README.md

Data, code, and paper for "Agreement Attraction in Turkish: The Case of Genitive Attractors"

Folder Structure

.
├── LICENSE
├── README.md
├── code
│   ├── analysis
│   ├── fits
│   ├── models
│   └── scripts
├── data
│   ├── all_ud_tr.conllu
│   ├── exp_data.rds
│   ├── exp_form.rds
│   └── frequencies_exp1_and_lago.xlsx
├── dataraw
│   ├── experiment
│   └── lagoetal
└── full_analysis.R

Run Order and Details

• To run all code, open full_analysis.R file and run everything inside. It may take some time due to 2 Bayesian GLM Models.
• Stand-alone scripts and data analysis can be found in R directory. Follow the order of the scripts to replicate our analysis.
  1. packages
  2. read_ibex
  3. eda
  4. misc_scripts
  5. data_wrangling
  6. word_frequencies
  7. descriptive_plots
  8. bayesian_models
  9. model_plots
  10. text_inputs
• Stan models exported from the brms package can be found in code/stan/ folder.
• Raw data from both Lago et al. (2019) and our experiment can be found in dataraw/.

Requirements

List of packages, softwares, fonts you need for replicating our paper and models:

• LaTeX
• R
• Stan
• and following R packages
  • languageR
  • tidyverse
  • gdata
  • MASS
  • magrittr
  • ggplot2
  • car
  • brms
  • xtable
  • rstan
  • knitr
  • dplyr
  • plyr