PSY 370 Lecture Notes - Lecture 29: Psy
Document Summary
Each of these deals with a different type of error variance. Need to identify which kinds of error we are most concerned with in a particular testing context. Goal: separate true score from error caused by temporary factors. Give the same test to the same group of people at 2 difference points in time. Need to think about the appropriate test-retest interval. What"s left over = proportion of the variance due to fluctuations over time. Sometimes, we don"t want to administer the same test twice to the same person. Certification exams, any time we are concerned about practice effects. Solution: create two alternate forms of the same test. Designed to sample the test content equally well. 10 items about addition, 10 items about multiplication, etc. Goal is to match items as closely as possible. To test: give both versions of the test to the same group. Proportion of the variance due to true score = the correlation between.