Psychological Testing and Assessment,

Psychological Testing and Assessment,

In the text of Psychological Testing and Assessment, we learned about three most common and different measures to determine the reliability of an assessment. They are the test-retest reliability, parallel and alternate forms of reliability and the internal consistency reliability. All three measures have coefficients that are interpreted differently from one another.

The test-retest reliability coefficient is the average of two scores of the same test (Cohen, Swerdlik, & Sturman, 2018). A coefficient of =.50 is a very ow reliability. This coefficient could mean that a score of 100 was graded and a score of 0 was graded on the second test, or vise versa, averaging a 50. A coefficient of =.50 means it is unacceptable because of the great variance between tests. A test-retest would be useful in determining descriptive traits.

The alternate forms of reliability is when two different version of a test were created (Cohen, Swerdlik, & Sturman, 2018). In order for the reliability to be high, the difficulty level must be equal to each other (Cohen, Swerdlik, & Sturman, 2018). The parallel forms and alternate forms reliability coefficient of =.82 means that the reliability is standard reliability. It is not low or not high. It could be used as a reference for good or low reliability for other tests. Alternate forms of tests would be useful to minimize plagiarizing or cheating among test takers. A source of error variance to this measure would be the variance of difficulty between versions of the tests.

The internal consistency reliability is a measure that doesn’t require different test versions or taking the same test twice (Cohen, Swerdlik, & Sturman, 2018). It measures how items on the test relate or differ from one another (Cohen, Swerdlik, & Sturman, 2018). A source of variance for this measure would be how the test was created. A coefficient of =.92 is a very high reliability and means that the items on the score are highly related to one another. This measure would be acceptable to use for tests such as rating or reviewing an item by using a scale.

Cohen, R. J., Swerdlik, M. E., & Sturman, E. D. (2018). Psychological testing and

assessment: An introduction to tests and measurement (9th ed.). New York, NY:

McGraw Hill.