Vertical Equating and Item Selection

George Ingebo believes that bank calibrations need to be all-purpose values. I have a specific purpose in mind with vertical test equating -- to evaluate a particular sequence of curriculum material. I have no intention of accounting for third graders' performance on eighth grade tests. If the scale is used to place new students in the curriculum sequence, I would only use the students' answers to items that were within a reasonable range of each student's estimated position on the scale.

All models engage data uncritically. Evaluation of model fit, or appropriateness of items for linking, comes after the model has been applied. Poor linking items are identified after an MFORMS analysis through the study of residuals and item fit. The method differs from looking at a plot of paired link-item calibrations, but the result is the same. Either way, the "noisy" items can be detected, deleted, and the analysis rerun. More work needs to be done to find the relationship between noise, as revealed by a plot of paired item calibrations (multistep procedure) and noise as revealed through the study of (one-step, concurrent) residuals.

In one sense, the removal of noisy items from the test linking data may be immaterial. Consider the plot of paired item calibrations. One tends to reject outlying items from both sides of the identity line, so the mean difference in difficulty between tests is hardly changed -- neither does the resulting scale except for the absence of the outlying items. If there is no "right" item for linking tests at different grade levels -- which is what some might argue -- we should not be much comforted by the removal of a few noisy items.

There are problems with vertical test equating, but I think it is too soon to give up. A vertically equated scale has a great deal of potential. How else are we to compare the difficulty of curricular material across grade levels and track growth in student ability? The extent to which the scale can be linear and unidimensional across grade levels is a matter for empirical research -- we have to construct scales in order to evaluate them.

Vertical equating and item selection. Schulz EM. … Rasch Measurement Transactions 2:2 p.18

Vertical equating and item selection. Schulz EM. … Rasch Measurement Transactions, 1988, 2:2 p.18

Please help with Standard Dataset 4: Andrich Rating Scale Model

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/rmt22b.htm

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