Panel - Diversity in Data Organizations: Building a Representative Workforce and Why it Matters
summary
Panel argues that representative teams and documented data practices stop harm and improve outcomes and gives concrete steps and examples. Public policing data like stop question and frisk shows 2% arrest rate expose one-sided collection and institutional incentives to miscount. Programs scale access such as Square Root Academy reaches 3,000 scholars/year with 97% scholars of color and firms publish headcount like Accenture 506000 employees 215000 women as accountability. Recommendations include equity impact reviews, community-involved analysis, data documentation, and bolting flawed surveys to focus groups.
AI-written summary · about
chapters (11)
- 0:00Start
- 6:20Vince Vu
- 8:17Nicholas Hastings
- 10:11Why It Matters
- 11:45New York City's Stop Question and Frisk Data Set
- 16:15How Does One Define a Representative Workforce and What Value Does that Represent Representative Workforce
- 37:39What Happens When Analytics Fail
- 40:39What Responsibility Do We each Have To To Ask those Questions about Our Data Models and Data Collection
- 1:14:12Unconscious Bias
- 1:22:47Scholars Playground
- 1:23:50A Toolkit for Centering Racial Equity within Data Integration
related talks
54mHow diversity drives excellence in our data driven tech world
54mPanel Discussion: Building Inclusive Data Communities
53mDiversity and Inclusion Best Practices: A panel discussion
01h 00mDX Design: Diversity Experience as an Accelerator
48mKeynote: Data Science Community and You
46mHow diversity is enriching the customer experience
44mThe Dark Side of D&I
46mBuilding Diverse Teams and Projects