Funding text 1: C.-F. Chen and Hannah Nelson were supported by (1) the Engineering Research Center Program of the U.S. National Science Foundation (NSF) and the Department of Energy under NSF award EEC-1041877 and the CURENT Industry Partnership Program; and (2) NSF award CMMI-1901740 .; Funding text 2: An ANOVA test with Tukey's HSD post-hoc was conducted to compare EUI across the income groups. The analysis demonstrated significant differences across income levels, F (2, 5683) = 145.60, p <.001. Post-hoc tests showed that LIHs had higher EUI (M = 34.18, SD = 31.35) than MIHs (M = 23.58, SD = 22.63) and HIHs (M = 18.61, SD = 24.93). The difference between MIHs and HIHs was also significant, p <.001. This result needs to be considered with the fact that the distributions of raw EUI were quite abnormal in all income groups (Fig. 1) and the variances were large. The results reported provide support for Hypothesis 1, which predicts that EUI will be higher among LIHs than MIHs and HIHs. Having empirically established that EUI differs across the income groups, this study next identifies the factors that influence EUI for each income group.These relationships differ across income groups, which provides support for Hypothesis 2. More importantly, appliance ownership and energy behavioral action explain a greater portion of the variances in EUI among LIHs than among higher-income households (Hypotheses 4 and 5). These results suggest great potential for reducing EUI and energy burdens for LIHs, which can be achieved through improvements in appliance efficiency and related energy behavioral changes. Appliance and behavioral variables related to space heating and cooling seem to be the most important. The number of Energy Star appliances was not associated with the EUI for LIHs or MIHs, but it increased the EUI of HIHs (Hypothesis 4). This is possibly due to more frequent use of such appliances among HIHs. In general, larger homes are associated with lower EUI. Contrary to our expectations in Hypothesis 3, the age of the housing unit and insulation levels have no significant influence on EUI across any income group, and this might be due to other important behavioral and social-economic factors. Finally, the EUI of different income groups vary across regions, which supports Hypothesis 7; LIHs in the South have lower EUI than LIHs in the other areas, while HIHs in the West have higher EUI than HIHs in the different regions.C.-F. Chen and Hannah Nelson were supported by (1) the Engineering Research Center Program of the U.S. National Science Foundation (NSF) and the Department of Energy under NSF award EEC-1041877 and the CURENT Industry Partnership Program; and (2) NSF award CMMI-1901740.