跨层次实证研究中的理论基础--以三份国际期刊的最新论文为例

Theoretical Foundations of Cross-level Empirical Studies——Evidence from Three Decent International Journals

在管理学研究领域,跨层次的实证研究成果越来越多。但不少学者在提出变量之间的跨层次关系假设时,往往遇到理论支撑不足的问题,从而影响了研究结果的可信性和有效性。为了帮助学者进一步认识和应用跨层次理论,本文以组织行为学领域的《应用心理学期刊》(Journal of Applied Psychology)、《组织行为学杂志》(Journal of Organizational Behavior)及《人事心理学》(Personnel Psychology)最近五年(2010—2014)的最新实证研究成果为分析对象,搜寻了应用广泛的跨层次理论,识别了情感事件理论、社会学习理论、社会认知理论以及团队和个人动机的整合模型等为跨层次理论模型的典型代表,并展现了已有实证研究具体运用这些理论模型的应用实例。本文希望可以促进中国学者对于跨层次理论的学习和应用,并推动新的跨层次理论的开发。

To have a comprehensive and rich understanding of the nested organizational phenomena, scholars increasingly adopt the cross-level paradigm in their organizational research. Many statistical techniques and software for multi-level analysis have been developed and advanced to facilitate the empirical testing of multi-level relation- ships. Equally important, appropriate multi-level theories are needed to clearly define the construct, variable and their levels, and explain the rationales behind the relationships between the variables from different levels. Yet, it is not uncommon for scholars to encounter the difficulty of finding sufficient theoretical supports when proposing cross- level hypotheses, which weakens the credibility and validity of the findings. Therefore, seeking the right multi-level theory as the basis of multilevel empirical research is particularly critical. To apply the multi-level theory and use it appropriately, it is primarily crucial for us to know existing multi-lev- el theories and have an in-depth understanding of them. By taking stock of existing knowledge of multi-level theo- ries,we can understand each theory＇ s key constructs and logics behind,which can inspire the identification of new research question and guide the selection of the major variables and the explanation of these variables for new em- pirical multi-level research. Furthermore, reviewing current multi-level theories can be helpful to avoid the wasteful repetition when developing new multi-level theories, and ensure the substantial novelty of the new efforts in muhi- level theory developments. In this paper, we review the multilevel theoretical bases of the empirical studies published in three international decent journals, i. e. ,Journal of Applied Psychology, Journal of Organizational Behavior, and Personnel Psychology, in a period of recent five years （2010--2014）. We identify three major,prevalent cross-level theories： affective e- vents theory, social learning theory, and the integrated model of team and individual motivation. We offer a brief in- troduction of each of these theories and then present the application of each theory in the empirical studies by analy- zing 30 exemplary articles ,for the purpose of showing how the general arguments of a muhi-level theory can be ap- propriately applied to the specific context of a cross-level organizational phenomena. In general, although they are not exhaustive in listing the multi-level theories, the three theories reviewed in this paper focus on different levels, have different features and thus can complement each other. Specifically, affec- rive events theory is often used to explain the effect of individual-level variables on within-individual variables, with some occasions being applied explain the effect of the working environment or other higher level factors on the indi- vidual-level attitudes and behaviors. Promoting the level of independent variables, social learning theory （and its sister theory social cognition theory） focuses on the effect of the organization-or team-level factors on the individual level. Rather than accounting for the one-way effect as affective events theory and social learning theory usually do, the integrated model of team and individual motivation highlights the interaction between individual factors and team factors, by integrating the top-down effect of team factors on individual elements and the bottom-up effect of individ- ual factors on team factors. In sum, these three theories each represent different types of multi-level relationships and as a whole can be applicable for a broad range of empirical research settings. By offering an overview of some major multi-level theories and their application in the recent empirical studies published in three international decent journals, we hope to stimulate the scholars to pay special attention to the the- oretical foundation of cross-level empirical studies and to understand its significance of multi-level theories for such empirical research. Hopefully, the multi-level theories reviewed by this paper can help some scholars better develop the empirical research design. Surely, the further test and development of these multi-level theories are also of neces- sity. Of equal importance is to build new multi-level theories, probably by relying on methods such as case study and/or grounded theory.