**INTRODUCTION**

When examining a health crisis or epidemic, it is beneficial to diagnose which areas or sects of people experience the strongest association with high or low numbers of a disease. This way, city and county health officials can combat the pandemic effectively. For this reason, correlational studies are used. Studying fifteen cities within four counties in Northern New Jersey, strong associations between different variables and high AIDS rates were found. Factors studied whose results showed strong relationships were: percentages of different races in given cities, percentages of households with inadequate indoor plumbing and on public assistance, median household income, and violent crime rates in given cities. Other factors such as different levels of education, percentages populations where no English is spoken, and distance to Manhattan, New York did not correlate as well.

In
order to fully understand the study, correlations and correlation coefficients
must be operationally defined. Correlation coefficients are calculated to
measure how well two variables co-vary. They are assigned values between the
numbers negative one and positive one. A negative correlation implies that as
one variable gets bigger, the other gets smaller while a positive correlation
shows that the two variables increase or decrease together. The closer to
negative or positive one the value is, the stronger the relationship. While
correlations allow for conclusions about *associations* between variables,
they do not imply *causation* (i.e. one variable is causing the other to
get bigger or smaller). A statistically significant correlation depends on how
confident the analyst wants to be in his or her conclusions. Correlations
significant at 95% confidence (with a p value of 0.05) were found in this study.
In order for a correlation to be significant at this level, the correlation
coefficient must be less than 0.514 or greater than -0.514. Often times, it is
useful to construct an XY scatterplot to depict the relationship between the two
variables and identify outliers.