IqScreenr

About

IQScreenr is a privacy-first browser extension that analyzes Twitter/X tweets in real-time and displays estimated verbal IQ scores using research-validated linguistic analysis.

✨ KEY FEATURES

• Real-Time Analysis: Automatically analyzes tweets as you scroll through your Twitter/X feed
• IQ Scoring: Estimates verbal IQ from 60-145+ based on comprehensive linguistic analysis
• Confidence-Based Colors: Visual gradient indicators from red (low confidence) to green (high confidence)
• Privacy-First: All processing happens client-side in your browser - no data sent anywhere
• Interactive Game Mode: IqGuessr - Guess the IQ of tweets before revealing scores
• Smart Filtering: IqFiltr - Filter tweets by IQ range to focus on specific content quality
• Developer Mode: Press CTRL+I+Q for detailed debugging and analysis breakdowns
• Hover Details: Comprehensive linguistic breakdown on badge hover

🔬 RESEARCH-VALIDATED METHODOLOGY

IQScreenr uses a research-validated estimator that analyzes four key dimensions of written language:

1. Vocabulary Sophistication (35-45% weight)
   - Uses Age of Acquisition (AoA) ratings from a 31,766-word dictionary
   - Higher AoA = words learned later in life = more sophisticated vocabulary
   - Based on research from Kuperman et al. (2012) and Nnamoko et al. (2024)

2. Lexical Diversity (25% weight)
   - Analyzes Type-Token Ratio (TTR), Mean Segmental TTR (MSTTR), MTLD, and Yule's K
   - Higher diversity = more varied vocabulary = higher IQ indicator
   - Research-validated correlation: r = 0.40

3. Sentence Complexity (15-20% weight)
   - Analyzes average words per sentence with optimality factors
   - Moderate complexity is optimal (extremely long sentences aren't always better)
   - Twitter-adjusted baseline accounts for character limits

4. Grammatical Precision (15-20% weight)
   - Approximates dependency depth from punctuation patterns, clauses, and sentence structure
   - Analyzes punctuation entropy, subordinate clauses, and connective density
   - Research-validated correlation: r = 0.45

🎨 UNDERSTANDING THE SCORES

Badge colors reflect confidence levels (0-100%), not IQ scores:
• Red (0-30%): Low confidence - estimate may be inaccurate
• Yellow (30-50%): Moderate confidence
• Green (50-100%): High confidence - reliable estimate

IQ scores range from approximately 60-145+:
• 60-85: Simple, straightforward language with basic vocabulary
• 85-105: Average complexity, typical of general population
• 105-125: Above average, sophisticated vocabulary and structure
• 125-145: High complexity, advanced vocabulary and complex sentence structures
• 145+: Exceptional complexity, rare sophisticated writing patterns

⚙️ CUSTOMIZATION OPTIONS

• Toggle Badges: Show/hide IQ scores on demand
• Adjustable Scale: Customize minimum and maximum IQ display range (default: 60-145)
• Confidence Display: Badge colors always reflect confidence levels
• Linguistic Breakdown: Enable/disable detailed analysis tooltips
• Cache Management: Clear cached scores for recalculation

🔒 PRIVACY & SECURITY

IQScreenr is privacy-first:
• ✅ 100% Client-Side Processing: All analysis happens in your browser
• ✅ No Data Collection: No tweets, scores, or personal data are sent anywhere
• ✅ No Tracking: No analytics, no telemetry, no external requests
• ✅ Local Storage Only: Settings and cache stored locally in your browser
• ✅ Open Source: Full code available for inspection

Your data stays on your device. Always.

⚠️ IMPORTANT DISCLAIMER

IQScreenr provides estimates based on linguistic analysis, not actual IQ test results. These scores:
• Are based on written language patterns only
• May not reflect actual intelligence accurately
• Should not be used for important decisions
• Are for entertainment and educational purposes

Use responsibly and remember: intelligence is multifaceted and cannot be fully captured by text analysis alone.

📚 RESEARCH SOURCES

• Kuperman, V., Stadthagen-Gonzalez, H., & Brysbaert, M. (2012). Age-of-acquisition ratings for 30,000 English words. Behavior Research Methods, 44(4), 978-990.
• Nnamoko et al. (2024). Age of Acquisition proxies for vocabulary sophistication estimation.
• Wolfram (2025). Linguistic complexity and intelligence correlation.
• Abramov (2018). Type-Token Ratio and lexical diversity metrics.
• Hendrix, A., & Yampolskiy, R. (2017). Automated IQ Estimation from Writing Samples.

🌐 SUPPORT

For questions, issues, or feature requests, visit: https://iqscreenr.win