Correlations
It is common knowledge that poverty is associated with the high incidence of HIV/AIDS across the African continent. Even though research findings have successfully proven that this association does exist, others believe that these findings do not provide a strong basis upon which to argue that poverty is directly linked to the high incidence of HIV/AIDS across the whole continent1. Proponents argue that because most research of this nature is done on a national level, a lot of essential factors are not taken into consideration; thus, these results cannot be generalized for an entire continent2. For example, Botswana has one of the highest per capita incomes in Sub-Saharan Africa; yet, it also has one of the highest rates of HIV/AIDS infection across the continent3.
Without knowledge of the above, when I set to find data that would correlate with the AIDS rate, I assumed that the crux of the work would be in actually finding data. Finding the data I needed wasn’t very hard; however, interpreting the results of my correlations proved to be much harder than I had expected. In the end, I realized that whether or not the data would correlate with the AIDS rate depended heavily on other factors such as the culture of the people, their geographic location, and possible external influences.
Let’s take a step back. The term correlation simply refers to the degree to which two or more variables are related. Therefore, if something correlates strongly with AIDS, it doesn’t mean that the thing causes AIDS; it simply means that wherever you find AIDS, it is very likely that that thing will be there too. Also, a correlation is determined to be significant or insignificant not only by the correlation coefficient but also by the number of datasets being correlated. In my case, because I had data for 10 regions, I needed correlation coefficients of no less 0.549 and 0.632, that is, at confidence intervals of 90% and 95% respectively.