Affect and managerial performance: a test of the sadder-but-wiser vs. happier-and-smarter hypotheses - includes appendix

Administrative Science Quarterly, June, 1993 by Barry M. Staw, Sigal G. Barsade

1988 "Depressive realism: Four theoretical perspectives." In Lauren B. Alloy (ed.), Cognitive Process in Depression: 223-265. New York: Guilford.

Alloy, Lauren B., Lyn Y. Abramson, and Donald Viscusi

1981 "Induced mood and the illusion of control." Journal of Personality and Social Psychology, 41: 1129-1140.

Alloy, Lauren B., and Anthony H. Ahrens

Managers have long believed that the happy worker is a productive one, but decades of research on whether job satisfaction influences productivity have generally revealed a weak to nonsignificant relationship (Brayfield and Crockett, 1955; Iffaldano and Muchinsky, 1985). Only turnover and absenteeism have shown reliable linkages to satisfaction (Mobley, 1982). As a result of these findings, organizational researchers have had the unfortunate task of being the bearer of bad news, offering managers consolations such as, "satisfaction may predict decisions to participate if not decisions to perform" (March and Simon, 1958), or "satisfaction may follow from performance if contingent reward systems are in place" (Cherrington, Reitz, and Scott, 1971). It is unknown whether these consolations have been persuasive or whether we have been successful in convincing managers that job attitudes and performance are loosely coupled. Nonetheless, most researchers have long relegated the satisfaction-performance linkage to the folklore of management, as an unsubstantiated claim of practitioners and the popular press (Staw, 1986).

Recently, something interesting has happened that could reopen the issue of whether people's attitudes and performance are linked in organizations. While most of the field has assumed that the attitude-performance question was safely put to rest, the variables making up this relationship have undergone metamorphosis (Staw, Sutton, and Pelled, 1993). No longer is job satisfaction the only operationalization of attitudes at work. Instead, a number of researchers have been concerned with the expression of emotion on the job (Rafaeli and Sutton, 1989), positive and negative moods (Isen and Baron, 1991), and dispositional affect (Staw, Bell, and Clausen, 1986). Likewise, instead of considering job performance simply as a combination of work quantity and quality, other researchers have explored extrarole behavior (O'Reilly and Chatman, 1986), citizenship (Organ, 1988), and task revision (Staw and Boettger, 1990). These expansions of the construct space for both attitudes and performance now make it possible to test new linkages between these variables. In our view, one of the most promising reformulations of the attitude-performance question involves the study of affect in organizations. Before posing theoretical arguments or hypotheses, however, it is necessary to consider some definitional issues.

The Nature of Job Attitudes

Attitudes have historically been a broad construct used to denote cognitive, affective, and behavioral aspects of the relationship between a person and social, physical, or ideological objects (e.g., Katz and Stotland, 1959). Using this definition, one's attitudes toward his or her job could include a set of beliefs about the work (e.g., it is easy or challenging), an affective reaction to the work (e.g., one likes it or not), and behavioral intentions (e.g., likelihood of leaving or recommending the job to a friend). An obvious problem with this definition is that it is difficult to know where the attitudinal construct leaves off and behavior begins. As a result, many researchers have followed Fishbein and Ajzen's (1975) recommendation that cognition, affect, and behavior be separated as much as possible and that attitudes primarily reflect the affective component of the person-object relationship. Many organizational researchers have also followed this convention. In an influential treatise on job satisfaction, Locke (1976: 1300), defined job satisfaction as "a pleasurable or positive emotional state resulting from the appraisal of one's job or job experiences." Smith, Kendall, and Hulin (1969: 6), in their development of the JDI indicator of job satisfaction, similarly proposed that "job satisfactions are feelings or affective responses to facets of the situation."

When job attitudes are explicitly equated with affective states, it is not much of a logical extension to argue that job attitude research should be expanded by incorporating more varied research on affect. This implies that job attitudes mean more than job satisfaction. Candidates for the study of affective reactions at work should therefore include the range of moods, emotions, and dispositions experienced by people in the organization. Whereas emotions, such as anger or fear, generally denote a strong reaction to a specific object or cause, moods usually refer to a milder, more diffuse affective state that may not be directed toward any single attitudinal object (Lazarus, 1991). Likewise, an affective disposition commonly refers to a general tendency to experience a particular mood (e.g., to be happy or sad) or to react to objects (e.g., jobs, people) in a particular way (Lazarus, 1991). Thus, one might propose that job attitude research can be broadened by considering a wider range of emotions directed toward the work situation, such as anger at supervisors, frustration with policies, or enthusiasm for the product. Alternatively, the study of job attitudes could be enriched by delving more deeply into the affective dispositions and moods that people bring to the work situation, the route we follow in this paper.