Sequence analysis of political trajectories over BR�F data

• Copyright 2021-23 No�mie F�vrat & Vincent Labatut

Bref is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation. For source availability and license information see licence.txt

• Lab site: http://lia.univ-avignon.fr/
• GitHub repo: https://github.com/CompNet/Bref
• Contact: Noemie Fevrat noemie.fevrat@gmail.com, Vincent Labatut vincent.labatut@univ-avignon.fr

This set of scripts takes advantage of the BR�F database compiled from research previously carried out by Fevrat et al. at Avignon University [LFM'20]. The results concerning the re-election of French elected officials presented in No�mie Fevrat's PhD dissertation [F'24] are based on these scripts.

If you use these scripts or the data they produce, please cite reference [F'24]:

@PhdThesis{Fevrat2021,
  author           = {F�vrat, No�mie},
  title            = {Le mandat de trop ? La r��lection des parlementaires et des maires en France et les conditions de sa remise en cause},
  school           = {Universit� d'Avignon},
  year             = {2024},
  type             = {PhD Thesis},
  address          = {Avignon, FR},
  url              = {[tel-04550896](https://hal.science/tel-04550896)},
}

The scripts are meant to be applied to v1.0.2 of the BRÉF database, that describes representatives elected in France during the Fifth Republic (1958-present). This version is available within this GitHub repository, as well as on Zenodo.

Here are the folders composing the project:

• Folder data: state Legislative elected officials, legislature after legislature.
• Folder plots: plots generated by the scripts.
• Folder src : set of scripts.
  • Folder rotations : scripts relating to the study of re-election and rotation of members of the french Parliament from one legislature to the next under the Fifth Republic.
  • Folder sequences : scripts relating to the study of the longitudinal elective careers of each of the elected representatives in the database, using the sequence analysis method.

1. Install the R language
2. Download this project from GitHub and unzip the archive.
3. Use the src/install.R script to install the required packages.

In order to replicate the experiments from the thesis, perform the following operations:

1. Open the R console.
2. Set the current projet directory as the working directory, using setwd("my/path/to/the/project/SignedBenchmark").
3. Run the script src/main_rotations.R for the rotation-related processing and src/main_sequences.R to perform the sequence analysis.

• igraph package: used to build and handle graphs.
• readr package: read rectangular text data.
• dplyr package: data manipulation.
• TraMineR package: sequence analysis.
• GDAtools package: geometric data analysis.
• plot.matrix package: matrix heatmap visualizatio.
• viridis package: color scale.
• scales package: scale functions for visualization.

• [LFM'20] V. Labatut, N. Févrat & G. Marrel, BRÉF – Base de données Révisée des Élu·es de France, Technical Report, Avignon Université, 2020. ⟨hal-02886580⟩
• [F'24] Noémie Févrat. Le "mandat de trop" ? La réélection des parlementaires et des maires en France et les conditions de sa remise en cause. PHD Thesis, Avignon Université, 2024. ⟨tel-04550896⟩