‹ Consumption & Efficiency

Commercial Buildings Energy Consumption Survey (CBECS)

2007 CBECS Large Hospital Building Methodology Report Main Report | Methodology | FAQ | List of Tables

CBECS 2007 - Release date: August 17, 2012

Data Collection

The data in the Energy Characteristics and Energy Consumed in Large Hospital Buildings in the United States in 2007 report and accompanying tables were collected in the 2007 round of the Commercial Buildings Energy Consumption Survey (CBECS). CBECS is a quadrennial survey is conducted by the Energy Information Administration (EIA) to provide basic statistical information about energy consumption and expenditures in United States commercial buildings and information about energy-related characteristics of these buildings. The survey was conducted in two phases, the Building Characteristics Survey and the Energy Supplier Survey.

The Building Characteristics Survey collects information about selected commercial buildings through voluntary interviews with the buildings’ owners, managers, or tenants. In the 2007 survey, these data were collected using Computer-Assisted Personal Interviewing (CAPI) techniques. Respondents were asked questions about the building size, how the building was used, types of energy-using equipment and conservation measures that were present in the building, the types of energy sources used, and the amount and cost of energy used in the building in the year 2007.

If the respondents to the Building Characteristics Survey could not provide the energy consumption and expenditures information, or if the information provided failed certain edits within the survey instrument, the Energy Supplier Survey was initiated. This Suppliers Survey obtained data about the building’s actual consumption of and expenditures for energy from records maintained by energy suppliers. These billing data were collected in a mail survey conducted under EIA’s mandatory data collection authority. A survey research firm, under contract to EIA, conducted both the interviews for the Building Characteristics Survey and the mail survey for the Energy Suppliers Survey.

The large hospitals are a subset of data collected in the 2007 CBECS. The 2007 data as a whole did not meet EIA’s standards for quality, credible energy information due to difficulties with a less expensive, experimental method for updating the main sampling frame. However, nearly all of the large hospitals were sampled from high-quality and well-established lists of hospitals, federal buildings, and other large buildings separate from the main sampling frame. EIA did a thorough evaluation of these frames and the data collected from their sampled units and deemed them to be representative of the large hospital building population as a whole. For more information see the large hospital FAQs.

Sample Size

A total of 157 large hospital buildings responded to the 2007 CBECS, representing an estimated population of 3,040 total large hospital buildings.

Response Rate

The response rate for large hospital buildings in the Building Characteristics Survey was approximately 86 percent1 .

Unit Nonresponse

An in-scope sample building, otherwise eligible for interview, for which no information is obtained, is called a unit nonresponse. The primary causes of unit nonresponse among large hospital buildings in the 2007 CBECS were the respondent’s refusal to participate in the interview (43 percent of total nonresponding buildings) and the inability to contact and interview someone knowledgeable about the building (43 percent of total nonresponding buildings).
Adjustments for unit nonresponse were made by redistributing the baseweights of nonresponding buildings to responding buildings with similar propensities for nonresponse. A predictive model for response propensity was developed to identify subgroups of the population with differential response rates.

Item Nonresponse

Item nonresponse occurs when a building respondent does not know the answer to a question or refuses to answer it. EIA uses imputation to fill in this missing information in order to have a complete dataset and unbiased estimates. The hot deck method is used to impute building characteristics data and a non-linear statistical model is used to impute consumption and expenditures. Most building characteristics survey items for large hospitals have very low item non-response rates, 10 percent or less. Consumption and expenditure item non-response rates are higher. For example, total major fuel consumption item non-response was 24 percent.

Sampling Error

The large hospital estimates are based on data reported by a statistically-designed subset of the entire large hospital building population in the United States, or a "sample." Consequently, the estimates differ from the true population values. However, the sample design permits us to estimate the sampling error in each estimate.
It is important to understand that these estimates should not be taken simply as point estimates, but as estimates with some associated error in each direction. EIA publishes a Relative Standard Error (RSE) for each estimate, which is a measure of the estimate’s precision. The RSE is defined as the standard error (square root of the variance) of a survey estimate, divided by the survey estimate and multiplied by 100. Items with RSEs greater than 50 are deemed not statistically meaningful and are not published, and are indicated in the tables by a “Q”.

Confidentiality of Information

The names and addresses of individual respondents or any other individual identifiable data that could be specifically linked to an individual sample building or building respondent are seen only by employees of EIA, its survey contractor, and EIA-approved agents. The data are kept secure and confidential at all times in compliance with the Confidential Information Protection and Statistical Efficiency Act of 2002 (CIPSEA). The legislation gives EIA both the authority and the responsibility to protect from disclosure identifiable data which respondents have been promised would be kept confidential and used exclusively for statistical purposes. The CBECS meets these criteria, and the 2007 cycle was collected under CIPSEA protection.

1 The response rate for this population cannot be determined with certainty because the exact building activity for the non-responding buildings on the lists other than the hospital list is unknown. However, the response rate from the hospital list is a known quantity, 86 percent, and the response rate from the remaining sample is assumed to be similar.

For further information, contact:

Katie Lewis

Jay Olsen