A weighting scheme is proposed to construct a new index of environmental quality for different countries using an approach that relies on consistent tests for stochastic dominance (SD) efficiency. The test statistics and the estimators are computed using mixed integer programming methods. The variables that are considered include countries greenhouse gas (GHG) emissions, water pollution and forest bene ts, as from the dataset of the World Bank. In the overall index of environmental quality land without forest contributes the most (with a weight around 71%), GHG emissions contribute with around 25% and water pollution contributes less (with around 4%). Moreover, countries are ranked according to their index of environmental quality and their rankings are compared with those of the Kyoto Protocol and alternative environmental indices. Then, employing a complementary SD approach, pairwise SD tests are employed to examine the dynamic progress of each separate variable over time, from 1990 to 2010, within 5-year horizons. Furthermore, pairwise SD tests are used to examine the major industry contributors to the GHG emissions and water pollution at any given time, to uncover the industry which contributes the most to total emissions and water pollution. It turns out that the components that are assigned high (low) weights in the SD approach are the ones that are the driving/fast-moving (holding back/slow-moving) variables in the sub- indices of GHG emissions and water pollution.