Using social technologies to increase sharing and communication around household energy consumption in low-income and rental communities

The behavioral impacts of providing users with real-time energy use feedback—even at the aggregate level—can reduce energy use by 10-15%. Though comparison is a feedback method shown to encourage additional savings, home-energy research studies exploring social communication around feedback devices are limited. As a result, the relationship between feedback technologies and energy-conservation behaviors has become an increasingly important focus in the field of human-computer interaction. Furthermore, few studies explore this phenomenon among households, specifically, low-income households. To bridge this gap, we conducted 26 photo-elicitation interviews with low-income tenant households across two locations, Pittsburgh, PA and eastern NC. Our studies of low-income households show that comparison across households can have an important impact on how energy is used (or saved). These studies also reveal conflict between various internal and external stakeholders such as family members and landlords. To better understand this conflict, we re-analyzed data from the first study for specific landlord / tenant conflicts and resolutions, held semi-structured interviews with landlords to understand their perspectives, and held a role-play exercise with tenants to understand ways to resolve the conflicts between landlords and tenants. Finally, based on our qualitative study results, we developed an Android-based application called the Community Monitor that supports comparison and provides household energy-use information. After conducting a longitudinal energy use study with 15 collocated U.S. households in Pittsburgh, PA, we were able to better understand the impact of engagement around social sharing and community energy monitoring in residential communities, identify energy-related and personal concerns, and provide design implications for home-energy applications that share consumption data among community members.