More insightful scientific estimates about population members’ gains

Because they sometimes conflate basic and applied research goals, social scientists rely on problematic null hypothesis statistical testing (NHST) to decide their empirical studies' theoretical and practical implications. For basic social science research, population predictions based on single studies provide insufficient evidence for deciding a tested theory's truthfulness and usefulness. Whether or not such studies can contribute to practical decision-making, their predictions must meaningfully augment that theory's support. In contrast, applied social science research for public policymakers and regulators must present predictions in a way that improves practical decision-making. As a result, gain-probability (G-P) analyses rather than NHST and effect size calculations can better inform theoretical and practical decisions regardless of research type. Furthermore, the theoretical and practical issues addressed by adopting G-P analysis can be considered within a larger philosophical context. • Basic social science research based on single studies provides insufficient evidence for deciding a theory's truthfulness and usefulness. • Applied social science research for public policymakers and regulators must present predictions that improves practical decision-making. • Gain-probability analyses beat null hypothesis statistical testing and effect size calculations for theoretical and practical decisions. • Instead of dichotomous findings about statistical significance, gain-probability analyses consider probabilities of varying gains or losses.