Neighborhood Change Database
1970-2000 Tract Data















Neighborhood Change Database Features

To be released in the next few days! Neighborhood Change Database (NCDB) gives you access to US Census data from 1970, 1980, 1990 and 2000 at the census tract level. It is the only source of census data with variables and tract boundaries that are consistently defined to the 2000 Census tract boundaries. It is an invaluable resource for policy makers, community organizations, and researchers who want to analyze changes that have occurred in U.S. neighborhoods over four census decades.

The NCDB is available on CD and can be accessed using the menu-driven mapping and analysis software, or it can be extracted for use in external database, mapping, and statistical analysis packages.

The NCDB contains 1970, 1980, 1990, and 2000 Long and Short Form data. The latest release of the NCDB includes responses to the Census 2000 Long Form questions that were asked of about one out of every six households in the census: detailed population, household, and housing characteristics, including income, poverty status, education level, employment, housing costs, immigration, and other variables.

An earlier version of this product was named CensusCD 40 Year Tract Series. The new NCDB completely replaces all previous versions of the dataset with a new more powerful search capacity, the 2000 Long Form data, and corrections to a few minor problems that users have reported.

Selecting Data Format - Regular vs. Normalized

You can access the data for 1970, 1980, 1990, or 2000 in their census year tract boundaries. In this way, the data will appear as it did for that census year and the maps will be drawn according to that year's boundaries. With Neighborhood Change Database, however, you can access data for all four decades normalized to 2000 tract boundaries.

The All years normalized to 2000 selection allows you to compare data for various years. The data for years 1970 to 1990 is recalculated and normalized, and the report uses the 2000 tract ID. This allows you to compare data for various years within the exact same boundary definitions. By selecting "All years normalized to 2000" you can do actual apples-to-apples comparisons of historic data in 2000 tract definitions.

Source of data

The data in the Neighborhood Change Database are compiled from the U.S. Census Long and Short Form tract-level tabulations for census years 1970, 1980, 1990, and 2000. These data include income, education, housing value, and commute time. All of the data are available at the tract level only. In 1970 and 1980, tracts covered only some areas of the U.S., and thus the data coverage will be incomplete for those years. Beginning in 1990, the entire U.S. was covered by census tracts. Also, the 1970 and 1980 censuses have suppressed fields. For those two years, the Census Bureau did not release data if there were fewer than 5 to 15 people in a tract.

What are the differences between the Neighborhood Change Database product and the individual census products for 1970, 1980, 1990 and 2000?

The Neighborhood Change Database has two main advantages: normalized data and cost. If you want to do time series analysis (looking at the changes of a given location across time periods) then the Neighborhood Change Database is the ideal product. Because the NCDB data is normalized, you can measure just the change in population without having to control for changes in boundary definitions. The year specific boundaries are also included and if you only need tract level geographies, this is a much more cost-effective way of purchasing 4 decades of Census data for the price of one.

The Neighborhood Change Database is not as exhaustive as the full census because the data are only expressed at the tract level and include a subset of the complete variable list for each decade. While the complete set of census variables is not included, the Neighborhood Change Database does have an extensive subset - about 1,000 variables for each decade. Furthermore, these variables have been defined to make it very easy to compare data across censuses. The complete set of census variables is available on the individual CensusCD products. The CensusCD 1980, 1990, and 2000 have an additional geographic level of block group (which is about 3-4 times smaller than a tract). We also offer data for 1990 and 2000 that go down to the Block level (40 times smaller than a Block Group). The other main advantage to the census year specific products is that you can easily run reports for areas larger than tract level. For example, if you wanted state numbers for each of the 50 states with the NCDB you would need to run Summary Reports 50 times or agglomerate up the tract level data to the state (summing 1000 rows per state on average).

Geographical selections

The Neighborhood Change Database is based on the geographic unit of the Census Tract. The Census Tract is the Census Bureau's statistical equivalent of a large neighborhood (with an average of about 4,000 people).

You can select any of the 4 geographical areas: Nation (US), States, Counties, and Tracts to gather data from. The 2000 data (and the normalized data) also allow for the geographies of MSA/CMSA and PMSA. The data is accessible at the tract level only.

You can select a single or multiple states or counties and get all of the component tracts. For example, if your area of interest is New York City you could select the 5 counties: New York (Manhattan), Richmond (Staten Island), Bronx, Queens, and Kings (Brooklyn), which contain the 5 boroughs of New York. If you think of New York City as including its outlying suburbs then selecting New York PMSA would add Rockland, Putnam, and Westchester Counties. Alternatively, you could select a point within New York City area and get all the tracts within a certain distance from this point (Radius).

Exporting the data and boundaries

You can use extracted data files as inputs to other systems like statistical (e.g. SAS, SPSS), database (e.g. Access, Oracle), spreadsheet (e.g. Excel, 1-2-3), or mapping (e.g. ArcView, MapInfo) packages.


Neighborhood Change Database Geography

The Neighborhood Change Database is based on the geographic unit of the Census Tract. The Census Tract is the Census Bureau's statistical equivalent of a large neighborhood (with an average of about 4,000 people).

You can select any of the 4 geographical levels:

  • Nation (United States)
  • States (50 + District of Columbia)
  • Counties (3,141)
  • Tracts (depends upon the census 35-65,000)
  • MSA/CMSA and PMSA (are also included for 2000)

The data will be presented at the tract level only, but each record has additional geographic identifiers for larger geographies (such as county, metropolitan areas, places) so it is possible for you to summarize the tracts in other database or statistical programs.

The 1970 and 1980 censuses did not have full tract coverage of the US. The tracts are predominantly located in urban areas. By 1990 the US was fully tracted and all parts of the country were covered.

The actual remapping procedure for converting data from 1970, 1980, and 1990 tracts to 2000 tract boundaries is quite complicated. Those wishing a technical explanation of this task should consult Appendix J of the Data Users' Guide. The basic procedure was to use geographic information system (GIS) software to overlay the boundaries of 2000 tracts with those of an earlier year. This allowed us to identify how tract boundaries had changed between censuses. We then used 1990 block data to determine the proportion of persons in each earlier tract that went into making up the new 2000 tract. For example, if a 1990 tract split into two tracts for 2000, the population may not have been divided evenly. Our method allows us to determine the exact weight to allocate to each portion.

These population weights were then applied to the various 1970, 1980, and 1990 tract level NCDB variables to convert them to 2000 tract boundaries. The population weights were used to convert all variables based on counts of persons, households, and housing units, all counts based on subpopulations (such as black persons or elderly households), and all aggregate data (such as aggregate household income). Proportions (such as the proportion of Hispanic persons) were remapped by first converting the respective numerator and denominator values (Hispanic persons and total persons, respectively) and then recalculating the proportion.


Neighborhood Change Database Variables

The data in the Neighborhood Change Database are constructed from an extensive subset of the Census variables for each decade (about 1,000 variables for each year).

The data in the Neighborhood Change Database is:

For a more complete description of the variables, geographies and methodologies please refer to the Data Users' Guide. The individual sections are pdfs, their size is indicated in parentheses:


All products come with a 30 Day Money Back Guarantee.
We accept purchase orders and most major credit cards.
You can place an order by fax (print out an order form) and fax the printed copy to 732-651-2721. You can also call us at 1-800-577-6717 and place an order.


Footer
Phone: 1-800-577-6717   Fax: 1-732-651-2721