Differential Item functioning analysis of the Preschool Language Scale-4 between English-speaking Hispanic and European American children from low-income families

Cathy Huaqing Qi, Scott Marley

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

The study examined whether item bias is present in the Preschool Language Scale-4 (PLS-4). Participants were 440 children (3-5 years old; 86% English-speaking Hispanic and 14% European American) who were enrolled in Head Start programs. The PLS-4 items were analyzed for differential item functioning (DIF) using logistic regression and a one-parameter item response theory (IRT) model. The logistic regression analysis identified Auditory Comprehension (AC) Item 35 as having non-uniform DIF. Using IRT, AC Item 55 and Expressive Communication (EC) Items 30 and 31 displayed statistical evidence of DIF. AC Item 55 and EC Item 30 were more difficult for Hispanic children than for European American children, whereas the opposite was true for EC Item 31. These findings did not find strong evidence of cultural bias for the PLS-4 against English-speaking Hispanic children from low-income families.

Original languageEnglish (US)
Pages (from-to)171-180
Number of pages10
JournalTopics in Early Childhood Special Education
Volume29
Issue number3
DOIs
StatePublished - Nov 2009
Externally publishedYes

Fingerprint

item analysis
Hispanic Americans
speaking
low income
Language
comprehension
Communication
language
communication
Logistic Models
logistics
model theory
trend
evidence
regression analysis
Regression Analysis
regression

Keywords

  • Hispanic
  • Item bias
  • Low income
  • PLS-4
  • Preschool Language Scale-4
  • Validity

ASJC Scopus subject areas

  • Education
  • Public Health, Environmental and Occupational Health

Cite this

@article{5e0fb58c95cb4604bb88cd5279eabb79,
title = "Differential Item functioning analysis of the Preschool Language Scale-4 between English-speaking Hispanic and European American children from low-income families",
abstract = "The study examined whether item bias is present in the Preschool Language Scale-4 (PLS-4). Participants were 440 children (3-5 years old; 86{\%} English-speaking Hispanic and 14{\%} European American) who were enrolled in Head Start programs. The PLS-4 items were analyzed for differential item functioning (DIF) using logistic regression and a one-parameter item response theory (IRT) model. The logistic regression analysis identified Auditory Comprehension (AC) Item 35 as having non-uniform DIF. Using IRT, AC Item 55 and Expressive Communication (EC) Items 30 and 31 displayed statistical evidence of DIF. AC Item 55 and EC Item 30 were more difficult for Hispanic children than for European American children, whereas the opposite was true for EC Item 31. These findings did not find strong evidence of cultural bias for the PLS-4 against English-speaking Hispanic children from low-income families.",
keywords = "Hispanic, Item bias, Low income, PLS-4, Preschool Language Scale-4, Validity",
author = "Qi, {Cathy Huaqing} and Scott Marley",
year = "2009",
month = "11",
doi = "10.1177/0271121409332674",
language = "English (US)",
volume = "29",
pages = "171--180",
journal = "Topics in Early Childhood Special Education",
issn = "0271-1214",
publisher = "SAGE Publications Inc.",
number = "3",

}

TY - JOUR

T1 - Differential Item functioning analysis of the Preschool Language Scale-4 between English-speaking Hispanic and European American children from low-income families

AU - Qi, Cathy Huaqing

AU - Marley, Scott

PY - 2009/11

Y1 - 2009/11

N2 - The study examined whether item bias is present in the Preschool Language Scale-4 (PLS-4). Participants were 440 children (3-5 years old; 86% English-speaking Hispanic and 14% European American) who were enrolled in Head Start programs. The PLS-4 items were analyzed for differential item functioning (DIF) using logistic regression and a one-parameter item response theory (IRT) model. The logistic regression analysis identified Auditory Comprehension (AC) Item 35 as having non-uniform DIF. Using IRT, AC Item 55 and Expressive Communication (EC) Items 30 and 31 displayed statistical evidence of DIF. AC Item 55 and EC Item 30 were more difficult for Hispanic children than for European American children, whereas the opposite was true for EC Item 31. These findings did not find strong evidence of cultural bias for the PLS-4 against English-speaking Hispanic children from low-income families.

AB - The study examined whether item bias is present in the Preschool Language Scale-4 (PLS-4). Participants were 440 children (3-5 years old; 86% English-speaking Hispanic and 14% European American) who were enrolled in Head Start programs. The PLS-4 items were analyzed for differential item functioning (DIF) using logistic regression and a one-parameter item response theory (IRT) model. The logistic regression analysis identified Auditory Comprehension (AC) Item 35 as having non-uniform DIF. Using IRT, AC Item 55 and Expressive Communication (EC) Items 30 and 31 displayed statistical evidence of DIF. AC Item 55 and EC Item 30 were more difficult for Hispanic children than for European American children, whereas the opposite was true for EC Item 31. These findings did not find strong evidence of cultural bias for the PLS-4 against English-speaking Hispanic children from low-income families.

KW - Hispanic

KW - Item bias

KW - Low income

KW - PLS-4

KW - Preschool Language Scale-4

KW - Validity

UR - http://www.scopus.com/inward/record.url?scp=70350543651&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350543651&partnerID=8YFLogxK

U2 - 10.1177/0271121409332674

DO - 10.1177/0271121409332674

M3 - Article

AN - SCOPUS:70350543651

VL - 29

SP - 171

EP - 180

JO - Topics in Early Childhood Special Education

JF - Topics in Early Childhood Special Education

SN - 0271-1214

IS - 3

ER -