Abstract
This paper analyzes some aggregation aspects of the procedure for constructing a composite index on a multidimensional socio-economic
phenomenon such as development, the main focus being on the unbalance among individual dimensions. First a theoretical framework
is set up for the unbalance adjustment of the index. Then an aggregation function is proposed that takes unbalance among development
dimensions into account; a separate index is also introduced that measures the unbalance itself. Finally the dataset of the
Index of African Governance for the year 2007 is used to test this method and compare it against the weighted arithmetic mean
of variables with relation to the measured values of unbalance, yielding significantly different results for ratings and rankings,
which in addition show negative correlations with the unbalance adjustment values. The changes ensuing from the adjustment
are commented for some countries.
phenomenon such as development, the main focus being on the unbalance among individual dimensions. First a theoretical framework
is set up for the unbalance adjustment of the index. Then an aggregation function is proposed that takes unbalance among development
dimensions into account; a separate index is also introduced that measures the unbalance itself. Finally the dataset of the
Index of African Governance for the year 2007 is used to test this method and compare it against the weighted arithmetic mean
of variables with relation to the measured values of unbalance, yielding significantly different results for ratings and rankings,
which in addition show negative correlations with the unbalance adjustment values. The changes ensuing from the adjustment
are commented for some countries.
- Content Type Journal Article
- Pages 1-27
- DOI 10.1007/s11205-012-0070-4
- Authors
- Enrico Casadio Tarabusi, Dipartimento di Matematica “G. Castelnuovo”, Sapienza Università di Roma, Piazzale A. Moro 5, 00185 Rome, Italy
- Giulio Guarini, Dipartimento per lo Sviluppo e la Coesione Economica, Ministero dello Sviluppo Economico, Via Sicilia 162/C, 00187 Rome, Italy
- Journal Social Indicators Research
- Online ISSN 1573-0921
- Print ISSN 0303-8300