p值操纵 [p-hacking]
定义: p值操纵是指利用某些技术手段，人为提高研究结果达到统计学显著性标准（通常为α = .05）的可能性。比如，通过多次分析后只报告p < .05的结果，选择性地删除数据直到p < .05，或者仅基于统计显著性选择用于分析的变量。
相关术语: Analytic flexibility, Fishing, Garden of forking paths, HARKing, Questionable Research Practices or Questionable Reporting Practices (QRPs), Selective reporting
参考文献:
- Hardwicke, T. E., Jameel, L., Jones, M., Walczak, E. J., & Weinberg, L. M. (2014). Only human: Scientists, systems, and suspect statistics. Opticon1826, 16, 25. https://doi.org/10.5334/OPT.CH
- Neuroskeptic. (2012). The nine circles of scientific hell. Perspectives on Psychological Science, 7(6), 643–644. https://doi.org/10.1177/1745691612459519
原稿作者: Mahmoud Elsherif
审阅者: Tamara Kalandadze, William Ngiam, Sam Parsons, Martin Vasilev
翻译者: AI-driven translation tool "TransFlow" (developed by Jinbiao Yang and COSN OpenTransfer team)
译稿审阅者: Zixi Wang, Liangjie Chen, Ruoting Liu, Shuxian Jin