The Child Labor Risk Index

How the score is built

Each indicator is min-max normalized to a 0–100 scale within this 6-country sample — this is a relative ranking among the six countries in the founding research, not a universal global score. The five sub-scores are averaged with equal weight (20% each), by design: with only six countries in the dataset, fitting statistically-derived weights would be overfitting dressed up as rigor. Equal weighting is the honest default until a larger sample justifies anything else.

Does the index actually predict anything?

Spearman rank correlation between the Risk Index and the real child labor rate: ρ ≈ 0.66. With only six countries, this is descriptive, not a statistically significant test — but a moderate positive correlation is exactly what you’d hope to see from a model that isn’t circular.

The more interesting finding is where the model is wrong: India and Bangladesh score as moderate-to-high risk on raw economic fundamentals alone, yet both have the lowestactual child labor rates in the study. That gap lines up with something the original research documents directly — both countries run large, targeted interventions (India’s Right to Education Act and Mid-Day Meal Scheme; Bangladesh’s primary education stipend program) that economic indicators alone don’t capture. Nigeria and Ethiopia, without comparable programs at scale, land close to what their fundamentals alone would predict.