Ph.D. student Jared Langevin and faculty members Drs. Patrick Gurian and Jin Wen have recently published a paper in the journal Applied Energy that examines residential energy consumption behaviors in low-income housing – a context that has not received enough attention in spite of its significant contribution to energy consumption in the United StatesThe paper, titled “Reducing energy consumption in low income public housing: Interviewing residents about energy behaviors”, uses a semi-structured interview format to explore key behavioral tendencies, energy knowledge gaps, and attitudes amongst residents of the Philadelphia Housing Authority (PHA), the 4th largest public housing authority in the United States.  The goal of the work was to demonstrate “a process for developing, scoring, and analyzing the interviews that will be useful to other researchers when first engaging complex subjects like behavior in contexts that are not well covered by the existing literature.”

In total, 50 interviews were recorded with PHA residents in multiple waves between July 2011 and April 2012.  The interviews followed a guide that was designed to engage the respondents for 20-40 minutes in discussions about key comfort, behavior, and energy use issues.  To ensure that the open-ended discussions could be interpreted quantitatively, the authors also developed a response scoring method that first created plausible response categories based on existing literature, subsequently examined the response audio for relevant bits of text, and finally assigned the text excerpts to the appropriate response category, reformulating and expanding the categories where necessary throughout the first wave of interviews.  This iterative approach allowed the scoring framework to remain flexible and to accommodate new themes and threads of discussion as they began to emerge in the resident responses.

Upon analyzing the interview audio, the authors discovered several interesting themes in the resident responses.  For example, residents often hold varying definitions of “comfort” in their homes, with some discussing the issue in terms of physical conditions (temperature, humidity, etc.) and others equating the term with concepts of privacy, security, and good maintenance.  Complaints about physical comfort are frequently associated with a reported lack of control over the environment in their homes, either because the PHA ultimately controls heating/cooling equipment, equipment controls are broken or hard to understand, or there are concerns about the comfort levels of others in their home.

Regarding the use of household equipment and appliances, energy consumption was found to run highest in service of cool comfort, as the three items on the interview with the largest proportion of “high consumers” in the response scoring are thermostats in the summer (44% of responses), window air-conditioners (54%), and opening windows when the heat was on (62%).  Air-conditioning use is high in particular because many residents describe leaving their units running on high throughout warmer months with little adjustment – including when they aren’t home.  On the opposite end of the spectrum, lighting consumption is quite low amongst PHA residents, with just 20% falling into the “high consumer” category.  Response context suggests that in some cases this is due to a high light sensitivity – many PHA residents are elderly – but more often it is a matter of common sense, as residents don’t see the need to leave lights on in areas of the house that aren’t being used.

Responses to a series of questions about 13 Energy Conservation Measures (ECMs) revealed that residents tend to evaluate these measures in terms of costs, savings, comfort, or a combination of those criteria.  The consideration of whether or not the resident pays their energy bill (some have it paid by PHA) was an important one when examining resident responses about the various ECMs.  For example, while those who do not pay bills often evaluate the ECM of programmable thermostats purely in terms of comfort, those who do pay for energy discuss their willingness to use a programmable thermostat in terms of potential energy cost savings.  Likewise, while those who do not pay bills tend to evaluate the attractiveness of EnergyStar appliances in terms of product appearance, performance, and initial cost, those who do pay the bills typically cite long-term reductions in their energy bill as justification for purchasing appliances with an EnergyStar label.

In general, many residents simply do not know enough about the various ECMs to feel comfortable using them in their homes – just 3 of the 13 ECMs asked about are familiar to more than half of the respondents who do not already practice these measures.  The authors suggest that increasing resident familiarity with these ECMS through energy education is a good opportunity, as the majority of residents expressed a high willingness to participate in energy education programs if PHA or another reputable organization offered them.

Results from the study will be used to inform the development of future, larger scale survey instruments for the PHA.  For these future surveys, the current set of interview responses and scoring outcomes provide invaluable context.  For example, given the finding that residents assess ECMs through the multiple lenses of energy savings, comfort, and initial cost, a future survey question might assess resident willingness to pursue an ECM alongside the following questions: 1) “How much do you believe this will cost?” 2) “How much do you believe this will affect your (physical comfort, privacy, safety)?” 3) “How much energy do you believe this would save in the long run?”  An additional “Don’t Know response could be included to expose areas of poor ECM knowledge.

The authors also intend that the process used to arrive at the current findings will be broadly useful to future studies of the building occupant and his/her energy-related behaviors.  In particular, the reported approach yields results that are interpretable from both a quantitative and qualitative perspective, allowing trends that may be ambiguously interpreted in the analysis of traditional occupant survey response data to be checked against direct commentary from the occupants, yielding a clearer understanding of what observed outcomes really signify.

The full paper can be found here.