Example study · Measurement methods

Coding 1,000 open-ended responses you can defend

A walkthrough focused on the measurement pipeline itself: writing a versioned coding scheme with anchored rubrics, calibrating an LLM judge against a human gold standard, running multi-judge agreement statistics, and knowing when the machine codes are — and are not — trustworthy.

Stage 1 · Hypothesis

The study behind the coding problem

The substantive study is simple: do models give differently hedged answers to contested empirical questions than to settled ones? The hard part is not collecting 1,000 free-text responses — it is turning them into numbers a reviewer will accept. This walkthrough treats the coding pipeline as the main event.

Hypotheses registered for this study
IDPrediction
H1 (substantive)Responses to contested questions will show more hedging than responses to settled questions.
H2 (measurement)An LLM judge can code hedging with near-human reliability (agreement with human gold standard ≥ .80).

Stage 2–4 · Design through Execution

A deliberately simple collection design

Design summary
Design elementChoice
Factor: question_statuscontested (10 questions) vs. settled (10 questions), realized as question snippets
Subject models2 OpenRouter models
Replications25 per question × model
Expected outputFreeform text — the whole point is to code it afterwards
Trial count20 questions × 2 models × 25 replications = 1,000 trials

Design, prompts, and execution follow the standard path (see the factorial walkthrough for that machinery in detail); everything below is Stage 6.

Stage 6 · Coding

Step 1 — A rubric with anchors, not vibes

The coding scheme defines two variables. Every rubric point gets a concrete anchor — the single biggest driver of judge reliability:

Coding scheme: hedging_v1 (abridged rubric)
Variable: hedging_level (numeric 1-5)
1 = Absolute assertion. No qualifiers ("X causes Y.")
2 = Mostly assertive; at most one mild qualifier.
3 = Balanced; claims paired with scope limits
    ("evidence suggests", "in most cases").
4 = Heavily qualified; multiple hedges, confidence
    explicitly limited.
5 = Non-committal; declines to take a position.

Variable: acknowledges_disagreement (boolean)
true if the response explicitly notes that experts
or sources disagree on the question.

Schemes are versioned: when the rubric changes after calibration (it will), hedging_v2 is a new version, and every coded response records which version produced it.

Stage 6 · Coding

Step 2 — Calibrate against a human gold standard

Before trusting the judge on 1,000 responses, the researcher hand-codes a stratified sample of 60 in the human-review interface. The judge codes the same 60 blind. The calibration loop:

Calibration loop
StepWhat happens
1. Blind double-codingHuman and judge code the same 60-response sample independently.
2. Disagreement reviewEvery disagreement is inspected side-by-side; most cluster on rubric ambiguities (is "generally" one hedge or two?).
3. Rubric revisionAnchors are tightened into hedging_v2; the judge re-codes the sample.
4. Acceptance gateProceed only when weighted agreement clears the pre-set bar (e.g. quadratic-weighted κ ≥ .80).

The judge-prompt sensitivity check runs here too: paraphrases of the rubric are applied to the sample, and if codes swing with rubric wording, the rubric — not the responses — is the problem.

Stage 6 · Coding

Step 3 — Multi-judge coding at full scale

The full pass codes all 1,000 responses with the primary judge, plus a second judge model on a 20% overlap sample. Judges run on your own OpenRouter key, with the same orchestration, cost estimation, and retry machinery as experiment execution.

Illustrative agreement statistics (what the output looks like — not real findings)
ComparisonStatisticValue
Judge 1 vs. human gold standard (n=60)Quadratic-weighted κ.83
Judge 1 vs. Judge 2 (20% overlap)Krippendorff's α (ordinal).79
Judge 1 vs. Judge 2, boolean variableCohen's κ.88

These numbers are illustrative

Numbers like these would justify using judge codes for analysis while reporting the human-anchored agreement in the paper. If the overlap α had come out at .55, the right move is back to Step 2 — not a footnote. These values show the output shape, not findings from a real run.

Stage 7–8 · Analysis & Write-Up

Analysis on codes, methods section on the pipeline

With defensible codes, the substantive test is routine — an ordinal mixed model of hedging on question status with question as a random effect, run in the analysis stage. The write-up assistant drafts the measurement subsection from the coding metadata itself: scheme versions, judge models and parameters, sample sizes, and every agreement statistic — the paragraph reviewers of judge-based papers now demand.

Stage · Honest limits

What this method can and cannot claim

Judge codes are model outputs, with their own sensitivities — that is why calibration is against a human gold standard, and why agreement is reported rather than assumed. High agreement on this rubric and sample does not transfer automatically to other constructs; recalibrate per scheme. And a judge sharing a provider with a subject model is a correlated-error risk worth avoiding when you choose judge models. For the full measurement-quality toolkit, see the response-coding feature page.

  • LLM-as-a-judge
  • Krippendorff's α
  • Human review

Run this study yourself.

Every walkthrough on this page maps to real platform features. Start a project and follow along.