Hambleton speaks at two national conferences
Professor Ronald Hambleton has recently given two invited addresses at national conferences in October. At the Winemiller 2006 Conference on Methodological Developments of Statistics in the Social Sciences at the University of Missouri in Columbia his address was entitled, "Item Response Theory Models and Methods for the Social Sciences."
At the Family and Health Sciences Methods Conference at the University of Massachusetts, Professor Hambleton spoke on "Applications of Item Response Theory to Improve Health Outcomes Assessment". Information about Professor Hambleton's talk can be found here, and below is the abstract for this presentation.
"The models, methods, and practices associated with item response theory, researched and developed heavily over the last 30 years in the field of educational testing to overcome many important shortcomings of traditional psychometric methods, are now being extended to assessment in the social and health sciences. After a review of classical measurement models and methods and their shortcomings, IRT models and their properties will be introduced. Of special interest in the presentation will be several of the models and their important features that seem especially useful to social and health scientists. The application of these models to health outcomes assessment will be illustrated with an example drawn from responses to 50 items that constitute the Patient Satisfaction with Medical Care Questionnaire (developed by RAND for the Health Outcomes Study). Respondents had chronic medical conditions such as diabetes, depression, and heart disease. This application highlights the critical role of IRT modeling in four areas: (1) selecting the best IRT model based on the characteristics of the data; (2) developing an instrument to match the population of interest for the study; (3) linking two or more instruments measuring similar constructs on a common metric or scale; and (4) creating item banks that provide the foundation for short fixed-form and computer adaptively administered instruments. The overall goal of the presentation is to highlight recent advances in measurement theory that can help researchers to develop assessment instruments that meet the criteria for validity, reliability, and sensitivity to change in health status, while minimizing the burden of long questionnaires for respondents."