Measurement error and misclassification

Aim

Increase the awareness of the problems caused by measurement error and misclassification in statistical analyses and remove barriers to use statistical methods that deal with such problems by publishing scientific articles and presenting papers and workshops at conferences. See also Freedman and Kipnis (2018).

Key messages

Only a minority of published papers present estimates that are adjusted for measurement error.
Considering measurement error is necessary because it may have an impact on the study results.
Special statistical methods are used to account for measurement error
Additional information is required about the type and size of the measurement error to adjust for measurement error.

Recent publications

Videos: STRATOS TIPS on Measurement Error

Part 1: What is measurement error?

Part 2: Why does measurement error matter?

Part 3: What can we do about measurement error?

Scientific Papers

Philipps V, Freedman L, Deffner V, Helmer C, Jacqmin-Gadda H, Boshuizen H, Thiébaut ACM, Proust-Lima C (2026). Including an infrequently measured time-varying error-prone covariate in survival analyses: a simulation-based comparison of methods. American Journal of Epidemiology. Link

Thiébaut AC, Boshuizen HC, Midthune D, Wallace MP, Kipnis V, Gustafson P (2026). Five misconceptions about categorizing exposure variables measured with error in epidemiological research. American Journal of Epidemiology. Link

Thiébaut ACM, Perperoglou A, Sedki M, Ferreira Guerra S, Gustafson P, Harrell FE, Sauerbrei W, Abrahamowicz M, Freedman LS. Methods for adjusting for covariate measurement error in flexible modelling of functional form: designing a blinded, controlled neutral comparison simulation study. Biometrical Journal. (provisionally accepted)

STRATOS Initiative

The topic group "Measurement error and misclassification" is a member of the STRATOS Initiative (STRengthening Analytical Thinking for Observational Studies) which is a large collaboration of experts in many different areas of biostatistical research. Ongoing research, discussions and activities within STRATOS are conducted in nine topic groups and several cross-cutting panels.