Research on teaching methods for popularizing Bayesian statistics in psychological research

2023 Fiscal Year Final Research Report

• Project/Area Number: 19K03269
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 10020:Educational psychology-related
• Research Institution: Waseda University
• Principal Investigator: Toyoda Hideki 早稲田大学, 文学学術院, 教授 (60217578)
• Project Period (FY): 2019-04-01 – 2024-03-31
• Keywords: ベイズモデリング / 帰無仮説 / 統計的仮説検定 / 特異項目機能 / 項目反応理論 / PHC

Outline of Final Research Achievements

Currently, psychological research is shaken by the problem of reproducibility. Even though these papers have passed rigorous peer review, they have not been reproduced even after repeated attempts, and there are many papers whose knowledge in the relevant research field is dubious. The biggest reason for this at present is that p-values in significance tests are automatically used as the main criterion for acceptance or rejection of papers. In this study, in order to overcome these disadvantages, we will create a teaching and learning series of psychological statistical methods using a Bayesian approach. The area covers the intermediate level of statistical methods that contribute to psychological research, and it is a teaching-learning series that is equivalent to two credits immediately after completing the first two credits. The research results were published as a book.

Free Research Field

心理統計学

Academic Significance and Societal Importance of the Research Achievements

尤度を使って現象を考えるという心理学研究のパラダイムを構築した。これは有意性検定の手続きを暗記し、当てはめ、有意水準を超えたか否かを判定するという、現在主流の
心理学研究のパラダイムを180度転換する思考法である。ベイズか頻度論かという選択とも違う。この興味深い現象は、どのように生成され、データとして自分の眼前に現れたのだろう」という疑問に尤度を使って答えることである。確率分布によってデータの生成過程をモデル化することを可能にした。