Miskey Cues

Multiple-choice tests are straightforward to answer, straightforward to optically scan, and deceptively straightforward to analyze. As an analyst, it is easy to delude oneself that all is well, when, in fact, things have gone awry.

A frequent source of self-deception is the answer key or code book. All multiple-choice answer keys tend to look the same, just a string of letters or numbers: "ACDEBCDA". A garbled or misaligned key looks just as valid as a correct key. Even a key from a different test appears as good as the genuine key. It's extremely easy to perform an entire analysis and report the results without noticing there's been a major flaw in the process.

How can we avoid this kind of embarrassing misadventure? What must we watch for? Let's compare some results using correct and incorrect keys. The figures show the variable maps for two analyses of the same data. The first analysis uses the correct key. In the second analysis, every other item has the wrong key. What has happened?

MEASURE                  PERSONS     ITEMS
 4.0                            . +
 3.0                        .#### + P Q
 2.0                 .########### + R
 1.0                .############ + S T
  .0    .######################## + U
-1.0                 .########### + V W
-2.0                  .########## + X
-3.0                        .#### + Y Z
-4.0                           ## +

Analysis with correct key. Person and items spread out. Persons and items aligned.

MEASURE                  PERSONS  +  ITEMS
 4.0                              +
 3.0                              + Y
 2.0                              +
 1.0                          .## + P Q R U W
  .0                       .##### + S T
-1.0 .########################### +
-2.0                              + V
-3.0                      .###### + X
-4.0                          .## + Z

Analysis with incorrect key. Persons and items bunched up. Persons below items.

Impact on Items:
The intended item order was "P" to "Z". With the correct key, the items are ordered in that intended descending order of difficulty. With an incorrect key, the construct validity of the items collapses. The incorrect key still has half the items keyed correctly. Even so the item hierarchy is lost in a confusing jumble. Of course, to detect this problem, the analyst must be familiar enough with the items on the test to predict their order of difficulty. Statistics alone can not reveal item-content disorder. Statistics, however, do give warning signs: observe how the items in the figure bunch up.

Impact on Persons:
With a wrong key, getting an item correct becomes random. This causes examinees to bunch up in ability. In addition, their performance is somewhat below the mean apparent item difficulty because examinee responses are unlikely to match the incorrect key even 50% of the time.

The moral: Without construct validity, you have nothing. Don't leave home without it!

Miskey cues. Garcia HI. … Rasch Measurement Transactions, 1992, 5:4 p.185

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/rmt54j.htm

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