Challenges in Using the National Broadband Map’s Data

As part of the American Recovery and Reinvestment Act of 2009, Congress set forth an ambitious goal of establishing a definitive National Broadband Map. Understandably, researchers are now anxious to use the data received to quantify the extent of broadband availability and to explain the relationships between availability and socio-economic factors. While the National Telecommunications and Information Administration should be lauded for its inaugural effort to complete this Herculean task, it is nonetheless important to recognize that the mapping data currently has many known and yet-to-be-discovered defects. These errors include, but are not limited to, measurement errors and sample selection, both of which can cause severe problems with empirical analyses. In light of the known defects in the data and the lack of a robust data verification process, the analyst must proceed with caution and considerable modesty. Indeed, until a data verification process is put into place and the known defects remedied, all statistical and econometric results using the National Broadband Map data should be viewed with skepticism as such work permits no strong causal and policy-relevant conclusions. Over time, as some of the measurement error and sample selection problems are resolved, econometric models may offer more robust conclusions. As always, empirical research using the Map data should begin with a sound, rational framework explaining observed outcomes