Breaking up is Hard to Do: The Heartbreak of Dichotomizing Continuous Data

By D. Streiner. Canadian journal of psychiatry. Revue canadienne de psychiatrie, 2002

412 citations recorded by the source corpus.

Subfields: Experimental and Cognitive Psychology, Food Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/d818eb1890d2301629a495843b9fd0515eb0dc92

How researchers see this paper

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

In Experimental and Cognitive Psychology

In Food Science

In Statistics and Probability

Citations in the corpus

Factors Influencing Domestic and Marital Violence against Women in Ghana (2013, methodology): “Under these circumstances, Streiner (2002) advised categorizing or dichotomizing continuous variables and applying non-linear techniques where model assumptions are relaxed.”
Assessing the reporting of categorised quantitative variables in observational epidemiological studies (2017, background): “efficiency [9], reliability [7] and higher type I [10] and type II [11] errors, leading to potential misleading estimates and clinical interpretations [8, 12–20].”
Person-environment congruence in relation to career change and career persistence (2006, background): “These assumptions were met and indeed Tabachnick and Fidell (2001) have stated that ANCOVA is quite robust to any violations.”
A researcher's guide to regression, discretization, and median splits of continuous variables (2015, background): “…we note they caution dichotomization is “a practice to avoid” (Dawson&Weiss, 2012), “a bad idea” (Royston, Altman, & Sauerbrei, 2006), and a “heartbreak” (Streiner, 2002); call for its “death” (Fitzsimons, 2008); point out its “cost[s]” (Cohen, 1983), “perils” (Lemon, 2009), and “negative…”
Loss of Power in Logistic, Ordinal Logistic, and Probit Regression When an Outcome Variable Is Coarsely Categorized (2006, background): “This kind of coarse categorization frequently occurs in psychiatric settings, where an effectively continuous scale is used to measure a clinical variable, and it is then dichotomized into “case” or “noncase” (Streiner, 2002).”
What's under the ROC? An Introduction to Receiver Operating Characteristics Curves (2007, background): “T hose of you who have read this series of articles religiously know that, because of the tremendous loss of information incurred, you should never dichotomize continuous variables.(1) Never! Nohow! Ever! Under no circumstances! Except, of course, when it makes sense to do so.”
Emotion-regulation strategies across psychopathology: A meta-analytic review. (2010, background): “correlational) over the dichotomized data, because splitting a variable into categories results in loss of information and might increase the probability of type II errors (Altman & Royston, 2006; Rosenthal & DiMatteo, 2001; Streiner, 2002).”