| Summary: | Thesis (M.Sc.)--Memorial University of Newfoundland, 1996. Psychology Bibliography: leaves 58-68. Eustress was conceptualized as the positive affect resulting from positive events. A moderator model was proposed which postulated that neuroticism moderates the relationship between eustress and health. Due to the possibility of specification error (i.e., data may reflect linear or mediating properties as opposed to interactive or moderating properties), an exploratory mediation model was developed to test for any mediating effects eustress may have in the neuroticism-health relationship. Specifically, the direct effects hypotheses for this mediation model proposed that uplifts lead to eustress, which in turn reduces reported symptoms of poor health. Neuroticism reduces eustress, which in turn leads to symptoms (i.e., eustress mediates the relationship between neuroticism and health). Neuroticism also leads to symptoms via other mechanisms not involving eustress (e.g., cognitive interpretation (Harkins, Price & Braith, 1989), or physiological processes (Friedman & Booth-Kewley, 1987)). Three hundred and twenty-two participants completed measures of positive affect, uplifts, neuroticism, and somatic complaints in the first phase of a two phase prospective study. One hundred and ninety-six participants from the original subject pool completed the positive affect, uplifts and somatic complaints measures two weeks later. The results provided support for conceptualizing eustress as the positive affect arising from positive events. With respect to the moderator model, the results failed to support the hypothesis that neuroticism moderates the relationship between eustress and health. The mediator model was not found to have a good fit to the data. The hypothesis that uplifts lead to eustress which in turn reduces symptoms was supported. However, the hypothesis that neuroticism leads to somatic complaints through reducing eustress (i.e., that eustress mediates the relationship between neuroticism and health) as well as through other mechanisms, was not supported. Analysis on transformed data showed some support for the hypothesized model. However, since this result was based on transformed data, it should be interpreted with caution. Alternative mediator models fit the data and supported the finding that uplifts lead to eustress resulting in low symptoms. Since mediator models fit the data, specification error could have resulted from only testing a moderator model. In other words, the data may have largely reflected linear (mediator) relationships as opposed to interactive (moderator) relationships. Possible explanations for the present findings and suggestions for future research are discussed.
|
|---|