The interviews with rental agents or their representatives were conducted in early fall. Altogether, 186 landlords or rental agents were interviewed; these interviews encompassed 382 households. The 382 households represented 59 percent of the 650 total households who were eligible for inclusion in the rental agent survey. Comparisons were made between rental agents' and household respondents' reports: on their building's year of construction; main space-heating and water-heating fuels; main space-heating equipment; fuel for cooking; central airconditioning information; and how the fuels for all of these uses are paid. Each discrepancy was examined and changes were made to the household data whenever it was judged that the rental agent was more knowledgeable than the household respondent on the different items of information. Generally, the person who paid for a specific fuel for a specific use was deemed the more knowledgeable person. However, error resolutions were made only after careful examination and consideration of all available sources of information, including the rental-agent questionnaire, the household questionnaire, and questionnaires of other households located in the same building. Landlords and rental agents were usually judged knowledgeable about the year the building was built and the type of main heating equipment; household respondents were typically deemed reliable concerning central air-conditioning and fuel for cooking. Energy Supplier Survey The overall objective of the energy supplier survey was to provide data to estimate the annual energy consumption and expenditures of sample households. Five energy sources were covered in the supplier survey--electricity, natural gas, fuel oil, kerosene, and LPG.' For each of the energy sources, the goal was to obtain complete consumption records from January 1, 1997, through December 31, 1997. Toward the end of the household interview, each household reported for each use of the energy source, whether or not it was paid for by the household, included in the rent, or paid another way. For the households that paid directly, the respondent was asked for the names, addresses, and telephone numbers of the energy suppliers; these respondents were also asked to sign a form, authorizing the contractor to collect consumption data from the suppliers. Altogether, the energy supplier survey included initial contact attempts with 887 companies (Table A5). Data-Collection Procedures Data-collection procedures for electricity and natural gas companies included at least the following steps: An initial letter from the Director of the Office of Energy Markets and End Use, addressed to the president or other official in the company, outlining the general nature of the request for participation. Enclosures in the letter included a printed statement, "About the Residential Energy Consumption Survey," specimen copies of reporting and authorization forms, and a postage-paid postcard with a checklist of available publications and data tapes. Publications that were checked on the postcard were sent to the energy supplier. A telephone contact to determine the name of the person to whose attention the survey materials should be sent. The mailing of survey materials to the person named as contact person. A letter from the survey contractor thanking the company for its effort. 1Households using LPG only for outdoor cooking grills were not included in the LPG data collection; LPG used by these households is excluded from consumption and expenditures estimates. Data on usage of wood fuel were reported by the household, because it was not practical to collect these data from suppliers as is done with the major home fuels. Unless otherwise noted, consumption of wood is not included in the tables for this report. Energy Information Administration "The total number of companies in the survey was 887; 38 supplied both electricity and natural gas; 2 supplied natural gas and LPG; and 23 supplied fuel oil or kerosene and LPG. "Households were asked for names of their "fuel-oil or kerosene" suppliers. Since most companies who have supplier records furnish both types and often supply both types of fuel to the same household, these companies are reported together. If a respondent reported only "cash and carry" purchases of a bulk fuel (fuel oil, LPG, or kerosene), they were not asked to furnish the name of the supplier. Source: Energy Information Administration, Forms EIA-457A-G of the 1997 Residential Energy Consumption Survey (RECS). RECS Public Use Data Files. A follow-up telephone contact a few days later to answer questions or discuss survey procedures as necessary. The personal contacts established at an early point largely precluded mailings of materials to an inappropriate person and the delays that might develop from such mailings. Procedures for fuel oil or kerosene and LPG suppliers were the same as for electric and natural gas suppliers up through and including the mailing of survey materials to the company person named as the contact. These suppliers, however, most often had only one or two households for which information was to be supplied, and data collection was generally completed by telephone. A pretest of the procedure conducted earlier had indicated a somewhat greater likelihood that suppliers would respond by telephone than as a result of a request to complete and return the forms by mail.2 After the supplier returned the information, additional contact with the suppliers and the households was sometimes required to identify the correct record in the company files. Data-Collection Dates The first set of advance letters was mailed to the energy suppliers in January 1998. The cutoff date for receipt of usable information was June 30, 1998. The last data were received in July 1998 and were used. Data Processing The energy consumption and expenditure statistics presented in this report are based on the individual annual consumption and expenditures amounts for each household. Individual consumption and expenditure amounts are calculated for each household for each of five energy sources (electricity, natural gas, fuel oil, kerosene, and LPG). None of the households that participated in the 1997 RECS used all five energy sources, but the majority of the households did use two or more energy sources. When possible, the annual consumption and expenditure amounts were calculated by using data obtained from the Energy Supplier Survey. 2The test is described in RECS: Consumption and Expenditures--April 1980 Through March 1981, Part 1: National Data, DOE/EIA-0321/1 (Washington, DC, September 1982), Appendix A, "How the Survey Was Conducted." Suppliers that chose to return the forms by mail, however, were not discouraged from doing so. Energy Information Administration The Energy Supplier Survey was conducted for households that paid their own fuel bills directly to the supplier and signed a form to authorize access to their billing records. These limitations meant that imputations of fuel consumption and expenditures were required for households whose fuel bills were included in the rent and for households that did not permit access to their records. Imputations were also required for households when the supplier failed to produce usable billing records. The billing records for a given fuel and a given household were considered missing (and hence nonusable) if: (1) the supplier refused to participate, (2) the supplier did not keep records, (3) the supplier could not find the householder's records, (4) the information provided by the household was insufficient to locate the supplier, or (5) the supplier was no longer in business. Available but nonusable billing records occurred when: (1) the household recently moved into the dwelling unit; (2) the amount of the bill that could be attributed to the housing unit was unknown; or (3) the billing records did not cover the entire amount used by the household. For cash-and-carry purchases, households were asked to provide estimates of consumption and expenditures for kerosene only. In addition, if the household indicated that it had the ability to use LPG, fuel oil, or kerosene but yet planned no purchases during 1997, they were assigned a consumption of zero. See "Annual Consumption and Expenditures" in Appendix B, "Quality of the Data," for more details on the annualization of monthly billing records and imputations for the energy supplier data. Weather and Price Data Weather and price data were added to the household data file. The weather data were taken from National Oceanic and Atmospheric Administration weather station data files of daily minimum and maximum temperatures for 1997. Weather stations were selected on the basis of their proximity and appropriateness in representing the weather experienced by sample households in the secondary sampling unit. In selecting an appropriate, nearby weather station, distance was the major consideration but intervening mountain ranges and the presence of bodies of water were taken into account. Every household record contains weather data, whether or not the household did any space heating or air-conditioning. For each household that used electricity (99.9 percent), the cost of 1,000 kWh reported by the household's electricity supplier on Form EIA-861 was added to the household record. For each household that used natural gas (57.9 percent), the cost of 1,000 cubic feet of natural gas reported by the household's natural gas supplier on Form EIA-176 was added to the household record. Special Data Collection for the Administration for The EIA collects supplemental data during the RECS interview for the Administration for Children and Families (ACF) for their use in program administration of the Low-Income Home Energy Assistance Program (LIHEAP). In the 1997 RECS, most of this information was in Section K of the Household Questionnaire (Form EIA-457A). Respondents with annual incomes and number of household members that might qualify them for assistance were asked a series of questions about the receipt of home energy assistance and lack of heat during the previous winter. RECS also supported the LIHEAP through an ACF-funded oversampling of low-income homes. An annual report to Congress is produced by LIHEAP, which contains data from the RECS. Energy Information Administration EIA does not receive or take possession of the names or addresses of individual respondents or any other individually identifiable energy data that could be specifically linked with a household respondent. All names and addresses and identifiable information are maintained by the survey subcontractor for verification purposes only. The household records that are placed on the public use data file do not have name or address information. Additional measures have been taken to mask the data for further confidentiality protection. Public-Use Data File Preparation The publication, Housing Characteristics 1997, was issued as an electronic report in two phases. First, the housing characteristics tables were produced with the survey data file received in March 1998 and placed on the Web in May 1998. The associated building survey information and analysis were placed on the Web in August 1998. The March 1998 data file contained data from the Household Survey and the adjunct Rental-Agent Survey. The Energy Suppliers Survey data were added and the data on the March 1998 file was updated. The final data file was received in February 1999. This data file was used to produce the tables in this report. The consumption and expenditures tables were placed on the Web on July 2, 1999. The public-use file is a subset of the data on the February 1999 file. Variables that might compromise the confidentiality of the individual respondents were not placed on the public-use file or an error term was added to the variables. The public use file was placed on the Internet in October 1999. Energy Information Administration Survey Estimates and Quality of the Data Introduction Survey Estimates All the statistics published in this report are estimates of population values, such as the number of households using natural gas. These estimates are based on a randomly chosen subset of the universe, the entire population of households. The universe includes all households in the 50 States and the District of Columbia, including households on military installations. The two major types of nonresponse are unit nonresponse and item nonresponse. Unit nonresponse occurs when a sampled household does not participate in the survey. Item nonresponse occurs when a particular item of interest is missing from a completed questionnaire. The next two sections provide details on the procedures followed for each type of nonresponse. Adjustments for Unit Nonresponse Weight adjustment was used to reduce unit nonresponse bias in the survey statistics. A weight was calculated for each sample household, which reflects the selection probability for that household. Adjustments were made to correct for potential biases arising from the failure to list all housing units in the sample area and from the failure to contact all sample housing units. Contacts were not successful with 19.0 percent of the eligible units. Six factors are used in the processing of RECS results to develop an overall weight for each household for which a completed questionnaire, either a personal interview or mailed questionnaire, is obtained. The factors are: the basic weight, a noninterview adjustment, a first-stage ratio estimate, and three second-stage ratio adjustments. The overall household weight is the product of these six factors. The weighting process for 1997 differed from the 1993 procedures for many of the computations. In particular, the weighting process in 1997 was conducted prior to the imputation procedures for item nonresponse. As a result, some of the questionnaire data were missing on the variables used in the weighting process, which required some modifications to the 1993 procedures. With the elimination of the new construction and the lighting supplements present in 1993, the weighting procedures for the non-interview adjustment could be simplified by reverting to the procedures used for 1990. Furthermore, the control totals from the Current Population Survey (CPS) public use files used in the second-stage adjustment process differed slightly from the characteristics used in 1993. The Basic Weight The basic weight is calculated and applied to households at the SSU level. For the 1997 RECS, all households in the same SSU had the same probability of selection and hence the same basic weight: The noninterview adjustment factor (NIAF) compensates for non-response households and for non-household units that were identified during the survey. Basically, this adjustment reflects the ratio of the number of completed and incomplete Energy Information Administration |