Validity of Psychological Tests

Introduction

            Along with reliability, validity is the second important psychometric property of a psychological test. The test standardization contains a validation process against an accepted criterion. A test to be valid must be reliable too. The low or zero reliability yields the inconsistence results that naturally negate the validity. Hence, all valid tests are compulsorily reliable but all reliable tests may not be valid. It is important to note that a test is valid only for a specific trait or purpose. 

Meaning

            The coefficient of validity shows the degree to which a test measures what it purports to measure, when compared with the accepted criterion (Freeman, 1965).

Needs of Validity

(i)        To standardized a test,

(ii)       For testing the predictions of a test against objective or actual outcomes, and

(iii)      For assessing the efficiency of a test by evaluating it against an independent criteria.

Types of Validity

(i)        Predictive Validity – Prediction means to forecast. The efficiency of a test to forecast and differentiate a behaviour or performance in real conditions.

(ii)       Face Validity – It is used to characterize test materials that appear to measure what the test’s author desires to measure. In other words test contents looks to fulfil the purpose of measurement. The face validity is a kind of subjective judgment.       

(iii)      Content Validity – It is estimated by evaluating the relevance of the test items. The contents (items) are supposed to measure the desired behaviour, trait or performance. Generally used for the testing the efficiency and educational achievements. It is achieved by expert analysis followed by statistical analysis.

(iv)      Factorial Validity – The factorial validity is established by a statistical technique known as factor analysis. It indicates the exclusive ability of differentiating among functional unities or primary mental abilities. It helps in devising a test which measures only one functional unity. Factorial validity ensures purity of a psychological test.

(v)       Construct Validity – A method of test validation that is employed to assess or measure the psychological process or traits as defined by the author. It is dependent upon the degree to which selected test items sample the class of activities explained by the mental process or traits. The construct validity ensures that the test scores should represent the trait or process under study.

(vi)      Concurrent Validity – It indicates the process of validating the new test by comparing it for agreement with some recent information. Recent means that the information must be obtained shortly before or very shortly after the test is given.

(vii)     Cross Validation – It means validation of a test on the different sample than the sample on which test was standardized drawn from the same population. It is the attestation of standardization process adopted for the test.

Conditions Affecting Validity

(i)      Sample Heterogeneity,

(ii)     Form of the relationship between test and criterion,

(iii)    Nature of the population or group,

(iv)    Selection of criterion for validation of test,

(v)      Sample size, and

(vi)    Length of test.

Methods of Calculating Validity

(i)        Simple Correlation – To calculate the validity the scores are correlated with each criteria.

(ii)       Biserial Correlation – This method is applied when one of the variables is in binary or categorical form i.e. pass or fail etc. and other variable can be in scores.

(iii)      Tetrachoric Correlation – Used when both the variable between which correlation is to calculated is given in dichotomous or binary form.

(iv)      Multiple Correlation – It is employed to when correlation is to be calculated among more than two variables. In other words this method is used when two or more variables are used as predictors.

(v)       Expectancy Tables – A table containing estimates base on calculated probabilities. These probabilities predicts that the individual who has been tested will achieve a specified score.

(vi)      Cut off Scores – A test score used as a point of demarcation between subjects (testees) who will be accepted & those who will be rejected.

References:

1.         Anastasi, A (1976). Psychological Testing. New York: Macmillan Publishing Co.

2.         Freeman, F. S. (1965). Theory and Practice of Psychological Testing. New Delhi: Mohan Primlani for Oxford & IBH.