File Download

There are no files associated with this item.

Supplementary

Conference Paper: Practical Consequences of Model Misfit in Assessing Academic Growth

TitlePractical Consequences of Model Misfit in Assessing Academic Growth
Authors
Issue Date2010
PublisherThe American Educational Research Association (AERA).
Citation
The Annual Meeting of the American Educational Research Association (AERA), Denver, Colorado, USA, 30 April-4 May, 2010 How to Cite?
AbstractThe study was to investigate the practical consequences of IRT model misfit in assessing achievement scores, academic growth, and pass rates in a typical state-wide assessment program. This study was designed to reflect common equating practices, and the major variable was model misfit. Results indicated that the typical level of model misfit would result in about 3% of the examinees being placed into different performance categories. The differences due to the choice of a non-fitting model are of major, practical consequence. The key messages from the study are that practical ways are available to study model fit, and, model fit or misfit can have practical consequences that should be considered when choosing an IRT model for a state testing program.
DescriptionConference Theme: Understanding Complex Ecologies in a Changing World
Session: Expanded Applications of Item Response Theory
Persistent Identifierhttp://hdl.handle.net/10722/187348

 

DC FieldValueLanguage
dc.contributor.authorZhao, Yen_US
dc.contributor.authorHambleton, R Ken_US
dc.date.accessioned2013-08-20T12:38:31Z-
dc.date.available2013-08-20T12:38:31Z-
dc.date.issued2010en_US
dc.identifier.citationThe Annual Meeting of the American Educational Research Association (AERA), Denver, Colorado, USA, 30 April-4 May, 2010en_US
dc.identifier.urihttp://hdl.handle.net/10722/187348-
dc.descriptionConference Theme: Understanding Complex Ecologies in a Changing World-
dc.descriptionSession: Expanded Applications of Item Response Theory-
dc.description.abstractThe study was to investigate the practical consequences of IRT model misfit in assessing achievement scores, academic growth, and pass rates in a typical state-wide assessment program. This study was designed to reflect common equating practices, and the major variable was model misfit. Results indicated that the typical level of model misfit would result in about 3% of the examinees being placed into different performance categories. The differences due to the choice of a non-fitting model are of major, practical consequence. The key messages from the study are that practical ways are available to study model fit, and, model fit or misfit can have practical consequences that should be considered when choosing an IRT model for a state testing program.-
dc.languageengen_US
dc.publisherThe American Educational Research Association (AERA).-
dc.relation.ispartofAnnual Meeting of the American Educational Research Association (AERA)en_US
dc.titlePractical Consequences of Model Misfit in Assessing Academic Growthen_US
dc.typeConference_Paperen_US
dc.identifier.hkuros216691en_US
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats