File layout for File 7 (Records are comma-delimited. First line contains a
list of variables on the file. The remaining lines contain the values of the
variables with fixed column positions.)
Variables on all files
DOEID '4-digit identification number' Columns 1-4
NWEIGHT 'The Final Weight' Columns 6-11
MQRESULT 'Mail Questionnaire Codes' Columns 13
REGIONC 'Census Region' Columns 15
DIVISION 'Census Division' Columns 17
LRGSTATE 'Four Largest States' Columns 19
CDD65 'Cooling Degree-Days to base 65, 1-97 TO 12-97' Columns 21-24
HDD65 'Heating Degree-Days to base 65, 1-97 TO 12-97' Columns 26-30
TYPEHUQ 'Type of Home: as report by Respondent' Columns 32
Variables in File 7 only.
SIGNFORM 'Authorization Form Signed' Columns 34
KFUELOT 'Fuel Charges For Non-Household Use' Columns 36
FARM 'Bills Include Farm/Machine Use' Columns 38
TENANT 'Bills Cover Fuel Of Other Household' Columns 40
BUSINESS 'Bills Cover Business/Office Use' Columns 42
OTHERUSE 'Bills Cover Fuel For Other Use' Columns 44
BILLUG 'Natural Gas Used For Other Purposes' Columns 46
BLPUSE 'LPG Used For Other Purposes' Columns 48
BILLFOIL 'Fuel Oil Used For Other Purposes' Columns 50
BILLKER 'Kerosene Used For Other Purposes' Columns 52
BILLEL 'Electric Used For Other Purposes' Columns 54
BILLUGP 'Percent Natural Gas Bill For Other Purposes' Columns 56
BILLLPGP 'Percent LPG Bill For Other Purposes' Columns 58
BILLFOLP 'Percent Fuel Oil Bill For Other Purposes' Columns 60
BILLKERP 'Percent Kerosene Bill For Other Purposes' Columns 62
BILLELP 'Percent Electric Bill For Other Purposes' Columns 64
Codebook for File 7
Variables DOEID, NWEIGHT, MQRESULT, REGIONC, DIVISION, LRGSTATE CDD65, HDD65,
and TYPEHUQ are on all files.
DOEID '4-digit identification number' Columns 1-4
Included in all files
Numerical Variable
Variable length: 4
The variable DOEID will be on all files. Each respondent is assigned an
unique value of DOEID. It can be used to merge data from one file with
data from another file.
Minimum Value: 1,001
Maximum Value: 6,903
NWEIGHT 'The Final Weight' Columns 6-11
Included in all files
Numerical Variable
Variable length: 6
Minimum Value: 3,841
Maximum Value: 79,878
The sample design for RECS is not a simple random sample. Consequently,
to estimate the number or percent of households/housing units with a
certain characteristic, the sampling weight should be used. The
variable NWEIGHT is the sampling weight for the observation. It is
approximately the inverse of the probability of selection. It gives the
number of households that the response represents in the sample. For
most categorical variables, the codebooks contain both the unweighted
frequency and the weighted frequency. The unweighted frequency gives
the number of observations, the weighted frequency gives survey estimate
of the number of households/housing units for each possible value of the
variable. The weighted frequencies are rounded to the nearest 100,000
households. This is the limit of the accuracy for RECS. When the
unweighted frequency is less than 10 or the weighted frequency is less
than 150,000; then weighted frequency is suppressed. A 'Q' is placed in
the frequency to denote that the number is suppressed. This indicates
that there are too few observation upon which to determine an accurate
estimate.
MQRESULT 'Mail Questionnaire Codes' Columns 13
Included in all files
Categorical Variable
Variable length 1
Mail Questionnaire (MQ)
Authorization Form (AF)
Personal Interview (PI)
1 = MQ Completed, AF Signed
2 = MQ Completed, AF not required
7 = MQ returned after a PI was completed
8 = PI was completed after MQ was returned
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
1 173 3,400,000
2 8 Q
7 1 Q
8 12 200,000
9 5,706 97,600,000
Q = Data withheld because fewer than 10 observation in the category.
When MQRESULT equals 1 or 2, the household and housing unit data was not
obtained via a personal interview. The data were obtained using a
shortened mail questionnaire. A hot-deck imputation procedure was used
to match each mail respondent to a personal interview respondent. The
critical matching variables were FUELHEAT, FUELH2O, TYPECOOL, TYPEHUQ,
and REGIONC. Additional variables used in a scoring function to
determine the best match were MONEYPY, NHSLDMEM, VEHICLES, HHAGE, and
TOTROOMS. The remaining household/housing unit characteristics for the
mail respondents were determined by the values of the matched personal
interview respondent. The energy consumption and expenditures data for
the mail respondents was based upon data obtained from their energy
suppliers. It was not imputed using a hot deck procedure.
Questionnaire data from personal interview respondents sometime is
missing and needs to be imputed. 'Z' variables are used to indicate the
presence of imputed data. For example, if ZURBRUR = 0 then URBRUR was
not imputed. If ZURBRUR = 1 then URBRUR was imputed.
Census Division/Region: A geographic area consisting of several States
defined by the U.S. Department of Commerce, Bureau of the Census. The States
are grouped into nine divisions and four regions.
Region Division States
Northeast New England Connecticut, Maine, Massachusetts,
New Hampshire, Vermont, and Rhode Island
Middle Atlantic New Jersey, New York, and Pennsylvania
Midwest East North Central Illinois, Indiana, Michigan,
Ohio, and Wisconsin
West North Central Iowa, Kansas, Minnesota, Missouri,
Nebraska, North Dakota, and South Dakota
South South Atlantic Delaware, the District of Columbia,
Florida, Georgia, Maryland, North
Carolina, South Carolina, Virginia,
and West Virginia
East South Central Alabama, Kentucky, Mississippi,
and Tennessee
West South Central Arkansas, Louisiana, Oklahoma,
and Texas
West Mountain Arizona, Colorado, Idaho, Montana,
Nevada, New Mexico, Utah, and Wyoming
Pacific Alaska, California, Hawaii, Oregon,
and Washington
REGIONC 'Census Region' Columns 15
Included in all files
Categorical Variable
Variable length 1
1 = Northeast Census Region
2 = Midwest Census Region
3 = South Census Region
4 = West Census Region
Number Weighted
Of Number of
Value Observations Households
1 1,332 19,700,000
2 1,234 24,100,000
3 1,979 35,900,000
4 1,355 21,800,000
DIVISION 'Census Division' Columns 17
Included in all files
Categorical Variable
Variable length 1
1 = New England Census Division
2 = Middle Atlantic Census Division
3 = East North Central Census Division
4 = West North Central Census Division
5 = South Atlantic Census Division
6 = East South Central Census Division
7 = West South Central Census Division
8 = Mountain Census Division
9 = Pacific Census Division
Number Weighted
Of Number of
Value Observations Households
1 490 5,300,000
2 842 14,400,000
3 783 16,900,000
4 451 7,200,000
5 871 18,700,000
6 527 6,300,000
7 581 10,800,000
8 466 6,200,000
9 889 15,600,000
LRGSTATE 'Four Largest States' Columns 19
Included in all files
Categorical Variable
Variable length 1
0 = Other
1 = New York State
2 = California
3 = Texas
4 = Florida
Number Weighted
Of Number of
Value Observations Households
0 4,258 70,300,000
1 399 6,800,000
2 609 11,500,000
3 369 7,000,000
4 265 5,900,000
CDD65 and HDD65 are weather data. The values of CDD65 and HDD65 were based
upon data obtained from the NOAA. A random error was added to both HDD65 and
CDD65 to mask the location of the weather station from which the data was
obtained.
CDD65 'Cooling Degree-Days to base 65, 1-97 TO 12-97' Columns 21-24
Included in all files
Numerical Variable
Variable length 4
Minimum Value: 0
Maximum Value: 5,954
HDD65 'Heating Degree-Days to base 65, 1-97 TO 12-97' Columns 26-30
Included in all files
Numerical Variable
Variable length 5
Minimum Value: 0
Maximum Value: 11,672
Cooling degree days are the number of degrees the average daily
temperature (ADT) is above the base temperature. Heating degree days
are the number of degrees the ADT is below the base temperature. To
calculate daily degree days (base 65 degrees):
ADT = (daily high + daily low)/2
CDD(daily) = 0 if ADT <= 65
= ADT - 65 if ADT > 65
HDD(daily) = 0 if ADT >= 65
= 65 - ADT if ADT < 65.
Annual degrees days are calculated by summing the daily degree days.
The .5 is carried over in the summation. The total is rounded up after
the summation.
TYPEHUQ 'Type of Home: as report by Respondent' Columns 32
Included in all files
Categorical Variable
Variable length: 1
Question (A-2B)
different than RECS93
Variable was constructed form responses to questions
A-2A, A-2B, A-5, and A-9
Imputed for Nonresponse
There is a fine line between the definitions of various types of housing
units. The distinction between a single-family attached unit and a unit
in an apartment building is particularly complex. The collection and
editing of the data on housing type changed from the paper-and-pencil
questionnaire for the 1993 RECS to the CAPI questionnaire for the 1997
RECS. The change in the data collection and editing procedures may have
contributed to changes in the survey results. For example, the estimated
number of occupied single-family attached units increased from 7.3
million for the 1993 RECS to 10.0 million for the 1997 RECS. Conversely,
the number of occupied housing units in buildings with two to four units
decreased from 8.0 million for the 1993 RECS to 5.6 million for the 1997
RECS.
1 = Mobile Home
2 = Single-Family detached
3 = Single-Family Attached
4 = Apartment in Building containing 2-4 units
5 = Apartment in Building Containing 5 or more units
Number Weighted
Of Number of
Value Observations Households
1 402 6,300,000
2 3,661 63,800,000
3 552 9,900,000
4 351 5,600,000
5 934 15,800,000
The remaining variables a present only in File 7.
SIGNFORM 'Authorization Form Signed' Columns 34
Categorical Variable
Variable Length 1
Question I-1
Question same as RECS93
0 = Authorization Form Not Signed
1 = Authorization Form Signed
6 = Don't Know
7 = Refused
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 480 8,500,000
1 5,131 88,500,000
6 1 Q
7 1 Q
9 287 4,500,000
Q = Data withheld because fewer than 10 observation in the category.
KFUELOT 'Fuel Charges For Non-Household Use' Columns 36
Categorical Variable
Variable Length 1
Question I-8A
Question different than RECS93
In RECS 93 Used Exhibit Card
0 = No
1 = Yes
6 = Don't Know
7 = Refused
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 5,487 94,800,000
1 109 1,900,000
6 5 Q
7 7 Q
8 4 Q
9 288 4,500,000
Q = Data withheld because fewer than 10 observation in the category.
FARM 'Bills Include Farm/Machine Use' Columns 38
Categorical Variable
Variable Length 1
Question I-8B
Question same as RECS93
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 77 1,400,000
1 31 500,000
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
TENANT 'Bills Cover Fuel Of Other Household' Columns 40
Categorical Variable
Variable Length 1
Question I-8C
Question same as RECS93
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 67 1,100,000
1 41 800,000
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
BUSINESS 'Bills Cover Business/Office Use' Columns 42
Categorical Variable
Variable Length 1
Question I-8D
Question same as RECS93
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 77 1,400,000
1 31 500,000
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
OTHERUSE 'Bills Cover Fuel For Other Use' Columns 44
Categorical Variable
Variable Length 1
Question I-8E
Question same as RECS93
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 89 1,600,000
1 19 300,000
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
BILLUG 'Natural Gas Used For Other Purposes' Columns 46
Categorical Variable
Variable Length 1
Question I-8F
Question same as RECS93
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 68 1,200,000
1 40 700,000
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
BLPUSE 'LPG Used For Other Purposes' Columns 48
Categorical Variable
Variable Length 1
Question I-8G
Question same as RECS93
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 93 1,700,000
1 15 200,000
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
BILLFOIL 'Fuel Oil Used For Other Purposes' Columns 50
Categorical Variable
Variable Length 1
Question I-8H
Question same as RECS93
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 103 1,800,000
1 5 Q
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
BILLKER 'Kerosene Used For Other Purposes' Columns 52
Categorical Variable
Variable Length 1
Question I-8I
Question same as RECS93
RECS 97 EXCLUDES COAL OIL From CATEGORY
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 107 1,900,000
1 1 Q
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
BILLEL 'Electric Used For Other Purposes' Columns 54
Categorical Variable
Variable Length 1
Question I-8J
Question same as RECS93
0 = No
1 = Yes
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
0 34 700,000
1 74 1,200,000
8 1 Q
9 5,791 99,600,000
Q = Data withheld because fewer than 10 observation in the category.
BILLUGP 'Percent Natural Gas Bill For Other Purposes' Columns 56
Categorical Variable
Variable Length 1
Question I-8K
Question different than RECS93
RECS 93 Anchors Question With Date
1 = Very Little (1-4%)
2 = Some (5-33%)
3 = About Half (34-66%)
4 = About Three-Quarters(67-95%)
5 = Most Of It (96-99%)
6 = Don't Know
7 = Refused
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
1 12 200,000
2 6 Q
3 14 200,000
4 3 Q
5 1 Q
6 1 Q
8 3 Q
9 5,860 100,700,000
Q = Data withheld because fewer than 10 observation in the category.
BILLLPGP 'Percent LPG Bill For Other Purposes' Columns 58
Categorical Variable
Variable Length 1
Question I-8L
Question same as RECS93
RECS 93 Anchors Question With Date
1 = Very Little (1-4%)
2 = Some (5-33%)
3 = About Half (34-66%)
4 = About Three-Quarters (67-95%)
5 = Most Of IT (96-99%)
6 = Don't Know
7 = Refused
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
1 3 Q
2 4 Q
3 3 Q
4 2 Q
5 3 Q
9 5,885 101,200,000
Q = Data withheld because fewer than 10 observation in the category.
BILLFOLP 'Percent Fuel Oil Bill For Other Purposes' Columns 60
Categorical Variable
Variable Length 1
Question I-8M
Question same as RECS93
RECS 93 Anchors Question With Date
1 = Very Little (1-4%)
2 = Some (5-33%)
3 = About Half (34-66%)
4 = About Three-Quarters (67-95%)
5 = Most Of IT (96-99%)
6 = Don't Know
7 = Refused
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
3 3 Q
6 1 Q
8 1 Q
9 5,895 101,400,000
Q = Data withheld because fewer than 10 observation in the category.
BILLKERP 'Percent Kerosene Bill For Other Purposes' Columns 62
Categorical Variable
Variable Length 1
Question I-8N
Question different than RECS93
RECS 93 Anchors Question With Date
1 = Very Little (1-4%)
2 = Some (5-33%)
3 = About Half (34-66%)
4 = About Three-QUARTERS (67-95%)
5 = Most Of IT (96-99%)
6 = Don't Know
7 = Refused
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
1 1 Q
9 5,899 101,500,000
Q = Data withheld because fewer than 10 observation in the category.
BILLELP 'Percent Electric Bill For Other Purposes' Columns 64
Categorical Variable
Variable Length 1
Question I-8O
Question different than RECS93
RECS 93 Anchors Question With Date
1 = Very Little (1-4%)
2 = Some (5-33%)
3 = About Half (34-66%)
4 = About Three-QUARTERS (67-95%)
5 = Most Of IT (96-99%)
6 = Don't Know
7 = Refused
8 = No Answer
9 = Not Applicable
Number Weighted
Of Number of
Value Observations Households
1 18 300,000
2 15 300,000
3 18 300,000
4 9 Q
5 1 Q
6 3 Q
8 10 200,000
9 5,826 100,300,000
Q = Data withheld because fewer than 10 observation in the category.