Details

Autor(en) / Beteiligte

• Hummel, Thomas J

Titel

Tables of Critical Values for Testing Clinical Hypotheses

Erscheinungsjahr

1993

Quelle

ERIC

Beschreibungen/Notizen

• Tables for testing clinical hypotheses about counseling clients are presented, and their use is described. Primary uses are determining a critical value or cutoff score on a predictor that can be used to classify an individual as a member of a clinical group, and instructing those unfamiliar with the limitations of tests about the conditions under which tests are inadequate for classifying individuals. The researcher states as the initial clinical hypothesis that the client is not a member of a particular clinical group, and decides on the maximum probability of incorrectly classifying any individual into that group. Table 1 is used when the criterion variable is normally distributed. Table 2 is for use when the criterion variable is dichotomous. Once the appropriate table is selected, the researcher must select an alpha-level, the maximum probability of error that will be accepted when classifying any client as a member of a clinical group. Specific directions are given for the use of both tables. The use of the tables in training counselors is also described. These tables allow a quick survey of the circumstances under which tests have little or no practical value. In general, they provide a practical view of a test's ability to classify individuals for clinical purposes. A technical appendix explains how the tables were developed, and both tables are included. (SLD)