Guidelines for reporting studies using Rasch analysis

Over the past decade the use of the Rasch measurement model has become widespread in rehabilitation, either in the development of new scales, or in reviewing existing scales (1).  Fitting data to the model, the process known as Rasch analysis, is now well established.  Consequently, when writing a paper which includes Rasch analysis, it is no longer necessary to give a full explanation of the mathematical basis of the model, nor a detailed explanation of the process of Rasch analysis. Rather refer readers to published texts which describe the models and the process (1–7).

For articles using Rasch analysis, readers will require a certain amount of information to allow them to judge the quality of the analysis presented, just as is required, for example, with any statistical procedure. Consequently authors will need to specify which parameterisation of the model was chosen, whether thresholds in the polytomous model were re-ordered, whether local response dependency, unidimensionality and Differential Item Functioning was examined and, if so, what values were used to specify satisfactory levels of these attributes, and what procedures, if any, were used to deal with any problems.  Likewise, with fit of items, the acceptable fit levels need to be specified, and what strategies were used in the presence of misfit. In this context, it is possible that some of the ideal values associated with, for example, fit may be incorporated into Tables, with a text explanation to say that these are specified in the Table. Reliability and targeting should also be reported, along with justification for the sample size. Always specify the software package used.

Any matters that may relate to issues of debate, e.g. fit range, may be appropriately placed in the discussion.

1.     Rasch G. Probabilistic models for some intelligence and attainment tests. Copenhagen: Danish Institution for Educational Research; 1960.

2.     Andrich D. Rasch models for measurement. Newbury Park CA: Sage, 1988.

3.     Tesio L. Measuring behaviours and perceptions: Rasch analysis as a tool for rehabilitation research. J Rehabil Med 2003; 35: 105–115.

4.     Tennant A, Conaghan PG. The Rasch Measurement Model in Rheumatology: What is it and why use it? When should it be applied, and what should one look for in a Rasch paper? Arthritis Rheum 2007; 57: 1358–1362.

5.     Bond TG, Fox CM. Applying the Rasch model: fundamental measurement in the human sciences, 2nd edn. Mahwah: Lawrence Erlbaum Associates; 2007.

6.     Hagquist C, Bruce M, Gustavsson JP. Using the Rasch model in nursing research: an introduction and illustrative example. Int J Nurs  Stud 2009; 3: 380–393.

7.     Ehlan AH, Kucukdeveci AA, Tennant A. The Rasch Measurement Model. In: Franco Franchignoni (Ed). Research Issues in Physical & Rehabilitation Medicine. Pavia: Maugeri Foundation, p. 89–102, 2010.

 

These guidelines have been prepared by Alan Tennant, Department of Rehabilitation Medicine, The University of Leeds, Faculty of Medicine and Health, The General
Infirmary at Leeds,
Leeds, UK