Delaney William P, Grube Joel W, Greiner Birgit, Fisher June M, Ragland David R
Prevention Research Center, Berkeley, California 94704, USA.
J Stud Alcohol. 2002 Jul;63(4):420-9. doi: 10.15288/jsa.2002.63.420.
This study tests the spillover model of the effects of work stress on after-work drinking, using the variable "length of time to unwind" as a mediator.
A total of 1,974 transit operators were contacted and 1,553 (79%) of them participated in a personal interview. Complete data on the variables in this analysis were available for 1,208 respondents (84% men). Using latent variable structural equation modeling, a model was tested that predicted that daily job problems, skipped meals and less social support from supervisor would increase alcohol consumption through the mediator, length of time to unwind and relax after work. Increased alcohol consumption was, in turn, hypothesized to increase drinking problems.
As predicted, skipped meals and daily job problems increased length of time to unwind and had an indirect positive relationship with overall drinking, even when controlling for drinking norms and demographic variables. Overall drinking was positively associated with drinking problems. Supervisor support at work, however, did not significantly influence length of time to unwind. Difficulty unwinding (longer time to unwind) did not have direct effects on drinking problems; however, indirect effects through overall drinking were observed.
These results provide preliminary support for the mediating role of length of time to unwind and relax after work in a spillover model of the stress-drinking relationship. This research introduces a new mediator and empirical links between job problems, length of time to unwind, drinking and drinking problems, which ground more substantively the domains of work stress and alcohol consumption.
本研究以“放松所需时间”这一变量作为中介变量，检验工作压力对下班后饮酒影响的溢出模型。
共联系了1974名公交司机，其中1553人（79%）参与了个人访谈。本分析中1208名受访者（84%为男性）有这些变量的完整数据。使用潜在变量结构方程模型，对一个模型进行了检验，该模型预测日常工作问题、不规律饮食以及来自上级的社会支持减少会通过中介变量——下班后放松所需时间，增加饮酒量。反过来，饮酒量增加被假定会增加饮酒问题。
正如预测的那样，即使在控制饮酒规范和人口统计学变量的情况下，不规律饮食和日常工作问题仍会增加放松所需时间，且与总体饮酒量呈间接正相关。总体饮酒量与饮酒问题呈正相关。然而，工作中上级的支持对放松所需时间没有显著影响。放松困难（放松所需时间更长）对饮酒问题没有直接影响；但通过总体饮酒量产生了间接影响。
这些结果为下班后放松所需时间在压力与饮酒关系的溢出模型中的中介作用提供了初步支持。本研究引入了一个新的中介变量，并建立了工作问题、放松所需时间、饮酒和饮酒问题之间的实证联系，为工作压力和酒精消费领域提供了更坚实的基础。
