求教： 怎样操作化well-being以检验其维度间的相互抵消效应

我的模型里要检验三个构面（construct:HRM Wellbeing OP organizational performance，企业绩效/表现）之间的关系，其中一个构面，wellbeing(WB)有三个维度（心理感受，人际关系，生理感受），为了方便大家理解，就用维度下的一些题目来代表：满意度，工作压力，信任公平(在国内念书的时候没有看这方面的中文文献，所以不知道这些概念合适的中文翻译，中文表达里只是粗略用词代替，没有斟酌，肯请大家批评指正)。

根据理论及现有文献里的实证结果，HRM 与满意度，信任公平是应该是正相关的，与压力是负相关的。HRM 会给员工带来正面影响也有消极效应。打个比方就是，HRM对员工有副作用（压力），如果这个副作用太大，整体的WB就会降低，从而导致OP也降低（当然还有一种情形是WB降低，OP却提高，如果有对这HRM-WB-OP模型感兴趣的朋友，我们可以进一步详细讨论）。这是我的理论模型中很重要的一点。关于HRM, OP操作化的文献很多，也可比较和选择，但是我不知道怎样操作化WB，才能实现这个逻辑，怎样去说明正效应和负效应会抵消。

如果 满意度+信任公平>工作压力  WB+

如果  满意度+信任公平<工作压力  WB-

根据罗老师他们的多维度构面分类标准，这应该是一个Aggregate的构面，但是文章里对操作化的建议，如工作满意度，每个维度都是正向的，相加就好，这个就不能实现我想检验构面内维度相互抵消效应的初衷。压力再小，也是有分值的，不会减少WB的值。

如果不行，我只能把它当作一个Profile的构面，分别检验单个维度与HRM, OP构面的关系。你们怎么认为呢？期待大家的建议和帮助。非常感谢。

WB 为 job satisfaction, job related affect(anxiety–content;depression-enthusiasm), 英国WERS 的问卷应该是按此维度来设计的，比较成熟，偏向心理测量，但是不能实现我相互抵消效应的检验，难道Negative ‘ tense, worried, uneasy’是用负分数体现的？求指教~~~

Eg. : Warr’s anxiety–comfort scale (1990),which is based on answers to the question: ‘Thinking of the past few weeks, howmuch of the time has your job made you feel    ?’, for each of six emotional states, thesebeing three positive states – relaxed, calm, and contented  and three negative ones – tense, worried, anduneasy. Anxiety–comfort is measured by the mean scores on the six emotionalstates。

Grant, A. M.,Christianson, M. K., & Price, R. H. (2007). Happiness, health, orrelationships? Managerial practices and employee well-being tradeoffs. The Academy of Management Perspectives, 21(3),51-63.   

Law, K. S., Wong, C.-S.,& Mobley, W. M. (1998). Toward a taxonomy of multidimensional constructs. Academy of management review, 23(4),741-755.   

Van De Voorde, K.,Paauwe, J., & Van Veldhoven, M. (2012). Employee Well‐beingand the HRM–Organizational Performance Relationship: A Review ofQuantitative Studies. InternationalJournal of Management Reviews, 14(4), 391-407.      

Warr, P. (1990). Themeasurement of well‐being and other aspects of mental health. Journal of Occupational Psychology, 63(3),193-210.   

Employee well-being was first defined asthe overall quality of employee experience and functioning at workplace( Warr,1987).

then, specified as three dimensions byGrant et al.(2007), van de Voorde et al., (2012) in their remark and review:

‘Happiness’(job satisfaction,organizational climate)

‘Health’( stress, burnout)

‘Relationship’(perceived trust frommanagement, cooperation with colleagues)

The central idea of this approach is that HRMis positively related to happiness, relationship dimension but negatively tohealth (increases satisfaction, trust etc. but undermine health, increasingstress). On the whole, the overall construct of employee well-being might playa mediating role in the HRM-organizational performance link. By this logic,here comes the tricky problem: how to operationalize well-being to demonstratethe trade-off (counteracting) generated by HRM.

By the taxonomy criteria ofmultidimensional constructs proposed by Law et al. (1998), well-being seems to bean aggregate model. It is an algebraic function of different dimensions but summingup would not realize my intention. As aforementioned, enhanced happiness, relationshipdimensions are accompanied with increased stress. If the stress outweighs thehappiness, relationship, then well-being becomes a mediator negativelyassociated with performance. If the stress is low or absent, well-being is a mediatorpositively connect the HRM-performance linkage.