How Does Validity Influence Psychological Tests?


For the test to be called standardized, the test should be valid. Validity is the second important characteristic of a good test. Validity can be evaluated by investigating how the test scores correspond to some criteria. A psychological test can either measure a person’s behaviour, his characteristics or traits, his personal accomplishments that distinctly reflects the attribute the test is designed to measure.Validity of any standardized test highlights the fact, how persistently a test can measure, what it is meant to measure. Reliability alone is not enough to call a test standardized and hence,validity is call forth. Validity and reliability are all interdependent. A test which is not reliable cannot be valid. A reliable test is the one which reports similar events in the similar way consistently.

Let us have a closer look at how the validity game goes.

  • For example, the intrepid test constructors in different business organizations and industries work in close collaboration with theorganizationalpsychologists, devising different tests of aptitude, attitude for screening candidates. Several different validity criteria can be used for each kind of test.

  • In other words, the validity criteria can be better assessed by observing the test takers in their everyday interactions. This will definitely reveal individuals who score high on the measures also have the kind of anticipated behaviour directed towards them, as desired in the test.

  • It should be made clear that assessing of the validity of any test requires a careful selection of appropriate criteria measures, and that reasonable people may disagree as to which criteria measure will be the best. But the fact that the criteria should be selected without utmost care is a point of emphasis. This is extremely true of intelligence tests. Reasonable people may disagree as to whether the best criteria measure of intelligence is school grades, college grades, teacher ratings or some other measures.

  • Valid tests can saturate us with a lot of useful information about people. Nonetheless, if we are to check on the validity of the test, we must settle on one or more criteria measures of the attribute the test is designed to measure.Whether it is intelligence, attitude,aptitude or some other attribute. Soon after the criteria measures have been chosen, the individual scores on the measures can be checked and compared with their scores on the test and the degree of correspondence can be examined for what it tells us about the validity of the test.

  • There are many different types of validity. Although not a proper scientific measure of validity, face validity makes sure such validity questions are absolutely related with what the test is intended to measure, so that it perfectly assesses the ability of the candidate.

  • As already said validity of psychometric test is usually expected to be lower than the reliability coefficients. Where reliability coefficient shoot up more than 0.90, coefficients of validity are often found around 0.50. Validities of 0.30 and 0.40 are more commonly found and it seldom rises to 0.60. Thus we have to remain satisfied with validitiesaround 0.30 and 0.40 between the test scores and criteria measures even if the reliability coefficient is quite high.

  • Tests which have low validity, actually relate very poorly to the criteria. However several tests with low but important validity can sometimes be beneficial if they areadministered together as a battery, and their results are considered together. The criteria related validity usually predicts future results or even present performance. Such validity is usually used to measureand correlate how well and accuratethe assessment inferences gel with other aspects of interest of the candidate.

  • Sampling validity is also known as content validity and it measures only the specific concepts or ideas which are aimed to study. It is called a sample because small samples are to be derived from each functional area of the concept under study,as everything cannot be studied in a wholesome way first hand. Sampling validity breaks up validity into small portions and then studies each and individual portion separately.

  • One possible explanation why the validity coefficients are generally lower than the reliability coefficient is that, the reliability of a test sets limit on how valid the test can be. A test that cannot give us reliable scores from one testing to the next is not likely to show dependable correlations with any validity criteria stepeither.Although tests which are valid generally have high reliability but vice versa is not true. There is no such rule that a high reliability must guarantee a high validity.

  • It is a common notion that a test cannot be valid if it is not reliable, that is if it does not score similar test results from time to time. The construct validity is the simplest form of validity which says that a test evaluates what it is entailed to be evaluated.A test which does not have a clear, operationalized and distinct objective or goal cannot have a good validity score. The formative validity is usually applied on the outcomes of an assessment test. This measure to what extent a technique is capable of getting information in order to help the organizationor the program under examine.

But only reliability and validity does not delineate the strength of standardization of any test. Norms in addition to reliability and validity are a symbol of a good test.

About the author

Manmohan Yadav

Add Comment

Click here to post a comment

Your email address will not be published. Required fields are marked *

Subscribe To Our Newsletter

Enter your email address:

Subscribe To Our Newsletter

Enter your email address: