[摘要]  国际化战略是否有助于企业提高绩效？对于这个全球战略研究领域的焦点问题，不同观点的争议由来已久：两者之间为正相关、负相关、倒U形、U形、水平S形、N形关系，甚至不相关。本文采用定量文献研究的Meta回归方法，以企业绩效对国际化的回归系数为效应量，应用64篇重要文献中的176个样本观测值，从多个层面、多重角度对此争议问题进行了透彻的原因分析。通过对影响国际化-企业绩效实证研究结果的全面系统误设定偏倚考察，及14个相关假设的检验，本文从样本企业特征、变量操作特征、研究方法特征、文献发表特征四个层面，获得了一系列有价值的调节因素发现：正是这些调节因素在各研究中的纳入不同和作用不同，导致了国际化与企业绩效关系结果产生了种种差异。这一发现能够更为有效地解释国际化-企业绩效的内在关系机制，并提供Meta回归研究的参考框架。

[关键词] 国际化； 企业绩效； 调节因素； Meta回归； 误设定偏倚

A study of influencing factors of internationalization-performance relationship: Mis-specialization bias analysis based on Meta-regression technique

CHEN Limin, WANG Xiaoxia

Abstract: Whether international strategy will help firms improve their performance? It is a focus question in the global strategic management field, and there are long-standing disputes among different viewpoints: The relationship between internationalization and firm performance is positive linear, negative linear, inverted U-shaped, U-shaped, horizontal S-shaped, N-shaped, or even irrelevant. In this paper we use a quantitative literature research approach Meta-regression to address this debate. We use the regression coefficient as effect size, utilizing 64 important relevant papers, involving 176 effect size observations, and conducting a thorough study from multiple levels and multiple perspectives for this controversial issue. We investigate systematically the mis-specialization bias influencing this relationship. After testing 14 hypotheses we obtain many valuable moderating factors from four aspects: sample firms, variable features, research methods and literature characteristics. This result can better illustrate the inner internationalization - performance mechanism, and give a reference variable framework for Meta-regression studies.

Key words: internationalization; performance; moderating factors; Meta-regression technique; mis-specialization bias.

（该论文发表于武汉大学奖励期刊《中国工业经济》2014年第11期。）