#Running all scripts #download packages and create folders #generate Glottolog table (based on Glottolog 4.4) #calculate metric scores (based on Grambank 1.0) #generate population table (all sociodemographic variables in one dataframe) #wrangling EDGE tree #generating AUTOTYP areas table source("get_external_data.R") source("generating_GB_input_file.R") source("set_up_general.R") #setup for INLA analysis source("install_and_load_INLA.R") #choosing whether to use the full dataset (possible only for reviewers and if one has own access to Ethnologue and saved the dataset in the data folder on their own) or to only to the subset of Ethnologue with transformed variables made available in this repostiory after running create_pop_table.R #sample <- "full" sample <- "reduced" #default source("make_ethnologue_SM_and_merging_tables.R") source("create_pop_table.R") source("set_up_inla.R") #run all INLA models + extract main results tables #Note that previously "fusion" was called "boundness", and this is how it is referenced in all scripts #predictors: random effects - phylogenetic and spatial (same scripts for "full" and "reduced" versions) source("models_Boundness_phylogenetic_spatial.R") source("models_Informativity_phylogenetic_spatial.R") if(sample == "full"){ #predictors: phylogenetic and spatial random effects + sociodemograhic variables as fixed effects source("models_Boundness_social.R") source("models_Informativity_social.R") #predictors: sociodemographic variables as fixed effects source("models_Boundness_social_only.R") source("models_Informativity_social_only.R") #conduct sensitivity testing + extract the corresponding table source("runs_sensitivity.R") #extract tables from INLA analyses source("table_INLA_summary_all_models_SI.R") source("variance_top_ranking_models.R") #plotting main results source("plot_social_effects_combined.R") } if(sample == "reduced"){ #predictors: phylogenetic and spatial random effects + sociodemograhic variables as fixed effects #(on reduced set of social variables: without log10 transformed L1 speakers) source("models_Boundness_reduced_social.R") source("models_Informativity_reduced_social.R") #predictors: sociodemographic variables as fixed effects source("models_Boundness_reduced_social_only.R") source("models_Informativity_reduced_social_only.R") #conduct sensitivity testing + extract the corresponding table source("runs_sensitivity_on_reduced.R") #extract tables from INLA analyses source("table_INLA_summary_all_models_SI_reduced.R") source("variance_top_ranking_models_reduced.R") #plotting main results source("plots_social_effects_combined_on_reduced.R") } #measure phylogenetic signal in two fusion and informativity source("measuring_phylosignal.R") #plotting source("plot_maps_main.R") #maps of scores source("plot_heatmap_B_I.R") #phylogenetic tree with a heatmap source("plot_map_Africa.R") source("plot_map_Eurasia.R") source("plot_heatmap_informativity_Uralic.R") #Uralic tree (informativity) + combined plot with two maps from above source("plot_spatial_parameters_linear_distances.R") #SI figure for visualizing how covariance under different kappa and phi parameters corresponds to spatial distances #additional analyses on WALS data source("make_ethnologue_SM_for_morphological_complexity_reanalysis.R") source("WALS_sparseness.R") source("WALS_reanalysis_setup.R") source("WALS_reanalysis_controlled_setup.R") source("WALS_reanalysis_controlled_setup_high_coverage.R") #analysis + summary table