Healthy universities: Exploring the relationship between psychosocial needs and work-related health among university employees

Abstract

The present study explores psychosocial needs among university employees and the extent to which these needs influence employee perceptions of how work positively or negatively affects their health. Structural equation modeling (SEM) analyses among Norwegian faculty members (N = 11,533) suggest that needs differ in importance to the two work-related health outcomes. Multi-group analyses suggest gender differences in the level of these needs and in their degree of relationship with positive/negative work-related health. Among women, the strongest predictors of positive and negative work-related health are work engagement and autonomy, respectively. Among men, the strongest predictors of positive and negative work-related health are meaning and social community, respectively. Although significant differences were found in the level of the psychosocial needs across different university groups (faculty, PhD students, administrative/technical staff), their predictive value for how work affects their health positively or negatively is basically equivalent across groups. Study findings raise two implications: (1) the mechanisms and characteristics of the work environment that promote versus detract from health in the university setting do not appear to be two sides of the same coin and suggest different sets of interventions for improving employee health, and (2) gender differences should be taken into account in designing interventions to improve health and well-being in universities.

Keywords:

• Gender
• healthy university
• multigroup SEM
• needs

Notes

1 Norwegian acronym for work environment and climate study.

2 To account for potential bias introduced by listwise deletion, we also estimate our main SEM using the MLMV estimator in Stata, which accounts for missingness. Our results are unchanged when using this estimator with two notable exceptions. First, trust in unit management is a statistically significant predictor of positive (β = 0.08, p < 0.05) and negative (β = −0.11, p < 0.05) work-related health, while recognition is no longer a statistically significant predictor of negative health. Full results available upon request.