Data Management for Citizen Science
Download as PPT, PDF2 likes1,247 views
The document discusses challenges and opportunities for data management in citizen science projects. It identifies developing data management plans, establishing data policies, developing supporting cyberinfrastructure or technology platforms, and ensuring data quality as key issues. A survey of citizen science projects found the greatest dissatisfaction with processes for sharing data and presenting results, but that data management planning was better than average. Top priorities for improvement included tools for analyzing, visualizing, documenting and describing data, as well as training. The presentation calls on USGS to lead by example in promoting data sharing, developing clear and reusable policies and platforms, and demonstrating best practices for data quality.
More Related Content
Similar to Data Management for Citizen Science (20)
Data Management for Citizen Science
- 1. Data Management for Citizen Science Challenges & Opportunities for USGS Leadership Andrea Wiggins Postdoctoral Fellow DataONE & Cornell Lab of Ornithology 12 September, 2012 USGS CDI Citizen Science workshop
- 2. DataONE PPSR Working Group Purpose: • Improve quality, quantity, and accessibility of PPSR data • Advance integration of PPSR data in conventional science Products: • Data Management Guide for PPSR - coming soon! • Articles in August FREE special issue • Data quality & validation paper 2
- 3. How long will it What is a data take to get management enough data? plan? Plan Analyze Collect How can I assure quality of volunteers’ What tools data? do I use? Integrate Assure What data about volunteers should I Who can help keep or share? me? Discover Describe Preserve Should I share What if the data are raw data with used for commercial known errors? profit?
- 4. How long will it What is a data take to get management enough data? plan? Plan Analyze Collect How can I assure quality of What tools volunteers’ data? do I use? Integrate Assure What data about volunteers should Who can help I keep or share? me? Discover Describe Preserve Should I share What if the data are raw data with used for commercial known errors? profit?
- 5. Citizen science data challenges Data policies Cyberinfrastructure Data quality 5
- 6. Policy? What policy? Data policies = boring http://www.flickr.com/photos/escapist/107455718/ 6
- 7. Policy? What policy? Data policies = boring Data policies = hard • Ownership, sharing, use, access, challenge, etc. • Lots of decisions, vague consequences 7
- 8. Policy? What policy? Data policies = boring Data policies = hard • Ownership, sharing, use, access, challenge, etc. • Lots of decisions, vague consequences Need examples of carefully crafted policies • Story of the data + policy that resulted • USGS is way ahead of the game! 8
- 9. Cyberinfrastructure Technology is a major pain point 9
- 10. Cyberinfrastructure Technology is a major pain point Platforms needed • Transcription, observation, processing • Ongoing support & development required 10
- 11. Cyberinfrastructure Technology is a major pain point Platforms needed • Transcription, observation, processing • Ongoing support & development required Who is going to pay? • <insert sound of crickets here> http://www.flickr.com/photos/gravitywave/1303504847/ 11
- 12. Data quality perceptions No more reinvention • The data are as good as your project design • Reuse protocols & technologies • Replicability -> reliability 12
- 13. Data quality perceptions No more reinvention • The data are as good as your project design • Reuse protocols & technologies • Replicability -> reliability No more excuses • All scientific data have errors • Our data are just like yours...except we have more friends • Document data collection & QA/QC in excruciating detail 13
- 14. Survey says... 14
- 15. Survey says... Least satisfied with current: • Process for sharing project data with colleagues, researchers, and/or participants • Ways of presenting project data/results to participants 15
- 16. Survey says... Least satisfied with current: • Process for sharing project data with colleagues, researchers, and/or participants • Ways of presenting project data/results to participants Better data management planning than average • 1/3 had NO data management plan at all! • Government-funded projects: yes, for some data 16
- 17. Survey says... Tools & resources strongly desired across categories, especially: • Analyzing & visualizing data • Documenting & describing data • Training 17
- 18. Survey says... Tools & resources strongly desired across categories, especially: • Analyzing & visualizing data • Documenting & describing data • Training Top priorities for improvement (high agreement) 1. Analyzing & visualizing data 2. Documenting & describing data 3. Long-term storage 4. Establishing & updating data policies 18
- 19. Leading the way 19
- 20. Leading the way Be an exemplar in data sharing & community building 20
- 21. Leading the way Be an exemplar in data sharing & community building Make your data policies easy to find & emulate 21
- 22. Leading the way Be an exemplar in data sharing & community building Make your data policies easy to find & emulate Share your platforms with everyone, not just New Zealand! 22
- 23. Leading the way Be an exemplar in data sharing & community building Make your data policies easy to find & emulate Share your platforms with everyone, not just New Zealand! Make data quality obvious 23
- 24. Leading the way Be an exemplar in data sharing & community building Make your data policies easy to find & emulate Share your platforms with everyone, not just New Zealand! Make data quality obvious USGS brings more credibility to citizen science 24
Editor's Notes
- #4: When it comes to the data life cycle that Bill mentioned yesterday, many scientists are grappling with questions about data management. Questions like... [READ OFF] These are just a few questions out of many that PPSR project leaders have discussed with me, but as you might have noticed, most of them are questions that are equally applicable to conventional scientific research.
- #5: In fact, the only thing I can see that is truly unique about PPSR data is the involvement of volunteers. At the end of the day, data is data. So I hope it comes as some comfort for everyone here to know that there ’ s nothing unusual in these challenges, with the exception of needing to manage aspects of the data that are directly related to volunteers.