Common Error 7: Converting a Continuous Variable into Categorical Without Appropriate Justification
Researchers sometimes dichotomize continuous moderators to facilitate analysis (Aguinis, 2017; Becker et al., 2023; Rasoolimanesh et al., 2021). Our review found 10.95% of studies used methods like median-splits or clustering (Rasoolimanesh et al., 2021; Aguinis et al., 2017). This practice risks significant information loss, reduces statistical power, and can alter the significance of the interaction term (Cohen, 1983; Irwin & McClelland, 2001). It may also create artificial subgroups without justification (Dawson, 2014). We advise against dichotomization. Instead, use SEM-based approaches to test the continuous product term (Aguinis et al., 2017). If categorization is unavoidable, compelling justification and rigorous measurement invariance tests are required (see Error 8).
常见问题七：缺乏充分理由将连续变量转换为分类变量
为了便于进行多组分析（MGA），部分研究（约10.95%）会将连续的调节变量（如“信任度”）武断地二分（如“高信任组”和“低信任组”，通常以中位数为界）。这种人为的二分法会导致信息严重丢失、统计功效降低，并可能扭曲交互项的真实显著性（Cohen, 1983）。更重要的是，这种分割点往往缺乏理论和实证依据。
问题实质：
将连续谱系（如从1到7的评分）硬生生切成两类，就像把丰富的色彩世界简化为黑白两色，丢失了大量细节与渐变信息，结论可能过于粗糙甚至误导。
解决建议
Ÿ 优先保留连续性：强烈建议使用基于结构方程模型（SEM）的方法来检验连续变量构成的乘积项（X*M），以充分利用数据的完整信息（Aguinis等, 2017）。
Ÿ 如必须分组，需严格论证：如果因特定理论或实践原因必须分类，研究者必须：（a）提供强有力的分割依据；（b）严格检验分组后的测量不变性（见问题8），确保工具在不同组间的测量意义一致。
Reference
Aguinis, H., Edwards, J. R., & Bradley, K. J. (2017). Improving our understanding of moderation and mediation in strategic management research. Organizational Research Methods, 20(4), 665-685.
Becker, J. M., Cheah, J. H., Gholamzade, R., Ringle, C. M., & Sarstedt, M. (2023). PLS-SEM's most wanted guidance. International Journal of Contemporary Hospitality Management, 35(1), 321-346.
Carte, T. A., & Russell, C. J. (2003). In pursuit of moderation: Nine common errors and their solutions. MIS Quarterly, 27(3), 479-501.
Cohen, J. (1983). The cost of dichotomization. Applied Psychological Measurement, 7(3), 249-253.
Dawson, J. F. (2014). Moderation in management research: What, why, when, and how. Journal of Business and Psychology, 29(1), 1-19.
Xu, Y., & Shiau, W. L. (2026). Moderation analysis in business and management research: Common issues, solutions, and guidelines for future research. International Journal of Information Management, 86, 102995.
转载本文请联系原作者获取授权，同时请注明本文来自萧文龙科学网博客。
链接地址：https://wap.sciencenet.cn/blog-3444471-1522888.html?mobile=1
收藏
