Researchers use geography in two different ways:

  1. To merge variables from external sources of data to add context or policy. Typically, following the merge, these geographic variables are removed. However, if you need to keep these variables for analysis purposes as well, please explain so in your proposal.
    Example: Neighborhood characteristics from Census can be added to examine their relationship to obesity.
  2. To answer a research question for a smaller geographical area, such as region.
    Example: To examine regional differences in the prevalence of asthma.

It is important to note that although smaller levels of geography are available for NCHS surveys, the majority of surveys (excluding SLAITS, Natality, and Mortality) are only representative at the regional and national level. It is inappropriate to make estimates based on NCHS data for areas smaller than the area for which the sample frame was designed. These smaller levels of data only exist for the purpose of adding contextual information from external sources of data.

Geocodes by NCHS Survey

Urban Rural

Some NCHS surveys include a measure of urban/rural, whereas others do not (please see the specific geocodes listed by survey above for details). If you are interested in this measure and the survey does not include the variable, it is possible to use state and county to merge NCHS to other sources of data that have a definition of urban/rural. One example of where you can obtain this measure is the NCHS Urban-Rural Classification Scheme for Counties . If you are interested in using urban/rural, please include state and county in your list of restricted variables and the NCHS urban/rural continuum as an additional NCHS data source. State and county will be removed after the urban/rural codes are merged.

Page last reviewed: October 13, 2015