Correlations

Paper # 2

http://www.scu.edu/stclaregarden/ethno/images/Sant_Clara_map.gif

In the first paper, it was seen that Palo Alto, California does not have a tremendous problem with Acquired Immune Deficiency Syndrome (AIDS) as its’ AIDS rate per 100,000 is 199, which is only slightly above Santa Clara County’s 194 and well below the national average of 339 per 100,000 (Note: in the previous paper the numbers are slightly different because different population estimates were used) (See Spreadsheet 1). As 54 percent of the AIDS cases in Santa Clara County resulted in death, much of Palo Alto’s AIDS rate can be attributed to gay men, especially those during the initial outbreak of the epidemic in the 1980s in the San Francisco area (See Santa Clara County Data). Looking at the data from the last paper, I found that the concentration of AIDS cases in gay men is due to the following correlations: Unmarried Partner Households, Male Householder and Male Partner, Percent of Population in Families, and Median Income. Other correlations that I expected to be significant but, in fact, resulted in weak correlations include: Percent of Population born in another state in the United States, Percent of Population Foreign Born, Percent of Males Age 35-44 years with an Associate Degree, Percent Black of African American, and Percent Hispanic. A few outliers, such as San Jose and Gilroy, stand at variance with the other data that significantly correlates lifestyles of gay men with AIDS rate. Strong correlations point towards relatively affluent, white, gay men at the center of the AIDS epidemic in Santa Clara County.

            Thirteen cities make up Santa Clara County, giving it 11 degrees of freedom, and requiring a correlation coefficient of 0.553 at p = .05 to be 95 percent confident of a correlation.  I found two factors that positively correlated significantly with AIDS rate: Unmarried-Partner Households and Male Householder and Male Partner Households. The Unmarried-Partner Household and AIDS Rate correlation had a correlation coefficient of r = +0.731 (See Data and Graphs). Such a strong correlation means that towns in Santa Clara County that have many households consisting of unmarried partners have high AIDS rates. Although it is not a causal relationship, the higher percent of unmarried partner households Santa Clara County has, the higher the AIDS rate is because the correlation is positive and the linear path is upward sloping (See Data and Graphs and Maps).  Since gay marriage is illegal in the State of California, many unmarried partner households may be gay couples living together. A strong correlation between AIDS rate and Unmarried Partner Households implies a relationship between AIDS rate and gay couples living together.

            The second strongest positive correlation is between AIDS rate and the Percent of Male Householder and Male Partner Households, which has a correlation coefficient of r = +0.695 (See Data and Graphs and Maps). This correlation shows that towns with many male-male households have high AIDS rates. The fact that areas that are home to high percentages of male-male households have high AIDS rates suggests that the AIDS epidemic in those regions focuses on gay men, many of whom live in households together. This correlation ties back to the original outbreak of the AIDS epidemic within gay communities in San Francisco. Many of the AIDS cases may date back to the 1980s or may be found in gay men who move to the San Francisco area because of the prominent and accepting gay community.

            The strongest correlation I found is a negative correlation relating AIDS rate to the Percent of the Population in Families, which has a correlation coefficient of r = - 0.82. Towns with higher percents of the population in families have lower AIDS rates. Similarly, towns with low percentages of the population in families have higher AIDS rates. This correlation may relate to gay men, in particular, contracting AIDS. Many single gay men or gay couples do not live in families and their lifestyles increase their chances of getting the disease. In this correlation, San Jose is an outlier, because it has a high percent of the population, 83, living in families and a high AIDS rate of 265 per 100,000 (See Spreadsheet 1).

            Another strong negative correlation, though just below the needed value of 0.553 needed to have 95 percent confidence, is the relationship between AIDS rate and Median Income, which has a correlation coefficient of r = - 0.517.  Towns with higher median incomes in Santa Clara County have lower AIDS rates than towns with lower median incomes. This result surprised me, as I had expected many of the AIDS cases to be found mostly in wealthy, gay communities. The fact that many towns that have low incomes have high AIDS rates is displayed by Campbell, which has the second lowest median income of $67,214 and the highest AIDS rate of 301 per 100,000. Gilroy, however, is an outlier with the lowest median income of $62,135 and the third lowest AIDS rate of 120 per 100,000 (See Data and Graphs).

            Other factors that I expected to correlate, but did not, included the percent of the population that was born in another state in the US, which has a correlation coefficient of r = – 0.07, and the percent of the population that was foreign born, which has a correlation coefficient of r = – 0. 08. Also, the percent of the male population aged 35 to 45 years with an Associate Degree, which has a correlation coefficient of r = +0.219, as well as the Black and African American population, which has a correlation coefficient of r = +0.46 and the Percent Hispanic, which has a correlation coefficient of r = - 0.09, did not correlate significantly (See Spreadsheet 1). I expected a high percentage of non-natives of Santa Clara County to correlate with AIDS rate because so many gay men move to the San Francisco area for the gay community, and in the past, for the infamous bath houses that spread AIDS so easily. I expected the percentage of the population with higher educations to correlate with AIDS rates, because I imagined that AIDS was common among the educated, wealthy, gay men in the area.

            Looking at the outliers, especially San Jose, and relating higher incomes to lower AIDS rates exposes another side of the AIDS epidemic in Santa Clara County. Not only is there a problem with the relatively affluent gay men, there appears to be a problem, unnoticeable when just looking at the vague AIDS rates numbers, of AIDS in poor sections of the cities. Because San Jose has such a large population of 893,889 many of the AIDS rates numbers from the disadvantaged neighborhoods, where people may contract AIDS through other methods, such as intravenous drug use, may be weighed out by the lower numbers from the wealthy neighborhoods, giving it an average AIDS rate per 100,000 of 265 (See Spreadsheet 1).

            In conclusion, looking at the correlations between AIDS rate and Unmarried-Partner Households, Male-Male Households, and the Percent of the Population in Families, it appears that the AIDS rate in Santa Clara County correlates mostly with relatively well-off gay men. Many of the cumulative cases may still remain from the initial outbreak of the AIDS epidemic in the gay communities of the San Francisco area in the 1980s.  In Santa Clara County, AIDS is not largely a result of impoverished minorities using intravenous drugs, although there may be another, smaller, overlooked front of the disease in such communities in San Jose.

 

Correlation Data & Graphs

Correlation Maps

Home

Perceptions--Paper 1

Solution--Paper 3