PSY 440 Chapter 2.4: PSY 440 Chapter 2.: PSY440 Chapter 2.4
Document Summary
Where a test is given more than once to see how stable the scores are. Where two forms of a test are given to the same people at the same time and the scores on both measures are correlated to provide a reliability estimate. Measures of reliability that assume the test or measure assesses one thing. If reliable, the test will be internally consistent and the items will be intercorrelated. An internal consistency estimate in which two halves of the test are treated as two small tests and measures how consistent the scores are between the two tests. An index of the intercorrelation among scale items. The average of all possible split-halves reliability estimates. Where ratings of one rater are correlated with the ratings of another rater. The extent to which the measure is actually measuring what it is supposed to measure. A process demonstrating a measure was developed in a way that sampled the domain of interest.