Validity
From Wikiversity
|
Test validity refers to the extent to which a psychological test measures what it is purported to measure. Validity is commonly discussed in terms of there being three main types: content validity, construct validity, and criterion validity. In classical test theory, validity is limited by reliability.
[edit] Content validity
Content validity refers to the degree to which the content of the test matches the content domain of the construct. For example, a test of ability to add two-digit numbers should cover the full range of combinations of digits. A test with only one-digit numbers, or only even numbers, would not have good coverage of the content domain.
Content validity evidence typically involves subject matter experts evaluating test items against the test specifications. Expert-based testing of content validity is distinguished from face validity which refers to prima facie estimates of whether a test appears to measure a certain criterion. Note that face validity does not guarantee that the test actually measures phenomena in that domain. Indeed, when a test is subject to faking (malingering), low face validity might make the test more valid.
[edit] Construct validity
Construct validity refers to the totality of evidence about whether a particular operationalization of a construct adequately represents what is intended by theoretical account of the construct being measured.
Construct validity involves empirical and theoretical support for the interpretation of the construct. Lines of evidence include statistical analyses of the internal structure of the test including the relationships between responses to different test items (see internal consistency). It also includes relationships between the test and measures of other constructs. Construct validity is not distinct from the support for the substantive theory of the construct that the test is designed to measure.
There are two main approaches to construct validity:
- Convergent validity: refers to the degree to which a measure is correlated with other measures that it is theoretically predicted to correlate with.
- Discriminant validity: describes the degree to which the operationalization does not correlate with other operationalizations that it theoretically should not correlated with.
[edit] Criterion validity
Criterion validity involves the correlation between the test and a criterion variable (or variables) taken as representative of the construct.
There are two types of criterion validity:
- Concurrent validity: The test data and criterion data are collected at the same time, e.g.,
- Self-report and observer measures collected at the same time
- Predictive validity: The test data is collected first in order to predict criterion data collected at a later point in time, e.g.,
- employee selection tests are often validated against measures of job performance.
- measures of risk of recidivism among those convicted of a crime can be validated against measures of recidivism.
[edit] Readings
- Cronbach, L. J. & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52, 281-302.
- Validity (statistics) (Wikipedia)
