Spring survey analysis - #32
Merged
Merged
Conversation
survey_analysis_v2 is a scripted, rerunnable pipeline that turns the Spring 2026 FutureEval participant survey into a publishable per-question report plus parsing-decision and parsing-review docs, all generated from config so the docs stay in sync with the code. - Per-question distribution charts split by everyone / non-winners / winners / top-10, with an aggregated "Other (write-in)" bar for free-text answers - Correlations against each bot's average spot peer score (computed from the SPOT_PEER leaderboard), restricted to bots with >=100 scored questions - Frontier-model classification from a dated, reviewable model registry - Privacy-preserving score charts: aggregate bars only, min 5 respondents per bar; review doc + audit CSV flagged internal-only (name<->score linkage) - Robust Google-Forms parsing (multi-select vocab matching, write-in capture) with unit tests for parsing, joins, and frontier logic Also includes the previously-staged survey_analysis_v1 and a run_spring_simulations.py update. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Adds a self-contained pipeline that analyzes the Spring 2026 FutureEval bot-maker survey, correlating each structured survey answer with tournament performance (average spot peer score), and generates a report, charts, two audit docs, and a landing page. Also refreshes project dependencies to their latest compatible versions.
Scope: 31 files changed (~6.7k insertions; ~1.6k of that is
poetry.lock). New module ataib_analysis/survey_analysis_v2/with 11 test files.What it produces
Run with
poetry run python -m aib_analysis.survey_analysis_v2.run, thenbash aib_analysis/survey_analysis_v2/render_html.shfor HTML/PDF:spring_survey_analysis.md— the report: per-question answer distributions (split non-winners / winners / top-10), a flat evidence summary ordered by |r|, and one correlation chart per measured feature.parsing_decisions.md— every parsing rule, the model registry with release dates, vocabularies, and midpoint maps.parsing_review.md— the audit: all columns accounted for, the full survey→leaderboard join table, per-bot model classification, and every write-in that was set aside.index.html— landing page linking the three.