Recoding and Pivoting: An Example

In a study of personal efficacy, a seven category rating scale was employed from "1", "Strongly Agree", to "7", "Strongly Disagree". The data are summarized in the Table. Two types of items were used, positively worded, "I like most parts of my personality", and negatively worded, "The demands of everyday life often get me down." Since I wanted to measure respondents on one variable of "personal efficacy" on which a higher score was to indicate a higher measure, I reversed the scoring of the positive items. Now a rating of "7" always indicates greatest efficacy.

Even with all rating scales oriented in the same direction, the construct definition (item ordering along the variable) was vague. Category fit statistics (RMT 9:3 450-451) signaled that respondents had been asked to discriminate an excessive number of levels of agreement.

Consideration of the categories, together with inspection of category counts and fit statistics, suggested that treating the central "Don't Know" category as missing data would remove substantial irrelevant off-variable behavior. So I omitted this category. Combining other categories in an intelligible way to maximize data-to-model fit and measurement efficiency produced the recodings in table 2. For instance, responding "Strongly Disagree" to the negative items is making a more forceful claim for personal efficacy than responding "Strongly Agree" to the socially-conforming positive items.

The item hierarchy was still irregular. Further thought solved the problem. The positive items were so socially desirable that any lack of agreement signaled a lack of efficacy. I set the dichotomization pivot between "Agree Strongly" and "Agree Somewhat" by anchoring the step in the recoded positive rating scale from category 2 to category 3 at 0.0 relative to item difficulty. The negative items were more even-handed, so I set the pivot point in the center of their recoded 4-category rating scale, by also anchoring the step scale from category 2 to category 3 in the same way. This yielded improved measurement and a clear construct.

Jeff Mosenkis

Recoding and Pivoting: An Example. Mosenkis J. … Rasch Measurement Transactions, 1997, 11:3 p. 578.

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/rmt113g.htm

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