Analysis of rapid prosody transcription experiment with Stefan Baumman
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
data
.DS_Store
001_RPT_individual_analysis_preprocessing.R
002_mixed_model_analyses.R
003_random_forest_analyses.R
004_random_forest_visualization.R
005_mixed_model_interpretation_results.R
006_by_individual_analysis.R
CODEBOOK.md
README.md
summarySE_function.R

README.md

Statistical analysis of rapid prosody description

  • Study design: Stefan Baumann
  • Data collection & preparation: Stefan Baumann & Janina Kalbertodt
  • Statistical Analysis: Bodo Winter

Libraries required for this analysis:

  • lme4
  • party
  • DMwR
  • dplyr
  • reshape2
  • xlsx

Script files contained in this analysis:

  • 001_RPT_individual_analysis_preprocessing.R
    The main preprocessing script, works on Excel files and outputs tidy csv files.
  • 002_mixed_model_analyses.R
    Computes mixed models (but does not interpret and visualize them). Warning: Takes a lot of time to run.
  • 003_random_forest_analyses.R
    Computes random forests and variable importances (but does not interpret and visualize them). Warning: Takes a lot of tim to run.
  • 004_random_forest_visualization.R
    Interprets and visualizes random forests.
  • 005_visualizations.R
    Interprets and visualizes mixed models and other analyses.

Data files contained in this analysis:

  • rpt-Daten-15juli2015.xls Contains all summary data, that is, prominence score averages (overa all listeners) for each word
  • rpt-Daten-31juli2015_spectral_tilt.xls This is the most up-to-date file of the summary data
  • rpt_Einzelwerte-25juli2014-1.xls Contains individual level data, that is, all prominence ratings from each listener (wide format)
  • RPT_summary_processed.csv The summary data, cleaned and in English.
  • RPT_individual_processed.csv The individual level data, cleaned (long format) and in English.
  • CODEBOOK.md description of all columns.
  • listener_gender_info.csv is needed to map genders onto listeners.
  • speaker_gender_info.csv is needed to map genders onto speakers.
  • block_order_information.csv is needed to map block orders (there were two block orders) for each participant.