WHAT ABOUT GROUP COHESION? Group cohesion is a construct widely studied in the group dynamics literature. Extensive literature reviews show that there is a lack of consistency and agreement regarding the concept and its use among the researchers who created this literature (Friedkin, 2004; Greer, 2012). Despite the abundance of literature on this concept, there are still divergent opinions on its nature, its typology and its effect on performance and productivity. Research on group cohesion has continued to garner interest and popularity (Greer 2012) as a concept widely studied in the academic literature. The group cohesion-group performance relationship has been studied extensively and the findings of early researchers did not present a methodical link between performance and cohesion (Forsyth, 1990; Stogdill, 1972). Although two meta-analytic studies have claimed that there is a positive relationship between group cohesion and group performance (Evans and Dion, 1991; Mullen and Copper, 1994). However, subsequent studies differed from these meta-analyses in whether the cohesion-performance relationship was moderated by other variables such as level of analysis, task interdependence, goal acceptance, and task norm. group (Gully, Devine, & Whitney, 1995; Langfred, 2000; Podsakoff, Mackenzie, & Ahearn, 1997). One explanation for this uncertainty in the literature was the non-uniformity of definitions and measures of cohesion and performance (Cota, Evans, Dion, Kilik, & Longman, 1995; Mudrack, 1989a, 1989b). The context of group cohesion remains broad, and researchers have continued to propose varying definitions and conceptual models of group cohesion. For example, group cohesion has been defined as "the total field of forces causing... middle of paper ...... the group in three weeks." This observation leads us to question the role of performance during the first week following the formation of a group. Taylor, Castore, and Tyler (1983) examined how performance can be maintained when groups experience success or failure. They observed that the impact of negative performance feedback on cohesion depends on the attributions that group members make about the causes of failure. Attributions can be described as the explanations that people formulate regarding their own and others' behavior (Taylor et al. 1983). Their data demonstrate that cohesion can remain high, despite repeated failures and little success. This may be due to group diffusion attribution biases. However, there appears to be little research examining the role of initial levels of group cohesion and the impact of performance on subsequent levels of group cohesion..