Examining Linking Language and Causal Implications in Observational Psychological Capital Research

The present study examines whether the issues with implied causal inference identified by Haber et al. (2022) in observational health research extend to the field of industrial-organisational (IO) psychology. Haber et al. found that causal relationships were implied through the language used to link exposures and outcomes and through recommendations for action. This small-scale conceptual replication applies Haber et al.’s approach to psychological capital (PsyCap), an increasingly popular construct in the domain of IO psychology. Specifically, we evaluate the causal strength of the language used to describe the nature of the connection between given variables when reporting findings and when making recommendations for action in 50 PsyCap articles. The majority of action recommendations were found to have strong causal implications (85% in abstracts, 84% in discussion sections), and most linking sentences implied some level of causality (64% in abstracts, 58% in discussion sections). This was despite very few articles explicitly stating an intention to estimate causal effects (10%) and many explicitly warning against drawing causal inferences (42%). No significant relationship was found between the causality implied in the linking sentences and the strength of causal implication of the action recommendations. These findings have important implications for how research is implemented, and very real consequences for those who are the focus of such implementations. Increased transparency about causal interests and reasoning is needed to uphold the credibility of findings and the integrity of applications of observational PsyCap research.