UBC Theses and Dissertations
Mood, motivation, and task me Zerbe, Wilfred Joachim
Theorists in organizational behavior have generally ignored emotional determinants of behavior. A task of this dissertation was to extend the use of emotions for understanding organizational behavior in general and work motivation in particular. Two theories, expectancy theory and network theory, are used to make predictions about the relationship between mood and perceptions of the relationship between effort and performance. According to expectancy theory, the effort that people choose to expend at tasks is a function of their belief about the degree to which effort and performance covary. Network theory predicts that memories are connected by a network of associations. The accessibility for recall of a memory is a function of the activation of these associations. In this way positive events are more accessible for recall when individuals are in a positive mood state because of associations based on the affective valence of memories. Such accessibility of events for recall has been shown to be a determinant of probability judgements. On this basis it was predicted that mood would bias individuals' judgements of the probability that specific levels of effort lead to specific levels of performance. In other words, that mood affects expectancy. Specifically, it was predicted that individuals in an elated mood would report higher expectancy than individuals in a depressed mood. Mood was defined as a self-evaluative feeling state. Two other hypotheses were formed: that mood would influence how cause for behavior is attributed, and that individual differences in self esteem would moderate the relationship between mood and expectancy. Three studies were performed to provide a foundation for the testing of these hypotheses. In a fourth study they were tested. Study One assessed the psychometric properties of measures of mood states, individual differences, and task perceptions. Study Two concerned the experimental induction of mood. Mood manipulations used in the experimental literature were reviewed and one, a musical procedure, was chosen. The validity of this manipulation was then tested by having participants listen to the music of an elated, neutral, or depressed mood induction procedure. The results of Study Two provided strong evidence for the validity of the manipulation. Both self-report measures of mood and an unobtrusive behavioral measure were significantly affected. The results of Study Two also showed the utility of a conceptualization of mood as comprising two components: arousal and pleasure. It was shown that depression is characterized by low arousal and displeasure, and elation by high arousal and pleasure. Study Three reviewed the conceptualization and measurement of expectancy. It was argued that expectancy is properly conceptualized as the perceived covariation between effort and performance. This requires measurement of the relationship between multiple levels of effort and multiple levels of performance and calculation from these measures of an index of perceived effort -- performance covariation. Most prior measurement has only considered the relationship between high effort and high performance. Further, it was argued that such appropriate measurement allows predictions to be made about expectancy across individuals, in contrast to the argument that expectancy theory is a within-subjects theory. Previous authors have used such an approach to measure expectancy but have not demonstrated its validity. Study Three undertook such validation. Participants completed one of two experimental tasks: one with high objective expectancy, the other with low objective expectancy. As predicted, scores on the perceived covariation measure of expectancy were significantly higher in the high objective expectancy task. Measures of related constructs were influenced in a manner consistent with this finding. It was concluded that strong support for the expectancy measure existed. On the foundation of Studies One, Two, and Three, Study Four undertook to test the formal hypotheses of the dissertation. In each of three experimental sessions, participants completed a business decision-making task, underwent either an elation, neutral mood, or depression induction procedure, and then completed measures of their mood state, expectancy, and other task perceptions. The results of Study Four indicated that significant differences in mood resulted from the manipulation. However, none of the experimental hypotheses were supported. Mood did not influence expectancy or task attributions. A number of alternate explanations for this finding were considered, including failure of the mood manipulation, measurement error, and lack of statistical power. Of these, it was concluded that while Study Four lacked power to detect a large effect, this did not fully explain the failure to support the experimental hypotheses. Also compelling was the argument that the mood manipulation was not sufficiently powerful.
Item Citations and Data