Abstract
Although the main focus of this conference is on validity and not precision, data on precision underscore the challenges of detecting small effects. For example, to be able to demonstrate a relative risk of 5 in the context of a case-control study in which the prevalence of exposure is about five per cent, only seventy cases and seventy controls are needed. On the other hand, if the relative risk one is interested in is 1.1, about 35,000 cases and 35,000 controls are needed for the same exposure prevalence of five per cent. Sample size requirements can, therefore, pose a major initiation to the epidemiologic assessment of a small increase in the relative risk.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Szklo, M. (1998). Obstacles in evaluating small effects. In: Hoffmeister, H., Szklo, M., Thamm, M. (eds) Epidemiological Practices in Research on Small Effects. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-80463-2_2
Download citation
DOI: https://doi.org/10.1007/978-3-642-80463-2_2
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-80465-6
Online ISBN: 978-3-642-80463-2
eBook Packages: Springer Book Archive