Reporting Biases in Empirical Management Research: The Example of Win-Win Corporate Social Responsibility

Abstract

Reporting biases refer to a truncated pool of published studies with the resulting suppression or omission of some empirical findings. Such biases can occur in positive research paradigms that try to uncover correlations and causal relationships in the social world by using the empirical methods of (natural) science. Furthermore, reporting biases can come about because of authors who do not write papers that report unfavorable results despite strong efforts made to find previously accepted evidence and because of a higher rejection rate of studies documenting contradictory evidence. Reporting biases are a serious concern because the conclusions of systematic reviews and meta-analyses can be misleading. The authors show that published evidence in win-win corporate social responsibility (CSR) research tends to overestimate efficiency. The research field expects to find a positive association between corporate social performance (CSP) and corporate financial performance (CFP), and findings meet that expectation. The authors explain how this pattern may reflect reporting bias. The empirical results show strong tentative evidence for a positive reporting bias in the CSP–CFP literature but only weak tentative evidence for CSP efficiency. The study also examines which factors, such as time trends, publication outlet, and study characteristics, are associated with higher reporting biases within this literature.