Actor-observer bias and perceived sensitivity to internal and external factors in competitive swimmers.

Wolfson, Sandy

Research into social perception has demonstrated that people are generally inclined to focus on internal attributions when explaining behaviour, although they tend to be more willing to use external attributions when analysing their own behaviour as opposed to the behaviour of others. For example, in the classic study by Nisbett et al. (1973), students made internal attributions to explain their friends' reasons for studying a particular university course, but relatively more external attributions to explain their own reasons.

Although there is considerable evidence in mainstream psychology for such actor-observer differences, the bias has been given very little attention in a sports context. Yet the process has potential implications for performance and for relationships between sports participants and their teammates and coaches. Firstly, if people do indeed focus relatively more on situational factors for themselves, they may give insufficient weight to internal influences such as effort or skill, thus limiting improvement-related strategies. Secondly, coaches might systematically overlook situational influences on their players and come to underestimate the impact of such factors as venue, climatic conditions, and crowd behaviour. Most importantly, the disparity in actor-observer explanations might very well lead to interpersonal conflict and hostility between players and their teammates or coaches, as each person will believe that his or her analysis of an outcome is the correct one and that the other lacks knowledge and insight.

Explanations for actor-observer bias help to clarify the problem in relation to sport. From a motivational perspective, it may simply be less effortful and time consuming for the "cognitive miser" (Heider, 1958) to label and conceptualise people as having particular characteristics and dispositions. This view suggests that unless performers and coaches are particularly motivated to ponder someone else's situation, the easiest option is to ascribe traits and focus on their causal effects. Along these lines, Prager and Cutler (1990) found a positive relationship between level of acquaintance and the use of external attributions.

Alternative explanations suggest that people obtain and process information differently for others than for themselves, with situational obstacles and internal responses to external stimuli more available for self analyses. For example, a swimmer will be acutely aware of an unusually low water temperature, but such information is unavailable to the observer. Furthermore, situational features are more salient to the actor, causing these to be seen as the cause of an event, while other people are prominent for the observer, who thus sees others as the cause of their own behaviour. McArthur and Post (1977) increased internal attributions by making people more salient through illumination, movement or vividness, and Storms (1973) used videotapes to alter the vantage points and thus the attributions made by actors and observers.

Sande, Goethals and Radloff (1988) have suggested that actor-observer differences might emanate from people's views that they are more complex and multi-faceted than others. This re-conceptualisation of the attribution process contends that people believe that they have more traits than other people, not that they are less governed by internal factors, and thus that more factors, both internal and external, will affect them than other people. This is likely to be due to a combination of motivational and information processing effects.

Although little research has examined actor-observer differences in a competitive context, Luginbuhl and Bell (1989) provided evidence for sports performers' general tendency to focus on dispositional explanations for the behaviour of others. In this research, athletes tended to give internal reasons for another athlete's unexpectedly poor performance, particularly if the performance was not in their own specialty sport. The researchers made no attempt to examine self attributions and thus could not comment specifically on actor-observer bias. However, the researchers suggested that when competitors made attributions about someone in their own sport, they not only had increased information about potential situational factors but were also motivated to empathise with and thus consider the plight of the fellow sports participant.

The present study was designed to test the hypothesis that competitive swimmers would focus on internal attributions when analysing outcomes, but that they would be relatively more external in their attributions for themselves than for their fellow competitors. In keeping with Sande, Goethals and Radloff's (1988) contention, it was also hypothesised that swimmers would perceive a wider variety of factors affecting themselves than other people.

Method

Design

A 2x2 repeated measures factorial design varied the target of attribution (self/others) and the locus of causality of attribution (internal/external).

Subjects

The subjects were 34 male (N = 17) and female (N = 17) finalists participating in the British Swimming Grand Prix at Newcastle-upon-Tyne City Pool in England. The event was one of six meets in the British Grand Prix circuit, organised to bring together top swimmers for the accumulation of points which determine participation in the grand final at the end of the season. Subjects were finalists in a range of events and distances. The mean age was 17.46.

Attribution Questionnaire

A three-page questionnaire was developed to measure the swimmers' attributions. The first page included items about background and experience. The remaining two pages asked subjects to rate the importance of ten factors which might have determined the outcomes for themselves and for their fellow competitors.

The instructions for the page which referred to the subjects' self attributions stated, "Many things affect how people do in a competition. You are now going to be asked which things you think were important today. How important do you think each of the following was in determining how you did today?"

The other page asked subjects to indicate how the same factors affected other people who had participated in the event. The instructions were identical except for the last sentence, which was altered to read, "Think about the other competitors who took part in your event today. How important do you think each of the following was in determining how the other competitors did today?"

Five items on the scale (skill, effort, mood, training put in, and personality) were classified as internal and the remainder (luck, water temperature, people watching, time or day of event, and coaching received) as external. The items were selected after discussions with club and national coaches about common attributions used to explain outcome amongst swimmers. Ratings were made using a Likert-style format with a 7-point not at all important to very important scale. Subjects were also given the opportunity to describe any additional factors which they thought affected them or their competitors. The order of self-others presentation was counterbalanced.

Procedure

The competition took place over three days, with heats for events in the mornings and finals in the late afternoons.

After swimmers had competed in their finals and changed clothes, they were asked to volunteer to fill in a three-page questionnaire regarding their perceptions of the competition. Prospective subjects were assured of the confidentiality of their responses, and no names were taken. Fifty swimmers were approached, and of these, 35 agreed to participate. The main reason given for refusal involved lack of time. One subject failed to complete all three pages of the questionnaire and was eliminated from the analysis.

After completing the questionnaire, subjects were thanked for their help, debriefed, and reminded that their individual responses would remain anonymous. They were also asked not to talk to other swimmers about the questionnaire until after the competition came to a complete finish. Finally, they were told not to volunteer again if they qualified for a second final in a later event.

Results and Discussion

There were no effects due to target order or to sex of subject.

The total internal and external scores were separately calculated for each individual and subjected to a 2 x 2 repeated measures analysis of variance, varying target of attribution (self/others) and the locus of causality of attribution (internal/external). The results revealed main effects for both factors [ILLUSTRATION FOR FIGURE 1 OMITTED]. Overall, participants rated the factors as more important for themselves (M = 24.66) than for others (M = 22.81) (F(1,33) = 14.02, p[less than].001). The internal factors (M = 28.31) were more strongly endorsed than the external factors (M = 19.16) (F(1,33) = 140.37, p[less than].001). The interaction was not significant.

The results do not support the classic actor-observer effect, as subjects were not more likely to use external attributions for themselves than for others. Instead, the swimmers rated themselves as more affected than others by both the internal and external factors. This supports Sande, Goethals and Radloff's (1988) contention that people see themselves as personally more susceptible to a wide variety of influences.

The main effect for locus of causality indicates that the swimmers were more likely to endorse the internal factors on the scale. Given that all competitors had reached the final of their event, this is consistent with the findings of Mullen and Riordan's (1988) meta-analysis, where internal attributions generally occurred for successful outcomes but external explanations were not consistently given for failure in studies of performance in naturalistic settings.

Notwithstanding, the classification of attributions is not a straightforward matter. While it might have seemed straightforward for Weiner (1986) to label "luck" as external and uncontrollable, some competitors might perceive "luck" as internal ("I am generally an unlucky person") and even, with the use of rituals and good luck charms, at least partly controllable. Given the potential misinterpretations, some theorists (see Biddle, 1988) have suggested that it might be presumptuous for researchers to impose their own classifications onto others' attributions. Russell (1982) developed the Causal Dimension Scale, in which people are asked to indicate causes for an event and then to rate each cause's stability, externality and controllability. Similarly, the Sport Attributional Style Scale (Hanrahan, Grove, and Hattie, 1989) gives people the opportunity to classify their own attributions for positive and negative events. However, these instruments might be confusing to the average person, and their reliability has also been questioned (Biddle, 1988).

The main effect for locus of causality must, therefore, be interpreted with caution, and an examination of the impact of single attributions might be considered. In order to assess whether certain attributions differentially contributed to the self-others main effect, a 2 x 10 repeated measures analysis of variance was carried out on the ratings for the ten attributions for self and other. No interaction was found, indicating that single attributions did not contribute differently.

One example of this intriguing effect relates to the perceptions of water temperature. Swimmers are objectively aware that competitors in an event operate under identical thermostatically regulated conditions. Nonetheless, the swimmers in this study believed that water temperature had a greater effect on themselves (M = 4.26) than on their rivals (M = 3.00) who were swimming in the same water at the same time (t = 2.37, df=33, p[less than].02). This assumption is consistent with the suggestion that because actors have so much more information about the effect of situational factors on themselves, they come to believe that they are more sensitive and multi-dimensional than others.

An implication of the research is that accomplished competitors are unlikely to deny personal responsibility for an outcome. However, since actors do appear to feel uniquely susceptible to a wide range of influences, interpersonal conflict could result if competitors confer with their teammates and coaches about their analysis of a result. For example, swimmers whose own analyses of outcome are not shared by their teammates and coaches might feel resentful over the ostensible lack of sympathy. At the same time, as observers of the self-analysis, the teammates and coaches might feel that the swimmer lacks insight. This suggests that participants involved in the competitive process could benefit from an awareness of the biases inherent in attributional processes. If people are alerted to the need to empathise with others and imagine their perspective, many potential misunderstandings might be prevented.

References

Biddle, S. (1988). Methodological issues in the researching of attribution-emotion links in sport. International Journal of Sport Psychology, 19, 264-280.

Hanrahan, S.J., Grove, J.R., & Hattie, J.A. (1989). Development of a questionnaire measure of sport related attributional style. International Journal of Sport Psychology, 20, 114-134.

Heider, G. (1958). The Psychology of Interpersonal Relationships. New York: John Wiley and Sons.

Luginbuhl , J., & Bell, A. (1989). Causal attributions by athletes: role of ego involvement. Journal of Sport and Exercise Psychology, 11, 399-407.

McArthur, L.Z., & Post, D.L. (1977). Figural emphasis and person perception. Journal of Experimental Social Psychology, 13, 520-533.

Mullen, B., & Riordan, C.A. (1988). Self-serving attributions for performance in naturalistic settings: a meta-analytic review. Journal of Applied Social Psychology, 18, 3-22.

Nisbett, R.E., Caputo, C., Legant, P., & Maracek, J. (1973). Behavior as seen by the actor and by the observer. Journal of Personality and Social Psychology, 27, 154-164.

Prager, I.G., & Cutler, B.L. (1990). Attributing traits to oneself and others: the role of acquaintance level. Personality and Social Psychology Bulletin, 16, 309-316.

Russell, D. (1982). The Causal Dimension Scale: a measure of how individuals perceive causes. Journal of Personality and Social Psychology, 42, 1137-1145.

Sande, G.N., Goethals, G.R., & Radloff, C.E. (1988). Perceiving one's own traits and others': the multifaceted self. Journal of Personality and Social Psychology, 54, 13-20.

Storms, M.D. (1973). Videotape and the attribution process: reversing actors' and observers' points of view. Journal of Personality and Social Psychology, 27, 165-175.

Weiner, B. (1986). An Attributional Theory of Motivation and Emotion. New York: Springer-Verlag.