Psychometric Effects of Altering the Ceiling Criterion on the Passage Comprehension Test of the Woodcock-Johnson Psycho-Educational Batftery-Revised

Abstract

The Passage Comprehension Test of the Woodcock-Johnson Psycho-Educational Battery-Revised (WJ-R) was administered to 77 elementary and middle school students and was scored according to ceiling criteria contained in the test manual (i.e., six consecutive errors). The protocols were then rescored using the following ceiling criteria: (a) five consecutive errors, (b) five errors in six consecutive responses, and (c) five errors in seven consecutive responses. There was a strong relationship between raw scores obtained using all four scoring methods, and there were consistent correlations between each group of raw scores and an external measure, full-scale intelligence quotient. The investigators found that relaxing the ceiling criterion reduced, to a statistically significant degree, the number of administered items needed to establish a ceiling without negatively affecting the psychometric properties of the test. Regression equations were presented so that raw scores obtained using the alternate scoring methods could be used to access the WJ-R norm tables. Practical implications and limitations of the results were discussed.