From multiple choice questionnaires to synthetic indicators: jointly use of Rasch analysis and NonLinear PCA
Paola Annoni, Dep. of Economics, Business and Statistics - University of Milan
Pieralda Ferrari, Dep. of Economics, Business and Statistics - University of Milan
Download the Paper (PDF format) - October 19, 2006
Tell a colleague about it.
Printing Tips: Select 'print as image' in the Acrobat print dialog if you have trouble printing.
ABSTRACT:
The jointly use of two latent factors methods is proposed to assess a measurement instrument for underlying phenomena. In particular the decay status of valuable Italian buildings is quantified on the basis of a wide set of observed categorical variables. At the purpose, Rasch analysis is used to properly calibrate the data, to discard non informative variables and redundant categories. As a second step, an optimal scaling technique, Nonlinear PCA, is applied to quantify variable categories
and to compute a quantitative decay indicator, which could be independently used for units of future inventory. On the overall, similarities and different capabilities of the two techniques are analyzed and discussed.
SUGGESTED CITATION:
Paola Annoni and Pieralda Ferrari,
"From multiple choice questionnaires to synthetic indicators: jointly use of Rasch analysis and NonLinear PCA"
(October 2006).
UNIMI - Research Papers in Economics, Business, and Statistics.
Statistics and Mathematics.
Working Paper 17.
http://services.bepress.com/unimi/statistics/art17
| 
