Attributions are explanations about why particular performances or behaviors have occurred. When faced with important, negative, novel, or unexpected events, individuals search for meaningful explanations for the causes of those events. In this regard, it is widely acknowledged that attributions are an area of importance in the field of applied psychology because of their implications for motivation and emotion. This entry discusses the historical theories that have provided the framework for research into, and application of, attribution theory in sport and exercise settings, then moves on to highlight recent theoretical advances, assessments of attributions, empirical evidence for attribution theory, and the application of attribution theory.
If a particular individual were to be singled out as the founder of the scientific study of attributions, it would be Fritz Heider. He proposed that people explore explanations for events or behaviors to increase their control over the environment and to satisfy a desire to understand and gain knowledge about the world. His insights provided the impetus for numerous theories and investigations. Three of the most well-known are Edward Jones and Keith Davis’s correspondent inference theory, Harold Kelley’s covariation model, and Bernard Weiner’s attributional theory of achievement motivation and emotion.
Correspondent inference theory proposes that people infer dispositions and intentions of others, as a result of observing their behavior. That is, behavior is seen as corresponding to or reflecting an underlying disposition of the actor. In general, correspondence is high when behavior is atypical and has clear implications. In other words, behavior is informative to the extent that it is seen to involve choice among alternatives. Whereas correspondent inference theory focused mainly on person perception (or attributing to others), the covariation model made contributions to our understanding of self-perception. The covariation model suggests that people arrive at a cause for an event by processing information about whether accompanying conditions and circumstances vary or not as the event changes. According to the model, people use three types of information— consistency, distinctiveness, and consensus—to verify whether they have correctly linked causes and effects.
The approach that has had most influence on attribution research in sport and exercise psychology is the attributional theory of achievement motivation and emotion. A central premise of this theory is that there is a dimensional structure underpinning the explanations people give for events. Building on previous works in this field, Bernard Weiner, a key figure in conceptualizing the attributional categories into dimensions, initially identified four main attributions (or attribution elements), namely, ability, effort, task difficulty, and luck; these were then classified into two dimensions, locus of causality and stability. Locus of causality refers to whether the cause is inside (internal) or outside (external) the person; stability refers to whether the cause will (unstable) or will not (stable) change over time. Later deductive theorizing led to the identification of a third dimension, controllability, referring to whether the cause could be viewed as controllable or uncontrollable.
Recent Theoretical Advances
Despite warnings by a number of researchers that the three dimensions of locus of causality, stability, and controllability may not be appropriate for all types of situations (e.g., sport and exercise), until recently, there has been little effort to examine alternative approaches. In 2005, Tim Rees, David Ingledew, and Lew Hardy published an article that examined the congruence between theory, research, and practice of attributions in sport. On the basis of their observations, they proposed a broader conceptual approach to our understanding of attributions. They encouraged (a) a focus on the controllability dimension, (b) an expanded conceptualization of generalizability attributions (in addition to stability, examining globality and universality attributions), and (c) exploring interactive effects of controllability and generalizability attributions upon performances or behaviors.
The focus on controllability is based on empirical and theoretical grounds. Empirically, studies in a wide array of domains have shown that perceived personal control is an important psychological predictor. For example, within general social psychology, perceived personal control has been shown to affect a person’s level of depression, loneliness, and shyness, and in sport an athlete’s level of subsequent self-efficacy. Theoretically, the need to exert control over future events was foundational to early attribution theorizing, with Harold Kelley commenting that the purpose of causal analysis is effective control. Further, drawing upon attribution theory, the expectancy of future uncontrollability is at the heart of Lyn Abramson, Martin Seligman, and John Teasdale’s reformulation of the learned helplessness model: The expectancy of future uncontrollability is the most direct determinant of helplessness.
The second proposal is that, in addition to the stability dimension, attribution research in sport and exercise should examine the dimensions of globality and universality. The addition of globality refers to whether the cause affects a wide range of situations with which the person is faced (a global attribution) or a narrow range of situations (a specific attribution); universality refers to whether the cause is common to all people (a universal attribution) or unique to the individual (a personal attribution). This leads to an expanded conceptualization of generalizability: In addition to whether causes generalize across time (stability), attribution research should examine whether causes generalize across situations (globality) or all people (universality).
The final proposal is that researchers need to move beyond testing solely main effects of attributions to exploring interactive effects of controllability and generalizability attributions upon performances or behaviors. Until recently, the focus of much attribution research had been upon main or independent effects of attribution dimensions. To model generalizability, however, implies the need to consider interactive effects. Interactions of attribution dimensions may well be important because attributing, for example, less successful performances to uncontrollable causes may only lead to negative effects when causes are also considered to be stable (unlikely to change over time), global (likely to affect a wide range of situations), or personal (unique to the individual). For example, an athlete attributing poor performance to poor concentration might say, “There was nothing I could do about it” (an uncontrollable attribution), together with “and this will never change” (a stable attribution), or “and this affects a lot of situations I find myself in” (a global attribution), or “and this only happens to me” (a personal attribution). In this instance, the athlete might well be expected to experience poorer subsequent performance. Conversely, higher levels of performance would be expected if the athlete were to combine this uncontrollable attribution with “but this will change” (an unstable attribution), or “however, this only affects a few situations I find myself in” (a specific attribution), or “but this affects everyone, not just me” (a universal attribution).
Various methods have been employed in the measurement and categorization of attributions. Open-ended methods involve the researcher categorizing the oral replies of participants to openended questions. Derived score methods require the participant to rate reasons for a success or failure on five-point scales for different elements (e.g., ability, effort, task difficulty) related to the attribution dimensions. Problems can arise, however, when researchers try to summarize attributions along dimensions or otherwise assume the dimensional categories of attributions. Recent attribution measures, such as the Causal Dimension Scale (CDS), the Causal Dimension Scale II (CDSII), and the Measure of Controllability, Stability, Globality, and Universality Attributions (CSGU), have used direct rating methods. These methods require the participants to state their reasons for the event and then map those reasons onto items referring to attribution dimensions.
The CDS was developed to reflect Weiner’s three-dimensional model of attributions. A number of methodological criticisms have been leveled at the CDS, in particular concerns over the nature of the controllability subscale, which contains items referring to controllability, responsibility, and intentionality. The revision of the CDS (the CDSII) focused on changes to the controllability items, with all six of the items from the locus of causality and stability subscales left unaltered. In the CDSII, the controllability dimension was subdivided into personal control (control by the actor) and external control (control by others). Based upon recent proposals in the sport and exercise psychology literature, measures of attributions should include four scales for controllability and the generalizability dimensions of stability, globality, and universality. Following this, researchers developed the CSGU, a 16-item, 4-factor measure of attributions that assesses the four dimensions of controllability, stability, globality, and universality.
Empirical Evidence for Attribution Theory
A number of research articles have been published, exploring the following key propositions of the attributional theory of achievement motivation and emotion: (a) the locus-of-causality dimension affects intrapersonal emotions including pride and shame; (b) the stability dimension affects expectations of future success and feelings of hopefulness; and (c) the controllability dimension affects interpersonal emotions including blame, anger, and pity. Despite empirical support, these propositions are limited to the extent they explore the effects of attribution dimensions in isolation. In other words, the proposals focus solely upon main effects of attribution dimensions upon outcomes.
Recent research has reported the interactive effects of attribution dimensions upon outcomes. Collectively, the research demonstrates that the independent effects of attribution dimensions upon outcomes—such as emotions, self-efficacy, and performance—may be conditioned by interactive effects. For example, research has demonstrated that, following failure, higher levels of perceptions of controllability may only lead to beneficial effects if causes are also considered to generalize, such as across time (interaction of controllability and stability).
Application of Attribution Theory
Attributional retraining involves attempts to change maladaptive explanations for outcomes (attributing failure to uncontrollable or stable causes) toward more adaptive explanations (attributing failure to controllable or unstable causes). Although there is discrepancy in the literature regarding whether negative behaviors or outcomes should be perceived as inside the individual (internal locus of causality) or outside the individual (external locus of causality), there is agreement in regard to the need to perceive such causes as controllable. For example, following failure like a loss in sport or a relapse from a positive behavior like smoking a cigarette following a period of abstinence, an individual is encouraged to attribute the negative outcome or behavior to causes that are within one’s control. In situations where such causes are perceived as outside one’s control (i.e., a maladaptive attribution), the procedure of attributional retraining involves practitioners manipulating causes toward aspects that are within one’s control.
Attribution theory has a rich history in general (social) psychology, and has significantly influenced applied practice in sport and exercise settings. Recent theoretical advances, supported by empirical evidence, suggest a promising future for attribution theory to continue as a principal theory in sport and exercise psychology.
- Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49–74.
- Biddle, S. J. H., Hanrahan, S. J., & Sellars, C. N. (2001). Attributions: Past, present, and future. In R. N. Singer, H. A. Hausenblas, & C. M. Janelle (Eds.), Handbook of sport psychology (2nd ed., pp. 444–471). New York: Wiley.
- Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley.
- Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The attribution process in person perception. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 2, pp. 219–266). New York: Academic Press.
- Kelley, H. H. (1967). Attribution theory in social psychology. In D. Levine, Nebraska symposium on motivation (Vol. 15, pp. 192–240). Lincoln: University of Nebraska Press.
- Rees, T., Ingledew, D. K., & Hardy, L. (2005). Attribution in sport psychology: Seeking congruence between theory, research and practice. Psychology of Sport and Exercise, 6, 189–204.
- Weiner, B. (1986). An attributional theory of achievement motivation and emotion. New York: Springer.