Repository navigation
Four new case studies: TaRL, Seguro Popular, Tusome, in-line chlorination - #16
Merged
Merged
Conversation
204 to 208. Teaching at the Right Level (India), Seguro Popular (Mexico), Tusome (Kenya) and in-line chlorination (global). The library's own standard is that a case carries four or five real citations, so the citations were the constraint rather than the prose. Each candidate reference was resolved against Crossref before a word of the study was written, and the record that came back was read: title, year, journal, authors. Seventeen references verified across the four studies. Three further candidates were dropped for failing that test rather than rewritten around it: Colombia's Escuela Nueva, Rwanda's Umuganda, and the Philippine sin tax reform, where only two of the four references resolved. A case with two verifiable sources is not a case this library can hold. The chlorine study is the one worth reading twice. Kremer et al. (2011) established take-up; the 2023 meta-analysis (NBER w30835) pooled child mortality across 15 randomised trials and found roughly a quarter reduction, a result no single trial in the set was powered to detect. Clasen's 2006 Cochrane review and Wolf 2022 are both in the study because they are the reason the earlier literature read as null. Derived files regenerated by `npm run build`; `npm run validate` passes at 208. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MrvR2NXsJFVRCeJZFCPuNL
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The library goes from 204 to 208.
india-teaching-at-the-right-levelmexico-seguro-popularkenya-tusomeglobal-chlorine-dispensersHow the citations were handled
Every study in this library carries four or five references, and that is the
binding constraint on adding one, not the prose. So the references came first:
each candidate was resolved against Crossref and the returned record read back
(title, year, journal, authors) before the study was written. Seventeen
references verified across the four.
Three other candidates were dropped rather than written around:
A case study with two verifiable sources is not one this library can hold, and
padding it with a plausible-looking citation would be worse than not shipping it.
One finding worth stating
The chlorination study is about a literature that read as null and was not.
Kremer et al. (2011, QJE) established that free point-of-collection
dispensers take up where a purchase decision does not. The 2023 meta-analysis
(NBER w30835) pooled child mortality across 15 randomised trials and found
roughly a quarter reduction, which no single trial in the set was powered to
detect on its own. Clasen's 2006 Cochrane review and Wolf et al. (2022,
Lancet) are in the study for that reason: they are the record of the earlier
reading.
Derived files
npm run buildregenerateddata/master-list.json,data/search-index.json,data/stats.jsonandsitemap.xml.npm run validatereports 208 studiesprocessed cleanly.
🤖 Generated with Claude Code
https://claude.ai/code/session_01MrvR2NXsJFVRCeJZFCPuNL
Generated by Claude Code