同源主观数据是否膨胀了变量间的相关性——以战略人力资源管理研究为例

Does Single Source Subjective Data Influence Correlations between Variables?——Evidence from SHRM Studies

研究变量之间的相关性是现代管理学实证研究的基本范式。然而,变量之间统计上的相关性极易受到变量的测量方法和数据类型的影响。特别是,基于同源数据的实证研究会夸大变量间的相关程度甚至带来虚假相关,这已经给管理学实证研究带来了严峻挑战。不幸的是,目前仍有大量的实证研究都在使用同源主观数据。本文以人力资源管理和企业绩效的关系为例,检验了在同一样本中,对变量的不同测量和数据类型如何导致变量间相关性结论的差异。结果显示,当自变量和因变量是"同源主观数据"时,自变量和因变量之间的相关程度最高;当自变量和因变量是"非同源主观数据"时,相关程度有所下降;当自变量采用主观数据,而因变量采用客观数据时,这种相关没有达到显著性。本文的研究例子表明,在管理学实证研究中一定要谨慎使用同源的主观测量数据。

Examining the relationships between variables is the basic paradigm of management empirical studies. However,correlations between variables can be influenced by measuring methods and data sources in empirical studies,which even results in exaggerated or false correlations.In this paper,we take the HRM-performance relationships as examples,examining how different measuring methods and data sources lead to different correlation outcomes using the same sample. The finding shows that when the IV（ HRM） and DV（ firm performance） are ＂single source subjective data＂,the correlation is the highest; when the variables using ＂different source subjective data ＂,the correlation decreased; when HRM is measured subjectively while firm performance data is objective,the correlation is not significant.The results indicate that,more attention should be paid to data collecting source in management empirical research.