UEQ Data Analysis Tool

Purpose & User Intent: The tool solves the problem of transforming raw UEQ questionnaire responses into standardized, interpretable metrics for UX researchers and product teams. It supports users after data collection to quantify user perceptions across six UEQ scales and to compare cohorts.
Core Logic & Features: Required features include dataset import (CSV/TSV with 1–5 responses per item and respondent metadata) and automated calculation of the six UEQ scales. Optional features include reliability checks (Cronbach's alpha), bootstrap confidence intervals, and cohort comparisons. The tool also offers per-item statistics and export of analysis results. It enforces 1–5 Likert scale validity and handles missing responses via pairwise exclusion by default.
Inputs & Outputs: Inputs include dataset, chosen item mapping, and optional metadata; outputs are per-scale scores, item-level statistics, reliability metrics, and exportable reports in CSV/JSON. Formatting rules standardize scores to a 0–1 or 1–5 scale alignment, depending on settings.
Algorithms & Calculations: It computes scale scores as the mean of constituent items, applies standard UEQ scoring conventions, and derives overall UX quality indexes. Optional bootstrapping yields 95% confidence intervals. It also provides cohort-based comparisons using effect sizes (Cohen's d).
Error & Edge Cases: Missing data is handled with predefined rules; invalid item mappings, unsupported formats, or incongruent respondent counts raise precise errors. Large datasets trigger performance-optimized batch processing. Industry/Region & Localization: Uses standard UEQ scoring (1–5) and supports GDPR-compliant data handling; locale formatting for numbers is configurable. Assumptions & Exclusions: Assumes standard UEQ item order and Likert scale; qualitative analysis and benchmark databases are outside scope.