An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)
Identifieur interne : 000426 ( Istex/Corpus ); précédent : 000425; suivant : 000427An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)
Auteurs : Barbara B. Ellis ; Peter Becker ; Herbert D. KimmelSource :
- Journal of Cross-Cultural Psychology [ 0022-0221 ] ; 1993-06.
Abstract
The measurement equivalence of an English-language version of the Trier Personality Inventory (TPI) was evaluated, using statistical methods based on item response theory (IRT) to identity items that displayed differential item functioning (DIF). In Study 1, the TPI was administered to 295 American and 213 West German subjects. From among the 120 items tested, 11 DIF items were detected. A follow-up study using a new sample of 203 Americans was conducted to replicate the original study and to evaluate the post hoc explanation that translation error was the source of DIF for one item. DIF was replicated for 6 of the 11 DIF items found in Study 1, and DIF was eliminated for one item by retranslation. Overall, there was significant agreement between the DIF indexes found in Studies I and 2 for all 120 items analyzed. Some problems in replicating significant indexes of DIF are discussed.
Url:
DOI: 10.1177/0022022193242001
Links to Exploration step
ISTEX:9A524158A96265D324ED0B68013F920B0449BE4CLe document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)</title><author wicri:is="90%"><name sortKey="Ellis, Barbara B" sort="Ellis, Barbara B" uniqKey="Ellis B" first="Barbara B." last="Ellis">Barbara B. Ellis</name><affiliation><mods:affiliation>University of Texas at El Paso</mods:affiliation></affiliation></author><author wicri:is="90%"><name sortKey="Becker, Peter" sort="Becker, Peter" uniqKey="Becker P" first="Peter" last="Becker">Peter Becker</name><affiliation><mods:affiliation>Universitait Trier, Germany</mods:affiliation></affiliation></author><author wicri:is="90%"><name sortKey="Kimmel, Herbert D" sort="Kimmel, Herbert D" uniqKey="Kimmel H" first="Herbert D." last="Kimmel">Herbert D. Kimmel</name><affiliation><mods:affiliation>University of South Florida</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:9A524158A96265D324ED0B68013F920B0449BE4C</idno><date when="1993" year="1993">1993</date><idno type="doi">10.1177/0022022193242001</idno><idno type="url">https://api.istex.fr/document/9A524158A96265D324ED0B68013F920B0449BE4C/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000426</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">000426</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)</title><author wicri:is="90%"><name sortKey="Ellis, Barbara B" sort="Ellis, Barbara B" uniqKey="Ellis B" first="Barbara B." last="Ellis">Barbara B. Ellis</name><affiliation><mods:affiliation>University of Texas at El Paso</mods:affiliation></affiliation></author><author wicri:is="90%"><name sortKey="Becker, Peter" sort="Becker, Peter" uniqKey="Becker P" first="Peter" last="Becker">Peter Becker</name><affiliation><mods:affiliation>Universitait Trier, Germany</mods:affiliation></affiliation></author><author wicri:is="90%"><name sortKey="Kimmel, Herbert D" sort="Kimmel, Herbert D" uniqKey="Kimmel H" first="Herbert D." last="Kimmel">Herbert D. Kimmel</name><affiliation><mods:affiliation>University of South Florida</mods:affiliation></affiliation></author></analytic><monogr></monogr><series><title level="j">Journal of Cross-Cultural Psychology</title><idno type="ISSN">0022-0221</idno><idno type="eISSN">1552-5422</idno><imprint><publisher>Sage Publications</publisher><pubPlace>Sage CA: Thousand Oaks, CA</pubPlace><date type="published" when="1993-06">1993-06</date><biblScope unit="volume">24</biblScope><biblScope unit="issue">2</biblScope><biblScope unit="page" from="133">133</biblScope><biblScope unit="page" to="148">148</biblScope></imprint><idno type="ISSN">0022-0221</idno></series><idno type="istex">9A524158A96265D324ED0B68013F920B0449BE4C</idno><idno type="DOI">10.1177/0022022193242001</idno><idno type="ArticleID">10.1177_0022022193242001</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0022-0221</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The measurement equivalence of an English-language version of the Trier Personality Inventory (TPI) was evaluated, using statistical methods based on item response theory (IRT) to identity items that displayed differential item functioning (DIF). In Study 1, the TPI was administered to 295 American and 213 West German subjects. From among the 120 items tested, 11 DIF items were detected. A follow-up study using a new sample of 203 Americans was conducted to replicate the original study and to evaluate the post hoc explanation that translation error was the source of DIF for one item. DIF was replicated for 6 of the 11 DIF items found in Study 1, and DIF was eliminated for one item by retranslation. Overall, there was significant agreement between the DIF indexes found in Studies I and 2 for all 120 items analyzed. Some problems in replicating significant indexes of DIF are discussed.</div></front></TEI><istex><corpusName>sage</corpusName><author><json:item><name>Barbara B. Ellis</name><affiliations><json:string>University of Texas at El Paso</json:string></affiliations></json:item><json:item><name>Peter Becker</name><affiliations><json:string>Universitait Trier, Germany</json:string></affiliations></json:item><json:item><name>Herbert D. Kimmel</name><affiliations><json:string>University of South Florida</json:string></affiliations></json:item></author><articleId><json:string>10.1177_0022022193242001</json:string></articleId><language><json:string>eng</json:string></language><originalGenre><json:string>research-article</json:string></originalGenre><abstract>The measurement equivalence of an English-language version of the Trier Personality Inventory (TPI) was evaluated, using statistical methods based on item response theory (IRT) to identity items that displayed differential item functioning (DIF). In Study 1, the TPI was administered to 295 American and 213 West German subjects. From among the 120 items tested, 11 DIF items were detected. A follow-up study using a new sample of 203 Americans was conducted to replicate the original study and to evaluate the post hoc explanation that translation error was the source of DIF for one item. DIF was replicated for 6 of the 11 DIF items found in Study 1, and DIF was eliminated for one item by retranslation. Overall, there was significant agreement between the DIF indexes found in Studies I and 2 for all 120 items analyzed. Some problems in replicating significant indexes of DIF are discussed.</abstract><qualityIndicators><score>1.773</score><pdfVersion>1.3</pdfVersion><pdfPageSize>396 x 612 pts</pdfPageSize><refBibsNative>true</refBibsNative><abstractCharCount>896</abstractCharCount><pdfWordCount>9</pdfWordCount><pdfCharCount>54</pdfCharCount><pdfPageCount>16</pdfPageCount><abstractWordCount>147</abstractWordCount></qualityIndicators><title>An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)</title><refBibs><json:item><author><json:item><name>P. Becker</name></json:item></author><host><volume>9</volume><pages><last>38</last><first>13</first></pages><author></author><title>Zeitschrnft für Differentielle und Diagnostische Psychologie</title></host><title>Seelische Gesundheit und Verhaltenskontrolle: Zwei replizierbare, vananzstarke Persönlichkeitsfaktoren [Mental health and behavior control: TWo replicable, primary personality factors]</title></json:item><json:item><host><author></author><title>Der Trierer Persönlichkeitsfragebogen (TPF) Handanweisung [The Trier Personality Inventory (TPI) Manual]</title></host></json:item><json:item><author><json:item><name>A. Birnbaum</name></json:item></author><host><pages><last>497</last><first>397</first></pages><author></author><title>Statistical theories of mental test scores</title></host><title>Some latent trait models and their use in inferring examinee's ability</title></json:item><json:item><author><json:item><name>G. L. Candell</name></json:item><json:item><name>C. L. Hulin</name></json:item></author><host><volume>17</volume><pages><last>440</last><first>417</first></pages><author></author><title>Journal of Cross-Cultural Psychology</title></host><title>Cross-language and cross-cultural comparisons in scale translations: Independent sources of information about item nonequivalence</title></json:item><json:item><author><json:item><name>R. B. Cattell</name></json:item></author><host><volume>1</volume><pages><last>276</last><first>245</first></pages><author></author><title>Multivariate Behavioral Research</title></host><title>The scree test for the number of factors</title></json:item><json:item><author><json:item><name>F. Drasgow</name></json:item></author><host><volume>72</volume><pages><last>29</last><first>19</first></pages><author></author><title>Journal of Applied Psychology</title></host><title>Study of the measurement bias of two standardized psychological tests</title></json:item><json:item><author><json:item><name>F. Drasgow</name></json:item></author><host><volume>13</volume><pages><last>90</last><first>77</first></pages><author></author><title>Applied Psychological Measurement</title></host><title>An evaluation of marginal maximum likelihood estimation for the two-paramneter model</title></json:item><json:item><author><json:item><name>B. B. Ellis</name></json:item><json:item><name> </name></json:item><json:item><name> </name></json:item></author><host><author></author><title>Ethnic psychology: Research and practice</title></host><title>Hofstede's culture dimensions and Rokeach's values: How reliable is the relationship?</title></json:item><json:item><author><json:item><name>F. Drasgow</name></json:item></author><host><volume>13</volume><pages><last>90</last><first>77</first></pages><author></author><title>Applied Psychological Measurement</title></host><title>An evaluation of marginal maximum likelihood estimation for the two-parameter model</title></json:item><json:item><author><json:item><name>B. B. Ellis</name></json:item><json:item><name> </name></json:item><json:item><name> </name></json:item></author><host><pages><last>274</last><first>266</first></pages><author></author><title>Ethnic psychology: Research and practice with immigrants, refugees, native peoples, ethnic groups, and sojourners</title></host><title>Hofstede's culture dimensions and Rokeach's values: How reliable is the relationship?</title></json:item><json:item><author><json:item><name>B. B. Ellis</name></json:item></author><host><volume>74</volume><pages><last>921</last><first>912</first></pages><author></author><title>Journal of Applied Psychology</title></host><title>Differential item functioning: Implications for test translations</title></json:item><json:item><author><json:item><name>B. B. Ellis</name></json:item><json:item><name>B. Minsel</name></json:item><json:item><name>P. Becker</name></json:item></author><host><volume>24</volume><pages><last>684</last><first>665</first></pages><author></author><title>International Journal of Psychology</title></host><title>Evaluation of attitude survey translations: An investigation using item response theory</title></json:item><json:item><host><author></author><title>Manual of the Eysenck Personality Questionnaire</title></host></json:item><json:item><host><author></author><title>Das Freiburger Persönlichkeitsinventar FPI Revidierte Fassung FPI-R und teitweise geanderte Fassung FPI-AI. Handanweisung. 4. revidierte Auflage rlThe Freiburger Personality Inventory, FPI, revised version, FPI-R, and partially modified, FPI-Al. Manual. 4th edition]</title></host></json:item><json:item><host><author></author><title>MinnesotaMultiphasic Personality Inventory: MMPI Deutsche Kurtzform für Handauswertung. Handbuch (Minesota Multiphasic Personality Inventory: MMPI. German short form for hand scoring. Handbook]</title></host></json:item><json:item><author><json:item><name>H. Goldstein</name></json:item><json:item><name>R. Wood</name></json:item></author><host><volume>42</volume><pages><last>167</last><first>139</first></pages><author></author><title>British Journal of Mathematical and Statistical Psychology</title></host><title>Five decades of item response modeling</title></json:item><json:item><author><json:item><name>R. K. Hambleton</name></json:item></author><host><volume>29</volume><pages><last>106</last><first>96</first></pages><author></author><title>Buletin of the International Test Commission</title></host><title>Constructing tests with item response models: A discussion of a method and two problems</title></json:item><json:item><host><author></author><title>Item response theory: Principles and applications</title></host></json:item><json:item><host><author></author><title>Differential item performance and the Mantel-Haenszel procedure</title></host></json:item><json:item><author><json:item><name>P. W. Holland</name></json:item><json:item><name>D. T. Thayer</name></json:item><json:item><name> </name></json:item><json:item><name> </name></json:item></author><host><pages><last>145</last><first>129</first></pages><author></author><title>Test validity</title></host><title>Differential item performance and the Mantel-Haenszel procedure</title></json:item><json:item><author><json:item><name>C. L. Hulin</name></json:item></author><host><volume>18</volume><pages><last>142</last><first>115</first></pages><author></author><title>Journal of Cross-Cultural Psychology</title></host><title>A psychometric theory of evaluations of item and scale translations: Fidelity across languages</title></json:item><json:item><author><json:item><name>C. L. Hulin</name></json:item><json:item><name>F. Drasgow</name></json:item><json:item><name>J. Komocar</name></json:item></author><host><volume>67</volume><pages><last>825</last><first>818</first></pages><author></author><title>Journal of Applied Psychology</title></host><title>Applications of item response theory to analysis of scale translations</title></json:item><json:item><host><author></author><title>Item response theory: Applications to psychological measurement</title></host></json:item><json:item><author><json:item><name>C. L. Hulin</name></json:item><json:item><name>L. J. Mayer</name></json:item></author><host><volume>71</volume><pages><last>94</last><first>83</first></pages><author></author><title>Journal of Applied Psychology</title></host><title>Psychometric equivalence of a translation of the Job Descriptive Index into Hebrew</title></json:item><json:item><author><json:item><name>R. G. Lim</name></json:item><json:item><name>F. Drasgow</name></json:item></author><host><volume>75</volume><pages><last>174</last><first>164</first></pages><author></author><title>Journal of Applied Psychology</title></host><title>Evaluation of two methods for estimating item response theory parameters when assessing differential item functioning</title></json:item><json:item><host><author></author><title>Applications of item response theory to practical testing problems</title></host></json:item><json:item><host><author></author><title>Statistical theories of mental test scores</title></host></json:item><json:item><author><json:item><name>M. E. McLaughlin</name></json:item><json:item><name>F. Drasgow</name></json:item></author><host><volume>11</volume><pages><last>173</last><first>161</first></pages><author></author><title>Applied Psychological Measurement</title></host><title>Lord's chi-square test of item bias with estimated and with known person parameters</title></json:item><json:item><author><json:item><name>M. L. Stocking</name></json:item><json:item><name>F. M. Lord</name></json:item></author><host><volume>7</volume><pages><last>210</last><first>201</first></pages><author></author><title>Applied Psychological Measurement</title></host><title>Developing a common metric in item response theory</title></json:item><json:item><host><author></author><title>Deutsche Personality Research Form (PRF)</title></host></json:item><json:item><host><author></author><title>MULTILOG user's guide</title></host></json:item><json:item><author><json:item><name>D. Thissen</name></json:item><json:item><name>L. Steinberg</name></json:item></author><host><volume>104</volume><pages><last>395</last><first>385</first></pages><author></author><title>Psychological Bulletin</title></host><title>Data analysis using item response theory</title></json:item><json:item><author><json:item><name>D. Thissen</name></json:item><json:item><name>L. Steinberg</name></json:item><json:item><name>M. Gerrard</name></json:item></author><host><volume>99</volume><pages><last>181</last><first>118</first></pages><author></author><title>Psychological Bulletin</title></host><title>Beyond group mean differences: The concept of item bias</title></json:item><json:item><author><json:item><name>D. Thissen</name></json:item><json:item><name>L. Steinberg</name></json:item><json:item><name>H. Wainer</name></json:item><json:item><name> </name></json:item><json:item><name> </name></json:item></author><host><pages><last>170</last><first>147</first></pages><author></author><title>Test validity</title></host><title>Use of item response theory in the study of group difference in trace lines</title></json:item><json:item><host><author></author><title>Detection of differential item functioning using the parameters of item response models</title></host></json:item></refBibs><genre><json:string>research-article</json:string></genre><host><volume>24</volume><publisherId><json:string>JCC</json:string></publisherId><pages><last>148</last><first>133</first></pages><issn><json:string>0022-0221</json:string></issn><issue>2</issue><genre><json:string>journal</json:string></genre><language><json:string>unknown</json:string></language><eissn><json:string>1552-5422</json:string></eissn><title>Journal of Cross-Cultural Psychology</title></host><categories><wos><json:string>social science</json:string><json:string>psychology, social</json:string></wos><scienceMetrix><json:string>health sciences</json:string><json:string>psychology & cognitive sciences</json:string><json:string>social psychology</json:string></scienceMetrix></categories><publicationDate>1993</publicationDate><copyrightDate>1993</copyrightDate><doi><json:string>10.1177/0022022193242001</json:string></doi><id>9A524158A96265D324ED0B68013F920B0449BE4C</id><score>0.5296517</score><fulltext><json:item><extension>pdf</extension><original>true</original><mimetype>application/pdf</mimetype><uri>https://api.istex.fr/document/9A524158A96265D324ED0B68013F920B0449BE4C/fulltext/pdf</uri></json:item><json:item><extension>zip</extension><original>false</original><mimetype>application/zip</mimetype><uri>https://api.istex.fr/document/9A524158A96265D324ED0B68013F920B0449BE4C/fulltext/zip</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/9A524158A96265D324ED0B68013F920B0449BE4C/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a" type="main" xml:lang="en">An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)</title></titleStmt><publicationStmt><authority>ISTEX</authority><publisher>Sage Publications</publisher><pubPlace>Sage CA: Thousand Oaks, CA</pubPlace><availability><p>SAGE</p></availability><date>1993</date></publicationStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a" type="main" xml:lang="en">An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)</title><author xml:id="author-1"><persName><forename type="first">Barbara B.</forename><surname>Ellis</surname></persName><affiliation>University of Texas at El Paso</affiliation></author><author xml:id="author-2"><persName><forename type="first">Peter</forename><surname>Becker</surname></persName><affiliation>Universitait Trier, Germany</affiliation></author><author xml:id="author-3"><persName><forename type="first">Herbert D.</forename><surname>Kimmel</surname></persName><affiliation>University of South Florida</affiliation></author></analytic><monogr><title level="j">Journal of Cross-Cultural Psychology</title><idno type="pISSN">0022-0221</idno><idno type="eISSN">1552-5422</idno><imprint><publisher>Sage Publications</publisher><pubPlace>Sage CA: Thousand Oaks, CA</pubPlace><date type="published" when="1993-06"></date><biblScope unit="volume">24</biblScope><biblScope unit="issue">2</biblScope><biblScope unit="page" from="133">133</biblScope><biblScope unit="page" to="148">148</biblScope></imprint></monogr><idno type="istex">9A524158A96265D324ED0B68013F920B0449BE4C</idno><idno type="DOI">10.1177/0022022193242001</idno><idno type="ArticleID">10.1177_0022022193242001</idno></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>1993</date></creation><langUsage><language ident="en">en</language></langUsage><abstract xml:lang="en"><p>The measurement equivalence of an English-language version of the Trier Personality Inventory (TPI) was evaluated, using statistical methods based on item response theory (IRT) to identity items that displayed differential item functioning (DIF). In Study 1, the TPI was administered to 295 American and 213 West German subjects. From among the 120 items tested, 11 DIF items were detected. A follow-up study using a new sample of 203 Americans was conducted to replicate the original study and to evaluate the post hoc explanation that translation error was the source of DIF for one item. DIF was replicated for 6 of the 11 DIF items found in Study 1, and DIF was eliminated for one item by retranslation. Overall, there was significant agreement between the DIF indexes found in Studies I and 2 for all 120 items analyzed. Some problems in replicating significant indexes of DIF are discussed.</p></abstract></profileDesc><revisionDesc><change when="1993-06">Published</change></revisionDesc></teiHeader></istex:fulltextTEI><json:item><extension>txt</extension><original>false</original><mimetype>text/plain</mimetype><uri>https://api.istex.fr/document/9A524158A96265D324ED0B68013F920B0449BE4C/fulltext/txt</uri></json:item></fulltext><metadata><istex:metadataXml wicri:clean="corpus sage not found" wicri:toSee="no header"><istex:xmlDeclaration>version="1.0" encoding="UTF-8"</istex:xmlDeclaration><istex:docType PUBLIC="-//NLM//DTD Journal Publishing DTD v2.3 20070202//EN" URI="journalpublishing.dtd" name="istex:docType"></istex:docType><istex:document><article article-type="research-article" dtd-version="2.3" xml:lang="EN"><front><journal-meta><journal-id journal-id-type="hwp">spjcc</journal-id><journal-id journal-id-type="publisher-id">JCC</journal-id><journal-title>Journal of Cross-Cultural Psychology</journal-title><issn pub-type="ppub">0022-0221</issn><publisher><publisher-name>Sage Publications</publisher-name><publisher-loc>Sage CA: Thousand Oaks, CA</publisher-loc></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.1177/0022022193242001</article-id><article-id pub-id-type="publisher-id">10.1177_0022022193242001</article-id><article-categories><subj-group subj-group-type="heading"><subject>Articles</subject></subj-group></article-categories><title-group><article-title>An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)</article-title></title-group><contrib-group><contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Ellis</surname><given-names>Barbara B.</given-names></name><aff>University of Texas at El Paso</aff></contrib></contrib-group><contrib-group><contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Becker</surname><given-names>Peter</given-names></name><aff>Universitait Trier, Germany</aff></contrib></contrib-group><contrib-group><contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Kimmel</surname><given-names>Herbert D.</given-names></name><aff>University of South Florida</aff></contrib></contrib-group><pub-date pub-type="ppub"><month>06</month><year>1993</year></pub-date><volume>24</volume><issue>2</issue><fpage>133</fpage><lpage>148</lpage><abstract><p>The measurement equivalence of an English-language version of the Trier Personality Inventory (TPI) was evaluated, using statistical methods based on item response theory (IRT) to identity items that displayed differential item functioning (DIF). In Study 1, the TPI was administered to 295 American and 213 West German subjects. From among the 120 items tested, 11 DIF items were detected. A follow-up study using a new sample of 203 Americans was conducted to replicate the original study and to evaluate the post hoc explanation that translation error was the source of DIF for one item. DIF was replicated for 6 of the 11 DIF items found in Study 1, and DIF was eliminated for one item by retranslation. Overall, there was significant agreement between the DIF indexes found in Studies I and 2 for all 120 items analyzed. Some problems in replicating significant indexes of DIF are discussed.</p></abstract><custom-meta-wrap><custom-meta xlink:type="simple"><meta-name>sagemeta-type</meta-name><meta-value>Journal Article</meta-value></custom-meta><custom-meta xlink:type="simple"><meta-name>search-text</meta-name><meta-value> The measurement equivalence of an English-language version of the Trier Personality Inventory (TPI) was evaluated, using statistical methods based on item response theory (IRT) to identity items that displayed differential item functioning (DIF). In Study 1, the TPI was administered to 295 American and 213 West German subjects. From among the 120 items tested, 11 DIF items were detected. A follow-up study using a new sample of 203 Americans was conducted to replicate the original study and to evaluate the post hoc explanation that translation error was the source of DIF for one item. DIF was replicated for 6 of the 11 DIF items found in Study 1, and DIF was eliminated for one item by retranslation. Overall, there was significant agreement between the DIF indexes found in Studies I and 2 for all 120 items analyzed. Some problems in replicating significant indexes of DIF are discussed. AN ITEM RESPONSE THEORY EVALUATION OF AN ENGLISH VERSION OF THE TRIER PERSONALITY INVENTORY (TPI) BARBARA B. ELLIS University of Texas at El Paso PETER BECKER Universitait Trier, Germany HERBERT D. KIMMEL University of South Florida It is well-known among test developers that the use of a test in a culture other than the one in which it was developed requires evidence of the test's reliability and validity in the new setting and, usually, the development of new cultural norms. Recent research has shown that it is also necessary to analyze culturally transported tests to identify items that do not function equivalently for both cultural groups (Candell & Hulin, 1987; Drasgow, 1987; Ellis, 1989; Ellis, Minsel, & Becker, 1989; Hulin, 1987; Hulin, Drasgow, & Komocar, 1982; Hulin, Drasgow, & Parson, 1983; Hulin & Mayer, 1986). As defined by Drasgow (1987), a test can be said to possess the property of measurement equivalence "when individuals with equal AUTHORS' NOTE: Correspondence concerning this article should be addressed to Barbara B. Ellis, Department of Psychology, University of Texas at El Paso, El Paso, TX 79968-0553. JOURNAL OF CROSS-CULTURAL PSYCHOLOGY, Vol. 24 No. 2, March 1993 133-148 1993 Western Washington University 133 134 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY standing on the trait measured by the test but sampled from different sub- populations have equal expected observed test scores" (p. 19). The comparison of individuals from different cultural groups requires that measurement instruments be used that are equivalent for the two groups. Statistical methods based on item response theory (IRT; Hambleton & Swaminathan, 1985; Hulin et al., 1983; Lord, 1980; Lord & Novick, 1968) permit the identification of test items that do not function similarly in the two cultures. IRT is composed of mathematical models that describe the relation- ship between an individual's observable response to an item and the individual's standing on the unobservable trait measured by the test. This relationship is represented by an item characteristic curve (ICC) or trace line that depicts the response probability as a function of the trait (3 measured by the test. In general, ICCs assume the form of monotonically increasing functions or s-shaped curves as shown in Figure 1. The mathematical forms of this function vary in terms of the number of parameters used to describe the ICC. One of the more well-researched IRT models is the three-parameter logistic model (Birnbaum, 1968) Pide)=c,+(l-c) {1+exp[-Da,(O-b,)]) (1) The parameters in this model are (a) the a parameter, which is proportional to the slope of the curve at the point of inflection; (b) the b parameter, which is the location on the trait dimension at the point of inflection; and (c) the c parameter, the lower asymptote of the ICC. D is a scaling factor set to 1.7 or 1.702. When the ICCs estimated separately for the same item for two different groups (i.e., the reference and focal groups) are the same, the item functions equivalently for both groups. When the ICCs differ by more than sampling error, the item is said to display dfferential itemfiuctioning (DIF; Hambleton & Swaminathan, 1985; Hulin et al., 1983; Lord, 1980; Thissen, Steinberg, & Wainer, 1988, 1989). Because the value of the ICCs represent the conditional probability of a correct response at each level of ability or proficiency, the ICCs should be the same if the item is measuring the same thing in the reference and focal groups. If the ICCs are not the same, it is an indication that the item is measuring something different or not functioning in the same fashion for the reference and focal groups. In other words, the item displays DIF. By definition, DIF items do not possess the property of measurement equivalence, and, therefore, should not be used to compare the reference and focal groups on the trait measured. It should be noted that, in the past, DIF was referred to as "item bias" (Hambleton & Swaminathan, 1985; Hulin et al., Ellis et al. / TRIER PERSONALITY INVENTORY 0.5 0.0 -3 -2 -1 0 +1 +2 +3 0 Figure 1: Hypothetical Item Characteristic Curve (ICC) 1983; Lord, 1980), but the more recent terminology has gradually replaced the earlier because DIF is more accurate and less evaluative (Holland & Thayer, 1986, 1988; Thissen et al., 1988, 1989). The use of an IRT analysis is particularly important when tests are translated into different languages. Numerous studies have shown that back translation alone is not enough to ensure a high-fidelity translation and that an IRT analysis can be used to assess the effectiveness of the translation itself (Candell & Hulin,1987; Ellis, 1989; Ellis et al., 1989; Hulin 1987; Hulin et al, 1982; Hulin et al., 1983; Hulin & Mayer, 1986). Items that have been inaccurately translated are likely to exhibit DEF. When translated items display DIF, there are three possible sources: (a) translation error, (b) differences in cultural relevance or meaning of the item, or (c) differences in culturally specific knowledge. Any one or all of these may exist. Once translation errors have been identified and corrected, the remaining DIF items may be content analyzed for the purpose of identifying the nature of the cultural differences involved. This additional use of IRT, although introduced only relatively recently (Ellis, 1989; Ellis et al., 1989), may prove to be an effective tool for the study of cultural differences and similarities. Several authors have recently encouraged IRT applications of this sort (Goldstein & Wood, 1989; Thissen & Steinberg, 1988). For example, rather than simply using IRT as a statistical tool for the identification and elimination of DIF items, the goal of a DIF analysis should include gaining 135 136 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY insight into and understanding the psychological processes that underlie the item responses. In any IRT analysis, there is some question about the error associated with the parameter estimates and the indexes of DIF. As recommended by Lord (1980), one way to assess the DIEF analysis itself is to replicate the study using new samples. In this article, we present the results of an IRT analysis of an English-language translation of a personality test originally developed in the West Germany, the Trierer Personlichkeitsfragebogen (TPF) (Becker, 1989), hereafter referred to by its English name, the Trier Personality Inventory (TPI). Translated items that did not function equivalently for Americans and Germans were identified by means of an IRT analysis. A follow-up study replicated the original study and sought to eliminate translation error as the source of DIF in the case of one item. Discrepancies between the findings of the two studies were then analyzed. Finally, the DIF items from both studies were examined to determine possible sources of DIF. STUDY 1 TEST INSTRUMENT The Trier Personality Inventory is a theoretically based measure of several relatively broad personality attributes (Becker, 1989). The test measures two independent higher-order factors of personality: behavior control and posi- tive mental health (Becker, 1988). In contrast with conventional, clinically developed measures of personality, the TPI was designed to focus on non- symptomatic behavior, and to assess variations in positive mental health. It was conceptualized primarily as a research instrument, but was also expected to be employed in a variety of applied settings. The 120-item TPI contains nine subscales: Behavior Control (17 items), Mental Health (20 items), Fulfillment versus Depression (12 items), Low Self-Centeredness (8 items), Well-Being versus Nervousness (11 items), Expansiveness (12 items), Autonomy (15 items), Self-Esteem (12 items), and Capacity to Love (13 items including 2 experimental items that were not in the original scale). These scales were developed by means of factor analysis of an original pool of 299 items administered to 285 subjects. The resulting 120-item scale was subsequently administered to a stratified random sample of male and female West German citizens ranging in age from 18 to 80 years. Census data were used to select individuals from all regions of West Ger- many. A token payment was made to 961 individuals who completed and Ellis et al. / TRIER PERSONALITY INVENTORY 137 returned the questionnaire (5,180 unsolicited surveys were mailed out). The return rate was approximately 19%. TPI examinees are instructed to respond to a series of statements regarding certain styles of behavior, thinking, feeling, and attitudes that individuals may display. The test format requires that respondents indicate how frequently they experience the behavior, thought, or feeling. An example is the following item from the Behavior Control scale: "I . . . prefer to dance to the beat of a different drummer." The subject must select from "always," "often," "some- times," or "'never" to fill in the blank in each statement. Classical test statistics indicate that the German version of the TPI is a reliable and valid measurement instrument (Becker, 1989). Internal consis- tency was measured using Cronbach's alpha in several studies using the developmental and other samples. The alpha values for the nine scales ranged from a low of .70 to a high of .92. A study of test-retest reliability following an interval of 11 months yielded reliability coefficients ranging from .69 to .78. The correlation between the nine scales of the TPI and a number of other personality measures has been determined, including the Freiburger Pers6nlichkeitsinventar (FPI-R) (Fahrenberg, Hampel, & Selg, 1984); the Personality Research Form (PRF) (Stumpf, Angleitner, Wieck, Jackson, & Beloch-Till, 1984); the German versions of the Minnesota Multiphasic Personality Inventory (MMPI) (Gehring & Blaser, 1982); the 16 Personality Factor Test (Schneewind, Schroder, & Cattell, 1983); and the Eysenck Personality Questionnaire (Eysenck & Eysenck, 1975). Details of these relationships are given in the test manual (Becker, 1989). TRANSLATION OF THE TPI The German version of the TPI was translated into English by a bilingual American psychologist, whose first language is English. A back translation of English version into Gennan was made by a bilingual German psycholo- gist, whose native language is German. Obvious translation errors revealed by the back translation were corrected. SUBJECTS For the purpose of the present DIF analysis, the TPI was administrered to a group of 295 Americans (85 males and 210 females). The subjects were college freshmen and sophomores in a large university in the northeastern United States. They received class credit for their participation in the study. A sample of 213 West German university students (91 males and 122 females) 138 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY were employed as the reference group in the DIF analysis. This particular sample of West Germans was use to have subjects of approximately equal age and educational level as the American sample. The mean age was 18.7 years of American subjects and 21.2 years for German subjects. DIF ANALYSIS The DIF analysis used in the present study requires that the items be dichotomously scored. For this purpose, responses of "always" or "often" were scored as 1 and responses of "sometimes" or "never" as 0. For reverse- scored items, the dichotomization was reversed. The procedure for dichoto- mously scoring items was made on the basis of logic, and, obviously, was somewhat arbitrary. We are in the process of reanalyzing these data with an IRT model that permits polychotomous scoring (Thissen, 1988; Thissen et al., 1988; Thissen, Steinberg, & Gerrard, 1986). This will make it possible to assess the influence of artificially dichotomizing responses in DIF detection. Another assumption of the IRT model used in this study is that the test is relatively unidimensional. The unidimensionality of each of the nine sub- scales was assessed separately by examining the first five eigenvalues generated by a factor analysis of the reduced matrix of item tetrachoric correlations (Lord, 1980). The largest off-diagonal was used as the commu- nality estimate. A pattern of eigenvalues such as (a) the first eigenvalue is large compared to the second, and (b) the second eigenvalue is not apprecia- bly larger than the other is taken as evidence that the items are relatively unidimensional (Lord, 1980, p. 21). The eigenvalues and the percentage of variance accounted for by the nine TPI scales are presented in Table 1. For eight of the nine subscales, the assumption of unidimensionality was sup- ported. For the Behavior Control scale, unidimensionality was questionable. The nine subscales were analyzed separately for DIF items using a constrained version of the three-parameter logistic model (Birnbaum, 1968; Hulin et al., 1983; Lord, 1980). The c or pseudoguessing parameter was constrained to a value estimated using both groups combined, as recom- mended by Lord (1980). Following the initial person and item parameter calibrations for each group separately, using the computer program BILOG (Mislevy & Bock, 1986), a procedure for linking ability metrics described by Stocking and Lord (1983) was applied iteratively. Lord's (1980) chi-square index was used to assess DIF; alpha was set at .05 throughout. In the case of the constrained three-parameter logistic model, Lord's index of DIF tests the hypothesis that for a given item (i) both biR = biF and aiR = aiF, where R and F refer to Ellis et al. / TRIER PERSONALITY INVENTORY 139 TABLE 1 Eigenvalues of Reduced Tetrachoric Correlation Matrices for the TPP Subscales Eigenvalue Number Subcale 1 2 3 4 5 Behavior control 3.75 2.41 1.13 .77 .49 (22) (14) (7) (5) (3) Mental health 13.00 1.29 .52 .35 .30 (65) (6) (3) (2) (2) Fulfillment versus depression 9.72 .24 .19 .13 .11 (81) (2) (2) (1) (1) Low self-enteredness 3.06 1.57 .36 .15 .10 (38) (20) (5) (2) (1) Well-being versus nervousness 4.23 1.24 .65 .41 .33 (38) (11) (6) (4) (3) Expansiveness 3.33 1.67 .57 .35 .19 (28) (14) (5) (3) (2) Autonomy 7.68 1.04 .64 .41 .23 (51) (7) (4) (3) (2) Self-esteem 4.58 1.37 .48 .37 .33 (38) (11) (4) (3) (3) Capacity to love 9.00 .64 .34 .21 .18 (69) (5) (3) (2) (1) NOTE: Largest off-diagonal correlations were used as communality estimates. Numbers in parentheses are the percentages of variance accounted for by each factor. a. TPI = Trier Personaity Inventory. the reference and focal groups, respectively. The index is calculated as follows: 2 r1 x = vi F, vi, (2) where v = - , -a], and 9 refer to the estimated discrimination and difficulty parameters and r, is the inverse of the asXmptotic sampling variance-covariance matrix for [8, - 9j] and [a, - a,]. This index is distributed as chi-square with two degress of freedom. Details of the proce- dures required to evaluate this statistic are given in Lord (1980, pp. 212-224) and in Hulin et al. (1983, pp. 174-176). Recent research using simulated data has examined the properties of Lord's chi-square index of DIF and the accuracy of vanous parameter estimation procedures, for example, joint maximum likelihood, marginal maximum likeli- 140 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY McLaughlin & Drasgow, 1987). In general, marginal maximum likelihood estimation, such as that used by the computer program BILOG, has been demonstrated to be more accurate than joint maximum likelihood estimation (Drasgow, 1989). Examination of the maximum likelihood estimation proce- dure also indicated that when using Lord's chi-square as an index of DIF, Type I error rates were inflated because standard errors of item parameter estimates were spuriously low (McLaughlin & Drasgow, 1987). More recent examina- tion of marginal maximum likelihood estimation indicated that actual alpha rates were close to nominal rates, although this varied as a function of sample size (Lim & Drasgow, 1990). Because the purpose of this study was to maximize the identification of all DIF items, and because marginal maximum likelihood estimation was employed, the alpha rate was not adjusted. RESULTS AND DISCUSSION Eleven of the 120 items tested (9.2%) were found to display DIF.' From among the 11 DIF items, 7 were from the Behavior Control scale, and there was 1 DIF item from each of the following scales: Low Self-Centeredness, Well-Beings versus Nervousness, Expansiveness, and Self-Esteem. There were no DIF items from the Mental Health, Fulfillment versus Depression, Autonomy, or Capacity to Love scales. A note of caution is in order in interpreting the seven Behavior Control DIF items, because the assumption of unidmensionality may have been violated, leading to an overidentification of DIF items. The item parameters (a = slope, and b = preference) and Lord's chi-square statistic for each DIF item are shown in Table 2. The 11 DIF items were examined by four bilinguals (2 West Germans and 2 Americans) for possible translation errors.2 Two possible translation errors may have caused Item 95 in the Behavior Control scale to show DIF. First, the phrase einen hohen Grad von (a high degree of), was inadvertently omitted in the English translation. Probably more crucial was the translation of the German word Eigenwilligkeit as "individuality." Thej udges concluded that a better translation would have been "willfulness." Americans may have found it difficult to respond to a question that characterizes an individual who exhibits the potentially incompatible traits of individuality and "lack of restraint." It is possible that Americans think of individuality as a positive attribute, whereas they interpret lack of restraint as negative. On the other hand, willfulness and lack of restraint both represent negative personal attributes for the trait of Behavior Control. Furthermore, a comparison of Germans' and American' rankings of Ro- keach's values (Ellis, 1988) has shown that Americans ranked individualism Ellis et al. / TRIER PERSONALITY INVENTORY 141 TABLE 2 Estimated Item Parameters and Chi-Square Statistics Study 1 Study 2 A x2 A 2 a a Item GM AM GM AM GM AM GM AM Ql 0.60 .24 1.64 4.66 62.3 0.67 .30 1.68 4.02 20.6 QlOR 0.30 .25 -2.89 -4.73 19.3 0.31 .32 -2.73 -3.96 n.s. Q17R 1.13 .49 0.46 3.18 7.7 1.13 .74 0.46 2.28 n.s. Q27 1.05 .89 -1.20 -2.23 n.s. 1.08 .57 -1.14 -3.52 10.6 Q28R 0.69 .21 -0.50 2.55 60.8 0.72 .20 -0.43 2.35 40.4 Q37 0.68 .47 1.06 1.74 6.3 0.74 .76 1.11 1.31 n.s. Q49R 1.56 .40 -1.36 3.30 8.4 1.54 .44 -1.39 -3.39 6.5 Q60 0.34 .34 -1.28 0.06 10.4 0.33 .44 -1.45 -0.36 n.s. Q64 0.49 .35 2.00 0.11 8.6 0.55 .30 2.05 0.85 n.s. Q83R 0.75 .40 -0.74 1.33 18.0 0.78 .35 -0.66 1.84 24.7 Q95R 0.56 .23 0.89 3.71 60.6 0.60 .23 0.95 3.68 36.4 Q107 0.21 .26 2.39 0.49 9.0 0.23 .24 2.53 0.42 18.5 Q95R (Retranslated) 0.60 .22 0.95 0.50 .3 ns NOTE: GM = German, AM American; R indicates an item that was reverse scored. significantly higher than Germans. It was therefore assumed that the coupling of a positive attribute with a negative one, in conjunction with the fact that Americans place a high value on individuality, may have caused Americans to focus on the positive characteristic of individuality. This may explain why it was more difficult for Americans to respond "sometimes" or "never" to this reverse-scored item (Americans, b = 3.71; Germans, b = .89). A sug- gested retranslation for Item 95 is as follows: "People who have held on to a high degree of willfulness and lack of restraint are . . . attractive to me." STUDY 2 A second study was conducted to replicate the DIF analysis of Study 1 and to determine whether or not it would be possible to eliminate DIF in Item 95 by retranslation. The original English version of the TPI, with the addition of the retranslated Item 95, was administered to 203 college students from the same university as those in Study 1. There were 101 males and 102 females in the second sample. Their mean age was 19.4 years. The German sample and the analysis were the same as that used in Study 1. 142 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY RESULTS AND DISCUSSION From among the 120 items, 7 (5.8%) were found to display DIF (see Table 2). The item parameters (a = discrimination and b = preference) and Lord's chi-square statistic for each DIF item are shown in Table 2. The reference group (German sample) was the same in both studies; however, the a and b parameters differed slightly. This was due to the fact that the c parameters were fixed to values based on the combined American and German samples. Although the c parameters for the DIF analyses in Studies 1 and 2 differed slightly, these differences were not significant. Six of the 11 items (54.6%) exhibiting DIF in Study 1 also displayed DIF in Study 2 (Qi, Q28, Q49, Q83, Q95, Q107). Five items (Q10, Q17, Q37, Q60, and Q64) that exhibited DIF in Study 1 did not have significant chi-square indexes in Study 2, although the chi-square for Q60 was almost significant (X2[2, N = 416] = 5.7, p < .10). One item, number 27 from the Capacity to Love scale, was found to display DIF in Study 2 but not in Study 1. Overall, the chi-square indexes were lower in Study 2 than in Study 1. Nine of the 11 items (81.8%) that exhibited DIF in Study 1 had lower chi-square indexes in the replication study than in the original DIF analysis. In its original form, Item 95 had a significant chi-square DIF index in Study 2 (e2[2, N = 461] = 36.4, p <.001), as it had in Study 1 (X2[2, N= 416] = 60.6, p < .001). When analyzed in its retranslated form, Item 95 no longer displayed DIF (X2[2, N = 416] = .3). To determine whether there were any similarities in item content, a factor analysis of the 12 DIF items from Studies 1 and 2 was conducted. Because the factor analysis was exploratory in nature, all 12 DIF items, as opposed to only 7 replicated items, were analyzed. On the basis of the scree test (Cattell, 1966), four factors were retained, and a varimax rotation was performed. The rotated factor matrix is presented in Table 3. Despite some complications in the factor analysis, for example, some items loaded on more than one factor, it is possible to speculate regarding what these factors may represent. Factor 1 consists of three items (Q1, Q64, and Q83) that deal with avoidance behavior. Germans were more likely to say that they always or often avoid fuss and excitement and that they sometimes or never are inclined to avoid unpleasant situations. Both of these items (Q1 and Q83) were also more discriminating for Germans than for Americans. On the other hand, Americans were more likely to say they always or often avoid arguments with others (Q64). Both groups show preferences for avoiding specific types of situations. Germans prefer to avoid fuss and excitement and Americans prefer to avoid arguments with others. Ellis et al. / TRIER PERSONALITY INVENTORY TABLE 3 Factor Loadings for 12 DIP Items (Varimax Rotation) Item Factor I Factor 2 Factor 3 Factor 4 Q1 AZ -.36 .32 -.13 Q10 -.15 -.04 .13 -6 Q17 .34 .03 -.12 Q27 .24 -.20 .39 Q28 -.01 .47 -A4 .07 Q37 .18 .08 64 .39 Q49 -.14 .01 66 -.03 Q60 -.25 -.04 .07 8 Q64 77 .19 -.04 .12 Q83 81 .03 -.03 -.07 Q95 -.04 .01 -.54 .31 Q107 -.07 .77 .14 -.02 NOTE: The underlined figures indicate items that have factor loadings >.04. a. DIF = differential item functioning. However, Germans are more likely than Americans not to avoid, but rather to face, unpleasant situations in general (Q83). Factor 2 is made up of three items (Q17, Q27, and Q107) that concern self-protective behaviors such as planning for the future, taking health precautions, or making oneself happy by being helpful to others. Germans were more likely than American to describe themselves as sometimes or never giving a lot of thought to their future (Q17R). Americans were more likely to say that it always or often makes them happy to be helpful to others; however, this item (Q27) was much more discriminating for Germans (Ger- mans, a = 1.08; Americans a = .57). Americans were also more likely to say that they are always or often interested in questions of health precautions (Q107). Factor 3 consists of four items (Q28, Q37, Q49, and Q95) that deal with reserved versus spontaneous behavior. Germans were more likely to say that they seldom or never are full of enthusiasm or attracted to people who have maintained their own individuality (willfulness) and lack of restraint (Q95R). When compared to Americans, German subjects were more likely to say that they always or often adopt a posture of caution (Q37). Americans, on the other hand, were more likely to respond that they are sometimes or never physically indisposed (Q49R). It should be noted, however, that this item is much more discriminating for Germans than for American (Germans, a = 1.54; Americans, a = .44). These four DIF items suggest that Americans behave more spontaneously than West Germans. 143 144 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY Two items (QlO and Q60) dealing with self-criticism make up Factor 4. Americans were more likely than Germans to say they sometimes or never behave in ways that others can take as selfish (QIOR), whereas Germans were more likely to say that they are always or often open to criticism about themselves (Q60). Together, these items indicate that Germans are more accepting of criticism than are Americans, even when it may stem from acting selfishly. GENERAL DISCUSSION Previous work has suggested that IRT analysis of translated tests is necessary, so that DIF items may be identified and either eliminated or modified, producing a test with the property of measurement equivalence. The results of the present research demonstrate that an IRT analysis may also be used in a somewhat more experimental fashion. Once DIF items have been identified, those whose DIF is thought to be due to translation error may be retranslated, readministered, and reanalyzed to determine whether or not DIF is eliminated. This procedure has two benefits beyond the more traditional ones of identification and elimination. First, if the measurement equivalence of some originally DIF items can be achieved by retranslation, the need to remove fewer items will tend to preserve the integrity of the original test. Second, the IRTprocedure provides an objective way to determine the correct translation for items that are in dispute. In the process of test translation, the decision regarding the final version of an item that fails the back translation process sometimes may be more or less a judgment call of the translators. The present results suggest that, when a translation is in dispute, the devel- opers of the test ought to administer both versions of the translation and analyze the items separately for DIF to determine which translation has the lower DIF index. The present results also indicate that there were differences between the outcomes of the original DIF analysis and its replication in Study 2. The total of 11 statistically significant DIF items identified in Study 1 fell to 7 in Study 2, and one of the seven DIF items identified in the replication had not been found to have DIF originally. In general, there appeared to be considerable shrinkage in the chi-square values for the significant DIF items in Study 2 as compared with those in Study 1. The broader implications of this phenome- non for IRT analysis of measurement equivalence are not immediately clear Hambleton (1989) describes the influence of "regression effects due to errors of measurement" on item parameter estimation in simulation studies. He concludes that "the amount of systematic error depends on sample size: large Ellis et al. / TRIER PERSONALITY INVENTORY 145 samples imply small errors; smaller samples imply large errors due to the regression effect" (Hambleton, 1989, p. 102). In the present case, the Amer- ican sample size in Study 1 was 50% larger than Study 2, but, even with equivalently large samples in the two studies, some shrinkage in the size of the chi-square would probably have been observed. From a practical point of view, this argues for some degree of caution in the identification of DIF items. Confidence in the parameter estimates and chi-square could obviously be increased by using samples larger than those used in the present study. This disagreement in identification of significant DIF items between the two studies, however, must be viewed in overall context of rather substantial agreement. The product-moment correlation between the chi-square values of the 120 items from Studies 1 and 2 was +.885, which was highly significant, p < .001. In other words, the present item response theory method generated high agreement in its DIF assessments of 120 items when two independent samples of American subjects were compared to the same German sample, even though there were some shifts in which items showed significant DIE. It is also possible to reduce the likelihood that items will be judged as DIF erroneously by using an alpha level of .01 or lower rather than .05. In the present study, if alpha had been lowered to .01, six items (Ql, Q10, Q28, Q60, Q83, and Q95) would have been identified as exhibiting DIF in Study 1, and six items would have been designated DIE in Study 2 (Q1, Q27, Q28, Q83, Q95, and Q107). In other words, with alpha at .01, four out of the original six DIF items (66.7%) were replicated in Study 2. On the other hand, if the goal is to identify as many DIEF items as possible, for example, in a study designed to identify all possible mistranslated items, it may not be wise to reduce alpha to this degree. One possible solution to this dilemma would be another replication study. Items displaying DIF in two out of three analyses would be considered "true" DIEF items. In analyzing the content of DIEF items for the purpose of examining differences between the attributes of two cultural groups, as has been done above, the problem of shrinkage is probably not a serious restraint. As long as groups of DIF items with somewhat similar content provide the basis of such analyses, there would appear to be little danger of serious error. It is, nevertheless, clear from the present results that replication with sufficiently large samples is required for statements regarding cultural differences to be made with any degree of certainty. The finding that retranslation of a DIF item from an originally inaccurate translation resulted in elimination of the DIF indicated that this use of IRT analysis has practical promise. In this case, what was thought initially to be a suitable translation was found to be inadequate. Instead of removing the item 146 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY from the test, in this instance retranslation proved to be an appropriate method for securing measurement equivalence. The fact that only a single item was found to show DIF due to translation error testifies to the effectiveness of the translation/back translation process. Finally, it is hoped that the results of this study will encourage cross- cultural researchers to employ an IRT analysis to identify and correct DIF items in translated tests. More important, as was suggested by Goldstein and Wood (1989). and by Thissen and Steinberg (1988), it is hoped that the application of IRT in cross-cultural testing will go beyond the statistical modeling of item responses, and will lead to insights regarding the culturally unique psychological processes underlying the responses to individual items on psychological tests and measures. NOTES 1. Pals of the results from Study 1 were reported in a paper presented in June, 1989 at the 2nd regional conference of the International Association for Cross-Cultural Psychology, Am- sterdam, the Netherlands. 2. Contents of these items in English and German are available on request from the first author. REFERENCES Becker, P. (1988). Seelische Gesundheit und Verhaltenskontrolle: Zwei replizierbare, van- anzstarke Personlichkeitsfaktoren [Mental health and behavior control: TWo replicable, primay personality factors]. Zeitschrnftfir Differentielle und Diagnostische Psychologie, 9, 13-38. Becker, P. (1989). Der Trierer Personlichkeitsfragebogen (TPF) Handanweisung [The Trier Personality Inventory (TPI) Manual]. G6ttingen, West Germany: Hogrefe. Birnbaum, A. (1968). Some latent trait models and their use in inferring examinee's ability. In F. M. Lord & M. R. Novick, Statistical theories of mental test scores (pp. 397-497). Reading, MA: Addison-Wesley. Candell, G. L., & Hulin, C. L. (1987). Cross-language and cross-cultural comparisons in scale translations: Independent sources of information about item nonequivalence. Journal of Cross-Cultural Psychology, 17, 417-440. Cattell, R. B. (1966). The scree test forthe number of factors. Multivariate Behavioral Research, 1,245-276. Drasgow, F. (1987). Study of the measurement bias of two standardized psychological tests. Journal of Applied Psychology, 72, 19-29. Drasgow, F. (1989). An evaluation of marginal maximum likelihood estimation for the two-paramneter model. Applied Psychological Measurement, 13, 77-90. Ellis, B. B. (1988). Hofstede's culture dimensions and Rokeach's values: How reliable is the relationship? In J. W. Berry & R. C. Annis (Eds.), Ethnic psychology: Research and practice Ellis et al. / TRIER PERSONALrIY INVENTORY 147 Drasgow, F. (1989). An evaluation of marginal maximum likelihood estimation for the two-parameter model. Applied Psychological Measurement, 13, 77-90. Ellis, B. B. (1988). Hofstede's culture dimensions and Rokeach's values: How reliable is the relationship? In J. W. Berry & R. C. Annis (Eds.), Ethnic psychology: Research and practice with immigrants, refugees, native peoples, ethnic groups, and sojourners (pp. 266-274). Lisse, the Netherlands: Swets & Zeitlinger. Ellis, B. B. (1989). Differential item functioning: Implications for test translations. Journal of Applied Psychology, 74, 912-921. Ellis, B. B., Minsel, B., & Becker, P. (1989). Evaluation of attitude survey translations: An investigation using item response theory. Intenational Journal of Psychology, 24,665-684. Eysenck, H. J., & Eysenck, S.B.G. (1975) Manual of the Eysenck Personality Questionnaire. San Diego: Educational and Industrial Testing Service. Fahrenberg, J., Hampel, R., & Selg, H. (1984). Das Freiburger Personlichkeitsinventar FPI Revidierte Fassung FPI-R und teitweise geanderte Fassung FPI-AI. Handanweisung. 4. revidierte Auflage rlThe Freiburger Personality Inventory, FPI, revised version, FPI-R, and partially modified, FPI-Al. Manual. 4th edition]. Gottingen, West Germany: Hogrefe. Gehring, A., & Blaser, A. (1982). MinnesotaMultiphasic Personality Inventory: MMPI Deutsche Kurtf,'ormftr Handauswertung. Handbuch (Minesota Multiphasic Personality Inventory: MMPI. German short forn for hand scoring. Handbook]. Bern: Huber. Goldstein, H., & Wood, R. (1989). Five decades of item response modeling. British Journal of Mathematical and Statistical Psychology, 42, 139-167. Hambleton, R. K. (1989). Constructing tests with item response models: A discussion of a method and two problems. Buletin of the International Test Commission, 29, 96-106. Hambleton, R. K., & Swaminathan, H. (1985). Item response theory: Principles and applica- tions. Boston: Kluwer-Nijhoff. Holland, P. W., & Thayer, D. T. (1986, April). Differential item performance and the Mantel- Haenszel procedure. Paper presented at the annual meeting of the American Educational Research Association, San Francisoo, CA. Holland, P. W., & Thayer, D. T. (1988). Differential item performance and the Mantel-Haenszel procedure. In H. Wainer & H. I. Braun (Eds.), Test validity (pp. 129-145). Hillsdale, NJ: Lawrence Eribaum. Hulin, C. L. (1987). Apsychometrictheory of evaluations of item and scaletranslations: Fidelity across languages. Journal of Cross-Cultural Psychology, 18,115-142. Hulin, C. L., Drasgow, F., & Komocar, J. (1982). Applications of item response theory to analysis of scale translations. Journal of Applied Psychology, 67, 818-825. Hulin, C. L., Drasgow, F., & Parsons, C. K. (1983). Item response theory: Applications to psychological measurement. Homewood, IL: Dow Jones Irwin. Hulin, C. L., & Mayer, L. J. (1986). Psychometric equivalence of a translation of the Job Descriptive Index into Hebrew. Jounal of Applied Psychology, 71, 83-94. Lim, R. G., & Drasgow, F. (1990). Evaluation of two methods for estimafing item response theory parameters when assessing differental item functiong. Joumal ofApplied Psychol- ogy, 75, 164-174. Lord, F. (1980). Applications of item response theory to practical testing problems. Hillsdale, NJ: Lawrence Eribaum. Lord, F. M., & Novick, M. R. (1968). Statistical theories of mental test scores. Reading, MA: Addison-Wesley. McLaughlin, M. E., & Drasgow, F. (1987). Lord's chi-square test of item bias with estimated and with known person paramers. Applied Psychological Measurement, 11, 161-173. 148 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY Stocking, M. L., & Lord, F. M. (1983). Developing a common metric in item response theory. Applied Psychological Measurement, 7, 201-210. Stumpf, H., Angleitner, A., Wieck, T., Jackson, D., & Beloch-Till, H. (1984). Deutsche Person- ality Research Form (PRF). Gottingen, West Germany: Hogrefe. Thissen, D. (1988). MULTILOG user's guide. Mooresville, IN: Scientific Software. Thissen, D., & Steinberg, L. (1988). Data analysis using item response theory Psychological Bulletin, 104, 385-395. Thissen, D., Steinberg, L., & Gerrard, M. (1986). Beyond group mean differences: The concept of item bias. Psychological Bulletin, 99, 118-181. Thissen, D., Steinberg, L., & Wainer, H. (1988). Use of item response theory in the study of group difference in trace lines. InH. Wainer&H. I. Braun (Eds.), Test validity (pp. 147-170). Hillsdale, NJ: Lawrence Erlbaum. Thissen, D., Steinberg, L., & Wainer, H. (1989, September). Detection of differential item functioning using the parameters of item response models. Paper presented at the Differential Item Functioning Conference, Educational Testing Service, Princeton, NJ. Barbara B. Ellis is an assistant professor of psychology at the University of Texas at El Paso. Her primary research interests are in cross-cultural testing and measurement, and personnel selection and organizational behavior in multinational organizations, Peter Becker is a professor in the Department of Psychology at the University of Trier in Germany. His current research interests are in the areas oftestdevelopment, multivariate personality psychology, and health psychology. Herbert D. Kimmel is a graduate research professor in the Department of Psychology at the University of South Florida He is a recipient of the Senior Scientist Prize from the Alexander von Humboldt Foundationfor research in Germany, and has published in a wide variety of areas including statistics and experimnental design and computer simulations of conditioning theory. </meta-value></custom-meta></custom-meta-wrap></article-meta></front><back><notes><p><list list-type="order"><list-item><p>1. Pals of the results from Study 1 were reported in a paper presented in June, 1989 at the 2nd regional conference of the International Association for Cross-Cultural Psychology, Amsterdam, the Netherlands.</p></list-item><list-item><p>2. Contents of these items in English and German are available on request from the first author.</p></list-item></list></p></notes><ref-list><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Becker, P.</surname></name> (<year>1988</year>). <article-title>Seelische Gesundheit und Verhaltenskontrolle: Zwei replizierbare, vananzstarke Persönlichkeitsfaktoren [Mental health and behavior control: TWo replicable, primary personality factors]</article-title>. <source>Zeitschrnft für Differentielle und Diagnostische Psychologie</source>, <volume>9</volume>, <fpage>13</fpage>-<lpage>38</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Becker, P.</surname></name> (<year>1989</year>). <source>Der Trierer Persönlichkeitsfragebogen (TPF) Handanweisung [The Trier Personality Inventory (TPI) Manual]</source>. <publisher-loc>Göttingen, West Germany</publisher-loc>: <publisher-name>Hogrefe</publisher-name>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Birnbaum, A.</surname></name> (<year>1968</year>). <article-title>Some latent trait models and their use in inferring examinee's ability</article-title>. In F. M. Lord & M. R. Novick, <source>Statistical theories of mental test scores</source> (pp. <fpage>397</fpage>-<lpage>497</lpage>). <publisher-loc>Reading, MA</publisher-loc>: <publisher-name>Addison-Wesley</publisher-name>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Candell, G. L.</surname></name> , & <name name-style="western"><surname>Hulin, C. L.</surname></name> (<year>1987</year>). <article-title>Cross-language and cross-cultural comparisons in scale translations: Independent sources of information about item nonequivalence</article-title>. <source>Journal of Cross-Cultural Psychology</source>, <volume>17</volume>, <fpage>417</fpage>-<lpage>440</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Cattell, R. B.</surname></name> (<year>1966</year>). <article-title>The scree test for the number of factors</article-title>. <source>Multivariate Behavioral Research</source>, <volume>1</volume>, <fpage>245</fpage>-<lpage>276</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Drasgow, F.</surname></name> (<year>1987</year>). <article-title>Study of the measurement bias of two standardized psychological tests</article-title>. <source>Journal of Applied Psychology</source>, <volume>72</volume>, <fpage>19</fpage>-<lpage>29</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Drasgow, F.</surname></name> (<year>1989</year>). <article-title>An evaluation of marginal maximum likelihood estimation for the two-paramneter model</article-title>. <source>Applied Psychological Measurement</source>, <volume>13</volume>, <fpage>77</fpage>-<lpage>90</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Ellis, B. B.</surname></name> (<year>1988</year>). <article-title>Hofstede's culture dimensions and Rokeach's values: How reliable is the relationship?</article-title> In <name name-style="western"><surname>J. W. Berry</surname></name> & <name name-style="western"><surname>R. C. Annis</surname></name> (Eds.), <source>Ethnic psychology: Research and practice</source></citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Drasgow, F.</surname></name> (<year>1989</year>). <article-title>An evaluation of marginal maximum likelihood estimation for the two-parameter model</article-title>. <source>Applied Psychological Measurement</source>, <volume>13</volume>, <fpage>77</fpage>-<lpage>90</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Ellis, B. B.</surname></name> (<year>1988</year>). <article-title>Hofstede's culture dimensions and Rokeach's values: How reliable is the relationship?</article-title> In <name name-style="western"><surname>J. W. Berry</surname></name> & <name name-style="western"><surname>R. C. Annis</surname></name> (Eds.), <source>Ethnic psychology: Research and practice with immigrants, refugees, native peoples, ethnic groups, and sojourners</source> (pp. <fpage>266</fpage>-<lpage>274</lpage>). <publisher-loc>Lisse, the Netherlands</publisher-loc>: <publisher-name>Swets & Zeitlinger</publisher-name>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Ellis, B. B.</surname></name> (<year>1989</year>). <article-title>Differential item functioning: Implications for test translations</article-title>. <source>Journal of Applied Psychology</source>, <volume>74</volume>, <fpage>912</fpage>-<lpage>921</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Ellis, B. B.</surname></name> , <name name-style="western"><surname>Minsel, B.</surname></name>, & <name name-style="western"><surname>Becker, P.</surname></name> (<year>1989</year>). <article-title>Evaluation of attitude survey translations: An investigation using item response theory</article-title>. <source>International Journal of Psychology</source>, <volume>24</volume>, <fpage>665</fpage>-<lpage>684</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Eysenck, H. J.</surname></name> , & <name name-style="western"><surname>Eysenck, S.B.G.</surname></name> (<year>1975</year>) <source>Manual of the Eysenck Personality Questionnaire</source>. <publisher-loc>San Diego</publisher-loc>: <publisher-name>Educational and Industrial Testing Service</publisher-name>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Fahrenberg, J.</surname></name> , <name name-style="western"><surname>Hampel, R.</surname></name>, & <name name-style="western"><surname>Selg, H.</surname></name> (<year>1984</year>). <source>Das Freiburger Persönlichkeitsinventar FPI Revidierte Fassung FPI-R und teitweise geanderte Fassung FPI-AI. Handanweisung. 4. revidierte Auflage rlThe Freiburger Personality Inventory, FPI, revised version, FPI-R, and partially modified, FPI-Al. Manual. 4th edition]</source>. <publisher-loc>Gottingen, West Germany</publisher-loc>: <publisher-name>Hogrefe</publisher-name>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Gehring, A.</surname></name> , & <name name-style="western"><surname>Blaser, A.</surname></name> (<year>1982</year>). <source>MinnesotaMultiphasic Personality Inventory: MMPI Deutsche Kurtzform für Handauswertung. Handbuch (Minesota Multiphasic Personality Inventory: MMPI. German short form for hand scoring. Handbook]</source>. <publisher-loc>Bern</publisher-loc>: <publisher-name>Huber</publisher-name>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Goldstein, H.</surname></name> , & <name name-style="western"><surname>Wood, R.</surname></name> (<year>1989</year>). <article-title>Five decades of item response modeling</article-title>. <source>British Journal of Mathematical and Statistical Psychology</source>, <volume>42</volume>, <fpage>139</fpage>-<lpage>167</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Hambleton, R. K.</surname></name> (<year>1989</year>). <article-title>Constructing tests with item response models: A discussion of a method and two problems</article-title>. <source>Buletin of the International Test Commission</source>, <volume>29</volume>, <fpage>96</fpage>-<lpage>106</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Hambleton, R. K.</surname></name> , & <name name-style="western"><surname>Swaminathan, H.</surname></name> (<year>1985</year>). <source>Item response theory: Principles and applications</source>. <publisher-loc>Boston</publisher-loc>: <publisher-name>Kluwer-Nijhoff</publisher-name>.</citation></ref><ref><citation citation-type="confproc" xlink:type="simple"><name name-style="western"><surname>Holland, P. W.</surname></name> , & <name name-style="western"><surname>Thayer, D. T.</surname></name> (1986, April). <article-title>Differential item performance and the Mantel-Haenszel procedure</article-title>. Paper presented at the <conf-name>annual meeting of the American Educational Research Association</conf-name>, <conf-loc>San Francisoo, CA</conf-loc>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Holland, P. W.</surname></name> , & <name name-style="western"><surname>Thayer, D. T.</surname></name> (<year>1988</year>). <article-title>Differential item performance and the Mantel-Haenszel procedure</article-title>. In <name name-style="western"><surname>H. Wainer</surname></name> & <name name-style="western"><surname>H. I. Braun</surname></name> (Eds.), <source>Test validity</source> (pp. <fpage>129</fpage>-<lpage>145</lpage>). <publisher-loc>Hillsdale, NJ</publisher-loc>: <publisher-name>Lawrence Eribaum</publisher-name>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Hulin, C. L.</surname></name> (<year>1987</year>). <article-title>A psychometric theory of evaluations of item and scale translations: Fidelity across languages</article-title>. <source>Journal of Cross-Cultural Psychology</source>, <volume>18</volume>, <fpage>115</fpage>-<lpage>142</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Hulin, C. L.</surname></name> , <name name-style="western"><surname>Drasgow, F.</surname></name>, & <name name-style="western"><surname>Komocar, J.</surname></name> (<year>1982</year>). <article-title>Applications of item response theory to analysis of scale translations</article-title>. <source>Journal of Applied Psychology</source>, <volume>67</volume>, <fpage>818</fpage>-<lpage>825</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Hulin, C. L.</surname></name> , <name name-style="western"><surname>Drasgow, F.</surname></name>, & <name name-style="western"><surname>Parsons, C. K.</surname></name> (<year>1983</year>). <source>Item response theory: Applications to psychological measurement</source>. <publisher-loc>Homewood, IL</publisher-loc>: <publisher-name>Dow Jones Irwin</publisher-name>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Hulin, C. L.</surname></name> , & <name name-style="western"><surname>Mayer, L. J.</surname></name> (<year>1986</year>). <article-title>Psychometric equivalence of a translation of the Job Descriptive Index into Hebrew</article-title>. <source>Journal of Applied Psychology</source>, <volume>71</volume>, <fpage>83</fpage>-<lpage>94</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Lim, R. G.</surname></name> , & <name name-style="western"><surname>Drasgow, F.</surname></name> (<year>1990</year>). <article-title>Evaluation of two methods for estimating item response theory parameters when assessing differential item functioning</article-title>. <source>Journal of Applied Psychology</source>, <volume>75</volume>, <fpage>164</fpage>-<lpage>174</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Lord, F.</surname></name> (<year>1980</year>). <source>Applications of item response theory to practical testing problems</source>. <publisher-loc>Hillsdale, NJ</publisher-loc>: <publisher-name>Lawrence Eribaum</publisher-name>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Lord, F. M.</surname></name> , & <name name-style="western"><surname>Novick, M. R.</surname></name> (<year>1968</year>). <source>Statistical theories of mental test scores</source>. <publisher-loc>Reading, MA</publisher-loc>: <publisher-name>Addison-Wesley</publisher-name>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>McLaughlin, M. E.</surname></name> , & <name name-style="western"><surname>Drasgow, F.</surname></name> (<year>1987</year>). <article-title>Lord's chi-square test of item bias with estimated and with known person parameters</article-title>. <source>Applied Psychological Measurement</source>, <volume>11</volume>, <fpage>161</fpage>-<lpage>173</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Stocking, M. L.</surname></name> , & <name name-style="western"><surname>Lord, F. M.</surname></name> (<year>1983</year>). <article-title>Developing a common metric in item response theory</article-title>. <source>Applied Psychological Measurement</source>, <volume>7</volume>, <fpage>201</fpage>-<lpage>210</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Stumpf, H.</surname></name> , <name name-style="western"><surname>Angleitner, A.</surname></name>, <name name-style="western"><surname>Wieck, T.</surname></name>, <name name-style="western"><surname>Jackson, D.</surname></name>, & <name name-style="western"><surname>Beloch-Till, H.</surname></name> (<year>1984</year>). <source>Deutsche Personality Research Form (PRF)</source>. <publisher-loc>Gottingen, West Germany</publisher-loc>: <publisher-name>Hogrefe</publisher-name>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Thissen, D.</surname></name> (<year>1988</year>). <source>MULTILOG user's guide</source>. <publisher-loc>Mooresville, IN</publisher-loc>: <publisher-name>Scientific Software</publisher-name>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Thissen, D.</surname></name> , & <name name-style="western"><surname>Steinberg, L.</surname></name> (<year>1988</year>). <article-title>Data analysis using item response theory</article-title><source>Psychological Bulletin</source>, <volume>104</volume>, <fpage>385</fpage>-<lpage>395</lpage>.</citation></ref><ref><citation citation-type="journal" xlink:type="simple"><name name-style="western"><surname>Thissen, D.</surname></name> , <name name-style="western"><surname>Steinberg, L.</surname></name>, & <name name-style="western"><surname>Gerrard, M.</surname></name> (<year>1986</year>). <article-title>Beyond group mean differences: The concept of item bias</article-title>. <source>Psychological Bulletin</source>, <volume>99</volume>, <fpage>118</fpage>-<lpage>181</lpage>.</citation></ref><ref><citation citation-type="book" xlink:type="simple"><name name-style="western"><surname>Thissen, D.</surname></name> , <name name-style="western"><surname>Steinberg, L.</surname></name>, & <name name-style="western"><surname>Wainer, H.</surname></name> (<year>1988</year>). <article-title>Use of item response theory in the study of group difference in trace lines</article-title>. In <name name-style="western"><surname>H. Wainer</surname></name> & <name name-style="western"><surname>H. I. Braun</surname></name> (Eds.), <source>Test validity</source> (pp. <fpage>147</fpage>-<lpage>170</lpage>). <publisher-loc>Hillsdale, NJ</publisher-loc>: <publisher-name>Lawrence Erlbaum</publisher-name>.</citation></ref><ref><citation citation-type="confproc" xlink:type="simple"><name name-style="western"><surname>Thissen, D.</surname></name> , <name name-style="western"><surname>Steinberg, L.</surname></name>, & <name name-style="western"><surname>Wainer, H.</surname></name> (1989, September). <article-title>Detection of differential item functioning using the parameters of item response models</article-title>. Paper presented at the <conf-name>Differential Item Functioning Conference, Educational Testing Service</conf-name>, <conf-loc>Princeton, NJ</conf-loc>.</citation></ref></ref-list></back></article></istex:document></istex:metadataXml><mods version="3.6"><titleInfo lang="en"><title>An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)</title></titleInfo><titleInfo type="alternative" lang="en" contentType="CDATA"><title>An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)</title></titleInfo><name type="personal"><namePart type="given">Barbara B.</namePart><namePart type="family">Ellis</namePart><affiliation>University of Texas at El Paso</affiliation></name><name type="personal"><namePart type="given">Peter</namePart><namePart type="family">Becker</namePart><affiliation>Universitait Trier, Germany</affiliation></name><name type="personal"><namePart type="given">Herbert D.</namePart><namePart type="family">Kimmel</namePart><affiliation>University of South Florida</affiliation></name><typeOfResource>text</typeOfResource><genre type="research-article" displayLabel="research-article"></genre><originInfo><publisher>Sage Publications</publisher><place><placeTerm type="text">Sage CA: Thousand Oaks, CA</placeTerm></place><dateIssued encoding="w3cdtf">1993-06</dateIssued><copyrightDate encoding="w3cdtf">1993</copyrightDate></originInfo><language><languageTerm type="code" authority="iso639-2b">eng</languageTerm><languageTerm type="code" authority="rfc3066">en</languageTerm></language><physicalDescription><internetMediaType>text/html</internetMediaType></physicalDescription><abstract lang="en">The measurement equivalence of an English-language version of the Trier Personality Inventory (TPI) was evaluated, using statistical methods based on item response theory (IRT) to identity items that displayed differential item functioning (DIF). In Study 1, the TPI was administered to 295 American and 213 West German subjects. From among the 120 items tested, 11 DIF items were detected. A follow-up study using a new sample of 203 Americans was conducted to replicate the original study and to evaluate the post hoc explanation that translation error was the source of DIF for one item. DIF was replicated for 6 of the 11 DIF items found in Study 1, and DIF was eliminated for one item by retranslation. Overall, there was significant agreement between the DIF indexes found in Studies I and 2 for all 120 items analyzed. Some problems in replicating significant indexes of DIF are discussed.</abstract><relatedItem type="host"><titleInfo><title>Journal of Cross-Cultural Psychology</title></titleInfo><genre type="journal">journal</genre><identifier type="ISSN">0022-0221</identifier><identifier type="eISSN">1552-5422</identifier><identifier type="PublisherID">JCC</identifier><identifier type="PublisherID-hwp">spjcc</identifier><part><date>1993</date><detail type="volume"><caption>vol.</caption><number>24</number></detail><detail type="issue"><caption>no.</caption><number>2</number></detail><extent unit="pages"><start>133</start><end>148</end></extent></part></relatedItem><identifier type="istex">9A524158A96265D324ED0B68013F920B0449BE4C</identifier><identifier type="DOI">10.1177/0022022193242001</identifier><identifier type="ArticleID">10.1177_0022022193242001</identifier><recordInfo><recordContentSource>SAGE</recordContentSource></recordInfo></mods></metadata><serie></serie></istex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Rhénanie/explor/UnivTrevesV1/Data/Istex/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000426 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 000426 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Rhénanie
|area= UnivTrevesV1
|flux= Istex
|étape= Corpus
|type= RBID
|clé= ISTEX:9A524158A96265D324ED0B68013F920B0449BE4C
|texte= An Item Response Theory Evaluation of an English Version of the Trier Personality Inventory (TPI)
}}
| This area was generated with Dilib version V0.6.31. |  |