Attribution Theory

An Organizational Perspective

Mark Martinko

• Publisher: Routledge
• Year: 2018
• ISBN:
  9781351465137
• Edition: 1
• Topic: Business
• Subtopic: Human Resource Management
• Index: Business

With Special Contributions from Bernard Weiner Ph.D. (UCLA) and Robert Lord Ph.D. (Univ. of Akron)
Attribution theory is concerned with peoples causal explanation for outcomes: successes and failures. The basic premise is that beliefs about outcomes are a primary determinant of expectations and, consequently, future behavior. Attribution theory articulates how this process occurs and provides a basis for understanding that translates into practical action.
Attribution Theory: An Organizational Perspective serves as a primary sourcebook of attribution theory as it relates to management and organizational behavior. The text provides an integrated explanation of the role and function of attribution theory in the organization. This important new book contains original empirical research relating attributions to leader evaluations, reactions to information technologies, management of diverse work groups, achievement, and executive succession and power. The contributors are from a variety of disciplines including management, psychology, education, educational psychology, and sociology.

Section III

APPLYING ATTRIBUTION THEORY TO ORGANIZATIONAL ISSUES

Section III

APPLYING ATTRIBUTION THEORY TO ORGANIZATIONAL ISSUES

Part A:

Intrapersonal Behavior

6

THE INTERACTIVE EFFECTS OF NEGATIVE AFFECTIVITY AND A CAREER-IMPACTING PERFORMANCE OUTCOME ON SELF-SERVING ATTRIBUTIONS OF CAUSALITY

L. Alan Witt, Dana M. Broach, Thomas F. Hilton, and Chan M. Hellman

ABSTRACT

Subjects were 114 federal employees in an eight-week resident training program. It was hypothesized that scores on a measure of negative affectivity (NA) collected on the first day of the program would moderate the relationship between a pass/fail job-training outcome and self-serving attributions of causality regarding that outcome collected on the final day. To rule out likely competing explanations for the hypothesized effect, a higher-order interaction was also examined to include positive affectivity (PA): NA × PA performance outcome. Initial regression analyses partially confirmed the hypothesis, and the plotting of the significant higher-order interaction suggested that the interaction between NA and the performance outcome on causal attributions varied substantially among employees at different levels of PA. Results support four models previously introduced to explain the self-serving bias and deviations from it: the classic, depressed, realist, and optimist models.

INTRODUCTION

Attribution theory is predicated on a “thinking person” assumption, in that the individual is seen as motivated “to attain a cognitive mastery of the causal structure of his environment” (Kelley, 1967, p. 193) and is based on three assumptions. First, individuals will attempt to assign causes for important instances of behavior (e.g., receiving an appraisal of poor performance) and, when necessary, seek additional information in order to do so (e.g., talk with peers about the problem). Second, individuals will assign causal explanations in a systematic manner. Third, the particular cause that an individual assigns to an event has important consequences for his/her subsequent behavior (Jones et al., 1972). Thus, attribution theory pertains to the processes used by individuals to interpret events by using causal explanations. The issue is “why” as opposed to “what” or “what if” (Weiner, 1972). In other words, the issue of interest is the relationship between the circumstance (e.g., being fired) and the reasons (i.e., causes) for the circumstance. The attribution process involves post hoc reasoning, whereby a person infers the causes of behavior from the observation of the behavior or event (Steers and Mowday, 1981). This process is important, because causal attributions play a vital role in providing the input to action and the bases for deciding among alternative courses of action (Kelley, 1973).

These “thinking person” assumptions do not imply that attributional errors are not possible. Rather, they reflect the desire of individuals to accurately comprehend their social environment (Kelley, 1971). Indeed, while the desire to accurately comprehend the environment is strong, individuals will ultimately act in accordance with their attributions, regardless of the accuracy of perceptions of the causes of circumstances. This is an important point, as understanding others’ attribution processes is a key to interpersonal success on and off the job.

Adapting different motivational styles to fit the personalities of subordinates is a critical element of managerial performance (Davis et al., 1992), and identifying how an individual makes attributions is an important element of determining the person-motivation style fit. One factor limiting managers is the failure to recognize the facts that (1) individuals differ in their attributional processes and (2) people act on the basis of their attributions rather than someone else’s “reality.” Once these facts are accepted, however, the difficult task is to identify and understand how each subordinate makes attributions about critical outcomes. The purpose of the present study was to assess the extent to which the predisposition to experience negative affect (i.e., “negative affectivity”) and positive affect (i.e., “positive affectivity”) influences causal attributions regarding an important work outcome.

OUTCOMES AND ATTRIBUTIONS

Although Ash (1980) reported that self-assessments of ability strongly predicted performance, a consensus developed years ago that self-appraisals of job performance are sufficiently riddled with measurement problems (Heneman, 1980) to render them useless for practical use in organizations. Levine (1980) stressed that research on the process of self-assessment is needed in order to determine where the problems lie. He suggested that self-appraisals may falter on the basis of an inadequate judgment process. Researchers in social psychology have been examining self-assessment problems since Heider (1958) argued that individuals, in explaining the causality of an event, will place great importance on protecting their own self-esteem.

Social psychologists have focused on the self-serving bias effect (Arkin et al., 1980). The notion is simple: individuals, in explaining success and failure outcomes, are more likely to attribute success to personal qualities such as ability or effort and to attribute failure to situational characteristics, such as luck or difficulty of the task (e.g., Weary-Bradley, 1978; Weiner and Kukla, 1970; Zaccaro et al., 1987). However, evidence suggests that individuals in some situations hold themselves more responsible for failures than for successes (Rotter, 1966; Ross et al., 1974). Furthermore, there is a basic antinomy between the desire to preserve familiar and valued cognitive schemas and the degree to which such schemas are open to change (James et al., 1978). Indeed, the cost of self-enhancing distortions is high, because misperceptions can render us less able to remedy situations that cause us problems than do accurate perceptions (Nisbett and Ross, 1980). Given this cost, why do people distort events to protect their self-esteem? Recent work on negative affectivity might shed some light on individual differences in the attribution process.

AFFECTIVITY

Personality influences on the attribution process have received relatively little emphasis in the literature. One notable exception is a study of high school students by Meyer (1970), who demonstrated the effect of individual differences in achievement motivation on the impact of success and failure on causal attributions.

In contrast, researchers (e.g., Weiner et al., 1978, 1979) have emphasized the effects of specific outcomes on state affect (i.e., mood). Golwitzer and Stephan (Golwitzer et al., 1982; Stephan and Golwitzer, 1981), however, focused on affect as an antecedent rather than as an outcome, showing that state negative affect influences causal attributions. Recent developments in the affect literature, however, suggest that affective predisposition influences responses to situations, which may also help explain some causal attributions.

In studies of self-reported mood, negative affect and positive affect have consistently emerged as two separate and relatively independent dimensions (Diener and Emmons, 1985; Watson and Tellegen, 1985). Watson and Clark (1984) argued that the tendencies to experience negative or positive affect reflect stable dispositions—negative affectivity (NA) and positive affectivity (PA). Studies on twins reared apart suggest that both PA and NA are inherited (Tellegen et al., 1988). Evidence suggests that PA and NA are key reasons why individuals experience positive and negative moods (Costa and McCrae, 1980; Watson and Clark, 1984). Watson and Clark (1984) noted that persons high in PA are characterized by high energy, full concentration, and active engagement, whereas low-PA individuals typically experience sadness and lethargy.

Marco and Suls (1993, p. 1060) suggested that “NA is first and foremost a prevailing state in that high NAs experience greater baseline levels of negative affect even in the absence of stress.” High-NA individuals focus on their disappointments, shortcomings, and mistakes, adopting a more negative view of life experiences. Low-NA persons, on the other hand, tend to be more self-secure, satisfied, and calm, focusing less on daily frustrations and adopting a more resilient approach to setbacks (Watson and Clark, 1984).

In contrast to researchers linking neuroticism with NA (e.g., Bolger and Schilling, 1991), we concur with Watson and Clark’s (1984) view that NA represents subjective differences in temperament, mood, and cognitive orientation rather than an index of psychological health.

Previous research suggests that in conditions of failure, high-NA individuals might make more internal attributions than low-NA persons. In a study of the affective states of 24 college students, Sharp and Tennen (1983) found that depressed students assigned greater personal responsibility for failure than those who were not depressed. They argued that high-NA personnel have a depressed attributional style, such that they identify causes that are most unfavorable to themselves. Other research, which indicates that persons high in NA are more reactive to situational cues (Larsen and Ketelarr, 1991; Marco and Suls, 1993; Parkes, 1990; Witt, 1991, 1994), also suggests that high-NA individuals would report disproportionately higher levels of internal causality for failure. Their findings, however, offer an alternative explanation: high-NA individuals experiencing failure might be more likely to overemphasize their own shortcomings and mistakes because of an overreaction to failure.

Unfortunately, it is not clear how NA might affect the self-serving bias in conditions of success. If the depressed attributional style of high-NA persons leads them to identify causes that are most unfavorable to themselves (Sharp and Tennen, 1983), then they are likely to attribute their success to external factors. In contrast, if individuals high in NA are more sensitive to situations (Larsen and Ketelarr, 1991), then they might tend to overreact and overestimate the impact of their efforts when succeeding (i.e., make more internal attributions).

Alternatively, Levin and Stokes (1989) argued that high-NA people may be able to make more realistic appraisals of their job experiences than low-NA people. Accordingly, they might be less prone to making self-enhancing distortions and thus be better able to address problems.

The literature, then, is not clear as to the effect of NA on the self-serving bias effect: depressed attributional style, greater sensitivity to situational cues, or greater accuracy in assessing events at work? The purpose of the present study was to investigate this issue.

Hypothesis 1: Following research showing moderating effects of NA on situation-behavior relationships (e.g., Parkes, 1990) and Sharp and Tennen’s (1983) college student study suggesting a “depressed attributional style,” it was hypothesized that NA would moderate the relationship between outcome and causal attributions. However, due to the conflicting findings in the literature, the form of the interaction was not predicted.

Assessing the “depressed attributional style” explanation requires the measurement of PA. The individuals predisposed to experience depressive states would likely report both low PA (low levels of activation) and high NA (high levels of anger and anxiety).

Hypothesis 2: Based on Sharp and Tennen’s (1983) notion of “depressed attributional style,” it was hypothesized that PA would moderate the NA × outcome interaction. Specifically, employees likely predisposed to experience depressive states (i.e., those high in NA and low in PA) will make comparatively more internal attributions for failure and external attributions for success than employees at other points on the PA and NA continua.

POSSIBLE CONFOUNDS

Consistent with previous findings of an expectancy-attribution link (see Weiner, 1986), the individual’s expectancy for success could potentially confound the effect that was hypothesized. An important component of expect...