Work–nonwork conflict and burnout: a meta-analysis

This study meta-analytically examines correlations between dimensions of work–nonwork conflict (work-to-nonwork and nonwork-to-work conflict) and burnout subscales (exhaustion, depersonalization/cynicism), with a special emphasis on the role of moderating variables. The meta-analysis is based on 220 coefficients from 91 samples with a total of 51,700 participants and employs a random-effects model. Primary studies relied on samples of working adults from different cultural backgrounds. Our results revealed that both directions of work–nonwork conflict were strongly related to emotional exhaustion as well as to cynicism (ρ between .34 and .61). The correlations were shown to be moderated differentially by gender, age, marital and parental status as well as by cultural background. Meta-analyses based on primary studies with multi-wave designs indicated that work interfering with nonwork and exhaustion have equal reciprocal effects when considering zero-order correlations. However, within meta-analytical structural equation modeling, cross-lagged relations between work-to-nonwork conflict and exhaustion across time did not improve the prediction of outcomes at Time 2 above the influence of stability coefficients.