Statistical Procedures for Identifying Possible Sources of Item Bias Based on χ 2 Statistics

Six statistical methods for identifying possible sources of bias in standardized test items are presented. For each method, the statistical null and alternative hypotheses are given and examined in terms of factors relevant to item bias. Issues relevant to and rationale for the statistical treatment...

Full description

Saved in:
Bibliographic Details
Published in:Journal of educational measurement 1981-12, Vol.18 (4), p.229-248
Main Authors: Marascuilo, Leonard A., Slaughter, Robert E.
Format: Article
Language:English
Subjects:
Applied statistics
Ethnic groups
Parallel lines
Post hoc
Probabilities
Statistical bias
Statistics
Test bias
Test scores
Unbiased estimators
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Six statistical methods for identifying possible sources of bias in standardized test items are presented. For each method, the statistical null and alternative hypotheses are given and examined in terms of factors relevant to item bias. Issues relevant to and rationale for the statistical treatment of the bias problem are discussed. The authors also provide a discussion of the relationship between χ 2 methods and item response theory methods. Following the presentation of the six methods, the discussion provides a summary and several recommendations for the use of the techniques in concert.
ISSN:0022-0655
1745-3984
DOI:10.1111/j.1745-3984.1981.tb00856.x