PSY 380 Lecture Notes - Lecture 11: Criterion Validity, Face Validity, Negative Number
Document Summary
Establishes that your test contains only good items that behave consistently. Draws on these quantitative methods with developers needs and expectations. Item difficulty index, item reliability index, item validity index, item discrimination index, item characteristic curve. Classic test theory and item response theory. Proportion of test takers who answered an item correctly. Computed on every individual item on test. The higher the number, the easier the number. Allows you to calculate an index of average item difficulty. Optimal average item difficulty is . 5, ranging from . 3 - . 8. Moderately difficulty level helps preserve reliability and validity of instrument as a whole. Enough potential for variance to degree of reliability and thus. Best at discriminating between high (or a lot of trait) performance or validity low performance (or not a lot of trait) To help you measure everyones performance level. Works best for right or wrong answers. Can incorporate effects of guessing for selected response items.