the origin of my problems; seeing the connection between my problems and underlying feelings or conflicts. Bs15 Insight into and corrective re–experiencing of problems from primary family relationships (n = 4): discussing what I experienced with the group therapist(s) or other group members has brought me insight into the problems I’ve had in the past or still have with my parents, brothers and sisters, and learning to handle them in a better way. Bs16 Becoming conscious of existential responsibility (n = 3): gaining insight that I myself have to give meaning to my life and assume responsibility for the life I lead, no matter how much support and guidance I get from others.

THERAPEUTIC FACTORS

215

TABLE 3. (continued) Ms5 Observational experiences (n = 13) consists of the following four basic scales: Bs17 Eliciting awareness by an event in the group (n = 3): some event processed in the group evoked many thoughts in me; learning by seeing another member working on some aspect of himself. Bs18 Discovering similarity with others (n = 3): discovering others had similar problems or experiences as me and deriving benefit from the work of the group with those similar problems. Bs19 Hope by progress in others (n = 3): seeing or hearing other group members making progress thanks to the group was encouraging to me. Bs20 Discovering the universality of problems (n = 4): recognizing that we’re all in the same boat, that I’m not the only one with problems, unhappy experiences or “unacceptable” thoughts or feelings, took away a great deal of negative feelings I had toward myself. Ms6 Getting directives (n = 15) consists of the following three basic scales: Bs21 Getting suggestions from members (n = 5): receiving practical tips, useful information, sugges−tions about how I could handle particular life problems, advice about how I could relate in a better way to an important person in my life. Bs22 Learning a method to master a problem (n = 4): learning to deal with my problems or complaints in a systematic way, getting explanations, and learning what I could do in a particular situation by seeing a group member reacting in an adaptive manner. Bs23 Guidance from group therapist (n = 6): receiving proposals from the group therapist to try out at home or at work something I had learned in the group, to deal in a particular way with a particular problem I discussed, or to try a particular technique or exercise. Ms7 Interactional confrontation (n = 10) consists of the following two basic scales: Bs24 Getting interpersonal feedback (n = 7): hearing what thoughts and/or feelings I evoke in others and how I am perceived by others, learning what they appreciate in my way of being and behaving and what they would like to be changed. Bs25 Expressing negative feelings toward others (n = 3): being able to say to an other group member that he or she was getting on my nerves; that I had been hurt by what he or she had done; being able to be really angry in the group. Separate basic scales with highest load on dimension 2 Psychological Work: Bs26 Modeling imitation/trying out (n = 3): taking advantage of seeing how other members react to find out how I want to relate to others; trying out particular behaviors or attitudes that appealed to me in other group members or therapist. Bs27 Recognizing one’s own transference reactions in the group (n = 4): recognizing that the sympathy or antipathy I felt for particular persons in the group, had to do not with these persons themselves but rather with other persons in the past, especially with my father or mother. Bs28 Expressing negative feelings toward the group therapist (n = 2): being able to directly express negative feelings or expressing my opinions toward the group therapist.

perception of satisfaction (varying from 1 = extremely dissatisfied to 8 = extremely satisfied); (2) Client perception of change (ranging from 1 = very obviously worse to 7 = very obviously better); and (3) Therapist perception of change (ranging from 1 = clearly changed positively to 5 = clearly declined).

216

DIERICK AND LIETAER

As shown in Table 4, the correlations of the two dimension scales and the main scales Ms1 to Ms4 (Group Cohesion, Interactional Confirmation, Cathartic Self–Revelation and Self–Insight and Progress) with these outcome measures are moderately high and significant (p ≤ .0001). A lower but not negligible and almost always significant positive correlation with each of the outcome measures (p < .001, mostly p ≤ .0001) is seen for Observational Learning, Getting Directives and Interactional Confrontation. The only correlation (p < .10) not reaching significance is between Interactional Confrontation and client rating of satisfaction. Client satisfaction correlates most strongly with Group Cohesion and the least with Interactional Confrontation, suggesting that client satisfaction rating has a hedonistic quality, with a positive group feeling at the one end of the continuum and the hard work of interactional confrontation at the other end. Client perception of change correlates the highest with Self–Insight and Progress, the main scale that also correlates the most with the dimension of Psychological Work. Finally, therapist perception of client change generally shows lower correlations with the process measures than the two client–rated measures. This does not surprise us as it concerns correlations between two different perspectives. Therapists’ perceptions do correlate most strongly with Cathartic Self–Revelation and relatively lower with Observational Experiences. Therapists apparently value the extent to which group members “throw themselves emotionally into the group” and seem to have very little sight on what happens behind the scenes as observational experiences. These findings concerning the connections with intermediate outcome measures and their meaningful differential pattern further support the validity of the therapeutic factors in our model. DISCUSSION

What do we learn from this empirically grounded model of therapeutic factors in group psychotherapy and growth groups? On the

53 .95 .76 2.35 .33 .48 .34

52 .95 1.06 2.39 .53 .53 .37

Number of items Cronbach’s Applicability (0—2): X Importance (1—3): X Correlations of Applicability with: Satisfaction with personal result of sessions Client perception of personal change Therapist perception of client change

— — — —

.65 .59 .48 .45

Ms4 Self—insight & progress Ms5 Observational experiences Ms6 Getting directives Ms7 Interactional confrontation

.46 .70 .67

—

— — —

.67

Ds2

Main scales Ms1 Group cohesion Ms2 Interactional confirmation Ms3 Cathartic self—revelation

Dimension scales Ds1 Relational climate Ds2 Psychological work

Ds1

.52 .45 .30

16 .91 1.18 2.40

.46 .47 .39 .22

— .64 .62

Ms1

.42 .49 .34

15 .89 .88 2.37

.66 .56 .44 .55

— .67

Ms2

.41 .45 .46

15 .90 .84 2.41

.67 .49 .42 .49

—

Ms3

.37 .53 .37

26 .93 .84 2.44

— .65 .65 .47

Ms4

.28 .33 .16

13 .87 .93 2.21

— .56 .32

Ms5

.34 .35 .24

15 .88 .63 2.26

— .27

Ms6

.08 .25 .21

10 .87 .74 2.36

—

Ms7

TABLE 4. Dimension Scales (Ds) and Main Scales (Ms): Intercorrelations, Number of Items, Reliability (Cronbach’s α), Mean Applicability and Importance Scores, Pearson Correlations Between Applicability Scores and Intermediate Outcome Measures (N = 489)

THERAPEUTIC FACTORS 217

218

DIERICK AND LIETAER

basis of a broad span of possible therapeutic experiences from many therapeutic orientations and types of group therapy, our cluster and factor analyses yielded three levels of abstraction: two dimensions, seven main scales, and 28 basic scales. Unlike previous attempts at taxonomies (Bloch & Crouch, 1985; Yalom, 1995) our statistically derived therapeutic factors are consistent with respect to level of abstraction and also explicitly show relationships to each other. Two dimensions of Therapeutic Factors

The two–dimensional structure we found is in line with a good deal of research on therapeutic processes that broadly distinguish relational–emotional and task aspects of psychotherapy. For example, referring to the general model on therapeutic processes in psychotherapy (Orlinsky, Grawe, & Parks, 1994), our two dimensions—Relational Climate and Psychological Work—seem to correspond to the notions of “therapeutic bond” and “therapeutic operations,” respectively. Our model of the factor analytically derived two dimensions and cluster analytically derived seven main scales gives us deeper insight into the coherence of and differentiation among its components. As to the Relational Climate dimension, the strongest and most purely loading basic scales were Acceptance and Cohesive Working Group (both from the main scale Group Cohesion) and Confidence in the Group Therapist. On the Psychological Work dimension, the most purely loading basic scale is Recognizing One’s Own Transference Reactions in the Group. The strongest loadings, however, came from the four basic scales that form the main scale Self–Insight and Progress. These scales, as do a majority of basic scales, have a higher loading on one of the two dimensions but also a considerable loading on the other dimension. However, as to the basic scales Acceptance, Cohesive Working Group and Confidence in the Group Therapist, the loading is clearly high on the Relational Climate dimension and rather low on the Psychological Work dimension, suggesting that a group

THERAPEUTIC FACTORS

219

member’s experience of a high level of bond–aspects does not necessarily mean that he or she has the experience of much work being accomplished. Comparison of Our Work to Yalom’s scales

How do our results compare to Yalom’s (1995) measure of 12 therapeutic factors? First, our data suggest that Yalom’s Catharsis scale may not be homogeneous. Several aspects appear to split off and become part of different main scales. Getting Things Off One’s Chest is part of the main scale Cathartic Self–Revelation and appears to encompass as much Relational Climate as Psychological Work. Expressing Mutually Positive Feelings is part of the main scale Interactional Confirmation and appears to contribute primarily to the Relational Climate dimension. In contrast, Expressing Negative Feelings Toward Others is part of the main scale Interactional Confrontation, while Expressing Negative Feelings Toward the Group Therapist appears to be a very specific and separate factor. Both these basic scales appear to reflect aspects of Psychological Work in contrast to Relational Climate. The different elements of Yalom’s concept of Interpersonal Learning are represented in our basic scales, but are split up into two main scales. The basic scales that refer to Yalom’s scales of Self–Understanding, Family Re–Enactment, Existential Responsibility, and Interpersonal Learning–Output are grouped in our statistically grounded model under our main scale Self–Insight and Progress. On the other hand, Yalom’s factor Interpersonal Learning–Input is found in our basic scale Getting Interpersonal Feedback, that forms part of the main scale Interactional Confrontation, since it is clustered with a second basic scale Expressing Negative Feelings Toward Others. As to Group Cohesion our model distinguishes the different concepts that in the research literature on therapeutic factors are confused. Yalom’s concept of Group Cohesiveness as the analogue of relationship in individual therapy we find in the dimension Relational Climate. Group Cohesion, the term we have re-

220

DIERICK AND LIETAER

served for the more specific therapeutic process in which we distinguish three differentiated basic scales: (1) Acceptance, incorporating the participant’s sense of belonging; (2) Cohesive Working Group, reflecting a team–spirit of dedication and genuineness; and (3) Empathy and Support, reflecting how others react to what a group member reveals. In these three basic scales we also (partly) recognize the three therapeutic basic ingredients described by Rogers (1957, 1959); namely, (1) unconditional positive regard, (2) authenticity, and (3) empathy, as the group member experiences them. Like other studies (cf. Piper, Marrache, Lacroix, Richardsen, & Jones, 1983) our research thus suggests that group cohesion is multifaceted. Our model of therapeutic factors reveals a main scale, Observational Experiences, composed of basic scales that appear to have in common the experience of the participant as a sympathizing and self–reflecting observer of the process in other members in the group. Here we recognize three traditional factors that are discussed separately in the literature: (1) Discovering the Universality of Problems, in which one feels liberated from the feeling of being an exceptional case; (2) Eliciting Awareness by an Event in the Group and Discovering Similarity with Others, where we recognize parts of Bloch and Crouch’s (1985) factor Vicarious Learning; and (3) an aspect of Instillation of Hope, called Hope by Seeing Progress in Others. We find three basic scales that focus on experiences about the group therapist: Confidence in the Group Therapist, Guidance from the Group Therapist and Expressing Negative Feelings Toward the Group Therapist. Furthermore, there remain two separate items that refer to the therapist and take a position apart: “being able to express positive feelings toward the group therapist” and “experiencing the group therapist as a model for particular values, behaviors or ways of being that one wants to adopt oneself.” Our cluster analyses suggest that members clearly distinguish therapeutic processes related to other members from those that relate to the therapist concerning similar contents (relationship qualities, guidance/suggestions, expressing negative feel-

THERAPEUTIC FACTORS

221

ings, expressing positive feelings, modelling). The finding that no main scale emerged for all therapist–related items suggests that the nature or content of the experience toward the therapist also makes a difference. For example, one can experience a high degree of confidence in the group therapist without experiencing guidance from the therapist. These findings are in line with Johnson, Burlingame, Olsen, Davies, and Gleave’s (2005) conclusion that process studies should examine member–leader, m e m b e r – m e m b e r , a n d me m b e r – g r o u p r e l a t i o n s h i p s simultaneously and across more than one construct at a time. In our analyses Yalom’s factor Altruïsm is reflected in two different basic scales: Finding Self–Confidence in Helping Others (part of the main scale Interactional Confirmation, showing relatively high correlations with other therapeutic factors) and Being Wrapped up in Helping Others (which appears to be a very specific experience). Furthermore, we find a third basic factor, Understanding Others, in which the group member is focused on others. Insofar as this factor encompasses a learning of social skills we can partly recognize this basic scale in Yalom’s Development of Socializing Techniques. We did not find empirical support for Yalom’s scale Existential Factors as such. In our results one central existential factor is found: Becoming Conscious of Existential Responsibility. This basic factor clusters the items that express the themes freedom (or self determination), responsibility, and meaninglessness (versus giving of meaning) (cf. Yalom, 1980). This basic scale is part of the main scale Self–Insight and Progress and shows a high loading on our dimension of Psychological Work. Limitations and Importance of this Research

We investigated client perception of therapeutic factors in group psychotherapy and growth groups. From this viewpoint, based on Yalom’s (1995) work; Crouch, Bloch, and Wanlass’ (1994) definition; and our own research, we propose to define therapeutic factors as: experiences the client in group can have, which can be a func-

222

DIERICK AND LIETAER

tion of the actions of the group therapist, the other group members, and the client himself, and which can contribute—in an intricate interplay—to improvement in the client’s condition. These experiences can occur in several degrees of consciousness: from actively present in consciousness to preconscious, implicit, but recognizable for the client. Within the perception or experience of the client, therapeutic factors can be viewed and concepts can be used at different levels of abstraction. Besides the client perception, therapeutic factors can be studied and conceptualized from other perspectives; for instance, therapist perception, independent observers, process–outcome research, and (quasi)experimental research. In our definition we find the term experiences that the client in group can have better reflecting what is meant in ours and other research studying the client perception of therapeutic factors, and more specific than Crouch et al.’s (1994) broad and vague term “an element of group therapy.” Their more general term we find suited to encompass all possible meanings a therapeutic factor can have from different perspectives. From the perspective of client perception we consider therapeutic factors as intervening variables, covering the whole range of: the client’s experience of climate elements in the group; how one has experienced particular interventions by therapists or members; how one has experienced his or her own interventions and behavior; and what one has experienced, thought, and felt even without its being verbalized. Some therapeutic factors can be situated as experienced context factors, others as learning experiences, and others as experienced consequences (in the flow of the process) of learning experiences (in the sphere of cognition, feeling and behavior). We believe that the implicit experience of therapeutic processes can be better evoked via a memory recognition task, implied in our structured questionnaire method, than when using a memory recall task, as in open–ended questions. In this way we try to trace also preconscious, implicit, but recognizable experiences of the client. Therefore, it was necessary to inventory and formulate in a

THERAPEUTIC FACTORS

223

well recognizable way all relevant therapeutic experiences, which we did as thoroughly as possible, grounded in literature study and in our own pilot research with open-ended questions (Dierick & Lietaer, 1990). On the basis of this substructure, via multivariate analyses we found a model that reveals the structure and connections of the therapeutic factors as they are differentiated in the experience of the group members. With this empirically-grounded integrative framework and the corresponding questionnaire, we hope that we have set a step further in the clarification and refinement of the multidimensional conceptions and measurement of client perceptions of therapeutic factors in group psychotherapy and growth groups. The therapeutic factors found already had good content validity, grounded in our literature study and pilot research. Construct validity is inherent in the statistical foundation of our model, but requires cross–validation to raise confidence in the stability of the findings. These therapeutic factors proved to be experienced by the group members as generally very “important, helpful, or instructive” and correlate very significantly with three measures of intermediate outcome. Validity is supported by these findings, but further research is needed to establish criterion validity with actual outcome measures. This kind of process research in general and our contribution in particular should be useful for the practice of group psychotherapy, for further research, and for theory building. To develop models of group therapy and to maximize the action of therapeutic factors for particular kind of clients, the group therapist can use our empirically derived model as a frame of reference for the client perception of therapeutic factors. Compared to the existing taxonomies (Yalom, 1995, recently operationalized in the TFI by Lese & MacNair Semands, 2000; Crouch et al., 1994), our model offers a more differentiated survey that distinguishes what to a certain degree is distinguished in the experience of the group members. Clinically, during a group session, the therapist might wonder:

224

DIERICK AND LIETAER

Which therapeutic factors might be active in the different group members? Which therapeutic factors that are inactive at the moment could be stimulated to foster the group process now? What does this group member need now, which therapeutic factors can best help his process further, and how can we advance this? Findings from this and other research comparing what therapists thought helpful for their group members and what clients reported as helpful for themselves (Bloch & Reibstein, 1980; Bonney, Randall, & Cleveland, 1986; Dierick & Lietaer, 1990; Hobbs, Birtchnell, Harte, & Lacey, 1989; Schaffer & Dreyer, 1982) raise the thought that group therapists may have less insight into process aspects that are experienced by the group members but are not directly reflected in what they say or do. Taking this into account, it seems important that the group therapist from time to time ask what is going on in the mind of each of the group members, what they experience as helpful or instructive in the group process, and what they are missing. Our questionnaire (Dierick & Lietaer, 2002) cannot only be used as a research instrument but also as a clinical tool by the practicing group therapist as a sort of monitoring of the experience of therapeutic factors in the clients after a session or a series of sessions. Further fundamental research can focus on investigating the stability of the therapeutic factor model in a replication study, contributing to the validity of these therapeutic factors in their connections with outcome measures, and studying connections with other process variables such as harmful or disturbing factors and therapeutic interventions. This kind of process research can contribute to theory building, as Bednar and Kaul (1994) stated, “to establish more consistency and depth in multidimensional conceptions and measurement of central therapeutic factors.” For example, Group Cohesion as a therapeutic factor we defined on the basis of our research findings in terms of its more specific basic components Acceptance, Cohesive Working Group, and Empathy and Support, and in its being a part of the more abstract and encompassing concept of Group Relational Climate. In doing this we delineate these concepts and relate them to each other, being

THERAPEUTIC FACTORS

225

aware that in other contexts these concepts may show different meanings and interrelations. Further progress in research and theory building can be made when unanimity grows concerning the specific meaning and interrelationships of the concepts used.

REFERENCES Aldenderfer, M.S., & Blashfield, R.K. (1984). Cluster analysis. Beverly Hills, CA: Sage. Bednar, R.L., & Kaul, T.J. (1994). Experiential group research: Can the canon fire? In A.E. Bergin and S.L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th ed., pp. 631–663). New York: Wiley. Berk, T. (1986). Groepstherapie: Theorie en techniek [Group therapy: Theory and technique]. Deventer: Van Loghum Slaterus. Berzon, B., Pious, C., & Farson, R.E. (1963). The therapeutic event in group psychotherapy: A study of subjective reports by group members. Journal of Individual Psychology, 19, 204–212. Bloch, S., & Crouch, E. (1985). Therapeutic factors in group psychotherapy. Oxford: Oxford University Press. Bloch, S., Crouch, E., & Reibstein, J. (1981). Therapeutic factors in group psychotherapy: A review. Archives of General Psychiatry, 38, 519–526. Bloch, S., & Reibstein, J. (1980). Perceptions by patients and therapists of therapeutic factors in group psychotherapy. British Journal of Psychiatry, 137, 274–278. Bloch, S., Reibstein, J., Crouch, E., Holroyd, P., & Themen, J. (1979). A method for the study of therapeutic factors in group psychotherapy. British Journal of Psychiatry, 134, 257–263. Bonney, W.C., Randall, D.A., & Cleveland, J.D. (1986). An analysis of client–perceived curative factors in a therapy group of former incest victims. Small Group Behavior, 17, 303–321. Burlingame, G.M., MacKenzie, K.R., & Strauss, B. (2004). Small group treatment: Evidence for effectiveness and mechanisms of change. In M.J. Lambert (Ed.), Bergin and Garfields’ handbook of psychotherapy and behavior change (5th ed., pp. 647–696). New York: Wiley. Butler, T., & Fuhriman, A. (1983). Level of functioning and length of time in treatment: Variables influencing patients’ therapeutic experi-

226

DIERICK AND LIETAER

ence in group psychotherapy. International Journal of Group Psychotherapy, 33, 489–505. Cabral, R.J., Best, J., & Patton, A. (1975). Patients and observers assessments of processes and outcome in group therapy: A follow–up study. American Journal of Psychiatry, 132, 1052–1054. Corsini, R., & Rosenberg, B. (1955). Mechanisms of group psychotherapy: Processes and dynamics. Journal of Abnormal and Social Psychology, 51, 406–411. Crouch, E., Bloch, S., & Wanlass, J. (1994). Therapeutic factors: Interpersonal and intrapersonal mechanisms. In A. Fuhriman and G.M. Burlingame (Eds.), Handbook of group psychotherapy: An empirical synthesis (pp. 269–315). New York: Wiley. Delbeke, L. (1986). Inleiding tot de Multivariate methoden in de psychologie [Introduction to the multivariate methods in psychology]. Cursustekst. Leuven: Acco. Dickoff, H., & Lakin, M. (1963). Patients’ views of group psychotherapy: Retrospections and interpretations. International Journal of Group Psychotherapy, 13, 61–73. Dierick, P. (2000). Cliëntperceptie van therapeutische factoren in groepspsychotherapie en groeigroepen: een structureel en vergelijkend onderzoek [Client perception of therapeutic factors in group psychotherapy and growth groups: A structural and comparative research]. Niet–gepubliceerde doctoraatsverhandeling, Katholieke Universiteit Leuven. Dierick, P. (2001). Cliëntbeleving van therapeutische factoren in groepspsychotherapie en groeigroepen: onderzoek en ontwikkeling van een structuurmodel [Client perception of therapeutic factors in group psychotherapy and growth groups: Research and development of a structural model]. In T.J.C. Berk et al. (Eds.), Handboek groepspsychotherapie (Q4: 1–63). Houten/Zaventem: Bohn Stafleu Van Loghum. Dierick, P., & Lietaer, G. (1990). Member and therapist perceptions of therapeutic factors in therapy and growth groups: Comments on a category system. In G. Lietaer, J. Rombauts, & R. Van Balen (Eds.), Client–centered and experiential psychotherapy in the nineties (pp. 741–770). Leuven: Leuven University Press. Dierick, P., & Lietaer, G. (2002). De Groeps Therapeutische Factoren—Cliënt Vragenlijst (GTF–CV); psychometrische en klinische karakteristieken van het structuurmodel [The Group Ther-

THERAPEUTIC FACTORS

227

apeutic Factories—Client Questionnaire (GTF–CQ): Psychometric and clininal characteristics of the structural model]. In T. J.C. Berk et al. (Eds.), Handboek groepspsychotherapie (Q6: 1–70). Houten/Zaventem: Bohn Stafleu Van Loghum. Drob, S., & Bernard, H.S. (1986). Time–limited group treatment of genital herpes patients. International Journal of Group Psychotherapy, 36, 133–144. Drob, S., Bernard, H., Lifshutz, H., & Nierenberg, A. (1986). Brief group psychotherapy for herpes patients: A preliminary study. Behavior Therapy, 17, 229–238. Eckert, J., Bierman–Ratjen, E.M., Tönnies, S., & Wagner, W. (1981). Heilfactoren in der gruppenpsychotherapie: Empirische Untersuchungen über wirksame Faktoren im gruppenpsychotherapeutischen Prozess [Curative factors in group psychotherapy: Empirical research on active factors in the group p s y c h o t h e r a p e u t i c pr o c e s s ] . G r u p p e n p s y c h o t h e r a p i e un d Gruppendynamik, 17, 142–162. Fagan, J., & Shepherd I.L. (1970). Gestalt therapy now. Palo Alto: Science and Behaviour Books. Farell, M.P. (1962). Transference dynamics of group psychotherapy. Archives of General Psychiatry, 6, 66–76. Freedman, S.M., & Hurley, J.R. (1979). Maslow’s needs: Individual perceptions of helpful factors in growth groups. Small Group Behavior, 10, 355–367. Freedman, S.M., & Hurley, J.R. (1980). Perceptions of helpfulness and behavior in groups. Group, 4, 51–58. Fuhriman, A., Drescher, S., Hanson, E., Henrie, R., & Rybicki, W. (1986). Refining the measurement of curativeness: An empirical approach. Small Group Behavior, 17, 186–201. Glass, C.R., & Shea, C.A. (1986). Cognitive therapy for shyness and social anxiety. In W. H. Jones, J.M. Cheek, & S.R. Briggs (Eds.), Shyness: Perspectives on research and treatment (pp. 315–327). New York: Plenum. Greene, L.R. (2003). The state of group psychotherapy process research. International Journal of Group Psychotherapy, 53, 130–134. Hill, W. F. (1957). Analysis of interviews of group therapists. Provo papers, 1, 1. Cited in W. F. Hill (1975). Further considerations of therapeutic mechanisms in group therapy. Small Group Behavior, 6, 421–429.

228

DIERICK AND LIETAER

Hill, W. F. (1975). Further considerations of therapeutic mechanisms in group therapy. Small Group Behavior, 6, 421–429. Hobbs, M., Birtchnell, S., Harte, A., & Lacey, H. (1989). Therapeutic factors in short–term group therapy for women with bulimia. International Journal of Eating Disorders, 8, 623–633. Johnson, J.H., Burlingame, G.M., Olsen, J.A., Davies, D.R., & Gleave, R.L. (2005). Group climate, cohesion, alliance, and empathy in group psychotherapy: Multilevel structural equation models. Journal of Counseling Psychology, 52, 310–321. Kahn, E.M., Webster, P.B. & Storck, M.J. (1986). Brief reports: Curative factors in two types of inpatient psychotherapy groups. International Journal of Group Psychotherapy, 36, 579–585. Kelman, H.C. (1963). The role of the group in the induction of therapeutic change. International Journal of Group Psychotherapy, 13, 399–432. Lese, K.P., & MacNair–Semands, R.R. (2000). The Therapeutic Factors Inventory: Development of a scale. Group, 24, 303–317. Leszcz, M., Yalom, I.D., & Norden, M. (1985). The value of inpatient group psychotherapy: Patients’ perceptions. International Journal of Group Psychotherapy, 35, 411–433. Lieberman, M.A., Yalom, I.D., & Miles, M.B. (1973). Encounter groups: First facts. New York: Basic Books. Lonergan, E.C. (1985). Utilizing group process in crisis–waiting–list groups. International Journal of Group Psychotherapy, 35, 355–372. MacKenzie, K.R. (1987). Therapeutic factors in group psychotherapy: A contemporary view. Group, 11, 26–34. Maxmen, J.S. (1978). An educative model for inpatient group therapy. International Journal of Group Psychotherapy, 29, 321–338. Nunnally, J.C. (1979). Psychometric theory. New York: McGraw–Hill. Orlinsky, D.E., Grawe, K., & Parks, B.K. (1994). Process and outcome in psychotherapy—Noch einmal. In A.E. Bergin & S.L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th ed., pp. 631–663). New York: Wiley. Papanek, H. (1958). Ethical change of values in group psychotherapy. International Journal of Group Psychotherapy, 8, 435–444. Perls, F. (1973). Gestalt therapy verbatim. Lafayette, CA: Real People Press. Piper, W.E., Marrache, M., Lacroix, L., Richardsen, A.M., & Jones, B.D. (1983). Cohesion as a basic bond in groups. Human Relations, 36, 93–108.

THERAPEUTIC FACTORS

229

Poey, K. (1985). Guidelines for the practice of brief, dynamic group therapy. International Journal of Group Psychotherapy, 35, 331–354. Rogers, C.R. (1957). The necessary and sufficient conditions of therapeutic personality change. Journal of Consulting Psychology, 21, 97–103. Rogers, C.R. (1959). A theory of therapy, personality and interpersonal relationships as developed in the client–centered framework. In S. Koch (Ed.), Psychology: A study of science. Vol. III: Formulations of the person and the social context (pp. 184–256). New York: McGraw–Hill. Rogers, C.R. (1970). Carl Rogers on encounter groups. New York: Harper and Row. Rohrbaugh, M., & Bartels, B.D. (1975). Participants’ perceptions of curative factors in therapy and growth groups. Small Group Behavior, 6, 430–456. Schaffer, J.B., & Dreyer, S.F. (1982). Staff and inpatient perceptions of change mechanisms in group psychotherapy. American Journal of Psychiatry, 139, 127–128. Slavson, S.R. (1954, 1979). A systematic theory. In M. Schiffer (Ed.) Dynamics of group psychotherapy. New York: Jason Aronson. Snijders, J.A., Van Der Spek, W.A., Colijn, S., Hoencamp, E., & Duivenvoorden, H.J. (1991). C uratieve factoren in groepspsychotherapie [Curative factors in group psychotherapy]. Groepspsychotherapie (Verenigingsblad van de Nederlandse Vereniging voor Groeps–Psychotherapie), 25, 16–25. Stone, M.H., Lewis, C.M., & Beck, A.P. (1994). The structure of Yalom’s curative factors scale. International Journal of Group Psychotherapy, 44, 239–245. Yalom, I.D. (1970). The theory and practice of group psychotherapy. New York: Basic Books. Yalom, I.D. (1975). The theory and practice of group psychotherapy (2nd ed.). New York: Basic Books. Yalom, I.D. (1980). Existential psychotherapy. New York: Basic Books. Yalom, I.D. (1983). Inpatient group psychotherapy. New York: Basic Books. Yalom, I.D. (1985). The theory and practice of group psychotherapy (3rd ed.). New York: Basic Books. Yalom, I.D. (1995). The theory and practice of group psychotherapy (4th ed.). New York: Basic Books. Yalom, I.D., Tinklenberg, J., & Gilula, M. (1968). Curative factors in group therapy. Unpublished study. Cited in I.D. Yalom (1995), The

230

DIERICK AND LIETAER

theory and practice of group psychotherapy (4th ed., pp.72–79). New York: Basic Books. Paul Dierick Boerenkrijglaan 21 B–3010 Kessel–lo (Leuven) Belgium E–mail: [email protected]

Received: October 3, 2006 Final draft: June 27, 2007 Accepted: July 15, 2007