A pitfall of empirically-based data analysis, into which the unwary easily stumble, is uncovered by even a cursory Rasch analysis of the Minnesota Multiphasic Personality Inventory-2 (Butcher et al. 1989).
Consider the "depression" scale of 57 true/false items. 20 items, like "I cry easily", are keyed "true". 37 items, like "I am happy most of the time" are keyed "false". The count of those responses by the subject that agree with the key is intended to indicate the severity of the subject's depression.
How were these items chosen? The selection criterion for each item was that it differentiated some sample of "normal" subjects from some other sample of patients exhibiting symptoms of depression. But this empirical method is precarious - accidental differences between samples will be mistaken for indicators of depression. "I sometimes keep on at a thing until others lose their patience with me" falls into this category because it emphasizes the mental state not of the subject, but of those with whom the subject comes into contact.
How are the items used to identify depression? The empirical assignment is built on the rule that the key is the direction (true or false) that the "depressed" patient sample endorsed more frequently. Here is the pitfall. Imagine an item to which none of the normal sample responded "true", but 49% of the "depressed" sample did respond "true". This item would differentiate the samples well, but it would be incorrectly keyed! Since 51% of the "depressed" sample responded "false", the key value would be "false", and consequently the normal sample would be reported as more depressed than the "depressed" sample! Of course, if we let ourselves be controlled by data, rather than by theory, ("What does it mean to be depressed?"), there are all manner of other ways to stumble into this pitfall.
In the MMPI-2 depression scale, the pitfall is encountered 8 times. Take an item such as "I sweat very easily even on cool days." Common experience indicates that "normal" people respond "false" to this. But in the MMPI-2 this is keyed "false" to indicate depression. This item can act as an indicator of depression, but it must be rekeyed "true" in order to add up with the other items into a score whose greater value indicates a more depressed state.
Empirically-based methods may provide some exploratory guidance when we are truly at a loss as to what to do. But theory-mandated choices must supersede accidental data patterns for meaning and knowledge to result.
Department of Psychology
University of Chicago
[After this Research Note was published, a member of the MMPI-2 development team remarked that this flaw in MMPI-2 was discovered early on, but ignored, because of the team's commitment to the "majority depressed" scoring rule.]
Which way is up? Correctly keying the MMPI-2. Chang C.-H. Rasch Measurement Transactions, 1994, 8:1 p.339
|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.
|Coming Rasch-related Events|
|Feb. 28 - June 18, 2022, Mon.-Sat.||On-line course: Introduction to Classical and Rasch Measurement Theories (D. Andrich, I. Marais, RUMM), The Psychometric Laboratory at UWA, Australia|
|Feb. 28 - June 18, 2022, Mon.-Sat.||On-line course: Advanced Course in Rasch Measurement Theory (D. Andrich, I. Marais, RUMM), The Psychometric Laboratory at UWA, Australia|
|May 20 - June 17, 2022, Fri.-Fri.||On-line workshop: Practical Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com|
|June 24 - July 22, 2022, Fri.-Fri.||On-line workshop: Practical Rasch Measurement - Further Topics (E. Smith, Winsteps), www.statistics.com|
|Aug. 12 - Sept. 9, 2022, Fri.-Fri.||On-line workshop: Many-Facet Rasch Measurement (E. Smith, Facets), www.statistics.com|
|Oct. 7 - Nov. 4, 2022, Fri.-Fri.||On-line workshop: Practical Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com|
|June 23 - July 21, 2023, Fri.-Fri.||On-line workshop: Practical Rasch Measurement - Further Topics (E. Smith, Winsteps), www.statistics.com|
|Aug. 11 - Sept. 8, 2023, Fri.-Fri.||On-line workshop: Many-Facet Rasch Measurement (E. Smith, Facets), www.statistics.com|
The URL of this page is www.rasch.org/rmt/rmt81f.htm