QueScript v1.0 Release Notes

QueScript v1.0 Release Notes

Release date: 2026-03-06

Highlights

QueScript v1.0 delivers a local-first survey workflow covering generation, simulation, measurement analysis, and auditability in one GUI:

• Text-to-HTML questionnaire generation
• Smart batch simulation with configurable sample size and response tendencies
• Branch-aware skip logic and per-sample path auditing
• Reliability/validity/discrimination analysis with EFA outputs
• Reproducibility artifacts for rerun and review

New in v1.0

Survey generation

• Added questionnaire text parsing to produce browser-ready HTML surveys
• Improved support for common question types (single choice, multiple choice, scale, matrix, text)

Simulation engine

• Added GUI-driven simulation flow (no command line required for routine use)
• Added response tendency modes for non-scale selection behavior:
  • random
  • positive
  • negative
  • central
• Added configurable latent dimension settings for scale-data generation scenarios
• Added strict skip-path execution support based on question visibility/jump rules

Measurement analysis

• Added built-in analysis report after simulation completion
• Added reliability analysis:
  • Cronbach's Alpha
  • Item diagnostics (CITC, Alpha-if-deleted)
• Added validity and structure checks:
  • KMO + Bartlett
  • EFA output (factor suggestions, variance contribution, loadings)
• Added item discrimination (critical-ratio style checks)

Auditability and reproducibility

• Added run metadata and reproducibility fields (run_id, seed)
• Added audit artifacts:
  • config.json
  • path_log.csv
  • analysis_meta.json
• Added report-level signatures for reproducibility trace

Packaging and delivery

• Added offline-oriented Windows packaging path
• Added local browser-runtime packaging support for Playwright-based execution layouts

UX and GUI updates

• Added modernized GUI layout with simulation controls and status/progress display
• Added integrated run log panel and report display workflow
• Added clearer analysis sections and structured report output

Compatibility

• Platform focus: Windows desktop usage
• Python runtime: project uses virtual-environment-based dependency management

Output files (typical)

After a simulation/analysis run, output files are generated in the target survey directory:

• survey_data_collected.csv
• config.json
• path_log.csv
• analysis_meta.json

Known limitations

• Branch-heavy questionnaires can reduce the number of globally comparable items in full-sample analysis.
• Structural metrics are sensitive to sample size and item coverage thresholds.
• Some advanced fit indicators may become unstable under small n/p conditions.

Upgrade and usage notes

• Recommended baseline for more stable structural analysis: larger sample sizes (commonly n >= 100)
• For strong branch logic, prefer branch-aware interpretation in addition to full-sample summaries
• Keep each latent dimension with enough comparable items for better model stability

Repository

• GitHub: https://github.com/SakuraMathcraft/QueScript