Test-retest

Reliability (of English "reliability ", German: . Reliability ) generally refers to the degree of accuracy with which a particular feature is measured by a psychological process. Because the majority of psychological tests assumes that each measurement has distortions that do not go back to the feature. Reliability therefore means simply, to what extent one can rely on a test result.

However, reliability refers to only a generic term for a range of concepts, each of which affect only certain aspects of the measurement accuracy. One of these concepts is the test-retest reliability. It denotes the degree of compliance of test results in the same subjects and the same test at multiple testing. The value for the reliability is expressed by the correlation coefficient of the two tests ( reliabilities are desired greater than 0.80 ).

The problem Reliabilitätsmessung on the test-retest reliability is because the stability of the measured characteristic is assumed. But change many psychological characteristics over time, as well as changed the size of a child over time. Another problem is the fact that a person can not remember for the new test on the test. This can lead to intelligence tests, learning effects, which the Reliabilitäts measurement influenced ( Scheinreliabilität ).

Other methods for Reliabilitätsmessung are parallel test reliability, the split-half method and the internal consistency.

• Method of Psychology