Examining sources of gender DIF : a confirmatory approach

A confirmatory approach based on a multidimensional model (Douglas, Roussos, & Stout, 1996; Shealy & Stout, 1993; Stout & Roussos, 1995) was used to identify sources of differential item functioning (DIF) and differential bundle functioning (DBF) for boys and girls on the British Colu...

Full description

Bibliographic Details
Main Author: Barnett, Sharon
Format: Others
Language:English
Published: 2009
Subjects:
Online Access:http://hdl.handle.net/2429/11965
id ndltd-UBC-oai-circle.library.ubc.ca-2429-11965
record_format oai_dc
spelling ndltd-UBC-oai-circle.library.ubc.ca-2429-119652018-01-05T17:36:11Z Examining sources of gender DIF : a confirmatory approach Barnett, Sharon Mathematics -- Sex differences. Mathematical ability -- Testing. A confirmatory approach based on a multidimensional model (Douglas, Roussos, & Stout, 1996; Shealy & Stout, 1993; Stout & Roussos, 1995) was used to identify sources of differential item functioning (DIF) and differential bundle functioning (DBF) for boys and girls on the British Columbia Principles of Mathematics Exam for grade 12 (PME12). Data consisted of a total of 9404 examinees; 4335 girls and 5069 boys. There were 45 multiple choice items in the exam. Analyses were completed in two stages. In stage 1, patterns present in the gender DIF research in mathematics were identified. Stage 2 was the statistical confirmation of these patterns. Sources of gender DIF were confirmed for the content areas: polynomial, quadratic relations, logarithms and exponents. Items tapping higher cognitive levels dealing with patterns and relation, word problems, and items containing visuals were also confirmed as a source of DIF. Exploratory analyses indicated that computation items for which no equations are provided may be a source of DIF along with trigonometry items. This study contributes to an increased understanding of sources of gender DIF that may assist test developers to ensure that mathematics items measure the construct that they are intended to measure and that the test as a whole measures that which it purports to measure. The findings of this research provide an additional source of information about the differential performance of boys and girls that may be used to develop guidelines and test construction principles for reducing gender DIF in mathematics. This research also contributes to a greater understanding of gender differences in mathematics learning and achievement. Education, Faculty of Educational and Counselling Psychology, and Special Education (ECPS), Department of Graduate 2009-08-06T23:46:30Z 2009-08-06T23:46:30Z 2001 2002-05 Text Thesis/Dissertation http://hdl.handle.net/2429/11965 eng For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use. 3565582 bytes application/pdf
collection NDLTD
language English
format Others
sources NDLTD
topic Mathematics -- Sex differences.
Mathematical ability -- Testing.
spellingShingle Mathematics -- Sex differences.
Mathematical ability -- Testing.
Barnett, Sharon
Examining sources of gender DIF : a confirmatory approach
description A confirmatory approach based on a multidimensional model (Douglas, Roussos, & Stout, 1996; Shealy & Stout, 1993; Stout & Roussos, 1995) was used to identify sources of differential item functioning (DIF) and differential bundle functioning (DBF) for boys and girls on the British Columbia Principles of Mathematics Exam for grade 12 (PME12). Data consisted of a total of 9404 examinees; 4335 girls and 5069 boys. There were 45 multiple choice items in the exam. Analyses were completed in two stages. In stage 1, patterns present in the gender DIF research in mathematics were identified. Stage 2 was the statistical confirmation of these patterns. Sources of gender DIF were confirmed for the content areas: polynomial, quadratic relations, logarithms and exponents. Items tapping higher cognitive levels dealing with patterns and relation, word problems, and items containing visuals were also confirmed as a source of DIF. Exploratory analyses indicated that computation items for which no equations are provided may be a source of DIF along with trigonometry items. This study contributes to an increased understanding of sources of gender DIF that may assist test developers to ensure that mathematics items measure the construct that they are intended to measure and that the test as a whole measures that which it purports to measure. The findings of this research provide an additional source of information about the differential performance of boys and girls that may be used to develop guidelines and test construction principles for reducing gender DIF in mathematics. This research also contributes to a greater understanding of gender differences in mathematics learning and achievement. === Education, Faculty of === Educational and Counselling Psychology, and Special Education (ECPS), Department of === Graduate
author Barnett, Sharon
author_facet Barnett, Sharon
author_sort Barnett, Sharon
title Examining sources of gender DIF : a confirmatory approach
title_short Examining sources of gender DIF : a confirmatory approach
title_full Examining sources of gender DIF : a confirmatory approach
title_fullStr Examining sources of gender DIF : a confirmatory approach
title_full_unstemmed Examining sources of gender DIF : a confirmatory approach
title_sort examining sources of gender dif : a confirmatory approach
publishDate 2009
url http://hdl.handle.net/2429/11965
work_keys_str_mv AT barnettsharon examiningsourcesofgenderdifaconfirmatoryapproach
_version_ 1718589019601764352