Abstract
Participants in utility-sponsored residential conservation programs are systematically different from nonparticipants. As a result self-selection is an important validity threat in studies of conservation program impact. Three approaches to dealing with this self- selection bias are reviewed: (1) designs that use participants as a control group, (2) construction of a matched sample on the basis of predicted energy consumption values, and (3) multiple regression analysis. The advantages and disadvantages of each approach are discussed.
Get full access to this article
View all access options for this article.
