AI Survey Design Workflow

Help me design a survey. Goal: [what decision will this data inform?]. Target audience: [who]. Sample size: [N or unknown]. Define: specific hypotheses to test, quantitative vs qualitative balance, survey length (target completion time: [X] minutes), distribution method (email/in-app/social), and incentive strategy. Recommend a sample size for statistical significance given expected response rate.

Write survey questions for: [research objective]. Write 15-20 questions including: 2-3 screening questions, 5-7 rating scales (NPS, Likert, satisfaction), 3-4 multiple choice, 2-3 open-ended, and 2 demographic. For each question: avoid leading language, provide balanced answer options, include "not applicable" where appropriate, and add skip logic rules. Order from easy to hard, personal questions last.

Review this survey for bias and quality: [paste questions]. Check for: leading questions, double-barreled questions, unclear terms or jargon, response bias (acquiescence, social desirability), order effects, missing answer options, survey fatigue (is it too long?), and mobile-friendliness. For each issue: explain the problem and provide the corrected version. Then create a pilot test plan for 5-10 respondents.

Analyze these survey results: [paste data or describe]. Respondents: [N]. Run: response rate analysis, quantitative analysis per question (means, distributions, cross-tabs), open-ended response coding (group into themes), statistical significance tests for key comparisons, and NPS calculation with segment breakdown. Create a findings summary with: top 5 insights, methodology notes, and recommended actions.