Perils and potentials of self-selected entry to epidemiological studies and surveys

Niels Keiding, Thomas A. Louis

Research output: Contribution to journalArticlepeer-review

81 Scopus citations


Summary: Low front-end cost and rapid accrual make Web-based surveys and enrolment in studies attractive, but participants are often self-selected with little reference to a well-defined study base. Of course, high quality studies must be internally valid (validity of inferences for the sample at hand), but Web-based enrolment reactivates discussion of external validity (generalization of within-study inferences to a target population or context) in epidemiology and clinical trials. Survey research relies on a representative sample produced by a sampling frame, prespecified sampling process and weighting that maps results to an intended population. In contrast, recent analytical epidemiology has shifted the focus away from survey-type representativity to internal validity in the sample. Against this background, it is a good time for statisticians to take stock of our role and position regarding surveys, observational research in epidemiology and clinical studies. The central issue is whether conditional effects in the sample (the study population) may be transported to desired target populations. Success depends on compatibility of causal structures in study and target populations, and will require subject matter considerations in each concrete case. Statisticians, epidemiologists and survey researchers should work together to increase understanding of these challenges and to develop improved tools to handle them.

Original languageEnglish (US)
Pages (from-to)319-376
Number of pages58
JournalJournal of the Royal Statistical Society. Series A: Statistics in Society
Issue number2
StatePublished - Feb 1 2016


  • External validity
  • Internal validity, Non-probability samples
  • Representativity
  • Transportability
  • Unmeasured confounders
  • Web-based enrolment

ASJC Scopus subject areas

  • Statistics and Probability
  • Social Sciences (miscellaneous)
  • Economics and Econometrics
  • Statistics, Probability and Uncertainty


Dive into the research topics of 'Perils and potentials of self-selected entry to epidemiological studies and surveys'. Together they form a unique fingerprint.

Cite this