sample-updating approach that was expected to be cost efficient. After supplementation with lists of buildings that were over 200,000 square feet in size, including federal buildings, colleges, hospitals, and airports, the 2003 area frame was updated using a commercial version of the U.S. Postal Service (USPS) delivery sequence file (DSF), purchased from a vendor licensed by USPS. The DSF is the USPS’s list of all delivery points in the United States. To use the DSF for updating, the list had to be matched to the addresses in the second-stage area frame and the duplicates removed. NORC reported that the removing of duplicates turned out to be a major challenge, in part because of imprecise address records. The use of this new approach to update the survey frame, along with implementation errors made by the contractor, ultimately led to the data quality problems that prevented EIA from being able to release the 2007 CBECS data (U.S. Energy Information Administration, 2011e).

The CBECS was initially conducted in the form of in-person interviews, using paper and pencil. Since 1995, however, it has been conducted using computer-assisted personal interviewing. The 1999 data collection was an exception because, in an effort to control costs, it was done by computer-assisted telephone interviewing. This was made possible by re-interviewing the sample from the 1995 CBECS. The supplier survey has always been conducted by mail.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement