|
Welcome to the memories page for IMPS 2010. If you were able to attend this conference, we hope you found it both educational and enjoyable. If you were not able to attend this conference, we hope to see you at future ones! The program is available here. If you would like to add your presentation, photographs, or news article about IMPS 2010 simply email it to pmetrika@uncg.edu. We would like to extend our sincere congratulations all of our prize winners for 2010.
Our thanks to everyone who came to and participated in IMPS 2010! We hope you had a great time, and hope to see you again in Hong Kong for IMPS 2011. Pictures of some of the individual attendees are posted below. If you recognize any unlabelled photographs please email pmetrika@uncg.edu with the name!
|
| IMPS 2010 Memories |
|
Annual Meeting
- 2011 Memories
- 2010 Memories
- 2009 Memories
- 2008 Memories
- 2007 Memories
- 2006 Memories
- 2005 Memories
- 2004 Memories
- 2003 Memories
- 2010 Memories
- 2009 Memories
- 2008 Memories
- 2007 Memories
- 2006 Memories
- 2005 Memories
- 2004 Memories
- 2003 Memories
For Members
Inferences about test takers in educational testing have been primarily based on their responses to the items. Information from the time needed to complete an item has been largely ignored because the recording of response times was unpractical with paper-and-pencil tests. Nowadays, computer based testing makes the collection of response times straightforward. Statistical methods will be discussed for the modeling of responses and response times on test items. It will be shown how separate measurement models for ability and speed of test takers can be used to describe the observed data. Subsequently, multivariate and multilevel model extensions account for dependencies between responses and response times and allow for the incorporation of covariates on both the person and the item side of the modeling framework. The flexibility of Bayesian Markov Chain Monte Carlo methods makes joint estimation of all model parameters feasible. Using real data examples, possibilities for making joint inferences from responses and response times are presented. 
Researchers are often interested in finding types or classes of people based on personality characteristics. This study compares several approaches to identifying classes using the personality items from the National Longitudinal Study of Adolescent Health (Add Health). First, the personality items were assigned to scales using exploratory and confirmatory factor analysis. Second, sum scores and IRT scores for each scale were calculated. Finally, mixture models were used to identify classes of personality traits for both the sum scores and IRT scores and the results were compared to the results from applying a factor mixture model directly to the latent personality factors. Implications for applied researchers are discussed.
In this work, the use of empirical prior information in computerized adaptive testing (CAT) is investigated. Besides the individual responses, background variables concerning the examinees may be available and can be introduced in the estimation process. Whenever a strong relationship between the ability and the covariates is identified, it is possible to include collateral information both in the initialization and in the ability estimation process within CAT. Commonly, item selection in CAT is performed by adopting the maximum-information criterion. A serious consequence of applying this criterion is over exposure of the first item. Moreover, while research has been oriented towards using CAT to shorten the test length, adaptive testing shows a peculiar weakness especially when dealing with short tests. In the current work, a simulation study has been carried out in order to compare the accuracy of the ability estimates in case prior information is included or not. Special attention is given to the case of short tests. The findings show that, when empirical information is introduced, ability estimates are more precise, i.e. mean standard deviations are lower. Furthermore, the introduction of individual prior information reduces the item over-exposure, because the ability is initialized with reference to the examinee's covariates.
|
To report broken links, please contact Gwen Exner at webmaster@psychometrika.org