Measuring food environments: Accuracy of commercial food store lists

Context: The majority of research studies assessing the retail food landscape draw on commercial (e.g., InfoUSA, Dun and Bradstreet) or public (e.g., telephone lists, state departments) sources to identify food vendors. The accuracy of such data in describing the residential food environment has not been systematically assessed.

Purpose: To examine how accurately commercial lists identify and categorize food retailers by comparing these stores availability, location, name, and type with data collected in ground truthing and visits to every store.

Methods: Using a standardized instrument to assess the residential healthy food environment, we visit all food retailers in 5 representative Connecticut towns. We verify each record in the commercial InfoUSA list of food vendors (N=383) via store visits and neighborhood combing. We assess discrepancies between the stores in the commercial list and the actual food environment, including the prevalence of commercially listed stores that cannot be located and verified (e.g., out of business), not listed stores that are open for business, and stores with errors in their names and/or contact information. We also compare how closely categories of convenience, grocery or supermarket stores in the InfoUSA list match the observed types of stores.

Results: Store visits will be conducted in March-April 2009.

Learning objectives: 1. Describe the accuracy of commercially provided lists of food vendors in measuring the residential food environment. 2. Present key problems with using commercial vendor lists and ways to adjust the records.