Type of Document	Dissertation
Author	Ingrisone II, James N.
URN	etd-11052008-150213
Title	Modeling The Joint Distribution Of Response Accuracy And Response Time
Degree	Doctor of Philosophy
Department	Educational Psychology and Learning Systems, Department of
Advisory Committee	Advisor Name Title Betsy Jane Becker Committee Chair Kai-Sheng Song Committee Co-Chair Akihito Kamata Committee Member Fred W. Huffer Outside Committee Member
Keywords	Item Response Theory Response Time Marginal Maximum Likelihood Maximum a Posteriori Weibull Distribution Joint Distribution
Date of Defense	2008-10-24
Availability	unrestricted
Abstract Making use of the observable response time data now available due to computerized testing offers exciting opportunities for the theory and practice of educational measurement. This study extends item response theory by developing a model that incorporates response time data into an IRT inspired model. A joint distribution is used to simultaneously model response accuracy and response time. The conditional distribution incorporates response time into a one parameter logistic model. The marginal distribution of response time uses a two-parameter Weibull distribution. Three simulation studies are conducted to evaluate the accuracy of the estimation techniques. Item difficulty parameters are estimated using marginal maximum likelihood (MML). Maximum a posteriori (MAP) estimation is used to estimate the latent ability and latent person speed parameters. The simulation studies illustrate that the estimation procedures recovered the item and person parameters quite well.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    IngrisoneJDissertation.pdf 499.13 Kb 00:02:18 00:01:11 00:01:02 00:00:31 00:00:02