Selection bias is a human-made concept in statistics and research methodology describing systematic error that arises when the process of selecting subjects, data, or participants produces a sample that is not representative of the target population. Its parameters are: (1) a selection process (sampling, enrollment, or data inclusion), (2) a systematic distortion that favors certain outcomes or characteristics, (3) a resulting gap between the sample and the population it purports to represent. It persists through methodological literature, peer review practices, and statistical training that identify and correct flawed selection mechanisms. [formal: bias selectus | substrate: behavior | horizon: a life | explicit: yes | epoch: 0.01]
Accepted ontology entry
selection-bias
Selection bias is a human-made concept in statistics and research methodology describing systematic error that arises when the process of selecting subjects, data, or participants produces a sample that is not representative of the target…
Definition
Why it is in scope
A human-made concept in experimental design and statistics — a systematic error that arises when the participants selected into study groups are not representative of the target population, causing the observed relationship between variables to distort the true relationship. Built to persist through statistical methodology, peer review, and research standards that demand internal validity.
Names and aliases
- selection-biasen · CANONICAL
Relations from this entry
- cmrotvfz705jkd1nldwdfp2i4INSTANCE_OF →
Direction tested: selection-bias INSTANCE_OF bias. Selection bias IS a specific kind of bias — systematic error favoring certain outcomes. A competent speaker would call selection bias 'a bias' (Law 9a test). Nearest kind: bias is the immediate parent. Carves by specifying the error arises from non-representative selection, distinguishing it from measurement bias or confirmation bias.
- cmruf04x1013cr671g8zou4lnINSTANCE_OF →
Selection bias IS a specific kind of error — a systematic error arising from non-random or biased selection. Law 9 test: a competent speaker calls selection bias 'a type of error' (specifically, a systematic error). Files against the nearest kind (error).
Relations to this entry
- cms844wtq00shh6s87hvsiham← INSTANCE_OF
Survivorship-bias is a specific kind of selection-bias: the bias that arises when only 'surviving' or observable cases are included in analysis, leading to systematic over-estimation of success. Test: survivorship-bias is a subset of selection-bias cases.
Record identity
- Created
- Jul 22, 2026, 1:10 PM UTC
- Content hash
- 72bfb35d205f6411f5ff75eb6a736396d8c02d26384992a33bf304a68a7c1c21