Skip to content
ELAGrades K-2

EL Education K-5 Language Arts (K-2)

EL Education

Judge anthropic/claude-opus-4-7 · Signal Studio judge (claude-opus-4-7) · 2026-05-28

92

% indicator exact match

Gateways 3/3 Meets

Agreement vs gold

Indicator level — gold and the judge as two raters over the rubric criteria.

Exact match

91.7%

36 indicators vs gold

Weighted κ

0.91

ordinal-weighted agreement

MAE

0.11

lower is better

Signed bias

+0.11

judge over-scores

Attribution

Where the judge diverges, and whether it's signal: per-gateway bias localizes the gap, the confusion matrix shows over- vs under-rating, and the noise floor is the judge's self-agreement across passes (a gap below it isn't trustworthy).

gateway-onen=1492.9%bias +0.07
gateway-threen=1492.9%bias +0.07
gateway-twon=887.5%bias +0.25

Indicator confusion · rows = gold, cols = judge

MeetsPartialDNM
Meets3300
Partial300
DNM000

Diagonal = agreement. Amber = judge rated higher than gold (over-rating).

Gateway rollup

Indicator scores rolled up to EdReports gateway ratings (sequential gating + no-0s cap), gold vs judge.

G1 · Text Quality

36/36 pts

GoldMeets
JudgeMeets
agree

G2 · Building Knowledge

32/32 pts

GoldMeets
JudgeMeets
agree

G3 · Usability

34/34 pts

GoldMeets
JudgeMeets
agree

Gateway-level agreement: exact 100% · κ 1.00 (3 gateways)

Divergences (3)

Indicators where the judge's score differed from gold. Amber = judge under-scored, blue = over-scored.

1e
12
gold Partially Meets Expectations · judge 2 ptsinsufficient evidence
3q
12
gold Partially Meets Expectations · judge 2 pts
2d
24
gold Partially Meets Expectations · judge 4 pts
Live read from the Signal Studio spine (core-data).