Test theory (psychology)

from Wikipedia, the free encyclopedia

The test theory is a part of the theory of measurement , the u. a. is used in differential psychology and psychological diagnostics .

In psychometry , a test theory is based on a mathematical model that allows certain statistical relationships between certain features and empirical test values ​​to be expected. After a test has been carried out, conclusions are drawn from the test results to the characteristics with the help of test theory. The test theory also provides quality criteria for psychodiagnostic methods by means of which the quality of the test can be assessed. The main field of application is test construction, i.e. the design of tests that optimize the quality criteria.

The most common are Classical Test Theory and Probabilistic Test Theory (Item Response Theory).

Names of tests for the same construct

Two tests for the same construct are structured as follows (in classical test theory ):

  • congeneric tests: the expected value of the test, the " true value " of one test is a linear function of the other:
  • Practically τ-equivalent tests: The "true values" of the tests only differ by one constant:
  • τ-equivalent tests: The "true values" agree:
  • parallel Tests: The tests are τ-equivalent and the scattering of the measured values is equal to: ;

Individual evidence

  1. Markus Bühner : Introduction to the test and questionnaire construction . Pearson Deutschland GmbH, 2011, ISBN 978-3-86894-033-6 , p. 152 ( limited preview in Google Book search).
  2. ^ Nicola Döring, Jürgen Bortz: Research methods and evaluation in the social and human sciences . Springer, 2015, ISBN 978-3-642-41089-5 , pp. 467 ( limited preview in Google Book search).
  3. Helfried Moosbrugger, Augustin Kelava: Test theory and questionnaire construction . Springer, 2011, ISBN 978-3-642-20072-4 , pp. 425 ( limited preview in Google Book search).

literature

Web links