Electrical energy consumption and the peak electrical demand were measured at two similar public housing sites in Selma, North Carolina. One site, designated as Preston Street (PST), had low cost energy retrofit measures combined with education of the occupants on energy efficient life styles. The other site, Canady Courts (CC), was originally planned as a control site. Both sites are single-storey wood-frame with brick-veneer construction. The sites consisted of a combination of efficiency and 1,2,3, and 4-bedroom units. PST consisted of 30 units with a total floor area of 22,000 ft2 and CC had 26 units with a total floor area of 18,800 ft2. Originally CC was intended to be used to quantify the impact of the energy conservation project undertaken at PST. This approach proved inadequate, and consequently an alternate methodology was employed. This method used the measured historical energy at both CC and PST to develop a database that could be used to predict the energy use that would be expected to occur at either site. If there are no changes at the site, then the predicted and measured energy use should agree. When conservation measures are instituted, the difference between predicted and measured energy use is the energy saving (or increase) due to the conservation measures. The report gives a description of the methodology that was used to analyse the historical data and the results and accuracy that were obtained. In addition, the problems with using one site as a control, as originally planned, are described. The Section on results gives a comparison of the predicted and measured energy use at both CC and PST for the calendar year 1985. The last part of the report deals with the investigation of the demand profiles and demand at PST. This analysis indicates the nature of the demand profiles at PST and thus the types of conservation measures that are likely to be the most effective in reducing demand.