Public opinion on energy sources: the power of reliability

· by Marco Stutz · in Master's and PhD students projects

The energy transition is not just a technological challenge – it’s a societal one. Public acceptance is crucial for the success of renewable energy projects. A master’s thesis analysed 2.6 million German-language tweets from 2008 to 2023 to categorise public sentiment towards nuclear, coal, solar, wind, water and gas/oil energy.

Heated public discussions about energy reflect the importance of both energy supply and sources in modern society. To mitigate climate change many countries are making a large-scale switch towards renewable energy sources. However, research has found that such energy projects are doomed to fail if there is a lack of public support for the underlying energy source.

To investigate public support for different energy sources in German-speaking countries, I built a language model to categorize 2.6 million German tweets to gain an overview of how Twitter (nowadays “X”) users felt about nuclear, coal, solar, wind, water, and gas/oil energy during the study period from 2008 until 2023.

The model labelled each tweet as either positive, neutral or negative towards each energy source. Daily average sentiment was then calculated and plotted on a sentiment timeline (Fig. 1). While positive values represent overall approval of the respective energy source, negative ones suggest rejection. The buffers around the graphs symbolize the uncertainty based on the number of tweets per day. While hydropower and especially solar power were generally positively perceived, Twitter users had rather negative feelings regarding nuclear and coal power. Wind and gas/oil were connotated with neutral sentiment at the beginning of the study period but lost popularity over time.

Figure 1: The sentiment timeline showing how German-speaking Twitter users perceived the six energy sources of interest over the study period (temporally smoothed plot).

Click on this link for an interactive display of the Sentiment timeline (Fig. 1): https://creativelab.geo.uzh.ch/wp-content/uploads/2025/03/sentiment-timeline-interactive-html.html

In terms of public support and proposing appropriate energy policies, it is interesting to have a closer look at the sentiment variations and their possible causes, by reading the content of Tweets to understand what is driving sentiment.

Fossil fuels: Constant supply, constant damage

Twitter users mostly perceived coal power as negative due to its nature and climate implications. These tendencies increased when global climate movements gained momentum starting in around 2016. In autumn 2018, the Hambacher Forst near Cologne gained a lot of attention. Since 2012, anti-coal activists had occupied the forest to protest against the regional coal extraction company RWE. When the state government decided to clear the occupation to enable coal extraction, the forest became a symbol for the coal opposition movement. Most users sided with the activists, rejecting coal power. In the following years, this strong disapproval diminished and fears of an energy shortage after the invasion of Ukraine somewhat improved the image of coal as necessary for energy supply security during the energy crisis.

Compared to coal, gas/oil enjoyed greater acceptance. Indeed, Twitter users even perceived these sources rather positively prior to 2019. This was mainly due to hype around Compressed Natural Gas (CNG) which was described as a “clean” fuel. However, CNG lost this reputation as the climate movement intensified. With the invasion of Ukraine, the urgency to move away from gas initially intensified in early 2022 before fears of energy shortage and rising energy prices took over, which broke the downward sentiment trend and revitalized gas as a possible energy source.

Renewable energy sources: Great technologies with one decisive downside

Increased climate awareness and technological progress in photovoltaics led to a growing appreciation of solar power over the years. Furthermore, Twitter users valued the possibility to produce their “own electricity” and, thus, be independent of external players. Nevertheless, the outbreak of the war in the Ukraine resulted in a strong downward sentiment trend. At times of energy shortages, the technology was criticized by many for not being able to constantly supply sufficient energy as it is heavily dependent on weather conditions. Still, most users continued to perceive it positively, focusing on its strengths.

While wind power had high acceptance until 2018, its popularity decreased thereafter. Back then, climate and environmental interests clashed when many users complained about a planned deforestation project in the legendary Reinhardswald to enable the construction of a large wind park. Sentiment recovered as attention on this event faded, and people demanded less strict rules for wind power. Resembling the solar power discussion, complaints about the dependency of weather and security of supply dominated after the Russian invasion of Ukraine.

Since hydropower is not a big player in Germany, the tweet volume was comparatively small and mainly came from Swiss users. Again, climate-related advantages were countered by environmental concerns, mainly regarding fish populations and interventions into landscapes. But compared to the other renewable energy sources, hydropower was praised for being reliable and weather independent.

Nuclear power: From a bloody villain to a versatile hero?

Even before 2011, Twitter users already rejected nuclear power. However, the sentiment further deteriorated in spring 2011 following the catastrophic accident at the Fukushima Daiichi nuclear power plant. The disaster tragically demonstrated the danger of the technology which caused fear and strong disapproval in the Twitter community. Although sentiment hit rock bottom as multiple media outlets incorrectly reported on an explosion in the Marcoule nuclear power plant in fall 2011, a rising long-term trend could be observed in the years after. Recurring health arguments were countered by growing voices pointing to the climate friendliness and energy supply security of the technology, which is a rare combination of strengths.

The clash of interests

All in all, Twitter users mostly referred to the most prominent advantages and disadvantages of the observed energy sources in expressing sentiment. These pros and cons can be summarised into four argumentative dimensions shown in Fig. 2. Since no energy source can accomplish all four demands, they compete each other. Depending on the time period, the Twitter community assigned more or less importance to these dimensions. While environmental protection and human health dominated in the beginning of the timeline, these factors were successively replaced by prioritized climate and – later – extraordinarily strong energy supply arguments. Here, it was primarily the war in Ukraine that led to such a shift of priorities. These results suggest that public support for renewable energy and the associated energy transition is heavily challenged by energy reliability and affordability. Thus, policy makers must be careful in assessing measures as too drastic approaches were found to alter public perception for the worse. Moreover, the limiting factors of renewable energy must be addressed, namely supply security and costs, and education on the long-term consequences of different energy projects must be continued.

Figure 2: The four dimensions identified that were used by Twitter users to argue for or against a certain energy source.

Finally, it is important to keep in mind that the Twitter community is not representative of the general population and opinion-making is a complex process. Nevertheless, most of the energy-specific developments are in line with existing research findings.

Further references:

Segreto, M., Principe, L., Desormeaux, A., Torre, M., Tomassetti, L., Tratzi, P., Paolini, V., & Petracchini, F. (2020). Trends in social acceptance of renewable energy across Europe – A literature review. International Journal of Environmental Research and Public Health, 17 (24), 9161.

Sokolowski, J., Lewandowski, P., & Frankowski, J. (2023). How to prevent another Yellow Vest Movement? Assessing public preference for a carbon tax with a discrete choice experiment (IBS working paper).

Vringer, K., & Carabain, C. L. (2020). Measuring the legitimacy of energy transition policy in the Netherlands. Energy policy, 138, 111229.

Martin, M., & Islar, M. (2021). The ‘end of the world’ vs. the ‘end of the month’: Understanding social resistance to sustainability transition agendas, a lesson from the Yellow Vests in France. Sustainability Science, 16, 601–614.            

Vrana, V., Kydros, D., Kotzaivazoglou, I., & Pechlivanaki, I. (2023). EU Citizens’ Twitter Discussions of the 2022–23 Energy Crisis: A Content and Sentiment Analysis on the Verge of a Daunting Winter. Sustainability, 15 (2), 1322.

Wiertz, T., Kuhn, L., & Mattissek, A. (2023). A turn to geopolitics: Shifts in the German energy transition discourse in light of Russia’s war against Ukraine. Energy Research & Social Science, 98, 103036.

The full thesis:

Stutz, Marco. Geospatial sentiment analysis of different energy sources using historical Twitter data. 2024, University of Zurich, Mathematisch-naturwissenschaftliche Fakultät. https://doi.org/10.5167/uzh-262474

This blog post was originally published in the Geography Creative Lab.

Schreiben Sie einen Kommentar

Ihre E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert