validity

An index of how well a test or procedure in fact measures what it purports to measure; an objective index by which to describe how valid a test or procedure is.

- concurrent v. an index of criterion-related v. used to predict performance in a real-life situation given at about the same time as the test or procedure; the extent to which the index from one test correlates with that of a nonidentical test or index; e.g., how well a score on an aptitude test correlates with the score on an intelligence test.

- construct v. the extent to which a test or procedure appears to measure a higher order, inferred theoretical construct, or trait in contrast to measuring a more limited, specific dimension.

- content v. the extent to which the items of a test or procedure are in fact a representative sample of that which is to be measured; e.g., items relating to ability in arithmetic and defining words are appropriate content for an intelligence test.

- criterion-related v. the degree of effectiveness with which performance on a test or procedure predicts performance in a real-life situation; e.g., a good correlation between a score on an intelligence test such as the Scholastic Aptitude Test and one's 4-year college grade point average.

- face v. the extent to which the items of a test or procedure appear superficially to sample that which is to be measured.

- predictive v. criterion-related v. used to predict performance in a real-life task at a future time. See construct v., criterion-related v..

* * *

n.

an indication of the extent to which a clinical sign or test is a true indicator of disease. Reduced validity can arise if the tests produce different results when conducted several times on the same person under identical conditions (i.e. reduced reproducibility, reliability, or repeatability). This may be because the same observer gets different results on successive occasions (intraobserver error) or because a series of different observers fail to obtain the same result (interobserver error). Such errors may arise because of a true difference in observation and/or interpretation or because of a preconceived notion (often unconscious) by the observer, which influences either his judgment or the tone and manner with which he questions the patient. Compare intervention study.

* * *

va·lid·i·ty (və-lidґĭ-te) the extent to which a measurement, test, or study measures what it purports to measure.