Рет қаралды 506
While data collection via the use of research instrument is very essential in research, such data may remain bias if validity of the instrument is not properly conducted or not conducted at all. Validity is the degree to which a research instrument, test, questionnaire, rating scale, checklist, e.t.c. measures what it is supposed to measure, and permit the appropriate interpretation of the results. Until these two conditions are satisfied, such an instrument is not valid for a particular study. While validity is situational, there are different types of validity. In this tutorial, there are five types of validity, and this include: face validity, content validity, construct validity, criterion-related validity, and consequential validity, These types of validity serve different purposes and has different methods. However, construct validity is often regarded as the most important validity, Construct validity is considered as the overarching concern of validity. While criterion validity is of two types, whcih are concurrent and predictive validity; construct validity is also of two types, which are convergent validity and divergent or discriminant validity. The several methods of establishing validity of a research instrument include experts judgement, sample of laymen, correlation, facto analysis, structural equation model (SEM).