Against quantiles: categorization of continuous variables in epidemiologic research, and its discontents.
Bennette C, Vickers A. · 2012
Read Against quantiles: categorization of continuous variables in epidemiologic research, and its discontents. by Bennette C, Vickers A. (2012) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc3353173.txt.
Description
Bennette and Vickers argued against a habit deeply embedded in epidemiology: chopping continuous measurements into groups such as thirds, quarters, or fifths before analysis. The practice feels intuitive, letting researchers contrast low against high, but the authors contend it distorts the evidence it is meant to clarify.
Categorising throws away information, hides the true shape of dose–response relationships, and can manufacture apparent thresholds that vanish under better analysis. The authors describe alternatives, including keeping variables continuous and modelling non-linear relationships directly, and they examine why the field has been slow to adopt them. Their summary verdict is that quantile groups are usually inadequate for the job.
Catalog record
| CREATOR | Bennette C, Vickers A. |
|---|---|
| YEAR | 2012 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 24 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |