What are the principles of employment tests?

[laji]

What are the factors that determine the success of an employment test?

[nbm]

1. Validity—The first important property which a test must possess is validity. Validity means that it should be able to measure what it purports to measure. The validity of a test is the degree to which it measures what it is intended to measure. In an employment situation, a valid test is one that accurately predicts the criterion of job success. There are five different types of validity.

(a) Content validity—A test has content validity to the extent that it is made up of the items which only measure situation or subject matter with the domain of interest. Content validity suffers when the test contains items which measure situations or subject matter outside the domain of interest. Content validity also suffers when the items in a test under represent the relevant domain.

(b) Face validity—Face validity relates to test whether the questions in a test are related to job or otherwise. However, face validity cannot be measured in statistical terms. Moreover, it does not really test the validity of test.

© Concurrent validity—This refers to the degree which test scores relates to the performance of employees presently on the job. This test is given to the present employees and their scores on the test are compared with their efficiency on the job. A high positive correlation shows the validity of the test.

(d) Construct validity—Predictive validity is evaluated by showing how well predictions made by the test are confirmed at some subsequent time. If those who made the highest score at the time of selection test also turned to be efficient later on, the test is valid.

(e) Predictive validity—Construct validity is evaluated by determining what psychological qualities a test measures, such as intelligence. Thus, it measures how well the test is measuring the hypotheses framed in this respect.

In business and industry, predictive and concurrent validity is the most important. However, before administering the test and testing its validity, one should be sure as to what qualities and traits these tests would measure. This is done through careful job analysis and determining the standard in this respect.

2. Reliability-The reliability of test is the consistency with which it yields the same scores throughout a series of measurements. Thus, if a test possesses high reliability, a person who is tested a second or third time with the same test and under same conditions will obtain almost the same result. There are three methods of testing reliability.

(a) Test — retest method — In this method, a group of subjects are given the same test twice after a certain interval in-between. The scores at two tests are compared. If both of these are positively associated, the test may be called reliable.

(b) Alternative form method — In this method, two separate, but equivalent, forms are given to the subjects. Such forms include items that are similar in nature. The scores on these tests are associated and a positive association shows the reliability.

© Split-halves method — This method divides the whole test into two similar and equal parts and the scores on these two parts are compared. A high association indicates internal consistency of the test.

3. Percentage of Present Employees Considered Satisfactory—A further probe about the usefulness of the tests may be made in terms of percentage of present employees who are considered satisfactory in a given employment situation. If this percentage is higher, the test may be called useful.

4. Test — A Supplementary Method —The tests should be used as an additional factor in selection procedure and should not be treated as the sole basis for selecting or rejecting a candidate. While administering a test, there should be proper control and standardisation so that results are comparable.

5. Standardisation. A test must be standarised — that administered understandard conditions to a large group of persons who are representative of the individual for whom it is intended. This is done to obtain norms or standards so that an individual's score can be compared to the scores of others in the representative sample of people or a defined group like college students, adults, etc.

6. Objectivity—A test must be constructed in such a fashion that two or more persons can score the responses to items questions or tasks in the same way.