Online Program

Complex Methods for Special Populations: The U.S. Commercial Building Energy Consumption Survey
Thomas J Leckey, U.S. Energy Information Administration 
*Joelle D Michaels, U.S. Energy Information Administration 
Eileen M O'Brien, U.S. Energy Information Administration 

Keywords: energy, buildings, commercial, area frame, multi-stage area probability sampling

The Commercial Buildings Energy Consumption Survey (CBECS) has been conducted by the U.S. Energy Information Administration (EIA) every 3 to 5 years since 1979. It is the only source of national and sub-national data on energy characteristics and use in commercial buildings. CBECS data are widely used for forecasting, benchmarking, and modeling domestic energy demand in commercial buildings and play a direct role in energy program and policy formation and evaluation. CBECS relies on a multi-frame list and area probability sample design to permit inferences from a small stratified sample to the population of commercial buildings. Because the unit of analysis in CBECS is unique, no list or frame exists from which to design the survey. This paper deals with the challenges that EIA has faced over the years in its choice of an atypical unit of analysis for the CBECS—the building.