Rasch Measurement SIG Abstracts for AERA 1994

The family assessment interviewer's rating
Many-facet Rasch was used to investigate item, rater, rater-training and child/family effects for a new measure of special needs child and family functioning. Ten raters (5 well-trained and 5 less well-trained) evaluated 30 children/families on 103 items. Four scaleable measures were defined using 67 of the original items. Items tended to be relatively easy for the raters to agree with. Levels of training produced no observable difference. Clinical ratings, interviewers' ratings, and an external measure demonstrated moderate concurrent validity.
Kathy Green, Lucy Miller (5.15)

Step fit analysis with polytomously scored items
The "step fit" procedure is an extension of item-fit diagnostics to response categories. Step fit statistics are computed and plotted to display the magnitude and pattern of deviations for each step of an item. The procedure is applied to a large set of performance-based items to explore its usefulness and limitations.
Huixing Tang (11.55)

Validating Guttman scaling using Rasch modeling
The Work Keys criterion-referenced assessments of general work-place skills have been developed using Guttman scaling. The items have been organized into strata of increasing difficulty and used to categorize examinees according to skill proficiency. Since ordering items or persons deterministically with Guttman methods is challenging, Rasch modeling was used to validate the ordering. Satisfactory results were obtained.
S. Lee, M. Schulz, T. Vansickle, J. McLarty (11.55)

Establishing common quantitative units for different brands of instruments purported to measure the same variable
This paper explores the implications of scale-free measurement for the co-calibrations of instruments that are widely held to measure the same construct, but do so in different units of measurement due to slight variations in item content, rating scale structure, or the framework in which the observations are recorded. A sample co-calibration is presented, and comments are offered on this technique's potential for enhancing the study of commonly measured variables and for improving research design.
William P. Fisher, Jr., Richard F. Harvey, Karl M. Kilgore, Patricia Taylor, Carol Kelly (11.55)

DIF detection for judge-awarded ratings
In the Mantel-Haenszel procedure, the more approximate the matching of abilities within level, the more misleading can be the statistical results. Mean differences in the leniencies of the judges for reference and focal groups can be falsely reported as DIF on the performance items. Three Rasch-based techniques are proposed that do not require ability grouping: (1) separate analyses for reference and focal groups, (2) splitting suspect response strings within one analysis, (3) post-hoc item-group interaction estimation.
John Michael Linacre (11.55)

Facets of influence in a state-wide performance-based assessment
When each student is only rated once by one rater, equity requires that the influence on the student rating of rater severity, rating site and special status of the examinee be investigated. The ratings given 100 8th grade students randomly selected from each of 15 different sites were analyzed to obtain (1) calibrations of the items in each of the assessment areas, (2) calibrations of rater severity, (3) calibrations of the influence of rating site, (4) calibrations of the influence of English-language fluency.
Robert K. Hess (15.48)

Constructing rater and writing task banks for the assessment of written composition
Following the views of Choppin (1968) on item banks, a writing task bank is a set of calibrated prompts. A rater bank is a calibrated set of raters whose severity, reliability and validity have been systematically examined and cataloged. Guidelines for constructing these banks with many-facet Rasch methodology are presented. Illustrations come from field test data of the Georgia High School Writing Test.
George Engelhard Jr., B. Gordon, D. Curtin (15.48)

Computerized adaptive testing for medical licensure examinations: a comparison of different adaptive strategies
Several CAT simulations were performed using examinee response data from paper-and-pencil administrations of two medical licensure examinations. Results suggest that CAT may be a feasible alternative to current fixed-length examinations. Adaptive mastery testing produces very accurate pass/fail classifications, and most examinees are administered one third to one half of the items on a conventional examination.
Carol A. Morrison, Ronald J. Nungester (42.53)

Using the Rasch model to validate a district-wide, curriculum-based mathematics assessment
This study describes a technique designed to help districts establish the instructional growth of their students using curriculum-based tests. Using Rasch methodology, the performances of 3rd and 6th grade students, pre-instruction (Spring) and post-instruction (Fall), were located on a calibrated variable, rather than reported as grade-equivalents, percentiles, or percent of outcomes mastered. Item and person fit patterns provided further group and individual information. These fit patterns were usefully categorized as item-related, person-related, or person-item-interaction characteristics.
Robert K. Hess (42.53)

Measuring values to apply the Golden Rule
This paper proposes a research program that would prepare the ground for a political morality based on the Golden Rule. This requires some way of discovering that "what I do unto others" is the same as "what I would have done unto me." To discover this requires a measuring system that keeps things in proportion by showing what counts as "the same thing" for different people. This measuring system sets up analogies between people's values and what is valued. The measurement system is based on the specification that "my values are to one aspect of a situation what yours are to that or another aspect", and that proportions of this kind hold constant no matter what particular persons are addressed and no matter which aspects of the situation are involved.
William P. Fisher, Jr. (42.53)

Rasch Measurement SIG Abstracts for AERA 1994. Rasch Measurement Transactions 1994 7:4 p.322-3

Rasch Measurement SIG Abstracts for AERA 1994. … Rasch Measurement Transactions, 1994, 7:4 p.322-3

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/rmt74d.htm

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