The 1993 Residential Energy Consumption Survey (RECS)
was the first to permit the estimation of annual kilowatthours
(kWh) used for lighting. The survey contained more detailed questions
about the number of indoor lights used for specific amounts of
time and more detailed questions about the use of outdoor lights
than did previous surveys. In addition to these basic questions
on the Household Questionnaire, the 1993 RECS also included a
supplementary questionnaire, administered to a subset of households,
that contained more detailed information about the types of lights
used in the household, the rooms in which they were located, and
the amount of time they were used.
The 1993 RECS is the ninth survey in a series of
nationwide household energy consumption surveys conducted since
1978 by the Energy Information Administration (EIA) of the U.S.
Department of Energy. The survey consists of three parts:
In the 1993 survey, more than 7,000 households were
surveyed, using complex sampling methods. These methods allowed
EIA to collect information from a small number of the 97 million
households in the nation, yet use that information to make inferences
about the total population. [6] The
information collected represents the 97 million households nationwide.
The data collected from the 1993 RECS are presented in two statistical
reports: Housing Characteristics 1993 and Household
Energy Consumption and Expenditures, 1993.
Figure 1.1. Definition of Light
For the purposes of both the RECS Household Questionnaire and the Lighting Supplement Questionnaire, EIA defines a light as every light bulb turned on by a single switch. If a light has only one switch, it is counted as one light, even if it has more than one bulb. A fixture or lamp with two switches controlling different bulbs is counted as two lights. If two switches control the same light (for example, switches at both ends of the hallway control a hall light), that is one light. In the Lighting Supplement Questionnaire, the wattage of lights containing more than one bulb is determined by adding the wattage of each bulb. Outdoor lights are counted on the RECS Household Questionnaire only if the light is controlled by the respondent.
The RECS Household Questionnaire (Form EIA-457A)
asked only a few questions about lighting in the household (see
Appendix A). The survey asked respondents
to consider a typical November weekday and report how many indoor
lights were on 1 to 4 hours a day, 4 to 12 hours a day, and more
than 12 hours per day, and how many of those lights were fluorescent.
Respondents were also asked to describe their outdoor light usage,
by giving information about the types of outdoor lights, the length
of time they were used, and the total wattage of all outdoor lights.
The RECS Lighting Supplement (Form EIA-457H) was
a short questionnaire completed by 474 of the 7,111 households
that responded to the RECS Household questionnaire. The 474 households
are a proportional sample of all households in the United States.
The oversampling that occurred in the 1993 RECS was not retained
in the Lighting Supplement, so all households selected in the
Lighting Supplement have equal sampling weights and represent
the same number of households nationwide. The data collected from
the Lighting Supplement were checked for logic and consistency
and were edited when necessary. However, there were no imputations
for missing data. See Table 1.1 for a comparison of the RECS sample
and the Lighting Supplement subsample.
The primary purpose of the Lighting Supplement was
to collect data to develop a model to predict electricity consumption
for lighting. The Supplement provides information on the number
of lights used 15 minutes or more, the hours used, the room where
each light is located, the type of light, and the approximate
wattage of each light. Although the Supplement provides more detail
than the main questionnaire, it is limited to indoor lights. See
Appendix A for a copy of the Lighting Supplement.
The Detailed Tables in this report contain data from
both the Household Questionnaire and the Lighting Supplement Questionnaire.
The RECS sample was designed to represent the entire U.S. population
of residential households. Consequently, tables that are compiled
from the Household Questionnaire include weighted data. On the
other hand, the main purpose behind the Lighting Supplement Questionnaire
was to gather enough representative data to develop a household-level
model of lighting consumption. Therefore, data from the Lighting
Supplement are not weighted and represent only the households
that participated in the Lighting Supplement. [7]
However, as Table 1.1 shows, the households that participated
in the Lighting Supplement have similar energy-related characteristics
as those households that participated in the RECS Household Questionnaire.
For the sake of simplicity, the remainder of the
report will refer to the RECS Household Questionnaire as the RECS
Survey. The Lighting Supplement Questionnaire will be referred
to as the Lighting Supplement.
Estimating Kilowatthours (kWh)
For each light used 15 minutes or more per day, the
Lighting Supplement asked respondents what types of lights they
had in each room in the house and the approximate wattage of each.
The possible responses for bulb type and wattage are listed in
Table 1.2.
The Supplement also asked respondents how long each
light was used in a day.
Based on the bulb type and approximate wattage, a
specific wattage was assigned to each light type. The bulb type
and the hours of use were used to calculate an estimate of annual
kWh for each household in the supplement. Appendix B gives the
details of this calculation.
Computation of End-Use Estimates
All of the RECS end-use estimates data are determined
by a nonlinear regression that allocates some portion of the total
consumption of each fuel to the specific end uses that consume
that fuel. In previous surveys, end-use consumption for space
heating, air conditioning, water heating, refrigeration, and appliances
was calculated. Because of the data collected in the Lighting
Supplement, it was possible to model lighting consumption of electricity
by using the Supplement and then applying the model to the total
sample to estimate the amount of electricity consumed by lighting
in U.S. households. Appendix B contains a complete description
of the end-use estimation methodology.
| Table 1.1 Comparison of Total RECS Sample and Lighting Supplement Subsample | ||
(n=7,111) | Supplement (n=474) | |
| Census Region | ||
| Northeast | 20.2 | 18.5 |
| Midwest | 24.1 | 24.9 |
| South | 34.7 | 36.9 |
| West | 21.1 | 19.6 |
| Type of Home | ||
| Single Family Detached | 61.5 | 61.1 |
| Single Family Attached | 7.6 | 7.2 |
| Apartment in Building with 2-4 Units | 8.3 | 7.8 |
| Apartment in Building with 5 or More Units | 16.8 | 16.5 |
| Mobile Home | 5.8 | 7.4 |
| Income | ||
| Less than $20,000 | 36.6 | 38.4 |
| $20,000 to $49,999 | 41.7 | 41.8 |
| $50,000 to $74,999 | 13.1 | 13.7 |
| $75,000 or More | 8.6 | 6.1 |
| Floorspace (square feet) | 1875.0 | 1855.0 |
| Electricity Rate (dollars per 1,000 kWh) | 89.2 | 86.1 |
| Number of Household Members | 2.6 | 2.6 |
| Number of Rooms | 5.5 | 6.1 |
Source: Energy Information Administration,
Office of Energy Markets and End Use, Forms EIA-457A-C, E, and
H of the Residential Energy Consumption Survey.
Table 1.2. Light Bulb Types in the RECS Lighting Supplement Questionnaire
| Low (10 to 40 watts) | Short (less than 24 inches) | One Type Only |
| Medium (41 to 149 watts) | Long (24 inches or more) | |
| High (150 watts or more) | Compact |
Source: Energy Information Administration, Office of Energy Markets and End Use, Form EIA-457H of the 1993 Residential Energy Consumption Survey.