Two Charts, Same Study
Designing for Different Readers
May 2, 2026 • 5 min read
A Center for Global Development working paper modelled what would happen to learning outcomes in lower and middle income countries if blood lead levels dropped to US rates. The original chart ran inside the paper. The Economist later redrew it for a magazine piece, and the two versions barely look related.
One of the paper's authors, Justin Sandefur, posted them side by side: "we spent an embarrassingly long time formatting this simple graph. Always interesting to see the choices good data viz people make to improve your work." You can read the full thread here.
What I find worth chewing on is the trade itself. Each chart picked a side and committed to it without trying to split the difference. One is a discovery chart. The other is a direction chart.
The discovery chart
![[object Object]](/_next/image?url=%2Fblogs%2Flead-learning-charts%2Facademic.png&w=3840&q=75&dpl=dpl_DitZx4n7qEGTRisxsSc9xQgNrvwu)
This is the version in the working paper. One dot per country for the current score, with a small "+23" next to it for the simulated gain if blood lead dropped to US levels. Russia 580 → +23. Serbia 521 → +30. Every number on the chart is exact, and the focus is on each country one at a time.
That exactness is the whole point. This chart lives inside a working paper, where the reader has the document open and is looking up specific countries to pull a number, cross-reference a table from earlier in the paper, or cite a value in their own work. The job of the chart is to make those lookups easy. The title just says what the chart shows: "Figure 3. The effect of eliminating blood lead levels on national learning outcomes." It's a label so a reader scanning the paper can find this specific figure. The note underneath gives the methodology straight, which is what an academic reader needs in order to inspect, replicate, or build on the result.
The direction chart
![[object Object]](/_next/image?url=%2Fblogs%2Flead-learning-charts%2Feconomist.png&w=3840&q=75&dpl=dpl_DitZx4n7qEGTRisxsSc9xQgNrvwu)
The Economist is answering a different question. Not what is Serbia's exact predicted score, but does this intervention work, and how big is the effect across the board. The job of the chart is to show that every dumbbell points the same way. Across dozens of countries, reducing lead moves every one of them in the same direction, all positive. Every other choice on the chart is in service of delivering that overall effect at a glance.
The title does most of the work. "Take a weight off their minds" is a power title, pitching the takeaway before you've read the chart. That split, an emotional headline over a dry subtitle, frames the overall effect of the intervention before the reader even looks at the dots.
The teal palette reinforces the pattern. Both dots are teal in two different shades, the lighter for each country's current score and the darker for the predicted score after lead is reduced. Same encoding on every row, so the reader stops reading individual countries and starts seeing the collective shift.
Why both are right
Discovery and direction aren't a hierarchy, they're answers to different questions. A peer reviewer asks what exactly is the predicted gain for Serbia. A magazine reader asks does this intervention actually move the needle, and in which direction. The same chart cannot answer both well, and trying to is usually how you end up with something nobody uses.
Ask whether your audience came to discover or to see a direction, and design for that.
