Revisiting Representativeness
Haidong Lu, Paul N Zivich, Jacqueline Rudolph, Zeyan Liew, Bhramar Mukherjee, Fan Li
Abstract
Open-access reader
Haidong Lu, Paul N Zivich, Jacqueline Rudolph, Zeyan Liew, Bhramar Mukherjee, Fan Li
Abstract
Open-access reader
Representativeness has been a topic of considerable discussion in epidemiology over the past decade, yet the concept of representativeness itself remains somewhat elusive. In this article, we revisit representativeness in the context of epidemiologic research by distinguishing between sample representativeness and estimate representativeness. Building on recent insights into selection bias, we explore the relationship between sample representativeness, estimate representativeness, and two types of selection bias. Specifically, we demonstrate through case studies how sample non-representativeness can lead to different types of selection bias, thereby resulting in estimate non-representativeness. Lastly, we describe scenarios where sample non-representativeness does not necessarily result in estimate non-representativeness.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Representativeness has been a topic of considerable discussion in epidemiology over the past decade, yet the concept of representativeness itself remains somewhat elusive. In this article, we revisit representativeness in the context of epidemiologic research by distinguishing between sample representativeness and estimate representativeness. Building on recent insights into selection bias, we explore the relationship between sample representativeness, estimate representativeness, and two types of selection bias. Specifically, we demonstrate through case studies how sample non-representativeness can lead to different types of selection bias, thereby resulting in estimate non-representativeness. Lastly, we describe scenarios where sample non-representativeness does not necessarily result in estimate non-representativeness.
Key concepts: Representativeness heuristic, Context (archaeology), Sample (material), Selection (genetic algorithm), Selection bias, Computer science, Sample size determination, Statistics