U.S. Energy Information Administration logo
Skip to sub-navigation
‹ Consumption & Efficiency

Residential Energy Consumption Survey (RECS)

1997 RECS Survey Data 2020 | 2015 |2009 | 2005 | 2001 | 1997 | 1993 |

Choose which format you would prefer to download:

1997 PUBLIC USE DATA FILES (ASCII FORMAT)

WHAT IS RECS?

The Residential Energy Consumption Survey (RECS) is a national sample survey of housing units. The survey collects statistical information on the consumption of and expenditures for energy in housing units along with data on energy-related characteristics of the housing units and occupants. The survey is restricted to housing units that are the primary residence of the occupants; the RECS does not cover vacant housing units, second homes, or vacation units. RECS is conducted by the Energy Information Administration of the U.S. Department of Energy. The RECS was conducted in 1978, 1979, 1980, 1981, 1982, 1984, 1987, 1990, 1993, and 1997. For the 1997 RECS, data were obtained for 5,900 housing units. Energy-related characteristics of the housing units and occupants are obtained in an on-site personal interview with the occupants. Energy consumption and expenditures information are obtained from the energy suppliers to the responding households during the Energy Suppliers Survey that follows the household personal interview.

WHAT ARE THE RECS PUBLIC USE FILES?

The 1997 RECS Public Use Files are microdata files that contain 5,900 records, representing housing units from the 50 States and the District of Columbia. Each record corresponds to a single responding, in-scope sampled housing unit and contains information for that unit about the size, year constructed, types of energy used, energy-using equipment, conservation features, energy consumption and expenditures (electricity, natural gas, fuel oil, kerosene, and LPG), and the amount of energy used for five end uses: space heating, air-conditioning, water heating, refrigeration, and other.

WHAT IS THE GEOGRAPHIC LEVEL OF DATA AVAILABLE?

RECS data are available for the four Census regions and nine Census divisions. State-level data are available for the four most populated States (California, Texas New York, and Florida).

WHAT IS THE FORMAT OF THE PUBLIC USE FILES?

The Public Use Files are constructed in two formats—ASCII and Microsoft ACCESS97. Both formats contain the same detail of information, with the notable exception that the ACCESS97 database has replaced all alphanumeric coding with English labeling. In ASCII files all records are comma-delimited with fixed column positions. The creation of comma-delimited ASCII files enables use of EIA's public-use files by a wide spectrum of data users. However, EIA realizes that some users are well versed in the use and manipulation of common database systems. Unfortunately, EIA does not have the resources to provide public-use files in multiple database formats. However, EIA has created an ACCESS97 version of the 1997 RECS because of the internal use of the Microsoft ACCESS97 software. The continuation of multiple format releases is highly dependent upon the use and feedback from our data users. Let us know if you find the ACCESS97 file helpful.

HOW ARE THE PUBLIC USE FILES ORGANIZED?

Because of the size of the RECS database, the variables were grouped into 12 files by section of Household Questionnaire:

  1. Section A: Housing Unit Characteristics
  2. Section B: Kitchen Appliances
  3. Section C: Other Appliances
  4. Section D: Space heating
  5. Section E: Water heating,
    Section F: Air conditioning,
    Section G: lights, doors, windows, and insulation
  6. Section H: Fuels Used and Fuels Payment Method
  7. Section I: Fuel Bill and Non-Residential Uses on Bill
  8. Section J: Household Characteristics
  9. Section K: Energy Assistance,
    Section L: EPA Energy Star Program
  10. Characteristic of Energy Supplier Data
  11. Energy Consumption
  12. Energy Expenditures

VARIABLES ON EVERY FILE

Several variables are frequently used in the analysis of residential energy data. These include the type of housing unit, the geographic location of the unit, and weather data for the location of the unit. The nine variables on all 12 files are:

  1. DOEID (unique housing unit identifier)
  2. NWEIGHT (household weight)
  3. MQRESULT (mail questionnaire identifier)
  4. TYPEHUQ (type of housing unit)
  5. REGIONC (Census region)
  6. DIVISION (Census division)
  7. LRGSTATE (indicator for California, Texas, New York, and Florida)
  8. HDD65 (heating degree-days to 65 degrees for 1997)
  9. CDD65 (cooling degree-days to 65 degrees for 1997)

HOW TO MERGE FILES

Each of these 12 files can be used by itself or be merged with other files. By merging files together, a new file can be created that contains, for each respondent, variables from two or more files. The variable DOEID can be used to link the files.

HOW TO USE WEIGHTS

The RECS sample was designed so that survey responses could be used to estimate characteristics of the national stock of occupied housing units. In order to arrive at national estimates from the RECS sample, base sampling weights for each housing unit, which were the reciprocal of the probability of that building being selected into the sample, were calculated. Therefore, a housing unit with a base weight of 10,000 represents itself and 9,999 similar, but unsampled housing units in the total stock of occupied residential housing units. The base weight is further adjusted to account for nonresponse bias. Finally, ratio adjustments were used to ensure that the RECS weights add up to Current Population Survey estimates of the number of households. The variable NWEIGHT in the data file is the final weight.

EXAMPLE 1: SINGLE RESPONSE
The respondent with DOEID = 5198 has NWEIGHT = 8,064. Hence this respondent represents a total of 8,064 households. The respondent used 820 gallons (GALLONFO = 820) of fuel oil. Hence, the respondent contributed 820 times 8,064 = 6,600,000 gallons to the estimated national total fuel oil consumption.
EXAMPLE 2: USING NWEIGHT TO ESTIMATE NUMBER OF HOUSEHOLDS
There were 710, out of the 5,900 RECS respondents, that used fuel oil in their homes (USEFO = 1). Most, but not all, of these households use fuel oil for space heating. The sum of NWEIGHT over these 710 cases is 9,957,479. Hence, the estimated number of households that use fuel oil is 10,000,000.
EXAMPLE 3: USING NWEIGHT TO ESTIMATE PERCENTAGE OF HOUSEHOLDS
The sum of NWEIGHT over all 5,900 cases is 101,481,171. This is also an estimate of the total number of households as of July 1997. Hence, the estimated percent of households that use fuel oil (for any use in the home) is (9,957,479/101,481,171) times 100 equals 9.8 percent.
EXAMPLE 4: USING NWEIGHT TO ESTIMATE TOTAL CONSUMPTION
To estimate the total fuel oil consumption, multiply NWEIGHT times GALLONFO for the 710 cases where fuel oil is used in the home (USEFO = 1), then sum the product over the cases where USEFO = 1. The resulting estimate is 7,273,294,433 gallons. This should be rounded to 7.3 billion gallons or 7,273 million gallons.
EXAMPLE 5: USING NWEIGHT TO ESTIMATE AVERAGE CONSUMPTION
The sum of NWEIGHT over cases where USEFO =1 is 9,957,479. Hence the estimated average fuel oil consumption, in homes that use fuel oil, is 7,273,294,433/9,957,479 = 730 gallons.

MAIL RESPONSES

If the field interviewers were not successful in obtaining a personal interview, a short mail questionnaire was mailed to the housing unit. Variables not on the mail questionnaire were then imputed for the housing unit using a hot deck procedure. There were 181 observations obtained via a mail questionnaire. These 181 records can be identified using the variable MQRESULT.

FUEL USAGE INDICATORS

The variables USEEL, USEFO, USEKERO, USELP, and USENG are indicator variables for the use electricity, fuel oil, kerosene, LPG, and natural gas in the housing unit. They are on three files. They were obtained using section H of the questionnaire and they are indicator variables that equal 1 if the households uses the corresponding fuel and 0 otherwise. In addition to being placed on the file with other section H data, they were also placed on the consumption data file and the expenditures data file.

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 missing values for respondents that are otherwise considered to be complete. Missing values for many, but not all, of the variables were imputed in 1997. The imputation flag indicates whether the corresponding non-Z variable was based upon reported data (Z variable = 0) or was imputed (Z variable = 1). There are no corresponding "Z variables" for variables from the RECS questionnaire that were not imputed, variables where there was no missing data, and variables that are not from the questionnaire. The missing data codes for the consumption and expenditure data are contained in the "Characteristics of Energy Supplier Data" file.

HOW IS THE SURVEY RESPONDENT'S CONFIDENTIALITY PROTECTED?

There are no respondent names and address on these files. 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 housing unit. Local geographic identifiers and National Oceanic and Atmospheric Administration Weather Division identifiers are not included on these data files.

In addition, values for HDD65, CDD65, ELECRATE, and UGASRATE were altered slightly to mask the exact geographic location of the housing unit.

LINKS TO EACH DATA FILE AND SUPPORTING DOCUMENTATION

For each data file, a codebook is provided (both files are in ASCII format). For files based upon the Household Questionnaire, the corresponding section of the questionnaire is provided (PDF format). To view and/or print PDF files (requires Adobe Acrobat Reader) Download Adobe Acrobat Reader .

Note: To DOWNLOAD one of the Text or PDF files below, click on the file of your choice to open it, then select FILE and SAVE AS, save file to your hard drive or a disk.

Microdata Files
by Topic Data Files Codebooks Questionnaire Release Date
File 1: Housing Unit Characteristics TXT TXT Section A 11/22/2009
File 2: Kitchen Appliances TXT TXT Section B 11/22/2009
File 3: Other Appliances TXT TXT Section C 11/22/2009
File 4: Space Heating TXT TXT Section D 11/22/2009
File 5: Water Heating, A/C, and Miscellaneous TXT TXT Sections E, F and G 11/22/2009
File 6: Fuels Used and Fuel Payment TXT TXT Section H 11/22/2009
File 7: Fuel Bills and Non-Residential Uses TXT TXT Section I 11/22/2009
File 8: Household Characteristics TXT TXT Section J 12/20/2009
File 9: Energy Assistance and Housing Unit Square Footage TXT TXT Section K and L 12/20/2009
File 10: Characteristics of Energy Supplier Data TXT TXT 12/20/2009
File 11: Energy Consumption TXT TXT 12/20/2009
File 12: Energy Expenditures TXT TXT 1/10/2000

1997 PUBLIC USE DATA FILES IN ACCESS MDB FORMAT

WHAT IS RECS?

The Residential Energy Consumption Survey (RECS) is a national sample survey of housing units. The survey collects statistical information on the consumption of and expenditures for energy in housing units along with data on energy-related characteristics of the housing units and occupants. The survey is restricted to housing units that are the primary residence of the occupants; the RECS does not cover vacant housing units, second homes, or vacation units. RECS is conducted by the Energy Information Administration of the U.S. Department of Energy. The RECS was conducted in 1978, 1979, 1980, 1981, 1982, 1984, 1987, 1990, 1993, and 1997. For the 1997 RECS, data were obtained for 5,900 housing units. Energy-related characteristics of the housing units and occupants are obtained in an on-site personal interview with the occupants. Energy consumption and expenditures information are obtained from the energy suppliers to the responding households during the Energy Suppliers Survey that follows the household personal interview.

WHAT ARE THE RECS PUBLIC USE FILES?

The 1997 RECS Public Use Files are microdata files that contain 5,900 records, representing housing units from the 50 States and the District of Columbia. Each record corresponds to a single responding, in-scope sampled housing unit and contains information for that unit about the size, year constructed, types of energy used, energy-using equipment, conservation features, energy consumption and expenditures (electricity, natural gas, fuel oil, kerosene, and LPG), and the amount of energy used for five end uses: space heating, air-conditioning, water heating, refrigeration, and other.

WHAT IS THE GEOGRAPHIC LEVEL OF DATA AVAILABLE?

RECS data are available for the four Census regions and nine Census divisions. State-level data are available for the four most populated States (California, Texas New York, and Florida).

WHAT IS THE FORMAT OF THE PUBLIC USE FILES?

The Public Use Files are constructed in two formats—ASCII and Microsoft ACCESS97. Both formats contain the same detail of information, with the notable exception that the ACCESS97 database has replaced all alphanumeric coding with English labeling. In ASCII files all records are comma-delimited with fixed column positions. The creation of comma-delimited ASCII files enables use of EIA's public-use files by a wide spectrum of data users. However, EIA realizes that some users are well versed in the use and manipulation of common database systems. Unfortunately, EIA does not have the resources to provide public-use files in multiple database formats. However, EIA has created an ACCESS97 version of the 1997 RECS because of the internal use of the Microsoft ACCESS97 software. The continuation of multiple format releases is highly dependent upon the use and feedback from our data users. Let us know if you find the ACCESS97 file helpful.

HOW ARE THE PUBLIC USE ACCESS97 TABLES ORGANIZED?

Because of the size of the RECS database, the fieldnames (581 unique names) were grouped into 26 tables by logical relationships within the RECS questionnaire:

  1. Air Conditioning Characteristics –
  2. Auxiliary Fuels Used –
  3. Bottled Gas Usage Characteristics –
  4. Electricity Usage Characteristics –
  5. Energy Assistance
  6. Energy Labels
  7. Final Sample Weights
  8. Fuel Billing Dates
  9. Fuel Oil Usage Characteristics –
  10. Household Characteristics
  11. Housing Structure
  12. Imputation Flags
  13. Interviewer Observations
  14. Kerosene Usage Characteristics –
  15. Kitchen Appliances
  16. Lights Windows and Insulation
  17. Location and Weather
  18. Natural Gas Usage Characteristics –
  19. Other Appliances
  20. Other Usage Characteristics
  21. Solar Usage Characteristics –
  22. Space Heating
  23. Survey Management
  24. Water Heating
  25. Wood Usage Characteristics –
  26. Pub Use Xwalk

Because we have renamed and reorganized the public use files into two formats, the historical user of RECS data may require further documentation on how the two formats link. The table named Pub Use Xwalk in the ACCESS97 file provides such linking; however, a detailed listing has been made available. Note: A "–" sign following a table name (i.e., a suffix) denotes a table with a record number of less than 5,900 housing units. A subset of the records are presented because the eliminated records are not applicable for the table. For example, only households that use the fuel kerosene are include in theKerosene Usage Characteristics table. Such modifications minimize the size of the ACCESS97 file while maintaining the analytical content of the RECS data. Field values that are blank are considered not applicable for that field name. Iin the case where a second refrigerator is not applicable to the household, for example, blank values have been place into the corresponding second refrigerator field name values.

FIELDNAMES and PRIMARY KEYS

Only one fieldname is common to each table: EIAEIAIDNum. This primary key fieldname represents the unique 4-digit identification number that EIA uses to identify a household record. Every attempt has been made to ensure an easy transition to the use of an ACCESS97-based public use file. Fieldnames have been renamed in "English" to guide the data user. In addition, captions for all fieldnames are available in the ACCESS97 file. These captions represent a 40-character definition of the fieldname. If this guidance is not sufficient for your data needs, then it is suggested that you employ the ASCII version of the public use files, along with the specified codebooks.

HOW TO USE WEIGHTS

The RECS sample was designed so that survey responses could be used to estimate characteristics of the national stock of occupied housing units. In order to arrive at national estimates from the RECS sample, base sampling weights for each housing unit, which were the reciprocal of the probability of that building being selected into the sample, were calculated. Therefore, a housing unit with a base weight of 10,000 represents itself and 9,999 similar, but unsampled housing units in the total stock of occupied residential housing units. The base weight is further adjusted to account for nonresponse bias. Finally, ratio adjustments were used to ensure that the RECS weights add up to Current Population Survey estimates of the number of households. The fieldname FinalWeight in the data file is the final weight.

EXAMPLE 1: SINGLE RESPONSE
The respondent with EIAEIAIDNum = 5198 has FinalWeight = 8,064. Hence this respondent represents a total of 8,064 households. The respondent used 820 gallons (EstFOPurGal = 820) of fuel oil. Hence, the respondent contributed 820 times 8,064 = 6,600,000 gallons to the estimated national total fuel oil consumption.
EXAMPLE 2: USING FinalWeight TO ESTIMATE NUMBER OF HOUSEHOLDS
There were 710, out of the 5,900 RECS respondents, that used fuel oil in their homes (UseFOinHome = Yes). Most, but not all, of these households use fuel oil for space heating. The sum of FinalWeightover these 710 cases is 9,957,479. Hence, the estimated number of households that use fuel oil is 10,000,000.
EXAMPLE 3: USING FinalWeight TO ESTIMATE PERCENTAGE OF HOUSEHOLDS
The sum of FinalWeight over all 5,900 cases is 101,481,171. This is also an estimate of the total number of households as of July 1997. Hence, the estimated percent of households that use fuel oil (for any use in the home) is (9,957,479/101,481,171) times 100 equals 9.8 percent.
EXAMPLE 4: USING FinalWeight TO ESTIMATE TOTAL CONSUMPTION
To estimate the total fuel oil consumption, multiply FinalWeight times EstFOPurGal for the 710 cases where fuel oil is used in the home (UseFOinHome = 1), then sum the product over the cases whereUseFOinHome = Yes. The resulting estimate is 7,273,294,433 gallons. This should be rounded to 7.3 billion gallons or 7,273 million gallons.
EXAMPLE 5: USING FinalWeight TO ESTIMATE AVERAGE CONSUMPTION
The sum of FinalWeight over cases where UseFOinHome = Yes is 9,957,479. Hence the estimated average fuel oil consumption, in homes that use fuel oil, is 7,273,294,433/9,957,479 = 730 gallons.

MAIL RESPONSES

If the field interviewers were not successful in obtaining a personal interview, a short mail questionnaire was mailed to the housing unit. Fieldnames not on the mail questionnaire were then imputed for the housing unit using a hot deck procedure. There were 181 observations obtained via a mail questionnaire. These 181 records can be identified using the fieldname MailCodes.

FUEL USAGE INDICATORS

The fieldnames UseELinHome, UseFOinHome, UseKeroinHome, UseLPGinHome, and UseUgasinHome are indicator variables for the use electricity, fuel oil, kerosene, LPG, and natural gas in the housing unit. They are on three files. They were obtained using section H of the questionnaire and they are indicator variables that equal Yes if the households uses the corresponding fuel and No otherwise. These indicator values are used to remove the household records from the ACCESS97 file. Note: A "–" sign following a table name (i.e., a suffix) denotes a table with a record number of less than 5,900 housing units. A subset of the records are presented because the eliminated records are not applicable for the table. For example, only households that use the fuel kerosene are include in the Kerosene Usage Characteristicstable. Such modifications minimize the size of the ACCESS97 file while maintaining the analytical content of the RECS data. Field values tha are blank are considered not applicable for that field name. Iin the case where a second refrigerator is not applicable to the household, for example, blank values have been place into the corresponding second refrigerator field name values.

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

The "FlagforZ fieldnames" are also referred to as "imputation flags." Imputation is a statistical procedure used to fill in missing values for respondents that are otherwise considered to be complete. Missing values for many, but not all, of the fieldnames were imputed in 1997. The imputation flag indicates whether the corresponding non-Z fieldname was based upon reported data (FlagforZ fieldname = No) or was imputed (FlagforZ fieldname = Yes). There are no corresponding "Z fieldnames" for fieldnames from the RECS questionnaire that were not imputed, fieldnames where there was no missing data, and fieldnames that are not from the questionnaire.

HOW IS THE SURVEY RESPONDENT'S CONFIDENTIALITY PROTECTED?

There are no respondent names and address on these files. 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 housing unit. Local geographic identifiers and National Oceanic and Atmospheric Administration Weather Division identifiers are not included on these data files.

In addition, values for HDDtobase651-97to12-97, CDDtobase651-97to12-97, ELRatelocal, and UgasRate were altered slightly to mask the exact geographic location of the housing unit.

DOWNLOAD THE ACCESS OFFICE 97 OR OFFICE 2000 FILES
Zipped ACCESS97 File (5 Megs)
Zipped ACCESS 2000 File (5 Megs)

Specific questions on this product may be directed to:

Consumption and efficiency data