Yannic Meier

Introduction

A couple of months ago, I published a novel questionnaire that measures people’s general skepticism toward science and scientists (you can find the full paper here). We defined general science skepticism as a universal negative disposition of doubt about and distrust toward science and scientists, including used methods, influence of third parties, and transparency in reporting results. The developed items cluster on two distinct, yet closely related dimensions: skepticism toward the scientific process itself and skepticism toward the independence of science. In four studies with nearly 4,000 German participants, we demonstrated that those who are generally skeptical of science are more likely to reject scientific evidence from various fields, such as global warming, vaccinations, genetically modified foods, and evolution theory. Additionally, we found that these individuals have reduced trust in science and science journalism and exhibit increased science-related populism, conspiracy mentality, and political disenchantment. While the paper primarily focused on testing the scale’s reliability and validity, we also measured several additional variables that are worth reporting.

In this blog post, I present differences in general science skepticism among supporters of different political parties in Germany. We asked participants which party they would vote for if tomorrow was a federal election in Germany. Here, I mainly focus on the six main parties which are regularly voted into the German national parliament (the “Bundestag”). These are the Social Democratic Party (SPD), the Christian Democratic Union (CDU), the Green Party (B90), the Left Party (Linke), the Liberal Democrats (FDP; currently not represented in the “Bundestag”), and the extreme right-wing Alternative for Germany (AfD). All other parties are subsumed under the “other” category.

In Figure 1, I have visualized how the means of both factors of general science skepticism are distributed among people who said they would vote for the respective party if there were a federal election in Germany tomorrow. Importantly, the paper comprises two datasets with these variables, allowing me to replicate the results and making these findings more robust. The first dataset (N = 1059) is from February 2022 while the second one (N = 1415) is from September 2025. The samples were recruited by two different German panel providers and are representative for the German population regarding age and gender quotas. Detailed descriptions of the samples can be found in the paper. Please note that differences across both time points should not be interpreted as changes over time as there are too few datapoints and the subsample sizes are too small to confidently make such statements. In Figure 2, the number of party supporters are shown across both datasets.

Results

Figure 1

Note. Dashed black lines represent the sample mean of both subdimensions. Black dots with accompanied numbers represent the mean of each party supporter group. The sample size of the 2022 study is N = 1059 and the sample size of the 2025 study is N = 1415. Both samples are representative of the German population regarding age and gender quotas.

Figure 2

The distributions in Figure 1 show that supporters of the democratic parties score around the respective sample means, but there is also one exception. Green Party supporters seem to be noticeably less skepticial toward science than the samples were on average. Another striking observation is that supporters of the extreme right-wing AfD are more skeptical about science than the sample average. These differences pertain to both dimensions of science skepticism, which means that AfD supporters are more likely to distrust and doubt scientists’ integrity and the whole scientific process while being at the same time more skeptical about the independence and transparency of science. Importantly, this pattern replicates across the time range of 3.5 years between both samples and can, therefore, be considered as robust.

Interestingly, similar divides between supporters of the Greens and the AfD have been found in other studies as well. For instance, a study by Weisskircher et al. (2026) found that Green Party supporters were the most likely to support renewable energies, while AfD supporters were the least likely. Similarly, Green Party supporters were found being most likely to accept vaccinations, while AfD supporters were least likely to get vaccinated (Magnus et al., 2025). This pattern may be explained by the very DNA of these parties. Whereas the Green Party’s main goal is to minimize the extent of global warming largely in concordance with scientific evidence, the AfD regularly attacks the integrity of scientists and the validity of scientific evidence (Küppers, 2024). Thus, persons who either trust and support or exhibit strong doubt and distrust toward science and deny empirical evidence may be drawn to these respective parties. At the same time, it is also conceivable that each party’s rethoric as either pro- or anti-science contributes to growing trust or skepticism among their electorate. The findings presented here contribute to existing evidence showing that not only are AfD supporters more likely to act contrary to the best scientific knowledge in certain areas, but they also appear to generally oppose science and scientists while the opposite is true for Green Party supporters.

These results are concerning since, although the share of Germans who trust and distrust science has stayed pretty stable across recent years (Science Barometer, 2025), the number of people who openly support the right-wing AfD has recently increased. Simultaneously, the number of Green Party supporters has decreased. This does, of course, not mean that persons who change their political affiliation immediately become more or less skeptical about science (note that the means of general science skepticism were stable across the range of 3.5 years). However, changes in political affiliation may be, for example, accompanied by changes in media consumption (e.g., following political candidates or party channels on social media), which may slowly but gradually lead to shifts in science skepticism and trust in science.

Implications and Conclusion

In conclusion, the results presented here indicate that supporters of democratic parties do not exhibit very strong science skepticism with supporters of the Green Party being least skeptical about science. Across all parties, people always seem to be somewhat more skeptical towards the independence and transparency of science than towards the scientific process and the integrity of scientists (we discuss this general pattern in the paper). Most skeptical about science are supporters of the AfD. Although these findings are certainly not very surprising, they are nonetheless concerning given the high (and seemingly growing) support for the far-right AfD across Germany and especially in Eastern Germany. While one may say that it is every person’s personal right to be skeptical about science and to not act according to the best scientific knowledge (e.g., by not getting vaccinated or not engaging in climate protection), the potential societal consequences should be of concern to everyone. With the possibility of becoming the strongest party in a federal state for the first time this year, and even having the chance to form the state’s government, the AfD has already announced “reforms” to the university system that constitute intolerable attacks on academic freedom (see Kwon, 2026). Therefore, if expressed in voting decisions, every person’s skeptical disposition towards science may eventually result in consequences that affect everyone such as declining academic freedom.

Accordingly, a key task for scholars should be examining ways to sustainably and independently of party affiliation, maintain or increase trust in science and reduce science skepticism. This does not seem to be an easy endeavor, as a current unsuccessful attempt to improve trust in science among conservative U.S. citizens demonstrates (Gligorić et al., 2025). Meaningful and sustainable change in people’s trust in science or science skepticism may require elaborate, time-intensive, dialogical, and repeated types of interventions. At the same time, policymakers should ensure that school curricula cover basic scientific knowledge to promote an understanding of (empirical) science to potentially counteract skepticism. Finally, defending science should be of importance to anyone who trusts it. Well-established scientific facts are indisputable. No one has the right to deny them without being challenged.

Final Note

There is no need to be skeptical about the results presented in this post ☺︎ You can reproduce the results and recreate the figures yourself! The datasets are publicly available on the related OSF page and in the box below, you’ll find the R-code to recreate the figures.

# Party preferences and general science skepticism
## load packages
packages <- c("tidyverse",   # data wrangling
              "Hmisc",       # miscellaneous functions
              "ggplot2",     # data visualization
              "ggpubr"       # arrange multiple plots
)
lapply(packages, require, character.only = T)
## load data
d_2 <- read.csv("dataset_2.csv")
d_4 <- read.csv("dataset_4.csv")
# study 2
study_2_vars <- d_2 |>
  select(skept_01, skept_06, skept_08, skept_09, skept_11, skept_14, 
         skept_16, skept_18, skept_20, skept_23, skept_24,
         polaffil_01)
f1_vars <- study_2_vars |>
  data.frame(factor_mean = rowMeans(study_2_vars |> select(skept_01:skept_14)),
             affiliation = study_2_vars$polaffil_01,
             factor = "Skepticism Toward the Scientific Process") |>
  select(-contains("skept"))
f2_vars <- study_2_vars |>
  mutate(factor_mean = rowMeans(study_2_vars |> select(skept_16:skept_24)),
         affiliation = study_2_vars$polaffil_01,
         factor = "Skepticism Toward the Independence of Science") |>
  select(-contains("skept"))
study_2_input <- rbind(f1_vars, f2_vars)
study_2_input <- study_2_input |>
  mutate(affiliation = case_when(affiliation == 1 ~ "Social Democrats (SPD)",
                                 affiliation == 2 ~ "Christian Democrats (CDU)",
                                 affiliation == 3 ~ "Green Party (B90)",
                                 affiliation == 4 ~ "Left Party (Die Linke)",
                                 affiliation == 5 ~ "Liberal Democrats (FDP)",
                                 affiliation == 6 ~ "Right-Wing Extremists (AfD)",
                                 affiliation == 7 ~ "other / non-voters"))
# means for each affiliation and subdimension
means <- study_2_input %>%
  group_by(affiliation, factor) %>%
  summarise(mean = mean(factor_mean, na.rm = TRUE), .groups = "drop")
# overall mean
subdimension_means <- study_2_input %>%
  group_by(factor) %>%
  summarise(mean = mean(factor_mean, na.rm = TRUE))
# set party colors
colors_1 = c("Social Democrats (SPD)" = "#E3000F",
             "Christian Democrats (CDU)" = "#000000",
             "Green Party (B90)" = "#1AA037",
             "Left Party (Die Linke)" = "#E3000F",
             "Liberal Democrats (FDP)" = "#FFEF00",
             "Right-Wing Extremists (AfD)" = "#0489DB",
             "other / non-voters" = "darkgrey")
# plotting
plot_1 <- ggplot(study_2_input,
                 aes(x = factor_mean,
                     y = reorder(affiliation, -factor_mean))) +
  geom_violin(aes(fill = affiliation),
              scale = "width",
              alpha = 0.4,
              width = 0.4) +
  geom_point(data = means,
             aes(x = mean, y = affiliation, color = "Mean"),
             size = 3) +
  geom_text(data = means,
            aes(x = mean, y = affiliation, label = round(mean, 2)),
            hjust = -0.35,
            color = "black",
            size = 3) +
  geom_vline(data = subdimension_means,
             aes(xintercept = mean),
             linetype = "dashed",
             color = "black",
             linewidth = 0.6) +
  facet_grid(~factor(factor,
                     levels = c("Skepticism Toward the Scientific Process",
                                "Skepticism Toward the Independence of Science"))) +
  scale_fill_manual(values = colors_1) +
  scale_color_manual(values = c("Mean" = "black")) +
  labs(x = "Score (1-7 Likert-scale)", y = "") +
  ggtitle("Dataset 1 (February 2022)") +
  theme_minimal(base_size = 15) +
  scale_x_continuous(breaks = 1:7) +
  theme(legend.position = "none",
        plot.title = element_text(hjust = 0.5, size = 15))
# study 4
study_4_vars <- d_4 |>
  select(contains("skept"),
         polaff)
f1_vars_2 <- study_4_vars |>
  data.frame(factor_mean = rowMeans(study_4_vars |> select(skept_01:skept_06)),
             affiliation = study_4_vars$polaff,
             factor = "Skepticism Toward the Scientific Process") |>
  select(-contains("skept"))
f2_vars_2 <- study_4_vars |>
  mutate(factor_mean = rowMeans(study_4_vars |> select(skept_07:skept_11)),
         affiliation = study_4_vars$polaff,
         factor = "Skepticism Toward the Independence of Science") |>
  select(-contains("skept"))
study_4_input <- rbind(f1_vars_2, f2_vars_2)
study_4_input <- study_4_input |>
  mutate(affiliation = case_when(affiliation == 1 ~ "Social Democrats (SPD)",
                                 affiliation == 2 ~ "Christian Democrats (CDU)",
                                 affiliation == 3 ~ "Green Party (B90)",
                                 affiliation == 4 ~ "Left Party (Die Linke)",
                                 affiliation == 5 ~ "Liberal Democrats (FDP)",
                                 affiliation == 6 ~ "Right-Wing Extremists (AfD)",
                                 affiliation == 7 ~ "non-voters",
                                 affiliation == 8 ~ "other"))
# means for affiliation and subdimension
means_2 <- study_4_input %>%
  group_by(affiliation, factor) %>%
  summarise(mean = mean(factor_mean, na.rm = TRUE), .groups = "drop")
# overall mean
subdimension_means_2 <- study_4_input %>%
  group_by(factor) %>%
  summarise(mean = mean(factor_mean, na.rm = TRUE))
# setting colors
colors_2 = c("Social Democrats (SPD)" = "#E3000F",
             "Christian Democrats (CDU)" = "#000000",
             "Green Party (B90)" = "#1AA037",
             "Left Party (Die Linke)" = "#E3000F",
             "Liberal Democrats (FDP)" = "#FFEF00",
             "Right-Wing Extremists (AfD)" = "#0489DB",
             "other" = "darkgrey",
             "non-voters" = "lightgrey")
# plotting
plot_2 <- ggplot(study_4_input,
                 aes(x = factor_mean,
                     y = reorder(affiliation, -factor_mean))) +
  geom_violin(aes(fill = affiliation),
              scale = "width",
              alpha = 0.4,
              width = 0.4) +
  geom_point(data = means_2,
             aes(x = mean, y = affiliation, color = "Mean"),
             size = 3) +
  geom_text(data = means_2,
            aes(x = mean, y = affiliation, label = round(mean, 2)),
            hjust = -0.35,
            color = "black",
            size = 3) +
  geom_vline(data = subdimension_means_2,
             aes(xintercept = mean),
             linetype = "dashed",
             color = "black",
             linewidth = 0.6) +
  facet_grid(~factor(factor,
                     levels = c("Skepticism Toward the Scientific Process",
                                "Skepticism Toward the Independence of Science"))) +
  scale_fill_manual(values = colors_2) +
  scale_color_manual(values = c("Mean" = "black")) +
  labs(x = "Score (1-7 Likert-scale)", y = "") +
  ggtitle("Dataset 2 (September 2025)") +
  theme_minimal(base_size = 15) +
  scale_x_continuous(breaks = 1:7) +
  theme(legend.position = "none",
        plot.title = element_text(hjust = 0.5, size = 15))
# combine plots into single figure
plot <- ggarrange(plot_1, plot_2,
                  ncol = 1)
# save plot
ggsave(plot = plot, "affiliation_plot.png", height = 12, width = 10, dpi = 650)
# participants per party
participants <- data.frame(affiliation = c("other / non-voters",
                                           "Right-Wing Extremists (AfD)",
                                           "Liberal Democrats (FDP)",
                                           "Left Party (Die Linke)",
                                           "Green Party (B90)",
                                           "Christian Democrats (CDU)",
                                           "Social Democrats (SPD)",
                                           "other / non-voters",
                                           "Right-Wing Extremists (AfD)",
                                           "Liberal Democrats (FDP)",
                                           "Left Party (Die Linke)",
                                           "Green Party (B90)",
                                           "Christian Democrats (CDU)",
                                           "Social Democrats (SPD)"),
                          participants = c(149, 105, 122, 95, 179, 179, 200, 276, 269, 38, 203, 227, 224, 178),
                          study = c("2022", "2022", "2022", "2022", "2022", "2022", "2022",
                                    "2025", "2025", "2025", "2025", "2025", "2025", "2025"))
plot_3 <- ggplot(participants, 
                 aes(x = participants,
                     y = reorder(affiliation, participants),
                     fill = factor(study, levels = c("2022", "2025")))) +
  geom_bar(stat = "identity", width = 0.5, position = position_dodge(0.7)) +
  scale_fill_manual(values = c("#c2e699", "#238443")) +
  xlim(0,300) +
  labs(fill = "study") +
  theme_minimal(base_size = 15) +
  ylab("") + xlab("number of party supporters per study")
ggsave(plot = plot_3, "supporters.png", height = 5, width = 10, dpi = 650)

Cited References

Gligorić, V., Van Kleef, G. A., & Rutjens, B. T. (2025). Political ideology and trust in scientists in the USA. Nature Human Behaviour, 9(7), 1501–1512. https://doi.org/10.1038/s41562-025-02147-z

Küppers, A. (2024). ‘Climate-Soviets,’ ‘Alarmism,’ and ‘Eco-Dictatorship’: The Framing of Climate Change Scepticism by the Populist Radical Right Alternative for Germany. German Politics, 33(1), 1–21. https://doi.org/10.1080/09644008.2022.2056596

Kwon, D. (2026). Scientists fight back against far-right plans to restrict academic freedom in Germany. Nature, 655(8122), 292–292. https://doi.org/10.1038/d41586-026-01496-7

Magnus, K. D., Dammann, N., & Lüdecke, D. (2025). Vaccination readiness and political party preference in Germany: Trust, collective responsibility, and the populist radical right. PLOS One, 20(7), e0328045. https://doi.org/10.1371/journal.pone.0328045

Science Barometer (2025). Wissenschaft im Dialog. https://wissenschaft-im-dialog.de/documents/528/WIBA_2025_en_Web.pdf

Weisskircher, M., Diermeier, M., Singer, L., Krondorfer, Z., & Oertel, E. (2026). Explaining Attitudes Towards Climate Action in Germany: AfD vs. the Greens, the East vs. the West, Wind vs. Solar Energy. German Politics, 1–27. https://doi.org/10.1080/09644008.2025.2587229