2nd International Conference on Measurement
in Health, Education, Psychology and Marketing: Developments with Rasch models
Murdoch University, Western Australia, 2004

A personal recollection of the Health Sciences sessions.

It is always a pleasure to visit Perth in the summer time, with the Indian Ocean and white sands to tempt one away from serious business (not to mention the Murdoch Club and its delicious lattes!). Nevertheless, the conference program was such that a full program of Rasch applications in Health kept ones full attention. Although it may be mentioned elsewhere, the opening plenary session by Professor David Andrich was based on health data, looking at female maturation. The data was provided by another speaker, Dr Stef van Buuren, from TNO in the Netherlands. David's point was methodological, looking at the difference between statistical approaches whereby the object was to combine items to provide a Maturation Index, and the measurement perspective, which looked at how an item could be used to inform on group differences, rather than be included in the Index. The presentation, using RUMM2020, laid out the inherent tensions which exist in much health outcome work, nicely highlighting the potential differences between measurement and explanation.

The range of applications in health was then illustrated by two presentations in the first parallel session. Carlyne Arnould from the Universite Catholique de Louvain presented her work on the development of the Abilhand-Kids questionnaire to measure manual ability in children with cerebral palsy. Taking an existing scale for adults as a starting point, the presentation reported on its adaptation, reliability and validity for use with children. There were some differences in the response to items made by the children and their parents (which is not unusual and has been reported in other measures where both perspectives are obtained), but, using Winsteps, the data fitted the Rasch model and good test-retest reliability was achieved (r=0.87). Following this Dr Stef van Buuren from TNO Prevention and Health in the Netherlands presented his work testing Rasch analysis to the extreme by taking data from many different countries, forming an (extremely) ill-conditioned data set, and forging links between countries (sometimes by using a special bridging study) to bring all the various country data together. The approach consisted of two steps, constructing the data matrix, taking items from national disability studies for constructs such as dressing or walking, then creating a conversion key using threshold estimates (from RUMM2020) to link all the items (for a single construct such as dressing) together.

In the next session, Ann Bjorkdahl from Goteborg University in Sweden presented her work on the European Brain Injury Questionnaire in stroke patients. Using Winsteps, she took data from 59 patients with median age 53 years, and looked at the 63 item scale. With an acceptable fit range of 0.6-1.4, seven items were found to misfit but all items showed stability over time. The Tukey Mean Difference plot was used, which is the same as the Bland & Altman plot commonly found in health studies. Irene Styles from Murdoch University then presented her work on the International Classification of Functioning Disability and Health (ICF) qualifier scale. A checklist of 48 impairments determined what were the patient's problems, and what were the extent of those problems. Data from patients with low back pain, breast cancer and stroke were fitted to the rating scale model. Although data tended to fit the model within diagnosis, extensive DIF was found across conditions. Elizabeth Betemps from the University of Cincinnati opened the next session with her work on a 13 item Psychiatric Distress Scale, looking at the difference in magnitude of effect by raw scores compared with transformed measures. Based upon 54 subjects on admission and discharge from an outpatient psychiatric treatment program in Ohio, USA, significantly more patients were found to have improved at discharge using the measure as opposed to the raw score. Eric Wong from the Chinese University of Hong Kong then presented his work on a sample of 1956 patients from two stroke databases, comparing the Functional Independence Measure (FIM) with the Barthel Index. With mean age of 72 years; 51% male and 51% with left hemi-paresis, data from the scales were fitted to the partial credit model using Winsteps. Bladder and bowel items were found to misfit on both scales, and the easiest and most difficult items were also consistent. Both measures showed invariance of items across admission and discharge. However, the FIM motor scale showed greater precision (person separation) than the Barthel Index.

The FIM in stroke and head injury formed the next joint presentation between Asa Lundgren Nilsson from Goteborg University in Sweden and Anita Slade for Sheffield Hallam University in the UK. The thrust of their presentation was similar to the discussion elsewhere in this issue with regard to the disordering of thresholds and DIF by country where pooled international data is required. For the stroke data, there were 2546 patients at admission from six countries. For head injury, 779 patients from 5 countries. The pattern of analysis was consistent across diagnosis; initially with considerable disordering of thresholds, a complex solution involving splitting of items was required to facilitate the pooling of data across countries.

Finally the author presented his work on the development of item banking for quality of life (QoL) across the rheumatic diseases. Using a strong theoretical model of QoL, namely the Needs-Based model of Hunt and McKenna, and the strong mathematical model of Rasch, an example of building an item bank was given for Psoriatic Arthritis and Systemic Lupus Erythematosus. The basis of this was common item equating, ensuring that links were free of DIF by diagnosis, and that the entire set gave adequate fit to the Rasch model.

Alan Tennant
Professor of Rehabilitation Studies
The University of Leeds. UK.

The three most interesting presentations at 2ICOM:

Although my stay at Perth was very short (from Tuesday morning to Wednesday evening), I enjoyed every minute of my stay in the conference sessions. Among the presentations which I attended (all of which are related to Education, because I am an educator), I found the following three most interesting and useful for my future research. They are:

1) Ted Brown's "The Scalability & Validity of Four Pediatric Visual Perceptual Instruments: A Comparison Using the Rasch Measurement Model", which gave me a hint to reconsider the construct validity of a language test.

2) Juho Loover's "Using Modern Psychometric Theory to Identify Differential Item Functioning in Polytomously-Scored Constructed-Response Items", which provided me with an illustrative methodology that can be applied to any examination which contains polytomously-scored items.

3) Sergij Garbrscek's "Taking another perspective: Matura examinations in Slovenia", which introduced me to a general ideal of Matura examinations in Slovenia with statistical and educational information.

Yuji Nakamura
Tokyo Keizai (Economics) University

Reports on the 2nd International Conference on Measurement in Health, Education, Psychology and Marketing, Tennant A., Nakamura Y. … Rasch Measurement Transactions, 2004, 17:4 p.956-957

Rasch Publications
Rasch Measurement Transactions (free, online)	Rasch Measurement research papers (free, online)	Probabilistic Models for Some Intelligence and Attainment Tests, Georg Rasch	Applying the Rasch Model 3rd. Ed., Bond & Fox	Best Test Design, Wright & Stone
Rating Scale Analysis, Wright & Masters	Introduction to Rasch Measurement, E. Smith & R. Smith	Introduction to Many-Facet Rasch Measurement, Thomas Eckes	Invariant Measurement: Using Rasch Models in the Social, Behavioral, and Health Sciences, George Engelhard, Jr.	Statistical Analyses for Language Testers, Rita Green
Rasch Models: Foundations, Recent Developments, and Applications, Fischer & Molenaar	Journal of Applied Measurement	Rasch models for measurement, David Andrich	Constructing Measures, Mark Wilson	Rasch Analysis in the Human Sciences, Boone, Stave, Yale
in Spanish:	Análisis de Rasch para todos, Agustín Tristán		Mediciones, Posicionamientos y Diagnósticos Competitivos, Juan Ramón Oreja Rodríguez

Forum

Rasch Measurement Forum to discuss any Rasch-related topic

Go to Top of Page
Go to index of all Rasch Measurement Transactions
AERA members: Join the Rasch Measurement SIG and receive the printed version of RMT
Some back issues of RMT are available as bound volumes
Subscribe to Journal of Applied Measurement

Go to Institute for Objective Measurement Home Page. The Rasch Measurement SIG (AERA) thanks the Institute for Objective Measurement for inviting the publication of Rasch Measurement Transactions on the Institute's website, www.rasch.org.

The URL of this page is www.rasch.org/rmt/rmt174f.htm

Website: www.rasch.org/rmt/contents.htm