Group Cohesion

A Historical View on Group Cohesion

While reviewing the cohesion articles that have appeared in Small Group Research over the years, several trends stand out. First, the topic of group cohesion has remained a remarkably popular research topic over the course of nearly half a century. While many topics come and go as hot topics for research, articles on group cohesion consistently find their place in academic journals. In total, 77 articles on the topic of cohesion have appeared in Small Group Research since the start of the journal in 1970, making it one of the most commonly appearing topics in the journal. At a broader cross-discipline level, entering the search term “group cohesion” in Google scholar yielded over 43,800 matches. Few topics in small group research can come close to rivaling the bulk of research on group cohesion.

Second, in addition to the sheer amount of research on group cohesion, the relative impact of articles published on the topic of group cohesion is remarkably high. Of the most-cited Small Group Research articles, the most commonly occurring topic is group cohesion (11 of the top 40 cited articles in the journal concern group cohesion). Indeed, all articles reprinted in this issue have been highly influential in the field; Evans and Dion’s (1991) article is the second most-cited article in Small Group Research, and Gully, Devine, and Whitney’s (1995) article the third most cited. ¹

One of the potential reasons why group cohesion has had such a high impact on the field of small group research and beyond is that it is one of the few areas of small group research where the main conclusion from the literature—that cohesion is moderately positive for group performance—has remained relatively constant over the years, as evidenced by multiple meta-analyses (e.g., Beal, Cohen, Burke, & McLendon, 2003; Chiocchio & Essiembre, 2009; Evans & Dion, 1991; Gully et al., 1995). While the relative strengths of effects may vary based on context and task, in general, cohesion is a remarkably robust process in teams, which researchers have been able to apply across a variety of contexts and disciplines.

As cohesion research has spread across different domains of research, inconsistencies in definition and measurement have inevitably occurred over time. While researchers have reliably agreed over time that attraction to the group is an important element of group cohesion (from Festinger, Schachter, & Back, 1950, to Chiocchio & Essiembre, 2009), the exact dimensionality of cohesion continues to be a source of debate. In addition, the generalizability of cohesion and its effects across different team types and contexts is also still called into question (see Pescosolido, 2012, in this special issue). Therefore, while our theoretical understanding of cohesion across a variety of contexts has grown and become more nuanced over the years, room for future research remains in continuing to understand the differences in the nature and effects of cohesion across different types of groups and group contexts.

Paralleling the advances in the theoretical development of group cohesion over the last decades, the measurement of group cohesion has also substantially evolved. The spread of cohesion research across disciplines has meant that methods for researching group cohesion have had to evolve and fit different contexts. In addition, with increasing awareness of the importance of identifying and accounting for levels of analysis in small group research (e.g., Bonito, Ruppel, & Keyton, 2012), cohesion research is increasingly conducted in a multi- or group-level manner, rather than utilizing individual-level analyses of individual perceptions.

While both theory and measurement of group cohesion have considerably developed over the years, important questions still remain. For example, since the beginning of cohesion research, theorists have pleaded for more theoretical and empirical attention to the dynamics by which cohesion evolves in groups and how cohesion may differ in different phases of group life. To this day, this is still cited as a future research direction in many articles on group cohesion (e.g., Chiocchio & Essiembre, 2009). The question then arises as to why this area of research has still yet to develop, and the answer undoubtedly lies in the complexity, both theoretically and empirically, of pulling apart the multilevel dynamics by which cohesion arises and changes over time in groups. However, with increasing methodological knowledge of how to model the evolution of dynamic group processes over time, there is hope that the next generation of cohesion research will begin to address this need in the literature.

You will undoubtedly see other trends, similarities, differences, and paradoxes when reading these classic cohesion pieces over the years (Carron & Brawley, 2000; Drescher, Burlingame, & Fuhriman, 1985; Evans & Dion, 1991; Gully et al., 1995), concluding with the present day review of cohesion in sports teams by Pescosolido (2012). Future submissions to Small Group Research that investigate these ideas would be welcome.

An Introduction to the Articles in the Cohesion Special Issue

The five articles contained in this issue each portray the field at a certain snap shot in time. Drescher et al.’s (1985) article offers an important overview of the first wave of research on cohesion in groups, identifying important parameters to consider when defining and operationalizing cohesion. Their framework helped provide clear boundaries and decision factors to consider that shaped cohesion research for years to come. Evans and Dion’s (1991) article presents an important meta-analysis of the literature, in which they find that cohesion was moderately positive for group outcomes. Gully et al.’s (1995) meta-analysis, published several years later, provides important nuance to Evans and Dion’s (1991) meta-analysis by showing that the relative strength of effects of cohesion on group outcomes may vary based on task interdependence and the level of analysis at which cohesion is operationalized and tested. Carron and Brawley’s (2000) article addresses the large range of cross-task and discipline research that spawned from Gully et al.’s (1995) meta-analysis. Importantly, Carron and Brawley (2000) provided useful guidelines on how cohesion research should best be applied and adapted when being tested across different disciplinary and contextual settings. Their guidelines have helped facilitate the ensuing waves of cross-discipline cohesion research that have been published in the last decade. Finally, Pescosolido (2012) discusses how work on cohesion has been applied across different disciplines in the past decade, and in particular focuses on the developments surrounding cohesion research in the area of sports teams. Each of these classic articles and Pescosolido’s contribution are described in more detail here.

Drescher, S., Burlingame, G., & Fuhriman, A. (1985). Cohesion: An odyssey in empirical understanding. Small Group Research, 16, 3-30.

Drescher et al. (1985) provide an important early review of theory and research on group cohesion. At that point in time, as the authors note, research on cohesion lacked theoretical and empirical integration, in part because of theoretical inconsistencies and measurement difficulties. For example, the authors point out that until that time, there had been a predominant focus in the literature on individual ratings of group cohesion (with little attention to leader, subgroup, or collective group ratings), the antecedents rather than consequences of cohesion, the use of self-report ratings rather than behavioral measurements, and cross-sectional studies that neglected the dynamic evolution of cohesion in teams over time. The authors go on to help address these past issues in the field by providing a theoretical framework that integrated past cohesion research into a more meaningful whole. The authors’ four-dimensional theoretical framework provided important guidelines that assisted future researchers in conducting cohesion research that better allowed for future tests of generalizability across different research domains. This blueprint for cohesion research was widely picked up by group cohesion scholars in the years to come and shaped how the field of research has evolved.

Evans, C. R., & Dion, K. L. (1991). Group cohesion and performance: A meta-analysis. Small Group Research, 22, 175-186.

After Drescher et al. (1985) lamented that group cohesion research had focused too much on the antecedents of group cohesion and not enough on the impact of group cohesion on team performance, a surge in research on the relationship between group cohesion and team performance occurred in the next few years. Evans and Dion (1991) documented the trends in this emerging line of research in their influential meta-analysis of the effects of group cohesion on group performance. Their widely cited conclusion was that cohesion has a moderately positive relationship with group performance.

Gully, S. M., Devine, D. J., & Whitney, D. J. (1995). A meta-analysis of cohesion and performance: Effects of level of analysis and task interdependence. Small Group Research, 26, 497-521.

In the years following Evans and Dion’s (1991) meta-analysis, a growing wave of research investigated the generalizability of the positive effects of team cohesion across different team contexts. The next meta-analysis by Gully et al. (1995) provided important nuance to the conclusions from Evans and Dion’s (1991) meta-analysis. In their meta-analysis, including 46 studies of group cohesion (compared to the 16 ultimately included in Evans & Dion’s [1991] meta-analysis), Gully et al. (1995) find that while cohesion generally has a positive relationship with performance, this relationship is more complex than commonly assumed in the literature at that point. They document that cohesion is most strongly linked to performance on interdependent tasks, and that the effects of cohesion in performance are also largely shaped by variations in the level of analysis used to conceptualize and operationalize cohesion. Namely, they find that the effects of cohesion on performance are stronger when both constructs are measured at the group level of analysis. Together, their meta-analysis marks an important turning point in the field, from which point onward more attention was given to task type and the importance of the level of analysis in cohesion research (and small groups research more broadly).

Carron, A. V., & Brawley, L. R. (2000). Cohesion: Conceptual and measurement issues. Small Group Research, 31, 89-106.

In the years following Gully et al.’s (1995) article, cohesion research continued to grow in a variety of different disciplines. The article by Carron and Brawley (2000) provided an important foundation piece in understanding and thinking about how existing cohesion research should best be applied across different disciplines and types of teams. They suggest that the discipline in which cohesion is investigated matters, and that researchers should not assume that definitions and operationalizations appropriate for one type of team will necessarily apply for other types of teams. Rather, when applying cohesion research to new settings, the onus is on the researcher to adequately pretest both theoretical and empirical aspects of cohesion in the setting they wish to examine. Carron and Brawley (2000) give explicit guidelines on how both definitions and measurements of cohesion can best be pretested and adapted to new contexts. Their guidelines have been helpful to researchers across a variety of disciplines, as cohesion has become a popular topic in a variety of team types, ranging from consulting project teams to musical bands. Their guidelines provide a useful reminder of the theoretical and empirical considerations researchers need to make when conducting interdisciplinary research on group cohesion and other topics.

Pescosolido, T. (2012). Cohesion and sports teams: A review. Small Group Research, 43.

In the final article in this special issue on group cohesion, Pescosolido (2012) reviews the effects of cohesion in research on sports teams up to the present day. He follows up on the query of Carron and Brawley (2000) to explore the emerging trends in specifically the discipline of sports research. He concludes that cohesion does not necessarily operate fundamentally differently in sports teams as Carron and Brawley (2000) suggested, but rather that in sports teams, compared to other team types, there are different contextual factors present. In other settings where these factors do not exist in the same manner, the effects of cohesion may be clouded, or masked. Therefore, in future research, Pescosolido advocates the importance of taking into account contextual factors, such as interdependence, team identification, colocation, and feedback immediacy, to gain a clear picture of the effects of cohesion in sports teams and beyond.

Conclusion

We hope that this special issue on group cohesion, encapsulating classic turning points in the literature and concluding with a state-of-the-art theoretical review of the role of cohesion in a cross-discipline setting—sports teams—offers you the opportunity to further your own thinking about the continuing role of cohesion in small group research. Your comments, ideas, and responses are welcome.