Submitting unidimensional questionnaire responses to correspondence analysis, a horseshoe effect (a.k.a. Guttman effect) shows up, if we plot the person-points on a graph. A unidimensional questionnaire analysis is hardly ever interesting, just because of this effect. However, such effect can be exploited to explore those cases when it does not show up for single persons-points. An index, based on the two first principal components of correspondence analysis, is presented, along with the computing formula. Some empirical evidence of its validity in detecting abnormal responding is presented and discussed.
Flebus, G. (2007). An Index to ascertain questionnaire response consistency based on the Guttman effect. Intervento presentato a: CARME, Correspondence Analysis and Related Methods, Erasmus University Rotterdam,, Rotterdam, The Netherlands..
An Index to ascertain questionnaire response consistency based on the Guttman effect
FLEBUS, GIOVANNI BATTISTA
2007
Abstract
Submitting unidimensional questionnaire responses to correspondence analysis, a horseshoe effect (a.k.a. Guttman effect) shows up, if we plot the person-points on a graph. A unidimensional questionnaire analysis is hardly ever interesting, just because of this effect. However, such effect can be exploited to explore those cases when it does not show up for single persons-points. An index, based on the two first principal components of correspondence analysis, is presented, along with the computing formula. Some empirical evidence of its validity in detecting abnormal responding is presented and discussed.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
