Socio-Environmental Development Index
for the State of Bahía
Leonardo Araújo, Elaine Fernandes and Patrícia Rosado
Environmental Conditions Index for the Municipalities of Bahía ( ...continuation )

The tests indicated that the sample provided adequate factorial analysis.6 The analysis was carried out using the principal component method, showing two factors with characteristic roots greater than one, as can be seen in Table 1. These results show that Factors 1 and 2 amount to 61.07% to explain the total variation of the indicators used. Orthogonal rotation of the data was chosen rather the Varimax method to ensure better interpretation of the results.

Table 2 shows which factors are related to which variables through the common factor loads. Factor 1 is strongly correlated with the following variables: persons living in permanent private homes with running water, persons living in permanent private homes with refuse collection and sewers; persons living in permanent private homes with refuse collection and industrial sector contribution to municipal gdp. As previously mentioned, the relative contribution of industrial gdp to municipal gdp was used as a proxy of industrial contamination; however, the factorial load of 0,442 is less than 0.65 in absolute values, which classifies this variable as irrelevant for building the eci. Factor 2 is strongly correlated with vegetation coverage in the municipality and persons living in homes without automobiles. In relation to this last item, it is important to note that the automobile is a significant air polluter, mainly because of carbon monoxide emissions.

It can therefore be said that Factor 1 summarizes the variables that encompass infrastructure in homes, mainly in relation to sanitation conditions (such as running water, sewage collection, refuse collection). These variables are strongly related to living conditions, so that where infrastructure conditions are improved in homes, so environmental quality is improved.

Factor 2 summarizes that variables related to environmental quality of ground, water and air, such as vegetation coverage and percentage of people living in homes without automobiles. Where vegetation coverage is better, there is a positive impact on soil fixation, replenishment of water springs, carbon sequestration from the air and oxygen is released for photosynthesis; in addition, the more homes without automobiles, the better the air quality and consequently the environmental quality.

After presenting the factors and factorial loads, the factorial points were estimated and the Environmental Conditions Index calculated.

6Bartlett’s sphericity test and the Kaiser-Meyer-Olkin test were used. Bartlett’s test obtained a value of 673,843, significant at 1% probability, which enables the null hypothesis to be rejected from the correlation matrix. It is an identity matrix, i.e. there is no correlation between the variables. The value obtained for the Kaiser-Meyer-Olkin test was 0,747, indicating that the sample is adequate for undertaking factorial analysis.