Special Issue on Group Development

Number 3 Spring 2001

Special Issue on Group Development

GROUP

FACILITATION: A RESEARCH &APPLICATIONS

JOURNAL

Editorial

A Superlative Task Sandor P. Schuman 1

Introduction to theSpecial Issue on Group Development

Sandor P. Schuman 2

Theory and Research

Group Newcomers:From Disruption to Innovation

Marie A. Cini 3

Group Development:A Review of the Literature and a

Commentary on Future Research Directions

George Smith 14

A Critical View of FacilitatingLabor-Management Collaboration

Grant T. Savage &Chadwick B. Hilton

46

Application and Practice

Facilitating Team Development:A View from the Field

John E. Jones &William L. Bearley

56

Classics for Group Facilitators

Developmental Sequence in Small Groups Bruce W. Tuckman 66

Book Reviews

The Dance of Change:The Challenges to Sustaining Momentum

in Learning Organizationsby Peter Senge, Art Kleiner, Charlotte Roberts,

Richard Ross, George Roth & Bryan Smith

Patricia R. Tuecke 82

The Logic of Failure:Why Things Go Wrong and

What We Can Do To Make Them Rightby Dietrich Dorner

Nancy S. Hewison 86

Success with Soulby Doris Pozzi and Stephen Williams

Judy Robb 89

Group FacilitationA Research & Applications

JournalNumber 3 Spring 2001 ©International Association of Facilitators ISSN 1534-5653

The mission of the International Associationof Facilitators (IAF) is to promote, support

and advance the art and practice ofprofessional group facilitation through

methods exchange, professional growth,practical research, collegial networking and

support services.

Chair:Greg Brittingham

Past Chair:Gary Austin

Chair Elect:Jo Nelson

Treasurer:Marieann Shovlin

Secretary:Jean-Anne Kirk

Vice Chair International:Gilbert Brenson-Lazan

Executive Coordinator:Robin Bailey

Association Coordinating TeamElsa Batica; Nadine Bell;

Cecil Carter; Lenny Diamond; FranciscoFernandez; Beret Griffith; Mirja Hanson; Cheryl

Kartes; Jean-Anne Kirk; Sue Laxdal; JakeMacDonald; Justine Marchant; Maria Begonia

Rodas; Edward Ruete; Sandor Schuman;Barbara Smith; Jim Troxel; Cynthia Vance

Group Facilitation: A Research &Applications Journal is published annually

by the:

InternationalAssociation of Facilitators7630 West 145th St., Suite 202

St. Paul, MN 55124Tel: 612-891-3541Fax: 612-891-1800

[email protected]

Editor: Sandor Schuman, University [email protected]

Associate Editors: Lynda Lieberman Baker,MeetingSolutions Inc.Daniel Mittleman, Depaul UniversityEdward (Ned) Ruete, ComputerSciences Corp.Michael Sabiers, SocioTechSolutions Inc.James Spee, University of RedlandsJean Watts, ICA New Orleans

Book Review Editor: Beret GriffithCopy Editor: Linda (Sunny) Walker, SunWalker

EnterprisesPast Editor: Mark Fuller (1997-2000)Reviewers: Mark Adkins; Michael Ayers;

Pierre Balthazard; Lew Barsky;Stephen Bather; Bob Bostrom;Robert Briggs; Aileen Buslig;Sharon Cannon; Peg Carlson;Erran Carmel; Dutch Driver;Sari Fried; William Gardner;Joey George; Bill Harris;Dale Hunter; Michelle Jackson;Jon Jenkins; John E. Jones;Gigi Kelly; John Keltner;Stephen King; Sue Laxdal;Don McCormick; Richard Milter;Deborah Pulak; Patricia Reagan-Cirincione; Bruce Reinig;Susan Rimkus; Judith Robb;Renee Rogers; John Rohrbaugh;Nicholas Romano; Edward Ruete;Michael Sabiers; David Stein;Robert Tobias; Teri Tompkins;Raymond Vles; John Walker;Ken Walsh; Tom Webler;Lonnie Weiss; Julia Young

Copyright 2001Printed in USA

No copying or excerpting of Group Facilitation:A Research & Applications Journal or any of itscontents is permitted, including copying for coursereadings, without the prior consent of the publisher.

Reprints Individual articles are available at $5.00 per article fromthe IAF Office. Quantity orders discounted.

Subscriptions& Back Issues

Current and back issues may be purchased from theIAF Office. Individuals: $25; Institutions: $40

Postmaster- Sendaddress changes

to:

Group Facilitation: A Research & Applications Journal,International Association of Facilitators, 7630 West145th St., Suite 202, St. Paul, MN 55124

A Superlative Task Editorial

Group Facilitation: A Research and Applications Journal - Number 3, Spring 2001 1

Sandor P. Schuman

One can hardly contemplate the passing scene of civilized society without a sense that the need ofbalanced minds is real and that a superlative task is how socially to make mind more effective.

- Chester Barnard, The Functions of the Executive, 1938, p. 322.

While some might say that group facilitation is just an ordinary task, I believe that group facilitatorstend to think of it as an important task, or even an extraordinary task. But who among us has thechutzpah — the self-righteousness — to assert that group facilitation is a superlative task? Better to turnto a venerated and impartial authority who can issue this bold proclamation!

Chester Barnard is such a person, a preeminent mid-twentieth-century corporate executive oftencalled the "father of organization theory." His classic The Functions of the Executive was required man-agement school reading for many decades following its 1938 publication. Though still in print, Barnard'soccasionally impenetrable prose has limited the use of his book to only the more rigorous graduate pro-grams, replaced elsewhere by more recent and easily-read authors. Nonetheless, Barnard still challengesus with pertinent ideas that have retained, if not increased, their relevance. In the concluding paragraphof this renowned book, Barnard highlights four very salient points.

Society is increasingly complex and organizations are more elaborate.Even more true than in 1938, the idea that society is increasingly complex now is accepted axiomati-

cally. Organizations are greater in number, size and geographical scope. We are more dependent thanever before on elaborate technologies and the equally elaborate organizations that create and rely on them.We are interconnected and interdependent; yet distinct and diverse.

The increasing specialization necessitated by such a society brings with it a diversity of methods and purposes thatmay be inconsistent and foster misunderstandings.

To manage our complex, technological world people must be specialized — in roles, expertise andskills. This makes effective communication, sharing of knowledge, and interpersonal understanding moredifficult. This difficulty occurs not only at the level of substantive issues but also at the underlying levelsof method (how people think about issues) and purpose (why they think about them). Misunderstandingsoccur between individuals, of course, and even more crucially between large groups of people.

What is needed are balanced minds that integrate feeling with reasoning, sense the net balance, and perceive theparts as well as the whole.

The difficulties brought on by the effects of complexity and specialization can be addressed. How? Byincorporating the views of multiple stakeholders with diverse interests and perspectives; perceiving thespecific parts of the system, as well as the system as a whole; and clarifying the expected results anddesired ends. We need to integrate analysis and intuition, facts and values, objective and subjective,thinking and feeling.

Meeting these challenges—which will help groups to be more effective cognitively and socially—is a superlative task.To meet these challenges we must be address the intellectual, analytical and cognitive demands of the

situation. This is necessary but not sufficient. At the same time, we must help groups engage interper-sonally, politically, emotionally and spiritually. As group facilitators we must, in Barnard's words, strive"socially to make mind more effective." Toward this accomplishment we devote ourselves as groupfacilitators and dedicate Group Facilitation: A Research & Applications Journal . Working together, we aimto strengthen our understanding — in organizations, communities and societies — of group facilitation, asuperlative task.--------------Barnard, Chester (1938). The Functions of the Executive Thirtieth Anniversary Edition (1968).

Cambridge, MA: Harvard University Press.

Here is the full quote: One can hardly contemplate the passing scene of civilized society without a sense that the need of balanced mindsis real and that a superlative task is how socially to make mind more effective. That the increasing complexity of society and theelaboration of technique and organization now necessary will more and more require capacity for rigorous reasoning seems evident; butit is a super-structure necessitating a better use of the non-logical mind to support it. “Brains” without “minds” seem a futile unbalance.The inconsistencies of method and purpose and the misunderstandings between large groups which increasing specializationengenders need the corrective of the feeling mind that senses the end result, the net balance, the interest of all, and of the spirit thatperceiving the concrete parts encompasses also the intangibles of the whole.

2 Special Issue on Group Development

Introduction to theSpecial Issue on Group Development

Group development is a recurring topic of discussion among group facilitators. It has been the subjectof research since the 1940s and continues to be an active area of inquiry. What have we learned aboutgroup development in the past 60 years? How can we apply that knowledge in our work as group facili-tators? What questions remain unanswered? The articles in this Special Issue on Group Development helpto answer these questions.

Were we to conduct a survey to assess the present state of knowledge regarding group development Isuspect that the response we would receive most often would include something about forming, storming,norming, performing and adjourning. We owe this memorable characterization of stages of group develop-ment to Bruce Tuckman who introduced this oft-cited naming scheme in 1965. With his permission, andthat of the American Psychological Association, we are pleased to reprint his hallmark article, Developmen-tal Sequence in Small Groups in our Classics for Group Facilitators section.

The development of a group is often viewed as occurring in a step-by-step progression that takes agroup from one stage to the next. Alternatively, group development can be viewed as shifting back-and-forth from one phase to another and back again, or in a repeating cycle of development and redevelop-ment. Yet another view is that a group's development does not occur in any particular pattern, but iscontingent at any point in time on contextual factors. In Group Development: A Review of the Literatureand a Commentary on Future Research Directions George Smith reviews the literature on group develop-ment, highlighting the similarities and differences between various perspectives, summarizing the currentstatus of thinking, and pointing to needs for future research.

Complicating any view of group development is that group membership can change. How do changesin membership affect group development? Focusing specifically on new members, Marie Cini applies theaccumulated research and provides interesting insights and specific suggestions for group facilitators inGroup Newcomers: From Disruption to Innovation.

In Facilitating Team Development: A View from the Field John E. Jones and William L. Bearley showhow they have come to terms with various views of group development and present their own framework.They provide numerous examples to illustrate how they facilitate team development.

A Critical View of Facilitating Labor-Management Collaboration, by Grant T. Savage and Chadwick B.Hilton, calls our attention to external processes -- what happens between meetings and relations withparties outside the group -- as well as internal group processes. The authors' place the role of the facili-tator in the context of Habermas' theory of communicative action and provide examples of facilitator inter-ventions in labor-management collaboration.

We hope you find these articles informative and helpful in your own practice, teaching, and research.

The Editorial Board and staff are themselves experiencing some "group development." On their behalfI extend appreciation and thanks to five individuals who were instrumental in founding the journal andgave years of dedicated service during its long gestation period and early years of publication:Mark Fuller, Editor-in-Chief; Peggy Runchey, Managing Editor; Beret Griffith, Book Review Editor;Vicki Wharton, Design Editor; and Jean Watts, Associate Editor. As they move on to other priorities weacknowledge that our current success is a tribute to their early and prolonged efforts. We thankEileen Ruete, Copy Editor, and wish her a complete and lasting recovery. Lynda Lieberman Baker, whoseservice as Associate Editor is much appreciated, will assume the post of Book Review Editor. To MichaelSabiers, who recently joined us as an Associate Editor, we extend our welcome and look forward toworking with you.

- Sandor Schuman, Editor


Group Newcomers: From Disruption to Innovation

Marie A. Cini

One of the ways that groups change over time is through the introduction of newcomers. Untilrecently, group research has primarily focused on the attempts of the existing group to socializethe newcomer, whereas the effect of the newcomer on the group has been relatively less explored.However, research on newcomer influence suggests that newcomers can influence the group undercertain conditions. Research on the power of the (numerical) minority in groups has also uncov-ered some intriguing findings regarding the positive effects a newcomer can have on a group. Forgroups seeking to be more innovative and effective, newcomers may be an overlooked source ofinnovation. Group facilitators can prepare the group and the newcomer to maximize the potentialof newcomer contributions.

Keywords

Innovation; Majority Influence; Minority Influence; Newcomer Influence

INTRODUCTION

Groups rarely remain as static entities. Theymay change over time in structure, activities, andmembership in order to adjust to internal changesor external demands (Worchel, Coutant-Sassic, &Grossman, 1992). For organizational groups inparticular, changes in membership are frequent ascurrent group members move to other organiza-tional units or leave for other jobs. New membersare added to work groups as new employees arehired, new skills become necessary to the group’sfunctioning, or an employee is motivated to gainexperience in a particular work group.

The addition of a new group member is a com-mon and often stressful occurrence because of theways the group may be affected. Descriptive evi-dence of the changes that newcomers cause ingroups is abundant. For example, a number ofobservers have commented on the regressiveeffects of newcomers to therapy groups. Goodman(1981) argued that the introduction of a newcomerinto a therapy group is threatening to old-timers.Kaplan and Roman (1961) and Leopold (1961)found that newcomers to therapy groups elicit avariety of reactions from old-timers (e.g., restless-ness, hostility, and resistance) until the groupaccepts the newcomer and rebalances itself. Yalom(1970) observed that newcomer socialization intherapy groups diverts members’ energy andinhibits therapeutic progress. Finally, Saravay(1978) argued that, because therapy groups havealready developed a complex structure prior to thenewcomer’s arrival, the group copes with the dis-ruption by either rejecting the newcomer orregressing to an earlier stage.

Observers of organizational life note that new-comers impact the organization and its employees

in a variety of ways. Nicholson (1984) and Schein(1971) argued that newcomers to an organizationoften attempt to change work demands and goals,rather than adapt to the status quo. Sutton andLouis (1987) expanded this line of reasoning andargued that every recruitment or socializationactivity (e.g., reading prospective employees’resumes, training new employees) influences thethoughts, feelings, and behaviors of currentemployees. Gadon (1988) argued that newcomerscause the group to recycle through the stages ofgroup development (Tuckman & Jensen, 1977) asthe group adapts to the newcomer. Finally, Levineand Russo (1985) argued that merely having new-comers present alters the composition of thegroup. For example, newcomers may divert some ofthe energy old-timers in the group devote to groupactivities. And, if newcomers are sufficiently moti-vated to change the group (perhaps because theyfeel deceived about what group membershipmeant), they can disrupt the group’s activities.

As a whole, this line of qualitative work sug-gests that newcomers can impact groups in anumber of ways. They may elicit reactions fromindividual group members, disturb relationshipsbetween members, alter others’ expectations ofthem, or even cause the group to return to an ear-lier stage of group development (Wanous, Reichers,& Malik, 1984). However, these observationalaccounts tend to focus on the disruptive effects ofnewcomers on groups, perhaps because of the per-ceived negative consequences that disruption cancause (e.g., lowered productivity, interpersonalconflict). On the other hand, the potentially posi-tive effects that newcomers can have on groupshave largely been ignored in empirical research.This is particularly interesting in light of the factthat most groups publicly state that they welcome



newcomers because they are looking for “newblood” and “fresh ideas.” Perhaps because of thisambivalence that groups feel aboutnewcomers—needing new members but fearing thedisruption they may bring—groups tend to defaultto exerting conformity pressures on the newcomer.Reflecting this reality, social psychologists havepredominantly focused their research efforts ongroup attempts to change a newcomer’s thoughts,feelings, and behaviors to make the individualmore similar to current members (i.e.,assimilation). They have tended to ignore thechanges that the individual may produce in thegroup’s structure, dynamics, or performance (i.e.,accommodation) (Moreland, 1985).

As just one example of this line of research thatignores newcomer influence, Lofland and Lejeune(1960) conducted a study that examined howexisting members of Alcoholics Anonymous (AA)socialized newcomers at meetings. Because of theresearchers’ focus on the group’s influence onnewcomers, they did not examine how newcomersmight affect old-timers in AA, although this wouldseem to be an important question for researchersand practitioners alike.

THE HISTORICAL FOCUS ON THE POWER OF THEMAJORITY

The preoccupation with assimilation ratherthan accommodation processes grew out of socialpsychology’s historical focus on the power of thegroup to sway the individual, even when themajority is clearly incorrect (Asch, 1956). This bodyof research is referred to as “majority influence,”and researchers in this tradition have explored themajor situational and individual variables thatfoster it. For example, classic work by Festinger(1950) uncovered strong pressures on group mem-bers to act in uniform ways. Festinger explainedthat group members (i.e., the majority) place pres-sure on other members to conform to group goalsfor two reasons: social reality and/or group loco-motion. Social reality refers to the need of groupmembers to validate their own opinions by havingothers agree with them. Group locomotion refers tothe group’s desire to move forward toward thecompletion of a goal. Thus, members of groupsoften experience strong pressure to agree with therest of the group and to act in ways that willensure movement toward the group’s goal.

Group members themselves respond to pres-sures from the majority with their own desires tofit in. Deutsch and Gerard (1955) posited that indi-viduals adopt the views of majority group membersfor two reasons: to gain information about reality(informational influence) and/or to gain approval

(normative influence). Newcomers, who are strug-gling to make sense of their new environment(Louis, 1980), may be more susceptible to majorityinfluence, especially if they are highly motivated toremain in the group. For example, Ziller andBehringer (1960) found that newcomers’ popularityratings declined markedly shortly after the “new-ness” wore off. The newcomer’s popularity roseonly after accepting the norms of the group andbehaving more like the majority. Thus, newcomerswho hope to gain the acceptance of the group arelikely to behave in ways that conform to groupnorms in order to be viewed as potentially “good”group members.

… members of groups often experiencestrong pressure to agree with the rest ofthe group and to act in ways that willensure movement toward the group’sgoal.

However, majority influence on the newcomeris only a part of the reality of group life. Facilitatorsof groups whose members collaborate on projectsor undergo significant experiences together knowthat newcomers are sometimes welcomed as indi-viduals who can contribute to the success of thegroup. For example, if a group is failing at a task,newcomers may arrive expecting to offer inputbased on their prior experiences and expertise, andold-timers may welcome the newcomer’s input toimprove the group’s viability. In other cases, thegroup and the newcomer exert mutual influenceover one another. For example, a newcomer to anongoing work team may adopt some of the normsof the group while also contributing new ideas.Thus, newcomers may be a rich source of newideas for a group seeking to become more creativeor productive.

THE VALUE OF NEWCOMER INNOVATION

In today’s rapidly changing economy, groupsand the organizations in which they function areincreasingly interested in becoming more innova-tive in order to stay competitive. West (1990)defined innovation as “the introduction and appli-cation, within a group, organization or wider soci-ety, of processes, products, or procedures new tothe relevant unit of adoption and intended tobenefit the group, individual, or wider society” (p.9). This definition casts innovation as an inten-tional behavior intended to make a change for thebetter. However, some innovations also arise fromunintentional behaviors. For example, a teamleader may be absent on a given day, which pro-

Cini


vides an unexpected opportunity for the team topractice shared leadership.

In addition to noting that important innova-tions may be produced unintentionally, it is alsoimportant to understand how innovation is fos-tered. For example, an individual’s level of creativ-ity is a strong predictor of how innovative he or shewill be on the job (Bunce & West, 1995). Oneimplication of this finding is that the quality ofinnovative ideas in a team will be determined by itsproportion of innovative members, and researchsupports this proposition (West & Anderson, 1996).

… group members are usually fearfulthat the new member will disrupt thegroup’s identity, interpersonal relation-ships, or ways of accomplishing tasks.

But creative individuals do not, by themselves,account for all of the innovation in a team ororganization. In fact, there is a social dimension tocreativity that is often overlooked. For example,Montouri and Purser (1995), in a study of researchand development organizations, found that themost creative researchers viewed themselves ascollaborative team players. Those researchers whoviewed themselves as loners were actually lesscreative. Interacting with others may provide newideas that a lone researcher might not consideralone (Brown, Tumeo, Larey, & Paulus, 1998).Indeed, West and Anderson (1996) concluded thatcreative ideas may come from individuals, but it isthe group climate that determines if the innovationis encouraged or derogated. Innovation increaseswhen it is rewarded (Amabile, 1983; Hackman,1992) and when it is fostered by group and organ-izational leaders (Burpitt & Bigoness, 1997).

However, given the forces toward conformityfound in most groups, innovation often lags as themembers become complacent and self-satisfied(Gadon, 1988). This complacency can lead todecreased performance and productivity becausethe group is focused more on maintaining relation-ships than on accomplishing the task (King &Anderson, 1990). Consequently, the group may failto reexamine and improve how it conducts its work(Gadon, 1988). Katz (1982) also found the ten-dency toward complacency and isolation in hisstudy of long-term work groups. He found that thelonger groups work together, the less they commu-nicate with key outsiders or scan their environ-ment for changing conditions, resulting indecreased performance.

One way to avoid the tendency toward compla-cency and isolation in groups is to seek input fromnewcomers. Newcomers are a particularly goodsource of innovation—both intentional and unin-tentional—because they can bring new perspec-tives from former experiences and may be knowl-edgeable of changes in the environment. For exam-ple, Ettlie (1980) found that major innovations aremore likely to occur when individuals move to newgroups and/or organizations. Without changes inmembership work groups run the risk of becomingless innovative over time (Bantel & Jackson, 1989;Jackson, 1996).

NEWCOMERS: DISRUPTION OR INNOVATION?

Group facilitators have a difficult role whennew members arrive. They must simultaneouslyassist the group in maintaining equilibrium, whilealso helping to integrate the newcomer in waysthat benefit both the group and the newcomer.Whether publicly acknowledged or privately felt,group members are usually fearful that the newmember will disrupt the group’s identity, interper-sonal relationships, or ways of accomplishingtasks. However, strong conformity norms imposedby the majority actually militate against the new-comer acting in a disruptive manner. And, at thesame time that “old-timers” in the group areattempting to “rein” in the newcomer, newcomersthemselves are concerned about being accepted bythe group (Heiss & Nash 1967; Nash & Wolfe,1957) and are aware of the pressure to conform togroup norms (Nemeth & Staw, 1989). For example,Feldbaum, Christenson, and O’Neal (1980) foundthat newcomers to groups were hesitant and non-assertive in their initial interactions. A newcomerto a group does not know which behaviors areacceptable and will therefore tend to act tentativelyuntil acceptable behaviors are learned eitherthrough observation or by trial and error.

The evidence suggests that, for the most part,newcomers will behave in ways that are largelynon-disruptive to the group. A skilled facilitatorshould be careful not to equate the naturalchanges that occur as a group and newcomer meldwith disruption that will cause the group to falter.Although some newcomers in some groups may bea disruptive influence, a facilitator should becareful not to become overly concerned with pro-tecting the group’s interests at the expense of thepotential contributions of the newcomer. In fact,given the need for innovation and creativity in mostgroups, the facilitator should try to help the new-comer contribute to the group in positive ways. Inorder to do that, the facilitator should be aware of



when groups are most likely to allow a newcomerto make changes.

WHEN NEWCOMERS CAN INNOVATE: THEEXPERIMENTAL EVIDENCE

Only a small number of empirical studies haveinvestigated newcomer innovation. These studiessuggest that characteristics of the group and char-acteristics of the newcomer can each affect howmuch change the group will “allow” from a new-comer.

Although some of the studies reviewed in thissection were conducted recently, many of thestudies were conducted in the 1940s, 1950s, and1960s. Though somewhat dated, these early stud-ies add to our knowledge base regarding whennewcomers to groups can be influential. Given thesparse number of topics on newcomer influence,even these early studies will be included to provideas inclusive a review as possible.

Group characteristics

Several studies suggest that certain character-istics of the group can affect how much influence anewcomer can exert. These characteristics includethe group’s level of (a) openness, (b) staffing, (c) co-hesion, (d) performance, and (e) development.

Group openness. Ziller (1962) defined opengroups as those that are fluid in their membership:Old-timers leave and newcomers arrive. In opengroups, membership is in flux. Closed groups havea static membership: Old-timers stay and newcom-ers are rare. In closed groups, membership doesnot change. Ziller, Behringer, and Goodchilds(1960) and Ziller, Behringer, and Jansen (1961)found that open groups are more likely to allownewcomers to influence the group than are closedgroups. Ziller explained that, in open groups, rela-tionships among members are temporary, andgroup structure is kept simple in anticipation ofpersonnel changes. Because they are prepared forchanges in membership, open groups allow new-comers to innovate more readily than do closedgroups.

Group staffing level. Barker (1968) argued forthe importance of studying staffing level in groups.Understaffed groups have too few members relativeto task requirements, overstaffed groups have toomany members relative to task requirements, andadequately staffed groups have a sufficient numberof members relative to task requirements. Under-staffed groups are presumably eager to recruit newmembers in order to ease the workload on existingmembers. In that case, understaffed groups maybe more willing than overstaffed or adequately

staffed groups to allow newcomers to influence thegroup. As predicted, Petty and Wicker (1974) andCini, Moreland, and Levine (1993) found thatunderstaffed groups were more likely than ade-quately and overstaffed groups to allow newcomersto make changes in the group.

… the facilitator should be aware ofwhen groups are most likely to allow anewcomer to make changes.

Group cohesion. Cohesion is a group’s senseof solidarity. Cohesion is high when members likeone another, enjoy the group’s activities, embracethe group’s values, and believe that membership isimportant to achieve personal goals (Cartwright,1968). Highly cohesive groups exhibit a greatertendency to reject members who deviate fromgroup norms than do less cohesive groups(Festinger, 1950).

The level of cohesion also affects how the grouptreats newcomers. Because old-timers in cohesivegroups tend to find satisfaction in their member-ship and tend to view newcomers as threatening(Brawley, Carron, & Widmeyer, 1988), they arereluctant to allow newcomers to make changes(Merei, 1949; Moreland & Levine, 1989). In con-trast, lower levels of cohesion result in decreasedconformity to group norms (Festinger, Schachter,& Back, 1950), suggesting that newcomers may beallowed to be more influential in less cohesivegroups. Indeed, research by Mills (1957) andSnyder (1958) found that newcomers in less cohe-sive groups were more influential than newcomersin more cohesive groups.

Group performance. All groups presumablywork toward achieving some goal. The process andoutcome of members’ efforts to achieve that goalconstitute group performance (Levine & Moreland,1990). Whether a group has succeeded or failed atan important task may have a strong impact onhow newcomers are subsequently treated. Mem-bers of successful groups may be reluctant to allowany changes because they fear that changes mayharm their success. Members of failing groups maybe eager to make changes in the hopes of becomingmore successful. Therefore, failing groups may bemore open to newcomer influence than are suc-ceeding groups. Indeed, Cini (1994), Fromkin,Klimoski, and Flanagan (1972), and Ziller andBehringer (1960) found that groups that were fail-ing on a task were more likely to allow newcomerinfluence than were successful groups.

Cini


Newcomers who imitated the behaviorsof the group leaders were more able toinitiate and direct activities at latermeetings than those newcomers whodid not.

Group development. The goal of group devel-opment research is to discover how and whygroups change over time (Levine & Moreland,1990). Tuckman and Jensen’s (1977) five-stagemodel of forming, storming, norming, performing,and adjourning has received wide acceptance(Moreland & Levine, 1988), possibly because it isbroad enough to encompass the many other stagemodels that have appeared over the years(Kormanski, 1988).

Several authors have speculated about when,in a group’s development, newcomer influence ismore or less possible. Moreland and Levine (1988)argued that newcomer influence can more likelyoccur during early stages of group developmentbecause the group is not structured enough to de-mand conformity from newcomers. And, in laterstages of development, the group possesses a clearset of norms and a desire to enforce these norms.Therefore, Moreland and Levine suggest that new-comer influence will be resisted during later stagesbecause of the potential disruption it represents.Cini (1994) explored the impact of group develop-ment on a newcomer’s ability to influence a group.In line with Moreland and Levine’s (1988) reason-ing, the results indicated that groups in earlystages of development asked for more input fromthe newcomer and held more positive attitudestoward a potential newcomer than did groups inlater stages of development.

Newcomer Characteristics

Several studies suggest that newcomer char-acteristics may also affect how much a newcomercan influence the group. These characteristicsinclude newcomer (a) dominance behaviors and (b)knowledge.

Newcomer dominance behaviors. Particularbehaviors of the newcomer may produce more orless influence in a group. In an early field study ofthe effect of dominant newcomers to playgroups,Merei (1949) observed groups of children, homoge-neous in sex and age, who met daily for two weeks.A dominant child was introduced after three to sixmeetings, when the groups had established strongnorms and interpersonal relationships. This newchild was usually ineffective in changing the

group’s activities. In some cases, however, thenewcomer was able to make minor changes in thegroup by initially imitating the old-timers and laterintroducing a minor variation such as increasingthe pace of the activity. This suggests that new-comers were able to produce some level of changeonly by first ingratiating themselves with thegroup. Similar results were also obtained byPhillips, Shenker, and Revitz (1951) who intro-duced nondominant newcomers into groups ofchildren. Newcomers who imitated the behaviors ofthe group leaders were more able to initiate anddirect activities at later meetings than those new-comers who did not.

Newcomer knowledge. Newcomers presuma-bly possess varying levels of knowledge pertainingto group activities. It is likely that groups valuenewcomers who possess higher levels of relevantknowledge. One study examined the effect of new-comer knowledge on ability to influence a group.Fromkin et al. (1972) found that in successful (all-white) groups, competent newcomers, regardless ofrace, were preferred to incompetent newcomers.However, in failing (all-white) groups, incompetentAfrican-Americans, competent African-Americans,and competent Caucasians were all preferred toincompetent Caucasians. It may be that in failinggroups, any indication of being different from theexisting group (e.g., greater competence, a differentrace) may be viewed as a cue that the newcomercan contribute to the future success of the group.Supporting this hypothesis that newcomers of adifferent race are perceived as potentially moreknowledgeable or innovative, Craig (1996) foundthat African-American newcomers to predomi-nantly Caucasian groups were judged as havingmore useful ideas than were Caucasian newcom-ers, Caucasian old-timers, and African-Americanold-timers.

RESEARCH CONCLUSIONS

Only a few studies have investigated the condi-tions under which newcomers are able to innovatewithin a group. Regarding group characteristics,the studies suggest that groups that are open (ver-sus closed), understaffed (versus adequately oroverstaffed), not cohesive (versus cohesive), failing(versus succeeding), and at an early (versus later)stage of development are more likely to allow new-comers to innovate. Regarding newcomer charac-teristics, newcomers who first ingratiate them-selves with the group members (versus behavingdominantly immediately) and who appear compe-tent or different (versus incompetent or similar tothe group) are more likely to be able to innovatewithin a group.



Clearly, in these studies, certain conditionswere created or uncovered in which groups wereless likely to consider newcomer innovation to bedisruptive. However, this line of research on new-comer innovation ended without culminating in acomprehensive understanding of when and hownewcomers can innovate. This was primarilybecause the studies were designed and conductedatheoretically. That is, the researchers exploredvariables of interest rather than designing researchthat followed from a theoretical framework. Fur-thermore, they neglected to study the possible con-sequences that newcomer innovation can have onthe group’s performance. However, in the 1970s, arelated research topic emerged that shed new lighton the study of newcomer innovation and itspotential effects on group members.

THE POWER OF THE FEW: MINORITY INFLUENCE

Minority influence research explores the waysin which an individual can influence and change agroup (e.g., Moscovici 1976; 1985). Perhaps mostimportantly, this research had as its foundation atheory of minority influence which led to a moresystematic program of study. Although thisresearch is not focused specifically on newcomers,it does subsume the study of newcomer influenceand provides a theoretical foundation for furtherresearch.

An early and influential researcher in this area,Serge Moscovici, argued that influence, whethermajority or minority, is directly related to the pro-duction and resolution of conflict. He argued thatdisagreement between numerical majorities andminorities produces both interpersonal andintrapersonal conflict. The way in which each sideresolves these conflicts is the key to understandingthe influence process of each.

Moscovici views majority and minority influ-ence as qualitatively different processes (1985). Heargued that in the face of majority pressure,minorities focus on the social consequences of thedisagreement. Because the minority views itself asthe weaker faction and wants to be accepted by themajority, the minority exhibits public changetoward the majority’s position. However, becausethe minority is anxious about being accepted,minority members do not turn their attention tothe actual issue and fail to engage in active infor-mation processing. Therefore, the minority isunlikely to experience true internalized influence.

It may be that in failing groups, anyindication of being different from theexisting group (e.g., greater competence,

a different race) may be viewed as acue that the newcomer can contribute tothe future success of the group.

In contrast, Moscovici argued that in the face ofminority pressure, majorities feel strong and arenot anxious about being accepted. This lack ofanxiety allows the majority to focus their attentionon the issue and engage in active informationprocessing, which, in turn, often produces privatechange toward the minority’s position. However,because the majority does not want to be viewed asweak or deviant, they are unlikely to admit theirchange of opinion publicly. In a meta-analysis ofthe minority influence literature, Wood, Lundgren,Ouellette, Busceme, and Blackstone (1994) foundevidence to support this claim. Majority groupmembers tend to avoid becoming aligned with adeviant minority and so are more willing to admithaving been influenced by the minority privatelythan they are publicly.

Moscovici, Lage, and Naffrechoux (1969) foundthat behavioral style is particularly important to aminority’s ability to influence a majority. Minoritiesthat exhibit consistency and commitment are mostlikely to be successful in exerting influence on amajority. Maass and Clark (1984), in a review ofthe minority influence literature, concluded that aconsistent behavioral style causes the majority toattribute certainty and confidence to the minority,especially when subjected to a great deal of socialpressure from the majority. According to Maass,West, and Cialdini (1987), the minority’s apparentcourage and commitment in the face of majoritypressure make them more persuasive than themajority. However, that persuasiveness functionsat a private, rather than a public, level. Indeed,Wood et al. (1994), in their meta-analysis, foundthat minorities that were perceived as consistentadvocates for their views were particularly influen-tial.

Nemeth has extended the line of research onthe differential effects of majority and minorityinfluence in some intriguing ways (1986). Nemethand her colleagues have found that majorities andminorities produce different effects on attention,thought, and problem solving. According toNemeth (1986), these differences in majority andminority influence lead to different consequencesfor thinking and problem solving. A numericalminority tends to produce divergent thinking,which in turn leads to more creative solutions toproblems. In contrast, a majority produces conver-gent thinking, leading to noncreative solutions toproblems. Interestingly, Nemeth (1992) found that

Cini


exposure to minority influence leads to more diver-gent thinking even when the minority position isincorrect. Thus, it appears that the presence of adissenting minority, whether correct or incorrect,can lead to more creative problem solving amongall the members of a group.

Furthermore, minorities who are members ofthe group tend to be treated more leniently thanare minorities who are from outside groups (Alvaro& Crano, 1996, 1997). That is, majority membersin a group resist real change when confronted byany minority position. However, if the minority is amember of the group, the majority is more likely tothink about the minority’s message and is lesslikely to derogate the minority than if the minorityis from another group. Crano and Alvaro (1998)and Wood et al. (1994) also found that exposure tominority influence may also lead to changes inmajority beliefs even indirectly associated with theactual topic under consideration.

The facilitator should realize that thegroup will naturally move towardassimilating the newcomer ratherquickly unless the facilitator assists thegroup in behaving differently.

As a whole, this research on minority influencesuggests that if newcomers are provided with anentrée into the group that encourages them tovoice their ideas in a constructive manner, groupscan benefit in some important ways. For example,group members exposed to a consistent and com-mitted newcomer may be encouraged to think morecreatively and this could potentially improve thegroup’s performance. Furthermore, a consistentand committed newcomer may induce changes ingroup members’ thinking about topics even indi-rectly associated with the topic under discussion.For example, if a majority of group members is infavor of purchasing a new computer system, andthen they listen to a persuasive newcomer’s alter-nate plan to upgrade the present system, they arelikely to a) privately shift their opinion towardupgrading the present system and b) also shifttheir thinking on associated topics such as whoshould be provided with laptop computers. How-ever, to maximize the potential of a newcomer’scontributions to a group, facilitators must prepareboth the newcomer and the group so that fears ofdisruption are dampened and openness to innova-tion is maximized.

Facilitating Newcomer Innovation

The earlier discussion on innovation in workgroups concluded that strong conformity pressuresfrom the majority can lead to decreased groupperformance and decision making. One of the rea-sons for this may be that in groups with a strongconformity norm, initiative from newcomers is keptat low levels (Nemeth & Staw, 1989). This isunfortunate, because when a newcomer joins agroup, it represents a rare opportunity for thegroup members to view themselves from a differentperspective, to gain new ideas, and to think inmore creative ways. This opportunity is lost oncethe newcomer is assimilated into the group. If thenewcomer is quickly assimilated into the groupbefore he or she is allowed to offer input, the groupis deprived of an opportunity to gain importantinsights (Gadon, 1988).

However, experienced group facilitators willrecognize the stress on the group that a newcomerrepresents. Groups, having once established anequilibrium, do not change willingly in most cases.Therefore, a newcomer, even one who is experi-enced in the group’s task and who behavesaccording to the norms of a “good” newcomer, willelicit ambivalent feelings from the group. Thegroup facilitator’s role is to help each party—theextant group members and the newcomer—to meldin ways that are positive and beneficial to both.

When a newcomer joins a group, a facilitatorwould do well to observe two processes: the level ofconformity pressures and the potential for new-comer innovation. First, the facilitator needs tounderstand and be aware of the strong pressurestoward conformity at work in a group when a new-comer enters. The facilitator should realize that thegroup will naturally move toward assimilating thenewcomer rather quickly unless the facilitatorassists the group in behaving differently. Thefacilitator must also understand the potential valueof the newcomer (e.g., increased divergent think-ing, enhanced creativity) and should seek out waysthat the group can capitalize on the newcomer’scontributions.

Research on group characteristics reviewedearlier in this paper suggests that when groups areopen, understaffed, less cohesive, failing, or earlyin their development, they are more likely to allownewcomer innovation. But “allowing” innovationdoes not mean that innovation will automaticallyoccur in the group. To ease this process, the groupfacilitator can assess the group’s level of theseconditions. A group that is understaffed, for exam-ple, may be less resistant to accepting a new-comer’s innovation than is a group with too many



members. Likewise, if a group is failing at a task,the members may be more open to a newcomer’sideas to become more successful.

However, this also means that groups that areclosed, adequately or overstaffed, more cohesive,successful, or later in their development are lesslikely to allow newcomer innovation because of thefear of disruption. However, these groups may verywell benefit from innovative ideas and increasedcreative thinking, nonetheless. Therefore, a facili-tator in groups with these conditions may have towork harder to help the group view the newcomeras a source of innovation.

Jones and Crandall (1985) suggested that agroup leader introduce the idea of the newcomer tothe group before the newcomer arrives, so that thegroup can deal with any negative reactions theymay have and adopt an accepting attitude. Bach(1954) suggested that facilitators should preparethe group by holding a discussion for the currentmembers to discuss how they felt when they werenewcomers in the group. Such a discussion canhelp to remind the group of the anxieties inherentin the newcomer experience. This discussion canalso serve as a time to educate the group about theimportance of the newcomer to group performance.

The group facilitator would also do well to cre-ate overall support for innovation in the group.Recall that the group climate for innovation was akey variable in predicting a group’s innovationlevel. Thus, a climate where innovation is sup-ported, welcomed, and rewarded is also most con-ducive to newcomer innovation.

Research has also demonstrated that certainnewcomer characteristics may allow the newcomerto be more influential in a group. For example,knowledgeable newcomers appear to be more influ-ential than do less knowledgeable newcomers(Fromkin et al., 1972). In addition, newcomers whoare consistent in their position also seem to pro-duce more influence than those who are less con-sistent (Wood et al., 1994). Group facilitatorsshould keep these characteristics in mind andactively search for knowledgeable and consistentnewcomers because they stand the best chance ofpositively influencing the group.

However, recall that dominant newcomers areactually less influential than those who first adoptthe norms of the group and then introduce inno-vations at a later point (Merei, 1949; Phillips et al.,1951). Therefore, a facilitator should also preparethe newcomer in terms of how and when to intro-duce new ideas. Certainly, newcomers ought to bereminded that early dominant behaviors may actu-ally work against them. However, research has

demonstrated that most newcomers tend to act innon-dominant ways, and so facilitators may actu-ally need to help newcomers learn when and howto speak up. Newcomers who are fearful of statingtheir opinion may benefit from a group facilitatorwho provides support. For example, the facilitatorcan publicly ask the newcomer for input, thusmodeling that the newcomer’s opinion is impor-tant. Even if the newcomer is wrong at times, thefacilitator should make sure that the newcomerfeels psychologically safe enough to take risks.Dissent or disagreement can lead to greater inno-vation in the entire group. As Nemeth (1992)found, even when a dissenter is wrong, that dis-senting voice often leads to more divergent think-ing among the entire group. And, it is divergentthinking that is the key to innovation in groups,particularly those that have become complacent intheir thought processes.

… dominant newcomers are actuallyless influential than those who firstadopt the norms of the group and thenintroduce innovations at a later point.

Too often, newcomers are viewed as individualswho must be subsumed into the current culture ofthe group. Instead, they should be viewed as con-sultants, whose fresh perspectives can enhance thegroup’s effectiveness. If group facilitators hope torecruit and retain knowledgeable and committednewcomers, and if they hope to improve the per-formance of their groups, they would do well tocreate conditions that allow for newcomer input.Once the newcomer becomes an old-timer, he orshe is no longer in a position to offer a similarlyfresh perspective. Newcomers are time-limitedresources; we should learn to capitalize on theirephemeral contributions to the group.

ReferencesAlvaro, E. M., & Crano, W. D. (1996). Cognitive

responses to minority- or majority-based com-munications: Factors that underlie minorityinfluence. British Journal of Social Psychology,35, 105-21.

Alvaro, E. M., & Crano, W. D. (1997). Indirectminority influence: Evidence of leniency insource evaluation and counterargumentation.Journal of Personality and Social Psychology,72, 949-964.

Amabile, T. M. (1983). The social psychology ofcreativity: A componential conceptualization.Journal of Personality and Social Psychology,45, 357-376.

Cini


Asch, S. E. (1956). Studies of independence andsubmission to group pressure: I. A minority ofone against a unanimous majority.Psychological Monographs, 70(9), (Whole No.417).

Bach, G. R. (1954). Intensive group psychotherapy.New York: Ronald Press.

Barker, R. G. (1968). Ecological psychology.Stanford: Stanford University.

Brawley, L. R., Carron, A. V., & Widmeyer, W. N.(1988). Exploring the relationship betweencohesion and group resistance to disruption.Journal of Sport and Exercise Psychology, 10,199-213.

Brown, V., Tumeo, M., Larey, T. S., & Paulus, P. B.(1998). Modeling cognitive interactions duringgroup brainstorming. Small Group Research,29(4), 495-526.

Bunce, D., & West, M. A. (1995). Self perceptionsand perceptions of group climate as predictorsof individual innovation at work. AppliedPsychology: An International Review, 44(3),199-215.

Burpitt, W. J., & Bigoness, W. J. (1997). Leader-ship and innovation among teams: The impactof empowerment. Small Group Research, 28(30),414-423.

Cartwright, D. (1968). The nature of group cohe-siveness. In D. Cartwright & A. Zander (Eds.),Group dynamics: Research and theory (3rd ed.,pp. 91-109). New York: Harper and Row.

Cini, M. A. (1994). Innovation by newcomers:Effects of group performance and group devel-opment. Unpublished doctoral dissertation,University of Pittsburgh, Pittsburgh, PA.

Cini, M. A., Moreland, R. L., & Levine, J. M. (1993).Group staffing levels and responses to pro-spective and new members. Journal ofPersonality and Social Psychology, 65, 723-734.

Craig, K. M., (1996). Are all newcomers judgedsimilarly? Distinctiveness of entry in task-ori-ented groups. Small Group Research, 27(3),383-397.

Crano, W. D., & Alvaro, E. M. (1998). The con-text/comparison model of social influence:Mechanisms, structure, and linkages thatunderlie indirect attitude change. EuropeanReview of Social Psychology, 8, 175-202.

Deutsch, M., & Gerard, H. (1955). A study of nor-mative and informational social influences onindividual judgement. Journal of Abnormal andSocial Psychology, 51, 629-636.

Ettlie, J. E. (1980). Manpower flows and the inno-vation process. Management Science, 26(11),1086-1095.

Feldbaum, C. L., Christenson, T. E., & O’Neal, E.C. (1980). An observational study of theassimilation of the newcomer to the preschool.Child Development, 51, 497-507.

Festinger, L. (1950). Informal social communica-tion. Psychological Review, 57, 271-282.

Festinger, L., Schachter, S., & Back, K. (1950).Social pressures in informal groups. New York:Harper.

Fromkin, H. L., Klimoski, R. J., & Flanagan, M. F.(1972). Race and competence as determinantsof acceptance of newcomers in success andfailure work groups. Organizational Behaviorand Human Performance, 7, 25-42.

Gadon, H. (1988). The newcomer and the ongoingwork group. In H. Gadon & N. Josefowitz(Eds.), Fitting in: How to get a good start in yournew job (pp. 161-175). Reading,Massachusetts: Addison-Wesley.

Goodman, M. (1981). Group phases and inducedcountertransference. Psychotherapy: Theory,Research, and Practice, 18, 478-486.

Hackman, J. R. (1992). Group influences on indi-viduals in organizations. In M. D. Dunnette &L. M. Hough, (Eds.), Handbook of industrial andorganizational psychology (Vol. 3, pp. 199-267).Palo Alto, CA: Consulting Psychologists Press.

Heiss, J., & Nash, D. (1967). The stranger in labo-ratory culture revisited. Human Organization,26, 47-51.

Jones, A., & Crandall, R. (1985). Preparing new-comers to enhance assimilation into groups: Agroup therapy example. Small Group Behavior,16, 31-57.

Kadis, A. A., Krasner, J. D., Winick, C., & Foulkes,S. H. (1965). A practicum of group psychother-apy. New York: Harper & Row.

Kaplan, S. R., & Roman, M. (1961). Characteristicprocesses in adult therapy groups to the intro-duction of new members: A reflection on groupprocesses. International Journal of GroupPsychotherapy, 11, 372-381.

Katz, R. (1982). The effects of group longevity onproject communication and performance.Administrative Science Quarterly, 27, 81-104.

King, N., & Anderson, N. (1990). Innovation inworking groups. In M. A. West & J. L. Farr(Eds.), Innovation and creativity at work (pp.81-100). West Sussex, England: John Wiley &Sons.

Kormanski, C. (1988). Using group developmenttheory in business and industry. Journal forSpecialists in Group Work, 13, 30-43.

Leopold, H. S. (1961). The new member in thegroup: Some specific aspects of the literature.International Journal of Group Psychotherapy,11, 367-371.



Levine, J. M., & Moreland, R. L. (1990). Progress insmall group research. In M. Rosenzweig & L. W.Porter (Eds.), Annual review of psychology (Vol.41, pp. 585-634). Palo Alto, CA: AnnualReviews.

Levine, J. M., & Russo, E. M. (1985). Majority andminority influence. In C. Hendrick (Ed.),Review of personality and social psychology:Group processes (Vol. 8: pp. 13-54). NewburyPark, CA: Sage.

Lofland, J. F., & Lejeune, R. A. (1960). Initial inter-action of newcomers in Alcoholics Anonymous:A field experiment in class symbols and sociali-zation. Social Problems, 8, 102-111.

Louis, M. R. (1980). Surprise and sense making:What newcomers experience in entering unfa-miliar organizational settings. AdministrativeScience Quarterly, 25, 226-251.

Maass, A., & Clark, R. D., III (1984). Hidden impactof minorities: Fifteen years of minority influ-ence research. Psychological Bulletin, 95, 428-450.

Maass, A., West, S. G., & Cialdini, R. B. (1987).Minority influence and conversion. In C.Hendrick (Ed.), Group processes: Review of per-sonality and social psychology (Vol. 8, pp. 55-79). Newbury Park, CA: Sage.

Merei, F. (1949). Group leadership and institution-alization. Human Relations, 2, 23-39.

Mills, T. (1957). Group structure and the newcomer.Oslo: Oslo University.

Montouri, A., & Purser, R. (1995). Deconstructingthe lone genius myth: Toward a contextual viewof creativity. Journal of Humanistic Psychology,35(3), 69-112.

Moreland, R. L. (1985). Social categorization andthe assimilation of “new” group members.Journal of Personality and Social Psychology,48, 1173-1190.

Moreland, R. L., & Levine, J. M. (1988). Groupdynamics over time: Development and sociali-zation in small groups. In J. E. McGrath (Ed.),The social psychology oftime: New perspectives(pp. 151-181). Newbury Park, CA: Sage.

Moreland, R. L., & Levine, J. M. (1989). Newcomersand oldtimers in small groups. In P. B. Paulus(Ed.), Psychology of Group Influence (pp. 143-186). Hillsdale, NJ: Erlbaum.

Moscovici, S. (1976). Social influence and socialchange. London: Academic Press.

Moscovici, S. (1985). Social influence and confor-mity. In G. Lindzey & E. Aronson (Eds.), Thehandbook of social psychology (Vol. 2, pp. 347-412). New York: Random House.

Moscovici, S., Lage, E., & Naffrechoux, M. (1969).Influence of a consistent minority on theresponses of a majority in a color perceptiontask. Sociometry, 32, 365-380.

Nash, D., & Wolfe, A. W. (1957). The stranger inlaboratory culture. American SociologicalReview, 22, 400-405.

Nemeth, C. (1986). Differential contributions ofmajority and minority influence. PsychologicalReview, 93, 23-32.

Nemeth, C. (1992). Minority dissent as a stimulantto group performance. In S. Worchel, W. Wood,& J. A. Simpson (Eds.), Group process and pro-ductivity (pp. 95-111). Newbury Park, CA: Sage.

Nemeth, C. J., & Staw, B. M. (1989). The tradeoffsof social control and innovation in groups andorganizations. In L. Berkowitz (Ed.), Advancesin experimental social psychology (Vol. 22, pp.175-210). New York: Academic Press.

Nicholson, N. (1984). A theory of work role transi-tions. Administrative Science Quarterly, 29,172-191.

Petty, R. M., & Wicker, A. W. (1974). Degree ofmanning and degree of success of a group asdeterminants of members’ subjective experi-ences and their acceptance of a new groupmember. Psychological Documents, 4, 1-22.(Ms. No. 616).

Phillips, E. L., Shenker, S., & Revitz, P. (1951). Theassimilation of the new child into the group.Psychiatry, 14, 1-22. (Ms. No. 616).

Saravay, S. M. (1978). A psychoanalytic theory ofgroup development. International Journal ofGroup Psychotherapy, 28, 481-507.

Schein, E. H. (1971). Occupational socialization inthe professions: The case of role innovation.Journal of Psychiatric Research, 8, 521-530.

Snyder, E. C. (1958). The Supreme Court as asmall group. Social Forces, 36, 232-238.

Sutton, R. I., & Louis, M. R. (1987). How selectingand socializing newcomers influences insiders.Human Resource Management, 26, 347-361.

Tuckman, B., & Jensen, M. (1977). Stages ofsmall-group development revisited. Group andOrganizational Studies, 2, 419-427.

Wanous, J. P., Reichers, A. E., & Malik, S. D.(1984). Organizational socialization and groupdevelopment: Toward an integrative perspec-tive. Academy of Management Review, 9, 670-683.

West, M.A. (1990). The social psychology of inno-vation in groups. In M. A. West & J. L. Farr(Eds.), Innovation and creativity at work:Psychological and organizational strategies (pp.309-333). Chichester, England: Wiley.

Cini


West, M. A., & Anderson, N. R. (1996). Innovationin top management teams. Journal of AppliedPsychology, 81(6), 680-693.

Wood, W., Lundgren, S., Ouellette, J. A., Busceme,S., & Blackstone, T. (1994). Minority influence:A meta-analytic review of social influence proc-esses. Psychological Bulletin, 115(3), 323-345.

Worchel, S., Coutant-Sassic, D., & Grossman, M.(1992). A developmental approach to groupdynamics: A model and illustrative research. InS. Worchel, W. Wood, & J. A. Simpson (Eds.),Group process and productivity (pp. 181-202).Newbury Park, CA: Sage

Yalom, I. D. (1970). The theory and practice ofgroup psychotherapy. New York: Basic Books.

Ziller, R. C. (1962). The newcomer’s acceptancein open and closed groups. PersonnelAdministration, 5, 24-31.

Ziller, R. C., & Behringer, R. D. (1960). Assimila-tion of the knowledgeable newcomer underconditions of group success and failure.Journal of Abnormal and Social Psychology, 60,288-291.

Ziller, R. C., Behringer, R. D., & Goodchilds, J. D.(1960). The minority newcomer in open andclosed groups. Journal of Psychology, 50, 75-84.

Ziller, R. C., Behringer, R. D., & Jansen, M. J.(1961). The newcomer in open and closedgroups. Journal of Applied Psychology, 45, 55-58.

Acknowledgements

I wish to thank Marianne Leister and Boris Vilic for their helpful comments on an earlier draft of thispaper.

Marie A. Cini is the Director of Academic Affairs at the James MacGregor Burns Academy of Lead-ership at the University of Maryland. She has published work on leadership development, onteaching leadership online, and on the psychological factors that draw individuals to groups andorganizations. Her work has been published in The Journal of Personality and Social Psychology,The Journal of Public Management and Social Policy, and The Journal of Leadership Studies. Dr.Cini earned her Ph.D. in social psychology from the University of Pittsburgh in 1994, with empha-ses in group processes and research methodology. Contact at: James MacGregor Burns Academyof Leadership, 1111 Taliaferro Hall University of Maryland, College Park MD 20742-7715; Phone:412-396-5839; [email protected]

Associate Editor: Sandor Schuman

Group Facilitation: A Research and Applications Journal - Number 3, Spring 200114

Group Development: A Review of the Literature anda Commentary on Future Research Directions

George Smith

The use of groups, and more specifically teams, in organizations has been on the rise during thepast decade. While many benefits have been attributed to these organizational arrangements, fewresearchers and practitioners have stepped back to look at the history and research underlyingmany of the models that are used to understand and anticipate group/team development. Thispaper takes a step in that direction as it reviews many of the developmental models, their rootsand patterns.

Keywords:

Groups, Teams, Group Development, Models, Theories

INTRODUCTION

Today’s organizational landscape is laden withsmall groups and teams. Organizations in the pub-lic, private and not-for-profit sectors are utilizingwork groups, task forces, blue ribbon commissionsand panels, and various types of teams and sup-port groups to accomplish their missions andobjectives. In addition, groups are being imple-mented at all hierarchical levels and have assumedmany roles and responsibilities central to thefunctioning of these organizations. Team andgroup work has become so prevalent thatHackman (1990) has stated that “Virtually every-one who has worked in an organization has been amember of a task-performing group at one time oranother” (p. 2).

Despite the presence of groups in today’sorganizations, there are still questions that need tobe addressed and dealt with. Central to such anexamination is an understanding and evaluation ofthe existing group development literature. Anunderstanding of this literature is important sinceone critical concern that has been raised byBettenhausen (1991) and Nahavandi and Aranda(1994) is whether our current knowledge andunderstanding is sufficient to warrant the use ofthese organizational types and more specificallywhat are our limits in understanding group andteam development. Therefore, it is important thatwe understand what we purport to “know” aboutthis phenomenon prior to any speculation aboutwhat should be done to improve the field.

This paper attempts to address this issue andothers by reviewing and categorizing many of theexisting group development models. In addition,the paper focuses on the commonalities across themodels (especially within categories) and provides adiscussion that raises issues for future research. Ifthis paper serves no other purpose, its primary

intention is to act as a stimulus to reopen investi-gation into group development and to inject someadditional doubt into the models, ideas and theo-ries that are currently being implemented on thejob, taught in the classroom and used in relatedresearch endeavors.

GROUP DEVELOPMENT DEFINED

As a point of entry into this work it is necessaryto define the phrase “group development.” Findingan acceptable definition of this phrase is difficultbecause it has been defined in many ways andwith varying degrees of depth and complexity.Berkowitz (1974), for example, has stated thatgroup development “refers to the fact that groupprocess undergoes modification which enables thegroup to have more alternative ways to solve prob-lems” (p. 311), while Ridgeway’s (1983) and Bennisand Shepard’s (1956) definitions of group develop-ment focus on the group’s need for improved com-munication patterns.

The definition that best fits the models dis-cussed in this paper was provided by Sarri andGalinsky (1974). In this definition these twoauthors have defined group development as“changes through time in the internal structures,processes, and culture of the group” (Sarri andGalinsky, 1974, p. 72). As outlined in this defini-tion group development entails changes withinthree different dimensions. The first dimension, thesocial dimension, is concerned with the organiza-tion of the group’s structure and the patterns ofthe participants’ roles and structures. The seconddimension focuses on the group’s activities, tasksand the operative processes of the group and islabeled the activity dimension. The final dimen-sion, group culture, includes properties such asgroup norms, values and a shared group purpose.

Smith


The definition is important to the presentdiscussion as it outlines the types of activities andevents that groups may deal with during theirdevelopment. More specifically, the definition issignificant in that it serves as a representation ofthe types of changes that occur in the variousdevelopmental models. As previously noted, theprimary reason for including this definition is toestablish an understanding of what is meant bythe phrase “group development.” While the defini-tion is relevant to the models discussed in thispaper, the crux of the review deals with an exami-nation of the similarities and differences betweenthe existing models with the intent of raising ques-tions and issues that relate not only to the modelsthemselves but to various methodological issuesused in gathering the data and developing themodels. It should however be noted that the mod-els in this paper do contain provisions for each ofthree dimensions outlined in the Sarri andGalinsky definition.

BRIEF OVERVIEW OF THE GROUP DEVELOPMENTLITERATURE AND HISTORY

While a large body of group development litera-ture exists today, Hare (1973) has noted that con-certed interest and attention to group develop-mental issues and trends really only began to growfollowing the publication of Bales and Strodtbeck’s(1951) work examining phases in group problemsolving and Bion’s (1961) work and experienceswith psycho-analytic groups of neurotic patients inEngland. Currently, groups, and more specifically,the use and development of work teams continueto be prominent issues in all types of organizationsand at different hierarchical organizational levels.

Because of the long history of interest andresearch in group development, only crude esti-mates can be made as to the number of groupdevelopment theories that exist. In addressing thisissue, Hill and Gruner (1973) discuss the fact thatHill was once an avid collector of group develop-ment theories. At one point, they report, he met upwith another collector of group development theo-ries and found that their combined collectionsyielded more than 100 distinct theories.

Hill’s meeting occurred in 1959. Today, one canonly speculate as to the total number of theoriesand models that exist. Therefore, the review pre-sented here should by no means be considered anexhaustive review of the group development litera-ture, since it would be nearly impossible to obtain,much less review all of the group developmenttheories that exist.

This paper has been written with two maingoals in mind. The first goal is to provide a con-temporary review of the group development litera-ture, examining some of the more recent modelsand theories that have been proposed and pre-senting an overview of the preceding work. Thesecond goal of this paper is to briefly critique andcomment on the state of the overall body of groupdevelopment literature. As will be seen, there isstill much to learn about group development.Furthermore, there are many problems associatedwith how the current body of group developmentresearch has been conducted and how “popular” orcurrently accepted, taught and practiced modelshave gained their notoriety. Finally, there are stillquestions that have not been answered oraddressed in the past and in some instances areonly now beginning to be examined andresearched.

GROUP DEVELOPMENT MODEL CLASSIFICATIONSCHEME

In organizing this voluminous body of research,an effort was made to develop or find a method forclassifying and placing the various models andtheories into a simplified framework. This processwas made difficult by the fact that while manyresearchers have stated that the theories sharecommon ideas and, in some cases, stages of devel-opment, they also often display slight and subtledifferences.

One critical concern that has been

raised ... is ... what are our limits in

understanding group and team devel-

opment.

During the course of this literature review, fiveclassification frameworks were encountered. Theearliest of these frameworks was Gibbard,Hartman and Mann’s (1974) and the most recentwas Mennecke, Hoffer and Wynne’s (1992). Theframeworks and classification categories can beseen in Table 1.

The five frameworks share several commonthemes. One shared themes is that each of theframeworks presents a classification that depictsgroup development as a linear, step-by-step proc-ess. Gibbard, Hartman and Mann (1974), forexample, apply the specific label of linear-progres-sive to this class of models while Mennecke et al.(1992) refer to the models as being progressive in

Group Development: A Review of the Literature and a Commentary on Future Research Directions

Special Issue on Group Development16

Table 1: Group Development Classification Frameworks

Gibbard et al.(1974)

Shambaugh(1989)

Poole(1989)

McCollom(1990)

Mennecke et al.(1992)

LinearProgressive

Sequential UnitaryPhase

Performance Progressive

Life Cycle RecurringPhase

Non Phasic Emotional Cyclical

Pendular orRecurring Cycle

ChangingCentrality

Contingency Revolt Non-sequential

nature as they build step-by-step toward higherlevels of productivity. In McCollom’s (1990) frame-work the argument could be made that two of hisclasses, performance and emotional, are linearprogressive. The emotional model category con-tains models that build hierarchically and his per-formance model category describes models that arefocused on improvements in performance overtime.

A second theme that emerges in three of thefive models, Gibbard et al. (1974), Shambaugh(1989), and Mennecke (1992), is a class of groupdevelopment models describing recurring, pendu-lar or cyclical phases. The models in this categorydepict the group development as a continual proc-ess of cycling or swinging between phases. Thecycling or recurrence of phases often results fromadditional knowledge or understanding of thegroup’s task. Poole’s non phasic models may alsofit this category. In his description of these models,Poole notes that groups and their members gothrough periods where they are in and out ofsynch. Thus, group development could be viewedas a constant cycling between these two phases.

Mennecke et al.’s (1992) framework was usedas a starting point for classifying the group devel-opment models reviewed in this paper. One reasonfor selecting this model was that it had a history inthat it built upon and updated Gibbard et al.’s(1974) work. A second reason for using this frame-work was that it included the two commonlyreported models types, linear-progressive andcyclical or recurring cycle, found in the four otherframeworks. The third reason for basing theframework used here on Mennecke et al.’s workwas that by virtue of it being the most recent, itrecognized and classified several of the newergroup development models.

Several modifications were made to Menneckeet al.’s framework. One notable change was themoving of the life-cycle models from under thecyclical model category to the linear progressivecategory. This change was made based on thebelief that life-cycles proceed in a linear fashion(birth-maturation-death). Even if an organism/group was born and immediately died it would stillhave progressed in a linear fashion from birth todeath.

Insights from the four other frameworks werealso applied in modifying Mennecke et al.’s frame-work. For example, Mennecke et al.’s category ofnon-sequential models was expanded to includehybrid models. Hybrid models come in two types.One type addresses the existence/affect that con-tingent factors may have on determining develop-mental patterns. This type was presented in Poole’scategory of contingency models. The second type ofhybrid model emerges from the combining of devel-opment models or theories. An example of this typeof model is Morgan, Salas and Glickman’s (1993)TEAM model that merged ideas found inTuckman’s (1965) model with the ideas of thecyclical models. The revised framework is depictedin Table 2.

The present framework divides group develop-ment models into three categories: linear progres-sive models, cyclical and pendular models, andnon-sequential or hybrid models. Linear progres-sive models are models that “imply that groupsexhibit an increasing degree of maturity andperformance over time” (Mennecke et al., 1992, p.526). Cyclical and pendular models are defined asmodels that “imply a recurring sequence of events”(Mennecke et al., 1992, p. 526). Non-sequentialmodels are models that “do not imply any specificsequence of events; rather, the events that occurare assumed to result from contingent factors that

Smith


Table 2: Categorization of Group Development Models

Linear Progressive Models Cyclical & Pendular Models Non-phasic/Hybrid Models

Bennis & Shepard (1956) Bales & Strodtbeck (1951) McGrath (1986)

Mills (1964) Schutz (1958) Gersick (1988)

Tuckman (1965) Hare (1973) Poole (1989)

Sarri & Galinsky (1974) Napier & Gershenfeld (1973) McCollom (1990)

Braaten (1974/1975) Srivastva, Obert & Neilsen(1977)

Morgan, Salas & Glickman(1993)

Heinen & Jacobson (1976) Bradford (1978)

Tuckman & Jensen (1977) Drexler, Sibbet & Forrester(1991)

Caple (1978)

Lacoursiere (1980)

Kormanski & Mozenter(1987)

Maples (1988)

change the focus of the group’s activities”(Mennecke et al., 1992, p. 526-7).Hybrid modelstend to be models that combine several differentmodels to form a new model. The hybrid modelsare grouped with the non-sequential modelsbecause they also do not propose a specific,ordered pattern of group development. These threegroupings and the specific theories that are classi-fied into each category will be examined in greaterdetail in the following sections of the paper.

LINEAR PROGRESSIVE MODELS

The linear-progressive models are perhaps thebest known and most widely cited type of develop-mental model. These models assume that groupsdevelop in a definite linear fashion and that thereis a “definite order of progression” from one phaseor stage to another. Table 3 provides a list of thevarious linear progressive models reviewed in thispaper, and also presents the sequential stages orphases as defined by the various researchers.According to these researchers, group developmentwould proceed in exact accordance with the hierar-chical lists located below each heading. For exam-ple, in Mills’ (1964) model a group would firstexperience the encounter, next test boundaries

and model roles, then proceed to negotiate a nor-mative system, followed by the production stageand finally arrive at a point in which group mem-bers would separate or exit the group.

Gibbard, Hartman and Mann (1974) add thatthe models are also considered to be sequential intime because passage to succeeding stages orphases implies that a group is becoming moredeveloped or mature. A good analogy to themechanics of these theories would be a flight ofstairs. In order to get to the top a group must walkup each individual step. In the case of groupdevelopment, the group must deal with eachsuccessive stage or phase.

Models that have been categorized as “life-cyclemodels” in prior reviews (Mennecke et al., 1992)are also included in this category. In previous work(Gibbard et al., 1974 and Mennecke et al., 1992),life-cycle models were grouped with the cyclical orpendular models. After reviewing these models, itwas apparent that the developmental patterns dis-cussed in the life-cycle models are more similar tothe patterns described in the linear progressivemodels since these models have definite linear

Table 3: Linear Progressive Model Theorists and Stages or Phases

Bennis andShepard (1956)

Mills(1964)

Tuckman(1965)

Sarri andGalinsky

(1974)

Braaten(1974/ 75)

Heinen andJacobson

(1976)

Tuckmanand Jensen

(1977)

Caple(1978)

Lacoursiere(1980)

Kormanski& Mozenter

(1987)

Maples(1988)

Phase 1DependenceSubphase 1:Dependence-

Flight

Encounter Forming Origin

Formative

Pregroup Forming Forming Orientation Orientation Awareness Forming

Subphase 2:Counter-

dependence-Flight

TestingBoundaries& Modeling

Roles

Storming Inter-mediate 1

InitialPhase

Differentia-tion

Storming Conflict Dissatisfac-tion

Conflict Storming

Subphase 3:Resolution-Catharsis

Negotiatingan

IndigenousNormative

System

Norming Revision Early Phase Integration Norming Integration Resolution Cooperation Norming

Phase 2Interdependence

Subphase 4:Enchantment-

Flight

Production Performing Inter-mediate 2

MatureWork Phase

FullMaturity

Performing Achieve-ment

Production Productivity Performing

Subphase 5:Disenchantment-

Flight

Separation Maturation Termination Adjourning Order Termination Adjourning

Subphase 6:ConsensualValidation

Termination



paths of development and progress sequentiallyfrom infancy to adulthood/ maturity and typicallyto death. One potential reason for the separation ofthese two models categories may be attributable tothe fact that life cycle models explicitly include atermination phases whereas many linear progres-sive models assume that such actions, if they areto occur, will happen at some point in the finaldevelopmental period. Tuckman and Jensen (1977)note for example that the original Tuckman (1965)model included such actions in its “original” finalstage, but later explicitly added the adjourningstage based on beliefs and publications of the life-cycle theorists.

The Development of Linear-Progressive Models.

The linear-progressive models operate underthe assumption that group development is basedon the completion of a definite and ordered set ofstages, phases or periods. As previously noted, themodels that are included in this specific categoryare some of the best known and tested of the groupdevelopment models. These models also tend to bethe models that are typically used in teachinggroup development and can be found in writtenmaterials ranging from academic textbooks toorganizational handbooks to articles appearing inthe popular press.

Linear progressive models have been developedfrom a number of varied approaches to studyinggroups and their development. The predominanttechnique used appears to have been observation(e.g., Bennis and Shepard, 1956; Mills, 1964;Braaten, 1974/75; Lacoursiere, 1980; and Maples,1988). The activities and settings in which groupswere observed have been diverse. For example, oneof the more prevalent types of groups used in thesestudies during the 1960s were undergraduate andgraduate students (e.g., Bennis and Shepard,1956; Mills, 1964; Maples, 1988).

Another frequently used group was groupscomposed of various types of medical and psychi-atric patients and individuals receiving counselingand/or training to deal with different addictions orspecial needs (e.g., Sarri & Galinsky, 1974;Lacoursiere, 1980). Many of these researcherscoupled their observational techniques with theirown personal experiences and insights from work-ing with groups as either facilitators, participantsor instructors (e.g., Braaten, 1974/75; Caple,1978; Lacoursiere, 1980).

Perhaps the next most common method ofmodel development has been development fromreviews of the existing body of group developmentliterature. The best known example of this type of

model development would be Tuckman’s (1965,Tuckman and Jensen, 1977) stages of small groupdevelopment. Other examples of these types ofmodels would include: Braaten (1974/75), Heinenand Jacobson (1976) and Kormanski and Mozenter(1987). Table 4 notes the basis of model develop-ment for each of the models included in this sec-tion.

The developmental processes in the various lin-ear progressive and life-cycle models exhibit manysimilarities in their depiction of group develop-ment. These similarities and some of the subtledifferences will be discussed in the remainder ofthis section.

Stage or Period of “Forming”.

In virtually all of the linear progressive models,the first developmental stage or period of develop-ment is an unspecified length of time in whichgroup members meet together in a physical space,become acquainted or familiar with each other andorient themselves to the task or work to be per-formed. Several different events and processesmark this stage. One specific and important ongo-ing process is a period of “boundary testing” (e.g.,Tuckman, 1965; Tuckman and Jensen, 1977;Heinen and Jacobson, 1976; Caple, 1978). During“boundary testing,” individuals actively attempt todefine the task, while at the same time work toestablish an identity within the group (e.g., Bennisand Shepard, 1956; Tuckman, 1965; Braaten,1974/75; Tuckman and Jensen, 1977; Kormanski& Mozenter, 1987). Often, each individual’s iden-tity is established upon the knowledge and skills(or lack of these qualities) that the individualbrings to the group.

Individual anxiety and disillusionment are twoother related themes that exist in many of thesemodels (e.g., Bennis and Shepard, 1956; Mills,1964; Caple, 1978; Kormanski and Mozenter,1987; Maples, 1988). In part, the anxiety resultsfrom meeting new individuals, being in a newphysical setting, and having a cursory understanding that to succeed (perform the task)all group members must find some way toovercome prejudices and similar problems. At thispoint in the development process, group membersoften deal with their anxiety by reverting to sociallyacceptable norms of behavior that have been usedin similar settings or situations (Bennis andShepard, 1956). Thus, individual group membersenact ideas and norms that they have used in theirexperiences with other groups and in dealing withunknown settings and people (Maples, 1988). Table5 lists the various theorists and names for stagesthat relate to a period of group formation.

Smith


Table 4: Theorists and Basis for Linear Progressive Model Development

Theorist(s) Basis of Theoretical Development

Bennis and Shepard (1956) Five years of observation of graduate students in a group dynamicscourse

Mills (1964) Observation of undergraduate students in training groups at Harvard

Tuckman (1965) Review of group development literature

Sarri and Galinsky (1974) Appears to have been developed from teaching social group workmethods and involvement directing this work

Braaten (1974/75) Review of fourteen existing models to develop composite model. Directobservation of 25 to 30 Scandinavian groups over a ten year period todevelop “his” model.

Heinen and Jacobson (1976) Review of existing literature and in particular a review of the literatureaddressing the issue of task group development

Tuckman and Jensen (1977) Review of group development literature applying Tuckman's 1965 model

Caple (1978) Personal experience facilitating groups, teaching group counseling andleading task groups

Lacousiere (1980) Observation and experience with various types of groups includingpsychotherapy and training groups at a VA hospital

Kormanski and Mozenter (1987) Review of current theories of group development. Based on Tuckman(1965) because work is a summarization

Maples (1988) Extension of Tuckman (1965) using feedback from student journals

Another event that occurs in groups that have aformally appointed leader is an attempt at deter-mining the leader’s ability to lead the group andtesting of his or her legitimacy in doing so. Sincethis a period of insecurity, group members areseen to be highly dependent on the appointedleader for direction (Bennis and Shepard, 1956;Mills, 1964). What often has been seen to occur isthat the members test the leader in terms of his orher ability to assess the situation and provide

sound judgment in dealing with the current groupcontext.

In sum, this period of a group’s development isa time in which group members are actively seek-ing a direction for action and attempting to estab-lish a sense of stability and purpose.

Stage or Period of Conflict and Unrest

The next general occurrence in these models is atransition to a period of unrest, disagreement and

conflict. The conflict(s) can arise for a number ofreasons. One reason is that in the process ofbecoming acquainted (spending time with oneanother), people have discovered others in thegroup who are similar to them and with whom theymost readily identify. When this occurs the groupcan be pulled into factions or cliques that mayfight among themselves for power and leadershipin the group. In other instances unrest becomesthe predominant behavior as some of thesubgroups identify with the appointed leader, whileothers stand in opposition to the leader (Braaten,1974/75). Table 6 lists the various theorists andnames for the corresponding stages that relate tothis period of group development.

Mills (1964) notes that if a formal leader is pre-sent in the group, group members may test thatindividual’s role and related responsibilities. InMills’ study, the leader was the instructor and theinstructor had more of a passive, nurturing rolethan a directive, authoritarian role. Thus, in Mills’



Table 5: Stages of Formation in Linear Progressive Models

Theorist(s) Corresponding Stage(s) Position of Stage(s) in Model

Bennis & Shepard (1956) Dependence-Flight First

Mills (1964 The Encounter First

Tuckman (1965) Forming First

Sarri & Galinsky (1974) Origin & Formative First and Second

Braaten (1974/75) Initial Phase Firsta

Heinen & Jacobson (1976) Forming First

Tuckman & Jensen (1977) Forming First

Caple (1978) Orientation First

Lacoursiere (1980) Orientation First

Kormanski & Mozenter (1987) Awareness First

Maples (1988) Form First

a In Braaten’s model, developed from his review of the literature, this would be the second stage. However,in the model that is his “own,” this stage is the first stage as he dropped his composite model’s prelimi-nary stage based on the premise that many groups do not have the ability to determine membership.

model, individuals pass through a time when theylook to the instructor to be directed to a time whenthe participant has become highly independent ofthe instructor. In similar fashion, Bennis andShepard (1956) have noted that frustration maymount with a leader’s apparent inability to provideanswers and direction to the group’s problems.This inability is seen to divide the group into “twowarring subgroups.” At the heart of the “warringsubgroups” concerns is disagreement over leader-ship and “structure” (Bennis and Shepard, 1956).

Contention and unrest may also arise from apolarization between individuals in a group whoare “task oriented” and members who have atendency to be more “people oriented” (Caple,1978). Essentially, what theorists adopting thisposition are saying is that groups must find a wayto balance group task needs with individual emo-tional needs.

A final reason for conflict in a group may bethat individual members are fighting to maintaintheir own individuality, rather than be swallowedup by the group and its developing identity(Tuckman, 1965). A related problem that exists ingeneral in implementing group structures (teams)in organizations is that many individuals areunwilling to forgo the individual recognition andrewards derived from their singular efforts for

group rewards and recognition. Thus, some groupmembers may increase their efforts to draw atten-tion to themselves and their accomplishmentswhile neglecting the need for a combined integra-tion of group member efforts.

Contention and unrest may also

arise from a polarization between

individuals in a group who are “task

oriented” and members who have a

tendency to be more “people

oriented”

Stage of Group Identity and Norm Formation

Linear progressive models indicate that follow-ing the period of unrest and contention, groupsenter a stage in which they begin to display cohe-sion and group identity. Table 7 provides a listingof the corresponding stages of group identity andnorm formation in the various models. This periodof time is one in which group members haveresolved many of their differences and more effortand energy is directed to engaging and accom-plishing the group’s assigned task. Tuckman

Smith


Table 6: Stages of Unrest or Conflict in Linear Progressive Models


Bennis & Shepard (1956) Counterdependence-Flight

Second

Mills (1964) Testing Boundaries andModeling Roles

Second

Tuckman (1965) Storming Second

Sarri & Galinsky (1974) Revision Fourth

Braaten (1974/75) Early Phase Seconda

Heinen & Jacobson (1976) Differentiation Second

Tuckman & Jensen (1977) Storming Second

Caple (1978) Conflict Second

Lacoursiere (1980) Dissatisfaction Second

Kormanski & Mozenter (1987) Conflict Second

Maples (1988) Storm Second

aThis stage would be later in Braaten's “own” model. See explanation for Braaten in Table 5.

(1965), for example, describes this period as a timein which a group is displaying cohesiveness andthere has been the creation of “ingroup” feeling (p.396). Mills (1964) notes that during this time “thegroup seeks to define and legislate what it shouldbe” (p. 245). Bennis and Shepard (1956) state thatthis is the “most crucial and fragile” phase of thegroup to this point in its development since thegroup “suddenly” shifts its whole “basis for action”(p. 423). In fact, Bennis and Shepard note that insome groups this phase may not even occur sincethe group may have become so splintered anddivided that it is unable to bridge existing differ-ences and develop a group identity and normativesystem.

Braaten’s (1974/75) model depicts this transi-tion as being one in which groups pass from aperiod of unrest to a period of resolution and per-forming. It is interesting to note in his model thatrather than have a set period of time to addressand remedy some of the problems and conflictsthat have occurred in the past, groups proceeddirectly to a stage in which the problems areresolved and in which the most productive taskwork occurs.

Caple’s sequential stage model is also some-what unique. During this third stage of a group’s

development, the group passes from a polarizedgroup atmosphere into a period in which membersevaluate the present needs of the group and thegroup’s past performance. This evaluation resultsin the group better understanding why norms areneeded and group members, in turn, work to buildadditional cohesion.

The Sarri and Galinsky (1974) model differsslightly from the other models. In this model, thetime period corresponding to the development of agroup identity and norm formation can be inter-preted as being composed of two stages that sur-round a conflictual period. What occurs in thismodel is that after groups “form” and developsimple operating procedures, they enter a stagedesignated as “Intermediate 1” in which there is alevel of cohesion and group behavior directed tocompleting the task. Following Intermediate 1,there is a developmental period in which the grouplapses into a period of conflict that challenges theestablished simple operating procedures and thegroup’s understanding of the task. The conflictualtime period is then followed by “Intermediate 2” inwhich high levels of integration and stability exist.Sarri and Galinsky note that the second intermedi-ate stage may be very short and, in manyinstances, hardly noticeable. The observability of



Table 7: Stages of Group Identity andNorm Formation in Linear Progressive Models


Bennis and Shepard (1956) Enchantment-Flight/Resoution-Catharsis

Third and Fourth

Mills (1964) Negotiating anIndigenous Normative

System

Third

Tuckman (1965) Norming Third

Sarri and Galinsky (1974) Intermediate 1 & 2 Third & Fifth

Braaten (1974/75) Mature Work Phase Thirda

Heinen and Jacobson (1976) Integration Third

Tuckman and Jensen (1977) Norming Third

Caple (1978) Integration Third

Lacoursiere (1980) Resolution Third

Kormanski and Mozenter (1987) Cooperation Third

Maples (1988) Norm Third

aThis stage would be later in Braaten's “own” model. See explanation for Braaten in Table 5

this stage is largely dependent on how good a jobthe group has done in defining the task and proce-dures early in its life.

Stage or Period of Production

Once group cohesion has been established andrules have been further clarified and defined,groups begin to actively produce or perform theirassigned task(s). Several theorists have noted thatthis is a period in group life in which the previouslyestablished rules become more flexible, pliable,and functional (e.g., Tuckman, 1965; Braaten,1974/75). The stages that correspond to thisperiod of production in the various models can beseen in Table 8.

Sarri and Galinsky (1974), in particular,describe this period as one in which the group hasstabilized its structure and purpose, as well as itsoperating and governing procedures. They alsonote that the group’s culture is expanding and thatthe group as a whole has developed effective

responses to the internal and external stressesplaced on it. Caple (1978) also notes that, duringthis time period, groups eagerly and mutuallyexplore and resolve problems.

Stage of Adjournment or Termination

The final stage that eight of the models includeis a point where the group dies off, terminates, oris disbanded. Table 9 provides a listing of the cor-responding stages of adjournment or terminationin the linear progressive models reviewed. ForTuckman, a stage specifically identified with groupbreakup was a late addition. In his and Jensen’s1977 article, they note that it was always assumedthat the stage occurred at some point in the per-forming stage because a group, at some point,would complete its work and no longer be neededby the organization within which it was embedded.

Sarri and Galinsky (1974) propose four reasonsgroups may terminate. The first reason is thatgroups have accomplished their work and thus areno longer needed. The second is that groups may

Smith


Table 8: Stages of Production in Linear Progressive Models


Mills (1964) Production Fourth

Tuckman (1965) Performing Fourth

Sarri & Galinsky (1974) Maturation Sixth

Braaten (1974/75) Mature Work Phase Thirda

Heinen & Jacobson (1976) Full Maturity Fourth

Tuckman & Jensen (1977) Performing Fourth

Caple (1978) Achievement Fourth

Lacoursiere (1980) Production Fourth

Kormanski & Mozenter (1987) Productivity Fourth

Maples (1988) Perform Fourth

aThis stage would be later in Braaten's “own” model. See explanation for Braaten in Table 5.

have been planned to exist only for a specificperiod of time. In this context, one can pictureteams that are hired for seasonal work responsi-bilities or special projects. Third, groups could failand thus die due to a lack of integration. In thiscondition, a group will fail if it is “unable to achieveessential conditions for endurance” (p. 78). Theseessential conditions are: a basic consensus amongmembers about goals, a high level of interpersonalties, a role system which permits sufficientpersonal satisfaction, and successful completion ofmajor tasks or effective operating procedures.Finally, groups could disband because of“maladaptation.” Maladaptation results whengroups cannot deal with internal changes andenvironmental fluxes. Maladaptation may occurwhen a group has not developed changemechanisms to cope with these events or when thegroup’s mechanisms have become rigid and do notallow the group to respond appropriately.

Braaten (1974/75) notes that the final period ofgroup life does not really start in what he identifiesas the final stage. Rather, events leading up to agroup’s termination often begin earlier in thegroup’s developmental sequence and may includethe withdrawal of group-member interest and

commitment, ultimately leading to disengagementfrom the group.

Caple’s (1978) model, at this point, perhaps moreclosely resembles models that are associated withrecurrent or cyclical stages. However, because ofhis position that the stages are sequential, it isgrouped with the other models in this section. Still,Caple does note that, in later stages, groups workto maintain their structure and do not rewardinnovative actions and activities as they did earlierin their lives. Therefore, one must question ifgroups terminate in Caple’s model. If one returnsto the groups he studies, it would seem fairly obvious that, for Caple, group life, at some point, mustend. Furthermore, unless a group is in an envi-ronment that is stagnant (as opposed to dynamic),the likelihood of failure increases as more andmore environmental fluctuations are heaped on agroup that is not interested in innovative ideas as ameans for dealing with these problems. As a result,one can envision such a group becoming trappedin its past and, to some extent, terminated by thatsame past. If, however, a group exists in a highlystable environment that does not require innova-tion, it is possible that the group could exist for anextended period of time, given appropriate condi-tions.



Table 9: Stage of Termination in Linear Progressive Models

Theorist(s) Stage(s) Position of Stage(s) in Model

Mills (1964) Separation Fifth

Tuckman (1965) Performing Fourtha

Sarri and Galinsky (1974) Termination Seventh

Braaten (1974/75) Termination Fourthb

Heinen and Jacobson (1976) No Corresponding Stage N/A

Tuckman and Jensen (1977) Adjourning Fifth

Caple (1978) Order Fifthc

Lacoursiere (1980) Termination Fifth

Kormanski and Mozenter (1987) Separation Fifth

Maples (1988) Adjourn Fifth

a As noted in the text‚ Tuckman (1965) did not originally have a specific end stage, but assumed at somepoint the group would complete its performance. Furthermore, Tuckman & Jensen (1977) note that theyadded "adjourning" largely in response to the life-cycle models of the day.

b This stage would be later in Braaten's “own” model. See explanation for Braaten in Table 5.c Caple indicates that termination need not necessarily occur‚ although it may

CYCLICAL AND PENDULAR MODELS

The second type of developmental modelreviewed here is the cyclical or pendular model.Models included in this category were developedbased on observations and the notion that groupsrevisit stages and phases over and over or swingbetween issues again and again during the devel-opmental process. According to the theorists andresearchers proposing these models, groups must

constantly address similar issues and problems atmultiple time periods and settings, for reasonsranging from changes in the external environment,to changes in group membership, to changes in thenature of the group task. Each cycle or swingserves to strengthen the group’s understanding ofits present situation and its assigned task and tomodify the group’s approach to dealing with thoseissues.The models discussed in this section can be foundin Table 10. Also included in this table are thevarious stages or phases that groups cycle andswing back and forth between as they develop. Thespecific ordering of these stages is not as important

to these models as was the case for the linear pro-gressive models. The authors assume, however,that for a group to fully develop and mature, itmust deal with the issues found in each of thedevelopmental stages or phases. Thus, while theBales and Strodtbeck (1951) model may appear toindicate that groups develop in an orderly fashionfrom the orientation stage through to the controlstage, a group may swing freely back and forthbetween any of the stages until it finds a workablesolution for achieving its objective(s).

Seven specific theories or models are discussedin this section of the paper (see Table 11). Onework that is conspicuously missing is Bion’s (1959)model. The decision to not include this particularmodel was based on several observations. The firstwas a reading and review of Bion’s classic work,Experiences in Groups. In reviewing this book, itbecame apparent that Bion did not write it with theintent that it would be used as a model of groupdevelopment. Rather, the work was intended toshare Bion’s observations and first-hand experi-ences in working with psycho-analytic groups.Furthermore, Hare (1973) states that Bion’s con-cern was with the various emotional states and

Table 10: Cyclical or Pendular Model Theorist and Stages of Phases

Bales &Strodtbeck (1951)

Schutz(1958)

Hare(1973)

Napier &Gershenfeld (1973)

Srivastva, Obert &Neilsen (1977)

Bradford(1978)

Drexler, Sibbet &Forrester (1991)

Orientation Inclusion Latent Pattern-Maintenance &

TensionManagement

The Beginning Inclusion Formation Orientation

Evaluation Affection Adaptation Movement TowardConfrontation

Inclusion→ Influence

Confrontation of aDifficult Problem

Trust Building

Control Control Integration Compromise &Harmony

Influence OvercomingProblem through

CooperativeProblem Solving

Goal or RoleClarification

Decision Making

Goal Attainment Reassessment Influence →Intimacy

GroupReorganization

Implementation

Resolution &Recycling

Intimacy High Performance

Renewal



swings in these groups, rather than with develop-ing a theory of group development. Gibbard,Hartman and Mann (1972) have also noted thatBion “explicitly rejected” placing his assumptionsand observations into a group developmental con-text (p. 86). Finally, Srivastva and Barrett (1988)have stated that Bion’s scheme “is not explicitlydevelopmental” in nature (p. 33). Thus, despite theexistence of other reviews that have includedBion’s work as a cyclical model, the present reviewdoes not because its principal focus is not directlyrelated to the issue of group development.

As can be seen in Table 11, there does notappear to be a strong pattern of similarity in termsof how the models were developed. If any similaritydoes exist, it is that the models tend to beexpanding upon or blending ideas from previouswork and applying that knowledge to the creationof new group development models (e.g., Hare,1973, Napier and Gershenfeld, 1973, Srivastva etal., 1977, Drexler et al., 1991).

The remainder of this section will examine thesimilarities and differences that exist among thesetheories. In general, these model have much incommon with the linear progressive models withrespect to the terminology used to describe groupdevelopment. However, there are differences notonly between these two sets of models, but alsoamong the theories within this section.

Formative Stage

This stage is similar to the first stage of the lin-ear progressive models in that it is a stage orperiod in which the group is seen to physicallycome together or form (see Table 12). Apart fromthe importance of the group converging in one spe-cific physical location, this stage also takes onimportance at the end of a developmental sequenceand will be discussed in more depth in the sectionthat discusses the advent of the recurrence ofdevelopmental cycles or swings in the patterns ofgroup development.

In this early stage, groups are primarily con-cerned with two central activities. The first is thatgroup members seek to define what type of groupthey are in and to define the task or problem thatis being dealt with (e.g., Bales and Strodtbeck,1951; Hare, 1973; Napier and Gershenfeld, 1973).In defining the nature of the group and its task,group members openly share information related tothe problem or task (Bales and Strodtbeck, 1951).The shared information is then used to define

group and task boundaries (Bales and Strodtbeck,1951). The sharing of information increases thegroup’s knowledge of the task’s requirements anddemands. With this understanding, the group pro-ceeds to establish goals and define work relation-ships and patterns that will allow it to begin workon its task (Hare, 1973; Bradford, 1978).

It is also during this stage that members beginto familiarize themselves with each other anddetermine if they desire to be a part of the group(provided such a choice exists) (e.g., Schutz, 1958;Srivastva et al., 1977; Bradford, 1978; Drexler etal., 1991). Srivastva et al. (1977) note that duringthis time period the familiarity with one anothermay be kept at a “superficial level” (p. 99). Groupmembers may only be interested in familiarizingthemselves with issues such as names, places oforigin, age, marital status, number of children andeducational and professional backgrounds(Srivastva et al., 1977, p. 99). The familiarity thatresults from this second central activity is impor-tant in determining the working relationships thatare necessary to accomplish the group’s work andin building the trust necessary to carry out thetask. This familiarity also helps to control the levelof anxiety group members may be experiencing.Furthermore, the knowledge permits each individ-ual an opportunity to find his or her unique place(identity) within the group.

Information Gathering, Goal and Role Clarification

As noted earlier, there tends to be some overlapof phases or stages depending on the specificnumber of stages authors have used to describetheir models and observations. For example, thereis some overlap in the activities included in theformative stage and subsequent activities involvinginformation gathering and role clarification. Theoverlap can be attributed to the fact that there is arecurrence of events, whereby each recurrencetheoretically enables the group to obtain a higherdegree of development and understanding of thegroup’s task, environment and composition.

Bradford (1978), for example, has noted thateach time a group meets, it essentially must reformand revisit the ideas and decisions that had beenpreviously accepted and adjust them. This positionis based on the idea that groups are open systemsand individual group members live outside of thegroups and can bring new ideas, possibilities, andproblems to a group in each successive meeting.

Smith


Table 11: Theorists and Basis for Cyclical or Pendular Model Development

Theorist How Model Developed

Bales and Strodtbeck (1951) Observations of all male, leaderless groups solving human relationsor construction problems at the Laboratory of Social Relations atHarvard University. Members had no prior knowledge of oneanother.

Schutz (1958) Developed from study of therapy groups, encounter groups and T-groups

Hare (1973) Model builds upon Parson’s functional theory of groups

Napier and Gershenfeld (1973) Developed from a review of the existing literature

Srivastva, Obert and Neilsen (1977) Developed from a review of the literature and some insights andobservations from personal experiences

Bradford (1978) Work draws on concepts of a previous paper prepared by authorand Thomas J. Mallinson for a summer session of the NTL in groupdevelopment.

Drexler, Sibbet and Forrester (1991) Blends Jack R. Gibb’s (Bradford, Gibb and Benne, 1964) researchon group behavior with process theories of Arthur Young (1976a, b).

Table 12: Formative Stage(s) or Phases in the Cyclical or Pendular Models

Theorist(s) Corresponding Stage or PhaseName

Stage in Author'sPresentation

Bales & Strodtbeck (1951) Orientation First

Schutz (1958) Inclusion First

Hare (1973) Latent Pattern-Maintenance &Tension Management

First

Napier & Gershenfeld (1973) The Beginning First

Srivastva, Obert & Neilsen (1977) Inclusion (Safety vs. Anxiety) First

Bradford (1978) Formation First

Drexler, Sibbet & Forrester (1991) Orientation &Trust Building First & Second

During this period of a group’s life the groupbecomes more concerned with clarifying its pur-pose and understanding the skills and resourcesnecessary to complete the task (see Table 13 forcomparable stages). The central issue of groupdevelopment is achieving a better understanding ofthe group's work.

Bales and Strodtbeck (1951) note that this is aperiod in group life where the group evaluates theavailable information that has been accumulatedthrough search and solicitation of ideas and opin-ions on the task. Hare (1973) notes that in thisstage, the group becomes more aware and knowl-edgeable about what skills are related to the

Smith


group’s goal or purpose and group members beginto openly display these skills for other members.

Srivastva et al. (1977) observe that group mem-bers during this stage begin to form dyadic rela-tionships based on the similarities that have beenshared and displayed by individual group mem-bers. Bradford (1978) notes that, among otherthings, this period is marked by a readjustment ofstructure which coincides with Srivastva et al.’sidea that this time period provides the group mem-bers with an opportunity to test the group’sboundaries and model specific roles.

In reviewing the work of Drexler et al. (1991),Bradford (1978), and Hare (1973), it can be seenthat this period marks a point in group life wherethe challenge of achieving a goal or overcoming aproblem becomes more “real” to the group. As aresult, the group alters its structural patterns,roles, and relationships in accordance with theincreased understanding. For some groups, thisstage may be more time consuming and difficultthan for others; thus it may indeed representBradford’s (1978) stage of “Confrontation of a Diffi-cult Problem” or at least a major challenge to thegroup’s initial understanding of its purpose andtask.

Schutz’s theory is an exception in that it reallydoes not appear to posit such a phase. The reasonfor this may be because his focus was more onsocial control in groups than on work or taskgroups. Thus, Schutz’s model posits that someindividuals have a need to control, some to beincluded (affection), and others to be controlled.Schutz’s second stage, “control,” could have beenplaced with the other theories’ first stage because,in the formative stage, group members areattempting to establish who they are in relation tothe other members and, in some cases, are lookingfor someone to lead them in completing the group’stask. This stage could also be included here, how-ever, since some members in the group will bemore powerful in shaping the group's direction inpursuit of its goal, while others are more willing tobe led and controlled, provided that the work isaccomplished.

Decision Making and Structural Stabilization

Some of the issues relating to individual rolesand group structure carry over into this next stage(see Table 14 for corresponding stages). Of centralconcern during this stage of group life is the crys-tallizing of work patterns and relationships and thespecification of structural arrangements that allowfor the completion of the group task or problem.

Table 13: Information Gathering/ Role Clarification Stage(s) or Phases in Cyclical or Pendular Models

Theorist(s) Corresponding Stage orPhase Name


Bales and Strodtbeck (1951) Evaluation Second

Schutz (1958) NAa NA

Hare (1973) Adaptation Second

Napier and Gershenfeld (1973) The Beginning/Movement TowardConfrontation

First and Second

Srivastva, Obert and Neilsen (1977) Inclusion _ Influence(Similarity vs. Dissimilarity)

Second

Bradford (1978) Confrontation of a Difficult Problem Second

Drexler, Sibbet and Forrester (1991) Goal or Role Clarification Third

a This model does not posit a stage that would be related to information gathering or role clarification.

Smith


Table 14: Decision Making and Structural Stabilization Stage(s) or Phases in Cyclical or Pendular Models



Bales and Strodtbeck (1951) Evaluation/Control Second/Third


Hare (1973) Integration Third

Napier and Gershenfeld (1973) Compromise and Harmony Third

Srivastva, Obert and Neilsen (1977) Influence (Support vs. Panic) &Influence _ Intimacy

(Concern vs. Isolation)

Third & Fourth

Bradford (1978) Overcoming a Problem throughCooperative Problem Solving

Third

Drexler, Sibbet and Forrester (1991) Decision Making Fourth

a This model does not posit a stage that would be related to decision making or the stabilization of groupstructure.

The primary group concern in this stage is toarrive at a decision on how to proceed in accom-plishing the work. Hare (1973) specifically notesthat at this time, group relationships are restruc-tured and group and social structure evolve in amanner that enables the group to pursue its goals.In addition, Bales and Strodtbeck (1951) state thatduring this stage, group members push for a deci-sion and, in doing so, the ideas of some memberswill be supported while those of others will beneglected and downplayed.

Srivastva et al. (1977) have also noted thatgroups at this point in their existence begin toplace value on the concepts of support, trust,affection, authority and influence. They note thatthis can be a conflictual time in a group’s life asdyads and cliques fight for power and recognition.This conflict however can have positive results as itserves to facilitate the group’s task accomplish-ment.

Implementation and Production

Once a group has passed through the preced-ing steps, it arrives at a point where the plans andstrategies that have been developed to deal withthe group’s goal or task are implemented and actedout. Essentially, during this stage, the group worksits plan.

Srivastva et al. (1977) note that during thisstage of group life, the members have established a“state of complex interdependency” (p. 106). Theinterdependency is based on highly differentiatedgroup members, tightly integrated around theissue of performing the group task well (Srivastvaet al., 1977).

The group thus becomes actively involved in itsproduction function and to some extent conductsmonitoring to determine its success or failure. Thependular swing or recurrence of previous cyclesbecomes very important at this point. Theincreased importance is derived from the fact thatthe group must produce something and meetexpectations. If the group is not meeting thoseexpectations the group must return to address theproblems and restructure in order to meet thetask’s demands and requirements more effectively.Table 15 shows the final stages that are related tothis period of group life.

When and How Recurrence of Cycles Fits into theModels.

For the most part, the theories discussed inthis section are very similar in terms of when arecurrence or swing may take place. Bradford(1978) notes that each successive group meetingprovides the possibility of such a swing. Hare(1973) notes that his model presents four func

Smith


Table 15: Implementation and Production Stage(s) or Phases in Cyclical or Pendular Models



Bales and Strodtbeck (1951) Evaluation & Control Second &Third


Hare (1973) Integration & Goal Attainment Third & Fourth

Napier and Gershenfeld (1973) Resolution and Recycling Fifth

Srivastva, Obert and Neilsen (1977) Influence _ Intimacy(Concern vs. Isolation) &

Intimacy(Interdependence vs.

Withdrawal)

Fourth & Fifth

Bradford (1978) Group Reorganization &Formation

Fourth & First

Drexler, Sibbet and Forrester (1991) Implementation & HighPerformance

Fifth & Sixth

aThis model does not posit a stage that would be related to implementation or production.

tional problems that groups must constantlydeal with during their existence. Drexler et al.(1991) and Srivastva et al. (1977) note that all theproblems that their models discuss are constantlyin play and may need to be addressed at any time.Bales and Strodtbeck (1951) make it explicit thatgroups are constantly involved in a struggle to bal-ance the demands of the task with the emotionaldemands of the group members.

These models also appear to have a secondarycyclical nature to them. Specifically, Hare (1973),Napier and Gershenfeld (1973), and Drexler et al.(1991) note that groups arrive at an end stage,normally occurring after the production stage, andmust determine if the group should continue ordisband. In Hare's model, this decision is madefollowing the goal attainment stage. At this time,the group returns to its latent-pattern mainte-nance stage and decides if the group should pre-pare to disband or start a new developmentalsequence. Some events that can prompt the cyclingto a new sequence are changes in the original goalor the group receiving a new goal. Drexler et al.’s(1991) model is similar to Hare’s model in thatwhen a group has implemented its plan andachieved high performance, it returns to theorientation stage and seeks to address the question

of “Why continue?” In addressing this question,the group assesses if there is still work that needsto be done, and, if so, whether it is still worthwhile.The members also specifically ask themselves if thework has some personal value or meaning.

Perhaps the greatest contribution these modelsmake to our understanding of groups and theirdevelopmental processes is that groups are adapt-able and flexible in dealing with environmentaldemands and constraints. This is evidenced by thefact that these models portray groups able toassess their external and internal environmentsand make either subtle or dramatic changes thatallow them to accommodate the changes they face.Thus, to some extent, these models appear to bemore capable of explaining and predicting groupdevelopment in the “real world,” especially adynamic and changing world, than the rigid andhierarchical linear progressive models, as they spe-cifically address a group’s ability to assess newinformation and make adjustments at variouspoints in its development.

NONSEQUENTIAL/ HYBRID MODELS

The models described in this section of thepaper do not have a prescribed pattern of develop-mental events. These models can be viewed as



“contingency models” of group development in thatthe observed patterns of development are largelythe result of environmental factors, such as time,that affect the development process. Five modelswill be examined in this section.

Poole’s Contingency Model

Poole’s model specifically examines how groupsdevelop as they deal with the task of decisionmaking. This model is based largely on the conceptof structuration or the “process of production andreproduction of social systems via the applicationof generative rules and resources” (Poole et al.,1985, p. 76). Structure refers to the rules andresources people use in interaction. Systems resultfrom the application of structures and lead to“regularized relations of interdependence betweenindividuals and groups” (p. 76). In other words,systems result as group members agree on therules and roles needed to be successful and thenapply them to group work.

Structure can thus be seen to have a dualnature. First, it is the set of rules people use asthey interact with others. The rules that are usedexist, however, only because they are acknowl-edged through their use in interaction. Therefore,structure only exists if group members acknowl-edge a similar (if not exact) set of rules to be usedin the group. Further confounding this distinctionis the fact that, since structure is a product ofhuman actors and interactions, it is continuouslyopen to change based on human understanding,interpretation, and creativity. Poole et al. (1985)reinforce this idea in the following quote:

The meaning of any one act depends on itsplace in the whole, its relation to the wholeensemble of rules that define not a static“game,” but an on-going practice that con-tinues to persist because of its use in thesocial system (p. 77).

Given the nature of the assumptions ofstructuration, Poole and Roth (1989) suggest that,in studying groups, researchers need to alter theirapproach to studying group decision making froma series of “phasic blocks” to a process that views itas a series of “intertwining threads of activity” (p.328). There are three specific threads that areintertwined: task process behavior, behaviors rep-resenting working relationships in the group, andtopical focus (Poole and Roth, 1989).

Task process behaviors are concerned withprocesses such as problem analysis and solutionevaluation. Behaviors that represent working rela-tionships are those that deal with conflict andintegration. Finally, topical focus deals with the

“substantive issues being dealt with in groupactivities” (Poole and Roth, 1989, p. 328). Pooleand Roth maintain that these are not the onlyissues with which a group may be concerned, butare the three principal threads with which groupsmust deal.

In comparing this model to other existing mod-els of group development, Poole and Roth explainthat the orderly patterns that have been observedby others are the result of the threads developingin a “coordinated fashion.” This, they assume,“probably happens most of the time” (p. 328). Thestrength of the model, however, is that it can alsoexplain why other interpretations of group devel-opment exist, in that groups did not deal with thethree main threads in an orderly fashion and/orthe development was also contingent upon otherfactors outside of the scope of the three mainthreads.

To help researchers understand transitions ingroups, Poole and Roth have also described threetypes of “breakpoints” (p. 328). The first type ofbreakpoint is the “normal breakpoint.” Normalbreakpoints are breaks within the group that donot disrupt the group's activities. Two examples ofnormal breakpoints are first, shifts in the topics ofgroup discussion and adjournments in groupactivity. The second type of breakpoint is “delays.”During delays, the group may cycle back, similarto recurring phase models, and repeat and reworkissues that have become problems, makingadjustments according to newly discovered contin-gency factors. The final type of breakpoint is “dis-ruptions.” Disruptions occur when events such asa major conflict “halt a group’s progress” or “whenfailures cause the group to reconsider its work” (p.329).

Poole and Roth also note that any sequence ofgroup development or decision making is compli-cated by two sets of contingency factors: taskcharacteristics and properties of group structure.Task characteristics are concerned with issuessuch as task difficulty and coordination require-ments, while properties of group structure involveproperties that “determine working relationshipssuch as involvement, leadership and consensualnorms” (p. 329). Poole notes that these factorsinfluence three aspects of group development: thesequence or type of group activity and order ofoccurrence, the complexity which is related to thenumber of cycles and the type and frequency ofbreakpoints, and the degree of organization orcoordination among the previously noted activitystrands (p. 329).

Smith


Time, Interaction and Performance Model.

The Time, Interaction and Performance (TIP)Model was developed from reviewing literature thatexamined group development and processes, time,communication, and prior work done by McGrathand colleagues. The model is established on thepremise that groups are “multi-functioned” andthat groups have relations with, and contribute to,systems at three levels: those systems in which thegroup is embedded, individual group members;and the group itself. In presenting these ideas,McGrath notes that while the functions are “ana-lytically distinguishable” (p. 151), they are insepa-rably intertwined . The functions that the groupserves in these relations are the production func-tion, well-being function and member-supportfunction. Each of these functions can be related toa mode of activity. A depiction of these relation-ships can be seen in Table 16.

The TIP model assumes that groups will alwaysbe acting in one of the modes outlined for eachfunction, but that they are not necessarily going tobe engaged in the same mode for each functionsimultaneously. In other words, a group may beworking in Mode I for the function of member sup-port, but be in Mode II for the function of well-being and Mode III for the function of production.In charting the course of a group’s developmental

sequence, the path is largely dependent on thepresence or absence of factors that either supportor interfere with the group’s ability to maintaineach of the three functions of the group. Furthercomplicating this model is the idea that while all ofthe modes are possible in group life, they are notrequired for a group to act and develop (McGrath,1991).

In explaining development using this model,McGrath notes that all groups must participate inModes I and IV. These modes are the beginningand end stages of group development. McGrathassumes that groups will consistently attempt touse this “least complex path” (I to IV) (p. 158). Inselecting any path, groups are seen to choose aphase sequence that is “satisfying” or “least effort”(p. 159). Groups will only choose more complexpaths (those with more modes) if the “conditionswarrant their use” (p. 159). Thus, it is possible forgroups to participate in any combination of modes(so long as I and IV are present) and in any order,depending upon the conditions a group mayencounter. It is even noted that groups may haveto return to prior modes to resolve problems thatdevelop in the course of a group’s maturation proc-ess. Some of the possible phase sequences inwhich a group may be involved are illustrated inFigure 1.

Table 16: Functions and Modes of McGrath’s TIP Model

Functions

Production Well-Being Member Support

Mode I

Inception(Goal Choices)

ProductionDemand/

Opportunity

InteractionDemand/

Opportunity

InclusionDemand/

Opportunity

Mode II

Problem Solving(Means Choices)

TechnicalProblemSolving

Role NetworkDefinition

Position/StatusAttainments

Mode III

Conflict Resolution(Political Choices)

Policy ConflictResolution

Power/PayoffDistribution

Contribution/Payoff

Relationships

Modes

Mode IV

Execution(Goal Attainment)

Performance Interaction Participation

Smith


One potential developmental sequence is that agroup may follow path ‘B’ which would require it tobegin at Mode I, move to Mode II to resolve oraddress a specific issue confronting the group,return to Mode I via path ‘B’ and complete itsdevelopment by going from Mode I to Mode IV onpath ‘A.’ Each potential path a group may follow isnoted by a letter and the various lettered paths canbe chained together to describe a group’s develop-mental pattern. An example of a longer develop-mental path would be a group proceeding fromMode I along path ‘B,’ to Mode II, from Mode II toMode III via path ‘F,’ and finally from Mode III toMode IV by way of path ‘E.’ While the path may belonger, the group has met McGrath’s necessaryconditions of participating in Modes I and IV, butdone so in a more circuitous route because of thespecial conditions or situations in which the grouphas found itself.

In the TIP model, group development must beconsidered in multiple dimensions. Thus, there isnot a single, simple linear path to explain the proc-ess of group development. Mennecke et al. (1992)note that this model specifically argues that groupsshould be looked at and examined in terms of notonly their task-related behaviors but also theirsocio-emotional behaviors. McGrath (1991) feelsthat the use of multiple levels of analysis is neces-sary since there may be underlying issues thatdrive the developmental process and determine itspattern. He also offers, however, the caveat that wecan only approximate group activities becausemuch of the work that is actually performed isdone outside of the group context or arena inwhich researchers have traditionally studiedgroups (McGrath, 1991). Furthermore, he notesthat much work performed under the guise of

group production is really done by individuals orsubgroups acting when the main group is not insession (McGrath, 1991).

The Punctuated Equilibrium Model

Of the six models reviewed in this section,Gersick’s (1988) model appears to have receivedthe most attention in the academic and popularliteratures (texts, journals, etc.). At the heart ofthis model are the issues of time and deadlinesand the concept of “punctuated equilibrium.”

Gersick (1991) notes that one of the paradigmsthat has affected how we view groups and changeis Darwin’s model of evolution. In this modelDarwin describes change as occurring in the formof “a slow stream of small mutations” (Gersick,1991). In contrast to the “Darwinian model,”Gersick cites work from natural historiansEldredge and Gould (1972) that provides an alter-nate theory of evolution. This alternate explanationis referred to as “punctuated equilibrium.”According to the Eldredge and Gould theory, “line-ages (species) exist in essentially static form (equi-librium) over most of their histories, and new spe-cies arise abruptly, through a sudden revolution-ary ‘punctuation’ of rapid change” (Gersick, 1991,p. 11).

...it is important that we understand

what we purport to “know” about

this phenomenon prior to any

speculation about what should be

done to improve the field.

The concept of punctuated equilibrium wasapplied in Gersick’s (1988) study of eight naturallyoccurring teams. Three of the eight teams werestudent teams working on class projects, while theremaining teams were a community fundraisingcommittee, a bank task force, hospital adminis-trators, psychiatrists and social workers, and uni-versity faculty and administrators. Gersick col-lected her data through observation of each meet-ing and generated a “complete transcript” of eachmeeting from her hand written notes and taperecordings. In addition, she interviewed somemembers of the final four groups mentioned aboveto “address aspects of each project's development”(Gersick, 1988, p. 14) that she did not directlyobserve (i.e., project’s history, member expecta-tions, perceptions and evaluation of the project,

Mode III

Mode IV

Mode II

Mode I

A A'

B

B'

C

C'

D

D'

E

E'

F

G

Figure 1: Potential Developmental Paths ofMcGrath’s TIP Model

(reproduced from McGrath, 1991, p.158)



and other events that occurred outside the meetingsetting).

The model resulting from this research wasbased on the observation that teams alternatedbetween periods of stasis and long periods of iner-tia that were “punctuated by concentrated revolu-tionary periods of quantum change” (p. 16). Ulti-mately, Gersick observed that the teams in herstudy passed through five phases or time periodsin their lives.

During the first time period, “first meeting,” thegroup members appeared to have an understand-ing of the group’s situation and how the groupwould behave. Gersick specifically notes in herarticle that the groups she observed did not exhibita period of “storming” as would be expected frommodels such as Tuckman’s (1965). The first meet-ing is where the group develops the plan underwhich it operates through the first phase. Duringthe “first phase” (second time period), teams con-centrate on only a few potential factors that definethe task at hand. In her commentary, the authornotes that there may be a parallel between thisobservation and Simon’s (1976) observation ofbounded and perfect rationality, in that people“must make simplifying assumptions in order totake any action at all” (p. 32).

The third time period occurred at approxi-mately the midpoint of the group’s life. This timeperiod marks a period of transition and greatchange for the group and its members. During thistime, the groups studied abandoned their phaseone agendas. Team members were also aware ofand commented on the amount of time remainingto complete the team’s task. In addition, theseteams were seen to have reached or developed newagreements on the ultimate direction they wouldfollow in completing their work. Overall, Gersicknotes that “changes in teams’ work tended to bedialectical” (p. 28). She observed that teams thatstarted out fast, paused in their work to evaluatewhat had been completed and addressed problemswith the work and processes; alternatively, teamsthat had started slowly experienced “exhilaratingperiods of structuring, making choices and pullingtogether” (p. 28). Groups were also seen to bemuch more open to and receptive to the idea ofusing outside influences during this period ofgroup life. Gersick sees the transition periods asthe “second chance” for the group in that it is atime to “rechart the course of its work” (p. 35).

The teams then proceeded to “phase two.” Theevents and outcomes of phase two were seen to begreatly affected by the activities and decisions ofthe transition period. Essentially, phase two is the

time period in which the team is operating underthe decisions and guidelines that were made andestablished during the transition period. Gersicknotes that a close parallel to the occurrences inthis phase are those described by Tuckman’s(1965) stage of performing.

Finally, the teams complete their lives. For themost part, deadlines arrived in which the team’swork had to be completed. Gersick describes threepatterns that characterized the final meetings inthis period of completion. The first patternobserved was that group task activity was lessinvolved with generating new ideas and materials,and more concerned with honing and editingexisting materials and ideas. The second patternwas that more attention was placed on the expec-tations and requirements of the outsiders whowould examine and review the group’s work. Thefinal pattern was that groups expressed more posi-tive or negative feelings about their work and otherteam members.

Dynamic Contingency Model

McCollom (1990) presents this model as analternative to the other models that have been pre-sented. He feels that a model of group developmentmust be “dynamic,” because groups develop overtime, and “contingent,” because the path to groupdevelopment is contingent upon complex interac-tions with a number of variables (p. 150).McCollom indicates that many of his ideas comefrom “open systems theory,” as discussed by Millerand Rice (1975), and from models of individualdevelopment.

There are three general factors that McCollomfeels can affect group development. The first factoris the group’s relationship with its environment.The second deals with internal group relations andthe third examines the group’s temporal bounda-ries.

The group’s environment is particularly impor-tant in light of the group’s relationship with thelarger institution or organization in which it isembedded. McCollom cites work that has shownthat the larger organization can affect group proc-ess in terms of authority relations and in thedefining of the group’s task.

Internal relations are also seen to potentiallyaffect group development in a number of ways.Some of these ways include the composition andsize of the group and the skills and personalities ofthe group members. McCollom also cites Bennisand Shepard’s (1956) work in support of the ideathat leadership dynamics and subgroup relations

Smith


within the group are potential forces that couldaffect group development.

Time is the final factor specifically discussed byMcCollom that could affect group development.McCollom notes that groups are like other livingsystems and therefore their lives are finite. Anotherimportant time-related factor is what McCollomrefers to as “temporal context. "Temporal contextrefers to things such as seasonal cycles or histori-cal events, presumably past, present, and future,that could affect how a group operates, develops,and perhaps even changes over time.

From these factors McCollom suggests thatgroups that develop in similar environments couldpossibly show great similarity in their developmentprocesses. He also suggests that even “if we cannotsay how groups develop...we should at least beable to identify general categories of factors thatwill shape development (analogous to the biolog-ical, psychological and social factors at theindividual level)” (McCollom, 1990, p. 151).

TEAM Model

Morgan, Salas and Glickman (1993) have pro-posed the TEAM model based on a review of theexisting literature and their work and researchwith Navy teams. The model is an amalgamation ofseveral prior authors’ work and their own workwith teams as can be seen in Figure 2. Despite theneat array of the nine stages in Figure 2, groupsare not expected to proceed through them in alinear fashion. Rather, it is expected that groupswill proceed through, and even start at, differentpoints in the model based on the past history andexperience of the group and its members, thenature of the task, and the demands andconstraints placed on the group by its environment(Morgan et al., 1993). Thus, it is not likely that allgroups will proceed to maturity following the samepath.

Another interesting idea proposed by thismodel is that groups develop along two separatepaths. The first path at the top of the chart indi-cates that groups must learn to understand thetask that they are being asked to perform. Thispath is specifically concerned with group or teammembers learning the skills, obtaining the knowl-edge, and possessing the ability to complete thetask. For example, it may involve learning how touse a specific type of machinery or to follow a pre-determined order of assembly.

The second path is concerned with the specificinteractions that occur among team members.These are labeled “teamwork skills” because theydeal with issues relating to the interactions among

team members and relationships that exist amongthe members. Morgan et al. felt that in some waysthis is where much of the effort is spent in “teambuilding,” as it is assumed that the peopleassigned to teams have already acquired the com-petency to complete the activities required to dealwith the task (Morgan et al., 1992). Morgan et al.’smodel also indicates two other interestingpossibilities for groups as they develop. The firsthas to do with the group’s ability to recyclethrough various phases of its life. During themiddle to later period of a group’s life, the groupand its members can return to issues that hadpreviously been resolved and correct for errors inthe group’s original solution or make adjustmentsbased on new knowledge. The other interestingpoint related to this model is that, as indicated bythe two sets of circles, as the group membersbecome more skilled in both task-related andteam-related skills, the two sets of skills merge.The more the two sets of circles overlap, the moreskilled and able is the group to effectively deal withand complete its task.

The main reason that this model is included inthis section is that the model was primarily derivedfrom Tuckman’s (1965) linear progressive modelideas and Gersick’s (1988) punctuated equilibriumideas. In addition, some of the ideas from Balesand Strodtbeck’s (1951) work, specifically dealingwith the issue of group development being taskrelated and socio-emotional, can also be seen inthis model. Thus, this model is a hybrid as itrepresents one of the few attempts to combineexisting theories and ideas into a new model ofgroup development.

GENERAL OBSERVATIONS

Several insights regarding this body of litera-ture emerged during the course of this review. Oneof the primary insights is that many of the modelsexhibit strong similarities on a number of vari-ables. Coupled with this insight is a rudimentaryunderstanding of some of the potential reasons asto why there may be differences and discrepanciesbetween the models. This section will include adiscussion of some of the factors and reasons formodel similarity and dissimilarity. This discussionis not intended to serve as an exhaustive listing ofthe possible explanations for the apparent simi-larities and dissimilarities between the reviewedmodels. Instead, this portion of the paper shouldbe viewed as an attempt to understand some of thereasons for the noted and proposed differences andsimilarities in models derived from the informationcontained within the various journal articles andreadings reviewed for this paper.

Smith


Figure 2: Graphical Presentation of Morgan et al.’s TEAM Model

This section will conclude by addressing tworecurrent, problematic issues: model generalizabil-ity and semantics. While it could be argued thatsuch a presentation is not necessary, it is includedhere to stress the importance of noting and defin-ing the limitations of this work and its application,and to encourage communication so that researchmay be conducted more as a “community” with acommon interest. The final reason for includingthis section is that the two issues still requireattention, despite a relatively long period of study.

Plausible Explanations for Similarities among theDevelopmental Models

As previously stated, many of the models,regardless of their classification scheme, exhibitsimilarities in terms of their form, patterns of pro-gression, terminology, and even the nature of thephases or stages that are posited by the theorists.There are several plausible explanations that haveemerged from this review that could be used toaccount for these similarities. Some of these expla-nations will be examined in the following para-graphs.

One potential explanation involves the locationin which the model was conceived and incubated.This explanation has multiple implications for theapparent similarities. At one level, several of themodels were developed at the same academicinstitution (Harvard). Therefore, many of the theo-rists were contemporaries who were aware of eachothers work and built upon that work or injectedportions of that work into their own models. Inaddition, as contemporaries, they were exposed tomany of the same class lectures, informal discus-sions, and laboratory sessions and shared thosecommon experiences and backgrounds.

Student-professor mentoring may also haveplayed some role. Ultimately, the mentoring profes-sor would have the ability to interject ideas into thedevelopmental process as that individual assistedthe student in not only developing and resolvingproblematic issues in the study, but interpretingthe resulting data.

A second explanation revolves around theoristsadopting certain schools of theoretical thought.While Bion’s work does not appear in this reviewfor reasons that were previously explained, many

Pre-forming

First Meeting

Forming

Phase I

Storming Norming Performing

Transition

Reforming

Phase II Completion

Performing II Conforming De-forming

Developmentof Task

Assignments

Taskwork Skills

Teamwork Skills

Investigation

of Group

Orientationto Task

Testing ofDependence

Review of

Accomplishments

Withdrawal

from Task

Exiting from Group

Remembering Group

Adjustmentto Environmental Demands

Fulfillment of Roles

Refinementof Roles

Developmentof Role

Relatedness

Completionand Delivery

of Task

Drive toCompletion

Adjustment of Framework

Emergenceof Solutions

Emotional Response toTaskDemands

Open Exchange of

Relevant Interpretations

Intragroup Conflict

Developmentof GroupCohesion

Recycle

*

**

* Adopted from Gersick (1988) **Adopted from Tuckman (1965) (From Morgan et al. , 1993 p. 281)



theorists in the field adopted his ideas and createdresearch instruments that were strongly tied tothose ideas. In a similar fashion, it could also beargued that many others have tied their work tothe review conducted by Tuckman (1965). Thus, byadopting many of Tuckman’s ideas, these theoristshave adopted and grounded their theories in theideas of those who preceded him, becauseTuckman’s model relies heavily on literature priorto 1965.

The third explanation arises from the types ofgroups that have been observed and studied. Table17 provides some information relevant to thisissue. Specifically, Table 17 indicates that many ofthe groups that have been studied were similar innature on a number of variables. For example,many groups existed in highly controlled environ-ments that ranged from treatment for addictionand dependency to more serious psychologicalproblems. It could also be argued that to someextent even the research laboratories that wereused in these studies were highly controlled. Inaddition to gathering data in these controlled set-tings, many other studies were developed arounddata obtained from observing undergraduate andgraduate students in group settings or reviewingstudent journal entries that were written during asemester long course.

One final explanation for the observed similari-ties is that the theorists were responding to and inharmony with the popular organizational and theo-retical trends and interests of the time period. Per-haps the most poignant example of this isTuckman and Jensen (1977) who note that theyadded the final stage of “adjourning” largely inresponse to the work of the “life-cycle” theorists ofthe period. Thus, although these authors arguethat a specific adjourning stage was not necessarysince it existed in the performing stage, their workultimately can be seen as being grounded in theresearch and theoretical interests of the time (perTuckman and Jensen 1977). In a similar fashion,many of the more recent models that have beendeveloped reflect the idea that groups are opensystems that must deal with and respond to multi-ple environments, whereas much of the previouswork modeled groups as closed systems.

Why are the Models Different?

Despite the many similarities among thesemodels, there also differences. As with the simi-larities, some educated speculation as to the rea-sons for these differences can be conducted. Often,the differences can be explained by noting that the

models are not similar based on one of the expla-nations for similarity discussed in the previoussection.

One explanation for reported differences can beaddressed by examining Table 17 again and look-ing at the types of groups that were studied. Thetable shows that while many of the groups were ofthe same type, they were still different in terms ofpurpose, environment, and even the meaningful-ness of the task to the group. For instance, studentgroups could be seen as very different fromLacoursiere’s treatment groups or Heinen andJacobsen’s work groups. Furthermore, Braaten’sstudy introduces the possibility that differentcultures (nationalities) may give rise to differenttrends in group development. Thus, while thisparticular model has strong ties to Tuckman’s andother researchers’ work, its differences may beattributable to the fact that the people observed indeveloping the model have different values, beliefsand cultural heritages.

A second explanation for the differences in themodels is that, while many of the researchers inthis field of research have psychology back-grounds, taken as a cohort, the researchers per-forming this work represent a very broad anddiverse set of interests. For example, in the litera-ture reviewed here, there are theorists with psy-chological, sociological, social work, communica-tion, medical and educational backgrounds. Theseresearchers, as expected, work in a divergent groupof institutions and settings that range from aca-demic institutions to VA hospitals to consultationpractices. Given the breadth of interests and back-grounds represented by this body of researchers,the differences in model content, presentation, andform should really come as no surprise.

Related to the issue of divergent interests andbackgrounds is the explanation that the researchmethods and instruments used to study this phe-nomena are not always the same. While the pre-ferred method for performing research has beenobservation, there are researchers who have usedquestionnaires and surveys as well as researcherswho have reviewed group documents, interviewedindividual group members, and even coded groupmember journals. In addition, while most of thesemethods have been used in isolation from oneanother, a few researchers have combined some ofthese methods to better understand, and attemptto gain greater insight into, the developmentalprocess.

Smith


Table 17: Sources of Model/Theory Development

Theorist(s)StudentGroups

Therapy &Treatment

GroupsLiteratureReviews

PersonalExperience

TheoryBlending &Combining

NaturallyOccurring

Work Groups

Bales & Strodtbeck (1951) X

Theodorsen (1953) X

Bennis & Shepard (1956) X

Schutz (1958) X

Mills (1964) X

Tuckman (1965) X

Napier & Gershenfeld (1973) X

Hare (1973) X

Sarri & Galinsky (1974) X X

Braaten (1974/1975) X X

Heinen & Jacobson (1976) X

Tuckman & Jensen (1977) X

Srivastva, Obert & Neilsen(1977)

X X

Bradford (1978) X

Caple (1978) X

Lacoursiere (1980) X

Poole (1985) X

Kormanski & Mozenter(1987)

X

Gersick (1988) X X

Maples (1988) X

McCollom (1990) X X

McGrath (1990) X

Drexler, Sibbet & Forrester(1991)

X

Morgan, Salas & Glickman(1993)

X

The main point being argued here is that thevarious methods and instruments could be attrib-uted to producing differences in the existing mod-els simply because of their sensitivity to the phe-nomena being studied. Furthermore, individualresearcher characteristics, such as differences intraining and backgrounds, may also have played arole in the interpretation of the data that could, in

turn, have allowed for differences in the existingmodels.

The prior sections raise concerns in two areas.The first area addresses the generalizability of theexisting models, while the second addresses theproblem of the semantics used in the group devel-opment models.



Model Generalizability

One of the primary purposes of this review wasto ascertain what types of groups had been studiedand what were the settings in which these groupsexisted and developed. After reviewing the litera-ture, it became apparent that the majority of thegroups that have been studied are groups of stu-dents (graduate and undergraduate), patients invarious types of hospitals (VA, mental institutions)and individuals involved in treatment or self-helpgroups (see Table 17).

While some researchers, such as Bettenhausen(1991), have argued that the types of groups usedin these studies “share many features with thetemporary, ad hoc task or decision-making groups”(p. 351) of today, others, such as Tuckman (1965),have maintained the argument that the existing“literature cannot be considered truly representa-tive of small-group developmental processes, sincecertain settings have been overrepresented” (p.395). This argument, while important, has fallenlargely on deaf ears over the last twenty-five years(see McCollom, 1990, for example). In addition tothis issue, there are yet other questions and issuesthat affect our current ability to make generaliza-tions across groups and settings.

Among the questions are issues concerning theeffect of group size on development and whether agroup is a “real” group or a simulated group (labo-ratory). These issues could affect the groupbecause the task may not be “high on motivatingpotential,” as discussed by Hackman (1990), or theincreased communication needed to function as agroup may impede the group's ability to clearlydefine its task and clearly understand thedemands of either laboratory observers or higherlevel organizational actors.

Cissna (1984), in a paper examining the evi-dence that refutes the existence of group develop-ment stages, has made the following statementabout this topic:

The convenient names — training, labora-tory, and natural groups — obscure signifi-cant and potentially important differencesamong different groups within each of thesecategories. For example, training groupsinclude 9-month-long “self-analytic” class-room groups of 25 members as well as 2-week, 1-week, and even weekend T-groupsof far fewer members, with different goals,degree of trainer directiveness, and so on(p. 9).

Cissna’s statement points to the fact that addi-tional research needs to be conducted. Among the

issues that should be examined are differences ingroup tasks (single or multiple tasks? simple orcomplex tasks?), group composition (How aremembers assigned? Does membership remain sta-ble during the group’s life?) and organizational andgroup settings. Another emerging area of studythat is in need of further attention addressesdevelopmental issues relating to the history of notonly the organization in which the group is embed-ded, but also the experiences and history of thegroup members themselves. McGrath (1991) notesthat a flaw in previous research has been that thegroups that have been used exist in a highly con-trolled environment, isolated from any otherorganization and exist for a limited period of time.Thus, the group and its members really have nopast to reflect upon or future to look forward to.

Semantics and Terminology

The second general problem is the issue ofsemantics and terminology. There is a real need forsome kind of agreement on terminology. Thisagreement is felt to be necessary to promote con-certed study and research in the area of groupdevelopment.

There is yet a need to know and define whatexactly a phase is, a stage is, and a trend is. Inaddition, some of the terminology used to describethese phases, trends or stages could be simplifiedand made more uniform since there is apparentgeneral agreement that groups tend to develop insimilar ways. This general developmental tendencyis supported by the fact that the models can begrouped and described as a cohort based on thecommonalities displayed in each model’s set ofstages. (For example, the linear-progressive modelstend to discuss formation, conflict, resolution, pro-duction, and termination or separation.)

Sarri and Galinsky (1974) have also noted thisproblem. In particular, they have stated that thecurrent state of the literature makes it difficult tomake comparisons between the various studiesthat have been published. Of specific concern tothese researchers is the need to develop a set ofcategories and concepts for use in developing andexplaining these models. If a standard vernacularfor labeling and describing these stages could beachieved, then more work like Maples (1988) couldbe performed to expose and expand the behaviorsand processes that operate and exist within a givenstage of development.

Any immediate resolution to this problem ismore likely wishful thinking and may prove to beproblematic. Part of the problem underlying thecurrent state of group developmental terminology

Smith


stems from the many different disciplines andinterested parties developing the models and con-ducting research. For example, as previouslynoted, theories and models of development havecome from the fields of psychology, sociology, andeducation and from psychiatrists, trainers, educa-tors, and combinations of the aforementionedgroups serving in consulting capacities. Theseindividuals also hail from diverse parts of theglobe. The diversity of researchers’ backgroundscould prove to be problematic in that seeminglysimilar ideas and terminology may indeed be sig-nificantly different. Kraiger (1995) has commentedon this general problem in terms of I/O psycholo-gists importing constructs from outside the disci-pline. His position is that great care needs to betaken when using these constructs and in properlydefining them. Another possible reason for the dif-ferences in terminology is that although the grouptypes studied are not as diverse in nature as wouldbe preferred, they are diverse enough to requireresearchers to learn new cultures and their relatedlanguages and symbols to understand them.

SUGGESTED FUTURE RESEARCH DIRECTIONS

While the prior section was concerned with lin-gering issues in the field, this section focuses onissues that have the potential to advance ourunderstanding of the group development phenom-ena and strengthen the research that is presentlybeing conducted. Specifically, this section willexamine and discuss three areas in which addi-tional work needs to be conducted to enhance ourunderstanding of the small group developmentalprocesses. The first two issues are concerned withincreasing our understanding of the affect thatprior history or experience with groups has onfuture group development and the multiple effectsthat “time” has, not only on the group, but also onthe researcher’s ability to accurately portray andgain access to data related to the developmentalphenomena. The final issue is a methodologicalissue that focuses on the virtues of conductinggroup development research using multi-methodapproaches.

Affect of Member Histories and Experiences onGroup Development Process

One particularly interesting area for futureresearch addresses the effect that “mental models”of group experience and other cognitive models ofgroups and teams (derived from past experiences)have on patterns of group development. Hinsz(1995) recently submitted the proposition thatmental models are important to group developmentbecause “our beliefs and expectations about sys-

tems and our interactions with those systems pro-foundly influence our actions with regard to thesystems” (p. 201). Because groups can be seen associal systems, it is important that we have someunderstanding of individual group members’ sub-jective perceptions of group reality. In other words,many of the models that have been developed todate appear to gloss over members’ prior groupexperiences and assume that all members entergroups with similar backgrounds and ideas per-taining to the group. What Hinsz is suggesting isvery different in that he has adopted a stance thatmaintains that all members have different experi-ences with groups and those experiences mayaffect the developmental processes and patterns ofany given group.

Hinsz also notes that mental models areacquired through learning and experience. Thislearning need not be direct but can be indirectand, perhaps, obtained through observation, hear-say, or contact with individuals currently involvedin groups. Furthermore, he states that initialmental models will be unstable for individuals withlimited interaction with groups, but, over time,these models will develop and stabilize. If thesestatements are true, then prior experience withgroups may indeed be a factor that is deserving ofmore attention, because prior experiences mayprevent the emergence of traditionally recognizedpatterns of development and may also be used toexplain why so many different models currentlyexist in the literature.

Rentsch, Heffner and Duffy (1994) have com-pleted some preliminary research on this topic andtheir findings appear to indicate that an individ-ual’s prior “teamwork” experience affects theapproach an individual takes to working in agroup. More specifically, the study indicates thatthose with prior team experience possess cognitiveframeworks of viewing teamwork and team devel-opment that are different than individuals withoutsuch experiences (Rentsch et al., 1994). This studyfurther reinforces the need to examine how priorexperiences, “teamwork schemas” and “mentalmodels” affect group development and the phasesor paths groups follow in reaching maturity. Inaddition, more attention should be paid to theeffect of culture, at the organizational, national andindividual levels, that is being brought to groupsand teams by the members. This issue is impor-tant since Gersick and Hackman (1991) have notedthat a possible source of habitual routines ingroups comes from patterns of behavior that areimported from outside the group.



Time: A Multifaceted Variable in the Study of GroupDevelopment.

Time is an interesting and problematic issue. Itis interesting because we have much to learnregarding its affect on the developmental process.For example, Gersick’s (1988) work provides someindication that groups are aware of the time limitsthat bind them and develop and react to themthrough periods of high activity and periods of lowactivity. This point is problematic because it sug-gests that groups may not be as rational as manyof the models have depicted. McGrath, for example,notes that, in his model, groups will always striveto proceed from “Mode I” to “Mode IV” (see Figure1) in the quickest manner and only deviate fromthis pattern when special issues or circumstancesarise. Gersick, on the other hand, essentially statesthat groups are not that rational; rather, they waitand react based on the time remaining to producesomething.

Kuypers, Davies and Glaser (1986) raiseanother time-related issue in terms of “develop-mental arrestations.” At the crux of this issue isthe seemingly accepted notion that all the groupsused in developmental models have reached apoint of maturity. What Kuypers, et al. argue isthat perhaps one of the reasons that so many dif-ferent models exist is that the groups that havebeen observed have become “arrested” in theirdevelopment and researchers have only assumedthat this lack of change signaled an end stage ingroup development. Essentially, the argument isthat further observation could possibly lead toadditional changes and present greater similarityamong the existing models. This argument mayhave some validity since one of the problems ofperforming this type of research is to be on site atthe specific moment a group is formed and to be inboth the formal and informal meetings of the groupas noted by McGrath (1991).

The prior time issue is one that researchersmust come to grips with and is, perhaps, more ofan issue for researchers studying “real,” naturallyoccurring groups. In addition to this, all research-ers’ efforts are to some extent affected by time andfinancial limitations and constraints. This is animportant issue to keep in mind, as one ofMcGrath’s (1986) list of “critical needs” for smallgroup theory and practice is to conduct researchthat is multi-occasion. The limitation placed on theresearcher and his or her timeliness in studyinggroup development could by itself possibly con-strain his or her ability to observe a group beyondits period of arrestation.

Coupled with this is the fact that, in real set-tings, groups cost money to the organization inwhich they are embedded and, therefore, come un-der pressure to perform. Obert (1983) builds onthis point by noting that it is likely that if “groupshave natural developmental tendencies, thenorganizations have structures and processes whichcontrol, channel and even block these tendencies”(p. 39). Some of the potential means organizationsmay have to control these tendencies include jobtechnology, the level of the group in the hierarchy,opportunity for interaction among the group mem-bers, the nature of the reward system, the neces-sity for cooperative work relations, and the style ofleadership used by management (Obert, 1983).Deadlines and time limits themselves may also bepotential means of controlling and channelinggroup activity and development. As a result, it isconceivable that the studies and models that havebeen reported to date are nothing more than con-trivances based upon organizational structuringand other related control mechanisms.

Use of Multi-method Research in Developing andTesting Developmental Models.

Many models have been developed and manymore will probably be developed in the comingyears. While it is conceded that the study of a phe-nomena must be undertaken using the methodappropriate to the data that are needed to answerthe research question(s), it was observed, duringthe course of this review, that many differentresearch methods and strategies have already beenapplied in singular fashion to the study of thisphenomena. For example, Maples (1988) has usedgroup member journals to augment and extendTuckman’s model; Gersick (1988) has used inter-views to assist her in interpreting her observations;and Drexler, et al. (1991), Morgan, et al. (1993) andWheelan and Hochberger (1996) have used ques-tionnaires to gather data pertaining to group devel-opment. In addition, it has also been suggestedthat insights could be gained from using docu-ments generated by the group being studied. Theseevidences of prior use of various research methodsraises the question of why researchers have notused more of the methods together in conductingtheir work.

While anecdotal and experiential data have pro-vided us with ideas and models, more rigorousprocedures need to be employed to advance thefield. One possibility is to combine observationaltechniques with data collected from group memberself-reporting techniques (questionnaires, surveys,interviews). The self-report data may better enableresearchers to interpret their observational data.

Smith


Furthermore, these data may explain events andoccurrences outside the observed group settingthat have shaped the group’s observed develop-mental pattern. A second data source largelyneglected to date is the use of archival analysis.One benefit to using these data is that they maycontain information that indicates how organiza-tional policies and procedures may have influencedthe group’s development.

At present, it is difficult to say what affect sucha research agenda would have on model develop-ment and refinement because there are few studiesthat have attempted to approach the phenomenain this way. However, if Gersick’s work is an indi-cation of the benefits that can be gained from suchan approach, there appears to be room for im-provement.

… there are a preponderance of

organizations currently employing

teams and promoting teamwork and

yet our understanding is largely

limited to models that have never

been empirically validated or studied

in dynamic, uncontrollable settings.

Overall, what is being suggested here is the useof multiple methods to augment the findings ofsingular, solitary methods such as observation. Byproceeding in such a manner, future researchersshould be able to uncover deeper and hiddenmeanings that underlie group and individualbehaviors. In addition, this approach can be usedin either a manner in which the various modelsbuild upon one another or in such a way that thedata gathered provide deeper meaning and insightinto the developmental process.

In general, it is hoped that, by pursuing aresearch agenda that uses multiple methods, therewould be some refining, combining or discarding ofthe current set of developmental models. Specifi-cally, the expectation is that the current models berefined or rejected. In particular, model refinementcould include improvements in the various mod-els’ability to deal with differences in group task,composition, membership stability, prior groupexperience, and the effect of time and deadlines ongroups. Furthermore, the additional research andstudy may help to better define the limitations of

previous studies and models and expose newopportunities for future research directions.

With regard to the overall use of multi-method,data-collection techniques in this body of litera-ture, it is interesting to note that of the articles andstudies reviewed here, only Gersick’s study hasemployed the use of multiple methods. In this par-ticular case, she combined observation with inter-views to attempt to better understand the eventsthat took place in group meetings and to obtaininformation about events and activities thatoccurred outside the formal group meeting setting.

CONCLUSION

On a positive note, the models that werereviewed are becoming more comprehensive andsystems oriented. Researchers such as Poole(1985), McGrath (1986, 1991), and McCollom(1990) are beginning to recognize groups as opensystems rather than closed systems. Theseresearchers, and particularly those promoting con-tingency models of group development, are point-ing out that groups do indeed find ways of dealingwith not only internal group problems, but withthe problem of obtaining resources from outsidethe group, thus, dealing with the environment notonly within the organization in which they are em-bedded, but also the environment outside of thatorganization’s boundaries. They are also beginningto examine how time and history affect groupdevelopment. In the future, it is hoped that thistrend will continue.

In retrospect, the cyclical and pendular modelsare more flexible, adaptable and capable of dealingwith a dynamic world than were the linear-pro-gressive models, while many of the recent nonse-quential/hybrid models are advancements overboth these sets of models in that they emphasize,to an even greater extent, the fact that groups areopen systems which are embedded in largerorganizations and environments. In terms of futuremodel development, it would appear that the pathto pursue is one which uses the frameworks andmodels established by the linear-progressive mod-els as guides, but focuses more energy and efforttoward understanding the cyclical patterns andideas proposed by the recurring cyclical and pen-dular models and the open systems ideas put forthby the nonsequential/hybrid models.

In conclusion, this paper suggests that, whilemuch has been written in the area of group devel-opment, relatively little advancement has beenmade in the past twenty to twenty-five years. Thisis disturbing because there are a preponderance oforganizations currently employing teams and pro-



moting teamwork and yet our understanding islargely limited to models that have never beenempirically validated or studied in dynamic,uncontrollable settings. Furthermore, as the priorsection of observations has indicated, there is roomfor additional research on the effect of time anddeadlines; organizational, group, and individualhistory and experience; culture; the nature of thegroup task; the type of group; and the group’scontextual arrangements.

While it is unlikely that a single, universalmodel of group development will ever exist, it ispossible for us to do a better job of testing our pre-sent models and examining how the factors out-lined in the preceding paragraph affect groupdevelopment. When we, as a research community,have conducted more research in these directions,perhaps we will be better able to not only under-stand the process of group development, but toassist individuals, groups and organizations inimplementing and using groups and teams.

ReferencesBales, R. F., & Strodtbeck, F. L. (1951). Phases in

group problem solving. Journal of AbnormalSocial Psychology, 46, 485-495.

Bennis, W. G., & Shepard, H. A. (1956). A theory ofgroup development. Human Relations, 9, 415-437.

Berkowitz, B. (1974). Stages of group developmentin a mental health team. Psychiatric Quarterly,48(3), 309-319.

Bettenhausen, K. L. (1991). Five years of groupsresearch: What we have learned and whatneeds to be addressed. Journal of Management,17(2), 345-381.

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

Braaten, L. J. (1974/75). Developmental phases ofencounter groups and related intensive groups:A critical review of models and a new proposal.Interpersonal Development, 5, 112-129.

Bradford, L. P, Gibb, J. R., & Benne, K. (1964).T group theory and laboratory method. NewYork: John Wiley & Sons.

Bradford, L. P. (1978). Group formation and devel-opment. In L. P. Bradford (Ed.) Group develop-ment (pp. 5-12). San Diego: UniversityAssociates.

Caple, R. B. (1978). The sequential stages of groupdevelopment. Small Group Behavior, 9(4), 470-476.

Chidambaram, L. (1989). An empirical investigationof the impact of computer support on groupdevelopment and decision making performance.Unpublished doctoral dissertation, IndianaUniversity, Bloomington.

Cissna, K. N. (1984). Phases in group development:The negative evidence. Small Group Behavior,15(1), 3-32.

Drexler, A. B., Sibbet, D., & Forrester, R. H.(1991). The team performance model. In W. B.Reddy (Ed.) Team building: Blueprints for pro-ductivity and satisfaction (pp. 45-61).Alexandria, VA: NTL Institute for AppliedBehavioral Science.

Eldredge, N., & Gould, S. (1972). Punctuatedequilibria: An alternative to phyletic gradual-ism. In T. J. Schopf (Ed.) Models in paleobiology(pp. 82-115). San Francisco: Freeman, Cooper& Co.

Gersick, C. J. G. (1988). Time and transitions inwork teams: Toward a new model of groupdevelopment. Academy of Management Journal,31, 9-41.

Gersick, C. J. G. (1989). Marking time: Predictabletransitions in task groups. Academy ofManagement Journal, 32(2), 274-309.

Gersick, C. J. G. (1991). Revolutionary changetheories: A multilevel exploration of the punc-tuated equilibrium paradigm. Academy ofManagement Review, 16(1), 10-36.

Gersick, C. J. G., & Hackman, J. R. (1991). Habit-ual routines in task-performing groups.Organizational Behavior and Human DecisionProcesses, 42, 65-97.

Gibbard, G. S., Hartman, J. J., & Mann, R. D.(1974). Analysis of groups. San Francisco:Jossey-Bass Publishers.

Habermas, J. (1970). Toward a theory of communi-cative competence. In H. P. Dreitzel (Ed.),Recent sociology no. 2: Patterns of communica-tive behavior (pp. 115-148). New York:MacMillan Co.

Habermas, J. (1979). What is universal pragmat-ics? In Communication and the evolution of soci-ety (pp. in process). (T. McCarthy, Trans.).Boston: Beacon Press.

Habermas, J. (1982). A reply to my critics. In J. B.Thompson & D. Held (Eds.), Habermas: Criticaldebates (pp. 219-283). (T. McCarthy, Trans.).Cambridge: The MIT Press.

Hackman, J. R. (1990). Groups that work (andthose that don’t). San Francisco: Jossey-BassPublishers.

Hare, A. P. (1973). Theories of group developmentand categories for interaction analysis. SmallGroup Behavior, 4(3), 259-304.



Heinen, J. S., & Jacobson, E. (1976). A model oftask group development in complex organiza-tions and a strategy of implementation.Academy of Management Review, (October), 98-111.

Hill, W. F., & Gruner, L. (1973). A study of devel-opment in open and closed groups. SmallGroup Behavior, 4(3), 355-381.

Hinsz, V. B. (1995). Mental models of groups associal systems. Small Group Research, 26(2),200-233.

Kraiger, K. (1995). Paradigms lost: Applicationsand misapplications of cognitive science to thestudy of training, Presentation at the 103rdAnnual Convention of the American Psycho-logical Association, August 13, 1995, New York.

Kuypers, B. C., Davies, D., & Glaser, K. H. (1986).Developmental arrestations in self-analyticgroups. Small Group Behavior, 17(3), 269-302.

Kormanski, C., & Mozenter, A. (1987). A newmodel of team building: A technology for todayand tomorrow. The 1987 Annual: DevelopingHuman Resources, 255-268.

Komorita, S. S., & Hamilton, T. P. (1984). Powerand equity in coalition bargaining. In S. B.Bacharach & E. J. Lawler (Eds.), Research inthe sociology of organizations: Vol. 3. The socialpsychological processes (pp. 189-212).Greenwich, CT: JAI Press Inc.

Lacoursiere, R. B. (1980). The life cycle of groups.New York: Human Sciences Press.

Maples, M. F. (1988). Group development:Extending Tuckman’s theory. Journal forSpecialists in Group Work, 13(Mr), 17-23.

McCollom, M. (1990). Reevaluating group develop-ment: A critique of the familiar models. In J. G.& M. McCollom (Eds.), Groups in context,(pp.134-154). New York: Adison-WesleyPublishing.

McGrath, J. E. (1986). Studying groups at work:Ten critical needs for theory and practice. In P.S. Goodman & Associates (Eds.), Designingeffective workgroups, (pp.362-391). SanFrancisco: Jossey-Bass.

McGrath, J. E. (1991). Time, interaction and per-formance (TIP) a theory of groups. Small GroupResearch, 22(2), 147-174.

Mennecke, B. E., Hoffer, J. A., & Wynee, B. E.(1992). The implications of group developmentand history for group support system theoryand practice. Small Group Research, 23(4), 524-572.

Miller, E., & Rice, A. K. (1975). Selections from‘systems of organization.’ In A. D. Colman andW. H. Bexton (Eds.), Group relations reader,Sausalito, CA: GREX.

Mills, T. M. (1964). Group transformation: Ananalysis of a learning group. Englewood Cliffs,N.J.: Prentice-Hall.

Morgan, B. B., Salas, E., & Glickman, A. S. (1993).An analysis of team evolution and maturation.The Journal of General Psychology, 120(3), 277-291.

Nahavandi, A., & Aranda, E. (1994). Restructuringteams for the re-engineered organization. TheAcademy of Management Executive, 3(4), 58-68.

Napier, R. W., & Gershenfeld, M. K. (1973). Theevolution of working groups: Understandingand prediction. In R. W. Napier & M. K.Gershenfeld, Groups theory and experience (pp.243-267). Boston: Houghton Mifflin Co.

Obert, S. L. (1983). Developmental patterns oforganizational task groups: A preliminarystudy. Human Relations, 36(1), 37-52.

Poole, M. S., Seibold, D. R., & McPhee, R. D.(1985). Group decision-making as a structura-tional process. Quarterly Journal of Speech, 71,74-102.

Poole, M. S., & Roth, J. (1989). Decision develop-ment in small groups IV. HumanCommunications Research, 15(3), 323-356.

Rentsch, J. R., Heffner, T., & Duffy, L. T. (1994).What you know is what you get from experi-ence. Group & Organization Management, 19(4),450-474.

Ridgeway, C. L. (1983). The dynamics of smallgroups. New York: St. Martin’s Press.

Sarri, R. C., & Galinsky, M. J. (1974). A conceptualframework for group development. In P.Glasser, R. Sarri, & R. Vinter, (Eds.), Individualchange through small groups (pp. 71-88). NewYork: Free Press.

Schutz, W. C. (1958). FIRO: A three-dimensionaltheory of interpersonal behavior. New York:Holt Rinehart.

Shambaugh, P. W. (1978). The development of thesmall group. Human Relations, 31(3), 283-295.

Simon, H. A. (1976). Administrative behavior. NewYork: Free Press.

Srivastva, S., & Barrett, F. J. (1988). The trans-forming nature of metaphors in group devel-opment: A study in group theory. HumanRelations, 41(1), 31-64.

Srivastva, S., Obert, S. L., & Neilsen, E. H. (1977).Organizational analysis through group proc-esses: A theoretical perspective for organizationdevelopment. In C. Cooper (Ed.), Organizationaldevelopment in the UK and USA: A joint evalua-tion (pp. 83-111). London: McMillan Press.

Tuckman, B. W. (1965). Developmental sequencein small groups. Psychological Bulletin, 63, 384-399.

Smith


Tuckman, B. W., & Jensen, M. C. (1977). Stages ofsmall-group development revisited. Group andOrganization Studies, 2(4), 419-427.

Wheelan, S. A., & Hochberger, J. M. (1996). Vali-dation studies of the group development ques-tionnaire. Small Group Research, 27(1), 143-170.

Young, A. (1976a). The geometry of meaning. NewYork: Delacourte Press.

Young, A. (1976b). The reflexive universe. NewYork: Delacourte Press.

Acknowledgements

The author would like to thank Dr. Sue R. Faerman and Dr. John W. Rohrbaugh for their time,encouragement, and contributions to this work.

George Smith is a doctoral student at the University at Albany’s (SUNY) Rockefeller College ofPublic Affairs. His current research interests focus on understanding the implications of individualmental models of group development for group/team development. Contact at: Rockefeller Collegeof Public Affairs, University at Albany, State University of New York, 493 Kenwood Avenue, DelmarNY 12054; [email protected]

Associate Editor: James Spee


A Critical View of Facilitating Labor-Management Collaboration

Grant T. Savage and Chadwick B. Hilton

Labor-management group facilitation is a complex but increasingly necessary skill. Facilitatorsneed both clear practice guidelines and an understanding of why those guidelines are legitimate.To meet these needs, this paper first provides a descriptive (structural-functional) framework forunderstanding the facilitator’s role and the communicative practices on which it is based. Acritique of this framework is then proposed using Habermas’ theory of communicative action.From this theoretical critique, group decision making is viewed as both a negotiative and adialogical process, entailing an expanded appreciation of the facilitator’s role. In congruence withthis theoretical stance, a set of directives for facilitating consensual decision making is proposed.A combined case and discourse analysis of two labor-management groups’ decision-makingprocesses illustrates the utility and implications of these directives.

Key words

Consensual Decision Making, Critical Theory, Facilitation

To conduct a conversation means to allow oneself to be conducted by the object to which thepartners in the conversation are directed. It requires that one does not try to out-argue theother person, but that one really considers the weight of the other’s opinion....It is not the artof arguing that is able to make a strong case out of a weak one, but the art of thinking that isable to strengthen what is said by referring to the object.

- Hans-Georg Gadamer, 1975, p. 330

Introduction

Much of the collaborative rhetoric in the man-agement field focuses on the synergy that canoccur through the building of cross-functionalteams—the creation of semiautonomous andautonomous work teams, and the participation ofemployee unions in management decisions viaownership or other power sharing arrangements.Each of these endeavors relies on some form ofintra-organizational collaboration to coordinate theactions of employees. That coordination typicallytakes the form of meetings to the extent thatorganizational meetings “represent a primarymeans of communication and coordination withinand across work units” (Niederman and Volkema,1999, p 330). Team meetings, however, arefraught with potential problems:

• An asymmetrical distribution of power (teamsconsisting of a mix of organizationally powerfuland less powerful individuals) that leads to thedomination of the powerful over the powerlesseven under the guise of seeking input and con-sensus. Consensus is the uncoerced agreementof all parties involved in a group decision-making process, whether task/project related orconflict related. Consensus does not assumeperfect accord from the outset nor is it expedi-ent agreement to end the process of evaluating

alternate decision options. Rather, consensusis the agreement that results from a group’sfree, rational, uninfluenced and thorough con-sideration of alternative propositions.

• Embedded hierarchical relations that impedethe consensus process—for example, an organ-izational hierarchy that enables threatenedmiddle managers to withhold or delay criticalinformation or a decision process within thehierarchy so convoluted that joint initiatives aredelayed or suppressed.

• False consensus—a consensus based on inten-tionally or unintentionally distorted communi-cation.

Unfortunately, as a result of these conditions,the outcome of many intra-organizational collabo-rations is a false consensus that too often resultsin alienation, distrust, and conflict. Avoiding andsurmounting this problem by engaging in a criticaldialogue is the primary focus of this paper. Cre-ating the opportunity for such a dialogue is one ofthe great challenges facing facilitators.

Both the increasing demand for team-baseddecision making and the problems inherent in theteam process have resulted in a need for groupfacilitators with highly developed skills. Asimportant as skilled facilitators are, though, there

Savage and Hilton


is relatively little research on the “formal role of theorganizational facilitator in preparing and execut-ing meetings” (Niederman and Volkema, p. 330).Particularly lacking is research that would lead touseful guidelines for working facilitators.

Guidelines formulated specifically for facilitat-ing consensual intra-organizational decision-mak-ing are hard to find. Perhaps one reason for thisdifficulty is because most of the research on con-sensual decision making (Destephen, 1983; Hill,1976; Knutson, 1972; Knutson & Holdridge, 1975;Knutson & Kowitz, 1977) has relied on self-reportmeasures of consensus gathered after a group hasreached a decision. Moreover, even dynamicmeasures of consensus (Spillman, Bezdek, &Spillman, 1979) rely on self-report instrumentsadministered while a group is making a decision.This line of research treats consensus as animportant outcome of group discussion, but it doesnot directly examine how the process of discussionaffects consensus. However, by applying adescriptive framework and a critical communica-tion theory perspective to selected cases of labor-management decision making, we derive and expli-cate a set of directives for facilitating consensualintra-organizational decision-making.

A DESCRIPTIVE FRAMEWORK FOR EXAMININGTHE PROCESS OF FACILITATING

First, we establish the definitions of someterms common to facilitation that will be usedthroughout this discussion. These terms assumethat consensual decision making often involvesconflicts over tasks, processes, and other matters.Second, we articulate a descriptive framework forintra-organizational facilitation that focuses on itsfunctions and how it is embedded within anorganizational structure, e.g., bureaucracy.

Key Terms

Facilitation. The least intrusive form of thirdparty intervention, facilitation typically involves aneutral or mutually trusted third party whofocuses upon ways to diffuse hostilities betweenthe parties and to help them become more con-ciliative in their communication. Simultaneously,the facilitator attempts to help the conflicting par-ties discuss their differences through a problem-solving process. The facilitator may help the dis-putants explore various solutions, but the twoparties in the conflict decide what, if anything, willbe done to resolve or manage the dispute.

Mediation. As a third party intervention, me-diation often includes not only process interven-tion, but also content intervention. As such,

mediation usually requires the neutral third partyto meet separately with each disputant to discusssubstantive issues and possible integrative solu-tions or compromises. As a result, the mediatorhelps the disputants to think through specific waysto resolve the conflict. Typically, the mediator alsoserves as a go-between, connecting the two or moreconflicting parties, and thus helping to dampen theemotions that might erupt during face-to-faceinteractions. Again, however, the final decisionrests with the conflicting parties.

Persuasion. In day-to-day language, persua-sion is the process of convincing others. Asapplied to group facilitation and decision making,this ordinarily means that a particular solution,predetermined by the persuader, should beadopted. This process is directional and assumesthat the persuader is success oriented from theoutset and that a variety of techniques, includingentreaty, logic, reasoning, and threat may be usedto ensure success. In contrast to ordinary lan-guage usage, we refer to persuasion, within thecontext of group facilitation, from a narrower andmore dialogical perspective. Following Johanne-sen’s (1971) logic, persuasion may be viewed as arhetoric that clarifies positions through the play ofargumentation. Furthermore, we view the facilita-tor as using persuasion primarily to influence theprocess of group decision making.

A Functional Model of Intra-OrganizationalFacilitation

A facilitator can affect the orientation of agroup (i.e., its internal process) and/or the group’srelationship to organizational sub-systems andsystems (i.e., its external process). At one time,much of the process-oriented research on decisionmaking focused directly upon a group’s internalprocess of communication (e.g., Fisher, 1970; seealso Cragan & Wright, 1980), ignoring how externalprocesses and even indirect internal processesmight affect decision making. However, recentwork in the fields of group decision making (Poole& Roth, 1989a, 1989b; Poole et. al, 1993; Poole &Holmes, 1995), mediation (Pruitt & Carnevale,1993), labor-management negotiations (Friedman& Podolny, 1992; Friedman, 1994), and conflictresolution (Wall & Blum, 1991; Wall & Callister,1995) highlights the importance of backstageinteractions, network linkages between stake-holders, and the effects of technology on the groupdecision-making process. Consequently, we seenow that understanding group decision makingand facilitation requires attention to the wholeprocess—both internal and external.



Facilitating internal processes includes twoforms of intervention: (1) “direct” interventionsduring meetings and (2) “indirect” interventionsafter and/or before meetings. Direct interventionsduring meetings, by their very nature, affect groupcommunication. Facilitators assess group interac-tion and, through their remarks, attempt either toguide or mediate the group’s decision-makingprocess. Indirect internal facilitation, on the otherhand, may take many forms, from counseling indi-viduals to mediating disputes between conflictingcoalitions within a group. Even though this inter-vention takes place outside the meeting, it may stillaffect the decision-making process of the groupsinvolved.

… from Habermas’ perspective, a facili-tator tries to create an ideal speechsituation and through the appropriateintervention strategies helps the par-ticipants to engage in a communicativedialogue that results in consensualdecision making.

External process facilitation also encompassestwo types of intervention: (1) “inward” interven-tions affecting group–stakeholder interactions and(2) “outward” interventions affecting group–organi-zation interaction. “Inward” interventions mayrange from suggesting how a group gathers infor-mation from its stakeholders to directly mediatinga group’s feedback sessions with its constituents.In other words, this form of intervention affectsboth the downward flow of communication fromand the upward flow to the group. “Outward”interventions affect both the upward and/or lateralflow of communication from the group and thedownward and/or lateral flow of communication tothe group. For example, the facilitator may act asa liaison—linking group “A” to group “B” in order tobroaden group A’s information base on a specificissue. Or the facilitator may help a group negoti-ate the implementation of a new work schedulewithin the organization, thus actively affecting thegroup’s (and the organization’s) decision making.

A THEORETICAL PERSPECTIVE FORUNDERSTANDING THE PROCESS OF FACILITATING

The descriptive framework for examiningfacilitating may be supplemented with criticalcommunication theory. Habermas’ theory of com-municative action (1979, 1982, 1984) provides auseful theoretical basis for understanding andexplaining the processes of facilitation and group

decision making. Although volumes have beenwritten on the theory of communicative action, thefollowing discussion will serve our purposes.

Using Habermas’ terms to describe the process,a facilitator, in the purest sense, is attempting tocreate an “ideal speech situation.” Habermasdescribes the ideal speech situation as one that,essentially, ensures fair play and dialogue. In anideal speech situation, the participants must:

• Possess communicative competence — be capa-ble of using and understanding the requiredlanguage.

• Understand the context of the discourse so as torelate to others appropriately.

• Be free from any form of dominance or coercionso that they may speak truthfully, freely, andparticipate fully.

If a facilitator is able to establish an idealspeech situation, then he or she further attemptsto bring about through that situation a dialogue ofall the participants that will result in whatHabermas terms “communicative action.”Habermas characterizes communicative action as“whenever the actions of the agents involved arecoordinated not through egocentric calculations ofsuccess but through acts of reaching understand-ing. In communicative action participants are notprimarily oriented to their own individual suc-cesses . . .” (Habermas, 1984, pp. 285-86). Com-municative discourse, then, is that which has asits ultimate goal mutual understanding, or consen-sus.

In contrast to communicative action, strategicaction is that in which the actions of the agentsinvolved are coordinated through “egocentric cal-culations of success” (Habermas, 1984, p.286).Strategic action has an explicit “success orienta-tion” as opposed to the “understanding orientation”of communicative action. In essence, strategic dis-course is concerned with “influencing the decisionsof a rational opponent” (Habermas, 1984, p.285)while communicative discourse is concerned withunderstanding, harmony, and reaching consensus.It is clear, then, that from Habermas’ perspective, afacilitator tries to create an ideal speech situationand through the appropriate intervention strategieshelps the participants to engage in a communica-tive dialogue that results in consensual decision-making.

A COMMUNICATIVE AND STRATEGIC ACTIONMODEL OF FACILITATION

Habermas’ theory of communicative actionsuggests a number of modifications to the descrip-

Savage and Hilton


tive framework for facilitating consensual decision-making. Although a facilitator usually acts com-municatively during interventions, some exigenciesrequire strategic action—using persuasion andmediation—in order to effect a consensus decision.For Habermas, communicative action is desirable,while strategic action, because it is self- and/orsuccess-oriented, is not. However, in a less thanperfect world, we believe facilitators sometimesshould act strategically in order to steer thegroup’s decision-making process toward true con-sensus. Importantly, during these occasions thefacilitator’s strategic actions should be orientedtoward directing the decision process, not towardinfluencing the decision outcome. Under theseconditions, strategic actions by facilitators not onlyare probable, but also desirable.

Decision Making as Negotiation

The model of facilitation so far elaborated issomewhat incomplete because, for analytical pur-poses, the social actions used to describe variousinterventions have been treated as discrete proc-esses. This atomized view is only slightly embel-lished by considering the varying purposes of andcontexts for persuasive and mediating interven-tions. To further complement the model of facili-tation, decision making must be perceived as anegotiative as well as a dialogical process. Twotypes of exigencies illustrate decision making asnegotiation.

(1) Whenever a group is attempting to reach aconsensual decision, it will necessarily employwhat it believes are communicative actions. Thefacilitator can help this dialogical process along byensuring that various exigencies (e.g., the internal-direct interventions mentioned previously) do notsidetrack this intent. Sometimes, however, thegroup will engage in unconscious self-deception—what Habermas calls systematicallydistorted communication—while making adecision. This disruption of the dialogical processof consensual decision making cannot, bydefinition, be facilitated solely by communicativeactions. In such a case, the facilitator alone maybelieve that the group is acting irrationally (or,perhaps, arrives at this diagnosis after a decisionhas been made already). To intervene, thefacilitator both must recognize the distortedcommunication and must be willing to reframe theprocess through which the group is makingdecisions. Such re-framing during a meeting willrequire the facilitator to act strategically in order tosteer the group back toward communicativedialogue. In this intervention, the facilitator mustachieve success—act persuasively—to change the

decision making process (otherwise, of course, theintervention will be rejected).

However, a re-framing intervention occurringduring the meeting may not adequately change thegroup’s self-understanding of its decision-makingprocess. Follow-up actions after the initial inter-vention may be necessary to help various groupmembers see how their actions during the priormeeting may have blocked open discussion, pre-maturely closed discussion, or been otherwisemisunderstood. For these types of problems, thefacilitator may use various communicative actionsto surface submerged conflicts among the groupmembers that systematically distort the group’sunderstanding of its own actions. And, of course,the group must again decide the issue if the inter-vention is to succeed. Clearly, this interventionnegotiates the group’s decision via either (a) dis-rupting decision making or reopening the decisionprocess, (b) showing different members within thegroup how their interests are not being harmonizedby the group’s actions, or (c) seeking to harmonizetheir interests in the future.

… in a less than perfect world, webelieve facilitators sometimes shouldact strategically in order to steer thegroup’s decision-making process towardtrue consensus. Importantly, duringthese occasions, the facilitator’s strate-gic actions should be oriented towarddirecting the decision process, nottoward influencing the decision out-come.

(2) Although such intervention portrays thefacilitator as playing a major role in negotiating adecision, many times the varied power relationsinherent in the decision-making context are theimpetus for negotiation. For example, in a labor-management context, if a work-site committeewishes to discuss alternative work schedules, itcannot make any viable decisions without at leastconsulting the general manager. Such consulta-tion may quickly lead to the committee making aproposal that it then negotiates with the manager.In essence, the committee does not make a singledecision, but rather it engages in a process orseries of decisions that take into account the powerand preferences of the general manager.

This simple example illustrates a process whichis often more intricate in practice. The general



manager seldom has complete autonomy, andother stakeholders—such as a labor union, a par-ticular division within the organization, or a factionwithin the committee’s constituency—will certainlyattempt to influence the committee’s decisionmaking. Under such circumstances, the facilitatormay perform a variety of roles, from liaison tomediator. As a liaison, the facilitator can conveyinformation to and from the committee, presentingthe committee’s proposal to stakeholders andfeeding their preferences back to the committee. Ifthe conflicting interests of stakeholders cannot besatisfied by the committee’s decision, the facilitatormay attempt to mediate the positions of the stake-holders vis-à-vis the committee. Seldom, however,are the roles of the facilitator well defined, andfacilitators will often find themselves performing aunique blend of roles within a specific decisionmaking context.

In summary, relationships and negotiationswith various stakeholder coalitions may influence across-functional team’s decision making. Thesestakeholder coalitions within the organizationmight include:

• Members within the cross-functional team;

• Other employees at the organization’s work site;

• Organizational members engaged in cross-func-tional programs such as continuous qualityimprovement;

• Divisions within the organization; and, possibly,

• The labor union’s leadership and its rank-and-file members.

While this example is drawn from a labor-man-agement context, its variety of coalitions and theneed to negotiate among them is typical of thegroup decision-making process in any organiza-tion. The specific coalitions will be different, butthey will exist, and the facilitator must recognizenot only their existence but also how their rela-tionships and negotiation processes may affect hisor her efforts to aid the groups to reach consensus.

DIRECTIVES FOR FACILITATING CONSENSUALDECISION MAKING

The model of facilitation based on reachingconsensual understandings—communicativeaction—has many implications for practitionersand researchers. To illustrate these directives, twocases of labor-management decision making areanalyzed. These two cases examine the decisionmaking that occurred in two Quality of WorkingLife (QWL) work-site committees, the DR committeeand the O committee. Each committee existed as

part of a QWL program supported by a large mid-western city and a labor union local. The seniorauthor served as a third-party facilitator for theprogram, and the DR and O committees were twoof the five work-site committees that were visitedon a regular basis (either every other week ormonthly).

The QWL program paralleled but inverted theorganizational structure of the city. The work-sitecommittees were empowered to make decisionsthat directly affected their working conditions, butthey could not violate city-, departmental-, or divi-sion-wide rules and policies. However, a work-sitecommittee could suggest experiments to the higherlevel QWL committees so that changes could beimplemented on a trial basis. Unlike the citybureaucracy, however, the higher level QWL com-mittees did not initiate changes unless pressed bythe work-site committees.

Work-site committees consisted of both fixed(for key management and union roles) and electedpositions (for supervisory and non-supervisoryemployees). Generally, the work-site manager andassistant manager had fixed positions, as did theunion steward and a designated union assistant.The elected positions were more variable in nature.Each committee set up guidelines for elections anddetermined what form of representation of theworkforce should occur in the committee.

DIRECTIVES FOR FACILITATING INTERNALPROCESSES

Four directives can help the process of facili-tating internal processes: (1) avoiding multipleconversations, (2) clarifying frames of reference, (3)preventing premature closure of discussions, and(4) mediating differences of opinion before and aftermeetings.

Avoiding Multiple Conversations

As simple as it may sound, the committeeshould avoid multiple, simultaneous conversa-tions. Although the role of regulating the commit-tee’s conversation usually falls in the hands of thechair of the committee, the facilitator must ensurethat the committee’s focus remains undivided.Such controlling actions remind the committee ofthe rules of procedure underlying the discussion;hence, these actions are oriented to reaching anunderstanding.

If multiple conversations are allowed to con-tinue, the committee’s consensus may be distorted.For example, during a meeting of the December DRcommittee that focused on flextime, the facilitatorsuggested that the committee table flextime dis-

Savage and Hilton


cussion until more information was gathered aboutthe workforce’s interest in flextime. Immediatelyafter this suggestion was made, VRG (the unionsteward) and BIL (an employee) began arguing overwho held the chair position on the flextime sub-committee (neither one seemed to desire the posi-tion since a survey of employee opinion would bethe subcommittee’s responsibility).

While they were arguing, ALF (the manager)addressed the rest of the committee, questioningthe practicability of a flextime program and con-cluding that the flextime discussion should betabled until another work-site committee imple-mented a flextime schedule. Near the end of themeeting, the committee unanimously agreed totable discussion of flextime; as the facilitatorlearned at the next meeting, however, neither ALFnor VRG and BIL recognized that they agreed totable flextime discussion for different reasons.

Clarifying Frames of Reference

A well-accepted activity of the facilitator is toclarify frames of reference by engaging participantsin a dialectical dialogue. Many misunderstandingsare caused by people thinking they are talkingeither (a) about the same thing, when, in fact, theyare discussing different things, or (b) about differ-ent things, when, in fact, they are conversingabout the same thing. Although this seems to be asimple form of intervention, it is often difficult todetect discrepant frames of reference until a mis-understanding does arise, especially if participantsinterrupt each other or otherwise violate implicitrules for turn-taking.

For example, during the same DR committeemeeting mentioned previously, VRG and DIK (chairof the committee and VRG’s immediate supervisor)argued about the virtues of various types of flex-time schedules. The transcript of their conversa-tion indicates that they both agreed that flextimeschedules in which employees did not have tonotify the supervisor of their starting times werenot practicable at the DR work site. VRG and DIK,however, did not mutually recognize this agree-ment because they continually overlapped theirtalk—simply put, they did not listen to each other.The facilitator should have intervened by summa-rizing their positions and asking if they agreed withthose summaries; however, this was not donebecause it seemed apparent from the content ofwhat they said that they were in agreement. Thisoversight suggests that facilitators should assessboth the illocutionary force (what is done) and thecontent (what is said) in order to clarify frames ofreference.

Preventing Premature Closure of Discussions

Perhaps the most important intervention that afacilitator can perform is to prevent prematureclosure of discussion. In short, the facilitatorshould rephrase decision proposals and test forconsensus. This discourse should occur not onlywhen a proposal is being considered positively, butalso whenever a proposal is being discussed nega-tively by the group.

The December DR committee meeting againprovides an example of the consequences of failingto intervene in this manner. Following DIK andVRG’s exchange, BIL made a number of concretesuggestions regarding how flextime could beimplemented. Even though DIK and VRGresponded positively to his proposals, BIL’s sug-gestions were ultimately discarded following ALF’sremark that flextime was not working out atanother work site. The facilitator intervened atthat point by suggesting that the committee assessthe employee desire for flextime. Although thisintervention was innocuous per se, it served toundermine the support that had been expressedfor BIL’s suggestions and to support ALF’s negativeimplications. This same intervention might havebeen more effective if (a) it had been prefaced witha summary of BIL’s ideas, (b) BIL had been askedto validate the summary, and (c) the committeehad been asked for an expression of support for oropposition to these “redefined” ideas. A dialecticalintervention of that sort would possibly have keptBIL’s ideas salient within the committee, furtheringthe concrete, positive examination of flextime bythe committee.

Mediating Differences of Opinion Before and AfterMeetings

Facilitating communicative action by interven-ing before and after meetings usually is a two-stepprocess: (a) Clarify the positions of each party indispute, and (b) explore different approaches thatmay mediate the dispute. Often individuals orcoalitions within a group will oppose an issuewithout clearly articulating the basis for theiropposition during a meeting. To reduce personalantagonism, the facilitator should approach eachparty separately after the meeting in order to clar-ify the positions on the issue. These encountersrequire the facilitator to engage in a dialecticaldialogue, similar to the action used to clarifyframes of reference. If the disputing parties appearto be fairly close in their positions, the facilitatormay meet directly with both parties in order toestablish a common ground before the next groupmeeting. Often, however, the parties are far apartin their views, and the facilitator may need to meet



a number of times with each party separately, lay-ing the ground rules for future group discussionsof the issue. Here, the facilitator engages inactions oriented to reaching an understanding.Such actions may be particularly needed wheninstrumental, issue-oriented conflicts have begunto produce expressive or procedural conflicts (orvice versa).

For example, during both the August and Sep-tember meetings of the DR committee, RPH andARP (supervisors) heatedly objected to consideringflextime at the work site. VRG and DEN (unionrepresentatives) countered these objections in anequally emotional fashion. To restore some calm tothe committee, the facilitator suggested that thefocus of the dispute—a flextime schedule in use atthe MR Work-site—should be examined morecarefully by the committee. As a result, the DRcommittee asked the MR committee to discuss itsflextime schedule. Although the MR committeerefused to send a delegation to discuss flextime, itdid invite the DR committee to visit the MR worksite to collect information about the flextimeschedule.

Many misunderstandings are caused bypeople thinking they are talking either(a) about the same thing, when, in fact,they are discussing different things, or(b) about different things, when, in fact,they are conversing about the samething.

During the October meeting, a four-membertask force consisting of BIL, DIK, RPH, and VRGwas appointed to visit the MR plant, and thefacilitator was asked to accompany the task forceas a “neutral” observer. After this meeting, thefacilitator met separately with BIL and VRG inorder to gather their views on flextime. Theyexpressed the opinion that (a) the MR flextimeschedule was practicable and (b) RPH and ARPwere opposed to flextime because of past abuses atthe MR work site that no longer occurred. Thefacilitator also met with DIK about two weeks afterthe committee meeting to obtain his view of flex-time. Surprisingly, DIK was fairly supportive of theidea, but he felt the flextime schedule practiced atthe MR work site would not succeed at the DRwork site. Moreover, he thought that any practica-ble schedule should be adopted on a work-crewbasis, according to the desires of the supervisorand his crew. Also, he added, the supervisor

should have the right to abolish flextime if he felt itwas not working out.

Because of construction work needing theirattention, DIK and RPH were not able to accom-pany VRG and BIL during a planned visit of theMR work site. BIL and VRG interviewed about 15of the 20 people affected by the flextime scheduleduring this visit and gathered a very favorablepicture of its operation. In order to “balance”VRG’s and BIL's survey results, DIK and RPHarranged another visit to the MR work site.

As supervisors, DIK and RPH were particularlyinterested in how the upper management felt theMR flextime schedule was working, so they metwith only the three top-ranking supervisory per-sonnel. RPH made it very clear that he felt thecrew he had worked fine, and he did not want to“fix it” if “it was not broken.” This approach led tothe disclosure that the present flextime scheduledid create some problems; for example, since mostemployees opted to come in early, only “skeleton”crews worked in the late afternoon. After thismeeting, I spoke with both RPH and DIK about thefeasibility of flextime at the DR work site. RPHstated that he would not participate in any type offlextime schedule, but that if a practicable sched-ule—one in which supervisors would know a weekahead of time when employees would be startingwork—could be introduced, he would not opposeits implementation.

The task force results were reported during theDecember committee meeting. As previously dis-cussed, this meeting led to a fairly open and calmdiscussion of flextime, albeit a discussion thatresulted in the tabling of flextime. Such a discus-sion would not have been possible without theclarification of positions and discussion of differentpossibilities that occurred outside of the committeemeetings.

DIRECTIVES FOR FACILITATING EXTERNALPROCESSES

The previous example illustrates the thin con-ceptual line between internal and external proc-esses. The facilitative efforts concentrated onmediating the DR committee’s conflict over flex-time, yet these efforts led to the involvement of theMR work site. Because the MR committee and thepersonnel at the MR work site did not overtly influ-ence the DR committee’s decision making, it seemsclear that the external process of negotiating adecision did not come into play. However, suchnegotiation processes usually do come into playwhenever a group seeks input from its constitu-

Savage and Hilton


ency and approval from the organization orthrough the labor-management program.

Facilitating group/constituent communicationinvolves a number of steps: (a) identify the inter-ests of different stakeholders, (b) explore differ-ences and similarities among these interests, and(c) seek an overarching interest that harmonizesthese multiple interests. Each group memberrepresents a set of constituents with certain needsand desires. Only if the member knows the inter-ests of his or her constituents, can that membertruly make informed decisions within the group.Moreover, the group’s very survival may hinge onconstituents believing that the group makes a dif-ference. Providing feedback and gathering prefer-ences and ideas are thus important activities ofmembers on the group. These activities can befacilitated, both within the group and within theconstituency.

Beyond facilitating information flow, the facili-tator may help the group establish more formallinks with its constituency via subgroups and taskforces. These subunits assure the group of firmerand more reliable links with its constituency. Yet,formal mechanisms for expressing and channelingthe interests of its constituency do not, per se,ensure that a group will further those interests.Only if the group seeks an overarching interestthat harmonizes seemingly competing interests willthe group’s constituency be fairly represented.Many of the techniques already discussed may beneeded to effect such an overarching interest.

The O committee—which used brainstorming,employee surveys, subcommittee reports, and aconsultant’s feasibility study to decide upon a flex-time schedule—exemplifies how the external-inward processes of decision making may befacilitated. The O committee, with the interven-tions of ART (a facilitator), conducted a number ofbrainstorming sessions to identify issues ofimportance. Among those issues identified weretardiness, flextime, and cross-training. To gathermore information on the flextime issue, ARTencouraged the O committee to contact the MRcommittee which was working on a flextimeexperiment proposal at that time. This contactprovided the impetus for the O committee to surveyits hourly workforce constituency about flextimeduring June. Hence, ART's interventions helpedthe committee better realize its own interests.

The survey results showed that a majority ofthe hourly workforce favored some form of flextime,and to address this interest, the O committeeformed a subcommittee to investigate the feasibilityof flextime. The subcommittee, though favorably

disposed toward flextime, did not produce any con-crete proposals. Its recommendation, accepted bythe O committee in early August, was to have athird party perform an extensive feasibility study.

A new facilitator, JIM, agreed to take on thethird-party role in April. JIM met with many of thecommittee members to gather their input beforepresenting a proposal for the feasibility study inMay. As one of his preliminary recommendations,JIM suggested that the committee create a newsubcommittee on flextime. This subcommittee wasformed in June after JIM presented the results ofthe feasibility study. The subcommittee—withJIM’s guidance—again surveyed the O workforceduring July and August. This second survey,moreover, included not only hourly employees butalso salaried supervisors. The results of the surveyshowed that most supervisors and most employeesfavored a flextime schedule with a core time from 9AM to 3 PM. Certainly, JIM’s interventions notonly helped the committee better understand itsconstituency’s interests, but also provided thecommittee with a means to focus those interestsupon a common goal.

Because of the positive information conveyedby the workforce, the O committee charged theflextime subgroup with drafting plans to implementflextime. JIM again lent his expertise to this effort.The draft plan was presented to the committee inlate November; one major objection surfaced fromsupervisors regarding the supervision of employeesduring non-core times. Rather than having theschedule of the supervisor determined through theformula specified by the subcommittee, the super-visors insisted that management should retain thisprerogative. This change was accepted at the nextmeeting in December. At the following meetinglater in December, another change was also made:participation of a work unit in the flextime sched-ule would be at the discretion of the supervisor.Even though this essentially unchanged version ofthe flextime schedule was not implemented untilMay, the committee—with JIM’s help—had recog-nized and had begun to achieve an overarchinginterest of its constituency.

SUMMARY AND IMPLICATIONS

Although facilitators may spend much of theirtime in meetings, interventions also often occureither before or after meetings. These acts offacilitating may include both mediation and per-suasion. In other words, facilitators may engage incommunicative or strategic actions, dependingupon the exigencies they encounter, their skill, andtheir moral judgment. From the stand point ofseeking consensual agreement, strategic actions



are the most problematic since they require thefacilitator to pursue interests that may be opposedby at least some group members. Nevertheless,strategic actions may help if the group enters intonegotiative relationships with organizational orlabor union stakeholders; moreover, such actionsmay be necessary if the group engages in system-atically distorted communication. In short, in animperfect world, strategic action is sometimes anecessary tool to ensure that decision-makingteams actually engage in a communicative actionprocess and end up with a true consensus asopposed to a false consensus.

Communicative actions, although grounded inthe ideal speech situation, are also problematic interms of their enactment, whether directed towardthe internal or the external decision making proc-esses of a group. To help the internal process, thefacilitator may intervene so as to (1) avoid multiple,simultaneous conversations; (2) clarify frames ofreference; (3) re-phrase decision proposals and testfor consensus; and (4) clarify the positions of each

party in dispute, and explore different approachesthat may mediate the dispute. Similarly, theexternal process of decision making may be facili-tated by (1) identifying the interests of differentstakeholders, (2) exploring differences and simi-larities among these interests, and (3) seeking anoverarching interest that harmonizes these multi-ple interests. While seemingly straightforward,each of the aforementioned interventions requiresthe facilitator to ground these actions in the ethicof seeking a consensual understanding.

One implication of this examination of facilita-tion is to underscore the difficulties that evencarefully conceived and implemented attemptstoward collaboration face within organizations withdeeply embedded hierarchical structures andasymmetrical power relations. Imagine the frus-tration (and steep learning curve) an organizationfaces that attempts to equalize power relations anddisplace the reliance on hierarchy with collabora-tion—without the help of third-party facilitators!

ReferencesCragan, J. F., & Wright, D. W. (1980). Small

group communication research of the 1970’s: Asynthesis and critique. Central States SpeechJournal, 31(3), 197-213.

Destephen, R. S. (1983). Group interaction differ-ences between high and low consensus groups.Western Journal of Speech Communication, 47,340-363.

Fisher, B. A. (1970). Decision emergence: Phasesin group decision-making. Speech Monographs,37 (1), 53-66.

Friedman, R. A., & Podolny, J. (1992). Differentia-tion of boundary spanning roles: Labor nego-tiations and implications for role conflict.Administrative Science Quarterly, 37(1), 28-47.

Friedman, R. A. (1994). Front stage, backstage:The dramatic structure of labor negotiations.Cambridge: The MIT Press.

Gadamer, H-G. (1975). Truth and method. NewYork: Seabury Press.

Greenhalgh, L., Neslin, S. A., & Gilkey, R. W.(1985). The effects of negotiator preferences,situational power, and negotiator personalityon outcomes of business negotiations.Academy of Management Journal, 28(1), 9-33.

Habermas, J. (1970). Toward a theory of commu-nicative competence. In H. P. Dreitzel (Ed.),Recent sociology no. 2: Patterns of communica-tive behavior. pp.115-148. New York:MacMillan Co.

Habermas, J. (1971). Knowledge and humaninterests. (J. J. Shapiro, Trans.). Boston:Beacon Press.

Habermas, J. (1979). What is universal pragmat-ics? In Communication and the evolution ofsociety. (T. McCarthy, Trans.). Boston: BeaconPress.

Habermas, J. (1982). A reply to my critics. In J.B. Thompson & D. Held (Eds.), Habermas:Critical debates. (T. McCarthy, Trans.). pp.219-283. Cambridge: The MIT Press.

Habermas, J. (1984). The theory of communicativeaction: Vol. 1. Reason and the rationalization ofsociety. T. McCarthy, Trans. Boston: BeaconPress.

Herrick, N. (Ed.). (1983). Improving government:Experiments with quality of working life sys-tems. New York: Praeger Publishers.

Hill, T. A. (1976). An experimental study of therelationship between opinionated leadershipand small group consensus. CommunicationMonographs, 43(3), 246-257.

Homans, G. C. (1974). Social behavior: Its elemen-tary forms. (2nd ed.). New York: HarcourtBrace Jovanovich, Inc.

Johannesen, R. L. (1971). The emerging concept ofcommunication as dialogue. Quarterly Journalof Speech, 57(4), 373-381.

Knutson, T. J. (1972). An experimental study ofthe effects of orientation behavior on smallgroup consensus. Speech Monographs, 39(3),159-165.

Savage and Hilton


Knutson, T. J. and Holdridge, W. E. (1975). Ori-entation behavior, leadership and consensus: Apossible functional relationship. SpeechMonographs, 42(2), 107-114.

Knutson, T. J., & Kowitz, I. C. (1977). Effects ofinformation type and level of orientation onconsensus-achievement in substantive andaffective small group conflict. Central StatesSpeech Journal, 28(1), 54-63.

Komorita, S. S., & Hamilton, T. P. (1984). Powerand equity in coalition bargaining. In S. B.Bacharach & E. J. Lawler (Eds.), Research inthe sociology of organizations: Vol. 3. The socialpsychological processes. pp. 189-212.Greenwich, CT: JAI Press Inc.

Neale, M. A., & Bazerman, M. H. (1985). Theeffects of framing and negotiator overconfidenceon bargaining behaviors and outcomes.Academy of Management Journal, 28(1), 34-49.

Niederman, F., & Volkema, R. J. (1999). The effectsof facilitator characteristics on meeting prepa-ration, set up, and implementation. SmallGroup Research, 30(3), 330-360.

Poole, M. S., & Holmes, M. E. (1995). Decisiondevelopment in computer-assisted group deci-sion making. Human Communication Research,22(1), 90- 127.

Poole, M. S., Holmes, M. E., Watson, R., &DeSanctis, G. (1993). Group decision supportsystems and group communication.Communication Research, 20(2), 176 – 213.

Poole, M. S., & Roth, J. (1989a). Decision develop-ment in small groups IV: A typology of groupdecision paths. Human CommunicationResearch, 15(3), 323-356 .

Poole, M. S., & Roth, J. (1989b). Decision develop-ment in small groups V: Test of a contingencymodel. Human Communication Research, 15(4),549-589.

Pruitt, D. G., & Carnevale, P. J. (1993). Negotia-tion in social conflict. Pacific Grove, CA:Brooks/Cole Publishing.

Rubin, J. Z., & Brown, B. R. (1975). The socialpsychology of bargaining and negotiation. NewYork: Academic Press.

Spillman, B., Bezdek, J., & Spillman, R. (1979).Development of an instrument for the dynamicmeasurement of consensus. CommunicationMonographs, 46(1), 1-12.

Wall, J. A., Jr., & Blum, M. W. (1991). Negotia-tions. Journal of Management, 17(3), 273-303.

Wall, J. A., & Callister, R. R. (1995). Conflict andits management. Journal of Management,21(3), 515-58.

Young, O. R. (Ed.). (1975). Bargaining: Formaltheories of negotiation. Urbana, IL: Universityof Illinois Press.

Zager, R., & Rosow, M. P. (Eds.). (1982). The inno-vative organization: Productivity programs inaction. New York: Pergamon Press.

Acknowledgements: Our thanks to the anonymous reviewers who sharpened our thinking on facilitationand critical theory and to Donald J. Cegala, Joseph Pilotta, and Don Ronchi who—at its inception—sharedtheir wisdom in seeing this work to its fruition.

Grant T. Savage is the Richard Scrushy/ HealthSouth Chair and Professor in Healthcare Management.Professor Savage has written extensively on conflict, healthcare, and stakeholder management issues injournals such as the Academy of Management Journal, Academy of Management Executive, Health CareManagement Review, and Hospital & Health Services Administration. He has co-authored five award win-ning papers, and published over 80 articles, chapters, and proceedings. He is a member of the Academyof Management, the International Association for Conflict Management, and the International Associationfor Business and Society. He has taught or conducted research on conflict, healthcare, and stakeholdermanagement in Finland, Germany, the Netherlands, and the United Kingdom. Grant received his B.A.from the University of Connecticut and an M.A. & Ph.D. from Ohio State University. Contact at: Univer-sity of Alabama, College of Commerce and Business Administration, Management and Marketing Depart-ment, Tuscaloosa, AL 35487-0225; Phone: 205-348-2926; Fax: 205-348-6695; E-mail:[email protected]

Chadwick B. Hilton is the Director of International Business Programs and Associate Professor of Man-agement in the College of Commerce and Business Administration. He has published extensively onreader-response to advertising copy, stylistics, and English language training in Japanese corporations injournals such as The Journal of Advertising Research, The International Journal of Advertising,The Journalof Business Ethics, the Journal of Business Communication, and The Journal of Business and TechnicalCommunication. His degrees include an A.B., UNC Chapel Hill; an M.A., N.C. State University; and aPh.D., University of Tennessee. Contact at: University of Alabama, College of Commerce and BusinessAdministration, Management and Marketing Department, Tuscaloosa, AL 35487-0225; Phone: 205-348-9432; Fax: 205-348-6695; Email: [email protected]

Associate Editor: Jean Watts


Facilitating Team Development: A View from the Field

John E. Jones and William L. Bearley

The purposes of this paper are to describe a guidance model for assisting groups of people to gen-erate the conditions of teamwork and to describe the types of interventions that we have made infacilitating team development in a wide array of groups internationally. We will lay out thegroundwork by commenting on differences between teams and groups and the appropriateness ofteam building.

Keywords:

Collaboration, Conflict, Group Development, Synergy, Team Building, Trust, Work Group

INTRODUCTION

In recent years there has been abundantattention to teamwork in the workplace. It isalmost as though executives want “teams by Mon-day.” Many organizations have experimented withself-directed work teams, and the number andvariety of “teams” has mushroomed. There hasarisen considerable confusion regarding teams andthe payoffs of creating them.

We were recently facilitating a team-buildingsession for a senior group within one of the largestcorporations in the U.S. when it occurred to usthat the concept of “team” had little or no meaningfor the participants. Just before a break we askedthem to write anonymously on pieces of paper thenumber of teams they either headed or were mem-bers of. During the break we tallied the results,which roughly described a normal curve—but witha mean of 16. When they returned to the sessionwe showed a graph of the distribution, and no onewas shocked. “It’s a way of life here,” they agreed.

WHAT IS A TEAM, AS DISTINCT FROM A WORKGROUP?

This story illustrates that many people, andmany organizations, are not critical in their think-ing about the word "team." We find it useful tomake a clear distinction between a group of peopleand a team. Most of what are called teams are, inour experience, not teams but groups. This isimportant because you can't get most of theadvantages claimed for teams simply by hangingthe label "team" on a group. The overarching goalof team building is to forge a group into a team.Having sharp definitions of these two terms helpsguide us as we plan and facilitate our interven-tions. Here is the distinction that we work from.

Group: a collection or aggregation of persons,places, or things that have at least one thing incommon. We are in this room together. That makes

us a group. Some of us are male. That makes us agroup -- or subgroup, if there are also females inthe group as a whole. Three of us support OptionA. That makes us a group or subgroup, dependingon whether remaining members support otheroptions.

Team: a functioning unit of people who meetthe following criteria of “teamness,” paraphrasedfrom Jones & Reilly (1974).

q The team has a “charter,” or an officialmandate form the organization;

q The members have work tasks that areinterdependent;

q The members are committed to collabora-tion and coordination;

q The team is held accountable (andrewarded) by the organization as a singleunit.

Team members are said to have spirit, or asense of loyalty to and belief in each other. Afterall, they need each other to do their work. Theyorganize to make work decisions, solve work-related problems, make plans, and managechange.

Work groups are imbedded within organiza-tional systems that may either support or obviateteamwork. Some organizations are ineffective withregard to communicating clear expectations towork groups. Others have reward systems thatfocus on the contributions of individuals, withphony performance-appraisal procedures andpractices. Unless the organization is ready, willing,and able to change its systems to emphasizeteamwork, building teams is a futile activity.

A group in the workplace may or may not be ateam—or even need to be one, as we discuss below.Their work may be parallel or unrelated to eachother. They may simply report to one boss as a

Jones and Bearley


matter of convenience. They may never make col-laborative decisions beyond those associated withthe annual Christmas party.

A note on size. It is probably not possible tosustain high levels of teamwork in groups of peoplewith more than about fifteen members. Carryingout the normal functions of an interdependentteam is made increasingly difficult with the addi-tion of any members beyond that threshold. Thesmallest team consists of two persons in partner-ship to accomplish common goals. A three-personteam is more complex dynamically. Adding moremembers dramatically increases the intricacy ofinterpersonal relations within the team. A five-member team has ten two-person relationshipswithin it, for example, but one of ten members hasforty-five pairs of two-person relationships.

U.S. football “teams” are not teams in the sensethat we are using the term here. They are simplytoo big to function collectively or efficiently. Inreality, they are several teams that operate more orless independently of each other: the offense; thedefense; and special teams such as the punt team,the kickoff team, and the onside-kick team. Callingall of the leaders of an organization “our manage-ment team,” when it comprises several dozen oreven hundreds of managers, is not what we meanby the term “team.”

When are teams either desirable or necessary?

When the work is organized in ways such thatthe people who do the work need to collaboratewith each other, working as a team is not onlydesirable but also necessary. If the work isassigned to individual contributors that are givendiscrete tasks to perform, forming a team is proba-bly unnecessary and may be undesirable.

The story of the development of the transistoris a case in point. We have had considerableexperience in consulting in research and develop-ment (R&D) components of various large organiza-tions. Pure research is sometimes a solitary activ-ity, headed by a person with a burning idea. Thatindividual may assemble a staff of researchers andassistants to carry out various functions, but thehead researcher controls decision making. In thecase of the transistor, a researcher at Bell Labo-ratories had a general idea about how to solve theproblems presented by the use of vacuum tubes.He recruited two other leading researchers to workwith him. The three worked at the idea, but, alongthe way, the group broke up because the lead took

credit for the contributions of the others. In thiscase, the unsuccessful and ultimately unnecessaryattempt to form a team probably both slowed thework and permanently damaged important workingrelationships.

A MATRIX MODEL OF GROUP DEVELOPMENT

One of the prerequisites for working as a teamis going through a process of becoming a team.This process is referred to in the literature asgroup development, and the process often appliesto other groups besides those trying to becometeams.

The fact that there are so many models ofgroup development may be due to the challenge ofcutting through the complexity of interactionswithin groups of people. We published a compara-tive chart of group development models (Jones,Bearley, & Watsabaugh, 1996) reproduced in Table1.

What is interesting about this comparison isthe relative sameness of the elements of the seem-ingly disparate models. The chart does not depictadequately the comprehensiveness of some of themodels, of course, because their major elementshave been reduced to a labeling of stages of groupdevelopment. The comparisons are, in that sense,“apples to oranges,” rather than “apples to apples.”However, each of these models shows that groups,including groups seeking to become teams, gothrough a series of challenges. Each of thesechallenges, if successully met, strengthens theirability to work together.

Each of these models is one dimensional. Ourfield experience has convinced us that groupsactually progress along two dimensions: howthey're working, or task behavior, and how they'reworking together, or relationship behavior. We findutility in thinking of groups as being on- or off-track. The processes of group development that weobserve are not necessarily linear. Sometimesteams sacrifice the task for internal calm, andsometimes teams sacrifice human relations in theservice of “getting the job done.” Consequently, wewanted our model to assist groups to diagnosetheir status in order to get back or stay on track todeveloping the conditions that characterize effec-tive teamwork. As shown in Figure 1, displayingthe team’s status at any given time on a two-dimensional matrix permits us not only to look atthe track but also to show the probable dynamicsin a variety of group conflict.



Table 1. Comparison of Group Development Models

Model Builder(s) Stages of Group Development

Shutz (1958) Inclusion Control Openness

Bion (1961) Flight Fight Unite

Tuckman (1965) Forming Storming Norming Performing

Francis &

Young (1979)Testing Infighting Getting Organized Mature

Closeness

Woodcock &Francis (1980)

RitualSniffing

Infighting Experimentation Effectiveness &Maturity

Mossbruker(1988)

Orientationto Group andto Task

ConflictoverControl

Group Formationand Solidarity

Differentiationand Productivity

Osburn, Moran,Musselwhite,Zenger (1990)

State ofConfusion

LeaderCentered

Tightly Formed Self Directed

Varney (1991) Formation Building Working Maturity

Figure 1. Team Development Matrix

Jones and Bearley


This model guides our facilitation of teamdevelopment. We see groups developing along twomajor dimensions, or axes. Organizations assemblework groups to get tasks done, not simply to havegood human relations. Since the task is usuallyprimary, groups often concentrate on it at theexpense of, in spite of, without sensitivity to, orunaware of the effects of group-task interactionson the group’s social system.

Task Behavior

This major dimension of group developmentincorporates four phases, or developmental tasks.Here are brief descriptions of the phases of TaskBehaviors.

Orientation. In the beginning, or when taskschange, members of the group need to learn whatis required of them as a group, what they have towork with, and the organization’s expectations ofthem with regard to quality, timelines, etc.

Organization. When the group comes to acommon understanding of the task’s requirements,it can then organize itself to achieve its goals.Unless the common goals are clear, however, thegroup can vacillate between Organization and Ori-entation, the first phase.

Open Data Flow. After the group agrees on awork plan and processes, it is time to ensure thatall available data within the group that are relevantto the task be available to all members. This meanssharing facts, opinions, feelings, hunches, connec-tions, networks, etc., with everyone.

Problem Solving. This advanced phase ofgroup development is the target for task comple-tion. The group members all know what the taskrequirements and goals are, they are living out anagreed-upon work arrangement for accomplishingthe group’s objectives, they are exchanging datafreely, and they are collectively managing theimplementation of the decisions they have made.

Unfortunately, groups are often composed ofpeople with different agendas. Some membersshow high integrity, some leaders behave as hardtaskmasters, some members are reluctant to speakup, and some are risk takers. The diversity of per-sonalities, motives, interpersonal histories, andother complexities combine to generate both func-tional and dysfunctional interpersonal events. Notethat in the case of Task Behaviors the four phasesare cumulative. That is, in order to engage ineffective Problem Solving groups must have OpenData Flow, and this is optimized by good Organi-zation, which, in turn, requires adequate Orienta-tion.

Relationship Behavior

This is the people side of the matrix. It alsoconsists of four developmental tasks, or phases, ofgroup development.

Dependency. In the beginning, members aredependent on the leader, or convener, for direction.Here the leader must explain the task require-ments and generate a common commitment toshared goals. Note that a reversion to this firstphase of development can also occur if the organi-zation changes its expectations of the group.

Conflict. Almost all observers of group devel-opment have noted that for a group to become ateam, it must go through a period of internal strife.Most often this is centered on a struggle for leader-ship or influence within the group. Unless thegroup faces this conflict, it will become “stuck” orgo “off-track.”

Cohesion. Surviving the difficulties of the Con-flict phase usually results in feelings of relief,increased trust in each other, and a sense that“we’re all in this together.” Notice that this is notthe optimal phase, however; it just feels better.

Interdependence. In this phase of Relation-ship Behaviors, the group’s trust level is so wellestablished that it can organize and reorganizeitself any way it wishes. It can operate in theabsence of leading members. The work tasks canbe knitted together because members are confidentin each other.

Up the Diagonal or “On Track”

Our Team Development Matrix indicates that,when groups progress toward teamness, they go upthe diagonal of the matrix, more or less. This is nota “stage” model, as almost all others are. Itdescribes patterns and sets of behaviors. Thediagonal of the matrix is not an “ideal” develop-mental sequence. It is simply the most efficient andeffective behavioral manifestation of team develop-ment. Sometimes teams “wobble,” and the facilita-tor assists them in returning to a productive line ofdevelopment.

“Square One.” Here we usually see phenom-ena that emerge from coping with the newness ofthe group, new tasks, new members, a new leader,etc. This is not so much a stage of group develop-ment as a set of conditions in which there is a lackof common orientation to the group’s task anddependency on the leader for direction.

Fractionation. This condition is marked bynegotiation on how to carry out the task and byinterpersonal conflict, particularly with regard to



leadership and influence. Again, this is not agroup-development “stage.” It is a general status inwhich groups “bounce around” with their internalstruggles.

Sharing, Dialog, and Trust. This conditionrepresents Open Data Flow regarding Task Behav-iors and Cohesion regarding Relationship Behav-iors. Members enjoy this condition because theyhave usually gone through conflict and stucktogether and are exchanging task-relevant datafreely.

Collaboration. Our behavioral definition of ateam, as opposed to a group, then, is a functioninghuman unit that knows what its task requirementsare, agrees on how to accomplish its goals,exchanges data openly, and engages in productiveproblem solving (the Task Behaviors side of thematrix) and who have risen to such a high level ofinterpersonal trust that they can capitalize on theirconfidence in and interdependence with eachother. It is in this condition that we observe bothshared decision making and camaraderie.

Becoming a true team is not a stopping place,however. Teams that are highly successful canbecome consistently synergistic, raising standardsabove expectations and achieving goals that areunexpected by the organization.

Synergy. This level of group development isseldom achieved, in our experience. Organizationsoften make what appear to be arbitrary changes,and these can have serious negative impacts onhighly functioning teams. When disruptive changeoccurs, some teams “backslide,” moving to a lowerstage of development. If the team is truly optimal,however, it can go back through the necessaryphases of development in Task and Relationshipbehaviors more quickly than originally. This canalso occur when the team’s membership (includingleadership) changes.

We have used the term “off-track.” By that wemean that groups can display combinations ofTask and Relationship behaviors off the diagonal ofthe matrix. For example, if a group were in theOrganization phase of Task Behaviors and theDependency phase of Relationship Behaviors, wewould probably observe inefficiency and wouldsearch for procedures. These off-diagonal condi-tions are important to know about as we planteam-building interventions. In fact, our mostcommon observation is that groups do not grow ineither a smoothly progressive manner or by “leapsand bounds.” They tend strongly to move both tothe right and left of the diagonal of the matrix. Pro-gressing up the diagonal is not a statement of idealteam development. It is descriptive of an efficient

use of group resources toward high levels of func-tioning, but one that even successful teams rarelyaccomplish.

Note that the Team Development Matrix doesnot simply give new names for the popular “Form-ing, Storming, Norming, Performing” sequence thatconstitutes the Tuckman (1965) model. We con-sider it unfortunate that that model (built on akind of meta-analysis of models existing at thetime of its formulation) uses rhyming labels for itsfour stages. The Tuckman model, in our judgment,is an oversimplification. What is unfortunate aboutthe rhyming is that it has become mnemonicallypowerful, to the point that a large proportion offacilitators consider it to be the truth. In our judg-ment, its only merits are that the model is easy toteach and Storming is legitimized as a subject forproblem solving.

The “classical” team-building intervention

The usual sequence of activities that we carryout in facilitating group development is linear.There is a beginning, middle, end, and followthrough. Here is the “normal” sequence.

Contracting. We work with the leadership ofthe work group to establish common expectationsand hoped-for outcomes. We generate an explicitagreement on fees and expenses, timing, etc.

Assessment. To prepare members of a workgroup for a team-building session, we intervieweach individual, beginning and ending with theleadership. These interviews are anonymous butnot confidential. We explain that we plan to pub-lish trends and themes from the interviews to thewhole group. We have each member complete ourGroup Development Assessment (GDA) (Jones &Bearley, 1986) instrument, which is based on anearlier expression of the Team Development Matrix.

Data Analysis. First, we study our verbatiminterview notes for common themes and prepare apresentation, often in PowerPoint, that is organizedaround themes and contains anonymous quota-tions from work-group members. Second, we scoreand profile the GDA and construct a group profilethat arrays the pairs of highest scores on TaskBehaviors and Process Behaviors.

Design. We outline the design of the team-building session, including statements of theagenda and the desired outcomes of the session.The overarching theme of the design is almostalways, “How can we work together better in orderto accomplish our goals more effectively.” Often theagenda includes interpersonal sharing of feedbackon the extent to which each member is being a

Jones and Bearley


team player, according to our behavioral defini-tions. Guiding our design is our analysis of wherethe team, or potential team, falls on the TeamDevelopment Matrix and how to move them back toor along the diagonal.

Facilitating the session. The leadership of thework group, ideally at least, opens the team-building session. We have usually coached thesemembers regarding setting a positive tone. Typi-cally, we then present the results of the Assess-ment phase and explain our role(s) in the session.Then the work group goes to work. We watch andlisten, verbally reinforcing productive behaviorsand making process suggestions for ways to attackchronic problems. We stay out of the content ofwhat is being discussed. Sometimes we interveneto explain effective group processes that couldmove the particular group forward. What they talkabout normally are the real problems that are fac-ing the group right now. The structured interven-tions that we usually include are interpersonalexchange of feedback and brief instruction on vari-ous problem-solving techniques (Harvey, Bearley,& Corkrum,1997).

Evaluation. We are not interested in gathering“smile sheet” data at the end of team-building ses-sions. (See Jones, 1990). We sponsor having thework group evaluate their own work during thesession. A simple method that we use is to ask twoquestions: (1) On a ten-point scale, how well did wework together as a team during this time together?and (2) What would it take to make it a 10 in ournext meeting? We then facilitate publishing thesedata, pushing for agreement on improved groupbehavior.

Follow-through. Work groups do not grow in asaltatory manner. There are many small steps,some of them backward, and few real leaps.Because teams can revert to group-like behaviorwhen their membership or the organizationchanges, it is important to have “booster shots.”The most dramatic outcome of quality-tools train-ing, for example, is using them for a while andslowly abandoning them. They are “overtaken byevents.” We attempt to include follow-through ses-sions in our contracts. In one large manufacturingplant, for example, we facilitated six senior-leaderoffsite sessions in three years. In a large research-and-development organization, we facilitatedmonthly team development sessions for the topleaders for three years, including three three-dayretreats.

Alternatives to the ideal intervention

It is often difficult to get members away fromthe workplace for any significant length of time, sothe facilitator needs to consider alternatives to theclassical off-site session. There are numerousactivities that can result in team development, andthe facilitator needs to be creative in finding waysto access the work group and methods for fosteringdevelopment.

Staff Meetings. Some work groups haveweekly staff meetings that can become the forumfor facilitating group development or team buildingif the group will accept the role and function of thefacilitator as adding value to their interactions. Inthis case we simply observe the group processesthat underlie how the group does its business,making interventions as necessary. There is nodesign; we remain ready to intervene to promoteinterdependent problem solving competencies andprocesses.

Startups. The best time to facilitate groupdevelopment is at the beginning of the group’s life.This practice can make going through the Conflictstage quicker and more thorough. The work groupcan mature more rapidly when it has the servicesof a disinterested interventionist. In this case, theuse of Assessment and Data Analysis are some-what different: instead of diagnosing where thegroup is, we are diagnosing the potential of thegroup to become a team and the ability of the envi-ronment to support true teamwork. We typicallydo this by engaging the group in SWOT analysis.

Training Sessions. Sometimes it is advanta-geous to put a work group through a series of brieftraining modules that are designed to result in thematuration of the group. The facilitator thenbecomes the trainer, and group members becomeparticipants. Debriefing such sessions must be, inour opinion, directly tied to the group’s work. Thisis not “entertrainment.” It is an alternative to teamself-diagnosis and improvement planning. Usingtraining activities that do not transfer to thegroup’s Task and Relationship Behaviors can be awaste of resources. While such training can be fun,the goal is not so much recreation as it is serious,task-relevant learning by both individual membersand the group as a functioning unit.

A note: The term “touchy-feely” is used bymany people to denigrate facilitator interventionswithout a standard of meaning. We suspect thatfor many who use it the term refers to anythingthat the user finds uncomfortable. Whether this istrue is less significant than the reality that teambuilding is both controversial and threatening tomany work-group members. An easy way for them



to deal with their discomfort is to put down theactivities they don’t like. As professional facilita-tors we must take this concern into account in allof our interactions with members of work groups,particularly before formal team-building sessions.We assure members that the work will focus onbecoming increasingly effective in achieving thegroup’s or team’s task goals. We do not facilitate“tree-hugging” or other activities that are notdirectly related to the group’s charter. All of ourcommunications emphasize the serious nature ofgroup development. We want members to show upready to work, and perhaps also to enjoy the proc-ess. Facilitating group development is not fun andgames. It is hardheaded attention to getting theresults that the organization expects.

“Noises” that team development facilitators make:The language of team building

For each place that a group or team might finditself on the Team Development Matrix, there areappropriate interventions, usually in the form ofone or more questions or statement-question pairs.In The New Fieldbook for Training (Jones, Bearley,& Watsabaugh, 1996) we catalog and brieflydescribe 86 facilitation interventions in teamdevelopment. Many of the interventions are tieddirectly to the Team Development Matrix.

Here is our generic process intervention (teamself-diagnosis):

q What’s going on?

q How do you feel about it?

q What would be better?

We use this intervention most often, because itgenerates team self-diagnosis that responds toconsiderations of both Task Behavior and Rela-tionship behavior simultaneously.

In TEAMBOOK (Jones and Bearley, 1994), welist typical things that facilitators say while facili-tating team development. Here is our list, alongwith links to the Team Development Matrix.

“What’s going on?”

This intervention asks the work group to diag-nose itself. If members understand the TeamDevelopment Matrix, they can answer this questionwith precision.

“Let’s pause for a moment to check our process.”

This as another way to sponsor work-groupself-diagnosis, and the Team Development Matrixprovides a convenient rubric for this analysis.

“Hold it. We seem to be moving away from the processwe agreed on at the beginning of the meeting.”

This intervention attempts to get the group to“get back on track” with regard to the Organizationphase of Task Behavior.

“Let’s see how people are reacting to what’s going onright now.”

Relationship Behavior is often ignored in theservice of getting tasks accomplished. This facili-tator intervention prods the work group to open upthis dimension for discussion and heightened sen-sitivity.

“I have a hunch that what’s happening right now is...”

The facilitator’s role does not include diagnos-ing the group’s development for them. Rather, wetest our hunches by promoting group self-diagno-sis, and we use the matrix as a map for the groupto determine where they are.

“Let me test an observation about what I’m seeing goingon.”

Here the facilitator is testing not a hunch butan observation. Since many observers of groupinteraction tend to be judgmental in their observa-tions, we use the matrix to denote the behavioralphenomena that we see. What we are after, how-ever, is to get the group to engage in this type ofanalysis spontaneously, especially when membersbecome uncomfortable with what they see is hap-pening in the group.

“I want each of you to rate this session so far on a ten-point scale, where 10 is high.”

Although this is a judgmental and simplisticassessment, it often causes the group to becomemore attentive to the Relationship Behavior side ofthe matrix.

“Then think about what it would take to make the meet-ing a ten or better.”

This is, in reality, an extension of the immedi-ately previous intervention. It can make it easy tolegitimize discussion and problem solving on Rela-tionship Behaviors.

“Let’s take time to share how we’re experiencing eachother in this meeting before we pick up the next agendaitem.”

This is another intervention that focuses thegroup’s energies on its Relationship Behaviors.

“Let’s review how well we’re using the six-step process.”

This intervention promotes discipline on TaskBehaviors. Groups tend to become sloppy with themethods they have been taught for problem solv-

Jones and Bearley


ing, decision making, and action planning. Thefacilitator needs to work with the group to use itsagreed upon methodology or agree consciously tomodify it.

“Where are we in the nine-step procedure?”

The comment directly above applies to thisintervention also. When groups show a lack ofcommitment to discipline in Task Behaviors, it istime for the facilitator to sponsor self-correctiveaction.

“I want us to have a ‘go-around’: make a statement thatbegins with, ‘Right now I...’ Maybe the roles in this ses-sion aren’t clear. ”

We are not reluctant to put people on the spotin facilitating team development. Usually membersare being paid to cooperate with each other. Wepromote accountability, including members’ takingpersonal responsibility to see that their task rolesare clear. Using the matrix, we help members toclarify their roles according to the four phases ofTask Behaviors.

“Let’s test that. What would be better for you in thismeeting today?”

This intervention is similar to one describedearlier, but this one is spontaneous. The facilitatorasks a member to “come clean” with preferencesregarding how the group’s meeting progresses.Often this intervention makes it easy to initiate agroup discussion of Relationship Behavior withinthe group.

“What is your sense of how well the group is workingtoward its goal today?”

The Team Development Matrix provides an effi-cient model for focusing this team self-assessment.The facilitator is careful not to “grade” the group,but members do so continuously. The matrix givesthem a way of becoming less evaluative and moredescriptive. The facilitator can work with thisresult to get the group to look at its own function-ing in a developmental way.

“Who feels like a winner or a loser right now?”

In group decision making, which is an integralpart of all problem solving, sometimes individualmembers’ feelings and opinions are suppressed ina quest for consensus. This can lead to a conditionof “groupthink” in which the group intimidatesmembers. Using the matrix as a diagnostic frame-work can obviate this phenomenon by making iteasy for members to describe how they are experi-encing the group’s processes before they becomedysfunctional.

“What have we learned in this meeting that can makethe next one better?”

The meta-goal of the facilitator is for the groupto “police” and reward itself in the absence of ahelper. Getting the group to commit to becoming a“learning group” can lead to a condition of con-tinuous improvement in team functioning. Thematrix provides a simple but powerful frame forthis learning.

Of course, facilitators may not talk exactly likethat. We adapt our language to each group.

The imperatives of facilitation

There are some rules that we apply to facilitat-ing group development. The guidelines that focusour wor arebased on our field experience withnumerous work groups and teams at all levels, inboth for-profit and not-for-profit organizations.

Show up prepared.

As Woody Allen once said, “Showing up iseighty percent of life.” We hold ourselves account-able for doing our homework in preparation forfacilitating group development. That means gath-ering and analyzing information about the group’spresent development pattern. The Team Develop-ment Matrix is an effective model for such datagathering and analysis. If there is a formal groupleader, we apply two rules: never surprise the boss,and never be surprised by the boss.

Be responsive to the group.

This means listening accurately to members,soliciting wants, needs, and expectations, and con-sidering the appropriateness of the interventionswith which we are familiar.

Promote success.

Organizations create work groups to get workdone. The work is primary, and, as facilitators, weattempt to assist groups to identify their desiredoutputs and to produce them efficiently, consistentwith the values that they espouse as a group.

Use your expertise when appropriate.

Group members often ask facilitators for theirexperience or opinions. “What do you think?” Weare usually tempted to respond, “I think youshould ask better questions,” but we fight thetemptation. When our expertise is directly relevantto the concern that a work group is addressing, weask permission to take off our “facilitator’s hat” fora moment in order to share what we know orthink. Then we get back into role as quickly aspossible.



When in doubt, confront; when all else fails, try honesty.

Group development is often uncomfortable. It isoften up to us to point out that “there is a mooseon the table.” When the group is not facing itsinternal conflicts squarely, we are prepared to openup the subject and facilitate its consideration. Foryouth, we call it “growing pains.” For adults, wewant to assist work groups to find ways of manag-ing conflict in the service of task and processeffectiveness.

Sponsor acceptance of interdependence.

Independence is a value prized in our largersociety, and many work-group members have tolearn the benefits of accepting this condition. “LoneRangers” have particular difficulties with this con-cept or value. In facilitating work-group develop-ment we ask members to analyze the dependencieswithin the group, such as customer-supplier rela-tionships.

Assist in movement toward problem solving capacity.

Since that interdependent problem solvingdefines “effective team” in our model, we con-sciously promote improvement in the group’s abil-ity to solve problems effectively. We often use qual-ity tools to work in this direction.

These are not hard-and-fast rules. They simplyguide our behavior as facilitators. They are ourstandards for ourselves as professionals when wefacilitate group development.

CONCLUSION

Businesses today recognize the need for thehigh-performance, collaborative efforts of teamsbut often do not recognize what it takes for a workgroup to actually achieve this level of performance.By working with these organizations and thepotential teams that are formed within them, wecan help them move through the developmentphases of task and relationship behavior to becomea true, synergistic team.

ReferencesFrancis, D., & Young, D. (1979). Improving work

groups: A practical manual for team building.San Diego, CA: University Associates.

Harvey, T. R., Bearley, W. L., & Corkrum, S. M.(1997). The practical decision maker: A hand-book for decision making and problem solving inorganizations. Lancaster, PA: Technomics.

Jones, J. E. (1999). Productive and counterproduc-tive behaviors of team members. In the “Free-bies” section of the Website http://ous.iex.net.

Jones. J. E. (1990). Don’t smile about smile sheets.Training & Development Journal, December, 19-21. Aslo available at http://ous.iex.net.

Jones, J. E., & Bearley, W. L. (1986). Group devel-opment assessment. King of Prussia, PA:HRDQ.

Jones, J. E., & Bearley, W. L. (1994). Teambook:Twenty-seven exercises for enhancing workgroups. King of Prussia, PA: HRDQ.

Jones, J. E., Bearley, W. L., & Watsabaugh, D. C.(1996). The new fieldbook for training: Tips,tricks, tools, & techniques. Amherst, MA: HRDPress & Minneapolis, MN: LakewoodPublications.

Mossbruker, J. (1988). Developing a productivityteam: Making groups and work teams work. InW. B. Reddy & K. Jamison (Eds.), Team build-ing: Blueprints for productivity and satisfaction.Alexandria, VA: NTL Institute for AppliedBehavioral Sciences & San Francisco: Jossey-Bass Pfeiffer.

Orsburn, S. D., Moran, L., Musselwhite, E., &Zenger, J. (1990). Self-directed work teams: Thenew American challenge. Homewood: BusinessOne Irwin.

Reilly, A. J., and Jones. J. E. (1974). Team build-ing. In J. W. Pfeiffer & J. E. Jones (Eds.), The1974 annual handbook for group facilitators.San Francisco: Jossey-Bass Pfeiffer.

Schutz, W. C. (1966). The interpersonal under-world. Palo Alto, CA: Science and BehaviorBooks.

Tuckman, B. W. (1965). Developmental sequencein small groups. In Psychological Bulletin, 63,384-389.

Varney, G. H. (1991). Building productive teams: Anaction guide and resource book. San Francisco:Jossey-Bass.

Woodcock, M., & Francis, C. (1980). Team build-ing: yes or no. In W. W. Burke (Ed.), Trendsand issues in OD: Current theory and practice.San Francisco: Jossey-Bass Pfeiffer.

Jones and Bearley


John E. Jones is President of Organizational Universe Systems. John co-edited numerous experi-ence-based training and consulting handbooks and annuals. As an experienced trainer, presenter,counselor, professor, entrepreneur, and consultant, he effectively bridges the theoretical and thepractical. He consults widely, with such clients as Air Canada, ARC International (U.S.A. & Japan),AT&T Bell Laboratories, AT&T Consumer Products, Coca-Cola, Coors Brewing Company, GeneralMotors, Holiday Inns, Kaiser-Permanente, McKesson, Network Management, Inc., PsychologicalAssociates, Public Service Electric & Gas, Rockwell Telecommunications, SAP Americas, Shearson-Lehman Mortgage, Tonka, Turner Construction, Unisys, Wallace Computer Services, Xerox, andnumerous not-for-profit organizations in education, government, and health care. Dr. Jones taughtat the University of Iowa for eight years, in Counselor Education. He has lectured widely on a widerange of topics in education, training, organization behavior, and leadership. His special interestsin development are training design and delivery, executive team building, intergroup problemsolving, organizational survey feedback, and management development. He has collaborated withDr. William L. Bearley on state-of-the-art management-training models and instruments. Contactat: Organizational Universe Systems, 5412 Barkla Street, San Diego CA 92122-4002; Phone: 858-452-0844; Fax: 858-452-0544; [email protected]; http://ous.iex.net

William L. Bearley is Vice President of Organizational Universe Systems. Bill has a variety ofexperiences as a teacher, professor, trainer, consultant and business person. He is unique in thathe in thoroughly educated in both computer science and behavioral science. Thus, he brings con-siderable breadth of perspective to the practice of developing and improving management systems.He has pioneered the fusion of organization development with the introduction of managementinformation systems in organizations. Bill consults with numerous clients, such as AT&T Commu-nications, AT&T Bell Laboratories, Honeywell, Xerox, L’oreal, McKesson, Kaiser-Permanente,Equifax, and many educational and health-care organizations. Dr. Bearley is a graduate of the UAlaboratory-education intern program and currently serves as a professor in the Doctoral Programin Educational Management at the University of La Verne. Contact at: Organizational UniverseSystems, P.O. Box 38, Valley Center CA 92082; Phone: 760-749-0737; Fax: 760-749-8041;[email protected]; http://ous.iex.net

Associate Editor: Edward Ruete

Classics for Group Facilitators


Developmental Sequence in Small Groups *

Bruce W. Tuckman

* Copyright 1965 by the American Psychological Association. Reprinted with permission. This article

appeared in Psychological Bulletin, Volume 63, Number 6, Pages 384-99.

Editor’s Note

As group facilitators we are often concernedabout the development of the groups with whichwe work. Frequently we make reference to “thestages of group development” and the stages mostfrequently cited are forming, storming, normingand performing. These stages were proposed byBruce Tuckman in 1965 based on his examinationof empirical research studies. In this classicarticle, Developmental Sequence in Small Groups,we find a rich description of these stages under avariety of settings as well as their applicability toboth group structure and task activity.

In a subsequent 1977 article, Stages ofSmall-Group Development Revisited, Tuckman andcoauthor Mary Ann Jensen noted thatsubsequent empirical studies suggested atermination stage which they named adjourning.While Table 1 below summarizes the stages with adescription of their associated group structuresand task activities, the original article provides amuch more complete understanding of theircontext, meaning, and limitations. Althoughother articles in this special issue suggest thelimitations of “stage models” such as this, thememorability and popularity of Tuckman’s modelmake this article required reading for every groupfacilitator.

- Sandor P. Schuman, Editor

Table 1: Stages of Group Development

Group StructureThe pattern of interpersonal

relationships; the way members actand relate to one another.

Task ActivityThe content of interaction as related

to the task at hand.

Forming:orientation, testing anddependence

Testing and dependence Orientation to the task

Storming:resistance to group influenceand task requirements

Intragroup conflictEmotional response to taskdemands

Norming:openness to other groupmembers

Ingroup feeling and cohesivenessdevelop; new standards evolveand new roles are adopted

Open exchange of relevantinterpretations; intimate,personal opinions are expressed

Performing:constructive action

Roles become flexible andfunctional; structural issues havebeen resolved; structure cansupport task performance

Interpersonal structure becomesthe tool of task activities; groupenergy is channeled into thetask; solutions can emerge

Adjourning:disengagement

Anxiety about separation andtermination; sadness; feelingstoward leader and group members

Self-evaluation

Tuckman


Author’s Note

My first professional job was as part of a small group of social psychologists in a think tank settingstudying small group behavior as the US Navy prepared for a future of small crew vessels and stations.Nine of us at the Naval Medical Research Institute were busy studying small groups from all perspectivesand under all conditions. I was fortunate to have an experienced and talented boss by the name ofIrwin Altman, who had been collecting every article he could find on group development. He turned hiscollection over to me and suggested that I look it over and see if I could make anything out of it.

The collection contained 50 articles, many of which were psychoanalytic studies of therapy or T-groups. The task of organizing and integrating them was challenging. After separating out two realms ofgroup functioning, namely, the interpersonal or group structure realm and the task activity realm, Ibegan to look for a developmental sequence that would fit the findings of a majority of the studies. I hiton four stages going from (1) orientation/testing/dependence, to (2) conflict, to (3) group cohesion, to (4)functional role-relatedness. For these I coined the terms: forming, storming, norming, andperforming—terms that would come to be used to describe developing groups for the next 20 years andwhich probably account for the paper’s popularity.

There still remained the task of getting the paper published and that was no mean feat. LloydHumphreys, then editor of the Psychological Bulletin, turned it down, offering me constructive editorialcriticism, but concluding that the reviewed studies themselves were not of sufficient quality to meritpublication. I was persistent, though, and rewrote the manuscript per his recommendations and sent itback to him, despite his initial outright rejection. I pointed out that I was not trying to justify thecollected articles but to draw inferences from them. Humphreys did a complete about-face and acceptedmy argument and my manuscript and, in short order, it appeared in print.

I ordered, thanks to the navy, 450 reprints and used them all to fill requests within the first three orfour years after the article appeared. Requests came from all over the world and from a wide range of dis-ciplines and I have saved some of the more exotic ones. Almost yearly, I receive a request from someoneto use parts of the article or at least the terms forming, storming, norming, and performing in print.Again, quotability may be the key to success.

In 1977, I published, by invitation, an update of the model in a journal called Group & OrganizationStudies—in collaboration with Mary Ann Jensen.1 We reviewed 22 studies that had appeared since theoriginal publication of the model and which we located by means of the Social Sciences Citation Index.These articles, one of which dubbed the stages ‘Tuckman’s hypothesis,’2 tended to support theexistence of the four stages, but also suggested a fifth stage for which a perfect rhyme could not befound. We called it ‘adjourning.’

1 Tuckman, B. W. & Jensen, M. A. Stages of small-group development revisited. Group Org. Studies

2:419-27, 1977.2 Runkel, P. J., Lawrence M., Oldfield S., Rider M. & Clark C. Stages of group development -- an empirical

test of Tuckman's hypothesis. Journal of Applied Behavioral Science. 7:180-93, 1971.

Abstract

Fifty articles dealing with stages of group development over time are separated by group setting, asfollows: therapy-group studies, T-group studies, and natural- and laboratory-group studies. The stagesidentified in these articles are separated into those descriptive of social or interpersonal group activitiesand those descriptive of group-task activities. Finally, 4 general stages of development are proposed, andthe review consists of fitting the stages identified in the literature to those proposed. In the social realm,these stages in the developmental sequence are testing-dependence, conflict, cohesion, and functionalroles. In the task realm, they are orientation, emotionality, relevant opinion exchange, and theemergence of solutions. There is a good fit between observed stages and the proposed model. Furtherstudy of temporal change as a dependent variable via the manipulation of specific independent variablesis suggested.

The purpose of this article is to review the literature dealing with the developmental sequence insmall groups, to evaluate this literature as a body, to extrapolate general concepts about groupdevelopment and to suggest fruitful areas for further research.

While small-group processes have been given great attention in recent years by behavioral scientists,the question of change in process over time has been relatively neglected. Perhaps the major reason forthis is the overwhelming tendency of the small-group researcher to run groups for short periods of timeand thus avoid the 'problems' created by temporal change. Laboratory studies of developmental phenom-

Developmental Sequence in Small Groups


ena are quite rare. The majority of articles dealing with sequential group development come from thegroup-therapy setting and human relations training-group setting, neither of which features strictexperimental control nor manipulation of independent variables. Moreover, the only major theoreticalstatements of group development which have appeared are those of Bach (1954), Bales (1953) andSchutz (1958).

In an attempt to bring the facts and the issues into sharper focus, existing research in the area ofsmall-group development will be cited, and a framework within which this phenomenon can be betterunderstood and further investigated will be presented. This framework will also serve to integrate thevariety of studies cited in a meaningful way.

CLASSIFICATION MODEL

The classification approach adopted for distin-guishing between and within developmental stud-ies is a threefold one. The delineations are basedon (a) the setting in which the group is found, (b)the realm into which the group behavior falls atany point in time, that is, task or interpersonal,and (c) the position of the group in a hypotheticaldevelopmental sequence (referred to as the stageof development). It is this last delineation thatallows not only for the separation and ordering ofobservations within each setting, but for thedevelopment of additional hypotheses as well.

Setting

Classification according to setting allows forthe clustering of studies based on their similarityof features, for example, group size, group problemarea, group composition, duration of ‘group life’,etc. More similarity between observations made inthe same setting than in different settings isexpected.

In the group-therapy setting the task is to helpindividuals better deal with their personal prob-lems. The goal is individual adjustment. Suchgroups contain from five to fifteen members, eachof whom has some debilitating personal problem,and a therapist, and the group exists for threemonths or more. The developmental data for suchgroups consist of the observations of the therapistand those professional observers that are present,usually as trainees. Such data are highlyanecdotal in nature and reflect the clinical biasesof the observers. Furthermore, such accounts areusually formulated after the fact and based on theobservation of a single group. Since the bulk ofthe literature reviewed comes from this setting, itsgenerality must be limited by the limitations of thesetting and the mode of data collection.

In the human relations training-group (T-group)setting, the task is to help individuals interactwith one another in a more productive, lessdefensive manner, and to be aware of thedynamics underlying such interaction. The goal isinterpersonal sensitivity. Such groups containordinarily from fifteen to thirty members, usuallystudents or corporation executives, and one

trainer or leader, and endure from about threeweeks to six months.

The most striking differences betweentherapy- and training-group settings are in theareas of group composition, task, goal, andduration of group life. Such differences canaccount for different findings in the two settings.The most striking similarity is in the manner ofdata collection. Data in the training-group settingare highly anecdotal, subjective, collected by thetrainer and his coworkers, and often based on theobservations of a single group. Again, this servesto limit the generality of these findings.

The natural-group setting is distinguished onthe basis that the group exists to perform somesocial or professional function over which theresearcher has no control. Members are notbrought together for self-improvement; rather,they come together to do a job. Such groups maybe characterized either by appointed or emergentleadership. Presidential advisory councils andindustrial groups represent examples of naturalgroups. Similar limitations to generalization basedon the manner of data collection and number ofgroups observed applies in this setting as in theprevious settings.

The laboratory-task setting features groupsbrought together for the purpose of studying groupphenomena. Such groups are small (generallyunder ten members), have a short life, and may ormay not have leaders. In this setting, groups aregiven a task or tasks which they are to complete.Quantitative data are collected and analyzedbased on multiple-group performances.

The last two settings have been combined dueto the small number of studies in each (the dearthof group development studies in the industrialarea is notable), and also because theoreticalstatements are reviewed which are generalized tocover both areas. All studies will be classified intoone of the three setting categories according tobest fit.

Realm: interpersonal v. task

Within the studies reviewed, an attempt willbe made to distinguish between interpersonalstages of group development and task behaviors

Tuckman


exhibited in the group. The contention is that anygroup, regardless of setting, must address itself tothe successful completion of a task. At the sametime, and often through the same behaviors,group members will be relating to one anotherinterpersonally. The pattern of interpersonalrelationships is referred to as group structure andis interpreted as the interpersonal configurationand interpersonal behaviors of the group at apoint in time, that is, the way the members actand relate to one another as persons. Thecontent of interaction as related to the task athand is referred to as task activity. The proposeddistinction between the group as a social entityand the group as a task entity is similar to thedistinction between the task-oriented functions ofgroups and the social-emotional-integrativefunctions of groups, both of which occur assimultaneous aspects of group functioning (Bales,1953; Coffey, 1952; Deutsch, 1949; Jennings,1947).

In therapy groups and T-groups, the task is apersonal and interpersonal one in that the groupexists to help the individuals deal with themselvesand others. This makes the interpersonal-taskdistinction a fuzzy one. A further problem withthis distinction occurs because the studies citeddo not distinguish between the two realms andoften talk about interpersonal development at onepoint in the sequence and task development atanother point. The distinction will be maintained,however, because of the generic differencebetween the reaction to others as elements of thegroup task versus the reaction to others as socialentities. Failing to separate stages by realmobscures the continuity of the developmentalprocess. While the two realms differ in content, aswill be seen, their underlying dynamics are similar.

Proposed developmental sequence

The following model is offered as aconceptualization of changes in group behavior, inboth social and task realms, across all groupsettings, over time. It represents a set ofhypotheses reflecting the author’s biases (ratherthan those of the researchers) and the perceptionof trends in the studies reviewed which becomeconsiderably more apparent when these studiesare viewed in the light of the model. The model ofdevelopment stages presented below is notsuggested for primary use as an organizationalvehicle, although it serves that function here.Rather, it is a conceptual statement suggested bythe data presented and subject to further test.

In the realm of group structure the firsthypothesized stage of the model is labeled as test-ing and dependence. The term ‘testing’ refers to anattempt by group members to discover what inter-personal behaviors are acceptable in the group,based on the reactions of the therapist or trainer

(where one is present) and on the reactions of theother group members. Coincident to discoveringthe boundaries of the situation by testing, onerelates to the therapist, trainer, some powerfulgroup member, or existing norms and structures ina dependent way. One looks to this person, per-sons, or standards for guidance and support inthis new and unstructured situation.

The first stage of task-activity development islabeled as orientation to the task, in which groupmembers attempt to identify the task in terms ofits relevant parameters and the manner in whichthe group experience will be used to accomplishthe task. The group must decide upon the type ofinformation they will need in dealing with the taskand how this information is to be obtained. In ori-enting to the task, one is essentially defining it bydiscovering its ‘ground rules’. Thus, orientation, ingeneral, characterizes behavior in both interper-sonal and task realms during this stage. It is to beemphasized that orientation is a general class ofbehavior which cuts across settings; the specificsof orientation, that is, what one must orient toand how, will be setting-specific.

The second phase in the development of groupstructure is labeled as intragroup conflict. Groupmembers become hostile toward one another andtoward a therapist or trainer as a means ofexpressing their individuality and resisting theformation of group structure. Interaction isuneven and ‘infighting’ is common. The lack ofunity is an outstanding feature of this phase.There are characteristic key issues that polarizethe group and boil down to the conflict overprogression into the ‘unknown’ of interpersonalrelations or regression to the security of earlierdependence.

Emotional response to task demands is identi-fied as the second stage of task-activity develop-ment. Group members react emotionally to thetask as a form of resistance to the demands of thetask on the individual, that is, the discrepancybetween the individual’s personal orientation andthat demanded by the task. This task stage will bemost evident when the task has as its goalself-understanding and self-change, namely, thetherapy- and training-group tasks, and will beconsiderably less visible in groups working onimpersonal, intellectual tasks. In both task andinterpersonal realms, emotionality in response toa discrepancy characterizes this stage. However,the source of the discrepancy is different in thedifferent realms.

The third group structure phase is labeled asthe development of group cohesion. Group membersaccept the group and accept the idiosyncrasies offellow members. The group becomes an entity byvirtue of its acceptance by the members, theirdesire to maintain and perpetuate it, and the



establishment of new group-generated norms toinsure the group’s existence. Harmony is of maxi-mum importance, and task conflicts are avoided toinsure harmony.

The third stage of task activity development islabeled as the open exchange of relevant interpreta-tions. In the therapy- and training-group context,this takes the form of discussing oneself and othergroup members, since self and other personalcharacteristics are the basic task inputs. In thelaboratory-task context, exchangedinterpretations take the form of opinions. In allcases one sees information being acted on so thatalternative interpretations of the information canbe arrived at. The openness to other groupmembers is characteristic in both realms duringthis stage.

The fourth and final developmental phase ofgroup structure is labeled as functionalrole-relatedness. The group, which was establishedas an entity during the preceding phase, can nowbecome a problem-solving instrument. It does thisby directing itself to members as objects, since thesubjective relationship between members hasalready been established. Members can now adoptand play roles that will enhance the task activitiesof the group, since they have learned to relate toone another as social entities in the precedingstage. Role structure is not an issue but aninstrument which can now be directed at the task.The group becomes a ‘sounding board’ off whichthe task is ‘played’.

In task-activity development, the fourth andfinal stage is identified as the emergence of solu-tions. It is here that we observe constructiveattempts at successful task completion. In thetherapy- and training-group context, these solu-tions are more specifically insight into personaland interpersonal processes and constructiveself-change, while in the laboratory-group contextthe solutions are more intellectual andimpersonal. Here, as in the three precedingstages, there is an essential correspondencebetween group structural and task realms overtime. In both realms, the emphasis is onconstructive action, and the realms come togetherso that energy previously invested in thestructural realm can be devoted to the task.

The next section presents a review of relevantstudies separated according to setting. The obser-vations within each study are separated accordingto stage of development and realm.

STAGES OF DEVELOPMENT IN THERAPY GROUPS

Stage 1

Group structure: testing and dependence.Of the twenty-six studies of development in

therapy groups which were reviewed, eighteenidentified a beginning stage as either testing ordependence or both. Bach (1954) speaks of initialsituation testing to determine the nature of thetherapy environment and discover the kinds ofrelationships the therapist will promote, followedclosely by leader dependence where group membersrelate to the therapist dependently. Barton (1953),Beukenkamp (1952) and Mann and Semrad (1948)identify an initial stage in which the group tests todetermine the limits of tolerance of the therapistand the group.

Researchers emphasizing the more dependentaspects of this initial stage are Bion (1961) whodescribes groups operating with the basicassumption of dependency, Cholden (1953) whohas observed dependency in therapy groups ofblind individuals, and Stoute (1950) who observeddependency in larger classroom therapy groups.

Others have observed this stage and haveused a variety of names to label it. Corsini (1957),in an integration of other studies, identifieshesitant participation as an initial stage, in whichmembers test the group and therapist to discoverhow they will respond to various statements.Grotjahn (1950) refers to an initial period oforientation and information, while King (1959)labels initial testing and orienting behavior inactivity-group therapy as acclimatization.Abrahams (1949) and Powdermaker and Frank(1948) describe the initial period as one oforientation and testing where group membersattempt to relate to the therapist and to discoverthe structure and limits of the therapy group.Schindler (1958), using bifocal-group therapy,labels the initial stage as attachment to the group,in which individuals discharge old ties and estab-lish new ones. Taylor (1950) talks about qualifyingfor acceptance by the group at the start of therapywhich implies both testing and conforming.

Four of the studies reviewed describe a stagepreceding the testing-dependence stage which willbe referred to as pre-stage 1. Thorpe and Smith(1953) and Osberg and Berliner (1956), in therapywith hospitalized narcotic addicts, describe aninitial stage of resistance, silence and hostility fol-lowed by a testing period where patients attemptto discover what behaviors the therapist deemsacceptable. Shellow, Ward and Rubenfeld (1958),who worked with institutionalized delinquents,described two such stages of resistance and hostil-ity preceding the testing stage, while Martin andHill (1957) theorized about a stage of isolation and‘unshared behavior’ preceding one of stereotypicresponding to fellow group members and adependent orientation toward the therapist.

Three of the four studies identifying apre-stage 1 were specifically based on observationsof groups of antisocial individuals (drug addicts

Tuckman


and delinquents) who probably must be won overto the situation and their initial extremeresistance overcome before the normal sequenceof therapy-group development can begin. Thiswould account for pre-stage 1.

The remaining studies did not identify aninitial stage of testing-dependence but dealteither with task development (to be discussedbelow), or offered an initial stage 1 which ispostulated here as a second stage. Finally, a studyby Parker (1958) described an initial stage ofcohesive organization in which sub-groups areformed, rules followed, and harmonymaintained—a description which is difficult to fitinto the testing-dependence category.

Task activity: orientation and testing. Dur-ing the initial stage, task development is charac-terized by indirect attempts to discover the natureand boundaries of the task, i.e., what is to beaccomplished and how much cooperation isdemanded, expressed specifically through (a) thediscussion of irrelevant and partially relevantissues (Bion, 1961; Coffey et al., 1950; Martin andHill, 1957; Osberg and Berliner, 1956), (b) the dis-cussion of peripheral problems (Stoute, 1950), (c)the discussion of immediate behavior problems(Abrahams, 1949), (d) the discussion of symptoms(Bach, 1954; Taylor, 1950), (e) griping about theinstitutional environment (Mann and Semrad,1948; Shellow, Ward and Rubenfeld, 1958; Thorpeand Smith, 1953), and (f) intellectualization(Clapham and Sclare, 1958; Wender, 1946).

This stage is also characterized by more directattempts at orientation toward the task as illus-trated in (a) a search for the meaning of therapy(Cholden, 1953), (b) attempts to define the situa-tion (Powdermaker and Frank, 1948), (c) attemptsto establish a proper therapeutic relationship withthe therapist through the development of rapportand confidence (Dreikurs, 1957; King, 1959; Wolf,1949), (d) mutual exchange of information(Grotjahn, 1950), and (e) suspicious of and fearful-ness toward the new situation which must beovercome (Corsini, 1957).

Stage 2

Group structure: intragroup conflict. Thir-teen of the twenty-six studies of group therapyreviewed identified a stage of intragroup conflict(in eleven cases as a second stage and in two as afirst stage). Abrahams (1949) identifies aninteraction stage typified by defensiveness,competition and jealousy. Bion (1961) discusses afight-flight period in which members conflict withthe therapist or attempt to psychologicallywithdraw from the situation. Grotjahn (1950)identifies a stage of increasing tension, whileParker (1958) talks about a crisis period wherefriction is increased, anxiety mounts, rules are

broken, arguments ensue, and a generalstructural collapse occurs. Powdermaker andFrank (1948) discuss a second stage featuringsharp fluctuation of relationships, sharp reversalsof feelings, and ‘intense but brief and brittle link-ages’. Schindler (1958) talks about a stage of psy-chodramatic acting-out and localization ofconflicts in the group, while Shellow, Ward andRubenfeld (1958) describe a stage characterized byambivalence toward the therapist which isexpressed through the formation of conflictingfactions in the group. Stoute (1950) describes asecond stage beginning with derogation andnegativity, while Thorpe and Smith (1953) describea stage beginning with disintegration, distance,defenses out of awareness and disruptedcommunication. King (1959), in activity-grouptherapy, describes a second stage of benignregression characterized by extreme acting-outand unacceptable behavior. Martin and Hill (1957)theorize about a stage of polarization featuringthe emergence of sub-groups following a stage ofinterpersonal exploration.

Coffey et al. (1950) identify an initial stage ofdefensiveness and resistance where membersclash with one another. However, these authorsalso see ‘pecking orders’ being established duringthis period; perhaps their initial stage includesstages 1 and 2 as postulated in this review. Mann(1953) describes an initial phase of ‘workingthrough of hostility,’ followed by a stage of ‘workingthrough of anxieties’. The hostility phase ischaracterized by disruption and fragmentationwhich are reduced gradually in the anxiety phase.

The remaining studies fail to identify thisstage. Some of them jump from stage 1 directly tostage 3, while others deal with task developmentas concerns the first two stages of therapy-groupdevelopment.

Task activity: emotional response to taskdemands. The outstanding feature of this secondtask stage appears to be the expression of emo-tionality by the group members as a form ofresisting the techniques of therapy which requirethat they ‘expose’ themselves and of challengingthe validity and usefulness of therapy (Bach, 1954;Barton, 1953; Cholden, 1953; Clapham and Sclare,1958; Mann, 1953; Mann and Semrad, 1948;Martin and Hill, 1957; Stoute, 1950; Wender1964). Furthermore, mention is made of the factthat this is a period of extreme resistance toexamination and disclosure (Abrahams, 1949;Barton, 1953), and an attempt at analysis of thisresistance is made (Wolf, 1949). Others emphasizeambivalence toward the therapist (Shellow, Wardand Rubenfeld, 1958), the discussion of sensitiveareas (Powdermaker and Frank, 1948), psycho-drama (Schindler, 1958) and resistance via ‘puttingone on’ (Thorpe and Smith, 1953).



Stage 3

Group structure: development of groupcohesion. Twenty-two of the twenty-six studiesreviewed identified a stage in which the groupbecame a cohesive unit and developed a sense ofbeing as a group. Bach (1954), Barton (1953) andClapham and Sclare (1958) identify a stage duringwhich in-group consciousness is developed andestablishment and maintenance of group bounda-ries is emphasized. Bion (1961) discusses the basicassumption of pairing in which the emphasis is oncohesion, but the unit is the pair as opposed tothe whole group. Coffey et al. (1950), and Taylor(1950) describe a stage following the stage ofintragroup hostility in which the group becomesunified and is characterized by the existence of acommon goal and group spirit. Parker (1958) andShellow, Ward and Rubenfeld (1958) see the stageof crisis and factions being followed by onefeaturing consensual group action, cooperation,and mutual support. Grotjahn (1950), Mann andSemrad (1948), and Powdermaker and Frank(1948) describe a third stage characterized bygroup integration and mutuality. Noyes (1953)describes a middle stage of group integration,while Stoute (1950) and Thorpe and Smith (1953)see the stage of intragroup hostility grading into aperiod of unity, support, and freedom ofcommunication. Martin and Hill (1957) theorizeabout a stage featuring awareness that the groupis an organism preceding the final stage ofdevelopment. Abrahams (1949) describes thedevelopment of ‘we-consciousness’ in the thirdstage, while Mann (1953) sees the third stage asone of personal mutual exploration and analysisduring which the group attains unity.

The notion that the group becomes a simula-tion of the family constellation (that is, throughtransference members react to one another asmembers of their family), with the unity and cohe-sion generally accepted in that structure, fits as aclose parallel to the stage of development of groupcohesion being postulated. Beukenkamp (1952)describes the middle stage of reliving the process ofthe family constellation where the group becomesa family-like structure, while King (1959) utilizes asimilar description (that is, family unity in thegroup) for the final stage in activity-group therapy.Wender (1946) and Wolf (1949) both describe astage preceding the final stage in which the groupbecomes the new family through the displacementof parent love.

Studies that fail to identify this stage are thosethat deal primarily with task development or thosethat integrate it as part of the final stage.

Task activity: discussing oneself and othergroup members. Many researchers observedprobing and revealing by group members at ahighly intimate level during this period and

labeled it as (a) confiding (Clapham and Sclare,1958; Coffey et al., 1950; Thorpe and Smith, 1953),(b) discussing personal problems in depth (Corsini,1957; Mann and Semrad, 1948; Osberg andBerliner, 1956; Taylor, 1950), (c) exploring thedynamics at work within the individual (Dreikurs,1957; Noyes, 1953), and (d) exploring the dynamicsat work within the group (Bach, 1954; Martin andHill, 1957; Powdermaker and Frank, 1948).

Beukenkamp (1952) observed that recalledmaterial was related to the family; Abrahams(1949) observed the process of common ideation;and Shellow, Ward and Rubenfeld (1958) and Wolf(1949) emphasized patients’ discussion of topicsrelated to transference to the therapist and toother group members which took place during thisperiod.

Stage 4

Group structure: functionalrole-relatedness. Only twelve of the therapystudies are at all explicit in their identification ofthis stage. Almost all of the therapists discuss thefinal stage of development of the therapy group intask terms as the therapeutic stage ofunderstanding, analysis and insight. The group isseen as serving a therapeutic function, but thenature of this therapeutic function is not spelledout. This is a stage of mutual task interaction witha minimum of emotional interference madepossible by the fact that the group as a socialentity has developed to the point where it cansupport rather than hinder task processesthrough the use of function-oriented roles.

Bach (1954) and Bion (1961) both refer to thegroup in its final stage as the work group. As suchit serves a function supportive of therapy.Abrahams (1949) and Wender (1946) see the groupas creating a therapeutic atmosphere in the finalstage, while Corsini (1951), Stoute (1950) and Wolf(1949) describe this stage as one of freedom andfriendliness supportive of insightful behavior andchange. Both Coffey et al. (1950) and Dreikurs(1957) see the group as a therapeutic force pro-ducing encouragement and integrating problemswith roles. Martin and Hill (1957) identify thegroup as an integrative-creative-social instrument inits final stage which facilitates problem solving,diagnosis, and decision making. Osberg andBerliner (1956) describe the self-starting stagewhere the group environment supports analysis,while Mann (1953) discusses a final stage of per-sonal mutual synthesis.

Other therapy researchers, failing to specifi-cally delineate this final stage in social develop-ment, have tended to lump the third and fourthstages together and not make the distinctionbetween the development of cohesion and the‘use’ of cohesion (via functional roles) as a

Tuckman


therapeutic force. Such descriptions wereincluded in the section on the third stage. Thesmall number of investigators identifying this finalstage is most likely due to the high visibility of taskfunctions occurring during this time period whichobscure and minimize social processes occurringsimultaneously.

Task activity: emergence of insight. Thereseems to be overwhelming agreement among theobservers of therapy-group development that thefinal stage of task development is characterized byattainment of the desired goal, insight into one’sown problems, an understanding of the cause ofone’s abnormal behavior and, in many cases,modification of oneself in the desired direction(Beukenkamp, 1952; Bion, 1961; Clapham andSclare, 1958; Coffey et al., 1950; Corsini, 1957;Dreikurs, 1957; King, 1959; Noyes, 1953;Schindler, 1958; Stoute, 1950; Thorpe and Smith,1953; Wender, 1946; Wolf, 1949). Others(Abrahams, 1949; Bach, 1954; Barton, 1953;Cholden, 1953; Grotjahn, 1950; Shellow, Ward,and Rubenfeld, 1958; Taylor, 1950) place moreemphasis on the processes of attempting todevelop insight and change during this last periodas opposed to the development of such insightand change itself.

Two additional therapy-group studies are wor-thy of inclusion, both of which utilized atechnique for collecting and analyzing data whichwas highly dissimilar to the approach used in theother therapy-group studies, namely,interaction-process analysis (Bales, 1950). Psathas(1960) found that groups phase from orientation toevaluation to control, based on an analysis of early,middle and late sessions. Talland (1955) failed toobserve this phase movement based on ananalysis of the first eight sessions.

STAGES OF DEVELOPMENT IN TRAINING GROUPS

Stage 1

Group structure: testing and dependence.Nine of the eleven training-group studies reviewedthat deal with the development of group structureidentify an initial stage characterized at least inpart by testing and dependence, with theemphasis on the dependent aspect of this stage.

Herbert and Trist (1953), Bennis and Shepard(1956), Bradford, (1964a), and Bradford andMallinson (1958) describe the initial group phaseas one characterized by the strong expression ofdependency needs by the members toward thetrainer, and attempts at group structuring to workout authority problems by the quick acceptance ofand dependence on such structure and arbitrarynorms. Thelen and Dickerman (1949) discuss ini-tial stage establishment of a leadership hierarchycatering to the dependency needs of the members.

Hearn (1957) sees group members making anattempt to structure the unknown and to findtheir position in the group in the earliest groupstage. Here again, structure reflects theexpression of dependency needs.

Miles (1953) describes a first stage character-ized by establishment of the situation throughinterpersonal exploration and testing, whileSemrad and Arsenian (1961) identify an initialphase during which group members ‘test’ the cen-tral figure and ‘test’ the situation.

Whitman (1964) describes a beginning stage inwhich the chief ‘vectors’ are dependency and hos-tility. It would appear that Whitman has identifieda first stage which combines the first two stagesproposed in this article.

The two studies that do not yield an exact fitto the proposed scheme are those of Barron andKrulee (1948) and the Tulane Studies in SocialWelfare (1957) which identify an initial periodcharacterized by the emergence of leadership andorientation, respectively. In so far as theseauthors see the authority area as being of centralconcern and emphasize the orientation aspects ofthe first stage, there is overlap with the schemeproposed herein. Moreover, orientation as a firststage fits the hypothesized initial stage for taskactivities; perhaps the observation in the Tulanestudies (1957) of a member orientation as aninitial stage is better classified in the task-activityarea.

Task activity: orientation. Bradford (1964b)identifies an initial stage of learning how to learnwhich is characterized by acceptance of thegroup’s goal and orientation to the techniques tobe used. Herbert and Trist (1953) label their initialstage as discovery, in which the members orientthemselves to the consultant or trainer whoserves an interpretive and educational role. Stockand Thelen (1958) discuss an initial stagecharacterized by little ‘work’ and a variableamount of ‘emotionality’, during which time themembers are concerned with defining thedirections the group will pursue.

As can be seen, initially interpersonalproblems are dealt with via dependence, whiletask problems are met with task-orientingbehavior (i.e., what is to be accomplished andhow).

Stage 2

Group structure: intragroup conflict. Tenof the eleven studies identify intragroup conflict asa second stage, while the remaining study(Whitman, 1964) describes an initial stageencompassing both dependence and hostility, inthat order.



Barron and Krulee (1948) and Bradford (1964a)discuss a second stage characterized by groupcleavage and conflict. Both studies identify theemergence of polarities during this stage—mem-bers favoring a more active, less defensiveapproach versus those who are more passive anddefensive and seek ‘safety’ via structure. Thelenand Dickerman (1949), Hearn (1957), the Tulanestudies (1957) and Bradford and Mallinson (1958),as well, identify a similar polarization and resul-tant conflict, frustration, and disruption duringthe second stage.

Herbert and Trist (1953) describe a secondstage characterized in part by resistance, whileMiles (1953) identifies anarchic rebellion duringthis stage of anxiety, threat, and resistance.Semrad and Arsenian (1961) identify rivalry for theposition of central figure and emotional strugglesin this period, while Bennis and Shepard (1956)see a similar power struggle in whichcounterdependents seek to usurp the leader,resulting in a conflict between counterdependentsand dependents.

There appears to be general agreement thatthe dependency stage is followed by a stage ofconflict between warring factions representingeach side of the polarized issue: dependenceversus independence, safe retreat into the familiarversus risky advance into the unfamiliar,defensiveness versus experimenting.

Task activity: emotional response to taskdemands. Bradford (1964b) identifies a secondstage in which individuals learn how to give helpwhich requires that they remove blocks tolearning about themselves, reduce anxiety, andexpress real reactions. Stock and Thelen (1958)see emotionality occurring in considerable excessof work during this period. The Tulane studies(1957) describe the second stage as one ofexperimental aggressiveness and hostility whereindividuals express themselves freely.

Thus, self-change and self-denial necessitatedby the learning task is reacted to emotionally, as isthe imposition of the group on the individual.Often the two (representative of the two realms)are difficult to separate.

Stage 3

Group structure: development of groupcohesion. All of the relevant T-groupdevelopment studies see the stage of conflict andpolarization as being followed by a stagecharacterized by the reduction of the conflict,resolution of the polarized issues, andestablishment of group harmony in the place ofdisruption. It is a ‘patching-up’ phase in whichgroup norms and values emerge.

Hearn (1957), Miles (1953), and Thelen andDickerman (1949) identify a third stage character-ized by attempts to resolve conflict and the conse-quent development of group cohesion and mutualsupport. Semrad and Arsenian (1961) and theTulane studies (1957) each describe two phases intheir temporal sequences which would beincluded in stage 3. In the case of the former, theirfirst cohesion phase is characterized by groupcohesion processes and their second by thedevelopment of affection bonds; in the latter, thefirst cohesion stage features the emergence ofstructure, roles and ‘we-feelings’, while the secondfeatures increased group identification on aconscious level and vacillation in role acceptance.Whitman (1964) talks about a middle phase,following conflict, described as the development ofa new group culture via the generation of normsand values peculiar to the group as an entity.Bradford and Mallinson (1958) describe stage 3 asone of reorganization, in which reforming andrepair take place and a flexible organizationemerges.

Bradford (1964a) describes a third stage inwhich the group norm of ‘openness’ emerges, anda fourth stage in which the group generates addi-tional norms to deal with self-revelation and feed-back. Furthermore, Bradford (1964b) identifies athird stage as one of developing a group climate ofpermissiveness, emotional support and cohesive-ness in which learning can take place. This de-scription would appear to subserve both interper-sonal and task realms.

Bennis and Shepard (1956) describe a thirdstage in which resolution of authority problemsoccurs, and a fourth stage characterized bysmooth relations and enchantment as regards theinterpersonal sphere of group functioning. Finally,Barron and Krulee (1948) identify the third stageas increasing member responsibility and changingfaculty role in which a definite sense of structureand goal orientation emerge in the group.

Task activity: discussing oneself andothers. Herbert and Trist (1953) identify a secondstage labeled as execution, in which the groupsettles down to the description of a single basicproblem and learns to accept ‘the examination ofwhat was going on inside of itself as a regular partof the task ....’ Stock and Thelen (1958) describe athird task phase in which the group shows a newability to express feelings constructively andcreatively. While emotionality is still high, it nowcontributes to work.

While the social function of the third stage isto cause a unique and cohesive group structure toemerge, the task function is to attempt to use thisnew structure as a vehicle for discovering personalrelations and emotions by communicating hereto-fore private feelings.

Tuckman


Stage 4

Group structure: functionalrole-relatedness. There is some tendency forT-groupers, as there was for the therapy groupers,to emphasize the task aspects of the final stage,namely, the emergence of insight into theinterpersonal process. In doing this, it is madeimplicit that the group as a social entitycharacterized by task-oriented role-relatednessmakes the emergence of such insight possible byproviding support and an opportunity forexperimentation and discovery.

Bradford (1964a) sees the group becoming awork organization which provides membersupport, mutual acceptance, and has strong butflexible norms. Hearn (1957) discusses mutualacceptance and use of differences in thecollaborative process during the fourth and fifthgroup stages, while Miles (1953) sees groupstructure as tending ‘to be functional and notloved for itself alone’ as it was in the precedingstage. The support function is further emphasizedby Miles when he says (p. 94),

in groups where the interpersonal bondsare genuine and strong ... members giveone another a great deal of mutual evalua-tive support, which seems to be a primerequisite for successful behavior change.

Semrad and Arsenian (1961) describe a finalphase of productive collaboration, while Thelenand Dickerman (1949) identify the group as aneffective social instrument during this period.Barron and Krulee (1948) see, as one groupfunction occurring during the final two meetings,the sharing and refining of feelings through thegroup process.

Bennis and Shepard (1956) see the stage ofgroup cohesion being followed by another periodof conflict, in which the issue is intimate socialrelations versus aloofness. The final stage is thenone of consensual validation in which groupinterpersonal problems are solved and the group isfreed to function as a problem-solving instrument.

The Tulane studies (1957) describe the stagefollowing the emergence of cohesion as one inwhich behavior roles become dynamic, that is,behavior is changed as a function of the accep-tance of group structure. An additional stage isalso identified in this study in which structure isinstitutionalized by the group and thus becomesrigid. Perhaps this stage, not identified by otherresearchers, would most apply to groups with along or indefinite group life.

The remaining T-group studies describe taskdevelopment exclusively during the final groupphase.

Task activity: insight. Bradford's (1964b)fourth phase is one in which members discoverand utilize various methods of inquiry as ways ofgroup development and individual growth while, inhis fifth and final stage, members learn how tointernalize, generalize and apply learnings toother situations. Herbert and Trist (1953) labeltheir final stage as evaluation. Stock and Thelen(1958) describe the fourth and final stage as onecharacterized by a high degree of work in theabsence of affect. The issues are dealt with in aless excited way.

The over-all fit between stages of developmentpostulated in this paper for application in all set-tings and those delineated by T-groupers is high-lighted in the fourfold scheme presented byGolembiewski (1962), based on his examination ofsome T-group development studies alreadyreviewed in this paper. Golembiewski describes hisstages as: (a) establishing the hierarchy, (b)conflict and frustratio, (c) growth of group securityand autonomy, (d) structuring in terms of problemsfacing the group rather than in terms ofstereotypic role prescriptions.

STAGES OF DEVELOPMENT IN NATURAL ANDLABORATORY GROUPS

Few studies or theoretical statements haveconcerned themselves with the developmentalsequence in natural groups or laboratory groups.

Stage 1

Group structure: testing and dependence.Modlin and Faris (1956), studying an interdiscipli-nary professional group, identify an initial stage ofstructuralization, in which members aredependent upon roles developed outside of thegroup, well-established traditions, and a fixedhierarchy of responsibility.

Schroder and Harvey (1963) describe an initialstage of absolutistic dependency, featuring theemergence of a status hierarchy and rigid normswhich reduce ambiguity and foster dependenceand submission.

Theodorson (1953) observed a tendencyinitially for only one leader to emerge and forgroup members to categorize one another so thatthey could define the situation and reduceambiguity.

Schutz (1958)1 sees the group dealing initiallywith problems of inclusion—to join or not to join,to commit oneself or not. The group concern,thus, is boundary problems, and the behavior ofmembers is individually centered. This descriptionis somewhat suggestive of testing.

Task activity: orientation. Bales (1953) andBales and Strodtbeck (1951), using Bales’ (1950)



interaction-process categories, discovered thatleaderless laboratory groups begin by placing majoremphasis on problems of orientation (as reflectedin Bales’ categories: ‘asks for orientation’ and‘gives orientation’). This orientation serves todefine the boundaries of the task (i.e., what is tobe done) and the approach that is to be used indealing with the task (i.e., how it is to be accom-plished).

Stage 2

Group structure: intragroup hostility.Modlin and Faris (1956) describe unrestcharacterized by friction and disharmony as thesecond stage, while Schroder and Harvey (1963)identify a second stage of negative independencefeaturing rebellion, opposition and conflict. In thisstage, the greater emphasis is on autonomy andindividual rights. Theodorson (1953) observedmore friction, disharmony, and animosity early inthe group life than during later periods.

Schutz (1958) postulates a second stage inwhich the group deals with problems of control.This entails a leadership struggle in which individ-ual members compete to establish their place inthe hierarchy culminating in resolution.

In the task area, the stage of emotionalresponse to task demands is not delineated, pre-sumably due to the impersonal and nonthreaten-ing nature of the task in these settings. When thetask does not deal with the self at a penetratinglevel, extreme emotionality in the task area is notexpected.

Stage 3

Group structure: development of GroupCohesion. Modlin and Faris (1956) identify changeas the third stage, characterized by the formationof the concept of the group as a functioning unitand the emergence of a team ‘dialect’. Schroderand Harvey (1963) refer to stage 3 as conditionaldependence, featuring a group concern with inte-gration and an emphasis on mutuality and themaintenance of interpersonal relationships.

Theodorson (1953) observed the followinggroup tendencies over time (i.e., tending to occurlater as opposed to earlier in group development):(a) discovering what is common to the membersand developing a within-group ‘parochialism’; (b)the growth of an interlocking network offriendship; (c) role interdependence; (d) mutualinvolvement and identification between memberswith a concomitant increase in harmony andsolidarity; and (e) the establishment of groupnorms for dealing with such areas as discipline.

Schutz (1958) postulated a third stage whereinproblems of affectation are dealt with. Characteris-

tic of this stage are emotional integration, pairing,and the resolution of intimacy problems.

Task activity: expression of opinions. Bales(1953) and Bales and Strodtbeck (1951) observedthat the orientation phase was followed by aperiod in which major emphasis was placed onproblems of evaluation (as reflected by categories:‘asks for opinion’ and ‘gives opinion’). ‘Evaluation’as a descriptor of the exchange of opinionsappears to be comparable to the third task stagein therapy- and training-group development whichwas heretofore labeled as ‘discussing oneself andothers’. Because the therapy and training tasksare personal ones, task opinions must involve selfand others. When the task is an impersonal one,the content of task opinions varies accordingly.

Stage 4

Group structure: functionalrole-relatedness. Modlin and Faris (1956) identifyintegration as the fourth and final stage in whichstructure is internalized and the group philosophybecomes pragmatic, that is, the unified-groupapproach is applied to the task.

Schroder and Harvey (1963) postulate a finalstage of positive interdependence, characterized bysimultaneous autonomy and mutuality (i.e., themembers can operate in any combination, or as aunit), and an emphasis on task achievementwhich is superordinate to social structure.

Theodorson (1953) sees the group as develop-ing into a subculture over time, along with thedevelopment of member responsibility to thegroup.

Schutz (1958) does not identify a fourth stage;rather, he sees his three postulated stages con-tinually cycling over time.

Task activity: emergence of solution. Thethird and final phase observed by Bales (1953) andBales and Strodtbeck (1951) is one in which majoremphasis is placed on problems of control (asreflected by categories: ‘asks for suggestion’ and‘gives suggestion’). The purpose of suggestions isto offer solutions to the task based on informationgathered and evaluated in previous developmentalperiods. This then represents an analogue of finalstages in therapy- and training-group task devel-opment where the emergence of insight yieldssolutions to personal problems.

These authors do not identify a period of taskdevelopment in laboratory groups comparable tothe second task stage in therapy- and train-ing-group development which features the expres-sion of emotional material. Again, because therapyand training tasks are personal ones, this will bereflected in the content of discussion, specificallyby the manifestation of resistance prior to dealing

Tuckman


with the personal task at a level of confidence andhonesty. This task stage does not appear to bequite relevant in laboratory discussion groups, andits existence has not been reported by Bales(1953) and Bales and Strodtbeck (1951).

Philp and Dunphy (1959) have further sub-stantiated the findings of Bales (1953) and Balesand Strodtbeck (1951) by observing the samephase-movement pattern in groups working on adifferent type of discussion problem.2

Furthermore, Philp and Dunphy (1959) presentevidence which indicates that sex of theparticipants does not affect the pattern of phasemovements.

Finally, Smith (1960) has observed thatexperimental groups show early concentration onmatters not related to the task, and, only later inthe development sequence, concentrate ontask-relevant activities. Again, this finding sug-gests a strong similarity between taskdevelopment in laboratory groups and in therapyand training groups, since, in the latter settings,constructive task-relevant activity appears onlylate in the developmental sequence.

DISCUSSION

The literature that has been reviewed can becriticized on a number of grounds. First, it may bepointed out that this literature cannot be consid-ered truly representative of small-group develop-mental processes, since certain settings havebeen over-represented, primarily thetherapy-group setting, and othersunder-represented, primarily the natural-groupand laboratory-group settings. This shortcomingcannot be rectified within the existing literature;rather, it must serve as a stimulus for furtherresearch in the latter group settings. Furthermore,the inequality of setting representationnecessitates caution in generalizing from this lit-erature. Generalization must, perforce, be limitedto the fact that what has been presented is mainlyresearch dealing with sequential development intherapy groups.

A second source of criticism concerns theextent of experimental rigor characteristic of themajority of studies cited in this review. Most of thestudies carried out in the therapy-group, train-ing-group and natural-group settings are based onthe observation of single groups. Furthermore,these observations are qualitative rather thanquantitative, and as such are subject to the biasesof the observer, ordinarily the therapist or trainer.This is not to suggest that the therapy-group set-ting is not appropriate for studying group proc-esses, but that the study of such processes shouldbe more subject to methodological considerations.A good instance of the application of such consid-erations is the study of Psathas (1960) conducted

in the therapy-group setting. Psathas coded groupprotocols using Bales’ (1950) scheme of interac-tion-process analysis. After satisfactory reliabilitieswere obtained, the data could be considered ashighly quantitative and objective, and could thenbe subjected to statistical analysis. Approaches ofequal rigor are recommended for other studiesconducted in the therapy-group setting and othersettings as well.

A final criticism concerns the description andcontrol of independent variables. Since most ofthe studies in the therapy-, training- and natu-ral-group settings used a single group, the controland systematic manipulation of independent vari-ables was impossible. In the absence of themanipulation of independent variables and theconsequent discovery of their differential effectswithin studies, these effects can only be approxi-mately discerned by comparing studies. However,many independent variables are likely to vary fromstudy to study, for example, group composition,duration, etc., and little light will be shed on theeffects of these variables on the developmentalprocess. Therefore, no conclusions about the spe-cific effects of independent variables on develop-mental phenomena will be drawn, and furtherwork along these lines is encouraged.

In order to isolate those concepts common tothe various studies reviewed (across settings), adevelopmental model was proposed. This modelwas aimed at serving a conceptual function as wellas an integrative and organizational one. Themodel will be summarized here.

Groups initially concern themselves with ori-entation accomplished primarily through testing.Such testing serves to identify the boundaries ofboth interpersonal and task behaviors. Coincidentwith testing in the interpersonal realm is theestablishment of dependency relationships withleaders, other group members, or pre-existingstandards. It may be said that orientation, testingand dependence constitute the group process offorming.

The second point in the sequence ischaracterized by conflict and polarization aroundinterpersonal issues, with concomitant emotionalresponding in the task sphere. These behaviorsserve as resistance to group influence and taskrequirements and may be labeled as storming.

Resistance is overcome in the third stage inwhich ingroup feeling and cohesiveness develop,new standards evolve, and new roles are adopted.In the task realm, intimate, personal opinions areexpressed. Thus, we have the stage of norming.

Finally, the group attains the fourth and finalstage in which interpersonal structure becomesthe tool of task activities. Roles become flexibleand functional, and group energy is channeled



into the task. Structural issues have beenresolved, and structure can now becomesupportive of task performance. This stage can belabeled as performing.

Although the model was largely induced fromthe literature, it would seem to withstand the testof common sense as well as being consistent withdevelopmental theory and findings in other areas.It is not unreasonable to expect ‘newness’ of thegroup-to-be greeted by orienting behavior andresultant unsureness and insecurity overcomethrough dependence on an authority figure, asproposed in the model. Such orienting responsesand dependence on authority are characteristic ofthe infant during the first year (Ilg and Ames,1955), the young child when first apprehendingrules (Piaget, 1932), and the patient when firstentering psychotherapy (Rotter, 1954).

After the ‘newness’ of the group has ‘worn off’,the members react to both the imposition of thegroup and the task emotionally and negatively,and pose a threat to further development. Thisproposal is mirrored by the rebelliousness of theyoung child following his ‘obedient’ stages (Ilg andAmes, 1955; Levy, 1955).

Such emotionality, if overcome, is followed by asense of ‘pulling together’ in the group and beingmore sensitive to one another. This sensitivity toothers is mirrored in the development of the child(Ilg and Ames, 1955; Piaget, 1932) and representsan essential aspect of the socialization process(Mead, 1934).

Finally, the group becomes a functionalinstrument for dealing with the task.Interpersonal problems lie in the group’s ‘past’,and its present can be devoted to realisticappraisal of and attempt at solutions to the taskat hand. This interdependence and ‘marriage toreality’ is characteristic of the ‘mature’ humanbeing (Erikson, 1950; Fromm, l941) and the‘mature’ nine-year-old child (Ilg and Ames, 1955).3

The suggested stages of group developmentare highly visible in the literature reviewed. The fitis not perfect, however. Some of the studiesidentify some, but not all, of the suggested stages.In some of these cases, two of the suggestedstages have been welded into one by the observer.For instance, Barton (1953) describes three stages;the first and second fit the first two conceptualstages closely, while Barton's third stage isdescriptive of the third and fourth conceptualstages in so far as it is characterized by both theemergence of cohesiveness and the workingthrough of problems. In other cases, one or moreof the hypothesized stages have been clearlymissing, and thus not recognized in the group orgroups being observed. For instance, Powdermakerand Frank (1948) identify three stages that fit thefirst three conceptual stages fairly closely, but they

do not identify any fourth stage. Perhaps caseslike this can be accounted for on the basis ofindependent variables such as duration of grouplife.

A few studies identify more than four stages.Some of these additional stages represent agreater degree of differentiation than that of themodel and are of less generality (i.e., highlyspecific to the independent conditions of thestudy). For instance, therapy-group studies withdelinquents and dope addicts identify a stage priorto conceptual stage 1 in which the antisocialgroup member must be won over to the pointwhere they will take the therapy seriously.

Some of the studies identify a stage that isclearly not in the model. Parker (1958) describes afirst stage of cohesive organization. Thisdivergence from the model may reflect a differentway of describing much the same thing or mayreflect an unusual set of independent conditions.Parker was observing a ward population of about25, rather than a small weekly therapy group. Itmay be that the hypothesized first stage issomewhat inappropriate for larger, living-togethergroups.

While the suggested sequence appeared tohold up under widely varied conditions of groupcomposition, duration of group life and specificgroup task (i.e., the sequence held up acrosssettings), it must be assumed that there is a finiterange of conditions beyond which the sequence ofdevelopment is altered, and that the studiesreviewed did not exceed this assumed range toany great extent. Setting-specific differences andwithin-setting differences may affect temporalchange as regards the specific content of thestages in the developmental sequence, the rate ofprogression through the sequence, or the order ofthe sequence itself. In the therapy-group setting,for instance, task information in the third stage isconsiderably more intimate than it is in thelaboratory-group setting, and this stage may beattained at a later chronological time in therapygroups than in laboratory groups.

Certainly duration of group life would beexpected to influence amount and rate of develop-ment. The laboratory groups, such as those runfor a few hours by Bales and Strodtbeck (1951), fol-lowed essentially the same course of developmentas did therapy groups run for a period of a year.The relatively short life of the laboratory groupimposes the requirement that the problem-solvingstage be reached quickly, while no suchimposition exists for the long-lived therapy group.Consequently, the former groups are forced todevelop at a rapid rate. The possibility of suchrapid development is aided by the impersonal andconcrete nature of the laboratory task.Orientation is still required due to the newness of

Tuckman


the task but is minimized by task rules, players’manuals and the like, that help to orient thegroup members. Emotionality and resistance aremajor features of therapy-group development andrepresent personal and interpersonalimpediments to group development and solutionattainment as a function of the highlyemotionally-charged nature of the therapy-grouptask. The impersonal laboratory task features nosuch impediments and, consequently, the stage ofemotionality is absent. The exchange of relevantinformation is as necessary to the laboratory taskas it is to the therapy task, but the information tobe exchanged is limited in the laboratory task bythe nature of the task and time considerations.The behavior of ‘norming’ is common to bothsettings, but not so salient in the laboratorywhere the situation is so task-oriented. Finally,the problem-solving or ‘performing’ stage is anessential stage in both settings.

One would expect the laboratory group tospend relatively more time in the fourth stagerelative to the first three stages because of thetask orientation in the laboratory setting. In thetherapy task, with its unavoidable deepinterpersonal penetration, we would expectrelatively equal time to be spent in each stage.This, however, can undoubtedly be furthermodified by group composition as well as by the

duration of group life and specific nature of thelaboratory task. Undoubtedly there is aninteraction between setting and developmentsuch that the sequence proposed here will bealtered.

Unfortunately, the above hypotheses cannotbe substantiated with available data, thoughcertain of the studies are suggestive of theexplanations offered. The articles reviewed do notdeal with rate of temporal change nor do they givesufficiently complete and detailed time dataassociated with each stage to make calculations ofrate possible. Furthermore, they do notsystematically describe their independentvariables nor relate them to the developmentalphenomena through systematic variation and theobservation of cause and effect. The major task ofsystematically studying the effects of a variety ofappropriate independent variables ondevelopment still remains. The value of theproposed model is that it represents a frameworkof generic temporal change within which theabove explorations can be tested and whichshould lead to the derivation of many specifichypotheses relating independent variables to thesequence of temporal change. Such quantitativeexplorations will undoubtedly lead to refinementsand, perhaps, major modifications of such a model.

ReferencesAbrahams, J. (1949), Group psychotherapy: impli-

cations for direction and supervision of mentallyill patients, in Theresa Muller (ed.), MentalHealth in Nursing, Catholic University Press,77-83.

Bach, G. R. (1954), Intensive Group Psychotherapy,Ronald Press, 268-93.

Bales, R. F. (1950), Interaction Process Analysis: AMethod for the Study of Small Groups,Addison-Wesley.

Bales, R. F. (1953), The equilibrium problem insmall groups, in T. Parsons, R. F. Bales and E. A.Shils (eds.), Working Papers in the Theory ofAction, Free Press, 111-61.

Bales, R. F., and Strodtbeck, F. L. (1951), Phases ingroup problem-solving, Journal of Abnormal andSocial Psychology, 46, 485-95.

Barron, M. E., and Krulee, G. K. (1948), Case studyof a basic skill training group, Journal of SocialIssues, 4, 10-30.

Barton, W. E. (1953), Group psychotherapy of thepsychoses, Digest of Neurology and Psychiatry,21, 148-9.

Bennis, W. G., and Shepard, H. A. (1956), A theoryof group development, Human Relations, 9,415-37.

Beukenkamp, C. (1952), Some observations madeduring group therapy, Psychiatric Quarterly.Supplement, 26, 22-26.

Bion, W. R. (1961), Experience in Groups, BasicBooks.

Bradford, L. P. (1964a), Trainer-intervention: caseepisodes, in L. P. Bradford, J. R. Gibb and K. D.Benne (eds.), T-Group Theory and LaboratoryMethod, Wiley, 136-67.

Bradford, L. P. (1964b), Membership and thelearning processes, in L. P. Bradford, J. R. Gibband K. D. Benne (eds.), T-Group Theory andLaboratory Method, Wiley, 190-215.

Bradford, L. P., and Mallinson, T. (1958), Groupformation and development, in Dynamics ofGroup Life, National Education Association,National Training Laboratories, Washington.

Cholden, L. (1953), Group therapy with the blind,Group Psychotherapy, 6, 21-9.

Clapham, H. I., and Sclare, A. B., (1958), Grouppsychotherapy with asthmatic patients, Interna-tional Journal of Group Psychotherapy, 8, 44-54.

Coffey, H. S. (1952), Socio and psyche group proc-ess: integrative concepts, Journal of SocialIssues, 8, 65-74.

Coffey, H., Freedman, M., Leary, T., and Ossorio, A.(1950), Community service and social research -group psychotherapy in a church program,Journal of Social Issues, 6 (1), 14-61.

Corsini, R. J. (195l), On the theory of changesresulting from group therapy, Group Psycho-therapy, 4, 179-80.

Corsini, R. J. (1957), Methods of Group Psychother-apy, McGraw-Hill, 119-20.



Deutsch, M. A. (1949), A theory of cooperation andcompetition, Human Relations, 2, 129-52.

Dreikurs, R. (1957), Group psychotherapy from thepoint of view of Adlerian psychology, Interna-tional Journal of Group Psychotherapy, 7, 363-75.

Erikson, E. H. (1950), Childhood and Society,Norton, 213-20.

Fromm, E. (1941), Escape from Freedom, Holt,Rinehart & Winston.

Golembiewski, R.T. (1962), The Small Group,University of Chicago Press, 193-200.

Grotjahn, M. (1950), The process of maturation ingroup psychotherapy and in the group therapist,Psychiatry, 13, 63-7.

Harvey, O. J., Hunt, D. E., and Schroder, H. M.(1961), Conceptual Systems and PersonalityOrganization, Wiley.

Hearn, G. (1957), The process of group develop-ment, Autonomous Groups Bulletin, 13, 1-7.

Herbert, E. L., and Trist, E. L. (1953), The institu-tion of an absent leader by a student’s discus-sion group, Human Relations, 6, 215-48.

Ilg, F. L., and Ames, L. B. (1955), Child Behavior,Harper & Row.

Jennings, H. H. (1947), Sociometric differentiationof the psychegroup and sociogroup, Sociometry,10, 71-9.

King, C. H. (1959), Activity group therapy with aschizophrenic boy follow-up two years later,International Journal of Group Psychotherapy, 9,184-94.

Levy, D. M. (1955), Oppositional syndromes andoppositional behavior, in P. H. Hoch and J.Zubin (eds.), Psychopathology of Childhood,Grune & Stratton, 204-26.

Mann, J. (1953), Group therapy with adults,American Journal of Orthopsychiatry, 23, 332-7.

Mann, J., and Semrad, E. V. (1948), The use ofgroup therapy in psychoses, Journal of SocialCasework, 29, 176-81.

Martin, E. A., and Hill, W. F. (1957), Toward a the-ory of group development: six phases of therapygroup development, International Journal ofGroup Psychotherapy, 7, 20-30.

Mead, G. H. (1934), Self, Mind and Society, Univer-sity of Chicago Press.

Miles, M. B. (1953), Human relations training: howa group grows, Teachers College Record, 55,90-96.

Modlin, H. C., and Faris, M. (1956), Group adapta-tion and integration in psychiatric team prac-tice, Psychiatry, 19, 97-103.

Noyes, A. P. (1953), Modern Clinical Psychiatry,Saunders, 4th edition, 589-91.

Osberg, J. W., and Berliner, A. K. (1956), Thedevelopmental stages in group psychotherapywith hospitalized narcotic addicts, InternationalJournal of Group Psychotherapy, 6, 436-47.

Parker, S. (1958), Leadership patterns in a psychi-atric ward, Human Relations, 11, 287-301.

Philp, H., and Dunphy, D. (1959), Developmentaltrends in small groups, Sociometry, 22, 162-74.

Piaget, J. (1932), The Moral Judgment of the Child,Harcourt, Brace & World.

Powdermaker, F., and Frank, J. D., (1948), Grouppsychotherapy with neurotics, American Journalof Psychiatry, 105, 449-55.

Psathas, G. (1960), Phase movement and equilib-rium tendencies in interaction process in psy-chotherapy groups, Sociometry, 23, 177-94.

Rotter, J. B. (1954), Social Learning and ClinicalPsychology, Prentice-Hall.

Schindler, R. (1958), Bifocal group therapy, in J.Masserman and J. E. Moreno (eds.), Progress inPsychotherapy, 3, Grune & Stratton, 176-86.

Schroder, H. M., and Harvey, O. J. (1963), Con-ceptual organization and group structure, inO.J. Harvey (ed.), Motivation and Social Interac-tion, Ronald Press, 134-66.

Schutz, W. C. (1958), FIRO: A Three-DimensionalTheory of Interpersonal Behavior, Holt, Rinehart& Winston, 168-88.

Semrad, E. V., and Arsenian, J. (1961), The use ofgroup processes in teaching group dynamics, inW. G. Bennis, K. D. Benne and R. Chin (eds.),The Planning of Change, Holt, Rinehart &Winston, 737-43.

Shellow, R. S., Ward, J. L., and Rubenfeld, S.(1958), Group therapy and the institutionalizeddelinquent, International Journal of GroupPsychotherapy, 8, 265-75.

Smith, A. J. (1960), A developmental study ofgroup processes, Journal of Genetic Psychology,97, 29-39.

Stock, D., and Thelen, H. A. (1958), EmotionalDynamics and Group Culture, National EducationAssociation, National Training Laboratories,Washington.

Stoute, A. (1950), Implementation of group inter-personal relationships through psychotherapy,Journal of Psychology, 30, 145-56.

Talland, G. A. (1955),Task and interaction process:some characteristics of therapeutic group dis-cussion, Journal of Abnormal and SocialPsychology, 50, 105-9.

Taylor, F. K. (1950), The therapeutic factors ofgroup-analytic treatment, Journal of MentalScience, 96, 976-97.

Thelen, H., and Dickerman, W. (1949), Stereotypesand the growth of groups, Educational Leader-ship, 6, 309-16.

Theodorson, G. A. (1953), Elements in the progres-sive development of small groups, Social Forces,31, 311-320.

Thorpe, J. J., and Smith, B. (1953), Phases ingroup development in treatment of drug addicts,International Journal of Group Psychotherapy, 3,66-78.

Tulane University, (1957), The Use of Group Meth-ods in Social Welfare Settings, Tulane UniversitySchool of Social Work.

Tuckman


Wender, L. (1946), The dynamics of group psycho-therapy and its application, Journal of Nervousand Mental Disease, 84, 54-60.

Whitman, R. M. (1964), Psychodynamic principlesunderlying T-group processes, in L. P. Bradford,J. R. Gibb and K. Benne (eds.), T-Group Theoryand Laboratory Methods, Wiley, 310-35.

Wolf, A. (1949), The psychoanalysis of groups,American Journal of Psychotherapy, 3, 16-50.

Notes1. The classification of Schutz’s theory as one primarily descriptive of natural and laboratory groups is

arbitrary. Some would argue that Schutz is working in the T-group tradition.2. As mentioned earlier, Psathas (1960), working with therapy groups, observed the same phase

movement, namely, orientation to evaluation to control. However, Talland (1955) failed to get thisphase movement in therapy groups.

3. A more detailed model of individual development (similar to the group model proposed here), alongwith many citations of supporting literature, may be found in Harvey, Hunt and Schroder (1961).

Bruce Tuckman is Professor of Education at the Ohio State University College of Education. His currentresearch focuses on motivation and educational achievement. He has served on the education facultiesat Rutgers University, the City University of New York, and Florida State University. His major booksinclude Conducting Educational Research, Theories and Applications of Educational Psychology, andEvaluating Instructional Programs. Professor Tuckman was kind enough to provide us with his personalcomments about this oft-cited article nearly thirty-five years since its original publication.

Book Reviews


The Dance of Change:The Challenges to Sustaining Momentum in Learning Organizations

by Peter Senge, Art Kleiner, Charlotte Roberts,Richard Ross, George Roth and Bryan Smith

Doubleday, New York, 1999, 596 pages. ISBN 0-385-49322-3, US $35.00.

Book Review by Patricia R. Tuecke

Why is it so difficult to sustain change? Overthe past two decades, organizations haveembarked upon change journeys to meet the chal-lenges of increased global competition, new mar-kets, and pace of the technological development.In their attempts to respond quickly to externalchanges in the environment, most have failed,even after some initial success. Ten years ago,Peter Senge wrote The Fifth Discipline: The Art &Practice of The Learning Organization, a seminalbook in the field of organizational change andsystems theory. In 1994 Senge and five otherauthors wrote a follow-up book, The FifthDiscipline Fieldbook: Strategies and Tools forBuilding a Learning Organization. This book wasfilled with practical ways to initiate new ways ofthinking in organizations. After its publication, asthey met and talked with people, the six authorsbecame increasingly aware of the difficultiespeople were having in sustaining organizationalchange.

The Dance of Change: The Challenges to Sus-taining Momentum in Learning Organizations is theauthors’ response to these difficulties. A basicpremise of this book is that organizations areproducts of the ways that people in them thinkand act. Organizational learning results fromindividuals participating in activities that embodynew ways of thinking and acting and relatingtogether, leading to an increasing and enduringorganizational capacity for change. The book grewout of conversations at the MIT Center forOrganizational Learning among leaders involvedin change efforts. They were responding to thequestion: “What forces seemed to propelorganizational learning efforts forward or to slowthem down?”

This book is written for those involved inchange initiatives, managers, organization execu-tives, directors, consultants, and facilitators, tohelp them make sense of the organization’sresponse to change. It is written from a systemsperspective. In the sense that to facilitate is ‘tomake easier,’ then understanding the authors’framework, and utilizing their models and sugges-tions to meet the challenges of change would helpfacilitate a change effort. Facilitators working withorganizations going through profound change willfind this book a major resource. Anyone reading itwill gain theoretical understanding of the systemic

dynamics at work in change initiatives, and surelyrecognize some of the challenges in their ownexperiences. The thought provoking questionsand exercises can be utilized in many groupsituations.

Like its predecessor, The Fifth Discipline Field-book, this book is written as a collection of “notesfrom the field.” The book is structured around thethree major phases of change: 1) initiating change,2) sustaining momentum, and 3) redesigning orredirecting the organization. The myriad personalstories it contains illustrate what often happensin a system undergoing change, and in meetingchallenges of sustaining change inherent in eachphase. Perhaps the most compelling aspect of thebook is the stories, gleaned from conversationsand interviews, relating experiences of changeefforts. The stories are written in a first-personstyle. The leaders openly share theirvulnerabilities and the mistakes they made. Theyrelate the excitement, the frustration, theconfusion, the surprise, and the satisfaction oftheir experiences. They tell of their struggle tounderstand what was going on, of theorganizational resistance they met, the tacticsthey used, and the creativity catalyzed in meetingthe challenges of leading their organizationalchange efforts.

The Dance of Change is the authors’ attempt tomake sense of change efforts, both the failuresand the successes, and to create a framework thatholds the dynamics of the change journey. Theybring into a coherent whole the fragments ofexperience from these stories recounted by organ-izational change leaders. The Dance of Change isseen by its authors as an atlas of organizationalchange. They tell how early mapmakers drewmaps based on the tales told by early explorers,descriptions of the terrain and the coastlines, thedangers experienced, the discoveries made. Later,early atlas makers gathered the rough-drawncharts, notes, and experiences of early explorersinto bound collections. Later still, GeradusMercator, with longitudinal and latitudinal gridlines created a framework of the earth as a whole,an image that is very familiar to us. In a similarway, the six authors of this book describe theterrain, the dangers and discoveries for changeagents to follow. They offer their framework as asimple grid that allows the reader to put into

Book Reviews


perspective the diverse experiences oforganizational change.

The Dance of Change is not just a collection ofintriguing stories. Equally compelling are theimages, frameworks, models, and definitions ofaspects of the change journey that the storiesillustrate. Here are a few of these key concepts.

Profound Change: The authors consciouslychose the phrase “profound change” to describethe scope and quality of the change this bookdeals with. They rejected the more ubiquitouslabel of “transformation” because they fear it hascome to mean “really big change,” which is muchless than the change they are attempting todescribe. The authors describe a fundamental,profound change incorporating both an internalshift in people’s values, aspirations, andbehaviors, and external changes in thefundamental thinking patterns of organizationsthat underlie organizational choices of strategy,structures, and systems.

Think Like Biologists: A major premise ofthe authors is that more expert advice or morecommitted managers can’t solve the problems thatface organizations today. These problems havebecome the targets of “flavor of the month” changeprograms. Senge urges us to think less likemanagers and more like biologists! All growth innature arises out of interplay between reinforcinggrowth processes and limiting processes. Heillustrates this with the scenario of a seed whichholds the possibility of a tree, and through anatural reinforcing growth process, begins torealize that possibility. Initial feelers take in waterand nutrients, then small roots develop, thenlarger roots and more growth. How successful theseed is in becoming a tree, or what progress itmakes depends on natural limits, the amount ofwater, nutrients, sunshine, space for expansion,and insects. Growth that doesn’t reach fullpotential has encountered some constraint that isa natural balancing process in living systems.Senge likens most change leaders to gardenersstanding over their plants shouting “Grow! Tryharder! You can do it!” What can we learn fromthis analogy? That “sustaining change requiresunderstanding the reinforcing growth processesand what is needed to catalyze them, andsimultaneously addressing the limits that keepchange from occurring.” This interplay betweenthe two processes is the “dance of change” referredto in the book’s title. Most leaders deal only withthe growth processes and don’t pay attention tothe limiting processes. Those involved in changeor considering how to change must learn to workwith nature’s way and give attention to both thegrowth processes and the limiting processes.Drawing on the insights of practitioners, the bookshows how leaders at many levels of anorganization can nurture both the reinforcing

growth processes of organizational change, andstrategically deal with the limiting processes thatcan impede or stop that growth.

Leaders of Profound Change: Senge viewsleadership as the “capacity of a human communityto shape its future, and specifically to sustain thesignificant processes of change required to do so.”The book focuses on “leadership communities”rather than hero-leaders. Organizations that relyon a hero-leader will never achieve the profoundchange the Dance of Change describes. Threelevels of leadership must help generate andsustain a kind of creative tension between thevision and current reality of the organization.They are 1) imaginative, committed local lineleaders, 2) enthusiastic mid-level communitybuilders or network leaders, and 3) executiveleaders who, in addition to their accountability fororganizational performance, must create anenvironment of innovation and knowledgegeneration. The book explores these roles withemphasis given to the actions the leaders take,not their personal “style” characteristics.

Critical Learning Capacities: The underlyingcause of failure in most change initiatives is thatthe organization has not developed the learningcapabilities necessary to reflect on, inquire about,and talk openly together about impeding struc-tures, practices, and mindsets that are blockingchange. People can’t raise the serious issues thatneed to be addressed without invoking defensive-ness. The authors still see that the five disciplinesdescribed in The Fifth Discipline (personal mastery,mental models, shared vision, team learning, andsystems thinking) support the crucial learningcapabilities of aspiration, reflective conversation,and understanding complexity. To change, organi-zations must develop their learning capabilities,“the skills and proficiencies that among individu-als, teams, and larger communities, enable peopleto consistently enhance their capacity to produceresults that are truly important to them.”

Three Phases—Ten Challenges

As changes in ways of working and thinkingbecome evident, challenges arise predictably ineach of three phases in the life cycle of change.They are manifestations of how the system itself ispushing back, the natural process of movingtoward homeostasis. Most change initiatives focusonly on the growth processes and not the limitingprocesses. The ten challenges described in thebook are initially phrased, not in systems jargon,but in everyday language heard in most changesituations:

Initiating Change 1. We don’t have time for this stuff! 2. We have no help! 3. This stuff isn’t relevant!

Book Reviews


4. They’re not walking the talk!

Sustaining Momentum 5. This stuff is ___! [fill in the blank] 6. This stuff isn’t working! 7. We have the right way/They don’t under-

stand us!

Redesigning the Organization 8. Who’s in charge of this stuff? 9. We keep reinventing the wheel! 10. Where are we going? What are we here for?

The thickness of the book (570 pages) may beoff-putting to some, but it was not written to beread front to back. Initially, I scanned the book,stopping at particularly intriguing spots, jumpingfrom one section to another. Later I went throughthe book from front to back for the purpose of thisreview. I confess it was overwhelming. There is somuch here, it is impossible to hold it all in yourhead at once. Some sections can be confusing tothose unfamiliar with systems thinking. To theircredit, the authors try to clarify each diagram withan explanation. The structure of the book facili-tates remembering the three major categories ofinitiating, sustaining, and redesigning. The sub-structure of each challenge section gives a mapwith increasingly familiar landmarks (thechallenge description, the system diagram for thatchallenge, strategies that have worked in dealingwith the challenge, stories the illustrate thosestrategies, and, exercises to do as a team orindividually) that makes it easy to navigatethrough the very rich landscape. Particularlyhelpful features are:

§ margin icons, small graphic symbols, identifyingdifferent types of material, such as suggestionsfor the three leadership roles, or an exercise fora group or one that you do alone – a soloexercise, practical techniques, or guiding ideasthe authors found meaningful.

§ systems diagrams for each of the ten challengesto change, and an initial tutorial in how to readthem

§ a lexicon that gives the roots of words in currentuse, especially the jargon of organizationalchange, and a thoughtful description of theauthor’s meaning and use of the word—wordslike assessment, governance, time, and tacitknowledge.

§ sets of reflective questions to consider using tohelp groups think systemically together abouttheir situation or their experience, a virtualtreasure chest of dialogue questions. It wasgratifying to me to see the continual emphasison profound communication and to find thepractical help to enable this to happen.

§ well-written paragraph-long descriptions on rec-ommended books on related topics

§ brief intriguing introductions to the authors ofeach illustrating story.

My advice to readers: begin anywhere yourinterest or need takes you and go forward or back-ward from there. Read it in chunks and give your-self some time in between to soak up the ideas.Look through the Orientation section that brieflyreviews the five disciplines of The Fifth Disciplineand how to read a systems diagram. This sectionalso describes the margin icons found throughoutthe book that will help you navigate through tosections that are of particular interest to thereader. The Dance of Change is resource rich. Itmakes quite an interactive read, with links identi-fying related topics in other sections of the book,and in other books. Reading it and following thesuggested resource topic threads is rather likeclicking on links on the Internet. In fact, theauthors encourage readers’ feedback and com-ments via their website address,www.fieldbook.com.

Sometimes I found it hard to put down thebook, as the intriguing titles were seductive. Someof my favorites: Barking and Nonbarking Dogs,Climbing Out of the Muck, The Pinecone Strategy,“As One Of The First Black Engineers In SouthAfrica,” Heretical Tactics, and The Perils of SharedIdeals. The first-person style and crisp writing ofthe illustrating stories also drew me in. I experi-enced it not unlike passing by a TV set and gettingsucked into the program on my way to doingsomething else.

This book is a valuable addition to the libraryof any facilitator working on a systems level withclient organizations. The review of systemsdynamics was very helpful. Perhaps you, likemyself, find it difficult to “read” systems diagrams,or to remember them without a visual picture.After going through the diagrams that accompa-nied each of the ten challenges, I felt much morecomfortable in interpreting them. It was informa-tive to read about how the challenges show up ineach phase of change; I was reminded of severalinstances in past work situations. The resourcesthat are included—both exercises to do withteams and books written by other authors on thetopic of each challenge—are likely to be greatlyappreciated; facilitators will find some favoriteshere and some new ones to explore. I hadimmediate use for two sections of the book, andrecommend them highly. One is an extensivediscussion on coaching and mentoring, one oftoday’s “hot” topics. The other has very helpfulinformation about assessment and measurementof change, along with suggested ways to help agroup devise assessment criteria that isappropriate for them.

In the final section of the book Sengereiterates the dynamics of the dance of change,

Book Reviews


the interplay between the reinforcing factors andthe limiting factors to growth. The challenges ofchange are dynamic, non-linear, andinterdependent. He illustrates the balancingprocess and compensating feedback that is anatural aspect of the growth process, anillustration of how a living system works tomaintain its internal balances, conserving thestatus quo. Organizational leaders should payattention to what the balancing process is con-serving. In this summary, Senge again speaks ofchange leaders as constituents of a leadership“ecology” which includes, but is not limited to, thelocal line leaders, the executive leaders, and thenetwork leaders described earlier.

He concludes with the point that in learningto initiate and sustain change, and in developingthe capabilities that support organizationallearning, we build a foundation to beginaddressing core issues of society. These are thedeep problems facing our planet that won’t befixed by a few great leaders. As more peoplebecome capable of thinking systemically aboutproblems and learn to talk openly about the deep,underlying issues that face us, we will gainconfidence in our ability to search for ways toaddress complex issues together. Hopefully, hewrites, this capability and confidence will begin tospill over into other aspects of society. In thisendeavor, we will be guided by a few core images

of the world we want to create, based on how thenatural world operates rather than those ofindustrial society that has given rise to theseproblems. Perhaps the guiding image may be “thatof the Earth itself, the living system that is ourhome.”

In summary, reading this book was like open-ing a treasure chest of resources, ideas, models,and provocative, reflective questions for helpinggroups journey through profound change. Themetaphor the authors suggest, an atlas of change,is definitely appropriate. With this book in hand,a change leader, whether facilitator or manager,will be able to navigate the journey withawareness and confidence. Advance warning ofinherent challenges along the way andsuggestions of ways of addressing them alleviatesome of the anxiety of embarking on such ajourney.

ReferencesSenge, P. (1990). The Fifth Discipline: The Art &

Practice of The Learning Organization. NewYork: Doubleday

Senge, P. (1994). The Fifth Discipline Fieldbook:Strategies and Tools for Building a LearningOrganization. New York: Doubleday

Patricia R. Tuecke is the president of Sierra Circle Consulting in Reno, Nevada. She is afacilitation consultant in organizational development and assists organizations in creating andimplementing effective strategies for initiating important changes. Mrs. Tuecke has consulted topublic, private, and volunteer organizations for over twenty years in the US, Europe and Asia. Shehelped develop the Technology of Participation® (ToP) and trains facilitators in those methods. Amajor focus of her work is designing and facilitating group processes to help management teamsplan systemically for the future of their organizations, including developing a shared vision andstrategic resolution of critical issues that may impede success. She designs interactive trainingprograms, delivers presentations and seminars on management topics, coaches managers, anddoes shadow consulting with practitioners using ToP® methods and other large group processes.She is a founding member of the International Association of Facilitators. Contact at: SierraCircle Consulting, 514 Island Avenue, Reno, NV 89501; Phone: 775-333-6998; Fax: 775-333-1088;[email protected]

Book Reviews


The Logic of Failure:Why Things Go Wrong and What We Can Do To Make Them Right

by Dietrich DornerMetropolitan Books, Henry Holt and Co., New York, 1996, translated by Rita and Robert Kimber, ISBN:

0805041605, $25.00. Paperback edition: The Logic of Failure: Recognizing and Avoiding Error in ComplexSituations, Perseus Press, Cambridge, MA, 1997, 222 pages, ISBN: 0201479486, $16.00.

Book Review by Nancy S. Hewison

Lots of things go wrong in Dietrich Dorner’sThe Logic of Failure. Nuclear reactors fail, commu-nities break down and entire societies crash.Complex situations go awry despite, or because of,the planning and subsequent actions ofintelligent and well-intentioned people. Dornerwould have much to talk about with Peter Senge(1990) and others working in the area of systemsthinking. The Logic of Failure is packed withillustrations of the ways humans approachcomplex situations and fail to solve them becauseof an inability to think systemically. Dorner, aprofessor of psychology at the University ofBamberg, Germany, utilizes computer simulationexercises to study problem solving, decisionmaking, human information processing, and thepsychological aspects of planning, and to elucidatethe cognitive behaviors that lead to failure inthese arenas. Rather than burden the reader withtedious detail about the computer simulations,however, he interweaves learnings from them withnumerous real life examples which enhance thereader’s comprehension, contribute to readability,and provide illustrations of potential use tofacilitators.

The first chapter presents a number of scenar-ios that capture the reader’s attention. In oneexample, each participant in a computer simula-tion experiment was given dictatorial powers toincrease the well being of the occupants ofTanaland, a fictitious region in West Africa. Overthe course of ten computer-simulated years, a par-ticipant had six opportunities to intervene in theregion’s agricultural practices, medical care, andaccess to water, electricity, and motorized equip-ment. At each intervention point, informationcould be gathered to feed into planning. Of twelveparticipants, only one succeeded in stabilizingpopulation growth while creating overallimprovement in the standard of living. Thedecisions of the average participant, on the otherhand, initially improved life in the region and thenled to one or another catastrophe. In one case, avigorous campaign to eradicate the rodents andmonkeys who were eating the crops deprived thelocal leopards of their normal food supply. Theythen turned to feeding on the farmers’ cattle. Inanother case, crop yield

increased due to motorized plows and artificial fer-tilizer, with the result that the population grewuntil it exceeded the capacity of the food supply.

Why did one participant succeed while all theothers failed? None of them had any particularexpertise and the experiment’s problems did notrequire any specialized knowledge. The answers,according to Dorner, lie in the way humans thinkand in the fact that, when dealing with complexsystems, “...we cannot do just one thing. Whetherwe like it or not, whatever we do has multipleeffects.”

In discussing a real life example, the 1986catastrophe at the atomic energy plant in Cherno-byl, Dorner suggests, “We cannot find a singleexample of [human] failure. No one who shouldhave stayed awake fell asleep...overlooked a sig-nal...[or] accidentally flipped a wrong switch.” Theplant operators, however, exhibited many of thecharacteristics of participants in the Tanalandexperiment. They thought in terms of linear net-works of causation rather than considering poten-tial side effects of their decisions and actions, andthey did not understand that exponentially devel-oping processes move extremely rapidly. Further-more, under time pressures, they applied over-doses of established measures.

Having stripped away any hope the reader mayhave harbored that the cognitive pitfalls ofdecision making reside only in such simulatedsituations as Tanaland, Dorner sets out, inchapter two, to demystify the demands ofcomplexity, dynamics, and intransparency whichplanners and decision makers regularly face. Thatis, there are many interrelated variables whichtogether form a system that moves on its own,whether or not we act (or even acknowledge themovement), and in which it may be impossible todiscern the information we most need.

Because the setting of goals is critical in plan-ning and problem solving, the author spends sometime in the third chapter delineating various typesof goals (positive/negative, general/specific, clear/unclear, simple/multiple, and implicit/ explicit)and the kinds of interdependencies betweengoals. Practical steps are suggested, as in anapproach to tackling the multiple problems thatmay be uncovered in the process of turning an

Book Reviews


unclear goal into a clear one: make a list of theproblems, determine the interdependenciesamong them, rank them regarding importance andurgency, and consider delegating any that aresufficiently independent to be worked on apartfrom the rest.

In the process of examining goal setting,Dorner holds up a mirror in which facilitators mayrecognize some of our clients or even, painfully,ourselves. An example is “repair service behavior,”which occurs when, faced with an unclearcomplex goal (such as “a user-friendly library”), weset about finding whatever seems to bemalfunctioning, and then make its repair ourimmediate goal. Sometimes we select problems towork on based on the ones that seem mostobvious or the ones we have the expertise tosolve. It is also possible to fall prey to “flowsituations” in which “the fascination exerted bywork that constantly poses new challenges ofmoderate difficulty” leads a problem solver awayfrom a major goal that has been inadequatelydefined and understood.

The information we collect and the models ofreality we construct based on it form another vitalpart of planning and action. If the garden poolstinks, he suggests, it’s all too common to spendthe weekend removing the fish and plants, drain-ing the water, peeling back and discarding theliner, and starting over with fresh ingredients.After all this work, we relax, thinking we’ve solvedthe problem, only to find weeks later that the poolstinks again. Of course, we’ve failed to take intoaccount the complex interrelationships betweenthe organisms, oxygen, temperature, and organicand inorganic substances in the pool. Had wetaken the systemic view, and collected key bits ofinformation, we might have determined aneffective solution to the underlying problem,rather than jumping to treat the symptom.

By this point in The Logic Failure, a reader withsome background in Peter Senge’s work on sys-tems thinking will be expecting his appearance inthis book with the turning of every page. Itappears, however, that the two wrote in ignoranceof one another’s work, perhaps because Dorner’sbook first appeared in print (in German, publishedin Germany) in 1989, almost simultaneously withthe 1990 publication (in English, in the USA) ofSenge’s The Fifth Discipline. Be that as it may, forfacilitators seeking a better understanding of sys-tems thinking, or looking for examples and analo-gies to use in explaining this approach to others,Dorner’s book is a gold mine.

The chapter on time sequences is particularlyrich with illustrations of systems thinking, or itslack. The author describes an experiment inwhich participants are asked to assume (in acomputer simulation) manual control of

temperature regulation in a refrigerated foodstorage area. This scenario nicely illustrates theproblems in dealing with developments that showchanges of direction (oscillations, suddenreversals) due to a delay in the system’s responseto human intervention. Dorner then eases into anexplanation of exponential growth and the pitfallsin failing to understand it. He tells the tale of theinventor of chess who, upon presenting the gameto his king and being promised a reward in acondescending fashion, asks “only” for one grain ofrice for the first square on the chess board, two forthe second, four for the third, eight for squarefour, etc. The smug king discovers to his dismaythat, through the power of exponential growth,the reward he owes for the last square aloneamounts to 153 billion tons of rice.

The penultimate chapter deals with planningand presents some ways to avoid failure incomplex situations. In Dorner’s view, planning“consists of examining the consequences ofindividual actions, then of stringing individualactions together into sequences and examiningthe possible consequences of these sequences ofaction.” The resulting plan may branch in a varietyof directions depending on circumstances.However, because most problems exist in a vastand complicated reality, we often must narrow the“problem sector.” The author offers guidance onwhich of a number of alternative planningapproaches to select for a particular situation.When goals are clear, he suggests reverseplanning: beginning with the goal and workingbackwards through the action needed to achievethat goal, then the conditions necessary tosupport that action, then what will create thosesupporting actions. When there appears to be nodirect path to the goal, planning may involve iden-tifying decisions which will lead to situations fromwhich one can efficiently proceed in a variety ofdirections, as in the game of chess.

In addition to reverse planning and the “effi-ciency diversity” method, the chapter on planningpresents the “hill climbing” approach, with itsattendant danger that we may be climbing thewrong peak as we seek to shorten the distancebetween desired goal and present reality. Anapproach suggested when one is at a total loss isto act on the basis of what’s proven successful inthe past, but without succumbing to methodism.Because, to Dorner, the planning process is largelyone of narrowing and expanding the problemsector through a variety of carefully selectedapproaches, he also offers ways to expand a prob-lem sector when a lengthy period of effort hasproven unsuccessful, thinking has become circu-lar, or possibilities seem exhausted. In the lattercase, he suggests thinking by analogy, and themany examples in The Logic of Failure provide arich stock of possibilities for analogous thinking.

Book Reviews


The book’s final chapter begins by identifyingthe psychological reasons for what Dorner calls“the many inadequacies of human thought” indealing with complex situations. First, becauseconscious thought takes time, and it is difficult toprocess multiple and varying information simulta-neously, we take shortcuts to economize. Forexample, we leap ahead to action planningwithout stopping to set goals, or we operate onthe basis of a single principle rather than thinkingstrategically. A second psychological process thatcan doom our endeavors (and one that is ratherchilling for any professional, including a facilitator,to contemplate) is the unconscious attempt topreserve a positive view of our own competence.Behaviors that may arise from this self-protectioncan include developing an overly simple hypothe-sis; too readily labeling a problem as an old friend,and therefore applying a tried and true solutionthat may not really fit; or selecting only thoseproblems we feel confident about solving. Dornerfurther suggests that, while the human memorycan store a great deal, it takes in new material ata relatively slow rate. This creates a storehouse ofpast events that lack the clarity and richness ofdetail with which we perceived them at the timethey occurred. One result of this is our difficulty,explored earlier in the book, in dealing withtemporal configurations. Finally, we typically “don’tthink about problems we don’t have,” leading usto overlook potential undesirable side effects ofplanned actions.

What, then, does Dorner suggest that we do,now that we are aware of the cognitive behaviorsleading to failure in complex situations? His adviceis to draw upon “operative intelligence" by beingaware of the ways we think and deliberatelychoosing among them. We should,” the authorasserts, “reflect on our own thinking, but withsome guidance,” an idea that certainly recom-mends itself to facilitators. Talking about complexproblems and how to approach them is notenough, however, in Dorner’s view. He believesstrongly in the value of computer simulations, fol-lowed by discussion of behavior and mistakes, asan excellent way of learning the essentials of sys-tems thinking. While some readers may feel thatthe final chapter fails to deliver on its title (“SoNow What Do We Do?”), the entire book is, inessence, the answer to that question.

Whether or not facilitators take Dorner’sadvice regarding computer simulations, a reflectivereading of his book should provide the insightsafforded by holding up a mirror to our ownpractice. That is, the author may help us considerthat in facilitation, as in other arenas of action,“whatever we do has multiple effects.”Additionally, we may see some of the dangers offailing to ask ourselves whether our selection of aparticular course of facilitation might be

influenced by the lure of fixing the obvious, or thecomfort in applying a familiar approach that is notreally a good fit for the client’s situation. Anotherbenefit of the book for facilitators is the expandedtreatment of systems thinking, which sheds lighton the challenges of problem identification,situation analysis, planning, and decision makingencountered in complex organizations. The book’srich collection of examples, presented to promoteunderstanding of concepts, can also provide thefacilitator with illustrations and analogies for usewith clients. In my case, I was not content to relysolely on the book’s index for future reference,and I covered the blank pages at the back of mycopy with lists of concepts and the exampleswhich Dorner uses to illustrate them: grains ofrice on a chessboard representing exponentialgrowth, the pitfalls of methodism as seen in failedbattle strategies, the experience of driving a car ina busy city as an example of attending to “supersignals” instead of every detail. It is difficult toisolate “must read” portions of The Logic of Failure.However, any abbreviated reading should beginwith some of the book’s many examples(Tanaland, Chernobyl, predator and prey relation-ships, etc.) because they provide eye-openingillustrations of the challenges involved in acting orintervening in complex systems.

Finally, a few words of praise are in order forthe felicitous combination of Dorner’s writing andthe work of the translators, Rita and RobertKimber. The result is so readable that I was wellinto the book before I realized just how denselypacked it is with concepts. Many of these meritfurther exploration and, fortunately, Dorner’schapter-by-chapter notes include manyreferences, some in English and some in German,for additional reading. On several levels, therefore,The Logic of Failure is well worth reading forpracticing facilitators and those teachingfacilitation, for leaders of complex organizations,and for individuals leading their own complexlives.

Book Reviews


Dietrich Dorner’s book, The Logic of Failure wasoriginally published by Rowohlt Verlag in Germanyin 1989, under the title Die Logic des Misslingens,and has been issued in English by a number ofpublishers, with various subtitles.

ReferencesSenge, P. (1990). The Fifth Discipline: The Art &

Practice of The Learning Organization. NewYork: Doubleday

Nancy S. Hewison is Professor of Library Science and Director of Administrative Services for thePurdue University Libraries. Her work in the Libraries over the past ten years has included thefacilitation of strategic and action planning. She has written about the Purdue UniversityLibraries’ organizational transformation in the journal Library Administration and Management,and, with Donald O. Bushman, in a chapter in Government Works: Profiles of People Making aDifference (edited by James P. Troxel; Miles River Press, 1995). Nancy holds an M.S. in LibraryScience from Simmons College, Boston, and a B.S. in Sociology from the University of NewHampshire. Contact at: Purdue University Libraries, 1530 Stewart Center, West Lafayette IN47907-1530; Phone: 765-494-2900; Fax: 765-494-0156; [email protected]

Success with Soul

By Doris Pozzi and Stephen WilliamsDorian Welles Pty Ltd, Melbourne, Australia, 1997, 282 pages, ISBN 0646286927

Book Review by Judy Robb

Success with Soul contains many key ingredi-ents to help a person facilitatechange—personally, organizationally, anduniversally. The authors, Doris Pozzi and StephenWilliams, begin the book with the assumption thatintegrating the successful side of our lives withthe side concerned with meaning—the soulful—isone of the key issues we struggle with today. Manyof us have worked long and hard at careers thathave provided many good things—recognition,money and comfortable lifestyles—but for somereason we still feel unsatisfied. “We have come toa point where we know we could go further, butdon't know how. Somehow our lives are lacking.We ask ourselves, could this be all that life has tooffer?” (Pozzi & Williams, 1997, Introduction xii)

Whether or not this theme rings true for you,Success with Soul contains information and activi-ties with which to examine professional success ina deeply personal way. If you are looking for a wayto examine your personal and professional suc-cess, you may enjoy the book. The book’sstructure may create a challenge for those whoare application-oriented in their thinking. Theauthors examine both theoretical and concreteconsiderations, sequentially rather than together.Readers with a predominantly theoreticalorientation may not find this a burden in readingthe book and may be able to effectively gleanhelpful information from the very start. Only afterfinishing the book did I completely understand it’smessage and see it as a book to help work througha process of personal reflection.

In the introduction, the authors ask readers to“… be content to say that, for our purposes, thesoul is that part of our non-physical being that isthe home of real personal meaning.” (Introductionxiv) I found this definition helpful because mymind kept jumping to the various meanings thatexperience and society have given to the word“soul.”

Pozzi and Williams have used theory and appli-cation to aid readers in integrating their views ofsuccess with soul. To see how they have accom-plished this, we will take a closer look at thestructure and content of the book. It is dividedinto three sections: Knowing, Doing and Being.The authors tell us early on that much of thebook is a synthesis drawn from their ownprofessional and personal experiences. The bookis a situational study of various concepts, not acritical analysis of them. I found that in readingthe book the most effective mind set for me was touse their premises and assertions as hypotheticalstarting points.

Knowing (approximately 100 pages) providesreaders with theoretical explanations, historicalreferences, research results and examples.

Initially, the authors discuss their perceptionsof how, up to now, people have thought about therelationship between success and soul. Readershave the opportunity to explore traditional andmodern definitions of success. Pozzi and Williamstouch on early writings on soul, including the rela-tionship of soul to culture, religion, philosophy,

Book Reviews


psychology, and art. They describe their views con-cerning the progression of soul—when it wentmissing (in Western thought), when soul returned,and their perception of the real need for a newway of talking about success.

Pozzi and Williams say this new way of under-standing success must begin from within—thatpersonal growth requires personal change. Theytake this concept a little further. First, they drawon new research about the way in which our per-sonal values shape our attitudes and actions.Secondly, they consider how our commitmentsand actions can, in turn, shape our attitudes andultimately change our values. (p. 27) They discussthe following topics: personality is not the answer;that the soul exists; where the mind goes thebody will follow; mental models underlie ourthinking; and chaos is part of our lives. Thechapter concludes with a discussion of wholesystems thinking and what can be drawn fromEastern and Western cultures as we strive toknow what we already know.

Next, the authors present the premise that “toadvance, any effective strategy must begin with anunderstanding of the underlying dynamics.” (p. 47)Pozzi and Williams indicate that the underlyingdynamics shape our experiences, beliefs andactions. As readers, we are shown the impact ofour tendency to focus outward. We sometimes letothers define what is important. As a result, theauthors indicate that today many of us are out oftouch with our personal values and are unclear asto our purpose. Are we simply to accept the valuesand purpose placed on us from outside by others,or the world around us? Sometimes we do thisconsciously; at other times we are not even awarethat we have done so.(p. 52)

Leading a balanced lifestyle is examined next.The reader is presented with an argument illus-trating why a balanced lifestyle (for example, bal-ancing competing commitments) is not thesolution to achieving personal success with realmeaning (soul). Pozzi and Williams pose andexplore two main notions. First, that a balancedlifestyle does not directly address the underlyingdynamics at work in our lives. It ignores what isreally going on. Secondly, that it creates newproblems that then need to be solved. (p. 61)

After Pozzi and Williams describe the problemswith the integration of success and soul, they pro-ceed to describe how success with soul cannot beachieved without talking about meaning in a newway—that meaning which provides our ownunique and personal answer to life. They discusstwo types of “meaning”—the big picture andpurpose, “Why are we all here?” and our particularrole, “Why am I here?” In this chapter, they alsodelve into another important aspect of meaning,“that two different people may ostensibly be doing

the same thing—in psychological terms, theiractions may be the same—but the two people willplace different meanings on their actions.” (p. 78)The consistent factor in these different views onmeaning is relationships.

We now learn about the authors’ “relationshipmodel of success.” “This model emerges from ‘sys-tems thinking’ which considers not only the indi-vidual components of a system, but also whatoccurs between them—the energies or relation-ships.” (p. 82) The model focuses on the system ofindividual, social and environmental relationships.In this chapter, readers see that system survival isnot about balance or equilibrium, but aboutchange, renewal and transformation. The authorspropose that “when you operate out of therelationship model of success, you will live aresponsible and meaningful life.” (p. 89)

The last consideration of the Knowing sectionof the book focuses on competing priorities and howthe tension they create impacts personal success,business matters and social issues. As the authorsstate, "Ultimately, each of us needs to understandour own personal creativity-sustainability tensionand manage it to achieve creativity whileremaining aware of the potential costs."(p. 100)

Through the nine chapters of approximately150 pages in section two, Doing, the authors pro-vide readers with practical activities and experien-tial descriptions to integrate success and meaninginto their lives.

For the authors, identifying personal valuesprovides the foundation for the integration of suc-cess and meaning into our lives. The authors setthe stage by discussing and defining the relation-ship between beliefs, attitudes, values andactions. They describe an examination of valuesby a group of international researchers. Theresearchers identified fifty-seven personal valuesclustered into ten motivational domains. This wasvalidated with inputs from hundreds of peopleacross forty-four countries. To the researchers,these domains represent the universal humangoals that shape our behavior. (p.108)

Using a manageable categorization of actualvalues, the authors provide readers with key stepsto begin identifying their own personal values.The easy-to-follow instructions focus on how tochart personal values, create a personal-journeytimeline, and build a values pyramid.

This is where the book changed for me. I clus-tered the researchers’ values into the “tempera-ment-oriented” frame of reference that I use in mywork (see Understanding Yourself and Others, AnIntroduction to Temperament, Linda V. Berens,).Therefore, with pen in hand and starting from mypoint of reference, I began using the authors’ proc-ess of identification. I quickly noted my personal

Book Reviews


values (freedom and action) and began developingmy personal-journey timeline. It provided a frame-work to link the changes in my life to majorevents, my moods and values. When I completedthe values pyramid, I had established a visualimage of the importance of my actual values. I wasexcited. I could see the possible application ofexercises in facilitating individuals and groupsthat are searching for meaning in their work.

With values as the foundation, Pozzi andWilliams move to defining and discussing the needto identify the important and meaningful areas ofone’s life. Using the helpful guidelines and exam-ples provided, readers can begin creating their lifecircle—a diagram that contains all the identifiedkey content areas of one’s life. The authors showhow the “life circle” tool can be used to “create acomplex network of linked and interconnected lifecircles for the various dimensions of life-a networkwith a central life circle and a series of ‘satellite’circles projecting out and perhaps linkingtogether.” (p. 135) As the authors indicate, “life cir-cles are perceptual maps. They allow us to create apicture of our lives, which is both orderly and cha-otic—representing our life system.” (p. 135) “Lifecycles are also an important systems thinking tool.They show the components, connections andrelationships that are important to us.” (p. 135)

Next, the authors provide a process to developstatements of purpose—why we exist. Activitiesinclude examining our response to life and life’sresponse to us. As the authors indicate, “Gettingout there and doing something is more likely toresult in some answers than ‘being’ and ‘knowing’alone.” (p. 143)

With a picture of the present, the authorsshow readers how to apply the “personal strategicplanning method” to create a vision or future view.Once established, readers repeat the process tocreate personal medium-term plans (five years out)and tangible plans (one year out).

Next, the authors take readers through activi-ties to turn personal plans into reality. Pozzi andWilliams provide an in-depth examination of effec-tive decision making, a personal decision makingchecklist and activities to build better habits fromeffective decisions. As the authors say, “Our val-ues, life’s purpose, decisions, plans, actions andpersonal reflection become key components of thedynamic system which is our mind, body and soul.Alignment and integration of these componentsprovide depth to everything we do. Meaning is cre-ated through the active structured dialogue ofthese components and through the establishmentof relationships with other people and the worldas a whole.” (p. 198)

With a foundation to make better choices andbuild better habits, the authors revisit systemsthinking. They introduce, discuss and apply the

five methods of systems thinking: integration,elimination, redefinition, substitution and accep-tance, in ways that take readers “beyond balanceto resolve competing priorities in our lives.” (p.200) Continuing to build on the knowledge andapplication presented, the authors provide activi-ties to help readers incorporate feedback, moni-toring, and reflection into the key relationships wehave with others, our environment and ourselves.

The section on Doing includes the authors’personal view. They explain what “creativechange” means to them, why they initiated it intheir lives, and how they went about applying theprocess. They describe their personally developedfour-stage creative-change process of disruption,incubation, transformation and action. Theauthors note that “during the implementation oraction stage in the creative process, your moodchanges completely; discouragement and fear turnto optimism and hope—smiles and energyreappear. You feel centered and focused. Often atthis stage, people experience a return of purposeand meaning into their lives.” (p. 240) The authorsclose the section on Doing with techniques theyhave found effective and that will help readersachieve personal success with soul.

The last section, Being (approximately 25pages), provides readers with points of reference tofacilitate the integration of success and meaninginto their lives. The authors explain that rediscov-ering soul is not about finding an object, like a lostset of keys. “Rediscovering soul is about engaging inthoughts and activities that are meaningful to us oradd meaning to our life.” (p. 252) Readers are givenideas on how to create and enhance their ownpersonal process of soul rediscovery.

As Pozzi and Williams express, “In many ways,we feel that at the heart of this book is a very sim-ple concept, that we can integrate personalsuccess and soul by improving the quality of ourrelationships. However, like most simple ideas, thetest is in the doing—this is where the challengelies.” (p. 270)

In conclusion, Success with Soul has severalworthwhile aspects. It offers a methodology toassist individuals in the search for success withmeaning in their lives. Secondly, Success withSoul contains tools and exercises a facilitatormight use with clients who are involved increating change. The authors present usefulexercises and applications to aid in achievingpositive results. Noteworthy, from my point of view,is the authors’ commitment to provide methodsdesigned to help you, as an individual, workthrough the process of initiating and obtaininglasting creative changes in your life throughpersonal reflection.

Book Reviews


ReferencesBerens, L. V. (2000). Understanding Yourself and

Others: An Introduction to Temperament. Hunt-ington Beach, CA: Telos Publications, www.tri-network.com

Judy Robb established her own performance consulting practice, the Robb Group of San Fran-cisco, California in 1994. Her services: performance consulting, training and facilitation range fromcoaching clients, conducting performance assessments, developing performance models, andfacilitating performance improvement processes and projects. She was a founding member of theorganization that won the 1997 American Society for Training and Development Best PracticeAward for Performance Consulting. She is the author of “The Job of Performance Consultant” inMoving From Training to Performance: A Practical Guide Book, edited by Dana Gaines Robinson andJames C. Robinson. She is the co-author of Temperament and Type Dynamics: The Facilitator'sGuide. Judy has a B.S. from the University of San Francisco and is an M.B.A. candidate from StMary's College. Contact at: Robb Group, 1 Sansome Street, Suite 2100, San Francisco CA 94104;Phone: 415-921-2466; [email protected]

Book Review Editor: Beret Griffith

Group Facilitation: A Research & Applications Journal

Aim and Scope

Group Facilitation: A Research and Applications Journal is a multi-disciplinary publicationfocused on the art and science of group facilitation. The aim of Group Facilitation is to advance ourknowledge of group facilitation and its implications for individuals, groups, organizations, and communi-ties. It is published semi-annually.

The Group Facilitation Journal is intended for facilitators, mediators, organizational developmentand training specialists, managers, researchers, and others who seek to use facilitation skills in theirpractice. Articles represent diverse perspectives, including organizational learning and development, groupand system dynamics, collaborative technology, negotiation, mediation, leadership, decision-making, con-flict resolution, cross-cultural contexts, and education. Possible topics include, but are not limited to,facilitator roles within the group, interventions for conflict management, descriptions of specific facilita-tion methods, approaches to facilitating specific tasks such as idea generation or priority setting, usingcomputer technology to support facilitation, increasing participation in organizations, exploring theunderlying values, beliefs and models of facilitation, and applying facilitation skills and concepts to vari-ous settings.

The journal is comprised of the following sections, which are described below in more detail: Applica-tion and Practice; Theory and Research; Edge Thinking; Book Reviews.

Application and Practice is devoted to articles that reflect on facilitator experience. Articles appropri-ate for this section include reports on experiences gained and lessons learned presented in a reflectivecase study, and discussion of facilitator roles, problems encountered by facilitators or their clients, andintervention methods and techniques. Studies should be both descriptive and evaluative and should drawon existing literature appropriately.

Theory and Research is devoted to articles that explore, propose, or test practices, principles, orother aspects of facilitation models. Such studies are typically based on survey, experimental, ethno-graphic, or other research methods.

Edge Thinking is intended to stimulate thinking on new concepts and issues. Contributions may beless formal than the other sections, and might include dialogues, essays, editorials, and proposals for newareas of inquiry.

Book Reviews presents critical and comparative reviews of recent and classic books related to groupfacilitation.

Submission Guidelines

Submission guidelines and other information about the journal may be obtained on the journal web-site www.iaf-world.org/Journal or from the Editor (see below).

Original manuscripts should be submitted via electronic mail (preferred) to:

Sandor P. Schuman, [email protected]

Paper submissions may be sent to:

Sandor P. Schuman, EditorGroup Facilitation JournalCenter for Policy ResearchUniversity at Albany, SUNYAlbany NY 12222

Group Facilitation: A Research & Applications Journal

Book Review Guidelines

Group Facilitation: A Research and Applications Journal presents critical and comparativereviews of recent and classic books related to group facilitation. These guidelines include recommenda-tions for writing and submitting a review, information about the review cycle and reviewer qualifications.

Writing the Review

When writing a review, please include:- overall impression of the book- the highlights and structure of the book- for whom the book would be appropriate- what you found particularly helpful, unclear, weak- your personal learning, if any- particular benefits to you in your facilitation, if any- value of the book for facilitators, if any- significant contribution of the book, if any, to the field of facilitation- your recommendation of "must read" portions of the book, if any- a summary or wrap-up of your reading experience.

In addition:- provide definitions of terms, acronyms, references, and background summary statements where appropriate.- where necessary, be sure to include complete citations and attributions.- identify specific texts (usually a sentence or phrase) for possible use in pull quotes.- publisher; ISBN designation; price ($US); where to purchase (i.e. all major bookstores or telephone number for direct

ordering if the book is not widely available commercially).- background about the book author: facilitation experience and/or other writings.

What we are looking for:- people familiar with the conceptual and practical sides of facilitation and who are willing to spend the time required to

write interesting and thought provoking reviews.- reviews of books that address facilitation and related issues, such as consensus decision making, participatory

problem solving and group decision-making.- in-depth and critical reviews that help readers decide whether or not the books reviewed are ones that they should

consider reading.- comparative reviews of two or more books that differentiate, compare and contrast the books and thoroughly exam-

ine the strengths and weaknesses.- reviews that place the book in the context of other literature.

Submitting a Review

Reviews are typically between 1,000 and 3,000 words. Submissions should be made electronically (viaemail or on IBM compatible diskette) in Microsoft Word, WordPerfect or Rich Text (RTF) format. If you areinterested reviewing a book for the Journal, please contact:

Lynda Lieberman BakerBook Review EditorGroup Facilitation JournalMeetingSolutions, Inc.Box 4062Austin, Texas 78765

Voice: 512-323-0583Fax: 512-323-9016Email: [email protected]