File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Improving person-fit assessment by correcting the ability estimate and its reference distribution

TitleImproving person-fit assessment by correcting the ability estimate and its reference distribution
Authors
Issue Date2008
Citation
Journal of Educational Measurement, 2008, v. 45, n. 2, p. 159-177 How to Cite?
AbstractThe standardized log-likelihood of a response vector (lz) is a popular IRT-based person-fit test statistic for identifying model-misfitting response patterns. Traditional use of lz is overly conservative in detecting aberrance due to its incorrect assumption regarding its theoretical null distribution. This study proposes a method for improving the accuracy of person-fit analysis using lz which takes into account test unreliability when estimating the ability and constructs the distribution for each lz through resampling methods. The Type I error and power (or detection rate) of the proposed method were examined at different test lengths, ability levels, and nominal α levels along with other methods, and power to detect three types of aberrance - cheating, lack of motivation, and speeding - was considered. Results indicate that the proposed method is a viable and promising approach. It has Type I error rates close to the nominal value for most ability levels and reasonably good power. © 2008 by the National Council on Measurement in Education.
Persistent Identifierhttp://hdl.handle.net/10722/228059
ISSN
2015 Impact Factor: 1.528
2015 SCImago Journal Rankings: 2.067

 

DC FieldValueLanguage
dc.contributor.authorDe La Torre, Jimmy-
dc.contributor.authorDeng, Weiling-
dc.date.accessioned2016-08-01T06:45:05Z-
dc.date.available2016-08-01T06:45:05Z-
dc.date.issued2008-
dc.identifier.citationJournal of Educational Measurement, 2008, v. 45, n. 2, p. 159-177-
dc.identifier.issn0022-0655-
dc.identifier.urihttp://hdl.handle.net/10722/228059-
dc.description.abstractThe standardized log-likelihood of a response vector (lz) is a popular IRT-based person-fit test statistic for identifying model-misfitting response patterns. Traditional use of lz is overly conservative in detecting aberrance due to its incorrect assumption regarding its theoretical null distribution. This study proposes a method for improving the accuracy of person-fit analysis using lz which takes into account test unreliability when estimating the ability and constructs the distribution for each lz through resampling methods. The Type I error and power (or detection rate) of the proposed method were examined at different test lengths, ability levels, and nominal α levels along with other methods, and power to detect three types of aberrance - cheating, lack of motivation, and speeding - was considered. Results indicate that the proposed method is a viable and promising approach. It has Type I error rates close to the nominal value for most ability levels and reasonably good power. © 2008 by the National Council on Measurement in Education.-
dc.languageeng-
dc.relation.ispartofJournal of Educational Measurement-
dc.titleImproving person-fit assessment by correcting the ability estimate and its reference distribution-
dc.typeArticle-
dc.description.natureLink_to_subscribed_fulltext-
dc.identifier.doi10.1111/j.1745-3984.2008.00058.x-
dc.identifier.scopuseid_2-s2.0-43949138622-
dc.identifier.volume45-
dc.identifier.issue2-
dc.identifier.spage159-
dc.identifier.epage177-
dc.identifier.eissn1745-3984-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats