Social Network Analysis on the movement of STDs in Neighborhoods in Brooklyn NYC

Introduction

My initial idea was to look at the drug networks in New York City in order to identify the potential dealers. It would have looked at how users clustered around a particular attribute and how identifying that attribute could be used to identify the person at the epicenter ‘i.e. the drug kingpin’. However, I stumbled across something more interesting in my search for data with the NYC census database.

I found a wealth of information on, education, income, crime, mortality—even the highest causes of mortality—in neighborhoods in Brooklyn. The initial analysis of the data was incredible. Neighborhoods that were more than 50% African American had higher levels of crime, lower levels of education and had higher percentages of STD transmissions (namely HIV) than predominantly white neighborhoods. Startled by the data, I decided to change the focus of my project. I want to understand what attributes connect these neighborhoods, beyond race, because there may be data that is being overlooked in considering why there is a large cluster of STD infected people in minority neighborhoods.

Question/Hypothesis

In 1996 a PHD student from the University of Chicago used social network analysis to prove that the homogeneity of African American communities in the United States in addition to the mixing patterns, explained why African Americans we 1.3 times more likely to contract an STD than other races. I am interested in doing a similar analysis in Brooklyn to understand what other factors contribute to the spread of STD’s in certain neighborhoods. I also want to see if I can use social network analysis to uncover a hidden institutional element that the student did not cover in her paper.

http://www.ncbi.nlm.nih.gov/pubmed/10333277

·         How can we use social network analysis to visualize the movement of STD’s between neighborhoods in Brooklyn, New York?

·         What attributes are associated with a ‘high risk’ neighborhoods and how are they connected to ‘low risk’ neighborhoods?

·         Could gentrification be a form of disease control or is it a possible cause for the increase? (Here I will be looking at data over the past 3-4 years for Williamsburg, Bushwick and Crown Heights—recently gentrified areas of Brooklyn)

Methodology

I will be looking at a variety of attributes:

·         STD's

·         Neighborhood

·         Proximity in miles from each other (this will be used to set up the matrix)

·         NYC zoning rules

·         Income

·         Education Level

·         Crime Rate

·         Number of Clinics in the Area

·         Population with healthcare

The most difficult part of this project will be to code the data in a way that it can be analyzed in Network Analysis tools. The main sources of my data will be The New York City Department of Health

Community Health Profiles: http://www.nyc.gov/html/doh/html/data/data2006.shtml

I am also hoping to find data on the individual STD network in Brooklyn. I would have to contact the CDC for the data. The data would not significantly affect my analysis if I cannot obtain in, but it would be interesting to have individual's data from each neighborhood (the people would probably be coded by numbers)

Possible Results

I am expecting to see many clusters in the data, namely separated by race. However, I could be surprised and find that other issues that connect these neighborhoods. While Laumann asserted in her paper that the homogeneity of African American communities contributed to the rapid spread of STI’s, I contend that limited access to health care and low level education can also be a deciding factor. Furthermore, zoning/district lines in New York City could possible serve as ad-hoc version of a quarantine. I will just have to analyze the data to find out!