Open - Ended Attributions in Team Competition

A total of 352 open-ended attributions were obtained in two field studies with volleyball teams and in two lab experiments, all involving team competition. All attributions were classified along the three causal dimensions of locus of causality, stability, and controllability. Attributions were also classified as referring to the self, to teammates, to the team as a whole, or to other factors and sorted into specific categories. A loglinear analysis revealed that attributions were predominantly internal, unstable, and controllable. A significant win/loss effect reflected the tendency for members of winning teams to use controllable, and particularly unstable, controllable, attributions more than members of losing teams. Overwhelmingly, attributions referred to the team as a whole rather than to individuals or other factors, and teamwork was an especially popular causal explanation. The findings suggest that research on attributions in team competition should focus on causal dimensions rather than the four traditional attributions of effort, ability, luck, and task difficulty, and that further attention should be given to team-referent causal explanations. Article: Investigations of success/failure attributions have been quite popular in the sport psychology literature and most of these sport attribution studies have drawn upon the theoretical work of Weiner, Frieze, Kukla, Reed, Rest, and Rosenbaum (1971). In that original work, Weiner and his colleagues identified the four standard causal attributions of ability, effort, luck, and task difficulty and proposed a twodimensional classification system with attributions classified as internal (ability and effort) or external (luck and task difficulty) and as stable (ability and task difficulty) or unstable (effort and luck). A number of predictions and relationships regarding achievement behavior emanate from that model, but to date, investigations of the implications for sport behavior are limited. The conventional research paradigm for sport attribution studies involves the assessment of postcompetition win/loss attributions by asking respondents to rate the importance of ability, effort, luck, and task difficulty. A number of studies using this approach have found that winners are more internal in their causal attributions than losers (Bird & Brame, 1978; Forsyth & Schlenker, 1977; Iso-Ahola, 1975, 1977; Lau & Russell, 1980; Roberts, 1975, 1978), and this trend is generally interpreted as a self-serving or egocentric bias. Several studies, however, suggest that even losers give predominantly internal attributions (Lau & Russell, 1980; Scanlan & Passer, 1980), and other investigators report that losers are actually more internal in their attributions than winners (Gill, 1980; Scanlan, 1977). The discrepant findings may reflect inadequate assessment of respondents' attributions and overreliance on a restricted attributional research paradigm. Despite sport psychology researchers' proclivity for Weiner's model and the four standard attributions, recent evidence indicates the inadequacy of that restricted approach for sport investigations. Weiner himself (Weiner, 1979; Weiner, Russell, & Lerman, 1978) has clearly stated that the four traditional attributions are not the only perceived causes of success or failure. In any achievement situation, other perceived causes may be equally or even more important. Researchers should consider the situation and investigate appropriate attributions. The findings of recent sport attribution studies support Weiner's admonitions. In an investigation of causal elements used in sport situations, Roberts and Pascuzzi (1979) noted that the four traditional attributions were used only 45% of the time. Bukowski and Moore (1980), using several alternative attributions along with the standard four, noted that while ability and effort were among the causes perceived as important, luck and task difficulty were not. They further suggested that more than the traditional reasons of ability, effort, luck, and task difficulty should be considered to adequately assess attributions for success/failure in athletics. Not only may success/failure in sport situations prompt different causal explanations than success/failure in other achievement situations, but group competition, a typical sport setting, likely elicits even more diverse attributions. The individual's relationship to teammates and the interplay of individual and team goals and responsibilities may prompt complex attribution patterns. Furthermore, if the achievement behaviors of the group are of concern, the investigation of attribution patterns and cognitive processes within groups is a logical approach. Just as the cognitive approach and attribution research have added to our understanding of individual achievement behavior, a cognitive approach may prove valuable in examining group achievement behavior. Indeed, the sport psychology research on group attributions indicates some differences between team and individual attributions even though that research seldom strays from the assessment of ability, effort, luck, and task difficulty. Generally the findings indicate that attributions for group success/failure are team-centered rather than self-centered. Iso-Ahola (1977) reported that team failure decreased the evaluation of team ability and effort but not individual evaluations. Discriminant analysis results of Bird and Brame's (1978) investigation indicated that three of the four attributions that discriminated between winning and losing teams were team attributions, with team ability being the most powerful discriminator. Gill (1980) observed that attributions of group success/failure to the own team or opponents were egocentric but assignment of responsibility within the team revealed a reverseegocentric pattern, and she interpreted these findings as reflecting a team-enhancing bias. The predominance of team-centered causal explanations in the group attribution studies and the implication that assessment of the four traditional causal factors is inadequate for sport attribution research prompted the current investigation. With the exception of Lau and Russell's (1980) investigation of attributional statements in newspaper reports of professional baseball and football games, researchers have not attempted to identify the specific causal explanations used by team sport participants. Thus, an exploratory investigation of win/loss attributions in team competition seemed reasonable. Open-ended attributions were obtained from members of winning and losing teams in two field studies and two lab experiments. As Elig and Frieze (1975, 1979) have noted, the open-ended format is especially appropriate for exploratory investigations and as a first step toward determining the relevant causal factors in a particular achievement situation. Ultimately, sport attribution research must go beyond the descriptive identification of causal factors to examine the theoretical and practical implications of varying attributions. Within the Weiner model, theoretical predictions and relationships depend on the causal dimensions rather than specific attributions. In the original model (Weiner et al., 1971), attributions were classified in terms of locus of causality or internality (within the person or external) and stability (varying over time or not). More recently, attribution theorists (Elig & Frieze, 1975; Weiner, 1979) have advocated inclusion of a third dimension of controllability, with attributions classified as under volitional control or not. The locus of causality, stability, and controllability dimensions relate to self-esteem, expectancy, and evaluation by others, respectively, and these relationships have implications for achievement behaviors. Thus, attributions obtained in the current study were classified along the three causal dimensions of locus of causality, stability, and controllability. Two major research questions were examined within the study; first, what types of causal explanations are given in team competition, and second, do the attributions of winning and losing team members differ? Because the study was exploratory, specific directional hypotheses were not examined either in terms of specific attributions or causal dimensions. In light of previous research, though, members of winning teams were expected to give more internal attributions than members of losing teams.