Although there is much research on the relationships of corporate social responsibility and employee-related outcomes, a systematic and quantitative integration of research findings is needed to substantiate and broaden our knowledge. A meta-analysis allows the comparison of the relations of different types of CSR on several different outcomes, for example to learn what type of CSR is most important to employees. From a theoretical perspective, social identity theory is the most prominent theoretical approach in CSR research, so we aim to (...) investigate identification as a mediator of the relationship between CSR and employee-related outcomes in a meta-analytical mediation model. This meta-analysis synthesizes research findings on the relationship between employees' perception of CSR (people, planet, and profit) and employee-related outcomes (identification, engagement, organizational attractiveness, turnover (intentions), OCB, commitment, and job satisfaction), thereby distinguishing attitudes and behavior. A total of 143 studies (N= 89,396) were included in the meta-analysis which was conducted according to the methods by Schmidt and Hunter (except of the meta-analytical structural equation model). Mean effect sizes for the relationship between CSR and employee-related attitudes and behaviors were medium-sized to large. For attitudes, the relationships were stronger than for behavior. For specific types of CSR, average effect sizes were large. Identification mediated the relation between CSR and commitment, job satisfaction, and OCB, respectively. Based on our results, we give recommendations concerning the design of CSR initiatives in a way that benefits employees. (shrink)