Validity is an important characteristic of a scientific instrument which means truth.
It is defined as that indicates to which an instrument measure what it is supposed to be measure. This is a scale is said to be valid when it correctly measures what it is expected to measure. Thus, validity refers to the degree to which a test measures what it claims to measure. Validity is concerned with generalizability.
Reality and validity are two-dimensional of the same thing, which is test efficiency. Validity is the correlation of the test with some outside independent criteria and reliability of the test in the self-correlation of the test.
High reliability means that represents arrows to shoot from the same bow would hit the target in essentially the same place but not necessarily the intended place.
High validity means that the bow would shoot true every time.
For example, Mr A obtained 80% marks in MBS 1st yr is reliable but not validity.
There are three types of validity.
1. Content validity
It is the representative of the measuring instruments as regards to the content, the substance and the matter and the topic of the universe dealt. Or it is also defined as the extent to which a measuring instrument provides adequate coverage of the topic under study. If the instrument contains a representative sample of the universe, the content validity is good.
Suppose U is the universe of items and u is the subset of U I,e u contains the items which are the substance of U; then u is said to be valid representation of U. Its determination is primary judgement and initiative.
Let A, B, C are contented representing arithmetic operations respectively given in U, then any sufficiently large amount of u of U is said to be content valid, if it represents A, B and C approximately equally. In order to be content valid, not only should the items selected contain the common trend, say of the attitudes under the study in the case of attitude test but also they should cover the full range of the attitude and cover it in a balanced way.
But unfortunately, it is not so possible to do so. Therefore, the determination of representative sample of the universe is primarily judgement and intuitive. It can also be determined by using a panel of persons who shall judge how well the measuring instrument meets the standards, but there is no numerical way to express.
2. Criterion-Related validity.
Criterion-related validity is ascertained by comparing the test or scale scores with an external variable known or believed to measure a phenomenon under the study.
Criterion-related validity relates to our ability to predict some outcomes or estimate the existence of some current condition. This form of validity reflects the success of measures for some empirical estimation purpose.
For example, if a questionnaire is designed to identify the psychiatrically distributed person, it’s validity can be ascertained by comparing its diagnosis items with those made by a psychiatrist on the basis of his/her clinical experience.
One concern of criterion validity is to predict the criterion on the basis of some measure to predict the success and the failure of students from academic aptitude test is one example of the use of criterion-related validity test. Prediction test is not necessary a forecast. One predicts about dependent variable from an independent variable. One predicts the exitance or non-existence of a relationship between variable. Also, this approach leads to two categories of validity.
1. Predictive validity.
This refers to the usefulness of a test in predicting some future performance.
2. Concurrent validity.
It refers to the usefulness of a test in closing relating to another measure of known validity.
Thus, predictive validity is concerned with how well the scale can forecast a future criterion and concurrent validity is concerned with how it can describe a present one.
3. Construct validity.
It is the most complete and abstract. Construct validity is defined as that which possess the degree that it confirms to predicted correlation with other theoretical propositions. Many of the social parameters are abstract and their measurements are therefore invariably indirect. Rigorous testing of the measurement of such parameters (Say the caring attitude) is very difficult. One approach is to postulates the types and degree of association to see whether they confirm the postulates formulation. The approach is the constant validity.
For instance, if one is concerned with rigours of the measurement of a scale, although one could not test the validity directly, one may judge the success of the measurement indirectly. For instance, one may be successful in discrimination those who are temple-goers and does not.
Kerlinger, F.N. Foundation of Behavioural Research. New Delhi: Surjeet Publication, 2000.
Kothari, C.R. Research Methodology. India: Vishwa Prakashan, 1990.
Singh, M.L. and J.M Singh. Understanding Research Methodology. 1998.
Singh, Mrigendra Lal. Understanding Research Methodology. Nepal: National Book centre, 2013.