Mapping Multi-Dimensionality

The physical universe is imagined to exist in 11 dimensions (and time) according to "string" or "(mem)brane" theory. Only three of these dimensions, length, depth, width, do we encounter directly. Very few can understand this complexity. But the mental universe of everyday experience is of a far higher dimensionality, and we encounter many of its dimensions simultaneously intertwined. Consider Figure 1 (excerpted from Figure 1 in De Mars C.E., "Measuring Higher Education Outcomes with a Multidimensional Rasch Model", Journal of Applied Measurement, 5, 4, 350-361, 2004).

Figure 1 shows student measures on two dimensions. Dimension 1 depicts the measures for the students on "American Experience" objectives and Dimension 2 on "Global Experience" objectives. Items 1-40 focus on American Experience and 41-78 on Global Experience.

Figure 1. Experience map in one dimension (DeMars, 2004)

The Paper reports the reliabilities of the student measures in the two dimensions of "Experience" to be .86 and .76, and the correlation between the student measures from separate analyses of the dimensions to be .64. This implies that the correlation between the dimensions, when disattenuated for measurement error, is .64 /?(.86*.76) = 0.78. This approximates 0.79, the correlation between the two dimensions in the joint multidimensional analysis. Thus analyzing the two Experiences together increases overall estimation accuracy. However, the accuracy for those students with atypical relationships between the dimensions may be reduced.

In wrestling with dimensional concepts, it is useful to consider parallel situations in physics. The two "Experience" variables are correlated .79. This is roughly the correlation of height and weight for growing children. So, looking at Figure 1 we could understand Dimension 1 to be height, and Dimension 2 to be weight. And items 1-40 to be marks on a height scale, and items 41-78 to be marks on a weight scale. When we don't know a child's weight, or measure it too sloppily, we can infer that weight from the child's height, and vice-versa. In fact, no matter how precisely we estimate a child's weight, the child's height estimate would indicate whether our weight estimate is on the heavy side (if the child is relatively short) or on the light side (if the child is relatively tall) and similarly for the child's height estimate. Further, our height and weight scales may be crude or uneven. In which case, the correlation between height and weight enables us to smooth out some of the irregularity in the scale markings. But, though these manipulations produce an overall improvement in accuracy, the accuracy for any particular child might be lessened.

On the map, however, these operational considerations, and the non-orthogonality of height and weight in our sample, must be left behind. There are two dimensions, so, to understand these data, a two dimensional map is needed. Figure 2 is Figure 1 redrawn with student locations simulated to match the distributions in Figure 1.

Figure 2. Experience map in two dimensions

On viewing Figure 2, an immediate question is, "Which students have substantively or statistically significantly different measures on the two dimensions?" Substantively, in many educational situations, one logit approximates one year of growth. Statistically, in the multi-dimensional analysis, the two sets of measures (and their standard errors) are not independent, so drawing confidence bands on Figure 2 is awkward. However those students whose two measures differ by more than a logit are statistically, and probably substantively, irregular.

John Michael Linacre

Mapping multidimensionality, DeMars C, Linacre J … Rasch Measurement Transactions, 2004, 18:3 p. 9990-991

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

To be emailed about new material on
please enter your email address here:

I want to Subscribe: & click below
I want to Unsubscribe: & click below

Please set your SPAM filter to accept emails from welcomes your comments:

Your email address (if you want us to reply):


ForumRasch 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,

Coming Rasch-related Events
June 23 - July 21, 2023, Fri.-Fri. On-line workshop: Practical Rasch Measurement - Further Topics (E. Smith, Winsteps),
Aug. 11 - Sept. 8, 2023, Fri.-Fri. On-line workshop: Many-Facet Rasch Measurement (E. Smith, Facets),


The URL of this page is