Academic Gender Search Analysis

📈 Gender Distribution by Project Count

Analysis of 2,675 Chief Investigators with 3+ Discovery Projects

Key Insights:

• Gender Gap Widens: Female representation decreases from 23.6% (3 projects) to 15.1% (6 projects)
• High-Impact Researchers: Among researchers with 6+ projects, males represent 76.9%+ of the cohort
• Leadership Pipeline: The data suggests a "leaky pipeline" where female participation diminishes at higher productivity levels

Note: Analysis excludes researchers with 1-2 projects to focus on established researchers. Data represents Chief Investigators in the Australian Discovery Projects system.

📊 Final Dataset Statistics

2,679 Chief Investigators with 3+ Discovery Projects analyzed using our comprehensive three-tier methodology.

• Male: 1,906 (71.1%)
• Female: 591 (22.1%)
• Unknown: 182 (6.8%)

Analysis Method Breakdown:

• Web Search (Tier 1): 2,004 researchers (74.8%)
• Name-Based AI (Tier 2): 493 researchers (18.4%)
• Manual Review (Tier 3): Available for remaining 182 researchers

Total Sources Found: 10,206 web sources (avg 3.8 per researcher in Tier 1)

🔬 Three-Tier Methodology

Selection Criteria: Analysis focuses on Chief Investigators with 3+ Discovery Projects, representing the most active researchers in the Australian academic system.

Tier 1 - Web Search Analysis (74.8%): OpenAI's search-enabled models (gpt-4o-mini-search-preview) perform web searches to find verified information about researchers, analyzing academic profiles, publications, and institutional pages for gender indicators.

Tier 2 - Name-Based AI Analysis (18.4%): For researchers where web search found no clear evidence, GPT-4o-mini analyzes name patterns and linguistic origins to make educated predictions. All Tier 2 results are clearly marked as speculative.

Tier 3 - Manual Review (6.8% remaining): Researchers can submit corrections through our GitHub Issues system for any misclassifications or to update remaining unknown cases.

Research Areas: Extracted from Tier 1 web search results including academic profiles, publications, and institutional pages.

Total Cost: AUD $72.11 for Tier 1 + ~$14.50 for Tier 2 = ~$86.61 total for comprehensive analysis.

⚠️ Important Disclaimer & Confidence Levels

We sincerely apologize for any gender misidentification. Gender identity is personal and complex, and our automated analysis may not always be accurate. Our methodology includes different confidence levels:

• High Confidence (Tier 1): Based on verified web information with clear evidence
• Medium-Low Confidence (Tier 2): Name-based predictions clearly marked as speculative
• Unknown (Tier 3): Insufficient information for any reliable prediction

Limitations:

• Gender cannot always be determined from names or public information
• Some individuals may use different pronouns or identify differently than our analysis suggests
• Non-binary and other gender identities may not be properly represented in our current classification system
• Tier 2 predictions are explicitly speculative and marked as such in the data

📝 Request Corrections: If you identify any errors in gender classification or research information, please:

• Email: tim.chen@adelaide.edu.au
• GitHub Issues: Submit an issue on GitHub

We welcome corrections from researchers themselves or anyone who notices inaccuracies in our dataset.

🛠️ Technical Details

Built with: Pure HTML, CSS, and JavaScript

Data format: JSON with comprehensive researcher profiles

AI Model: OpenAI GPT-4o-mini-search-preview with web search capabilities

Features: Client-side filtering, responsive design, interactive components

Last Updated: August 2024

🤝 Contributing

This is an open-source project aimed at promoting transparency in academic gender representation. We welcome contributions, corrections, and suggestions.

How to help:

• Report incorrect gender classifications
• Suggest improvements to methodology
• Contribute to data visualization features
• Help expand the dataset