UNESCO GRALE Analysis

Exploratory analysis of UNESCO's Global Report on Adult Learning and Education (GRALE) survey data, examining patterns in adult learning and education (ALE) systems across countries.

Project Overview

This project explores the GRALE dataset to understand global patterns in adult education systems. GRALE is UNESCO's primary mechanism for monitoring adult learning and education worldwide, with countries submitting detailed responses every 3-4 years.

Key Questions I'm Exploring

• How does ALE participation vary across regions and demographic groups?
• Where are the data quality gaps in global ALE monitoring?
• What patterns emerge in how countries report on their ALE systems?

Data Sources

• GRALE 5 Open Data (grale-5_opendata.xlsx): Country-level responses to UNESCO's ALE monitoring survey
• GRALE 5 Codebook (grale-5_codebook.xlsx): Variable definitions and coding schemes
• GRALE 4 Data (also included for longitudinal comparisons)

Dataset Structure

The GRALE 5 dataset includes 197 countries and covers:

• Policy & Governance (v1xx-v2xx variables): Legal frameworks, coordination mechanisms, stakeholder involvement
• Financing (v5xx variables): Budget allocations, spending levels, funding sources
• Participation (v6xx variables): Enrollment rates, demographic breakdowns, barriers to access
• Quality & Outcomes (v7xx-v8xx variables): Quality assurance, learning outcomes, program evaluation

Current Analysis

1. ALE Financing Exploration (In Progress)

Examining variable v520 (ALE spending as % of public education budget):

• 146/197 countries (74%) provided responses
• 40 countries (27%) reported "Do Not Know" - unable to quantify ALE spending
• Spending categories: <0.5%, 0.5-1%, 1-2%, 2-4%, 4%+
• DNK responses exported for manual coding to extract embedded information

Files:

• grale_analysis.ipynb: Main analysis notebook
• dnk_responses_for_coding.xlsx: DNK responses being manually reviewed

Upcoming Explorations

• Regional patterns in the data
• Participation and equity patterns
• Data quality assessment across all variables
• Longitudinal comparisons with GRALE 4

Methodology

This analysis uses Python (pandas, matplotlib, seaborn) for:

• Data cleaning and validation
• Exploratory data analysis
• Visualization of patterns
• Manual coding where automated analysis is insufficient

About GRALE

The Global Report on Adult Learning and Education (GRALE) is UNESCO's flagship report monitoring the state of adult learning and education worldwide. Published every 3-4 years since 2009, GRALE synthesizes country survey data to track progress on international commitments to adult education, including the Sustainable Development Goals (SDG 4).

Repository Structure

├── README.md
├── grale_analysis.ipynb          # Main analysis notebook
├── grale-5_opendata.xlsx         # GRALE 5 dataset
├── grale-5_codebook.xlsx         # Variable documentation
├── grale-4_opendata.xlsx         # GRALE 4 for comparison
├── grale-4_codebook.xlsx
└── dnk_responses_for_coding.xlsx # Manual coding work

Getting Started

import pandas as pd

# Load GRALE 5 data
df = pd.read_excel('grale-5_opendata.xlsx', 
                   sheet_name='GRALE 5_Dataset', 
                   skiprows=10)
df = df[df['iso3'].notna() & (df['iso3'] != 'Total')]

# Load codebook
codebook = pd.read_excel('grale-5_codebook.xlsx', 
                         sheet_name='Codebook', 
                         skiprows=8)

License

Data sourced from UNESCO Institute for Lifelong Learning (UIL). Analysis and code in this repository are for educational and research purposes.