🗳️ Climate Opinions and Elections

This repository contains the code and materials for the master's thesis "Elections and Climate Attitudes: How Do People’s Views on Climate Change and Policies Shift During an Election?" by Paraskevas K. Leivadaros, supervised by Kyuri Park at the University of Amsterdam (MSc in Data Science, 2025). The thesis examines how public perceptions of climate change, willingness to pay, and support for climate policies shift during election periods, using survey data and temporal analysis methods.

📌 Research Questions

1. Does support for climate policies (like carbon taxes or emissions standards) change during elections? And is this support influenced by personal or community-level perceptions of climate harm?
2. Does willingness to pay for climate solutions vary during elections and what factors influence it?
3. Does political ideology moderate the relationship between perceptions of harm and willingness to pay?

📊 Dataset Summary

• 📈 Observations: 5,667
• 📊 Variables: 20 numeric (11 float64, 9 int64)
• 📅 Time Range: 2020–2021 (Waves 2,3 and 4, centered on U.S. election)

Key Dimensions

Dimension	Variables
Climate concern	cc4_world, cc4_wealthUS, cc4_poorUS, cc4_comm, cc4_famheal, cc4_famecon
Willingness to pay (WTP)	ccSolve100, ccSolve50, ccSolve10, ccSolve1, ccSolve0 (merged into ccSolve)
Policy support	cc_pol_tax, cc_pol_car
Political orientation	pol_party, pol_lean, pol_ideology
Demographics	dem_income, dem_male, dem_age, dem_educ

🔍 Exploratory Data Analysis (EDA)

Key Insights

• Global and community-level concern about climate change is consistently higher than personal or economic concerns.
• Support for policies, particularly emissions standards (cc_pol_car), remains high across waves, while willingness to pay (WTP) drops when financial costs increase.
• Political identity strongly predicts policy support, but is weakly correlated with general concern.
• Dimensionality reduction (PCA, ICA, Factor Analysis) highlights three core constructs:
  1. Climate impact perception
  2. Willingness to pay
  3. Political/policy alignment

👉 Full visualizations are available in notebooks/1-eda.ipynb, including:

• KDE plots and histograms
• Correlation heatmaps (Spearman)
• Dimensionality reduction results (PCA, ICA, Factor Analysis)

⚙️ Methodology

1. Time Series Modeling

• Panel VAR (PVAR) is used to model temporal dynamics of climate perceptions, policy support, and WTP.
• Election-period effects are evaluated via subgroup comparisons and time-aligned estimations.

2. Causal Discovery

• PCMCI+ algorithm from tigramite identifies time-lagged causal relationships among variables.
• Results from PVAR and PCMCI+ are compared to validate robustness of causal inferences.

3. Moderation Models

• Interaction effects between harm perception and political ideology are tested to identify conditional influences on WTP and policy attitudes.
• Multicollinearity is addressed using a composite harm index.

📁 Project Structure

📁 data/
📁 docs/
    ├── paraskevas-leivadaros-master-thesis.pdf
📁 notebooks/
    ├── 1-eda.ipynb                             # Exploratory data analysis
    ├── 2-modeling-and-inference.ipynb          # Time Series and Causal Analysis
📁 results/
📁 scripts/
📜 LICENSE
📄 README.md
📄 requirements.txt
📟 setup.py

📦 Installation

This project requires Python 3.13.0 and pip 25.1.1. To avoid conflicts with system-wide packages, we recommend installing Python via a package manager (like Homebrew) and using a virtual environment.

🔧 Step 0: Install Python 3.13.0 and pip 25.1.1

🐍 Option A: Using Homebrew (macOS/Linux)

If you don’t have Homebrew installed, run:

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

Then install Python 3.13.0:

brew install python@3.13

Add it to your shell (if needed):

echo 'export PATH="/opt/homebrew/opt/python@3.13/bin:$PATH"' >> ~/.bash_profile
source ~/.bash_profile

Upgrade pip:

python3 -m pip install --upgrade pip

🪟 Option B: On Windows

1. Download Python 3.13.0 from the official website:
   https://www.python.org/downloads/release/python-3130/

2. During installation:

   • ✅ Check “Add Python to PATH”
   • ✅ Enable pip in the installer options

3. After installation, confirm with:

   python --version
   pip --version

1. Clone the Repository

git clone https://github.com/paraskevasleivadaros/climate-opinions-and-elections.git
cd climate-opinions-and-elections

2. Create a Virtual Environment

python3 -m venv venv

3. Activate the Virtual Environment

• macOS/Linux:

  source venv/bin/activate

• Windows (Command Prompt):

  venv\Scripts\activate

• Windows (PowerShell):

  .\venv\Scripts\Activate.ps1

4. Verify the Python Environment

which python

Expected output (macOS/Linux):

/Users/yourname/climate-opinions-and-elections/venv/bin/python

5. Install the Project Package and Dependencies

pip install .

This installs all dependencies from setup.py

📚 For more help:

• Python: https://www.python.org/
• pip: https://pip.pypa.io/en/stable/installation/
• Homebrew: https://brew.sh/

📈 Main Findings

• Election periods do not significantly alter general climate attitudes, but subtle shifts in WTP and policy support occur among independents.
• Support for carbon taxes is strongly predicted by previous policy support (e.g., emissions standards).
• Multicollinearity in interaction models affects precision, prompting dimensionality reduction through index construction.

🧠 Tools and Libraries

Tool/Library	Purpose
pandas	Data manipulation and panel structuring
numpy	Numerical operations and array handling
statsmodels	PVAR (Panel Vector Autoregression) estimation
tigramite	Time-lagged causal discovery using PCMCI+
matplotlib	Static plotting
seaborn	Statistical graphics for exploratory data analysis
plotly	Interactive network visualizations
networkx	Construction and layout of causal graphs
graphviz	Rendering directed acyclic graphs (DAGs)
skimpy	Quick summaries and data diagnostics

📜 License

GPL-3.0 license

🙏 Acknowledgments

• Supervisor: Kyuri Park
• Faculty of Science – University of Amsterdam
• The creators of open-source survey datasets and Python libraries

📬 Contact

Paraskevas K. Leivadaros
📧 paraskevasleivadaros@gmail.com
🌐 GitHub | LinkedIn

📚 Citation

If you use this repository, please cite the thesis:

Paraskevas K. Leivadaros, Elections and Climate Attitudes: How Do People’s Views on Climate Change and Policies Shift During an Election?, University of Amsterdam, 2025. PDF