Skip to main content

Table 2 Common types of systematic errors in prevalence studies

From: Methods used in prevalence studies of disrespect and abuse during facility based childbirth: lessons learned

Type of Systematic Error

Description

Systematic Error in the Selection of Study Population

 Selection bias

This is the degree to which the survey estimate differs from the true value of the phenomenon due to the fact that study was conducted in a non-representative sample of the target population.

 Response rate bias

This type of occurs when a substantial number of sampled study subjects refuse/decline to participate in the study or do not respond to a part of the study that is relevant to the outcome variable.

Measurement Errors (can be random or systematic)

 Imperfect test error

Values in the study data set do not reflect the true values of the variable of interest due to inaccurate measurement caused by poor data collection instrument

It reflects validity (the degree to which the instrument measures accurately what it purports to measure).

 Interviewer-related error

This is distortion of the responses given by subjects caused by behaviors of the interviewer, such as leading or cueing the subject, which influence subjects’ responses.

 Courtesy/Desirability-related error

This error occurs when respondents do not report accurately on any event of interest because they don’t want to offend the person seeking their opinion.

 Recall-related error

The degree to which the study value differs from the true value because of inaccurate recall of information about the variable of interest.