Files in this item



application/pdf8409947.pdf (4MB)Restricted to U of Illinois
(no description provided)PDF


Title:The Relationship of Passage Independent and Passage Dependent Test Items to Each Other and to Cultural Item Bias
Author(s):Hastings, Charles Nicholas
Department / Program:Education
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Education, Tests and Measurements
Abstract:A strategy was presented for investigating the contribution of pre-specified standardized achievement test item characteristics to culturally related item bias. The relative passage or context independence of certain reading comprehension items was hypothesized to be a potential source of such bias in an example test, the Science Research Associates (SRA, Form E, Blue Level, 1971 Edition). Passage free data were collected from 71 fifth grade students, and were used in conjunction with passage present data obtained from the ANCHOR Test Study (Bianchini and Loret, 1974) to rank order the test's 60 items according to three major context independence indices, Pyrczak's I, Tuinman's PDI, and Hanna-Oaster's CDI. A fourth index was also developed based on maximum likelihood theory, the MLI, and used to rank order the items. All indices tended to give very similar rank orderings, thus, the simplest of the four, I, was used to order items for most of the major analyses. ANCHOR data were used for both classical correlational analyses and LISREL confirmatory factor analyses to determine the relationship between subtests composed of relatively passage independent versus passage dependent items. Analyses performed on several variations of the subtests, and for several race-by-sex defined subgroups of students all indicated that the two types of items were highly related. Thus, it was concluded that the relative passage independence of certain test items was not likely to be a primary cause of culturally related items bias, at least within the SRA. However, the study did provide support for the viability of the general strategy as a means of investigating specific item characteristics as potential causes of culturally related item bias.
Issue Date:1983
Description:120 p.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1983.
Other Identifier(s):(UMI)AAI8409947
Date Available in IDEALS:2014-12-15
Date Deposited:1983

This item appears in the following Collection(s)

Item Statistics