SDG Internet & Education Dashboard

Overview

This project investigates how internet access impacts educational completion rates across South American countries between 2000 and 2023. Built in the first year of the Applied Data Science & AI programme at BUAS, it was developed for the BUas SDG Hub and aligns with two UN Sustainable Development Goals — SDG 4 (Quality Education) and SDG 9 (Industry, Innovation, and Infrastructure).

The final deliverable is a Power BI dashboard that makes the findings accessible and actionable for policymakers, educators, and NGOs.

Data & Methodology

Two datasets were merged and cleaned using Python:

  • World Bank — percentage of individuals using the internet per country per year
  • UN SDG dataset — educational completion rates by country, sex, location, and education level

Data preparation involved removing incomplete rows, filtering to 10 South American countries with sufficient data, standardising percentage formats, and removing extreme outliers. The World Bank dataset was reshaped from wide to long format using melt() before merging on country and year.

Analysis

Three modelling approaches were used:

Pearson Correlation Coefficient — overall R score of 0.58, indicating a moderate to strong positive relationship between internet access and educational completion rates. Country-level correlations were also exposed individually in the dashboard.

Linear Regression — a regression line was fitted to predict completion rates from internet access levels, validated with R-squared values.

Year-over-Year Growth — annual changes in both metrics were tracked to identify periods of significant improvement or decline.

Dashboard

The Power BI dashboard includes scatter plots with regression lines, heat maps, line charts, KPI cards, and geographical maps highlighting regional variation. Users can explore country-specific trends and correlations interactively.

Published on Power BI for the BUas SDG Hub and also released as a .pbix file on GitHub for public access.

Key Findings

  • Higher internet access is positively correlated with higher educational completion rates across South America
  • The relationship varies significantly by country — Suriname is a notable outlier
  • Both metrics showed gradual improvement over the 23-year period
  • Internet access alone explains approximately 34% of the variation in completion rates (R² = 0.34), suggesting other socioeconomic factors also play a significant role

Tech Stack

  • Python (pandas, data cleaning and merging)
  • Power BI (dashboard and visualisations)
  • Excel (year-over-year growth calculations)
  • World Bank & UN SDG open datasets