Book Review: Corrupt Research: The Case for Reconceptualizing Empirical Management and Social Science

Raymond Hubbard documents the statistical, social, and philosophical causes of the " reproducibility crisis " in Corrupt Research: The Case for Reconceptualizing Empirical Management and Social Science. Hubbard's diagnosis of the reproducibility crisis includes a detailed critique of Null Hypothesis Significance Testing (NHST), in particular testing nil null hypotheses (hypotheses of " no difference "). His critique is supported by an almost exhaustive reference list, making Corrupt Research a useful text even if you are very familiar with the issue. Less familiar to readers may be his discussion of how the widespread tacit acceptance of a philosophically naive form of Hypothetico-Deductivism (HD) as the Scientific Method permits and exacerbates the other complementary causes. According to Hubbard, HD is erroneously thought to legitimize the inappropriate use of NHST and other methodological flaws. The HD model in question prescribes that scientists first form a hypothesis, in some unspecific way, then deduce predictions of the hypothesis, and finally test those predictions. In these vague terms, the model does describe one important type of scientific reasoning, so if you ignore questions about the applicability of the HD method to many areas of psychology and how little the model tells you about the confirmatory value of experimental results, not to mention the scientists such as Darwin, Skinner, and many others who used very different inferences, HD does possess a prima facie plausibility. That along with its vague formulation, permits and seems to justify the use of NHST to apply HD by providing a method for calculating whether a hypothesis has been disconfirmed. In turn, NHST's dominance has given the illusion that all scientific practice fits the HD model. In Hubbard's account, NHST and HD together cause a range of problems, such as the widespread practice of pretending that post-hoc hypothesizing happens a priori. Hubbard tells a compelling story regarding the proposed connection between HD and NHST, and he cites many corroborating critics, but his argument consists of circumstantial evidence, as he doesn't cite any authorities using HD to explicitly endorse NHST, or vice versa, or provide a survey of literature or textbooks showing that it is standardly presented as doing so. After his discussion of the beliefs and behaviors associated with HD his strongest conclusion is, " Given the above beliefs, it is understandable why