Question

For this task you will create a chart describing the different methods of establishing the reliability of a psychological instrument. In one column write the method (e.g., test-retest method) and on another column write a brief description for each of the methods.

Answer 1

Reliability measures whether a test is able to produce consistent results over a period of time. For example, an intelligent test should show similar results for an individual who takes it at different periods of time. Reliability of a test can be assessed through the following ways

Test-Retest method	Determines consistency of test across different times. The test is administered two times on separate occasions to the same individual. If results are similar and have high correlation, indicates reliability. The time gap between the administrations is crucial as a shorter gap will have a higher correlation between the results.
Parallel forms reliability	Two different tests are used to measure the same variable. The two tests are administered to the same subject at the same period of time. Two different tests are administered even though they measure the same variable as the subject will not remember the questions as in the case of test-retest method. Examples of tests, GRE , GMAT
Inter-Rater reliability or Inter Observer Reliability	The test is scored with two independent judges and their scores are compared in order to assess the consistency between the two raters. The interrater reliability is assessed through Cronbach’s α for quantitative judgements and cohen’s kappa is used for categorical judgements.
Internal Consistency Reliability	It measures the consistency of individual’s responses across the items within a test. Internal consistency is measured through Split Half Reliability method. Here the items of the test are divided into two halves and the scores of the halves are compared with each other to determine the relationship between the sets. Cronbach’s α is the statistical test used to determine consistency