Data science project

Education

A Journey Through How the Share of Women in National Parliaments Impacts Environmental Policy Enforcement

The future of our Earth might depend on who sits in the seats of power.

Did you know that as of 2022, men occupied 74% of seats in parliaments worldwide (The World Bank, 2023)? At the COP28 Summit, this disparity was evident with a majority of male heads of state and government in attendance. But what if the key to unlocking more effective climate change legislation lies in changing this ratio? What if increasing female representation in parliaments could be a pivotal step towards enforcing robust and effective climate change laws to limit global temperature rise to 1.5ºC and prevent irreversible damage to our planet?

If you want to learn more, I invite you to read the full article here. This project was completed as part of the Statistics and Data Science course. We conducted all data cleaning, visualization, and analysis, as well as statistical modeling using Python. The entire process, including both code and text, was documented on Google Colab.

The most significant takeaway from this course is the critical distinction between correlation and causation. While correlation can indicate a relationship between variables, causation is essential for decision-making but requires extensive research to establish.

Process

Data Finding

Data cleaning

Data visualisation

Statistic analysis

Lessons Learned 

Through this project, I gained valuable insights and practical skills in several key areas:

Data Sourcing and DAG:

    • We learned how to identify and source high-quality data.
    • Understanding the role of confounders versus covariates was crucial, as was identifying meaningful heterogeneous variables.

Data Cleaning and Completion:

    • Proper data cleaning and completion were essential to ensure our datasets were comprehensive and ready for analysis.

Data Exploration and Visualization:

    • We explored our data extensively, using visualization techniques to detect outliers and gain a deep understanding of our data.
    • Visualization proved invaluable in understanding data distributions, trends over time, and geographic patterns.

Statistical Analysis:

    • Conducting statistical analyses helped us uncover relationships between variables.
    • It was important to use appropriate tools based on the nature of the data to ensure robust causal analysis.

Program used

Python

Google Colab

Medium