Understanding the Neighborhood Profiles

What are "neighborhood profiles"?
For the past several years PCA has been publishing information on the status of older Philadelphians. We have completed two major reports, The Socio-Economic and Health Characteristics of Philadelphia’s Elderly Population (2004) and Looking Ahead: Philadelphia’s Aging Population in 2015 (2006). These studies have focused on the city’s elderly as a whole.

We decided that the next major publication of information would differ from our previous reports in three important ways. First, it would be based on the web, rather than a paper publication (although a summary of the first printed report and full version of the second are available on our web site). Second, we would divide the city into smaller units so that we could examine the lives of the elderly at a more local level. Third, we would include information about these local areas that was not specifically about the elderly – crime rates for example, so we could place the elderly in their neighborhoods within the wider context of the areas in which they live.

We created twenty-six areas, which we call "neighborhoods" based on the division of the city by zip codes. Eight neighborhoods are each one zip code, and seventeen were created by combining two zip codes. Center City, which has eight zip codes, is treated as a single neighborhood. The reason for this division is discussed in the next section.

We hope that by providing information about the city’s elderly at this level, and linking the information about the elderly to information about the neighborhoods in which they live, we can help community leaders, people in the aging network, neighborhood activists, researchers and others better understand the lives of the city’s elderly population.

Where does the data come from?

We used three sources for the data. The first was the 2006 Southeastern Pennsylvania Health Survey. This survey, conducted every two years by the Philadelphia Health Management Corporation (PHMC), is an invaluable source of information on the health of people in the five county region, both young and old. The second source is data collected by the City of Philadelphia and provided by the Cartography Modeling Laboratory at the University of Pennsylvania. The data provided by the CML is also from 2006. The third is data from the 2000 United States Census provided by DataPlace.org.

Although we are matching the data from the three sources by zip code, it is important to understand that each provides a different type of information. The Census data, which is the oldest, counted every single person and institution in the city. The Cartography Lab data is from City records, so again it represents the entire number of events (such as criminal activity) in the city.

On the other hand, the PHMC data is a sample of the population. Using a method designed to ensure that the sample represents the entire population, they called random phone numbers in the City and completed interviews with as many people as possible. Then, through a technical process called "weighting," each person in the sample is assigned a number of persons to represent, so that the proportion of women, persons over the age of 75, etc. are correct. The standard we follow when using sample data is that there must be at least thirty (30) actual respondents in order to claim that the sample represents the whole group. In cases where we did not have thirty respondents in a zip code we combined that zip code with an adjacent one to create an area with enough respondents to make generalizations about the elderly in that neighborhood. So the 257,303 estimated elders in the City is based on a sample of 1,258 respondents. In zip code 19143 (Kingsessing), there were sixty respondents, so we treated that zip code as an individual neighborhood. On the other hand, in zip code 19142 (Paschall), there were only twenty respondents; so we combined them with zip code 19153 (Eastwick) which had seventeen respondents, to create a neighborhood with enough respondents to enable us to make general statements about all the elderly in the area.

What information is contained in the profiles?

We have selected items from all three data sources for our profiles. We hope that this selection provides a rich set of information which is not overwhelming.
The following table provides basic information on all the items contained in the profile. All items refer to persons age 60+ unless otherwise noted.
 
Name of the neighborhood
 
Zip code(s) contained in the neighborhood
 
Estimated total population
 
Estimated population of persons age 60+
 
60+ as proportion of entire population
 
Percentage of persons living below 200% of the poverty level.
 
Percent minority (including Black, Latino, Hispanic and American Indian)
 
Percent of the 60+ population who are age 75+
 
Percent Female
 
Percent reporting having a chronic illness
 
Percent saying their health status is fair or poor
 
Percent having one or more problems with instrumental activities of daily living
 
Percent having one or more problems with activities of daily living
 
Percent caregiving for a person age 60 or older
 
Percent reporting three or more depressive symptoms during previous week.
 
Percent saying they strongly agree or agree that they are part of their neighborhood.
 
Percent saying they strongly agree or agree that people in their neighborhood can be trusted.
 
Percentage who live alone.
 
Percentage reporting that they need roof repair.
 
Percentage reporting that plumbing needs repair.
 
Percentage reporting that heating and/or cooling system needs repair.
 
Average (mean) age.
 
Percent having one or more children at home.
 
Type of health plan (This item has three columns, for "fee for service", "HMO" and "PPO.").
 
Years of education. (This item also has three columns, 1 ) persons who did not complete a high school education; 2) persons who completed high school and persons with some college education; 3) persons who completed a bachelors or higher degree.
 
Crimes against persons (robbery, aggravated assault)
 
Number of crimes against persons per 1000 population
 
Crimes against property (Burglary, Theft, Auto Theft)
 
Number of crimes against property per 1000 population
 
Total number of properties as of 7/1/07
 
Total number of residential properties as of 7/1/07
 
Residential properties as proportion of total properties
 
Total number of commercial properties as of 7/1/07
 
Commercial properties as proportion of total properties
 
Total number of vacant properties as of 7/1/07
 
Vacant properties as proportion of total properties
 
#/1000 - physicians
 
#/1000 drug stores
 
#/1000 dentists
 
#/1000 social agencies
 
#/1000 supermarkets
 
#/1000 banks
 
#/1000 convenience stores
 
#/1000 restaurants

 
Because the data is often an estimate, we would caution anyone about making judgments about small differences between neighborhoods on a given item (but see below). For estimates that were 5% or below or were 95% or above we report that information rather than the actual estimate because extreme estimates such as this (for example, an estimate that says 100% of the elders in a given neighborhood are minority) need to be taken with caution.

We have placed the city data in the first column so that each neighborhood can be compared with the city as a whole.

How can I use the information?

There are two basic ways to use the information. The first is to explore a particular neighborhood by looking at the various types of information contained in the table. The second is to compare a neighborhood to one or more other neighborhoods, or to the city as a whole.

What we find is that no one item draws a full picture of the elders in a particular neighborhood. Nicetown residents can seem to be much worse off (in terms of income), about the same (in terms of chronic illness), or even better off (in the proportion of crimes against property per 1000 persons) than the elders in Center City. One must look at a range of items to gain an overall picture of the lives of the lives of elders in a given area.

Further, in order to identify either individuals or neighborhoods in greatest need of intervention we should look not at demographic correlates of health problems (such as living alone or age) but at how people report their ability to function (functional health, depressive symptoms) in their environment. Age is not a diagnosis; but the inability to perform the activities of daily living can be the direct cause of nursing home placement or other serious outcomes. While we need to look at a range of items to gain a picture of a neighborhood, we also need to be able to identify those issues that are the most critical in determining what interventions are needed and where.