Simulating Data from Marginal Scores

Simulating Rasch datasets from known item difficulties and person abilities is straightforward www.rasch.org/rmt/rmt213a.htm

More challenging is simulating Rasch datasets with known marginal scores. Verhelst et al. (2007) achieve this for complete rectangular dichotomous datasets using a Markov-Chain Monte-Carlo (MCMC) algorithm. Its operates on dichotomous data by shuffling observations between rows and columns. The program is implemented as an R package, RaschSampler, which can simulate matrices of up to 1024 rows and 64 columns.

Another approach is to simulate the data using Rasch measures. The dataset can be dichotomous or polytomous, complete or incomplete, and of any size.

1. Compute the marginal scores (persons, items) and counts (polytomous categories) in the generating dataset.

2. In the simulated dataset, impute the extreme observations corresponding to all extreme marginal scores in the generating dataset. Flag those observations as missing in the generating dataset. If there were extreme marginal scores, repeat from 1.

3. Estimate the measures (persons, items, polytomies) for the generating dataset. Rough estimates are good enough. Polytomies: allow for intermediate categories with sampling zeroes.

4. Select a non-missing observation at random in the generating dataset. Simulate its value based on the estimated measures and place it in the matching location in the simulated dataset. Flag that data-point as missing in the generating dataset.

5. Repeat the procedure from 1 until there are no active observations in the original dataset. For dichotomous data, only the measures for one person and one item will need to be re-estimated for each simulated observation. For polytomous data all measures may need to be re-estimated.

When this procedure completes, the simulated dataset will have the same marginal scores and category counts as the original generating dataset.

John Michael Linacre

Verhelst N., Hatzinger R., Mair R. (2007) The Rasch Sampler. Journal of Statistical Software, 20:6. www.jstatsoft.org/v20/i04

Simulating Data from Marginal Scores … J.M. Linacre, Rasch Measurement Transactions, 2008, 22:2, 1168

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/rmt222h.htm

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