Donald Truxillo

Date of Award


Document Type


Degree Name

Master of Science (M.S.) in Psychology



Physical Description

1 online resource (v, 125 pages)




As the workforces of industrialized countries around the world continue to age, research is needed to better understand how policies regarding retirement, and exit from the workforce, impact older workers. In particular, it is important to identify mechanisms that can be used to understand and promote the retention of older workers. Work ability (WA), a construct that has been predominately studied in Scandinavian and European countries, has been studied in this context, and identified as a predictor of exit from the workforce. Using the Job Demands-Resources model (JDR; Demerouti, Bakker, Nachreiner, & Schaufeli, 2001) as the theoretical basis, the goal of the present study was to compile and synthesize the existing literature on WA, quantitatively analyze its antecedents and outcomes (via meta-analysis), and assess potential moderators to these relationships. To my knowledge, this is the first quantitative synthesis of the WA literature. In total k = 158 studies including n =149,758 workers were included in this this meta-analysis. Results showed that WA was related to a number of antecedents including job demands (r ) Ì… = [-.15, -.30], job resources (r ) Ì… = [.19, .25], and personal resources (r ) Ì… = [.14, .45]; as well as to several outcomes including job satisfaction (r ) Ì… = .23, job performance (self-rated) (r ) Ì… = .23, and exit behaviors (r ) Ì… = [-.19, -.36]. Moderator analyses showed that when assessing the relationship between WA and some correlates (e.g., disability, retirement), studies that used the Work Ability Index (WAI) found stronger relationships with WA than studies that used measures of perceived WA. Additionally, studies that included workers from certain occupations (e.g., blue collar jobs) found weaker relationships between WA and some of its correlates (e.g., physical job demands, job control) compared to studies of workers in other occupation groups (e.g., white collar jobs). The mean age of the sample also moderated the relationships between WA and some of its correlates. Specifically, studies that consisted of older workers found stronger relationships between WA and certain correlates (e.g., job control, physical job demands), whereas for other correlates (e.g., musculoskeletal disease, disability) studies consisting of younger workers found stronger relationships with WA. This study contributes to the existing literature by synthesizing findings from existing work, identifying gaps in the existing literature, and determining how various measures of WA impact the relationships between WA and its correlates. In addition, this study helps to identify factors that can be considered in interventions aimed at retaining older workers and extending working careers.


A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Psychology

Persistent Identifier

Available for download on Wednesday, December 06, 2017

Included in

Psychology Commons