What Are “Good” Values of q2? Guidance Based on Experimental Accounting Researchers’ Assessments of Fit
研究了实验会计研究者如何评估q2值（衡量均值模式与自定义对比的拟合度），发现图形展示和个人特征影响评估，并基于模糊集理论提出一种事前评估方法，帮助研究者判断q2值的好坏。
ABSTRACT Although q2 measures how well a pattern of means fits a custom contrast, there is no guidance for what values are “good.” We survey experimental accounting researchers who assess the fit between plots of means and contrast weights as poor, acceptable, good, or excellent. We find that graphical presentation effects and researchers’ individual attributes influence their assessments. This suggests that research needs an ex ante method for evaluating q2, grounded empirically in the wisdom of the crowd across many different presentations, rather than relying solely on the idiosyncratic assessments of individual researchers. Using fuzzy set theory, we develop such a method that researchers can use to characterize q2 = 0.100, for example, as mostly good fit, leaning toward acceptable. Our approach has significant advantages over bright-line cutoffs commonly used for other statistical indices. Overall, our study can improve our discipline’s assessments of fit between experimental results and custom contrast weights. Data Availability: Data are available from the authors upon request. JEL Classifications: M40; M49.