The terms univariate, multivariate, univariable, and multivariable often appear in scientific medical publications. For example, Kanbaş et al., claim that they have evaluated factors prognostic for cervical cancer using multivariate Cox regression. However, Cox regression is a semi-parametric technique that they use to evaluate how a single response variable (survival time) is affected by one or more explanatory (potentially prognostic) variables. This is not a multivariate analysis.
Univariate and multivariate are terms that refer to the probability distribution upon which a statistical model is based. If a model only has one response variable, a univariate probability distribution works. But if the model has more than one response variable, a multivariate probability distribution is needed. In contrast, the model's univariable or multivariable nature is determined by the number of explanatory variables.
As an example, a model for evaluating the relationship between systolic blood pressure and sex is univariate and univariable. If the explanatory variable age is added to the model, the new model is still univariate but has become multivariable. If the two models are further developed for evaluating two response variables, both systolic and diastolic blood pressures, they become multivariate, the first one univariable and the second one multivariable.
References
1. Kanbaş CÖ, Keles E, Mert EÖ, Özdaş CE, Can HK. Prognostic impact of adenomyosis in cervical cancer: insights from machine learning-driven survival analysis. Rev Assoc Med Bras (1992). 2026 Jun 29;72(4):e20251656. doi: 10.1590/1806-9282.20251656. PMID: 42385036.
No comments:
Post a Comment