A signiﬁcant proportion of previous research in the occupational stress area has tended to treat the personality variable negative affectivity (NA) as a nuisance variable. This perspective has led researchers to routinely control for the effects of NA. However, P. E. Spector, D. Zapf, P. Y. Chen, and M. Frese (2000) have proposed a number of different mechanisms by which NA could have substantive effects. The current research used a longitudinal framework to test several competing mechanisms proposed by Spector et al. (speciﬁcally, the perception, hyperresponsivity, and causality mechanisms) on the relationship between work stressors and psychological well-being. Customs workers and dentists constituted the longitudinal sample (N 345). Results provided strong support for the perception mechanism, indicating that the effects of NA on psychological health were partially mediated by work stressors. The authors discuss the theoretical and practical relevance of the perception mechanism to occupational health. Keywords: negative affectivity, work stress, occupational health
Negative affectivity (NA) has been described as a personality trait reﬂecting individual differences in negative emotion and self-concept (Watson & Clark, 1984). It has been argued that NA may affect perceived levels of stress by pervasively inﬂuencing perceptions of the self or environJoseph E. Oliver, South London & Maudsley NHS Foundation Trust, London, United Kingdom; Angela Mansell and Paul E. Jose, Psychology Department, Victoria University of Wellington, Wellington, New Zealand. Appreciation is expressed to New Zealand Customs Service employees and dentists who took part in this research. We are grateful for all the advice and feedback provided by Andrew J. Hart in the preparation of this article. Correspondence concerning this article should be addressed to Joseph E. Oliver, South London & Maudsley NHS Foundation Trust, LEO Services, London, United Kingdom SW9 9HG. E-mail: firstname.lastname@example.org 56
International Journal of Stress Management 2010, Vol. 17, No. 1, 56 –77 © 2010 American Psychological Association 1072-5245/10/$12.00 DOI: 10.1037/a0017696
Negative Affect and Work Strain
ment (Chen & Spector, 1991). Accordingly, individuals found to be high in NA tend to report higher levels of dissatisfaction and perceived stress than individuals who have lower levels of NA (Eysenck, 1991). Early occupational-related research into the NA construct has focused on potential confounding effects between work-related stressors and strains. Some authors have suggested that the well-documented relationship between work-related stressors and strains is merely an artifact of a tendency of those individuals high in NA to respond to self-report methodologies with an overall negative perception, thereby artiﬁcially inﬂating correlations (Costa & McCrae, 1990; Watson, Pennebaker, & Folger, 1987). In support of this hypothesis, Brief, Burke, George, Robinson, and Webster (1988) found that controlling for the effects of NA reduced the stressor–strain relationship considerably. They found that six out of 15 correlations between stressor and strain variables were reduced to nonsigniﬁcance after the effects of NA had been removed. They concluded that NA had inﬂated the relationship between stressors and strain and suggested that future studies could control for NA to avoid this problem. Such results have led to an often routine treatment of NA as a nuisance variable that must be controlled to provide a more “accurate” estimation of the true relationship between work stressors and strains (e.g., Cavanaugh, Boswell, Roehling, & Boudreau, 2000; Wiesner, Windle, & Freeman, 2005). However, subsequent replications (e.g., Chen & Spector,...