Discuss the criteria according to which you would evaluate the psychometric properties of a psychological test before using it.

Authors Avatar

DISCUSS THE CRITERIA ACCORDING TO WHICH YOU WOULD EVALUATE THE PSYCHOMETRIC PROPERTIES OF A PSYCHOLOGICAL TEST BEFORE USING IT.

Devising tests is an arduous and difficult task. One which requires extreme attention to detail, and careful construction, if accurate and reliable results are to be drawn. Not only must the test be valid and reliable, it must also be user friendly. I would seek to evaluate the psychometric properties of the new test in accordance with the Professional Board for Psychology's criteria, which are very specific.

VALIDITY:

Validity can be described as the appropriateness, meaningfulness and usefulness of the test. That is to say, how effective and valid is the test as a measuring instrument, and does it test what it sets out to test? In assessing if the psychological test is valid, I would question what the test is measuring, and if this is infact what it is intended to measure.

If I were to check the validity of the psychological test merely on its face validity, I would not know how accurately the test really measures what it claims to measure, as face validity is not an accurate form of validity because it relies only on the appearance of validity. It is thus too subjective and lacks an empirical basis.

Therefore, the test would need to be administered to a test group and from the answers and scores of the test group, evaluation of the test's validity and reliability could be ascertained.  This would be done in the following way:

I would start assessing the validity of the test by checking how accurately the questions and tasks elicit information relevant to that on which the test aims to seek information/data. I would relate the outcome of the questions back to the original intention of the test. This is called content validity I might use a panel of experts (expert judges) to test if the questions have content validity.

This would be done as follows:

Items which the judges feel are of extreme relevance will be put in 'D' i.e. strong relevance

Items where judges disagree to the extreme will be put in 'B' i.e. weak relevance  

Items disagreed upon will be placed in 'C'

Items agreed not to be appropriate for the test go in 'A'

Content validity =         D

                  (A + B + C + D)

Unfortunately, calculating content validity can be seen as very subjective because it calls into question the judgement of 'experts'.

Criterion-related validity can be used to assess the validity of the test by determining the relationship between scores on a test and an independent, nontest criterion e.g. the test is validated when it correctly predicts an out-come measure/criterion such as an examinees average marks for a subject.

Predictive validity is a form of criterion-related validity. To ascertain criterion validity I would calculate a coefficient of validity based on a set of criterion scores. Predictive validity determines the degree to which scores on a test predict real performance. Unfortunately the criterion measures are often only realized in the future, (sometimes months or even years after the test scores are gathered).

Join now!

Concurrent validity has similarities to predictive validity, however, rather than predicting, it postdicts. The criterion measures and test scores are obtained simultaneously.  I would ascertain this validity by correlating the test scores with data already gathered e.g. I would correlate the new test with an existing test.  

To test Convergent validity I would see how highly the test correlates with other tests or factors where, in theory, there should be an overlap. To establish Divergent validity I would see how poorly the test correlates with other tests or factors where, in theory, there should be no overlap.

...

This is a preview of the whole essay