Self-selection (selection bias)

‘The choosing of oneself for something’ and refers to relationships between subject and environmental variables which arise when people (and animals) with different biological, behavioural, and psychological characteristics select different types of environments or outcomes.

An example of this comes from the manner in which the media (radio, TV, newspaper, magazines, news websites, etc.) informally conducts unscientific surveys about issues by extending open invitations to information consumers to call-in by phone, to write, or to register their response via a web-page. The sampling is therefore limited to those people who have listened to, watched, or read this invitation, from whom a sub-sample amongst that group then choose to respond. We might wonder why some people responded whereas others did not.

In consequence, it is unclear in what way or to what extent – if at all – results from any such self-selected samples might sensibly be generalized to a population.

Statistical studies may also be subject to a non-random response rate – meaning people who choose to return completed survey forms may hold opinions which don’t accurately reflect the group of people sent survey forms. Likewise, people who volunteer for a scientific study, for example, a health-focused study concerning heart disease, may already differ from the population with the condition by acting in ways which do not parallel that population. Perhaps they are more conscious of actions which lessen or moderate the condition and are more likely to act on this knowledge (appropriate diet and exercise).

Which U.S. state has the highest incidence of respiratory illness? what if the answer is not a state with very poor air quality which we might assume to be likely to be associated with such illness – might it be because people with respiratory illnesses have tended to move to that state (perhaps to retire) and subsequently die there?

Selection bias should always be carefully considered as a (potential) source of error when examining ‘interesting’ correlations between variables.

(see also: Studies – Scientific or Empirical, statistical evidence, statistical significance, experimental research, peer review, scientific approach, science, scientific methodology, pseudo-science, systematic evidence, anecdotal evidence, (conceptual) models, operational definition, critical thinking, theory (scientific), falsifiability, testability) .