Public Use Microdata

CBECS is a national sample survey that collects statistical information on the consumption of and expenditures for energy in U.S. commercial buildings along with data on energy-related characteristics of the buildings. CBECS is conducted by the Energy Information Administration of the U.S. Department of Energy. The first CBECS was conducted in 1979 and then triennially since 1983. In the 1995 CBECS, there were 6,590 sampled buildings of which 5,766 were successfully interviewed. Energy-related characteristics of the buildings are obtained in an on-site personal interview with the building managers, owners or tenants during the Commercial Buildings Characteristics Survey. Energy consumption and expenditures information are obtained from the energy suppliers to the responding buildings during the Energy Suppliers Survey. Commercial buildings are defined as enclosed roofed and walled structures used predominantly for commercial purposes with floorspace greater than 1,000 square feet. This definition includes buildings such as schools, health care buildings, and religious institutions, as well as office buildings and retail stores.

WHAT ARE THE CBECS PUBLIC USE FILES?

The Public Use Files are microdata files that contain 5,766 records, representing commercial buildings from the 50 States and the District of Columbia. Each record corresponds to a single responding, in-scope sampled building and contains information for that building about the building size, year constructed, types of energy used, energy-using equipment, conservation features, energy consumption and expenditures, and the amount of energy used for nine end uses: space heating, cooling, ventilation, lighting, water heating, cooking, refrigeration, office equipment and other.

WHAT IS THE GEOGRAPHICAL-LEVEL OF DATA AVAILABLE?

CBECS data are available for the 4 Census regions and 9 Census divisions. No state-level data are available.

WHAT IS THE FORMAT OF THE PUBLIC USE FILES?

The Public Use Files are constructed in ASCII and dBASE formats. The records are comma-delimited with fixed column positions.

HOW ARE THE PUBLIC USE FILES ORGANIZED?

The 21 data files are organized by subject matter. The first 15 files contain only building characteristics data, while files 16 through 21 contain consumption and expenditure information. Each file contains the following core variables: (1) building identifier, which is the link between files; (2) adjusted sampling weight; (3) variance stratum and pair member which can be used for calculating variances; (4) Census region and division; (5) square footage; (6) principal building activity; (7) year constructed category; and (8) a set of variables indicating whether electricity, natural gas, fuel oil, district steam or district hot water was used in the building.

The 21 files are grouped into 4 self-extracting files, each containing 5 or 6 files. Files ending in T are in ASCII format and those ending in D are in dBASE format.

Files 1 through 5 are located in C950105T.EXE (ASCII format) or C950105D.EXE (dBASE format) and contain the following files:

File 1: General Building Information and Energy End Uses (record length: 173 bytes) BC95F01T.TXT or BC95F01D.DBF
File 2: Building Activity Percents and Special Measures of Size (record length: 196 bytes) BC95F02T.TXT or BC95F02D.DBF
File 3: Heating Equipment (record length: 147 bytes) BC95F03T.TXT or BC95F03D.DBF
File 4: Cooling, Refrigeration and Water Heating Equipment (record length: 175 bytes) BC95F04T.TXT or BC95F04D.DBF
File 5: End Uses of Major Energy Sources, Electricity Generation and Transportation Gas (record length: 165 bytes) BC95F05T.TXT or BC95F05D.DBF

Files 6 through 10 are located in C950610T.EXE (ASCII format) or C950610D.EXE (dBASE format) and contain the following files:

File 6: Minor Energy Sources and End Uses for Minor Energy Sources (record length: 164 bytes) BC95F06T.TXT or BC95F06D.DBF
File 7: Multibuilding Facilities, Fuel Switching, and Special Space Functions (record length: 105 bytes) BC95F07T.TXT or BC95F07D.DBF
File 8: Lighting Percents, Lighting Equipment, Conservation, and Renewable Energy Sources (record length: 164 bytes) BC95F08T.TXT or BC95F08D.DBF
File 9: Imputation Flags for File 1 (record length: 132 bytes) BC95F09T.TXT or BC95F09D.DBF
File 10: Imputation Flags for Files 2 and 7 (record length:120 bytes) BC95F10T.TXT or BC95F10D.DBF

Files 11 through 15 are located in C951115T.EXE (ASCII format) or C951115D.EXE (dBASE format) and contain the following files:

File 11: Imputation Flags for File 3 (record length: 126 bytes) BC95F11T.TXT or BC95F11D.DBF
File 12: Imputation Flags for File 4 (record length: 148 bytes) BC95F12T.TXT or BC95F12D.DBF
File 13: Imputation Flags for File 5 (record length: 156 bytes) BC95F13T.TXT or BC95F13D.DBF
File 14: Imputation Flags for File 6 (record length: 160 bytes) BC95F14T.TXT or BC95F14D.DBF
File 15: Imputation Flags for File 8 (record length 150 bytes) BC95F15T.TXT or BC95F15D.DBF

Files 16 through 21 are located in C951621T.EXE (ASCII format) or C951621D.EXE (dBASE format) and contain the following files:

File 16: Electricity and Sum of Major Fuels (record length: 187 bytes) CE95F16T.TXT or CE95F16D.DBF
File 17: Natural Gas (record length 187 bytes) CE95F17T.TXT or CE95F17D.DBF
File 18: Fuel Oil and District Heat (record length: 222 bytes) CE95F18T.TXT or CE95F18D.DBF
File 19: Consumption of Electricity by End Use (record length: 187 bytes) CE95F19T.TXT or CE95F19D.DBF
File 20: Consumption of Natural Gas, Fuel Oil, and District Heat by End Use (record length 277 bytes) CE95F20T.TXT or CE95F20D.DBF
File 21: Consumption of Major Fuels by End Use (record length: 187 bytes) CE95F21T.TXT or CE95F21D.DBF

HOW ARE VARIABLES FROM SEPARATE FILES LINKED?

In some instances, a CBECS user may want to undertake analysis that requires using a variable from several different files. In those cases, the files have to be linked to one another. The common variable on each file that can be used to merge files is BLDGID6.

HOW ARE THE VARIABLES THAT BEGIN WITH A Z DIFFERENT FROM THE NON-Z VARIABLES?

The "Z variables" are also referred to as "imputation flags." Imputation is a statistical procedure used to fill in values for missing items. Many, but not all, of the variables were imputed in 1995. The imputation flag indicates whether the corresponding non-Z variable was reported, imputed, missing, or inapplicable.

HOW ARE THE SAMPLE DATA WEIGHTED TO ARRIVE AT NATIONAL ESTIMATES?

The CBECS sample was designed so that survey responses could be used to estimate characteristics of the entire commercial buildings stock nationwide. In order to arrive at national estimates from the CBECS sample, base sampling weights for each building, which were the reciprocal of the probability of that building being selected into the sample, were calculated. Therefore, a building with a base weight of 1,000 represents itself and 999 similar, but unsampled buildings in the total building stock. The base weight is further adjusted to account for nonresponse bias. The variable ADJWT6 in the data file is the adjusted weight. In order to obtain a weighted estimate, each sample building's value must be multiplied by the building's weight.

HOW IS THE SURVEY RESPONDENT'S CONFIDENTIALITY PROTECTED?

The EIA does not receive nor take possession of the names or addresses of individual respondents or any other individually identifiable energy data that could be specifically linked with a building respondent. All names and addresses are maintained by the survey contractor for survey verification purposes only. Geographic identifiers and National Oceanic and Atmospheric Administration Weather Division identifiers are not included on any data files delivered to EIA. Geographic location information is provided to EIA at the Census division level. In addition, specific building characteristics that could uniquely identify a particular responding building are masked on any data provided to EIA by the survey contractor.

These specific variables that have been modified to protect the confidentiality of respondents are:

Square Footage: For buildings over one million square feet, the numeric square footage was replaced with the weighted average square footage of all responding buildings over one million square feet. Separate weighted means were calculated for each of the four Census regions. For buildings one million square feet or less, the numeric square footage was rounded to within 5 percent of the upper limit of the buildings' square footage categories. If the rounded value fell below the lower limit of the category, the value was coded at the lower limit -- for example, buildings in the range of 5,001 to 10,000 square feet were rounded to the nearest 500 square feet (except that buildings rounding to 5,000 were coded as 5,001.)

Number of Workers: For buildings where the numeric number of workers was between 2,500 and 4,999, the reported number was rounded to the nearest 250. For buildings where the numeric number of workers was 5,000 or more, the reported numeric number of workers was replaced with the weighted average number of workers of all responding buildings with 5,000 or more workers. Separate weighted means were calculated for each of the four Census regions.

Number of Floors: The upper range of the number of floors was replaced with two categories: 15 to 25 floors (coded as 994 on the file) and over 25 floors (coded as 995 on the file).

Special Measures of Occupancy: Seven special measures of occupancy are included in the 1995 CBECS (seating capacity for religious buildings, public assembly buildings, education buildings, and food service buildings; licensed bed capacity for in-patient health care and skilled nursing buildings; and number of guest rooms for lodging buildings). These numbers were rounded to the following: Fewer than 25 units (no rounding performed); 25-49 units (rounded to nearest 5); 50-99 units (rounded to nearest 10); 100-249 units (rounded to nearest 25); 250-499 (rounded to nearest 50); 500-999 units (rounded to nearest 100); 1,000-2,499 units (rounded to nearest 250); 2,500-4,999 (rounded to nearest 500); 5,000 or more units (rounded to nearest 1,000).

Weather Variables: Heating and cooling degree-days (HDD656 and CDD656, respectively) that have a base 65 degrees Fahrenheit are included on the data files for each building. The heating degree-day variable has been inflated or deflated by a random percentage, normally distributed with mean zero and standard deviation of 2.0. The files also include annual mean (TEMPAVG6) and standard deviation (TEMPSTD6) of daily average temperatures. for each building. The mean and standard deviation of temperature and the base 65 degrees Fahrenheit cooling degree-days have been modified to be consistent with the modified heating degree-days.

SPECIAL NOTES

Reporting Period Shifts: For energy consumption and expenditures, EIA requested that energy suppliers provide data covering the period from December 1994 through January 31, 1996. However, some suppliers were unable to provide data for this period. Therefore, for electricity, natural gas, and district energy sources (steam and hot water), variables are included in the file to indicate reporting period shifts. A negative value indicates the number of days that the period was shifted back into 1994, and a positive value indicates the number of days shifted forwarded into 1996. A zero indicates that the data represent calendar year 1995.

Energy End-Use Estimation Methodology: The end-use estimates had two main sources: (1) survey data collected by the CBECS and (2) building energy simulations provided by the Facility Energy Decision Screening (FEDS) system. The CBECS provided data on building characteristics and total energy consumption, for all end uses, for a national sample of commercial buildings, Using data collected by the CBECS, the FEDS engineering modules were used to produce estimates of energy consumption by end use. The engineering estimates were then statistically adjusted to match the CBECS total energy consumption. (For more methodological detail see End-Use Estimation Methodology: Technical Note)

HOW ARE THE DATA FILES, FILE LAYOUT, AND VARIABLE CODES ACCESSED?

The 1995 CBECS microdata file layout and variable codes are located here in PDF format.

To view and/or print PDF files (requires Adobe Acrobat Reader ) - Download Adobe Acrobat Reader . The file layout contains the variable names, a brief description of the variables, and the variable position on the files.

The 1995 CBECS microdata files are located on the FTP site in ASCII and dBase formats :

To download files 1 through 5 click here for ASCII or click here for dBase

To download files 6 through 10 click here for ASCII or click here for dBase

To download files11 through 15 click here for ASCII or click here for dBase

To download files 16 through 21 click here for ASCII or click here for dBase

---

Questions about CBECS may be directed to:

Joelle Michaels
joelle.michaels@eia.gov
Survey Manager