Classify, extract, and summarize open-ended text (survey responses, social media, academic papers, political text, web content) at scale with LLMs. Python, R, and Stata.
-
Updated
Oct 20, 2026 - HTML
Classify, extract, and summarize open-ended text (survey responses, social media, academic papers, political text, web content) at scale with LLMs. Python, R, and Stata.
Effortlessly conduct chatbot-based interviews with customizable protocols and randomized condition assignments, tailored for social science research.
An Agent Skill for reviewing translated scales, adapted questionnaires, paired surveys, and pre-launch survey quality. For OB, management, marketing, psychology, and social science researchers.
Slides for my talk on surveydown at the 2025 posit::conf
Run synthetic focus groups and user research panels using AI personas. CLI tool, Python library, any LLM.
Extreme Recall: Which Politicians Come to Mind?
Web application for mobile phones to crowdsource data collection from geographically distributed locations
Pre-registered evaluation: a small amortized person-embedding model beats LLM personas at predicting individual survey responses; LLMs win at zero-shot questions and social-influence propagation. Paper included.
A privacy-aware SQLite survey explorer demonstrating branching logic, reproducible cohorts, and honest synthetic-data provenance.
Sampo Vesanen's Master's thesis for the University of Helsinki, Department of Geosciences and Geography.
Survey methodology project evaluating ASHA response behavior, postcard reminder effects, and postal mail vs. web survey mode differences using R, bootstrap inference, and logistic regression.
Audit of LLM synthetic survey respondents against 31,856 CGSS respondents; v2 adds fp32 option probabilities and a preregistered factorial experiment
Synthetic Population Intelligence - statistically-grounded synthetic personas anchored in public census/Eurostat data, with conformal-bounded insights and Hybrid PPI++. Public showcase; source on request.
🏆 Winner, Accenture GenAI Mavericks 2026. Synthetic population of Spain (25,000 profiles) that predicts survey answers by region, sex and age, with a validated confidence traffic light. CIS-trained models + local LLM.
Two PLS-SEM studies on Coca-Cola purchase intention using primary survey data in SmartPLS, covering measurement and structural models, bootstrapping, and mediation analysis.
General attitudinal post hoc segmentation analysis using survey research data to understand the market for social media/entertainment apps
Survey-grounded evaluation harness: does an LLM in government track the public's policy preferences, and does the evidence channel that steers it also break its rights floors? BSA microdata, five-rung steering ladder, judge-free scoring.
A Qualtrics toolkit for research staff working with AI-assisted development
AutoXR. End to end product design case study for an XR vehicle customization experience. Covers user research, conjoint driven tradeoffs, pricing, and launch strategy.
(SPSS) دراسة كمية عن تأثير الأفلام والمسلسلات الأجنبية على القيم الاجتماعية | Quantitative SPSS field study
To associate your repository with the survey-research topic, visit your repo's landing page and select "manage topics."