Correlation Analysis
Having completed a basic analysis of the AIDS issue in Onondaga County and of the greater Syracuse area, it is now time to examine correlations between demographic data and AIDS rates of the area. In total thus far, 25 zip codes have been examined in Onondaga County. A correlation coefficient is calculated between two sets of data to determine if there is any relation between the data sets. In order for this association to be significant, a critical probability value must be obtained based upon the number of data sets being compared. Therefore, our N (number) value will equal 25, and we will be using a two-tailed probability curve, with a p-value of .05. A p-value of .05 means that we can be 95% confident that there is a statistically significant correlation between the sets of data. At the p < .05 level, a correlation coefficient of at least ±.396 is required for the data to have any statistical correlation. Anything in between these two values would not be significant enough to conclude there is a correlation. Just how statistically significant the actual correlation is can be determined by where the correlation falls between -1 and 1. The closer to -1 or 1 the value of the correlation is, the more negatively or positively correlated the data points are, respectively. The following data presented is an examination of the AIDS data obtained during the first analysis and its correlation to certain demographic data of Onondaga County in central New York.