08级7_ItemAnalysis.pptVIP

Primarily purposes of item analysis Improve the degree of discriminability, for better quality of the test; Adjusted test difficulty; Deepen the teacher’s understanding of the test, and adjust the teacher’s view about and the use of test results. Item Analysis for its Uniqueness Item analysis is usually applied to norm-referenced tests and sometimes to criterion-referenced test. Professional testing services usually improve the reliability and validity of the test by means of item analysis. We test the test after its development but before its administration. But the test is considered to be no better than the items that go into its composition. Weak items should be identified and removed from the test. The single most important characteristic of an item to be accurately determined is its difficulty. Tests that are too difficult or too easy for a given group of examinees often show low reliability. Item difficulty is determined with the following formula: Item Difficulty p =∑Cr / N p = difficulty, proportion correct ∑Cr = the sum of correct responses N = the number of examinees What is meant by q = 1- p Notice The formula shows: the higher the difficulty, the lower the proportion correct and the higher the proportion incorrect. Thus, the difficulty index is inversely related to the actual difficulty of any given item. Question What is the appropriate item difficulty for a given set of examinees? Generally about the mid-point of the difficulty range, i.e. 0.33-0.67. (See p.49) Item Discriminability Difficulty alone is not sufficient information upon which to base the decision ultimately to accept or reject a given item. Another important characteristic of a test item is how well it discriminates between weak and strong examinees in the ability being tested. The necessity of item discriminability lies between the lines in the first paragraph in 4.4: ...a high score would be an indication of inability... The analysis has basically four issues to be addressed at An