We carefully administered questionnaires designed to measure respondents' degree of victimization. But soon discovered that analyzing responses to questions irrelevant to the respondent produced clinically misleading results.
-------------------------------------- | RAW OUTFIT |Subject | | SCORE MEASURE MEAN-SQUARE|Response| -------------------------------------- | 1 -2.47 7.79 | S 0001 |<-one severe experience | 1 -2.47 .20 | M 1000 |<-one mild experience --------------------------------------
The questionnaire includes 4 types of victimization. Respondent M had only
experienced one type, the relatively mild "verbal pressure". Respondent S,
however, had also experienced only one type, the more severe "physical force".
Both respondents rated their experiences as "1" on a scale from "0" to "3".
Since both raw scores are the same, 1 on the 4 items, their measures, based on
the pre-calibrated items, are also the same. Yet their fit statistics are quite
different, indicating the measure for Respondent S is problematic.
It is clear that "physical force" goes beyond "verbal pressure", so that we could edit the data by asserting that more extreme forms of victim abuse imply milder forms. Then remeasure the respondents, so obtaining more clinically reasonable measures and fit statistics.
A simpler approach is to bear in mind that the measure of interest corresponds
to each respondent's most extreme form of victimization. It is easy to identify
that measure on a map of the "Expected Ratings on Victimization Items".
Expected Ratings on Victimization Item Mild Severe -5 -4 -3 -2 -1 0 1 2 3 4 *lt;-Logits +-----+-----+-----+-----+-----+-----+-----+-----+-----+ 0 : X : 2 : 3 Physical Force 0 : 1 : 2 : 3 Threats 0 : 1 : 2 : 3 Intoxication 0 : X : 2 : 3 Verbal Pressure +-----+-----+-----+-----+-----+-----+-----+-----+-----+ ^Respondent M(ild) ^Respondent S(evere)
According to this Figure, Respondent M's mild victimization yields a measure of -3 logits, but Respondent S's severe victimization yields a measure of +1 logit. Now the measures make clinical sense. In other words, there is no need for us to be constrained by the printed form of the questionnaire, rather we can act as though each person was asked only one question: "Describe your worst victimization". Insisting that all data be analyzed in accordance with the ways they happen to have been collected manifests a "foolish consistency [which] is the hobgoblin of little minds" (Ralph Waldo Emerson, Conduct of Life: Self-Reliance).
Omit Inconsequential Responses. Karabatsos, G. Rasch Measurement Transactions, 1997, 10:4 p. 523.
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 | |
---|---|
Oct. 4 - Nov. 8, 2024, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
Jan. 17 - Feb. 21, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
May 16 - June 20, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
June 20 - July 18, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Further Topics (E. Smith, Facets), www.statistics.com |
Oct. 3 - Nov. 7, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
The URL of this page is www.rasch.org/rmt/rmt104a.htm
Website: www.rasch.org/rmt/contents.htm