Describe two of the four methods of establishing reliability. How are these methods similar? How are they different?

One method is the test-retest method.  Test/retest administrations occur at two different times with the same test presented to the same group. This method tests the same thing at two different times. The book uses the ruler example; however another example would be giving a group of students a geography/history test at the beginning of a semester and then giving them the same test at the beginning of the next semester, the results should be somewhat similar (Cohen & Swerdlik, 2010). Another method to establish reliability is alternate-forms or parallel-forms. Administrators will give a parallel test (second test) after they have administered the first test. This helps to check for errors amongst the test and deems it reliable or unreliable (Cohen & Swerdlik, 2010). The methods are similar because each one uses two tests. Each test has the initial test and a follow up or second test. Both methods are looking for the same type of measurements.  However, these two methods also differ because in alternate-forms/parallel-forms, the administrator and test are looking for the average or mean. With test-retest method, the administrators and tests are looking for the same answer as in the first test. Also with this method, there should not be many factors that can skew or differentiate the second test because it is the same method, just at a different place in time. For the alternate-forms/parallel-forms, there are several factors that can contribute for this second test to differ from the first (Cohen & Swerdlik, 2010).                            

 

 

Reference
Cohen, R. J., & Swerdlik, M. E. (2010). Psychological testing and assessment: An introduction to tests and measurements (7th ed.). New York, NY: McGraw-Hill.