Epidemiology · Reference
What is information bias?
Information bias is a systematic error arising from inaccurate measurement or classification of exposure, outcome or other variables in a study. Also called measurement or misclassification bias, it concerns how data are obtained from those in the study, as distinct from who is selected into it.
What information bias is
Information bias arises during the measurement and recording of data, once participants are already in a study. If exposure, outcome or confounders are assessed with error — through faulty instruments, imperfect memory, inconsistent diagnostic criteria or inaccurate records — some participants are classified incorrectly. When this misclassification is systematic rather than random, it biases the result. Information bias is therefore often described as misclassification: people are sorted into the wrong exposure or outcome categories. It sits alongside selection bias and confounding as one of the principal threats to a study’s validity.
Differential and non-differential misclassification
Misclassification comes in two forms with very different consequences. Non-differential misclassification is unrelated to the other variable — for example, exposure is measured with equal inaccuracy in those who do and do not have the outcome. For a binary exposure this typically biases the estimate toward the null, diluting a real association.
Differential misclassification depends on the other variable — for instance, exposure recorded more accurately in cases than controls. This can bias the estimate in either direction, exaggerating, hiding or reversing an association, which makes it the more dangerous of the two.
Common forms
Several named patterns are forms of information bias. Recall bias arises when participants remember past exposures with differing accuracy, a classic differential problem in case-control studies. Interviewer (observer) bias occurs when those collecting data probe or record differently depending on participants’ status. Reporting and social-desirability bias arise when respondents under- or over-report sensitive behaviours. Surveillance (detection) bias occurs when one group is monitored more closely and so has more outcomes detected. Each is a route by which measurement systematically misclassifies people.
How it is reduced
Information bias is mainly tackled through careful measurement design: using validated instruments and standard case definitions, drawing on objective records where possible, training data collectors and applying procedures identically across groups, and blinding assessors to exposure or outcome status so that knowledge cannot influence measurement. Unlike some confounding, it usually cannot be corrected by analysis after the fact, though validation sub-studies and quantitative bias analysis can help gauge its likely effect. Reporting standards such as STROBE ask authors to describe measurement methods so readers can assess misclassification. This page defines the concept in general methodological terms only.
Key facts
At a glance
- Definition: Systematic error from inaccurate measurement/classification
- Also called: Measurement bias; misclassification bias
- Non-differential: Usually biases toward the null (binary exposure)
- Differential: Can bias an estimate in either direction
- Reduced by: Validated tools, standard definitions, blinding
Common questions
FAQ
What is information bias?+
Information bias is a systematic error that occurs when exposure, outcome or other variables are measured or classified inaccurately, placing participants in the wrong categories. Also called measurement or misclassification bias, it distorts the estimated association and concerns how data are obtained rather than who is selected into the study.
What is the difference between differential and non-differential misclassification?+
Non-differential misclassification is measurement error unrelated to the other variable; for a binary exposure it usually biases the estimate toward the null. Differential misclassification depends on the other variable — such as exposure measured more accurately in cases than controls — and can bias the estimate in either direction.
How is information bias different from selection bias?+
Information bias arises from how variables are measured or classified once people are in the study, whereas selection bias arises from who is included in or retained by the study. They occur at different stages and call for different remedies, mainly careful measurement versus careful sampling and follow-up.
The step most authors miss
Doing CRediT right? Don’t stop at the statement.
A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.
Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.
