A historical and philosophical study on null hypothesis significance testing

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K13044
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 01080:Sociology of science, history of science and technology-related
• Research Institution: Health Sciences University of Hokkaido
• Principal Investigator: MORIMOTO Ryota 北海道医療大学, リハビリテーション科学部, 准教授 (70648500)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 科学哲学 / 統計的仮説検定 / 有意性検定 / 仮説検定

Outline of Final Research Achievements

Null hypothesis significance testing (NHST) is the most widely used statistical method. Nevertheless, misuse and misinterpretation of NHST are widespread, and severe criticisms have been levelled against NHST. NHST is an incoherent hybrid of two different testing theories: Fisher's significance testing and Neyman-Pearson hypothesis testing. Fisher on one side and Neyman-Pearson on the other were engaged in a heated controversy as to philosophy of testing theory. In this study, I review the primary textbooks on NHST and conduct a historical and philosophical survey to clarify the source of misuse and misinterpretation of NHST.

Free Research Field

科学哲学

Academic Significance and Societal Importance of the Research Achievements

生物学や心理学などの統計解析を使用する分野では、科学哲学的な問題に踏込むよりも実用的な要請への対応が優先される。本研究は、統計的仮説検定の誤解と誤用に正面から向き合い、史的考察と哲学的分析を通じて、統計的仮説検定の適切な使用法を示す点が創造的であり、科学史・科学哲学のみならず統計教育の領域にも貢献できることが期待できる。