Measurement with the Rasch Model

Description:

Outline

This two-day course aims to introduce participants to measurement theory and the Rasch model for construction and validation of measures. It covers the basic theory behind measurement, from an Item Response Theory perspective, focusing on the assumptions of the Rasch models, in particular.  The Rasch model provides the means to create measures (or score scales) from a combination of items in tests or questionnaires.  The principles governing the application of such models are shown through examples from educational measurement but are easily applicable to other areas in social and health sciences. Participants will have the chance to practice with various models of the Rasch family (Dichotomous, Rating Scale and Partial Credit) with specialised software (Winsteps).

Objectives

This two day course aims to introduce participants to measurement theory and the Rasch model for measures construction and validation (Day 1). Participants will get hands on experience with analysis and interpretation of the Dichotomous (Day 1), Partial Credit and Rating Scale (Day 2) Rasch models, with specialised software as well as packages freely available in R. 

The course will be delivered with a mixture of lectures and practical sessions, around the following themes:

• Introduction to (Rasch) Measurement
• The Dichotomous Rasch Model (with software application)
• The basics of the validation framework with the Rasch Model
• Examples and applications with the Rating Scale Model (which is appropriate for Likert type items) and the Partial Credit Model (which allows for items with different number of response categories to be analysed together). 
• Differential Item Functioning, Optimal functioning of rating scales and dimensionality. 

Prerequisites

Participants should have some basic knowledge of introductory statistics. Some familiarity or previous experience with syntax (commands) in statistical packages will also be useful. Examples of data sets from various contexts will be provided; however, participants might also bring their own datasets and problems to analyse, possibly with a view to preparing a validation report or paper for publication. In order to ensure that the datasets are appropriate for such analyses participants are advised to contact Maria Pampaka (at maria.pampaka@manchester.ac.uk) in advance of the course. 

Cost:

£195 (£140 for those from educational, government and charitable institutions).

Website and registration:

https://www.cmist.manchester.ac.uk/study/short/booking/

Region:

North West

Keywords:

Survey Research, Mixed Methods, Data Management , Quality in Quantitative Research, Regression Methods, Rasch modelling, Latent Variable Models, Rasch models, Quantitative Approaches (other), Mixed Methods Data Handling and Data Analysis, Winsteps , Rasch , Dichotomous Rasch Model

Related publications and presentations:

Survey Research
Mixed Methods
Data Management
Quality in Quantitative Research
Regression Methods
Rasch modelling
Latent Variable Models
Rasch models
Quantitative Approaches (other)
Mixed Methods Data Handling and Data Analysis

Back to archive...