TY - JOUR
T1 - Some considerations to carry out a composite indicator for ordinal data
AU - Zanarotti, Maria Chiara
AU - Pagani, Laura
PY - 2015
Y1 - 2015
N2 - Composite indicators (CIs) are important and useful tools in many elds
to assess, compare and rank performances, development stage, quality and
many other dierent targets. CIs are an overall measure of a multidimen-
sional, not directly observable, concept and are obtained by means of a set of
manifest variables (elementary indicators) that contribute to dene the over-
all measure. In this paper, some matters regarding methods to build CIs are
reviewed, assuming elementary indicators are ordinal and quantication is
necessary to convert observed data into a numerical form. Scoring methods,
aggregating functions and weighting systems are considered. In particular,
a scoring method based on the observed distribution or the use of dissim-
ilarity indices for quantication together with the Kendall- association or
a heterogeneity measure for weighting are suggested. Some of the reviewed
procedures are compared using students' satisfaction data.
AB - Composite indicators (CIs) are important and useful tools in many elds
to assess, compare and rank performances, development stage, quality and
many other dierent targets. CIs are an overall measure of a multidimen-
sional, not directly observable, concept and are obtained by means of a set of
manifest variables (elementary indicators) that contribute to dene the over-
all measure. In this paper, some matters regarding methods to build CIs are
reviewed, assuming elementary indicators are ordinal and quantication is
necessary to convert observed data into a numerical form. Scoring methods,
aggregating functions and weighting systems are considered. In particular,
a scoring method based on the observed distribution or the use of dissim-
ilarity indices for quantication together with the Kendall- association or
a heterogeneity measure for weighting are suggested. Some of the reviewed
procedures are compared using students' satisfaction data.
KW - ordered categorical variables, dissimilarity indices, combining functions, composite indicator, performance index
KW - ordered categorical variables, dissimilarity indices, combining functions, composite indicator, performance index
UR - http://hdl.handle.net/10807/78958
U2 - 10.1285/i20705948v8n3p384
DO - 10.1285/i20705948v8n3p384
M3 - Article
SN - 2070-5948
VL - 2015
SP - 384
EP - 397
JO - Electronic Journal of Applied Statistical Analysis
JF - Electronic Journal of Applied Statistical Analysis
ER -