Public Use Microdata File

The 2012 CBECS public use microdata file contains untabulated records about individual buildings. This file is published so that data users can create custom tables that are not available through EIA's pretabulated Detailed Tables.

The microdata file contains 6,720 records. They represent commercial buildings from the 50 States and the District of Columbia. Each record corresponds to a single responding, in-scope sampled building. The sample represents an estimated 5.6 million total buildings in the United States.

Data are available in two formats: a CSV (comma delimited) file and a SAS data file. The variable and response codebook contains the variable length, type, labels, and descriptions of the response codes for each variable. The SAS format codes can be used to assign formats to variables in SAS.

2012 CBECS microdata files and information

Relative standard errors (RSEs) and confidence intervals for survey estimates should be calculated using the replicate weights provided for each of the 6,720 CBECS building cases to correctly account for the complex probabilities of selection in the sampling.

Please note that results tabulated from this file will not exactly match the CBECS Detailed Tables, which use unmasked data.  For public data users who wish to verify their tabulations, a few selected tables are available that use the masked, public use data, both in Excel format (above) or printed in the Appendix of the User's Guide to the 2012 CBECS Public Use Microdata File.

Data users are strongly encouraged to read the User's Guide to the 2012 CBECS Public Use Microdata File. The documentation includes detailed examples of computing estimates and standard errors (RSEs), a description of the variable masking procedures, a list of changes from the 2003 public use files, and sample tables using with the public use data.

The CBECS was conducted using a computer-assisted interview. A professional interviewer administered the questionnaire to the building respondent using a laptop computer. A paper representation of the questionnaire is provided, specially annotated for public use file users, along with the Show Cards, questionnaire aids that contain response lists, examples, and illustrations that accompanied some of the survey questions.

The smallest level of geographic detail available is the Census division, of which there are nine in the United States. No state level identifiers are available on this dataset to ensure the privacy and confidentiality of data for sampled buildings.

Confidentiality

The confidentiality of building respondents is protected through a variety of means, ensuring that it is impossible to identify individual buildings, yet making sure the results are still useful. The public use files do not contain building names, addresses, or any information that might identify an individual building. Detailed information on masking procedures can be found in the User's Guide to the 2012 CBECS Public Use Microdata File.

* August 2016 revision to microdata

Please note an error in the data file that has been corrected as of August 2016. There were nine cases for which the square footage (SQFT) was recorded in earlier versions as 100,001 square feet instead of 10,001 square feet. The square footage category for these cases was correctly indicated as 10,001 to 25,000 square feet. The public use building identifiers (PUBID) for these cases are: 0134, 0890, 2329, 2974, 4800, 5033, 5887, 6524, and 6641.

Questions about CBECS may be directed to:

Joelle Michaels
joelle.michaels@eia.gov
Survey Manager