Structure of Domestic Energy Saving

The purpose of this study is to examine whether we can find, behind the richness of energy-saving behavior, more general behavioral patterns. Specifically, this study compares the one-factor confirmatory model, the two-factor confirmatory model, and the Rasch model, which embody alternative assumptions about the dimensionality of energy-saving behavior. Unlike previous studies, this study uses a confirmatory approach to the testing of alternative models, and covers 10 Organisation for Economic Co-operation and Development (OECD) countries ( N = 10,251). This makes our tests rigorous and generalizable beyond a country-specific context. Our results indicate that energy saving is a one-dimensional class of behavior that is a function of individual’s propensity to engage in energy saving and the difficulty of energy-saving behavior. The Rasch model provides a valid and parsimonious representation of energy-saving behavior that outperforms the one- and two-factor models. We also show that the curtailment–efficiency dichotomy is not empirically valid.