Linked from

The 28 pages that link to Selection bias, each with the reason it gives.

• Causal inferenceRelated: Selection into a study or analysis can create misleading exposure–outcome relationships.

• Comparative methodRelated: Choosing cases based on outcomes can create misleading comparative patterns.

• EpidemiologyRelated: Who enters a study can create or conceal population associations.

• MemoirRelated: Choosing which memories to include can make a memoir’s version of a life partial.

• ConfoundingCompared with: It arises through selective inclusion, rather than necessarily through linked exposure and confounder variation.

• Publication biasNarrower topic: Publication bias is a research-literature form of this broader selection problem.

• Statistical inferenceRelated: A distorted sample can make population conclusions systematically misleading.

• Sampling biasNarrower topic: Sampling bias is one form of selection bias, arising specifically from how a sample is drawn.

• Observational studyRelated: Who enters or remains in a study can distort its measured associations.

• Nonresponse biasNarrower topic: Nonresponse bias is a specific form of bias caused by selective participation.

• Natural experimentRelated: An apparently external event can still produce selected groups that undermine comparison.

• PaleopathologyRelated: Preservation and burial practices determine which individuals enter paleopathological samples.

• RandomizationCompared with: A flawed assignment or enrollment process can reintroduce selection bias.

• Anthropic principleNarrower topic: Observer-dependent sampling is a special case of selection bias.

• Quasi-experimentRelated: Differences in who receives treatment can undermine group comparisons.

• Social contagionCompared with: People often choose peers who already share their behaviors, mimicking contagion.

• Missing dataRelated: Analyzing only complete records can select a systematically different subset.

• Asymptomatic infectionRelated: Studies recruited through clinics may underrepresent infections without symptoms.

• AdditionalityRelated: Participants may already be more likely to produce the outcome, inflating apparent additionality.

• Control groupRelated: Unequal selection can make a control group unlike the population receiving the intervention.

• Data collectionRelated: Recruitment and sampling decisions can make collected observations unrepresentative.

• Probability samplingRelated: A probability design controls selection mechanisms, but implementation and nonresponse can still create bias.

• Sample (statistics)Compared with: It arises when inclusion in this sample is related to the outcome of interest.

• Low surface brightness galaxyRelated: Surface-brightness thresholds cause surveys to undercount diffuse galaxies.

• Regression toward the meanRelated: Selecting cases for extreme scores can make subsequent scores look improved even without a causal effect.

• Large quasar groupRelated: Uneven quasar detection can create or exaggerate apparent spatial concentrations.

• Observational cosmologyRelated: Survey selection can imitate or conceal trends in distant objects.

• Cinderella effectRelated: Cases drawn from official reports may overrepresent severe incidents and distort estimated risk.