Other measuring techniques, apart from self-report questionnaires, include personal interviews, in which employees discuss various aspects of the job with a supervisor or an interviewer from the personnel department and even, psychophysical tests in which the relationship between the attributes of physical stimuli and the psychological sensations that these stimuli produced are examined. The latter technique is, therefore, meant to assess the employee’s affective response toward a particular attitude object on the basis of his or her bodily response.
But, given all these techniques for measuring job satisfaction, is there a direct linkage, if any, with any scientific research about attitudes?
Relationship between Job Satisfaction Measures and Scientific Research on Attitudes
Job satisfaction measures and the tri-partite model of attitudes. Eagly and Chaiken (1993) argue how social scientists have often assumed that responses that express evaluation and therefore disclose people’s attitudes can be or should be divided into three classes, namely cognition, affect and behavior. Summarily, the cognitive category contains beliefs, theories, expectancies, cause and effect beliefs, and perceptions relative to the attitude object. The affective category, on the other hand, consists of feelings or emotions that people have in relation to the attitude object, whereas the behavioral category covers up basically people’s actions with respect to the attitude object.
Importantly, evaluations are often considered the central component of attitudes. Evaluations consist of the imputation of some degree of goodness or badness to an attitude object. When we speak of a positive or negative attitude toward an object, we are referring to the evaluative component, namely a function of cognitive, affect and behavioral intentions of the object. It is most often the evaluation that is stored in memory, often without the corresponding cognitions and affect that were responsible for its formation.
With regard to measures of job satisfaction and their potential link with the above-mentioned tri-partite model of attitudes, Arnold, Cooper and Robertson (1998) explain how scales like the JDI, the Job Satisfaction Scale (JSS) and the more recent job satisfaction scale of the Occupational Stress Indicator (OSI), for instance, “all involve questions or statements asking respondents to indicate what they think and/or feel about their job” (Arnold et al., 1998, p.204), thus mirroring the affective and cognitive components that, together with the behavioral component, had been theorised to contain attitudes. Interestingly, it might not always be the case that self-report questionnaires consist of items relating to all the three components of an attitude, but would rather include items pertaining to either one or two of the three components.
But, given the tri-partite theoretical model of attitudinal responding, does it make sense to have measures of job satisfaction that merely measure attitudes on account of one or two components, rather than on all three? Indeed, whereas it has been asserted that attitudes are manifested in cognitive, affective and behavioral responses, nonetheless, this tripartite view of attitudinal responding has raised a number of important questions. One question is whether attitudes must have all three of these aspects at the point of attitudinal responding.
Implications of the cognitive, affective and behavioral analysis on job satisfaction measures. In line with Eagly and Chaiken’s (1993) argument, although the older three-component definition of attitudes may have implied that these three aspects must be in place in order for a true evaluative tendency to emerge, attitudes can be formed primarily or exclusively on the basis of any of the components. In this regard, having job satisfaction questionnaires measuring one or two components of an attitude – for instance, the cognitive and the affective – rather than all three components, seems to parallel very well with the idea of consistency, that is to say, the notion that people tend to express about the same degree of evaluation of an attitude object through responses of each of the three classes.
Nonetheless, Eiser (1987) suggested that cognitive, affective and behavioral responses will be evaluatively consistent to the extent that all three response classes contributed to the initial formation of the attitude. This would mean that if a particular attitude was acquired via a cognitive route, for instance, it would be mistaken to measure one’s attitude toward the same object on the basis of affective or behavioral components.
However, Millar and Tesser (1986a) cited in Ajzen and Fishbein (1980) maintain that whereas an attitude acquired via a cognitive route might tend to elicit primarily cognitive responses, and one acquired via a behavioral route might tend to elicit primarily behavioral responses, any strong one-to-one relationship of this sort is quiet improbable. Indeed, it has also been suggested, in particular by research on the cognitive-affective interface, that different classes of evaluative responses have an impact on one another and exist in what might be described as a cooperative, synergistic relation. One may, for example, acquire beliefs about an attitude object, think about this knowledge, and thereby decide upon a course of action or generate an emotional response.
We can, therefore, infer that job satisfaction measures consisting of items relating to any one of the components of attitudes, rather than all the three aspects measured altogether, are fairly consistent with the theoretical idea that different classes of evaluative responses do not exist independently of each other but do, indeed, impinge on one another.
Theoretical basis for the response biases in measures of job satisfaction
The view that attitudes are ordinarily expressed in cognitive, affective and behavioral responses, leads one to expect that people’s attitudes are positively correlated with the evaluative implications of their overt behaviours. People who hold positive attitudes should engage in behaviors that approach, support, or enhance the attitude object, and people who hold negative attitudes should engage in behaviors that avoid, oppose or hinder the object.
Anderson et al. (2001) explain that no social scientist would expect consistently high correlations between attitudes and behavior and that as many researchers have noted, attitudes represent only one of the several important classes of variables that guide overt behaviors. Indeed, Fishbein and Ajzen (1975) cited in Brehm, Kassin and Fein (2001) suggest that the immediate cause of behavior is one’s intention to engage in the behavior. Attitudes influence behavior by their influence on intentions, which are decisions to act in a particular way. In interposing intention between attitudes and behaviors, Brehm et al. explain how Fishbein also chose to negate the possibility that attitudes sometimes elicit behavior with little or no intervening thought.
It is important to note that in Ajzen’s theory of planned behavior, the class of attitudes taken into account is attitudes toward behaviours, not attitudes toward the focal object. Attitude toward the behavior enters the model as one of the determinants of intention. The other determinant of intention, called subjective norm, consists of a person’s belief about whether significant others think that he or she should engage in the behavior. Hence, behavioral intention is a linear regressive function of the attitude toward the act or behavior and the subjective norm.
Therefore, given that one’s attitude toward a particular act or behavior is based both on one’s intention as well as one’s subjective norm, measuring one’s attitude toward a particular target on the basis of one’s behavioral evaluative responses might not be a very good predictor of one’s actual attitude toward the object.
Indeed, Eagly and Chaiken (1993) explain how challenges to issues of reliability and validity in relation to the techniques in attitude measurement, seem to result from the possibility that people may avoid answering questions or distort their reports to protect their privacy, to gain economic advantage, to obtain social approval and avoid social disapproval by co-workers, and to project or protect particular identities, amongst other possible reasons. This would inevitably imply that self-report instruments in measuring job satisfaction might not be measuring an employee’s attitude toward a particular object as such, but rather his or her attitude towards the behavior of the target attitude.
Other measures of job satisfaction, such as interviews or any of the psychophysiological tests mentioned above, seem to be also prone to this bias. For instance, one’s attempt to measure attitudes through bodily activity, presumed to be the physiological expression of one’s emotions toward the attitude object, might seem to be inadequate in that one’s emotional response might not only reflect a person’s emotions with respect to his or her attitude toward the focal object, but possibly also the orienting response that is set off by certain factors such as surprise, change, inconsistency, novelty, or by any other unforeseen variables.
Conclusion
Through what had been discussed thus far, it seems that measures of job satisfaction reside quiet comfortably within the common theoretical assumption that one’s attitude toward a particular aspect can be measured in terms of a person’s affective, cognitive and behavioural evaluations – or parts of it – toward the attitude object.
Nonetheless, while certain job measuring techniques, like the JDI, proved to be “reliable and has an impressive array of validation evidence behind it” (Anderson et al., 2001, p.32), I think one should not overlook the crucial value of taking into consideration any variables that might act as a go between the employee’s actual attitude toward a particular facet and his or her response on a job satisfaction scale.
References
Ajzen, I. & Fishbein, M. (1980). Understanding attitudes and predicting social behaviour. Englewood Cliffs, NJ: Prentice-Hall.
Anderson, N., Ones, D.S., Sinangil, H.K., & Visvewvaran, C. (2001). Handbook of industrial, work and organisational psychology (Vol. 2). London: Sage.
Arnold, J., Cooper, C.L. & Robertson, I.T. (1998). Work psychology: Understanding human behaviour in the workplace (3rd ed.). Essex: Pearson Education Ltd.
Brehm, S.S., Kassin, S.M. & Fein, S. (2002). Social psychology (5th ed.). New York: Houghton Mifflin Co.
Eagly, A.H. & Chaiken, S. (1993). The psychology of attitudes. New York: Harcourt Brace Yovanovich, Inc.
Eiser, J.R. (1987). The expression of attitude. New York: Springer-Verlag.
Fishbein, M. & Ajzen, I. (1975). Beliefs, attitudes, intention and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.
Locke, E.A. (1976). The nature and causes of job satisfaction. In M.D. Dunnette (Ed.), Handbook of industrial and organisational psychology. Skokie, IL: Rand McNally.
Millar, M.G. & Tessar, A. (1986a). Effects of affective and cognitive focus on the attitude-behaviour relation. Journal of Personality and Social Psychology, 51, 270-276.
Muchinsky, P.M. (1993). Psychology applied to work: An introduction to industrial and organisational psychology (4th ed.). Belmont: Wadsworth, Inc.
Schultz, D.P. (1978). Psychology and industry today (2nd ed.). New York: Macmillan Publishing Co., Inc.
