To identify the region’s hottest neighborhoods and towns, we began with year-end real estate price and transaction data for approximately 400 MLS areas and specific zip codes in the eight-county area (Bucks, Chester, Delaware, Montgomery and Philadelphia counties in Pennsylvania, and Burlington, Camden and Gloucester counties in New Jersey) derived from HomExpert Market Report, a product of Prudential Fox & Roach, Realtors, Research Division. The information provided was based on MLS data and deemed accurate, but not guaranteed.
Philadelphia magazine researchers then collected demographic data, including population figures, household income and educational attainment, for the same areas from the 2000 and 2010 U.S. censuses and the 2011 American Community Survey.
We sent our compiled data to George Recck, a statistical expert at Babson College in Wellesley, Massachusetts. To ensure statistical validity, he first removed from consideration any MLS area that did not have at least 30 transactions over each of the last four years. From the remaining towns, he calculated various year-over-year percentage increases and decreases, depending on the category. At this point he calculated a standardized value for each district from the average mean and standard deviation. Predetermined weights were then applied to each category to compute an overall standardized Z-score for each district; greater weights were applied to more recent data. From the overall standardized scores, a hotness determination was made using logical statistical cutoff points.
The following weights were used for each category:
Percent increase in number of properties sold from 2011-2012: 9%
Percent increase in number of properties sold from 2010-2011: 5%
Percent increase in number of properties sold from 2009-2010: 3%
Percent increase in median price of properties sold from 2011-2012: 18%
Percent increase in median price of properties sold from 2010-2011: 10%
Percent increase in median price of properties sold from 2009-2010: 6%
Percent decrease in average days on the market from 2011-2012: 18%
Percent decrease in average days on the market from 2010-2011: 10%
Percent decrease in average days on the market from 2009-2010: 6%
Percent increase in population 20-54 from 2000-2010: 2.5%
Percent decrease in population 65+ from 2000-2010: 2.5%
Percent increase in median household income from 1999-2011: 5%
Percent increase in households with a BA or higher from 2000-2011: 5%