**This project contains supplementary material for the manuscript “Legacy media as inhibitors and drivers of public reservations against science: Global survey evidence on the link between media use and anti-science attitudes”.** ______________ Here we share all materials needed to reproduce the results presented in the main article and the appendix, i.e. the complete survey data, code, outputs, and pre-computed <code>brms</code> and <code>lmer4</code> models. The project also includes all figures in higher resolution. Please find the following folders and files: - **01_data**: Contains the entire data used in the analyses: - <code>WVS_Cross-National_Wave_7_R_v1_6.rds</code>: World Values Survey data, publicly available at [worldvaluessurvey.org][1] (retrieved on 09/04/21) - <code>country-codes_csv.csv</code>: Country names and ISO codes, publicly available at [datahub.io][2] (retrieved on 11/04/21) - <code>All_data_FIW_2013-2021.xlsx</code>: Freedom of the press scores, publicly available at [freedomhouse.org][3] (retrieved on 11/04/21) - <code>Global Party Survey by Party SPSS V2_1_Apr_2020-2.csv</code>: Populism scores, publicly available at [Harvard Dataverse][4] (retrieved on 13/04/21) - <code>6-dimensions-for-website-2015-08-16.csv</code>: Uncertainty avoidance scores, publicly available at [geerthofstede.com][5] (retrieved on 12/04/21) <br> - **02_code**: Contains the R scripts used for the analyses. Should be executed in chronological order: - <code>01_setup.R</code>: Script to import and set up the data - <code>02_measures.R</code>: Script to set up, compute, and inspect all variables used in the analyses - <code>03_descriptive-analysis.R</code>: Script to run the analyses testing RQ1 as well as additional analyses (e.g., for creating the supplementary figures) - <code>04_multilevel-models.R</code>: Script to run analyses testing H1 and RQ2 as well as additional analyses (e.g., Frequentist procedures) <br> - **03_models**: Contains pre-computed regression models fitted with the <code>[brms][6]</code> (Bayesian models) and the <code>[lmer4][7]</code> packages (Frequentist models): - <code>brmsmeans_antisci_full.rds</code>: Bayesian null model used to estimate the prevalence of anti-science attitudes in the full sample - <code>brmsmeans_antisci_cntr.rds</code>: Bayesian null models used to estimate the prevalence of anti-science attitudes across countries - <code>brmsfit1.rds</code>: Bayesian multilevel model used to test H1 - <code>brmsfit2.rds</code>: Bayesian multilevel model used to test RQ2 - <code>brmsfit1_crits.rds</code>: <code>brmsfit1.rds</code> including LOO and WAIC fit indices - <code>brmsfit2_crits.rds</code>: <code>brmsfit2.rds</code> including LOO and WAIC fit indices - <code>lmerfit1.rds</code>: Frequentist multilevel model used to test H1 - <code>lmerfit2.rds</code>: Frequentist multilevel model used to test RQ2 <br> - **04_plots**: Contains all figures in high resolution (PNG format) _____________________ [1]: https://www.worldvaluessurvey.org/WVSDocumentationWV7.jsp [2]: https://datahub.io/core/country-codes [3]: https://freedomhouse.org/sites/default/files/2021-02/All_data_FIW_2013-2021.xlsx [4]: https://doi.org/10.7910/DVN/WMGTNS [5]: https://geerthofstede.com/research-and-vsm/dimension-data-matrix/ [6]: https://cran.r-project.org/web/packages/brms/index.html [7]: https://cran.r-project.org/web/packages/lme4/index.html