Major Types of Validity:
Different aspects or facets of validity questions are explained by the type of validity, and number of indirect and logical methods have evolved to assess validity.
a) Content Validity: This is concerned with the sampling adequacy of the content area being measured. It refers to the fact that the items included in -the tool represent a reasonable sampling of all possible itemshehaviours that make the domain being measured. The content area to be tested must first be studied to make certain that all major aspects or subcategories are represented in correct proportions. Experts in the content area are often called upon to analyse the items to see if they adequately represent not only the content universe (or domain) but also in cohect proportions. They also point out errors and ambiguities in items and make recommendations for changes. Content validity assessment also involves a more elaborate procedure by which items are developed, given to experts for judgment and determination of percentage of items for each category. Percentage of agreement between the judges is considered as the basis for inclusion or rejection of items. After item analysis, content validity needs to be assessed again to determine if the correct proportion of items for each subcategory has been maintained. If the proportion has not been maintained, exchange of items and sometimes addition of items are made until the content is sampled appropriately while retaining an acceptable level of reliability.
b) Criterion-related Validity: Seeks to establish that the measure correlates to another criterion rather than to ascertain how well the tool measure the particular phenomenon. For example, the instrument measuring anxiety state is said to be valid if the anxiety scores correlate with some criterion, such as, persons who are known to be anxious (e.g. student prior to a difficult examination) would be expected to have high scores on the anxiety scale. That is, groups known to possess or not to possess the characteristics being measured would be the criterion. Sometimes distinctions are made between two types of criterion validity, predictive and concurrent. Predictive validity refers to the adequacy of an instrument in predicting performance or behaviour of individuals on some fbture criteria, e.g. how well does the admission test predict the future performance of the student. This can be got by correlating admission test scores of first year nursing students with subsequent achievement in nursing in second, third and kourth year. Here, the predictive validity of admission test for nursing is being evaluated.
c) Concurrent Validity, on the other hand, involves concurrent (i.e. simultaneous) measurement of the scale being validated. The difference between predictive and concurrent validity, then, is the difference in timing of obtaining measurements on a criterion. One of the hardest aspects of criterion related validity is selection and measurement of criterion. It must be relevant, possess reliability and validity and be free from bias.