Necessary Condition Analysis: Type I Error, Power, and Over-Interpretation of Test Results. A Reply to a Comment on NCA. Commentary: Predicting the Significance of Necessity

Frontiers in Psychology

article

Jan Dul¹  Erwin van der Laan¹  Roelof Kuik¹  Maciej Karwowski² 

[1] Rotterdam School of Management, Erasmus University;Department of Historical and Educational Sciences, Institute of Psychology, University of Wrocław

关键词: Necessary Condition Analysis;    NCA;    null hypothesis testing;    alternative hypothesis;    significance;    power;    type I error;    p -value;   

DOI  :  10.3389/fpsyg.2019.01493

学科分类：社会科学、人文和艺术（综合）

来源: Frontiers

【 摘 要 】

We reply to Sorjonen and Melin (2019) article “Predicting the significance of necessity” that is acomment on a recently proposed statistical test for Necessary Condition Analysis (Dul et al., inpress). Necessary Condition Analysis (NCA) is a method that draws a ceiling line on top of thedata in an XY scatter plot (Dul, 2016). This line represents the level of X that is necessary but notsufficient for a given level of Y1. The empty space above the line is the necessity effect size. Thestatistical test for NCA is a null hypothesis test that detects the randomness of the empty space. Itis a permutation test2that produces an estimate of the p-value and “. . . is intended to answer thequestion: ‘Can the observed effect size be the result of random chance?’ by responding: ‘Yes, butwith probability smaller than p.”’ (Dul et al., in press, p. 2). Dul et al. (in press) show by simulationsand by referring to a mathematical proof that the test is valid for identifying randomness, hence forhelping researchers to avoid type I error (rejecting the null hypothesis when the null is true).

【 授权许可】

CC BY