<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE ArticleSet PUBLIC "-//NLM//DTD PubMed 2.7//EN" "https://dtd.nlm.nih.gov/ncbi/pubmed/in/PubMed.dtd">
<ArticleSet>
<Article>
<Journal>
				<PublisherName>University of Isfahan</PublisherName>
				<JournalTitle>Applied Research on English Language</JournalTitle>
				<Issn>2252-0198</Issn>
				<Volume>3</Volume>
				<Issue>2</Issue>
				<PubDate PubStatus="epublish">
					<Year>2014</Year>
					<Month>08</Month>
					<Day>01</Day>
				</PubDate>
			</Journal>
<ArticleTitle>A confirmatory study of Differential Item Functioning on EFL reading comprehension</ArticleTitle>
<VernacularTitle></VernacularTitle>
			<FirstPage>55</FirstPage>
			<LastPage>68</LastPage>
			<ELocationID EIdType="pii">15489</ELocationID>
			
<ELocationID EIdType="doi">10.22108/are.2014.15489</ELocationID>
			
			<Language>EN</Language>
<AuthorList>
<Author>
					<FirstName>Alireza</FirstName>
					<LastName>Ahmadi</LastName>
<Affiliation>Shiraz University, Iran</Affiliation>

</Author>
<Author>
					<FirstName>Touraj</FirstName>
					<LastName>Jalili</LastName>
<Affiliation>Shiraz University; PhD Candidate, University of Isfahan, Iran</Affiliation>

</Author>
</AuthorList>
				<PublicationType>Journal Article</PublicationType>
			<History>
				<PubDate PubStatus="received">
					<Year>2014</Year>
					<Month>04</Month>
					<Day>23</Day>
				</PubDate>
			</History>
		<Abstract>The  present  study  aimed  at  investigating  DIF  sources  on  an  EFL  reading  comprehension &lt;br /&gt;test.  Accordingly,  2  DIF  detection  methods,  logistic  regression  (LR)  and  item  response &lt;br /&gt;theory  (IRT),  were  used  to  flag  emergent  DIF  of  203  (110  females  &amp;  93  males)  Iranian &lt;br /&gt;EFL examinees’ performance on a reading comprehension test. Seven hypothetical DIF &lt;br /&gt;sources were examined in this regard: text familiarity, gender, topic/text interest, guessing, &lt;br /&gt;and  the  social  variables  of  location,  income,  and  educational  status.  Only  LR,  for  gender &lt;br /&gt;and  text  familiarity,  could  preempt  DIF  with  gender  supporting  the  gendered-text  effect &lt;br /&gt;while  text  familiarity  benefiting,  inversely,  the  participants  with  low  level  of  text &lt;br /&gt;familiarity.  For  interest  in  topic,  LR  found  a  single  item  favoring  the  group  with  higher &lt;br /&gt;levels  of  interest  and  the  IRT  model  detected  DIF  in  either  extreme.  Regarding  guessing &lt;br /&gt;and  income,  the  LR  indicated  DIF  supporting  the  low  guessers  and  high-income  group &lt;br /&gt;whereas IRT, conversely,  showed DIF  favoring the high guessers and low-income  group. &lt;br /&gt;For  location  and  education  both  methods,  correspondingly,  demonstrated  DIF  for  the &lt;br /&gt;expensive  location  and  educated  groups.  Finally,  the  differential  test  functioning  result &lt;br /&gt;made it clear that only three sources of DIF (gender, income, &amp; interests) were transferred &lt;br /&gt;to the test level. The findings could support a proportional effect of DIF sources.</Abstract>
		<ObjectList>
			<Object Type="keyword">
			<Param Name="value">DIF</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">LR</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">IRT</Param>
			</Object>
			<Object Type="keyword">
			<Param Name="value">Reading Comprehension</Param>
			</Object>
		</ObjectList>
<ArchiveCopySource DocType="pdf">https://are.ui.ac.ir/article_15489_9e33df94cd1abc40c3dc9bcb612ad100.pdf</ArchiveCopySource>
</Article>
</ArticleSet>
