|
?:abstract
|
-
"The issue addressed in this paper is how to obtain a composite measure of several indicators using benchmarking principles. While the exposition is only in two dimensions, and thus can be presented graphically, this is sufficient to capture the essence of the methodology and provide the basis for a critical examination of the assumptions. The data used is labour market statistics for the Member States of the European Union. The proposed approach comes from a technique originally used in production theory, namely efficiency frontiers. Here, however, we benchmark not efficiency but performance. There are two main problems. First, related to composite measures, how does one compare (weigh) indicators that are not obviously comparable? Second, related to benchmarking, how does one benchmark countries that may differ considerably as regards the mix of the various indicators. Both these issues concern weights and require that the weighting system should be parsimonious as regards assumptions and flexible, in that not all countries should necessarily be awarded the same weights. First we show how to construct the benchmark. It is constructed assuming only that the weights are positive, that we cannot discriminate between countries that are best in any single dimension and that a linear combination of these best performers is feasible, and also on the frontier. When benchmarking several indicators we arrive, not at a mark, but a frontier. The performance frontier is a multi-dimensional benchmark. Next we show how to measure the distance of other countries from the performance frontier. This requires the assigning of weights. The weights are determined by the location of a particular country to a particular segment of the frontier. The countries off the frontier are weighted in accordance with the weights of the countries on the frontier that have a similar mix of indicators. The essence of the method is presented by constructing a composite index of the unemployment and employment rates of the member states in 1999. We then present an example in three dimensions using various measures of unemployment. Finally, we extend the cross-sectional approach, i.e. the member states for a single year, to the construction of a pooled cross-sectional time series performance frontier. The pooled frontier is made up of the best performance of the best of countries at the best of times. In the final section we summarise the merits of this approach and comment on some possible criticisms." (author's abstract)
(xsd:string)
|
![[This page as RDF]](https://data.gesis.org/gesiskg/static/rdf-icon.gif){kind=icon}