Interesting facts from the 1993 RECS are presented graphically in the "Full Report" section:
Congress has mandated that EIA collect, analyze, and disseminate impartial, comprehensive data about energy--how much is produced, who uses it, and the purposes for which it is used. To comply with that Congressional mandate, the EIA conducts two types of surveys:2
Consumption surveys gather information every 2 or 3 years 3 directly from energy end users on the types of energy they use, along with information on the energy-related characteristics of households, commercial buildings, residential vehicles, and manufacturing establishments. The results of these surveys are published in energy-consumption reports, such as this report, and in special analytical reports.4
2 For information on the differences, see Energy Information Administration, Energy Consumption by End-Use Sector, A Comparison of Measures by Consumption and Supply Surveys, DOE/EIA-0533 (Washington, D.C., April 6, 1990). Appendix C of this report includes a summary of the differences for the residential sector.
3 Beginning with the 1994 data year, data are collected from the manufacturing establishments every 2 years.
4 See Appendix F, "Related EIA Publications on Energy Consumption," for a listing of publications from RECS and other EIA consumption surveys.
RECS Methodology
In 1984, EIA began conducting the RECS, a national sample survey of residential housing units and their energy suppliers, every 3 years. Previous surveys were conducted annually from 1978 to 1982. The RECS consists of three parts:
After obtaining permission from respondents, EIA mails questionnaires to their energy suppliers to collect the actual billing data on energy consumption and expenditures.
RECS Sample Revision
EIA made a major investment to update the sample design for the 1993 RECS to reflect the results of the 1990 Census.5 Updating the sample design for the 1,610 RECS sample clusters requires producing about 5,000 maps, identifying boundaries of the sample location, sending personnel to each location to make lists of the housing units there, creating data bases of the lists, and checking the accuracy of the data bases.
Within each RECS sample cluster, the number of housing units changes through construction of new housing units, conversion of commercial space to housing units, demolition of existing housing units, conversion of residential units to other uses, and division of existing housing units into multiple units. Over the 10-year period between major revisions of the sample design, the number of housing units in some sample clusters will decline dramatically, while the number of housing units in other clusters will increase dramatically. Revising the sample design is like preventive maintenance—if it is not done periodically, the machine may cease to function.
As part of the redesign effort, a provision to sample new housing units at a higher rate than older housing units was built into the sample. This oversample of new housing units was included to provide more accurate and more precise data for this important segment of the housing stock. Both the redesign effort and the new housing units oversample increased the number of observations for new housing units for the 1990 RECS (Figure 1.1).
Figure 1.1. Increase in New Home Sample Cases
Another feature of the 1993 sample design is a provision that allows for an increased geographic breakdown of the observations. Before the redesign, the observations could be classified by the nine Census Divisions. The 1993 design allows the observations to be classified by the nine Census Divisions and the four most populous States, which were California, New York, Texas, and Florida.
Data are presented in this report for the four Census regions, however, for the nine Census Divisions and four States, data will be available in the Household Consumption and Expenditures 1993 report. The data files for the 1993 RECS will contain geographic identifiers for the nine Census Divisions and for the most populous States mentioned above.
5 Energy Information Administration, Sample Design for the Residential Energy Consumption Survey, DOE/EIA-0555(94)/1 (August 1994).
RECS Data Used in This Report
The statistics published in this report are based on a sample of 7,111 households from the population of all primary, occupied residential housing units in the United States as of July 1993. 6 As a result, all of the statistics are estimates rather than exact measurements of the population. The 1993 RECS represents 96.6 million households in the 50 States and the District of Columbia. As described in Appendix B, "Quality of the Data," the accuracy of each estimate is indicated by the relative standard error (RSE). No estimates were published that were based on fewer than 10 sample households or that had an RSE greater than 50 percent. All the tables of estimates in the section titled "Detailed Tables" include corresponding RSE’s that are calculated using row and column RSE factors.
EIA gratefully acknowledges the cooperation of the respondents for supplying the information used to produce the estimates in this report.
6 This represents a change from past RECS where the date was November of the survey year. The change was made to give greater weight to the consumption and expenditures data, which are collected for calendar year 1993 for which July 1993 is the midpoint. The change means the time separation between the date used to estimate the universe of households for the 1993 RECS is 2 2/3 years after the date used for the 1990 RECS. When calculating estimates of annual change, one needs to consider that the period covered was less than 3 years.
Organization of the Report
This introductory chapter is followed by two chapters: Chapter 2 highlights some of the more interesting survey findings and presents some data that are not included in the "Detailed Tables", Chapter 3 contains 60 tables that report information at the National level and four Census regions and for various Climate zones. In addition to data on the year the housing unit was constructed and family-income categories, the tables also contain data on single-family homes, apartments in two-to four-unit buildings, apartments in five or more units buildings, and mobile homes.
The tables in Chapter 3 are organized by the following categories: household and housing-unit characteristics (including fuels used), space heating, air-conditioning, appliances, lights, usage indicators, conservation, and major equipment purchases.
These chapters are followed by six technical appendices:
- Appendix A, "How the Survey Was Conducted," describes the sample design and data collection procedures.
- Appendix B, "Quality of the Data," discusses nonsampling error and sampling error that affect the accuracy and precision of the estimates.
- Appendix C, "RECS Coverage Related to EIA Supply Surveys," discusses differences between the estimates from the RECS and estimates from supply surveys.
- Appendix D, "Survey Forms," contains reproductions of the forms used to collect the data presented in this report.
- Appendix E, "U.S. Climate Zones and Census Regions and Divisions Maps," contains maps showing the Census regions and divisions and the climate zones by which the data in this report are organized.
- Appendix F, "Related EIA Publications on Energy Consumption," provides a list of related energy consump-tion publications for readers interested in earlier RECS publications or reports for other sectors.
A Glossary of statistical and technical terms used in this report follows the appendices.
View and/or Print Full Report
EIA Introduces Energy
InfoDisc; New CD-ROM Provides Instant Access To Comprehensive Energy Data,
Analyses
File Last Modified: April 15, 1997
- Contact:
- stephanie.battles@eia.doe.gov
- Stephanie Battles
- Project Manager
- Phone: (202) 586-7237
- Fax: (202) 586-0018
URL: http://www.eia.gov/emeu/recs/recs2e.htm
If you are having any technical problems with this site, please contact the EIA Webmaster at
wmaster@eia.doe.gov
