In a previous mission (see Finding Obama in the smallest Census geography) we delved down to the see what data was available at the level of individual blocks. Unfortunately, as we noted there, the Census doesn’t provide a whole lot of useful data at the block-level, since the results will exclude sample data from the SF3 “long form” (or, post-2000, the American Community Survey). If we want to know more about a neighborhood we will need to think in slightly larger geographies, and seek data at the tract-level or higher.
For this mission, we’ll be zooming into to Park Slope neighborhood on Brooklyn, and gathering data on income, race, education, and the breakdown of owners and renters for a single census tract. Since its often helpful to be able to view data like this in the context of the surrounding neighborhood, subsequent missions will explore ways to make comparisons with this sort of data, either to other tracts or to larger geographies.
But for starters, our target: although defining the exact edges of a neighborhood is never easy – especially ones in dense, diverse areas, where even residents disagree over terminology and the continual processes of gentrification, urban decline, migration, and other demographic shifts continually redefine the categories – most observers would agree that the neighborhood extends roughly north and west from Bartel Pritchard Square, at the lower corner of Prospect Park, with both 15th Street and Prospect Park itself providing something of an “edge.” Since edges are often exciting places to observe change, we will select an address along 15th Street, near the corner of 5th Avenue.
Getting the data
Returning to the Census FactFinder website and selecting the
Geographies tab from the menu on the left, we can enter an address to search for the right tract: for this, let’s start with “545 5th Ave., New York, NY 11215.”1 FactFinder should return a list of matching geographies, including “Census Tract 141, Kings County, New York,” which you can add to your selections.
After closing the Geography selection, we can start to figure out the particular data we want to collect. For starters, let’s answer the “When?” question: we want the most recent data available. From the “Topics” menu, we select
Year-->2010, which will help focus our search: as you can see when the screen refreshes, most of the Tables that meet our two criteria (Geography equals this tract, Year equals 2010) come from the ACS 5-year estimates.2
As for the “What?” question, we’ve said we want to learn about the following:
- Income: There are dozens of ways to think about income in the census data: individual vs. family vs. household income; percentage above or below the poverty rate; number of individuals falling into different income categories; and so on. For now, let’s start with one of the more common tables: “B19013: Median Household Income in the Past 12 Months.” From this table, we can see that at the time of the survey the median income for households in this census tract was $78,674, plus-or-minus $15,376.
- Race: As with income, there is a dizzying assortment of tables on FactFinder concerning race and ethnicity. Let’s begin with one of the more straightforward ones: “QT-P3: Race and Hispanic or Latino Origin” which includes numbers and percentages for all individuals in the tract (and, since this particular table is derived from the SF-1 dataset, these figures are actual counts, not just estimates). From this table, we learn (for example) that there were 239 Asians living in the tract, and that 23.6% of the people in the tract were Hispanic or Latino.
- Education: A good multi-purpose table on education is “S1501: Educational Attainment.” In addition to providing the number of individuals in different categories of educational attainment (wisely limited to only those individuals over a certain age – it’s a little silly, for example, to include children under 18 when trying to say something about the percentage of residents with a high-school diploma), the table also breaks down this data to show how it intersects with age categories and also median income and poverty status. From this table, we can see that, for example, at the time of the survey, 35.5% of the population 25 and over had at least a Bachelor’s degree.
- Tenure: The most basic table on the breakdown between ownership and rental housing is “B25003: Tenure.” More complex tables further split this data based on a host of other characteristics (and also help note differences between, say, occupied and vacant units). In our selected tract, we can see that there were 1,185 occupied units in the tract, and that a little under half of them (509) were owner-occupied.
Two Cautions Concerning ACS Data:
In passing, it’s worth noting that some of the data we’ve gathered — the tables on income, education, and housing ownership — contain data derived from the American Community Survey. Unlike Decennial Census data, which is a complete census of the entire population (but is gathered only once every ten years, and is limited to a few specific questions on age, sex, race, and familial relationships), ACS data is the result of a small random sample of the population collected every month and aggregated over time to provide “rolling averages” for many more detailed questions. As a result of this difference, one should be aware of the following when interpreting tables from the ACS:3
- Margins of Error: Since it is the result of a sample, and not a complete count, all ACS data is presented with margins of error: these are estimates, not exact numbers. Elsewhere we will explore more statistically-appropriate ways to work with these kinds of numbers, but for now it’s worth at least looking at these 90% margins of error and thinking about them in terms of the numbers being measured. For example, Table B25003 indicated that there were 509 ownership units and 676 rental ones, but given the margins of error reported (110 and 125, respectively), it’s not inconceivable that the two numbers could be even closer (or conversely, further apart).
- Time Periods: As a result of the “rolling” nature of the ACS (which is necessary in order to capture a large enough sample to report data for small geographical units), these tables do not represent a point in time, but rather a blend of five-years’ worth of responses. This is a slight annoyance when trying to convey a portrait of the neighborhood for a particular year; even more problematic are the repercussions when we try to make statements about changes in the neighborhood over time (see next mission).
So far, all we’ve done is gather data on a census tract. We could use these tables to try to say some things about the people living in this neighborhood over this time period, as we’ve suggested above, but as you attempt this you will soon see that numbers alone are hard to render meaningful: we need some context. In our next mission, we’ll remain focused on this tract in Park Slope, but we’ll start to think about ways to add context to our data.
1 Note that there is nothing particularly interesting about this address – it’s just something that should get us into the right ballpark for this tract.
2 There are also some tables from the 2010 SF-1, but these will only contain the more basic demographic data we found at the block-level.