Table II: Pairwise correlations

However, if the correlation coefficient is close to zero (0), it suggests that the variables are linearly independent, and multi-collinearity will not be an issue. In addition, a correlation coefficient less than 0.5 is generally considered acceptable. The table reports that all the correlation coefficients are weakly correlated, showing correlation coefficients below 0.5 in most cases.

The study also shows that most of the correlation coefficients are statistically significant 10%. Theoretically, the study expects a positive relationship of Energy use per capita, credit to the private sector as a percentage of GDP, and CO2 emissions per capita; the results meet this expectation. Similarly, the study expects energy consumption per capita to be positively correlated with a net inflow of foreign direct investment, remittances % GDP, and trade % GDP.

A negative relationship runs from the inequality variables to energy consumption per capita. The weak correlation between the variables indicates that the study is insulated from the problem of multi-collinearity.