Title

CONCURRENT AND LAGGED EFFECTS OF LEADERSHIP BEHAVIOR ON SUBORDINATE STRESS AND HEALTH

Date of Award

2005

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Ph.D.)

Department

Psychology/Industrial-Organizational

First Advisor

Steve Jex

Abstract

Employee stress and health bear significant costs to both individual employees and to organizations. This study investigates the role of leadership behaviors and two job resource variables (i.e., task control and rest breaks) in the prediction of subordinates’ acute strains (i.e., perceived stress and tension-anxiety) and of self-reported and medically verified reports of musculoskeletal symptoms. Two theoretical models were presented and tested: an indirect effects framework, and buffering effects framework. Specific leadership behaviors were expected to buffer or directly impact theoretically matched stressors. Using a 3-wave longitudinal design, the initial sample included 416 office workers (e.g., claims processors, claims representatives, word processors) from a U.S. insurance organization. Structural equation modeling was used to test the proposed indirect effects model, and moderated hierarchical multiple regression was used to test the buffering effects model. Each hypothesis was tested for cross-sectional, concurrent, and lagged effects. Partial support was found for both the indirect and buffering effects frameworks. Leader initiating structure indirectly reduced levels of acute strains through its effects on role stressors. High frequency of negative criticism from the supervisor was associated with greater levels of acute strains, it strengthened the positive relationships between intra-group conflict and acute strains, and it had an indirect effect on increasing musculoskeletal symptoms through its effect on tension-anxiety. In addition, leader consideration buffered the effects of intra-group conflict on tension-anxiety and of tension-anxiety on musculoskeletal symptoms. Feedback/coaching from the supervisor received limited support as a moderator variable. Task control was found to buffer the effects of workload on stress, and of acute strains on musculoskeletal symptoms, while rest breaks received less consistent support as a moderator variable. Implications for research and theory in the areas of industrial-organizational psychology and public health are offered. In addition, practical implications are discussed, with recommendations for interventions such as leadership training, work redesign, and stress management to minimize or prevent acute strains and to reduce musculoskeletal symptoms.