关键词: 工作自主性, 工作要求-资源模型, 维生素模型, “过犹不及”效应

Abstract: Nowadays, more and more employees are granted job autonomy. According to the job characteristics model and the job demands-resources model, job autonomy is defined as the degree of freedom given to employees to decide on their own work schedules and procedures (Hackman & Oldham, 1976). The existing research showed mixed findings on the relation between job autonomy and employees' attitudes and behaviors. Some research has found that job autonomy fosters employees’ work engagement, work performance, organizational citizenship behavior, voice behavior, creativity, work-family balance, job satisfaction, mental health, and well-being, while others found that job autonomy decreases the turnover rate. However, a small body of empirical work indicates that too much job autonomy may be detrimental to employees and work-related outcomes, such as reducing employee well-being, job satisfaction, work performance, and even inducing their unethical behaviors.
The effects of job autonomy on employees’ attitudes and behaviors might be more complex than linear positive or linear negative effects. There may be curve relations between job autonomy and work-related results. However, it is difficult to explain the negative impact of job autonomy based on job demands-resources model. Then Warr (1990, 1994) proposed the Vitamin model, which challenged the assumption that job autonomy is generally beneficial. The model proposed that the effects of job autonomy on mental health are analogous to the effects of vitamins on physical health. Whereas some vitamins are generally positive regardless of the extent to which they are consumed (i.e., vitamins C and E), others become detrimental to people’s health with excessive consumption (i.e., vitamins A and D). The Vitamin model provides a new perspective to understand the positive and negative effects of job autonomy. However, the Vitamin model is phenomenon-oriented rather than theory-driven. It is difficult for researchers to clarify how the impact of job autonomy on work-related results changes from positive to negative.
Therefore, we propose a too-much-of-a-good-thing effect (TMGT effect) of job autonomy on work-related results, which means that the role of "positive" job autonomy is not always positive (i.e., the inverted U-shape). We further identify two types of underlying mechanisms of job autonomy’s TMGT effect. One is “additive benefit and cost” mechanism, the other one is “interactive motivation and ability/opportunity”. Specifically, TMGT effect of job autonomy on work performance and work well-being can be explained by “additive benefit and cost” mechanism; and TMGT effect of job autonomy on workplace deviance behavior can be explained by “interactive motivation and ability/opportunity” mechanism.
In addition, we discuss several suggestions for future research: (1) Clarify the scope of TMGT effect of job autonomy; (2) Improve the research design of TMGT effect of job autonomy; (3) Clarify the internal mechanism behind TMGT effect of job autonomy; (4) Identify the threshold points of TMGT effect of job autonomy based on the contextual factors; (5) Identify the moderators and boundary conditions of TMGT effect of job autonomy.

Key words: job autonomy, job demands-resources model, vitamin model, too-much-of-a-good-thing effect