Finding common development paths in voluntary national reviews reporting on sustainable development goals using...

Area of Science:

• Social Sciences
• Data Science
• International Relations

Background:

• Voluntary National Reviews (VNRs) are key for countries to report progress on United Nations Sustainable Development Goals (SDGs).
• Understanding the sentiment within VNR narratives is crucial for accurately assessing global SDG achievement.
• Existing assessments often lack a nuanced understanding of the qualitative data presented in VNRs.

Purpose of the Study:

• To analyze the sentiment expressed in VNRs across all 17 SDGs.
• To compare country-reported sentiment with official SDG indicators.
• To identify patterns in sentiment reporting across different country groups and over time.

Main Methods:

• Aspect-Based Sentiment Analysis (ABSA) applied to 232 VNRs from 166 countries.
• Sentiment scores for each SDG were extracted and compared to official UN SDG scores.
• t-distributed stochastic neighbor embedding (t-SNE) used to visualize sentiment patterns.

Main Results:

• Positive correlation found between sentiment and SDG scores for SDG 2 (Zero Hunger) and SDG 11 (Sustainable Cities).
• Negative correlation observed for SDG 5 (Gender Equality); no significant correlation for most SDGs.
• Developed countries show consistent sentiment; Emerging Markets, Frontier Markets, and Least Developed Countries (LDCs) exhibit greater sentiment variability.

Conclusions:

• VNR sentiment does not always align with official SDG progress, necessitating contextual interpretation.
• Sentiment reporting varies significantly across country economic classifications, especially LDCs.
• Future SDG assessments should consider the nuanced sentiment expressed in VNRs alongside quantitative data.