Why DC’s Low Graduation Rates?
Read the original on Slate Star Codex →
Summary
A data-driven investigation of why DC's projected graduation rate cratered to 42% (after fraud was exposed). Scott shows DC's test scores are low-tier but not 42%-low, its white students excel and its black students do poorly-but-not-uniquely, so 'terrible schools' (hyp 1) and 'uniquely disadvantaged population' (hyp 2) don't fully explain it — leaving hyp 3: DC over-corrected to stricter standards than anyone else, which implies other cities (Detroit/LA/Chicago/Baltimore) sustain their rates via 'fraud and really low standards.' The signaling-collateral-damage point: if education is signaling, the worst thing is to speak a different signaling-language, unfairly failing DC kids who'd graduate anywhere else. (His own EDIT/companion later credits absences as a bigger factor.) Parent post of the Highlights companion (SSC-657, 62).
Why this score
Quality 70 · Strong. A sharp data-driven investigation (the graduation-vs-test-score outlier -> widespread low-standards-elsewhere inference; the signaling-collateral-damage point), but topical and partly superseded by his own absences edit. Strong; scored above its Highlights companion (62).
Claude’s paradigm shift 42 · Moderate. Moderate. A sharp data-driven inference on a specific topical case.
Real-world impact 2 · Minor. Minor (2) — a topical education-policy investigation with within-sphere reach; no concrete real-world change.