Q Studies of Leadership and Learning

Presentation Prepared for

A CELEBRATION OF THE LIFE AND WORK OF WILLIAM STEPHENSON (1902-1989)

at the University of Durham, England December 12-14, 1997

Q-STUDIES OF LEADERSHIP AND LEARNING IN A SOCIAL MATRIX: Investigating Group Relations Conferences in the Tavistock Tradition (revised 1/5/98)

ROBERT M. LIPGAR

University of Chicago Medical Center and Chicago Center for the Study of Groups and Organizations of the A.K. Rice Institute

OVERVIEW
Group relations conferences in the A.K. Rice/Tavistock tradition provide opportunities for learning by experience about authority, leadership and covert processes of groups and organizations. These conferences are organized as temporary educational institutions. Their methods and outcomes have been intensively studied but not generally with empirical methods of the social sciences.

Although the conference design and purposes remain quite constant and there is a consistency in the qualifications and orientation of staff, the work and outcomes are uniquely responsive to the particular participants and conditions as these interact in the "here and now" events of each conference. Staff consulting behaviors as well as members‘ participation are influenced by complex interactions among many variables such as individual, group, inter-group and institutional system dynamics and processes.

In examining such institutions scientifically, the use of conventional questionnaires and self-report instruments are generally cumbersome and intrusive as well as incompatible with the educational goals and the primary learning tasks. Because Q-methodology allows for more subtle expressions of individual's subjectivity, collection of data is not disruptive of conference work. Further, Q-data can be submitted to sophisticated statistical analyses, notably factor analysis. This makes possible quantified representations of critical aspects of social matrices and the development as well as testing of hypotheses concerning their processes and dynamics. Q-methodology adapts to both the scientific requirements of steps of discovery as well as those involved in arguing the merits of rival interpretations. Further, Q-data can be collected and analyzed within the live context of conferences as they happen in ways which can be integrated with, and contribute to ‘learning by experience' -- the touchstone of the Tavistock tradition.

This presentation reports major findings about leadership, consultant roles, and the styles of participating and learning of conference staff and members developed over a ten-year period with Q-methodology. The consistency with which certain factors emerge in correlational matrices drawn from different conferences, the conceptual usefulness of the identified factors, and the similarity between conceptualizations founded on Q analyses and those uncovered by other methodologies, point to the power of Stephenson's ideas about research and about how knowing can best be advanced .

PROLOGUE

This is a happy occasion for me. It is an opportunity to participate in a celebration of one of the most influential teachers in my education as a psychologist, William Stephenson (1953). It is an occasion to present work which I believe shows the power of his ideas about research and to do so in a context which relates to the work of another Englishman who had great influence on me, although I never had the opportunity to study with him, Wilfred R. Bion (1961, 1962). For me, their visions are complimentary and inspirational.

My first study of leadership as a student of Stephenson's resulted in three factors which I identified as: 1) Group-centered Egalitarian; 2) Group-centered, Work-oriented; and 3) Leader-centered, Technical Expert. Such findings hardly were of historic import. However, eight years later during my clinical internship, my supervisor and mentor, Meyer Williams, began his seminar on group psychotherapy by outlining these same three possible leadership stances as a way of orienting us to the choices we must make as we take up our professional work as group therapists. My sense that Q-studies could identify useful and important categories and concepts was reenforced.

In my dissertation research (1965) in psychology, I was able to use Q-methodology again to produce factors that identified behavioral patterns operative under several experimental conditions. Specifically, the four factors which emerged from a person-to-person correlational matrix of subjective probability notions (patterns of guessing heads and tails to the flip of a coin) showed effects of how people approached random events and decision-making under other conditions of ambiguity and uncertainty. Effects of these factors could also be found in the structural features of stories told in response to Card #1 of the Thematic Apperception Test (TAT) and of response patterns to the Rorschach inkblots.

The first factor identified in this particular Q-study replicated a finding in another dissertation on subjectivity probability (Lawlor, 1956) at the University of Chicago. With Q-methodology, however, I was able to identify not only the "normal" adult pattern of response but also three additional factors or "deviant" patterns.

This exploration of subjectivity impressed me with the power of Q-methodology. I had grounds to believe that subjectivity -- what goes on inside the "black box" of individuals' psychological functioning -- and individuality itself could be studied scientifically, and that a psychology of subjectivity need not be a quixotic pursuit nor was it necessary to de-humanize the subject in the process of study. Quite the contrary. Scientific psychology need not be confined to a narrow behaviorism examined with rigidly experimentalist, mechanistic procedures.

It was not clear to me then, nor is it now, whether Q-studies limit us in the study of personality to searching only for typologies or archetypes. I think not. As this report of of a program of group relations conference research explains, I believe Q-methodology can effectively give us access to data which take us beyond typologies and archetypes. The scientific study of our complex inter-relatedness within social matrices is, with Q-methodology, within reach.

SMALL GROUP CONSULTANCY IN GROUP RELATIONS CONFERENCES Small Study Group Consultants as Psychological Work Leaders

In 1965 I was introduced by Marshall Edelson to the work of A.K. Rice (1965) and the work of his colleagues at the Tavistock Institute for Human Relations. Marshall Edelson, in addition to his work in groups and psychoanalysis (1970, 1988), had also done a highly original single subject Q-study (1954) with Stephenson during our graduate school days.

Beginning in 1986, together with colleagues in Chicago conducting group relations conferences in the Tavistock tradition, we began to study variations in orientations to the critical role of the small study group consultant. We composed a set of statements describing the consultant's role in small study groups conducted as part of the conference program and based on the work of Bion at the Tavistock Clinic. These 72 statements were sorted by 12 staff members ranging in years of Tavistock conference experience from a couple of years to nearly 20. How staff conceives of, and functions in the role of the small study group consultant is central to the work of the entire conference, to its culture as a learning institution and to its impact.

The first matrix, collected at Northwestern University in 1986 was analyzed under my supervision by David Bradley (1987). Seeking simple structure, orthogonal factors, he found four which he identified as Work, Educative, Nurturing, and Protective. He fund some evidence to support the notion that these were hierarchical in terms of levels of experience and some evidence to support the association of these factors with Bion's (1961) famous categories of Work, and the basic assumptions Dependency (baD), Pairing (baP) and Fight/Flight (baF/F) in that order. Later, I re-examined this matrix and slightly re-interpreted the factors (Lipgar, 1993), discussing them as complimentary components of good consultancy rather than primarily as points on a continuum. I also found that all the high ‘Work' factor loaders had been promoted to higher conference management roles (director or associate director) in subsequent conferences whereas only one of the high loaders in the other three factors had advanced in this way. Even though the size of the sample is very small, only a matrix of 12, these findings bear out Stephenson's view that much could be accomplished with small samples.

In re-examining the ‘factor-arrays" (the items as they were ranked in composite listings representative of the factor), it again appeared to me that these four factors can be substantially associated with Bion's categories. As I prefer to think of them, the four factors can be signify four essential aspects of the consultant's role: 1) Group-Interpretive Analyst; 2) Group Facilitator; 3) Participant-Collaborator; and 4) Protective Manager. I believe that this kind of interpretation and re-interpretation of factors is integral to Q-studies and can be considered a strength of Stephenson's approach to science rather than as a weakness, as I hope that this report will demonstrate further.

Although these factors suggest readily recognizable "types", their real merit and status as evidence of Q's power is better shown, I believe, in two other ways. These same four factors emerged in every one of eight subsequent conference staff matrices. That is, when the Q-sorts which represented each of the four factors, the"factor arrays", were added to the new matrices, a varimax solution again included four factors on which these factor arrays, or prototypes, emerged as very high "loaders". Such consistency in empirical findings is outstanding, I believe, and especially so when we take into consideration that the correlational matrices were composed of conference staffs of different sizes, different personnel and compositions, working under different conference directors.

Further, these factors bear a striking resemblance not only to Bion's categories based on his keen clinical observations, but also to the four functions of leadership, factors reported by Lieberman, Yalom and Miles (1973) in their study of leadership in encounter groups: 1) Meaning Attribution; 2) Caring; 3) Emotional Stimulation; and 4) Executive Functions. (These also can be related to Bion's categories.) The Q-factors and these two sets of categories also bear resemblance to the four essential functions of social systems described by Edelson (1970) in his adaptation of Talcott Parsons' "Theory of Action." Edelson (op. cit.) also draws attention to Freud's four major structural concepts Ego, Id, Ego-Ideal and Super-Ego in the organization of a personality system. I suggest that further studies of the inter-relationship among all these sets of four concepts (developed by using different methods to investigate different psycho-social systems operating under different conditions) would be very worthwhile.

Conference Members' Learning for Leadership

With the 1987 non-residential weekend group relations conference sponsored by the University of Chicago Department of Psychiatry, we began to explore members' views of leadership. We chose this as an initial focus for assessing educational outcomes. We developed a questionnaire and asked conference members to express their views before and after the conference by ranking 68 statements descriptive of leadership behaviors and attitudes (1 to 4) arranged in 17 four-item sets. These data from two conferences, 1987 and 1991, were submitted to multi-variate analyses (Lipgar & Struhl, 1993). Descriptions of good leadership changed during both conferences and in a direction consistent with the educational goals of these "working conferences" in the Tavistock tradition.

Members before the conference presented a portrait of idealized leadership -- leader as hero, the cardboard figure of a "cereal-box", latency-age ego-ideal. "The portrait of the preferred leader that emerges after conference participation . . . retains many of the initial idealized characteristics but is rounded out with greater appreciation and specification of what it takes to function in a leadership role. . . . Instead of a flat portrait in black and white . . . there emerges a description which has form, depth and color -- leadership connected to members at work and to group process." (op. cit. p. 59)

By taking into consideration only those statements which were ranked highest before the conference with those ranked highest after the conference, it is clear that members showed an increased appreciation of "on-the-job" functioning. Items describing the leader in terms of having the "right stuff" were replaced with statements describing leadership interactions with members working as a group.

It was of further interest to those of us responsible for conducting these conferences, that the 1987 conference members seemed more impressed with "true grit" and the importance of persistence and personal responsibility. The 1991 conference members, however, came to value leadership which was openly attuned to the emotional life of the group and actively engaged in protecting group functioning (op. cit. p. 65). Greater changes were associated with participation in the 1991 conference than in the 1987 conference, and these were in the direction considered by staff to be desirable. Since the 1987 conference consisted of two and not three of the "here and now" components of the conference design, and the staff in general were less experienced (probably less expert) these findings were not surprising.

For both conferences, staff orientations toward the small study group consultant's role were quantified in the form of Q-sorts, correlational matrices, and factor analyses. We can, therefore, state with some confidence in our objectivity, that the culture of the two staff groups differed. The conference staff in 1987 showed four factors in the following order of prominence: 1) Group Analyst; 2) Group Facilitator; 3) Collaborative-Participant; 4) Protective Manager. The rank order to these factors differed in 1991 (factors 1 and 2 changed places) and the 1991 staff culture was more complex (the Varimax solution revealed two additional factors). The change in the order of the factors in terms of how much of the variance each accounted for is consistent with the kind of learning about leadership represented in members' questionnaire responses. The staff culture had shifted toward a more group facilitative rather than strict interpretive mode and the members' views of leadership showed a parallel appreciation for leadership attuned to the to the dynamics of people working together as a group, to teamwork.

Next, in 1995 and 1996 we embarked on several Q-studies of members' and staff
concepts of leadership -- deriving the Q-sort of 41 statements from the set of 68 statements used in the questionnaire study just cited. Both staff and members sorted these statements before and after the conferences. Q-methodology enables us to diagram the culture of the conference and describe the social matrix in terms of sub-groupings, or constituencies with different points of view. This is important because it allows us to examine changes in the composition of the social matrix in terms of shifts in view points and the kinds of differences that mark the boundaries between sub-groups of opinion. Q also makes it possible to track the direction of particular sub-groups, persons with particular roles within the culture, either by explicit structural designation (i.e., the Conference Director) or by his or her position in a covert structure (i.e., "high loader" in particular factors of particular interest, theoretically or empirically).

Let me illustrate here with some exhibits (charts #1, #2, #3, #4, # 5, #6) which show similarities and differences between conference cultures, relations of member to staff factor loadings, and changes in members and staff views before and after the conference.

In studying the 1996 conference, we also collected conventional ratings of the value of the small group experience by members and the effectiveness of the consultant by members, peers and research observers. With this additional data we sought to address the link between staff orientation, conference culture and member learning. Having established that learning takes place in these conferences (at least in terms of meaningful changes in opinion) and that staff cultures can be described objectively in terms of particular sub-groups of orientations to the consultant's role, we wanted to gain objective evidence of how consultant behavior effects member learning.

This led us to examine individual small study groups, including the views members held of the experience, of their consultants, and the kind of movement members would show in terms of their factor loadings in the matrices before and after the conference. We found very few surprises. Peers on the Small Group Consulting team rated the designated head of their team as the most effective. The research observers were less emphatic in singling out the team head in this way. Member ratings, however, of their small study groups indicated a slight preference for the work of one of the male consultants (J.B.). Supporting the significance of this trend is the fact that the members of this small study group showed greater increases in their factor loadings on the dominant factor (leadership characterized as "Group Process Facilitation") than did the members of the other four small study groups.

It is of interest that the head of the small group team (H.C.), although known by staff
but not by members, to be much more senior in the work, was not rated higher by the members, nor was her small group rated higher. Members' Q-sorts from this small group were not noticeably different from those of three other small groups. The Q-sort changes, however, for the members of the group consulted to by the less experienced J.B. were slightly greater. Gender bias may be at issue here and other data, made accessible with Q-methodology, adds to the story and makes another hypothesis more credible.

Behavioral anecdotes of group process also enrich the story and contribute to building a multi-layered hypothesis, one in which unconscious and covert gender issues may play a part. During the staff meetings, two particularly dramatic incidents stood out in the Director's recollections: 1) he felt so under-supported at one critical point that he was driven to the brink of asking consultant H.C. to resign (and did explicitly asked for a restatement of her authorization of him in the role of director!); and 2) during a staff meeting in which the staff were processing their own dynamics and feelings on the second day of the conference, the consultant J.B. was moved to express with considerable emotion, his deep feelings of affection for the director.

This identification of J.B. with the director and this tension between H.C. with the director were reflected in the Q-factorial alignments (Varimax solutions obtain with "qmethod" software). Consultant J.B., both before and after, was one of only two staff loaded on the same factor as the director, whereas, H.C.'s loadings decreased on this factor during the conference.

Considering these data from different sources, the hypothesis that develops is that effectiveness in job performance may depend on covert aspects of authorization to work as well as training, skill and experience. That is, a kind of value orientation alignment, or unannounced identification with the values of the designated authority figure may be a critical factor in accounting for on-the-job effectiveness at a particular time and place in an organization's life. Without Q-factor data placing J.B. and the director in a particular sub-group within the staff, we would be less likely to consider competency in role in systemic and multi-dimensional terms, and be more inclined to consider competency only in terms of individual attributes, traits, and skills. Q-methodology gives us objective and quantified data with which to build and test hypotheses about multi-leveled interactions within social matrices. Q provides another way to study gender, as well as other aspects of individuals and groups, in relation to both contractual and covert authorizations in relations to competence in role performance, as these effect outcomes in live group and organizational settings.

Styles of Learning and Group Participation

In our most recent conference, conducted in Chicago on November 7-9, 1997, we began the use of Q to explore of the process of ‘learning by experience' -- how people engage in the work of these conferences. We want to learn more about how different participants' learning styles may effect educational outcomes.

We composed a Q-sort of 34 statements selected from a "concourse" of how people see themselves as participants and learners in educational activities which are important to them. Several steps forward were accomplished in our understanding how both members and staff engage in conference work. For the first time, we were able to collect Q-data not only before and after the conference, but also during the conference itself. A half-hour research activity during which time all staff and members sorted the Q-statements, was conducted as part of the conference program, structured and scheduled as another opportunity for learning.

Also, we were able to analyze the data and report initial findings in a plenary conference review session, making data collecting and analysis part of the discovery and discourse processes integral to the social system study objectives of the conference itself. This was achieved with the excellent work of a staff research team composed of four members, each of whom brought to the team some special expertise in Q-methodology, organizational and political psychology, psychoanalysis, and/or group research methods, as well as considerable firsthand experience with conference work.

The high level of cooperation obtained for this project should also be noted. Before the conference, 38 of 43 members and all staff ranked the set of 34 statements. During the conference, all 43 members and all staff sorted the Q-cards. The research team presented an initial report to the staff within approximately five hours. These initial findings were reported to members the next morning during a plenary review session.

The research team delivered their interpretation of the factors in the form of brief monologues, giving ‘voice' to each of the factors as though each were a member participating in the open dialogue and discourse of the plenary review session. No individual's factor loadings were announced and no reference was made at this time to methodological or theoretical matters. (Prior to the conference and during the plenary opening, members had been invited to address further inquiries about the research to Dr. Bair, as Assistant Director for Research, or myself as Conference Director.)

In the data collected before the conference three factors were found, two of which were bi-polar. Two additional factors emerged during the conference itself. To give you a sense of the texture of Q research findings, these are the five ‘factor-voices' as they were presented in the conference itself:

#1 The research has revealed a generalized approach to learning plus four sub-categories. . . With very few exceptions, everyone's Q-sort characterization of ‘how I learn' bears some resemblance to this generalization . . . From the most senior staff members to the most unsophisticated first-time conference members -- all participate in this understanding about ‘how I learn.' To give you an initial inkling into this mode of thought, we would like to give it voice by articulating some of the ideas which it embraces.

I am very open to learning: ‘I like to link feelings with experiences;' ‘I have a readiness for discovery and intuition;' and no matter how tough things might become, I would never consider ‘closing the group off.'

Crucial to my learning is receptivity to others: ‘I enjoy the give-and-take;' ‘I am intrigued by how others behave;' I learn through cooperation; and it would never occur to me to 'leave the group.' For me, learning is intrinsically gratifying: it is almost always fun; I enjoy sharing new ideas; and I am almost never bothered by conference pressure or my own feelings of anger and frustration.

As can be seen, this generalized approach to learning is thoroughly positive, is without self-doubt, and is a source of immediate gratification . . . The other four voices enter into and interact with this more general approach to create more complex and nuanced learning styles that begin to differentiate individuals within the staff and within the membership.

#2+ Hello. It is a bit difficult to be open with you regarding my learning style. I am a bit counter-dependent in my strongly held belief that when it comes to learning things that matter,'I am my own best authority'; I am self-authorizing and ‘I can become stubborn and closed in a group.' Yet there is an underlying vulnerability that I cover so that the truth that ‘bad vibes in a group make me lose focus' is covered by a bit of bravado as a defense against discomfort. It is my belief that ‘abstractions are not where it's at' because of my desire to be autonomous and emphasize the part of the me that values learning from experience. However, despite my expressed belief that ‘I work best when challenged and stressed', I have some passivity and a mild dependency.

#2- In general I like situations in which I can have rich engagement with peers and ‘firm and supportive structure from leadership.' I don't much care whether my engagement takes the form of cooperating or competing, but I have a good time when I'm out there, putting my ideas forward and engaging in give and take with others, as long as there's supportive leadership. Absent that engagement with peers and support and recognition from leadership, I'm likely to have a bit of problem knowing what to make of my reactions and experiences -- I can get a bit swamped and lost, but I'll keep trying to get what I need, never closing down or withdrawing.

#3+ I am very open to learning and am receptive to others and consider learning to be intrinsically rewarding, but in addition ‘I learn best with metaphors and symbols,' ‘I learn best when others struggle with their feelings,' and learning for me involves my having to ‘manage my own feelings and strong reactions.'

#3- I need to feel secure and in control; ‘people who interfere with my listening irritate me a lot;' ‘leadership support and structure are essential to my learning;' and ‘I learn best when I have time to take notes and review.'

These are the two new factors which emerged during the conference:

#4+ When I get a new idea I share it and I lead opening new issues in a go\group. It helps me to learn through good ‘give and take' with others. ‘Learning and fun often go together;' I am comfortable with abstractions. The quality of my learning does not ‘depend on the quality of the staff consultants' and ‘I do not feel annoyed with the staff.'

#4- ‘I feel annoyed with certain staff' and feel ‘the quality of my learning depends on the staff.' I do not like abstraction nor do I share my ideas when I get them or open up new issues. At this point in my participation, I'm feeling sullen, even a bit angry.

#5+ These members are coping with the currents and pressures of the conference by a type of cerebral flight/flight. They cling to strands of autonomy by endorsing strongly the notion that ‘readiness, discovery and intuition are keys to my learning', yet keep a hand on a life-raft by indicating that ‘ their learning depends on group cohesiveness.' They tend to minimized internal struggle and attempt to jettison their frustrations by their belief that they ‘lead in opening up new issues in a group' and ‘ that competition stimulates their learning' and that ‘fun and learning go together for them;' they try to get rid of their tensions by admitting that they feel ‘annoyed with certain staff.'

#5- On what may be regarded as the other side of the "struggle factor" are individuals who seem to be bending under the pain of the times. There is a sense of internalized struggle and pained introversion as these people acknowledge that they ‘have become stubborn and closed in a group' and that their ‘learning involves managing feelings and strong reactions.' They acknowledge that their ‘anger and frustrations sabotage their learning,' but try to maintain their hopefulness and their idealizations: ‘overall, I trust the staff to be helpful' and acknowledge that the ‘quality of my learning depends on the quality of the consultants.'

What I have included above is almost verbatim how the research team reported their findings to the plenary conference review session scheduled near the end of the conference program. Although one member, as he heard these factor-voices, blurted out "that's me!" the implications of these findings were not explored. Not surprising, since this was the first time such data were included as an integral part of the learning opportunities in a group relations conference. The staff gained increased awareness of the different ways in which members were joining in the work of the conference and better appreciation of styles of resistance.

In Q-studies each thread leads to new links and associations which further both theoretical and empirical inquiry for both the researchers and the subjects. Q-studies can, I believe, be made integral to the work of exploring authority and leadership and learning by experience, the educational goals of conferences in the A.K. Rice/Tavistock tradition.

After the conference another factor solution was sought to the same correlational matrix which sheds more light and how people engage in the process of ‘learning by experience.' From this analysis, the following seven dimensions or styles of participation and learning were identified: 1) Aware that learning requires engaging with others and with inner feelings; 2) Self-reliant (perhaps ounter-dependent), self-possessed and self-authorizing; 3) Engages thoughtfully with inner feelings and others in the "here and now;" 4) Cooperative and accepting of authority and structure, harmony-seeking ; 5) Assertive, competitive with authority, willing to express dissent and conflict; 6) Works with difficult emotions (i.e., anger), but requires structure and authority in order to be interactional; 7) Responsive, dependent on and reactive to situational factors. They are listed here in order of the amount of variance each accounted for in the matrix. These factors and their relative prominence provide a quantified representation of the composition of this conference and can be used to as a basis for exploring similarities and differences in subsequent conference cultures.

During this conference, we also collected data in the form of questionnaire-ratings by the members of their valuation of different parts of the conference program and different consultants. Research observers' and consultants' rankings of members as "workers" in the small study group events were recorded. Each consultant's small study group can be characterized in terms of the factor-loadings of the members; each consultant can be characterized in terms of orientation to the consultant's role; and the movement of members in terms of factor loadings before, during and after the conference can be used to indicate the kind of learning which has occurred. Together with other ratings and anecdotal reports by staff and members, significant case studies can be constructed.

Our aims are two-fold: 1) to develop hypotheses linking the functioning of consultants as psychological work leaders (in a Bionian sense) with the views members gain of good leadership (as an indicator of educational outcome) with the styles of participation and learning which characterize conference members; 2) to obtain objective evidence to support hypotheses about what kind of consultant unctioning is most effective in promoting the most effective learning for people who may take up their ‘work' as learners in quite different ways. This program of research (Lipgar, 1992), in which Q-methodology is critical, is progressing along the lines described in this presentation.

CONCLUSION

In light of our investigations to date, it seems clear that Q-methodology enhances our ability to understand how members learn, how consultants contribute to the psychological work of the conference, and how leadership functions in both members and staff. In order to understand the interactional functioning of different systems within social matrices, quantitative as well as ‘clinical' methods are required. Knowing and learning as well as leadership, psychological and managerial, take place in social matrices and are functions of them. People take up their work as learners and exercise authority and leadership in different ways, engaging in learning by experience along several spectrums: openness vs. defensiveness in relation to the social matrix; acceptance vs. defiance of authority; responsiveness to internal vs. external stimuli and states. Both deeper and more comprehensive ‘knowledge of acquaintance' (Rice, 1965) of authority, leadership, groups and organizations understanding as well as scientific knowledge which can be shared and testable hypotheses can be achieved with the application of Q-methodology.

The three major components of a "working conference" in the A.K. Rice/Tavistock tradition are: 1) the consultant's role as a psychological leader based primarily on the work of Wilfred Bion (1961) in groups; 2) learning for leadership as the primary task; 3) learning by experience as the method (Rice, 1965). The design of the conference events, the general philosophy of management and consultation, the emphasis on task and boundary management, as well as the way anxieties integral to learning by experience are encountered and managed -- all follow from these major components. Our research efforts, therefore, have been directed to understanding more about the consultant's behavior and values, leadership behavior and functions, and the processes of participation and learning.

The challenges of such inquiries are daunting, but the prize is worth the journey. The penetrating power of Q-methodology and its adaptability make the journey possible and the passage instructive. I know of no other methodological approach which enables us to combine the riches of the domain of feelings, fantasies, and values with the discipline of quantification and statistical analysis. Stephenson has brought us to the threshold of new ways of learning, new ways of sharing our learning, and new ways of creating knowledge and community.

Both Stephenson and Bion, in the most profound sense, were intensely interested in how people learn -- how personal experiences are observed and inform our thoughts and knowledge, how experiences are shared, worked and reworked as "common sense" and sophisticated knowledge. With different but not incompatible contributions, both bring us further along the journey of exploring the unknown, symbolizing the as-yet-unspoken, and giving voice to thoughts waiting to be born -- searching Stephenson's (1978, 1986) "concourses of communicability."

References

Bion, W.R. (1961) Experiences in groups. New York: Basic Books, Inc.

________ (1962) Learning from experience. New York: Basic Books, Inc.

Bradley, D. (1987) A Q-methodology study of the conceptual framework of small study group consultants. Doctoral dissertation, The Chicago School of Professional Psychology, Chicago, IL.

Brown, S. (1980) Political subjectivity -- applications of Q-methodology to political science. New Haven: Yale University Press.

________ (1994) Group psychology and leadership, an address delivered at a panel on Psychological approaches to studying leadership, Midwest Political Science Association, Chicago, April 14-16, 1994.

Edelson, M. (1988) Psychoanalysis: a theory in crisis. Chicago: the University of Chicago Press.

_________. (1970) Sociotherapy and psychotherapy. Chicago: the University of Chicago Press.

Edelson, M. & Jones. A. (1954) Operational exploration of the conceptual self system and of the interaction between frames of reference. Genetic psychological monographs, 50, 43-139.

Lawlor, W. (1956) Subjective probability in sequential uncertainty situations. Unpublished doctoral dissertation, University of Chicago.

Lieberman, M.A. & Yalom, I.& Miles, M.A. (1973) Encounter groups: first facts, New York: Basic Books.

Lipgar, R.M. (1993) Views of the consultant's role: a Q-methodology study, in Hugg, Carson & Lipgar (Eds.), Changing group relations: the next 25 years. Jupiter, FL: A.K. Rice Institute.

________ (1992) A programme of group relations research: emphasis on inquiry and the trial of techniques, Group analysis, 25, 365-375.

________ (1965) Subjective probability notions, guessing behavior, and their personality correlates. Unpublished doctoral dissertation, University of Chicago.

Lipgar, R.M. & Struhl, S. (1993) Learning for leadership: member-learning during group relations conferences. In West, K.L. & Hayden, C. & Sharrin, R.M. (Eds), Community/Chaos: Proceedings of the Eleventh Scientific Meeting of the A.K. Rice Institute. Jupiter, FL: A.K. Rice Institute.

Rice, A.K. (1965) Learning for leadership. London: Tavistock Publications, Ltd.

Stephenson, W. (1953) The study of behavior. Chicago: University of Chicago Press.

___________ (1978) Concourse theory of communication. Communication, 3, 21-40.

___________ (1986) Protoconcursus: The concourse theory of communication. Operant Subjectivity, 9, 37-58, 73-96.