ISSN 0439-755X
CN 11-1911/B

Acta Psychologica Sinica ›› 2016, Vol. 48 ›› Issue (12): 1561-1573.doi: 10.3724/SP.J.1041.2016.01561

Previous Articles     Next Articles

A meta-analysis of the relationship between organizational identification and turnover intention

ZHANG Shuhua1; LIU Zhaoyan2   

  1. (1 Institute of Human Resource Development and Management, Shenyang Normal University, Shenyang 110034, China) (2 School of Education Science, Shenyang Normal University, Shenyang 110034, China) (3 Basic Teaching Department, Shandong Maritime Vocation College, Weifang 261108, China)
  • Received:2016-01-14 Published:2016-12-24 Online:2016-12-24
  • Contact: ZHANG Shuhua, E-mail: zhangshuhua2000@126.com

Abstract:

Employee’ turnover behavior has been an important issue in management circles because it will inevitably bring certain loss for the enterprise. Turnover intention, as the precursor of employee’ turnover behavior, can effectively predict individual behavior of changing job. Importantly, organizational identification, as a core organization construct, can has a significant impact on the turnover intention. Since Riketta’s (2005) meta-analysis on the relationship between organizational identification and antecedents/outcomes of organizational identification, increasing numbers of empirical studies have been published in the field of organizational behavior. The present research to examine in-depth research on the relationship between organizational identification and turnover intention seems timely. Meta-analysis was used in this study to aggregate results from studies examining the relationship between organizational identification and turnover intention. We included studies in both English and Chinese which were conducted between 1985 to 2015. Fifty-nine studies, which included a total of 71 independent samples and 23180 participants, met the criteria for inclusion in the meta-analysis. This overall sample was also divided into subgroups for moderator analysis. Data was analyzed utilizing the Comprehensive Meta-Analysis (CMA) Version 3 program. There are four major analyses in this research, including heterogeneity test, publication bias test, main effect analysis and moderation effect analysis. The test for heterogeneity showed that there was significant heterogeneity in the 71 independent effect sizes, so a random effects model was used as the meta-analysis model. The publication bias test indicated that the impact of publication bias was not statistically significant. Results showed a strong positive correlation (r = −0.453) between organizational identification and turnover intention. Additionally, results indicated that different dimensions in the organizational identification scales and the participants’ industry would moderate the relationship between organizational identification and turnover intention, but the moderation effects through turnover intention scale of different cognitive structure and culture background were not statistically significant. Specifically, in the moderator of organizational identification scales, there was a strong link between organizational identification and turnover intention with the using of organizational identification scales that is composed of the cognitive, affective and evaluative dimensions. In contrast, the link through organizational identification scales with the cognitive dimension was weak. Secondly, for participants in the leasing and business services industry, the correlation between organizational identification and turnover intention was stronger than those in other industries. Results suggest that organizational identification can effectively predict turnover intention. Researchers are expected to emphasize the importance of measuring tools and the particularity of some industries in their further research of organizational identification. Overall, the findings provided empirical evidences on the importance of organizational identification in determining turnover intention, which helps to guide future research.

Key words: organizational identification, turnover intention, meta-analysis, moderate effect