Summary
Disruptive climate protests have gained widespread attention in recent years, yet their impact on the political system remains under explored. While climate protests can raise awareness and influence public discourse, it is less clear whether they also shift electoral preferences, through increasing support for pro-climate parties or triggering a backlash that benefits anti-climate parties. This report analyses voting intention polls taken on 916 different days from the UK, Sweden, and Germany over a two-year period. Our findings show that in all three countries, climate protests have a modest but measurable effect in a pro-climate direction, increasing voting intention for green and more environmentally conscious parties. This suggests that disruptive protests may have an important impact on electoral dynamics, with potential implications for future climate policy.
​
Key Findings
-
In Germany, disruptive climate protests were associated with a 0.19 percentage point rise in support for the pro-climate Linke Party and a 0.44 percentage point drop for the anti-climate AfD. Follow-up analysis supports this pattern: parties with stronger climate policies tended to benefit more from protests. Overall, pro-climate parties saw a combined boost of 0.33 percentage points due to the protests.
-
In Sweden, disruptive climate protests raised voting intention for the nominally pro-climate Centerpartiet by 0.45 percentage points and overall, parties with more favourable climate change policies tended to benefit more from protest, seeing a rise of 0.79 percentage points.
-
In the United Kingdom, disruptive climate protests raised voting intention for the Green Party by about 0.34 percentage points. However, no overall shift towards pro-climate parties was observed, largely because the anti-climate Reform UK trended towards a rise in support. This suggests that climate protests may have a more polarising effect in the UK.
​
Cover image by Jonathan Brady. Used under Creative Commons Attribution-Share Alike 4.0 International license. Like to image here.
This report was funded by the Joseph Rowntree Reform Trust (JRRT). JRRT has supported this work in recognition of the importance of the issue. The facts presented and the views expressed in this report are, however, those of the authors and not necessarily those of the Trust. www.jrrt.org.uk​
Introduction
A well-documented wave of climate protests in recent years has sparked extensive interest in their effects. There is strong evidence that they can influence public opinion on climate change, both positively and negatively (Fuller et al., 2025; Kenward & Brick, 2024; Kountouris & Williams, 2023), shape public discourse (Thackeray et al., 2020); affect media coverage in both volume and tone (Scheuch et al., 2024); and bolster support for more moderate environmental groups (Dasch et al., 2024; Ostarek et al., 2024; Simpson et al., 2022). A critical question is whether protests also influence climate policy. Research shows environmental protests have increased the likelihood of new environmental laws (Agnone, 2007) and have heightened parliamentary and governmental attention on climate change (Walgrave & Vliegenthart, 2012). Schürmann (2023) similarly found that Fridays for Future protests in Germany led to more climate-related discussions in parliaments. Our own research on the UK campaign group Insulate Britain revealed that mentions of home insulation in Parliament rose significantly during and after the campaign (Rogers et al., 2024).
Protests may also influence electoral outcomes. Valentim (2023) found evidence suggesting that Fridays for Future protests in Germany led to higher Green Party vote shares due to higher voter turn-out and shifts in attitudes towards climate issues. Fabel et al. (2022) similarly found environmental protests led to increased vote shares for the Green Party and additionally reported a shift in votes from the far-right AfD party to the center-right Christian Democrats (see Fabel et al., 2025 for insights into the mechanisms that bring about the shift towards the Green Party).
While electoral outcomes are a useful metric, they do not capture how political attitudes shift between elections - and given the infrequency of elections, this is most of the time. For this study, we looked at the impact of climate protests on voting intention as they occurred, enabling us to track political attitudes between election cycles. It is also plausible for changes in voting intention to have an impact on climate policy even in between-election periods, given that policymakers closely monitor opinion polls (e.g. Bernardi, Bischof, and Wouters, 2021), especially on issues that could affect their re-election prospects (Mayhew, 1974). If climate protests lead to pro-climate shifts in voting intention, they could exert indirect pressure on legislators, shaping policy decisions even outside the context of an election.
It is an open question, however, whether climate protests actually move voting intention in a more pro-climate direction. This is particularly true of disruptive climate protests (for example, blocking roads or disruptive sporting events); while such tactics have the potential to increase pro-climate voting intention due to the high levels of mediatic and public attention they reach, they also have the potential to irritate voters and ignite backlash in an anti-climate direction (Fuller et al., 2025), or to do both at the same time and increase polarisation on the issue.
Method
This research uses polling data from a two-year period to assess the impact of disruptive climate protests on voting intention in three countries: the United Kingdom, Sweden, and Germany. Specifically, we are interested in whether disruptive climate protests move voting intention in a pro-climate direction, by increasing voting intention for pro-climate parties and/or decreasing voting intention for anti-climate parties, or whether the opposite occurs and disruptive climate protests trigger a backfire effect, increasing support for anti-climate parties.
We study this using data from 916 days for which at least one voting intention poll was available. The country breakdown is: Germany (612 polls from a total of 406 polling days), Sweden (167 polls from 119 polling days), and the UK (514 polls from 400 polling days) over a two-year period (January 2022–December 2023). We use polls from PolitPro (https://politpro.eu), a platform that collects daily polls from across Europe. Our protest data comes from ACLED (Armed Conflict Location and Event Data), a database tracking protest and conflict data across the world. Both organisations’ data are of high quality, are widely used in analyses and both regularly update and monitor their databases.
The dates (January 2022–December 2023) were selected because JSO was founded early 2022 and thus is the earliest time that data were available for all groups. December 2023 was the most recent point for which protest data were available at the time of collection. These countries were chosen because they are all major economies with significant emissions, have made ambitious climate commitments, are home to high-profile disruptive climate organisations, and adequate polling data was available. All polls track voting intentions in the respective country’s next general election.
​
We focus on protests by a single group within each country: Just Stop Oil (JSO) in the UK, the Återställ Våtmarker (Restore Wetlands Movement; RWM) in Sweden, and Letzte Generation (Last Generation; LG) in Germany. Each of these groups has had a high profile in that country’s protest context. In the United Kingdom, Just Stop Oil’s past actions have included disrupting the 2023 national Snooker championships and throwing soup on a Van Gogh painting (Scheuch et al. 2024). In Germany, Last Generation staged hundreds of disruptive climate actions across Germany each year. In Sweden, the Restore Wetlands Movement has engaged in various high-profile actions, including road blockades and disruptions of live televised events. All three groups are members of the A22 Network, an international climate protest network. Our focus on disruptive protest allows us to look at the risk of a backfire effect: if there were a negative electoral effect from climate protests, we would expect it from disruptive groups who tend to be more negatively covered in the media (Scheuch et al. 2024) and are viewed more negatively by the public (Vandeweerdt, 2024).
In each country, we use a pre/post design that compares voting intention before and after climate protests by our three key groups. This research design is a variation of an Unexpected Event Study Design (UESD). UESDs typically use the occurrence of an unexpected event while a poll is in the field to measure the impact of that event on a certain variable, by comparing average levels of that variable before and after the unexpected event happens. The design is widely used to measure the impact of a variety of events, from terrorist attacks to political scandals to epidemics. It has previously also been used to study protests (Munoz et al. 2020). We vary the design slightly by comparing the results of polls taken within seven days after a protest to an average of three polls before the protest, to reduce the noise in baseline vote share numbers. We also note that sometimes several polls were available for the same day, in which case we simply used the average across them.
In our main specification, a poll is considered “treated” (affected by a protest) if the groups in our database held one or more protests in the seven days leading up to the poll being released [1]. We use a seven-day window for two reasons: first, we assume that protests have the strongest potential to affect voting preferences in the days following the protest since this is when they are most likely to be reported in the media, and second, to allow for the fact that polls are sometimes in the field (collecting data) several days before the results are released. To robustness-check this approach, the supplementary material contains variations on this seven-day window and shows that 1) reasonable variations of the window (5-15 days) lead to similar results, 2) the choice of seven days tends to be supported statistically because it leads to better model fits, 3) very short windows (e.g., 3 days) produce noisier estimates, as expected given their reduced likelihood of capturing the protest’s effects.
The outcome variable utilised in all analyses is the difference between the voting intentions on a polling day of interest minus the baseline; the average of the voting intentions on three recent polling days at least seven days earlier. We used two different complementary analyses, all run separately for each country, to get an overview of how disruptive climate protests affected voting intention:
(1) Per-party analyses. These analyses evaluate, for each political party, whether and to what extent protests affected voting intentions. The regression model takes the form:
​
where “Difference” is the difference between that party’s vote share in the poll in question and an average of the three polls making up our baseline control polls, and “Treated” is a binary variable recording whether or not there was a protest in the 7 days prior to the poll (henceforth the “protest dummy”). This binary variable does not make a distinction between polls that saw a single protest during that seven-day window and polls that saw multiple protests. We use a binary variable, rather than a continuous one, because our sample size in each country is too small to reliably estimate effects due to different numbers of protests. “Time Effects” are fixed effects representing the quarter and week that the poll occurred. Our inclusion of fixed effects is intended to account for the confounding factor of time, in that pro-climate protests might plausibly occur in time periods where pro-climate vote share may be independently changing, for example during major climate conferences or highly-publicised climate change–related disasters. Time effects account for the variation in vote share over time and help to rule out this explanation. All analyses were performed using Bayesian regression models as implemented in the R (version 4.3) package brms (version 2.19).
(2) Assessing the effect of a party’s favourability towards climate change policies. For this analysis, we assessed for each country whether the data support an overall shift of voting intentions in favour of, or against, more pro-climate parties. In considering parties’ favorability towards climate policies, we generated a classification using GPT-4o (ChatGPT, April 2025), via Deep Research, drawing on publicly available information including:
-
Policy commitments (manifestos, targets)
-
Legislative records (voting behavior, implementation)
-
Public statements and party communications
-
Media and expert assessments (e.g. Greenpeace, Carbon Brief, Naturskyddsföreningen)
-
Public perception (voter surveys and party reputation)
Parties were scored on a 7-point scale ranging from 1 (strongly anti-environmental) to 7 (strongly pro-environmental), and received labels to categorize their climate change stance more broadly: “Anti” (1–2), “Neutral/Mixed” (3–4), or “Pro” (5–7). A broader range (5–7) was allocated to the “Pro” category than other categories to reflect important nuance among supportive parties. A score of 5 includes parties that are at least nominally pro-climate and support some action—though often cautiously or inconsistently—while scores of 6 and 7 are reserved for parties that treat climate change as a core issue and align closely with climate science. The model’s outputs were cross-checked independently by team members, who confirmed the analysis and the classification’s validity. We then used these climate favourability scores to test whether this measure interacts with the protest dummy, assessing whether voting intentions for parties is influenced by protests as a function of their stance on climate change. The classification results and climate favourability scores are included in the tables for each country in the Results.
The full analysis scripts and data can be found on OSF.
​​Results
Germany​
​
In Germany, we ran analyses for five main political parties: the AfD, the Gruene (Green) party, the Linke Party, the ruling Social Democrats, and the Christian Union. Of these parties, the Linke and Green parties are the most explicitly pro-climate, the Social Democrats and Christian Union parties have championed climate policies at various points in the past, and the AfD has historically been strongly against climate policy. For example: the Die Linke and Green parties both include strong climate action as a core tenant in their party programmes; on the other end of the spectrum, the AfD has denied that humans contribute to climate change (Connolly, 2019). Specific areas of tension include how quickly Germany should develop renewable energy, with the AfD opposing some types of renewable energy such as wind power entirely (Thompson, 2025), while the Linke and Green parties support a rapid and total decarbonisation of the electricity grid. The Social Democrats and the CDU usually fall somewhere in between, with, for example, the CDU supporting carbon neutrality by 2045 but opposing a Social Democrat-backed law to regulate the greenhouse gas emissions of new buildings (a law backed by the Green and Linke parties) (Thompson, 2025).
Figure 1 plots the effects of Last Generation protests on voting intention for each of the five parties. Table 1 shows these effects in percentage points. In each case, we have 406 polling days: 330 with and 76 without protests occurring in the seven days prior. The per-party analyses show that LG protests were associated with an increase in voting intention for the Linke (estimate = 0.19, CrI [0.003, 0.37]) and with a decrease in voting intention for the AfD (estimate = -0.44, CrI [-0.73, -0.15]). Assuming that people would indeed vote as they indicated, this corresponds to an estimated 94.3K increase in votes for the Linke and a 218.5K drop in votes for the AfD.
The analysis directly assessing the link between a party’s climate favourability and the effects of protests shows an interaction between the two, such that a one-unit increase in climate favourability is associated with a 0.06 ppt gain in voting intention from climate protests (CrI [0.01, 0.12]). This would correspond to a 29.8K vote gain per one-unit increase in a party’s climate favourability. Overall, these results are consistent with a pro-climate shift across the electoral spectrum: the left-wing parties classified as “pro-climate” saw a combined boost of 0.33 ppts from protests, which would correspond to a 163.9K additional votes.​​​
Fig 1. Impact of Last Generation protests on voting intention per party. Dots represent the posterior median estimates for the effect of protests on voting intention. Thick horizontal bars indicate 66% credible intervals, and thin lines represent 95% credible intervals. The shape of the surrounding distribution visualizes the posterior probability density for each estimate.
Table 1. Germany: Voting intention effects by party and climate policy stance. For each party, the estimates are displayed along with the lower and upper limits of the 95% credible intervals. “Position” refers to whether a party is considered “pro-climate”, “anti-climate”, or “neutral”. “Score” shows each party’s overall stance towards climate issues and policies, expressed in a 1 (strongly anti-climate) to 7 (strongly pro-climate) Likert score.
​
As outlined in the Methods, we ran robustness checks with alternative specifications for the time window in which a protest must have occurred for a given poll to be considered treated (i.e. affected by it), varying it from 3-15 days. The results shown in the supplementary information are in line with the results from our main analysis using a seven-day window: a drop in voting intention for the AfD, and a small rise in voting intention for the Linke.
Sweden
​
In Sweden, we track results for eight major parties: Moderata Samlingspartiet, Centerpartiet, Kristdemokraterna, Liberalerna, Miljöpartiet, Socialdemokraterna, Sverigedemokraterna, and Vänsterpartiet. While many parties in Sweden have favored various types of environmental and climate action in the past, the two parties most associated with climate action are the Centerpartiet, which historically represents rural interests including environmental protection, and the Miljöpartiet (Green Party), which, like its peers in Germany and the U.K., is the most closely associated with adopting pro-environmental positions.
Figure 2 plots the effects of a protest in the last seven days on voting intention for each of the parties, showing effects as percentages. Not shown, but included in the model, are the intercept and time fixed effects. 28 polls belong to the treatment group and 91 to the control group. This smaller sample size, relative to Germany and the UK, leads to larger standard errors and is the reason some of the posterior probability distributions (the gray shaded areas in Fig. 2) are very spread out. Smaller sample sizes generally lead to higher statistical uncertainty.
The per-party analysis indicates that protests by Restore Wetlands on average lead to an estimated rise of 0.45 ppts in voting intention for the Centerpartiet/Center Party (CrI [0.1, 0.8]). For reference, this would correspond to a 29.1K increase in votes.
Fig 2. Impact of Restore Wetlands protests on voting intention per party. Dots represent the posterior median estimates for the effect of protests on voting intention. Thick horizontal bars indicate 66% credible intervals, and thin lines represent 95% credible intervals. The shape of the surrounding distribution visualizes the posterior probability density for each estimate.
Overall, there appears to be a pattern whereby more pro-climate parties benefit from RW’s protests relative to more anti-climate parties. Indeed, the analysis by climate favourability reveals an interaction between a party’s climate favourability score (1-7) and the protest dummy variable indicating whether protests happened, such that a one-unit increase in climate favourability is associated with a 0.07 ppt gain in voting intention from climate protests (CrI [0.003, 0.14]). This would correspond to a 4.5K gain in votes per one-unit increase in climate favourability. Collectively, the effects for parties classified as “pro-climate” represent a rise in voting intention of 0.79 ppts toward pro-climate parties. This would correspond to an additional 51.2K votes for pro-climate parties.
Table 2. Sweden: Voting intention effects by party and climate policy stance. For each party, the estimates are displayed along with the lower and upper limits of the 95% credible intervals. “Position” refers to whether a party is considered “pro-climate”, “anti-climate”, or “neutral”. “Score” shows each party’s overall stance towards climate issues and policies, expressed in a 1 (strongly anti-climate) to 7 (strongly pro-climate) Likert score.
UK
In the UK, we ran models for six political parties of interest: the Green Party, the Conservative Party, the Labour Party, the Liberal Democrats, Reform UK, and the Scottish National Party (SNP). The most explicitly and consistently pro-climate parties tend to be the Green Party, the SNP, and the Liberal Democrats (Scheuch, 2021). Reform UK has the most anti-climate stance, rejecting the scientific consensus on climate change and opposing net-zero targets (Sandford, 2024). Figure 2 plots the effects of a protest in the last seven days on voting intention for each of the parties. Table 2 shows these effects as percentages. Not shown, but included in the model, are the intercept and time fixed effects. 206 polling days belong to the treatment group and 194 to the control group. The full analysis scripts and data can be found on OSF.
The per-party analyses indicate that Just Stop Oil protests lead to a rise in voting intention for the Green Party by 0.34 ppts (CrI [0.06, 0.63]). This would correspond to an additional 98.3K votes for the Green Party.
​
Fig 3. Impact of Just Stop Oil protests on voting intention per party. Dots represent the posterior median estimates for the effect of protests on voting intention. Thick horizontal bars indicate 66% credible intervals, and thin lines represent 95% credible intervals. The shape of the surrounding distribution visualizes the posterior probability density for each estimate.
In contrast to the other two countries examined, in the UK we do not see an overall benefit in voting intention as a function of how pro-climate a party is. This is due to the finding that Reform UK, the most anti-climate party, trended towards a positive effect. This hints at some degree of polarisation whereby voting intention is shifted towards parties on opposite ends of the climate favourability (and political left-to-right) spectrum.
Table 3. UK: Voting intention effects by party and climate policy stance. For each party, the estimates are displayed along with the lower and upper limits of the 95% credible intervals. “Position” refers to whether a party is considered “pro-climate”, “anti-climate”, or “neutral”. “Score” shows each party’s overall stance towards climate issues and policies, expressed in a 1 (strongly anti-climate) to 7 (strongly pro-climate) Likert score.
Discussion and Conclusion
We have reported evidence that climate protests have a measurable impact on voting intentions in the UK, Germany, and Sweden. Our evidence suggests that disruptive climate protests tended to shift voting preferences in a pro-climate direction.
In Germany, protests by the disruptive climate group Last Generation were associated with increased voting intention for the strongly pro-climate Linke Party (Left Party) and with a decrease in voting intention for the strongly anti-climate far-right AfD. Moreover, we saw an overall pattern whereby parties tended to benefit more from the protests to the extent that they are favourable toward climate policies. While we cannot track the movement of individual voters, one plausible explanation is that climate protests shift AfD voters to the more pro-climate Christian Union Party, and shift Social Democratic voters to the more pro-climate Green and Linke parties. The resulting net effect may be a pro-climate shift larger than that implied by the 0.33 ppt gain in voting intention for pro-climate parties. This is particularly important for the half of the Bundestag chosen by proportional representation: since the effect that we observe here would grow the pro-climate share of representatives in either a right or left-wing coalition government. Notably, while these results dovetail with results from previous studies showing positive effects from German climate protests, ours do so for a group that adopts highly disruptive protest tactics. Last Generation adopts a very different approach from the previously studied group, Fridays for Future, studied by Valentim (2023), indicating that the positive electoral effects of climate protests may not be limited to groups that adopt non-disruptive tactics, in line with Brehm and Gruhl’s (2024) results. The February 2025 German election confirmed the relevance of protest tactics that can drive voters away from the AfD and toward pro-climate parties. The election saw record results for the AfD and a resurgent performance by the Linke party, confirming that both parties will be important actors in national politics in the years to come. Influencing them offers an opportunity to potentially shape German climate policy.
In Sweden, the Centerpartiet (Centre Party) received a voting intention boost from disruptive climate protests carried out by Återställ Våtmarker/Restore Wetlands. Moreover, similarly to Germany, there was an overall pattern whereby parties tended to see more gains from protests the more they have a favourable stance on environmental and climate issues. Although the Centerpartiet has traditionally supported various green policies, its record in Government has been mixed, drawing criticism from some environmental groups (see Naturskyddsföreningen, 2022). Still, its consistent support for a number of pro-environmental policies justifies its classification as ‘pro-environmental’. The most plausible explanation for the Centerpartiet effect is that they succeeded in attracting a portion of the right-wing vote share away from other right-wing parties, as they were the most vocal pro-climate voice on the right in 2022/23 in Sweden. Our sample size was by far the smallest for Sweden. Future higher-powered studies could usefully investigate whether the impression of a possible protest benefit for the Miljöpartiet de Gröna (Green Party) and the Vänsterpartiet (Left Party) reflects an existing effect that we failed to detect and to generally provide more certain estimates for all parties.
In the UK, Just Stop Oil protests were associated with increased voting intentions for the Green Party. In contrast to the other countries we studied, we did not see a general pattern of stronger gains for more pro-climate parties. This was mostly because the anti-climate Reform UK trended towards a protest-related increase in support, contrary to the results for Germany where the comparable right-wing AfD was most negatively affected by protests. The pattern of results in the UK hints at a degree of polarisation whereby voting intention may shift towards parties on opposite ends of the climate favourability (and political left-to-right) spectrum.
One explanation for this apparently more polarised response in the UK could relate to the introduction of the Police, Crime, Sentencing and Courts Act 2022. The Act expanded police powers to restrict protests and provoked widespread opposition, including the “Kill the Bill” movement, resulting in a more politicised environment around protest in the UK. A second factor may be the UK’s hostile media environment. Right-leaning outlets (e.g., The Daily Mail, The Sun) have long framed disruptive protests in highly negative terms (Scheuch et al., 2024). Such coverage may provoke backlash among conservative and moderate voters while simultaneously energising climate-concerned groups. However, this remains speculative and further research is needed to assess whether similar media dynamics were present in Germany and Sweden.
While the Labour Party adopted a strongly pro-climate stance during the period under investigation, protests were associated with a trend towards decreased support for the party. Though we can’t track individual voter movement, a plausible explanation is that some climate-concerned voters shifted to the more explicitly pro-climate Green Party, perceiving Labour’s stance as insufficiently bold.
​
Although the effect sizes observed in this study are generally small across all three countries, it is notable that disruptive climate protests had a measurable impact at all, especially given that only a small portion of the population is typically exposed to any single protest. This suggests that the actual effect of exposure may be larger than our reported figures imply. In some cases, the relative impact is quite substantial. For example, the estimated gain of 0.45 ppts for the Swedish Centerpartiet represents a nearly 7% shift relative to their vote share of around 6-7% in 2022/23 [2]. Our effect sizes are in line with other studies of protest impact. For example, Valentim (2023) finds that Friday for Future protests increase vote share for the Green party in Germany by up to 2 percentage points. The fact that our effect sizes are in line with other studies increase our confidence in our findings.
While our sample size (916 polling days) is reasonable, these polls are not evenly distributed across countries, with a particularly low number of Swedish polls. Future research could place an emphasis on collecting additional data across all three cases, and particularly in Sweden. Doing so would shrink the margins of error around our estimates and potentially help to recover a more precise estimate of which parties are losing how many voters as pro-green parties gain. Additionally, panel studies (that is, studies that track the same individuals over time), which are currently rare in these contexts, could provide further insights into specific voter movement patterns between parties.
These results should also be considered in the wider context of outcomes that disruptive protests can influence. For example, Brehm and Gruel (2024) found that disruptive protests increased climate concern, while Ostarek et al. (2024) found that disruptive protests increased support for moderate environmental groups. Additionally, we can consider these results in the context of other interventions aimed at influencing voting intention. For example, effects from experiments testing the effectiveness of political adverts find effects ranging from smaller than those we observe (e.g. Konitzer et al. 2019) to those slightly larger than ours, in the range of several percentage points (Goldberg et al. 2021). Research on in-person campaigning, an alternative approach to disruptive protests, also finds similar results to those we observe (Kalla and Broockman, 2018). Collectively, these findings suggest that, despite widespread criticism of disruptive protest, it can have an impact on voting intention in line with or slightly exceeding the impact of alternative interventions.
These results are encouraging for advocates of disruptive climate action for several reasons. First, they are quite consistent across three countries with different compositions of parties and different electoral systems. Second, we find little evidence of backfire effects; in all three countries, pro-climate parties are not negatively affected by disruptive climate protests in a significant way. Finally, as discussed above, even modest shifts in reported voting intention can have important impacts, indirectly influencing legislative agendas and shaping climate policy. Taken together, these findings suggest that disruptive climate protests can be a politically viable tactic, one that not only raises awareness but may also contribute to lasting political change.
​
Footnotes
[1] Note that data collection for polls is usually done a few days before the release date up until the release date.
[2] The % shift is obtained by dividing the protest effect (0.45 ppts) by the party’s vote share (6-7%).
About Social Change Lab
​
Social Change Lab conducts empirical research on disruptive protest and people-powered movements. Through research reports, workshops, and trainings, we provide actionable insights to help movements and funders be more effective. You can find all of our research projects and resources on our website. You can contact us at info@socialchangelab.org
References
Agnone, J. (2007). Amplifying public opinion: The policy impact of the US environmental movement. Social forces. 85, 1593–1620.
Bernardi, L., Bischof, D., & Wouters, R. (2021) The public, the protester, and the bill: do legislative agendas respond to public opinion signals?, Journal of European Public Policy, 28:2, 289-310,
Brehm, J., & Gruhl, H. (2024). Increase in concerns about climate change following climate strikes and civil disobedience in Germany. Nature Communications, 15(1), 2916.
Connolly, K. (2019). Germany’s AfD turns on Greta Thunberg as it embraces climate denial. Guardian. 14 May. https://www.theguardian.com/environment/2019/may/14/germanys-afd-attacks-greta-thunberg-as-it-embraces-climate-denial
Dasch, S.T., Bellm, M., Shuman, E., Zomeren, M. van. (2024). The Radical Flank: Curse or Blessing of a Social Movement? Global Environmental Psychology, 2, 1–33.
Fabel, M., Flückiger, M., Ludwig, M., Rainer, H., Waldinger, M., Wichert, S. (2022). The Power of Youth: Political Impacts of the “Fridays for Future” Movement (Working Paper No. 9742). CESifo Working Paper.
Fabel, M., Flückiger, M., Ludwig, M., Rainer, H., Waldinger, M., & Wichert, S. (2025). The relationship between the youth-led Fridays for Future climate movement and voting, politician and media behaviour in Germany. Nature Human Behaviour, 9(3), 481-495.
Fuller, K., Ferrali, R., Damesin, L., Gainsburg, I., Simpson, B., Willer, R. (2025). Extreme Protest Tactics Reduce Support for the Climate Movement and Climate Mitigation Policies. https://doi.org/10.31235/osf.io/n6aw2
Goldberg, M.H., Gustafson, A., Rosenthal, S.A. & Leiserowitz, A. (2021). Shifting Republican views on climate change through targeted advertising. Nat. Clim. Chang. 11, 573–577
Kalla, J. L., & Broockman, D. E. (2018). The minimal persuasive effects of campaign contact in general elections: Evidence from 49 field experiments. American Political Science Review, 112(1), 148-166
Kenward, B., Brick, C. (2024). Large-Scale Disruptive Activism Strengthened Environmental Attitudes in the United Kingdom. Global Environmental Psychology, 2, 1–35. https://doi.org/10.5964/gep.11079
Konitzer, T., Rothschild, D., Hill, S., & Wilbur, K. C. (2019). Using big data and algorithms to determine the effect of geographically targeted advertising on vote intention: Evidence from the 2012 US presidential election. Political Communication, 36(1), 1-16.)
Kountouris, Y., Williams, E. (2023). Do protests influence environmental attitudes? Evidence from Extinction Rebellion. Environ. Res. Commun. 5, 011003. https://doi.org/10.1088/2515-7620/ac9aeb
Mayhew, D. R. (1974). Congressional Elections: The Case of the Vanishing Marginals. Polity, 6(3), 295–317. https://doi.org/10.2307/3233931
Muñoz, J., Falcó-Gimeno, A., & Hernández, E. (2020). Unexpected event during survey design: Promise and pitfalls for causal inference. Political analysis, 28(2), 186-206.
Naturskyddsföreningen. (2022). Centerpartiets miljöpolitik – vad säger de och vad har de gjort? https://www.naturskyddsforeningen.se/artiklar/centerpartiets-miljopolitik/
Ostarek, M., Simpson, B., Rogers, C., & Ozden, J. (2024). Radical climate protests linked to increases in public support for moderate organizations. Nature Sustainability, 1-7.
Parkes, I., Thomas-Walters, L., Sabherwal, A., O'Dell, D., & Shreedhar, G. (2023). Social identification predicts behavioural engagement with environmental activist movements, but does not moderate the collective climate action intention-behaviour gap.
Rogers, C., Ostarek, M., Jones, M., & Özden, J. (2024). What was the impact of the Insulate Britain campaign? Social Change Lab. https://2ed91cf7-c25d-453e-a3d5-a90f73ccb6ab.usrfiles.com/ugd/2ed91c_4112e779eaf240ebb3f119082f423b40.pdf
Sandford, D. (2024). Reform UK election pledges: 11 key policies analysed. BBC News. 17 June. https://www.bbc.co.uk/news/articles/cqll1edxgw4o
Scheuch, E. G. (2021). Green Evolution: Lessons from British Climate. Policy Consilience, (23), 48-58
Scheuch, E. G., Ortiz, M., Shreedhar, G., & Thomas-Walters, L. (2024). The power of protest in the media: examining portrayals of climate activism in UK news. Humanities and Social Sciences Communications, 11(1), 1-12
Scheuch, E.G. Thomas-Walters, L. Goldberg, M.H. and Leiserowitz, A. (preprint). Repeated Messaging Increases the Durability of Political Persuasion, But Only In the Short Term.
Schürmann, L. (2023). The impact of local protests on political elite communication: evidence from Fridays for Future in Germany. Journal of Elections, Public Opinion and Parties, 34, 510–530. https://doi.org/10.1080/17457289.2023.2189729
Simpson, B., Willer, R., Feinberg, M., (2022). Radical flanks of social movements can increase support for moderate factions. PNAS Nexus 1, pgac110. https://doi.org/10.1093/pnasnexus/pgac110
Thackeray, S.J., Robinson, S.A., Smith, P., Bruno, R., Kirschbaum, M.U.F., Bernacchi, C., Byrne, M., Cheung, W., Cotrufo, M.F., Gienapp, P., Hartley, S., Janssens, I., Hefin Jones, T., Kobayashi, K., Luo, Y., Penuelas, J., Sage, R., Suggett, D.J., Way, D., Long, S. (2020). Civil disobedience movements such as School Strike for the Climate are raising public awareness of the climate change emergency. Global Change Biology, 26, 1042–1044. https://doi.org/10.1111/gcb.14978
Thomas-Walters, L., Scheuch, E. G., Ong, A., & Goldberg, M. H. (2025). The impacts of climate activism. Current Opinion in Behavioral Sciences, 63, 101498.
Thompson, E. (2025). Germany set to be cooler on global warming after election but won’t ignore climate. Green Central Banking, accessed from https://greencentralbanking.com/2025/02/18/german-election-cooler-on-climate/
Valentim, A. (2023). Repeated exposure and protest outcomes: How Fridays for Future protests influenced voters. Open Science Framework.
Vandeweerdt, C. (2024). The Activist’s Trade-off: Climate Disruption Buys Salience at a Cost.
Walgrave, S. & Vliegenthart, R. (2012). The Complex Agenda-Setting Power of Protest: Demonstrations, Media, Parliament, Government, and Legislation in Belgium, 1993-2000. Mobilization: An International Quarterly, 17, 129–156. https://doi.org/10.17813/maiq.17.2.pw053m281356572h