On the practice of dichotomization of quantitative variables.

By R. Maccallum, Shaobo Zhang, Kristopher J Preacher, Derek D. Rucker. Psychological methods, 2002

3,410 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/67646d3def337c40bbfa2476f0f1ff785ae18242

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Testing Moderator and Mediator Effects in Counseling Psychology Research. (2004, methodology): “When one or both variables are measured on a continuous scale, regression procedures that retain the continuous nature of the variables clearly are preferred over using cut points (e.g., median splits) to create artificial groups to compare correlations between groups or examine interaction effects ”
Visual word recognition of single-syllable words. (2004, background): “MacCallum, Zhang, Preacher, and Rucker (2002) have recently reported a review of the literature, along with a series of simulations, which nicely demonstrated that with a relatively small number of observations, such categorization can decrease reliability and lead to the inappropriate rejection of”
Proactive and Reactive Aggression (2016, background): “The grouping approach typically requires dichotomizing continuous variables, and this approach can attenuate associations or even create spurious relationships (Cohen 1983; MacCallum et al. 2002).”
Visual word recognition of multisyllabic words (2009, background): “There is also evidence that such categorization could yield misleading results, either by decreasing statistical power and reliability (Cohen, 1983) or by inflating the Type II error rate ( MacCallum, Zhang, Preacher, & Rucker, 2002 ).”
Predicting Interpersonal Competence and Self-Worth From Adolescent Relationships and Relationship Networks: Variable-Centered and Person-Centered Perspectives (2006, background): “Results may be biased by the arbitrary division of a continuous variable into a limited number of categories ( MacCallum, Zhang, Preacher, & Rucker, 2002 ), although a clear conceptual framework to guide hypothesis testing may ameliorate some of these concerns (Bergman, Magnusson, & El-Khouri, 2003;”
Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration (2007, neutral): “However, grouping loses information and may reduce statistical power [87] especially when dichotomization is used [82,85,88].”
Celebrity endorsement, self-brand connection and consumer-based brand equity (2015, methodology): “However, we decided against using the dichotomization approach as it purportedly leads to loss of information that leads to spurious results and is strongly discouraged (MacCallum et al., 2002).”