Gender differences in the causal direction between workplace harassment and drinking.

Data from a longitudinal study of university employees across four waves is used to determine the extent to which workplace harassment predicts drinking or conversely the extent to which drinking predicts workplace harassment, and to address gender differences in these relationships. Mixed effects regression models are used to test the effects of 1) harassment at the previous wave on drinking at the current wave, adjusting for drinking at the previous wave, and 2) drinking at the previous wave on harassment at the current wave, adjusting for harassment at the previous wave. For males, drinking at the previous wave predicts sexual harassment at the current wave, whereas for females, sexual harassment at the previous wave predicts drinking at the current wave.