Introduction to Rasch measurement model (and ACER's conquest)

Scope / Abstract

This workshop aims to introduce participants to measurement theory and the Rasch Model (RM) for the construction, validation and calibration of measures. The interactive session will include a mix of theory and hands-on activities.

The RM is the parsimonious item response theory (IRT) model having one parameter for the person (ability), and only one parameter corresponding to each category of an item (threshold). It is the only IRT model in which the total score across items characterizes a person completely. Thus, the RM provides a proper method for nonlinear transformation of ordinal raw scores to linear measures, is an efficient tool for post-hoc analysis of an instrument and for development of new instruments.

Some models of the Rasch family (Dichotomous, Ordered Response Categories: Rating Scale and Partial Credit) are introduced and analysed through ACER’s CONQUEST4.13 software. Model fit including differential item functioning is presented.

This workshop will be of interest to practitioners and researchers engaged with measurement, questionnaire/test designs, and in making objective and meaningful inferences from the scale or construct used. Hands on activities include computing item statistics and ability estimates, plotting item characteristic curves, and generating plausible values to estimate population characteristics.

Biography

Jessica is currently a PhD candidate in the School of Public Health, where she is investigating musculoskeletal symptoms in musicians. She holds an Associate Diploma in Music, Bachelor of Physiotherapy with Honours, Bachelor of Music and Graduate Certificate in Clinical Epidemiology. Prior to commencing her PhD, Jessica worked as a researcher at the Universities of South Australia, Tasmania and Western Australia.

Requirements

Participant should have some basic knowledge and appreciation of statistics, and have access to a laptop for use during the workshop.

Software

ACER conquest 4.13