A Comparison of Energy Consumption in American Homes by Climate Region

The present research analyzes the impact of nine factors related to household demographics, building equipment, and building characteristics towards a home’s total energy consumption while controlling for climate. To do this, we have surveyed single-family owned houses from the 2015 Residential Energy Consumption Survey (RECS) dataset and controlled the analysis by Building America climate zones. Our findings are based on descriptive statistics and multiple regression models, and show that for a median-sized home in three of the five climate zones, heating equipment is still the main contributor to a household’s total energy consumed, followed by home size. Social-economic factors and building age were found relevant for some regions, but often contributed less than size and heating equipment towards total energy consumption. Water heater and education were not found to be statistically relevant in any of the regions. Finally, solar power was only found to be a significant factor in one of the regions, positively contributing to a home’s total energy consumed. These findings are helpful for policymakers to evaluate the specificities of climate regions in their jurisdiction, especially guiding homeowners towards more energy-efficient heating equipment and home configurations, such as reduced size.

Surviving a Shut-Off: U.S. Households at Greatest Risk of Utility Disconnections and How They Cope

This is the first known study to estimate household characteristics and coping behaviors associated with utility disconnections in the United States. We capitalize on a measure of disconnections available in the Residential Energy Consumption Survey that is administered by the U.S. Energy Information Administration. Using the 2015 panel, we analyzed the prevalence of disconnection notices, disconnection of services, and related coping strategies, including: forgoing basic necessities, maintaining an unhealthy home temperature, and receiving energy assistance. Findings indicate that nearly 15% of U.S. households received a disconnection notice and 3%—more than three million households—experienced a service disconnection in 2015. Our results further demonstrate that more households resorted to forgoing basic necessities than maintaining an unhealthy temperature or receiving energy assistance, though many families used a combination of strategies to prevent or respond to the threat or experience of being disconnected. We discuss implications for future research on material hardships, survival strategies, and the health impacts of poverty.

Estimating Net Coverage of Segments in ABS Frames

The literature has no standard method for estimating the coverage of area probability segments in address-based frames. Versatility is desirable for different study needs, but standardization improves comparability. Many segment estimates are simple ratios of counts of frame addresses to control totals, or net coverage ratios. Challenges to segment ratios include geocoding error, outdated control totals, errors in the address frame, and systematic exclusion of types of addresses. We tested various net coverage ratios on segments selected for the 2015 Residential Energy Consumption Survey, and we share our results and recommendations for using net coverage ratios for estimating coverage for segments.

Total Error in a Big Data World: Adapting the TSE Framework to Big Data

While Big Data offers a potentially less expensive, less burdensome, and more timely alternative to survey data for producing a variety of statistics, it is not without error. The AAPOR Task Force on Big Data and others have called for researchers to evaluate the quality of Big Data using an approach similar to the total survey error (TSE) framework. However, differences in the construction of, access to, and overall data structure between survey data and Big Data make application of TSE difficult. In this article, we seek to develop the Total Error Framework (TEF), an extension of the TSE framework, to be (1) more inclusive and applicable to many types of Big Data, (2) comprehensive in that it considers “total” error, and (3) unified in that it allows researchers to compare errors in Big Data to errors in survey data. After outlining this framework, we then illustrate an application of TEF by comparing error in housing unit area (square footage) estimates collected in a survey (the 2015 Residential Energy Consumption Survey [RECS]) to those estimates found in three Big Data databases (Zillow.com, Acxiom, and CoreLogic).

Evaluating the Age-Energy Consumption Profile in Residential Buildings

Age is an important proxy for many life course trajectories. The relationship between energy consumption and age is complex and understudied. We evaluated the existence and determinants of an age-energy consumption profile in the U.S. residential sector, using microdata from four waves of the Residential Energy Consumption Survey (RECS) in 1987, 1990, 2005, and 2009. We constructed pseudo cohorts from Bayesian generalized linear model estimates to draw micro-profiles for energy consumption across the life course. Overall, we found that residential energy consumption increases over the life course. Much of the increase in energy consumption is due to housing size. Variations in the age-energy consumption micro-profiles can be described by concave and convex functions. In contrast to previous research that suggested that population aging would reduce energy demand, our results indicate that changing population age structure could amplify residential energy demand.

Calibration Weighting for Nonresponse with Proxy Frame Variables (So that Unit Nonresponse Can Be Not Missing at Random)

When adjusting for unit nonresponse in a survey, it is common to assume that the response/nonresponse mechanism is a function of variables known either for the entire sample before unit response or at the aggregate level for the frame or population. Often, however, some of the variables governing the response/nonresponse mechanism can only be proxied by variables on the frame while they are measured (more) accurately on the survey itself. For example, an address-based sampling frame may contain area-level estimates for the median annual income and the fraction home ownership in a Census block group, while a household’s annual income category and ownership status are reported on the survey itself for the housing units responding to the survey. A relatively new calibration-weighting technique allows a statistician to calibrate the sample using proxy variables while assuming the response/ nonresponse mechanism is a function of the analogous survey variables. We will demonstrate how this can be done with data from the Residential Energy Consumption Survey National Pilot, a nationally representative web-and-mail survey of American households sponsored by the U.S. Energy Information Administration.

Demystifying the Relationship Between Use-Phase Attributes and Energy Consumption: A Case Study of Personal Computers

Understanding the use-phase energy consumption of consumer electronics is of great importance, as it has significant effects on both policy and product designs. Inaccurate estimations of the use phase energy consumption can offset the results of the life cycle assessment and impeach the effectiveness of the energy intervention policies. The use phase energy consumption is governed by the consumers’ usage behavior. However, the relationship between consumers’ attributes and their usage behavior, and energy consumption is not clear. This paper analyzes two data sets, a data set of hard drives’ Self-Monitoring, Analysis and Reporting Technology (S.M.A.R.T) and the Residential Energy Consumption Survey (RECS) to shed light on the relationship between usage behavior and energy consumption. Several supervised and unsupervised machine-learning methods have been used to reveal possible trends in the consumers’ use-phase attributes. The results of the study suggest that various demographic properties and behavioral variables related to computer usage affect the energy consumption profile of households.

Constructing Strata of Primary Sampling Units for the Residential Energy Consumption Survey

The 2015 Residential Energy Consumption Survey design called for stratification of primary sampling units to improve estimation. Two methods of defining strata from multiple stratification variables were proposed, leading to this investigation. All stratification methods use stratification variables available for the entire frame. We reviewed textbook guidance on the general principles and desirable properties of stratification variables and the assumptions on which the two methods were based. Using principal components combined with cluster analysis on the stratification variables to define strata focuses on relationships among stratification variables. Decision trees, regressions, and correlation approaches focus more on relationships between the stratification variables and prior outcome data, which may be available for just a sample of units. Using both principal components/cluster analysis and decision trees, we stratified primary sampling units for the 2009 Residential Energy Consumption Survey and compared the resulting strata.