msa_bivaltyp.R

library(tidyverse)
library(lingtypology)
df <- bivaltyp.feature()


df |> 
  filter(family_WALS == "Nakh-Daghestanian") |> 
  writexl::write_xlsx("~/Desktop/daghestan_bivaltyp.xlsx")

df |> 
  filter(family_WALS == "Nakh-Daghestanian",
         valency_pattern != "")  |> 
  select(language_external, valency_pattern, predicate_label_en) |> 
  pivot_wider(names_from = language_external, values_from = valency_pattern) |> 
  column_to_rownames("predicate_label_en") |> 
  mutate_all(factor) ->
  for_mca

for_mca |> 
  MASS::mca() ->
  mca

for_mca <- cbind(for_mca, mca$rs)
variables <- as_tibble(mca$cs)
variables$var_names <- rownames(mca$cs)

for_mca  |>  
  rownames_to_column("verbs") |> 
  filter(`2` < 0.003) |> 
  ggplot(aes(`1`, `2`, label = verbs)) +
  geom_point()+
  geom_label()
  stat_ellipse(aes(color = Aux))+
  geom_text(data = variables, aes(`1`, `2`, label = var_names))+
  theme_bw()+
  scale_y_continuous(limits = c(-0.015, 0.02))