Multiple Regression via Measurement

Kyle Perkins sent me physical dimensions, blood chemistries and diagnoses for 96 clinically-relevant patients, half of whom were thought to have gout. Since the precise relationship between the physical indicators and their medical implications is not known, I linearly transformed the original 3 physical and 5 chemical metrics into rating scales with ten categories, 0-9. The 5 diagnoses, reflecting expert (but imperfect) medical opinion, were coded as present=1, absent=0.

My next step was to discover which pieces of this heterogeneous collection would cooperate together to tell a meaningful story. From the necessarily rough ordinal data, I constructed Rasch measures.

Factor analysis of the residuals reported that blood creatinine, uric acid and urea nitrogen clustered with diagnoses of gout, hypertension and diuretic. Excluded were triglycerides, cholesterol, height, weight, surface area and the diagnoses of diabetes and kidney stones.

This led to the specification of a blood chemistry variable, based on levels of creatinine, uric acid and urea nitrogen, on which I could regress all five diagnoses. This was easily done by (1) calibrating these three items by themselves along with their rating scales; (2) anchoring these calibrations and the matching person measures, and (3) introducing gout and other diagnostic dichotomies into the analysis. [Later, Ben simplified (2) and (3) by using zero-item-weights for the non-measurement variables.]

The first results bear on the diagnosis of gout. Figure 1, "The Complete Story in One Picture", shows how well my new 3-blood-chemistry variable predicts a gout diagnosis. This Figure has the meaning of a conventional "multiple regression" in which gout is regressed on blood chemistry, but without the usual statistical obfuscation.

In Figure 1, there is the usual and unavoidable region of uncertainty, here between measures of 48 and 59. Otherwise the discrimination of gout/not gout is quite clear.

Investigation of the two "gout" patients with measures of 16 and 39 raise strong doubts about the accuracy of their recorded diagnoses. Neither has any blood chemistry evidence of gout.

Investigation of the two "no gout" patients with measures of 66 and 64 urge reconsideration of their diagnoses, since both have blood chemistries indicative of gout.

The measurement-based "regression" analysis laid out in Figure 1 was also done for the other four diagnoses. In fact, since the blood chemistry items, rating scales and person measures are all anchored, multiple "multiple regressions" can be done simultaneously.

These results are listed in Table 1, Rasch "Multiple Multiple Regression". The "multiple regression" correlations are simply the correlations between the scores on the diagnostic items and the patient measures. Gout=.61, Hypertension=.51, Diuretic=.39, Kidney Stone=-.03 and Diabetes=-.06 are clear enough. But much more accessible and useful are the diagnostic measure values and their explicit regions of uncertainty for the clinically-relevant patients summarized in Table 2.

Gout turns on at 59 with doubt down to 48.

Hypertension turns on at 59 with doubt down to 49.

Diuretic turns on at 61 with doubt down to 52.

Kidney Stones and Diabetes cannot be predicted from this blood chemistry variable.

Analyses like this could be run on all blood chemistry data for all diagnoses. Then relationships could be detected and variables constructed to implement all predictable diagnoses. The indicative levels for each diagnosis could be updated continually to focus on local and current practice and to keep pace with changing ways.

How simple, convenient, timely and useful!

Benjamin D. Wright with K. Perkins and K. Dorsey, Southern Illinois University

Multiple regression via measurement. Wright B.D. … Rasch Measurement Transactions, 2000, 14:1 p.729

Rasch Books and Publications
Invariant Measurement: Using Rasch Models in the Social, Behavioral, and Health Sciences, 2nd Edn. George Engelhard, Jr. & Jue Wang	Applying the Rasch Model (Winsteps, Facets) 4th Ed., Bond, Yan, Heene	Advances in Rasch Analyses in the Human Sciences (Winsteps, Facets) 1st Ed., Boone, Staver	Advances in Applications of Rasch Measurement in Science Education, X. Liu & W. J. Boone	Rasch Analysis in the Human Sciences (Winsteps) Boone, Staver, Yale
Introduction to Many-Facet Rasch Measurement (Facets), Thomas Eckes	Statistical Analyses for Language Testers (Facets), Rita Green	Invariant Measurement with Raters and Rating Scales: Rasch Models for Rater-Mediated Assessments (Facets), George Engelhard, Jr. & Stefanie Wind	Aplicação do Modelo de Rasch (Português), de Bond, Trevor G., Fox, Christine M	Appliquer le modèle de Rasch: Défis et pistes de solution (Winsteps) E. Dionne, S. Béland
Exploring Rating Scale Functioning for Survey Research (R, Facets), Stefanie Wind	Rasch Measurement: Applications, Khine	Winsteps Tutorials - free Facets Tutorials - free	Many-Facet Rasch Measurement (Facets) - free, J.M. Linacre	Fairness, Justice and Language Assessment (Winsteps, Facets), McNamara, Knoch, Fan
Other Rasch-Related Resources: Rasch Measurement YouTube Channel
Rasch Measurement Transactions & Rasch Measurement research papers - free	An Introduction to the Rasch Model with Examples in R (eRm, etc.), Debelak, Strobl, Zeigenfuse	Rasch Measurement Theory Analysis in R, Wind, Hua	Applying the Rasch Model in Social Sciences Using R, Lamprianou	El modelo métrico de Rasch: Fundamentación, implementación e interpretación de la medida en ciencias sociales (Spanish Edition), Manuel González-Montesinos M.
Rasch Models: Foundations, Recent Developments, and Applications, Fischer & Molenaar	Probabilistic Models for Some Intelligence and Attainment Tests, Georg Rasch	Rasch Models for Measurement, David Andrich	Constructing Measures, Mark Wilson	Best Test Design - free, Wright & Stone Rating Scale Analysis - free, Wright & Masters
Virtual Standard Setting: Setting Cut Scores, Charalambos Kollias	Diseño de Mejores Pruebas - free, Spanish Best Test Design	A Course in Rasch Measurement Theory, Andrich, Marais	Rasch Models in Health, Christensen, Kreiner, Mesba	Multivariate and Mixture Distribution Rasch Models, von Davier, Carstensen

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/rmt141a.htm

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