Attitudes influence energy savings at home more than income, according to a study

The article uses standard meta-analysis techniques and includes a good number of studies, systematically selected.

I think the moderation analysis by study characteristics is particularly valuable. This shows that studies that use observed energy savings measures (not reported by the informant) report smaller effects.

On the negative side:

• It is not clear whether they included unpublished evidence. This is fundamental in meta-analysis: we know that journals tend to publish positive results rather than negative ones, which inflates the effects (the so-called success bias).
• They do not report whether they have issued calls for unpublished evidence, such as doctoral or master's theses, conference papers, etc., which do tend to report non-significant results.
• The study is based on correlational, not experimental, studies. Correlational studies are less suitable for establishing causal relationships, as intended here. At most, they can be interpreted as associations, rather than predictors, due to the common-method biases inherent in correlational studies.
• There is no assessment of the quality of the studies. This is typically done in all meta-analyses because the validity of the calculated effect depends on the quality of the studies. If the studies are of low quality, we should be suspicious of the reported effects.

The psychological factors that explain behaviors can be summarized as three: motivation, ability, and opportunity. This meta-analysis focuses primarily on motivation and partially on ability (including knowledge and perceived self-efficacy). The main conclusion is that intrinsic motivation to save energy (which is, in turn, explained by attitudes, environmental concern, or pro-environmental or ecological values) has a moderate to strong association with reported actions.

Extrinsic motivation (social pressure) has a moderate effect.

Ability has a greater effect: the variable with the strongest association is perceived self-efficacy, that is, whether I see myself as capable of saving energy. The more I do, the greater the reported savings.

The effects are smaller when actual energy savings are measured, in kilowatt-hours (kWh) consumed. This can be explained by two non-mutually exclusive reasons:

• Because we tend to exaggerate our behavior: if I perceive myself as concerned about the planet and saving resources and I'm asked if I save energy, I'll obviously say yes to remain consistent with my self-image.
• Because actions are taken that have little impact on actual energy savings. The typical example is turning off the light: it's very "visible" as an action, but its impact in kWh is marginal. The same can be said for not leaving appliances plugged in or on standby. These are the classic ideas one finds on the internet for saving energy, but they reduce kWh only slightly. What has the greatest impact on kWh is not using or using less of the most polluting electrical appliances: air conditioning, heaters, ovens, or washing machines. Furthermore, as appliances become more energy-efficient, the potential for energy savings by changing practices will decrease. This is perhaps why the study finds weaker correlations over time, as the authors explain.

On the other hand, it's worth asking how important energy savings really are. Wasting resources is always a bad practice, but as our percentage of green energy increases, energy savings will become less important as a decarbonization strategy. If my air conditioner is solar-powered, saving by turning off the air conditioning will not be a strategy for decarbonizing the system. Having more efficient air conditioning and an appliance that lasts longer or is repairable will likely be actions with a greater impact.

This meta-analysis focuses on analyzing how individual dispositions or capabilities are associated with actions leading to energy savings. By understanding these dispositions, we can design interventions that act on them or make them salient.

Other meta-analyses have focused on interventions—experiments that test the effectiveness of an action to promote energy savings—and the results are similar. Interventions that make environmental identity or commitment to saving salient increase efficiency. Similarly, personalized advice (what you can do at home with your specific equipment) increases savings. Volitional interventions that increase perceived effectiveness also improve savings.