Total square footage as presented in the 2001 Residential Energy Consumption Survey includes living space (living rooms, bedrooms, dining rooms, kitchens, dens, and other living space not separately reported), other enclosed interior space (space not included in any other category such as laundry rooms, hallways, and closets), attached garages, basements, finished and heated attics, and enclosed porches.  These same areas were last measured in the 1993 RECS.  The average total square footage of U.S. housing units, as measured in the 2001 RECS, was 2,066 square feet.  This is a 10.6 percent increase above the average total square footage of 1,875 square feet in 1993, when it was last measured (see Table 1).

This increase can be attributed to two factors:  first, the size of the average housing unit in the United States has, in fact, increased over the past 7 years; and second, the improved methodology employed in measuring the square footage of housing units in the 2001 RECS was more thorough than the previously used methodology, resulting in the inclusion of space (particularly garages and basements) that was previously overlooked by those performing the measurements.  (See the Appendix for a detailed discussion of the 2001 RECS housing measurements methodology.)

As shown in Table 1, the 1993-2001 changes varied by the type of housing unit.  Single-family attached housing units had the largest 1993-2001 change, increasing 31.9 percent from an average of 1,799 square feet in 1993, to 2,373 square feet in 2001.  In contrast, the average single-family detached housing unit increased by 9.2 percent growing from 2,337 square feet in 1993, to 2,553 square feet in 2001.

The increase in size of apartments and mobile homes was more modest than that of single-family housing units (see Table 1).  Apartments increased from an average of 972 square feet in 1993 to an average 1,043 square feet in 2001, a 7.3 percent increase.  Within this category, the average apartment in a 2-4 unit building increased by 16.3 percent, growing from 1,198 square feet in 1993, to 1,393 square feet in 2001.  In contrast, the average apartment in buildings with 5 or more units decreased by 1.6 percent from 861 square feet in 1993, to 847 square feet in 2001.  The floorspace in mobile homes increased by 8.9 percent, growing from 975 square feet in 1993, to 1,062 square feet in 2001.

As noted earlier, the methodology used in 2001 differed from that used in 1993 and may account for some of the change in average total square footage.  This is particularly the case for mobile homes and single-family housing units that have garages, and single-family attached housing units and apartments in 2-4 unit buildings that have basements.  Initially, the Computer Assisted Personal Interview questionnaire used for the 2001 RECS did a much better job of simply identifying those housing units with garages and with basements than did the paper-and-pencil questionnaire used in the 1993 RECS.  Once garages and basements were identified, the computers used by the interviewers provided them with instructions to actually measure them.

Table 2 presents the numbers of housing units reported to have both attached and detached garages in the 1993 and 2001 RECS.  (Prior to the 2001 RECS data about attached and detached garages were not collected separately, but only the square footage of attached garages was included in the square footage totals.  Accordingly, any 1993 versus 2001 comparisons of the number of garages must include both attached and detached garages.)  Overall, the number of mobile homes, single-family detached, and single-family attached housing units having a garage increased by 31.9 percent.  The number of mobile homes with a garage more than doubled from 1993 to 2001,  increasing by 142.6 percent.  Among single-family detached housing units the number with a garage increased by 22.3.  In contrast, the proportion of single-family attached housing units with a garage more than doubled, increasing by 134.5 percent.

What is remarkable about these findings is that in all categories, with the exception of mobile homes, the 1993-2001 increase in the number of housing units having a garage was larger than the increase in the total number of housing units.  For this to reflect reality not only must all the housing units built between 1993 and 2001 have had a garage, but also a large number of existing structures would have added a garage during the same period.  This is particularly implausible for single-family attached housing units, which have little ground space around them where a garage could be added.

The effect of identifying garages attached to the housing unit and including their square footage in the total is illustrated in Table 3.  This table shows that the 1993-2001 increases in total square footage for single-family detached and single-family attached housing units with garages were substantially larger than the changes for the housing units in these two categories without garages. 

Table 4 presents the numbers of housing units reported to have basements in the 1993 and 2001 RECS.  Overall, the number of single-family detached and single-family attached housing units and apartments in 2-4 unit buildings having a basement increased by 15.7 percent.  (Mobile homes were assumed to have no basement and respondents in apartments in buildings with more than 4 units were simply not asked on the assumption that they did not have exclusive use of any part of the basement.)

Among single-family detached housing units the number of units with a basement increased by 10.2 percent.  In contrast, the number of single-family attached housing units with a basement nearly doubled, increasing by 94.2 percent.  The number of apartments in 2-4 unit buildings with a basement decreased by 4.8 percent.  This change is likely due to the more precise determination of whether the respondent actually had exclusive use of some or all of the basement, which, in turn, determined if it should be included in the square footage measurements.

Among single-family detached housing units surveyed in 1993, 45.3 percent had a basement.  In the 2001 RECS, 47.1 percent of these units had a basement.  In contrast, among the single-family attached housing units surveyed in 1993, 40.0 percent had a basement.  In the 2001 RECS a much larger percentage, 53.7 percent, reported having a basement.  Among the 3.3 million single-family attached housing units built between 1993 and 2001 84.8 percent reported having a basement.

The effect of identifying garages and basements and including their square footage in the total is illustrated in Table 5.  The 1993-2001 increase in average total square footage for single-family attached housing units and apartments in 2-4 unit buildings with basements were substantially larger than the increases for the housing units in these two categories without basements.

Table 6 presents the average total square footage of single-family housing units for the four possible combinations of with and without garages and with and without basements.  (Respondents residing in apartments in 2-4 unit buildings were not asked about garages, those residing in apartments in buildings with more than 4 units were not asked about garages and basements, and those living in mobile homes were not asked about basements, so comparable data are not available for these housing units.) 

While Tables 3 and 5 illustrated that housing units with garages and basements, taken separately, tended to have larger 1993-2001 increases in square footage that those without, Table 6 suggests that garages were the principle contributor to the larger 2001 measurements compared to 1993 among single-family detached housing units.  Regardless of the presence or absence of a basement, the single-family detached housing units with a garage had larger 1993-2001 percentage change increases (5.0 percent and 7.3 percent, respectively) than those units without a garage.

This result contrasts with the more mixed findings for single-family attached housing units.  The largest 1993-2001 percentage change increase was among the housing units in this category with both basements and garages (39.0 percent), followed by the housing units with neither a basement nor a garage (17.7 percent).

In summary, users of the 2001 RECS square footage data are advised to take care when making comparisons between these data and those obtained in previous RECS.  Unquestionably, the “active living area” and the total square footage of the average U.S. housing unit have increased over the 1993-2001 period.  However, part of the observed 1993-2001 increases can be attributed to the improved measurements methodology used in the 2001 RECS resulting in the increased identification and inclusion of garages and basements in the housing unit measurements.

Square Footage Measurements and Comparisons: Caveat Emptor

Appendix:  Household Measurements Methodology

2001 Residential Energy Consumption Survey

The methodology for measuring the square footage of housing units used in the 2001 Residential Energy Consumption Survey is a departure from that used in previous RECS.  Accordingly, the data presented here are not directly comparable to those collected and presented in the past.  However, we believe that this new methodology has resulted in improved accuracy of the measurements.  And, for the first time, we present a breakdown of the total heated and unheated square footage of the housing units into its component parts—the attic, basement, attached garage, and all other floorspace (the living space of the housing unit).  Table A-1 presents the areas of U.S. housing units that were separately measured (and reported in the 2001 RECS data set) and included or excluded in total and heated square footage data in the 2001 RECS.  These same areas were measured in previous surveys.

In past RECS the interviewers were responsible for identifying all the floorspace in each housing unit, actually measuring all of it, and identifying hich portions of the measured floorspace were heated and unheated.  All the data, including drawings of each floor, were manually recorded in the questionnaire booklet and later transcribed and entered into the computerized data files and then edited.

In contrast, the CAPI household interview used in the 2001 RECS permitted most of the tasks previously assigned to the interviewer to be performed by the computer.  To accomplish this a number of conventions that were obtained from an examination of the data collected in previous RECS were adopted.  These conventions resulted in less reliance on the interviewer and allowed for enhanced statistical imputation of square footage data.

The large majority of housing units are square, rectangular, L-shaped, or T-shaped.  Also, for most units with a basement or more than one floor, the shape and size of the basement or floors above the main floor are the same as the main floor.  Only if the basement below the main level differed in size or shape from the main floor was it measured.  Similarly, the floor above the main floor was measured only if it differed in size or shape from the main floor.  Where a third floor was present, and its’ size and shape were the same as the second, it also was not measured.  Given that the basement or the floor above the main floor are the same size and shape as the main floor the measurements of the main floor, which was obtained for all housing units, imputations of the area of the floor above or below it could be made.

Garages are very standardized in size and have not changed over time.  Based on previous RECS, it is known that a one-car garage averages 250 square feet in size, a two-car garage is 400 square feet, and a three-car garage is 600 square feet in size.  Accordingly, garages were not measured as separate sections of the housing unit, as in the past.  Where the garage was an integral part of a floor containing living space or the basement, the entire floor or basement, including the garage, was measured.  In the editing of the data the square footage of the garage, depending on the number of cars it could hold, was deducted from the total basement or living space.  Garages that were attached to the housing unit, but not an integral part of the housing unit, were not measured and the square footage imputed based on the number of cars the garage could hold.

Based on this information, the CAPI provided instructions to the interviewers that specified what they were to measure.   The measurements collected were then entered into the CAPI system by the interviewers.  Included in the CAPI interview programs were range edits that questioned unusually small or large areas that were entered.  Only in cases where the housing unit was more than three stories high or some shape other than L-shaped or T-shaped, was the interviewer required to make drawings and record their measurements in a supplementary booklet.

Unlike previous RECS, the interviewers were not required to identify the space they measured as heated or unheated.  In Section A of the household questionnaire the householder was directly asked if any basements, attics, or garages that were part of their housing unit were heated or not (specifically, they were asked if the area was warm enough to sit, work, or play in during the winter months).  In those cases where only a portion of the basement or attic was reported to be heated, the householder was asked what portion was heated (specifically, very little (1-4 percent), some (5-33 percent), about half (34-66 percent), about three-quarters (67-95 percent), or most of it (96-99 percent)).  Also householders were asked the specific number of rooms in the housing unit that were not heated.  Based on the responses to these questions the total space in each category was allocated between heated and unheated space by imputation.

To Top

Specific questions may be directed to:

Michael Laurence
michael.laurence@eia.doe.gov
Author
Phone: (202) 586-2453
FAX: (202) 586-0018

Contacts:

Eileen M. O'Brien
RECS Survey Manager
Phone: (202) 586-1122
Fax: (202) 586-0018