Premium
A composite indicator dimension reduction procedure with application to university student satisfaction
Author(s) -
Marozzi Marco
Publication year - 2009
Publication title -
statistica neerlandica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 39
eISSN - 1467-9574
pISSN - 0039-0402
DOI - 10.1111/j.1467-9574.2009.00422.x
Subject(s) - dimension (graph theory) , principal component analysis , computer science , sample (material) , mathematics education , reading (process) , principal (computer security) , dimensionality reduction , quality (philosophy) , reduction (mathematics) , socialization , psychology , mathematics , artificial intelligence , social psychology , computer security , philosophy , chemistry , geometry , epistemology , chromatography , political science , pure mathematics , law
Universities play a central role within society and should provide high quality services to students. Therefore, a careful evaluation of university services is necessary. This evaluation is complex because it involves many partial aspects and can be assessed through a composite indicator. In this paper, we propose a simple method for reducing the number of partial aspects underlying a composite indicator. A practical application to data from a sample survey conducted on the last year students of the University of Padova is discussed. This survey considered the quality of many services, lecture rooms, library services, computer classrooms, reading rooms in libraries, study rooms, structure of exams, student socialization, reached skills and so forth. The method has been compared with principal component analysis. The results show that our method is worthy of consideration as it is markedly simpler to be applied than the other dimension reduction methods and requires milder assumptions.