Academic journal article Research Quarterly for Exercise and Sport

Calibrating Portfolios to Make the Scores Meaningful

Academic journal article Research Quarterly for Exercise and Sport

Calibrating Portfolios to Make the Scores Meaningful

Article excerpt

Weimo Zhu, University of Illinois at Urbana-Champaign

Since scoring rubrics in portfolios are often set on an ordinal scale, total scores generated from the scale may be misleading. Fortunately, ordinal scores can be transformed onto an interval scale by applying the Rasch calibration, an approach based on an advanced measurement model. The purpose of this presentation is to describe how the many-faceted Rasch model was applied to the calibration of nine motor skill knowledge portfolios developed and examine related advantages and disadvantages. The collected data on these portfolios (N = 1,940) were analyzed by the many-faceted Rasch model, with three defined facets: Portfolios, scoring rubrics, and students. The model-data fit was examined using Infit and Outfit mean-square statistics, with a value of 1 being considered as fit and values greater than 1.3 or less than .7 as unfit. It was found that overall the model fit the data well, with all Infit and Outfit statistics being close to one. One scoring rubric (Scoring Rubric 2 of Portfolio 1), however, was foun d misfit (Infit = 1.4 and Outfit = 2.0). Further, three of 15 scoring rubrics revealed disordered rating categories. After these scoring rubrics were recategorized, the scores were re-analyzed using the Rasch model. Again, the model fit data well and the problems of disordered categories were corrected. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.