Data Science Education in 2022
summary
Conference data shows 48% self-taught learners and a workshop program that runs 19 sessions with ~100 certified instructors teaching about 1,300 students this week. Claims include a data explosion where 90% of data is created in the last two years and computing growth (~4.5× in five years) far lags data growth (~46×), plus irreproducibility rates like 6/53 and 36/100 and an estimated US biomedical cost of $28 billion. Recommendations: teach statistics and sampling on messy real data, publish reproducible R Markdown workflows, curate shared datasets, and train domain experts to tell authoritative data stories.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:28Studio's mission and goals
- 2:51Current initiatives and challenges
- 3:53Self-taught growth
- 4:37Challenge: data explosion
- 6:24Challenge: irreproducible results
- 10:02Challenge: fake data and misinformation
- 13:28Solutions for education
- 15:31Democratizing via storytelling
- 17:39Q&A
related talks
02mBuilding data science capacity in your org
18mHow Data Science Can Unlock Teaching & Learning at Scale
36mCore of Data Science and Open Source Resources
48mHuman data interactions: experiments in how humans analyze data
42mPanel Discussion: Improving Industry-Academia Collaboration To Close The Data Science Talent Gap
04h 02mHands-on Session on Data Science for Beginners
24mData Centricity - Rethinking Introductory Computing To Support Data Science
12mOpening Address