Correlation with other simple indicators or indexes
Composite indicators often measure concepts that could be linked to well-known and measurable phenomena, e.g. productivity growth, entry of new firms. These links can be used to test the explanatory power of a composite indicator. Simple cross-plots are often the best way to illustrate such links. An indicator measuring the environment for business start-ups, for example, could be linked to entry rates of new firms, where good performance on the composite indicator of business environment would be expected to yield higher entry rates.
Note that correlation analysis should not be mistaken with causality analysis. Correlation simply indicates that the variation in two data sets is similar. A change in one indicator does not necessarily lead to a change in the other composite indicator and vice versa. Countries with high GDP might invest more in technology or more technology might lead to higher GDP. The causality remains unclear in the correlation analysis. More detailed econometric analyses can be used to determine causality, e.g. the Granger-causality test. However, causality tests require time series for all variables which are often not available.