Abstract
The energy crisis following the Russian invasion of Ukraine has prompted affected countries to increase efforts to save energy. Despite substantial energy conservation observed across sectors, research has inadequately explored the reasons behind these savings. This study bridges this gap by analyzing the discourse on energy conservation during the crisis in Germany, using Twitter data. We analyse a set of 136,837 tweets from January 2022 until May 2023. The analysis consists of two approaches based on natural language processing (NLP): First, we conduct word counts within the days with high tweet activity related to energy saving to identify specific triggers. For a broader quantitative analysis, we then use text classification to categorize the data set into different aspects as well as different stances and investigate how these evolve during the crisis. Our results show a heightened focus on energy conservation in the Twitter debate during the crisis, reflected in an increase of the average monthly tweets from 730 (before 2022) to 8,552 in 2022/2023. Regarding the motivations, ethical concerns linked to the war play a larger role in the initial months, while other factors such as energy costs, solidarity aspects, and climate change are consistently present throughout the debate. Additionally, we note significant negative discourse regarding the policy measures enacted during the crisis, indicating that policy backlash is a critical issue, at least in the realm of the Twitter debate. Given the vital role of energy efficiency in combating climate change, comprehending the motivations and de-motivations behind voluntary energy conservation during the crisis offers valuable insights. This understanding can inform future policies aiming to encourage energy savings.