Academic journal article Research Quarterly for Exercise and Sport

Calibration of the Barrier Instrument Using the Rasch Rating Scaling Model

Academic journal article Research Quarterly for Exercise and Sport

Calibration of the Barrier Instrument Using the Rasch Rating Scaling Model

Article excerpt

Weimo Zhu, University of Illinois at Urbana-Champaign

Most of existing exercise barrier scales are constructed by measurement models based on classical testing theory (CTT), which suffers from several psychometric disadvantages. For example, the calibration of barriers and personal measures under CTT is dependent on each other, which has made it very difficult to generalize calibrations from one data set to another. The purpose of this presentation is to describe how an advanced measurement model was applied to the calibration of the modified exercise barrier instrument and examine related psychometric advantages. More specifically, the collected response scores were analyzed using the Rasch Rating Scale model, with both barriers and person measures positively defined (i.e., the larger the raw scores, the higher the logits, therefore, representing more perceived barriers). The model-data fit was examined by Infit and Outfit mean-square statistics (i.e., a value [less than] .7 or [greater than] 1.3 was considered not fit), which were provided by FACETS, a Rasch c alibration computer software. It was found that overall the model fit data very well. Except one item whose Outfit = 1.4, all Infit and Outfit statistics were close to one. After the calibration, the severity of barriers and individuals' perceived barriers can be evaluated on the same scale. The top 10 most severe harriers were: "lack self-discipline" (logits = 1. …

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.