Guidelines for Validating Research and Evaluation Instruments

Abstract

The paper examined guidelines for validating research and evaluation instruments. Validity as the extent to which research and evaluation instruments measure what they were designed to measure, is a very important quality of any measuring instrument. The concepts of validity, research and evaluation were explained in the paper. The guidelines for validating an instrument were explored, and the paper identified and proposed the following: identification and definition of the construct derived from psychological theories, prior research, systematic observation and analysis of relevant behaviour domain; preparation of the items of the instrument to fit the construct definition; giving draft copies of the instrument items to experts in the  area of the construct identified; giving the draft copies of the instrument to other experts different from those experts used earlier to determine the suitability and relevance of the items; selection of the most relevant items from the initial item pools using the mean statistic; correlation of the scores of the instrument with the scores of already existing standardized instrument on the same construct; and other appropriate internal analysis such as factor analysis of item cluster or subsets. The paper concludes that it is necessary that research and evaluation instruments be validated using any of the steps outlined.