This paper argues that electricity 'curtailment' behaviors (i.e. frequent and/or low cost or free energy saving behaviors) in households are distinct from one another and they thus should be analyzed and promoted. We test this claim with data from telephone interviews with Greek households in the capital city of Athens (N=285), analyzing the impact of a number of demographical/structural, psychological (based on the Theory of Planned Behavior) and moral (based on norms' activation) predictors though hierarchical binary logistic regression modeling. We find that that each electricity curtailment behavior depends on a different mix of predictors with 'Age', 'Gender' and 'Perceived Behavioral Control' being statistically significant for most behaviors. Overall, the psychological and the demographical/structural clusters of variables substantially contribute to the explained variance of electricity curtailment behaviors. The moral cluster's contribution is not statistically significant since moral concerns are largely interwoven in the psychological constructs. (C) 2014 Elsevier Ltd. All rights reserved.