Last update:
04-Apr-2013

Questionnaires in epidemiological studies must have high validity and reliability. A preliminary pilot study with a few participants (roughly 30−50) should always be conducted in order to estimate the validity and the reliability of a questionnaire and to correct as far as possible any errors and omissions detected in this way. The validity of a questionnaire is the extent to which the questionnaire measures the study concept or, otherwise, the variate that it is supposed to measure. Increase in validity of a questionnaire is associated with a decrease in systematic error. Evaluation of the validity of a questionnaire includes evaluation of content validity, construct validity, criterion validity and face validity. Content validity is the extent to which a questionnaire covers the different dimensions of the concept that is being measured. Construct validity is the extent to which a questionnaire reflects the real theoretical meaning of the concept that it is supposed to measure. Estimation of construct validity can be performed by factor analysis, convergent validity, discriminant validity and the known groups' method. Factor analysis is a complex mathematical procedure, comprising explanatory factor analysis and confirmatory factor analysis. Criterion validity is about the use of a criterion, i.e., a questionnaire with proved validity and reliability, as a reference method or "gold standard" for estimation of the validity of a new questionnaire; the new and the proven questionnaires must measure the same concept. The reliability or precision of a questionnaire concerns the degree of stability or consistency with which the questionnaire measures the concept that it is supposed to measure. Increase in the reliability of a questionnaire is associated with a decrease in random error. The estimation of reliability of a questionnaire includes test-retest estimation, alternative forms reliability, split-half reliability, internal consistency reliability and inter-observer reliability.

Key words: Factor analysis, Questionnaire, Reliability, Validity.