A Comparison of Different Methods for the Construction of Composite Indicators

Composite indicators may be used to measure complex variables which are not directly measurable. The basic idea is to break down a latent variable into components which can be measured by means of simple (partial) indicators. The partial indicators are then combined to obtain a composite indicator. This paper discusses the development of composite indicators and their robustness. In particular, the comparison of different methods to construct composite indicators is addressed. A Monte Carlo simulation study is performed to evaluate the different methods