Evaluating the workplace fatalities impact of “right-to-work” laws: What matters is effect size, not statistical significance

A recent study on “right to work” laws (Zang et al. 2024) concluded that these laws have a “null” impact on occupational fatalities in the USA – a result that contrasts with previous research suggesting that the laws exacerbated the fatality rate. Their interpretation of results involves a conventional application of statistical significance and hypothesis testing. But a key assumption for hypothesis testing via significance is not met: the data are not taken from a random sample. When we consider the size of the effects revealed by their analysis (which consists of the latest techniques for analyzing panel data), a different and arguably more compelling interpretation is that adoption of right-to-work laws did lead to an increase in the rate of occupational fatalities, an increase even greater than was suggested in earlier research.