Open topic with navigation
Bias is a systematic error that leads to an incorrect estimate of effect or association. Many factors can bias the results of a study such that they cancel out, reduce or amplify a real effect you are trying to describe.
Epidemiology categorises types of bias, examples are:
- Selection bias - e.g. study of car ownership in central London is not representative of the UK.
- Observation bias (recall and information) - e.g. on questioning, healthy people are more likely to under report their alcohol intake than people with a disease.
- Observation bias (interviewer) - e.g. different interviewer styles might provoke different responses to the same question.
- Observation bias (misclassification) - tends to dilute an effect
- Losses to follow up - e.g. ill people may not feel able to continue with a study whereas health people tend to complete it.
Some strategies to combat bias:
- multiple control groups
- standardised observations (e.g. blinding (don't know if placebo or active intervention) of subject, observer, both subject and observer (double blind) or subject, observer and analyst (triple blind))
- corroboration of multiple information sources
- use of dummy variables with known associations
See also: confounding.
Copyright © 2000-2019 StatsDirect Limited, all rights reserved. Download a free trial here.