Massive amounts of new census data were released just after midnight on August 15th, which show among other things:
1) New Yorkers have become much better educated since 2000.
2) Though immigrants continue to stream into the city, the number of foreign born residing in the city only increased by about 160,000 in the five years through 2005, well below the increases of the 1990s.
3) The number of blacks has declined in the city.
4) The number of whites has increased in Manhattan and now Brooklyn.
The midnight data dump was the first release from the United States Census Bureau's new American Community Survey. Three more releases from the 2005 survey are planned, including one on economic status (August 29th), one on housing, and one on detailed characteristics by race and Hispanic status. In the future, the data will be constantly updated.
Some of the questions that the American Community Survey will ultimately answer include:
1) Has Manhattan become even more unequal? After the 2000 Census the richest 20 percent in Manhattan had incomes 50 times those in the lowest 20 percent. Could this have increased?
2) What changes are discernable in the various community planning districts?
3) If the net number of foreign born residents is increasing more slowly than it was in the 1990s, while the number of Mexicans is soaring, which groups are showing a decline?
4) Has the trend toward a more highly educated population in the city and especially Manhattan affected the housing opportunities of affluent and non-affluent New Yorkers?
5) How are the various immigrant groups faring in the post 9/11 world?
American Community Survey
The advent of the American Community Survey represents a sharp departure from the way the census bureau collected data on so-called social and economic characteristics in earlier decades.
It is important to realize that the census bureau’s clients are the United States Congress and the governments of the 50 states. The United States Constitution requires that there be a census every ten years in order to carve the United States into an appropriate number of Congressional Districts, each representing the same number of residents. This redistricting and reapportionment also occurs on a state and local level, for legislative, judicial and other districts.
To accomplish the redistricting, the Census Bureau need only collect a handful of variables – the location of each person, age, sex, race, Hispanic status, whether one owns or rents a house or apartment or lives in a group quarters (such as a college dorm, prison, or chronic care hospital).
Along with these so-called "short form" data, a sample of respondents in the past was asked many more questions about income, occupation, veteran status, citizenship, education, disability status, ancestry and ethnicity, migration, journey to work, and so forth. About one-sixth of the households in the United States, or about 18 million, received this "long form."
The American Community Survey is the replacement for the long form. It is a rolling survey with a set number of interviews each month totaling about 3 million per year. One of the largest advantages of this survey is that those doing follow-up will be professional interviewers, and methods have been established to count hard-to-reach groups effectively. Five years of testing ensures quality control. As more and more data become available; in all, the census bureau will publish about 15 million records. Information on tracts and block groups, the small areas that were released for the "long form" will be available in 2010.
The just-released 2005 American Community Sruvey includes some data for places and counties with population 65,000 or larger, including most of New York City's community planning districts, which are released as areas called Public Use Microdata Areas (PUMAs). PUMAs are required to contain at least 100,000 people, and there are about 2,100 across the country. New York chose to make them largely coincident with the community planning districts. However, since the sample is not quite three percent, one can expect that only about 2,500 people will be interviewed in each PUMA or planning district. This means that for very detailed variable categories, such as the 123 individual foreign birthplaces that are counted, most PUMAs may have a very small number of a given population type. When this occurs, the census bureau suppresses the data since no accurate estimate of such groups are possible.
So far the American Community Survey has provided a surprising new look at NYC. Though immigration is continuing, immigrants along with non-immigrants seem to be leaving the city. The increasing affluence of Manhattan seems to be spilling over into Brooklyn. These and other trends mean that the remaining data releases are of great interest. We will know now how New York City is evolving, we will not have to guess or wait until the results of the new census in 2012.
Andrew A. Beveridge has taught sociology at Queens College since 1981, done demographic analyses for the New York Times since 1993, and been in charge of Gotham Gazette's demographics topic page since 2000. The opinions expressed are his alone.
