The epistemic and pragmatic function of dichotomous claims based on statistical hypothesis tests

Researchers commonly make dichotomous claims based on continuous test statistics. Many have branded the practice as a misuse of statistics and criticize scientists for widespread application hypothesis tests to tentatively reject (or not) depending whether p-value is below or above an alpha level. Although are rarely explicitly defended, we argue they play important epistemological pragmatic role in science. The function consists transforming data into quasibasic statements, which accepted singular facts that can corroborate falsify theoretical claims. This transformation requires prespecified methodological decision procedure such Neyman-Pearson tests. From perspective falsificationism these procedures necessary, probabilistic statements (e.g., statistics) cannot falsifiers substantive hypotheses. facilitate scrutiny criticism among peers by generating contestable claims, process referred Popper “conjectures refutations.” We speculate about how surprisingly use 5% level might facilitated this function. Abandoning example because researchers p-values, would sacrifice their crucial epistemic functions.