Bayes, stat! Workthrough notebook

CLE Retreat 2023

Read in and visualise data

Because we set R’s working directory to the project directory with the code in line 10, to read in the data, we can just give the relative path to the data from the project directory.

acc <- read_csv('data/acceptance.csv')

Rows: 1476 Columns: 15── Column specification ──────────────────────────────────────────────────────────────────────
Delimiter: ","
chr (12): ppt_id, condition, nom_sfx, acc_sfx, task, agent, patient, verb, orientation, se...
dbl  (3): trial_idx, sentence_accepted, rt
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

head(acc)

acc %>% 
  group_by(ppt_id, condition, sentence_type) %>% 
  summarise(propn_accepted = mean(sentence_accepted)) %>% 
  ggplot(aes(x = sentence_type, y = propn_accepted, fill = sentence_type, colour = sentence_type)) +
  geom_violin(alpha = 0.5) +
  geom_jitter(alpha = 0.5, width = 0.1) +
  facet_wrap(~ condition) +
  ylim(-0.05, 1.05) +
  labs(y = 'Proportion sentences accepted',
       x = 'Sentence type') +
  theme(legend.position = 'none') +
  scale_x_discrete(labels = c('Case marking', 'Word order')) +
  NULL