Redefining Representativeness of a Sample in Causal Terms

TL;DR

This paper redefines sample representativeness in medical science using causal Bayesian networks to provide a clear and practical framework.

Contribution

A novel formal definition of sample representativeness using causal Bayesian networks is proposed.

Findings

A precise formal definition of sample representativeness is developed using causal Bayesian networks.

The definition translates into actionable methodological guidance for researchers.

Examples and a checklist are provided to illustrate the practical application of the concept.

Abstract

Despite its crucial role, sample representativeness remains a controversial topic in the methodology of medical science. There is an ongoing debate not only about how best to define and ensure the representativeness of a sample (e.g., Rudolph et al. 2023; Porta 2016), but also about whether representativeness is worth pursuing at all (e.g., Rothman et al. 2013).

Our aim is to construct a formalised, precise, and practical conceptualisation of sample representativeness.

We employ the established framework of causal Bayesian networks to develop such a conceptualisation.

We propose a precise formal definition of sample representativeness that translates into clear and actionable methodological guidance. Additionally, we provide examples and a checklist to illustrate the application of the proposed conceptualisation.

We believe that the presented definition will facilitate further discussion of the issue of representativeness and prove useful to scientists in practice.

Figures