Embedded Files
The RECVD Team has developed extensive protocols to improve locational and classification accuracy using systematic checks to identify and limit positional error, duplication, and misclassification; we have refined these protocols using comparison to alternative data sources.
Visual of NETS Categories and Overlaps
Visual of NETS Categories and Overlaps
NETS Categories
NETS Categories
Two category types have been created in the NETS datasets (with the UHCID or aggregated by CT and ZCTAs):
Auxiliary: Auxiliary categories are defined in relatively small subsets of NETS records that can be combined into versions that are more tailored to specific outcomes or allow for sensitivity analyses by including or not including certain Auxiliary categories. Not all Auxiliary categories are independent and may overlap. Additionally, some of the categories are not expected to be meaningful when used on their own (e.g. chain pharmacies used to construct total pharmacies).
Main: Main categories are groupings of Auxiliary categories that are most likely to be used across multiple projects or analyses, limiting the need to make customization each time, and increasing consistency. For most analyses/purposes the Main categories will be more useful than Auxiliary categories. Certain Main categories are broader than others: for example, the broad Main category "all food stores" encompasses the Auxiliary categories that are also contained in the narrower Main category "all supermarkets."
Cautions: Individual business records may be found in more than one category for several reasons, including due to word and name searches categorization of a business is assigned based on its standard industrial classification ( SIC), and also according to the results of a word or name search. The primary category for a business is assigned based on its 8- digit SIC code classification and also may incorporate a text search for words or chain names. Searches employing key words or names from chain name lists were used to assign establishments to categories, and this is thought to better represent the establishment type, than when relying on SIC code alone. For example, a “PIZZA HUT” with SIC code of 58120300 would be classified as “Fast Food – SIC code based definition (FFS)” based on SIC code, “Pizza (PIZ)” based on having the word pizza, and “Fast Food Quick Service (QSV)” based on Pizza Hut being included in the Technomics/R&I chain name list.
It is NOT reccomended to simply add together counts from several existing categories, since not all categories are independent. When combining categories, caution should be taken to ensure individual business records are not double counted. For analysts wishing to construct custom combinations we have created a “hierarchy” version.
NETS Hierarchy System
NETS Hierarchy System
Since records could be categorized into more than one category, a hierarchy system was established to classify each business record into the research defined “best fit” category for that record such that each record is only classified into no more than one Auxiliary category. The decision for “best fit” was defined by which categories were most likely to be used within the aims of the RECVD study grant where food stores, physical activity, and walkability were the main focus. The hierarchy uses the auxiliary categories to create non-overlapping counts grouped together using the definitions in the Main categories to produce the hierarchy-based or modified versions of the Main categories as needed.
Hierarchy technical definitions and general descriptions are based on auxiliary category descriptions found in an Appendix of the NETS GIS Codebook (see auxiliary categories section for these details). The variable name (3-letter code) for the hierarchy categories is the same as the auxiliary and main categories but with an “h” on then end of the 3-letter code.
The NETS Technical Document can also be found in an Appendix of the NETS GIS Codebook.
To learn more about NETS Datasets and NETS GIS Codebook, please visit the About the Datasets page.
Resources
Resources
- NETS_Catalogue_3LTR_Variables: Catalogue of the three letter variables for Main and Auxiliary Categories and colored coded according to the NETS Visual above. This document is a useful tool to help new users become familiar the NETS data.
- NETS_Categories_Diagram: Download the pdf of the NETS Visual and adjust the zoom to examine categories more closely.
- Priority Variables: A list of variables commonly used for analysis can be found in the Appendix of the NETS GIS Codebook
- Wall & Associates Documents: The NETS data is compiled by Walls & Associated in collaboration with Dun & Bradstreet, using annual snapshots from Duns Marketing Information files. When using the NETS Business Level data as derived by Walls and Associates, please reference the users guides and crosswalks in the Walls & Associates Documents portal.
This collection of citations represent different uses of the NETS datasets by the Urban Health Collaborative based team in collaboration with external partners from the Built Environment Research Group at Columbia University and the Multi-Ethnic Study of Atherosclerosis (MESA) study.
Updates 03/19/2019
Google Sites
Report abuse