Evaluation 2009 Banner

Return to search form  

Contact emails are provided for one-to-one contact only and may not be used for mass emailing or group solicitations.

Session Title: Measurement Issues: Standards for Concept Mapping and Optimization of Item Response Theory
Multipaper Session 771 to be held in Suwannee 18 on Saturday, Nov 14, 10:55 AM to 11:40 AM
Sponsored by the AEA Conference Committee
Chair(s):
Ann Doucette,  George Washington University, doucette@gwu.edu
The Optimization of Item Response Theory Model's Type to Evaluating Practice in Context of Different Item Format
Presenter(s):
Marina Chelyshkova, State University of Management, mchelyshkova@mail.ru
Victor Zvonnikov, State University of Management, zvonnikov@mail.ru
Abstract: In paper we represent the results of the researches which have been carried out within the project: 'Comparative Efficiency of Parametric and Nonparametric Item Response Theory Models for Combination Different Item Formats in Tests'. The goal of this project was connected with choice the optimum models for scoring examination's data in conditions of integration the quantitative and qualitative scores, which were obtained by varying formats of test items. The opportunities of models were compared by two basic criteria: high objectivity of measurement and high comparability of graduate's scores in different tests variants. Test information functions were used as basis of score's comparison for different models. The analysis has shown, that the Monotone Homogeneity Model of Nonparametric Item Response Theory is the best for scaling. As the results of the work the theoretical and technology requirements developing measurements in evaluation are formulated.
Establishing Standards in Concept Mapping: A Meta-review and Analysis
Presenter(s):
Scott Rosas, Concept Systems Inc, srosas@conceptsystems.com
Abstract: The use of concept mapping for planning, evaluation, and research has expanded considerably in recent years. Concept mapping, an applied multivariate methodology that integrates familiar qualitative group processes with multivariate statistical analyses encompasses six distinct phases: preparation, generation, organization, representation, interpretation, and utilization. Due to the emergent interest in and the variety of application of the concept mapping methodology in research and evaluation, it is vital to define rigorous and feasible standards of quality across all phases of the process. As an initial step in the development of benchmarks, a systematic meta-review of 30 concept mapping projects was conducted and data generated within each phase of the process was analyzed. The findings serve as a preliminary set of standards by which results from future concept mapping studies can be measured or judged. The implications of these findings are discussed relative to methodological expectations for concept mapping and mixed method approaches.

 Return to Evaluation 2009

Add to Custom Program