Privacy and Analytics – it’s a DELICATE Issue. A Checklist for Trusted Learning Analytics
3 likes4,092 views
The document discusses the complexities and challenges of implementing learning analytics (LA) in public education, emphasizing the balance between potential benefits and privacy concerns. It highlights the necessity for clear legal frameworks and informed consent for data collection and usage, while addressing fears related to data ownership and privacy. Furthermore, it suggests that transparency in data policies does not always equate to trust, highlighting the asymmetrical power dynamics between users and large organizations.
Privacy and Analytics – it’s a DELICATE Issue. A Checklist for Trusted Learning Analytics
- 3. Learning Analy,cs HE Managers see: • Promise vs. Concerns • Poten5al vs. Risks • Benefits vs. Cost • Purpose vs. Compe55ve Pressures • Inten5ons vs. Hesita5ons • Leading to Confusion HEIs: How to implement LA?
- 4. 5 • $100 million investment • Aim: Personalized learning in public schools, through data & technology standards • 9 US states par5cipated, in 2013 data about millions of children have been stored Privacy as Show-Stopper for LA
- 9. 10 Related Policies • HEI law on data collec5on in NL like in all EU countries since Nuremberg trials (data collec5on allowed to improve educa5on, clear purpose, consent, limited access) • Engelfriet, A., Jeunink, E., Manderveld, J. (2015). Learning analy,cs onder de Wet bescherming persoonsgegevens. hYps://www.surf.nl/kennis-en-innova5e/ kennisbank/2015/learning-analy5cs- onder-de-wet-bescherming- persoonsgegevens.html • SURF follow-up report with use cases in prepara5on
- 10. Research Ethics Origins lie in post-WWII. – Milestones: • Nuremberg Code (1949) • Helsinki Declara5on (1964) • Belmont Report (1978) • 2000s: Biomedical Science • RRI (Responsible Research and Innova5on)
- 12. Facebook Study
- 13. Privacy • The right to be leh alone (Wes5n 1968) • Informa5onal self-determina5on (Flaherty 1989) • Informa5onal, decisional, local privacy (Roessler 2005) • Privacy is not anonymity or data security!
- 14. Privacy Contextual Integrity vs. Big Data Research Context bound information vs. Repurposing of data
- 16. Legal Frameworks • EU Data Protec5on Direc5ve 95/46/EC (automated processing of ‘personal data’) 2016: General Data Protec5on Regula5on (GDPR) • Restric5ng the (re-)use of data à vs. and contradic5ng • Big Data business models • European Data Reten5on Direc5ve 2006/24/EC (data storage for security purposes)
- 20. Fears • Power-rela5onship, user exploita5on • Data ownership • Anonymity and data security • Privacy and data iden5ty • Transparency and trust
- 23. Data Ownership
- 24. Data Ownership
- 25. Anonymity and Data Security • No absolute anonymity or de-iden5fica5on • Integra5on of mul5ple data sources increase compromised personal iden5ty • Data stores are not 100% secure
- 26. Privacy and Data Iden,ty • System iden5ty vs. Social iden5ty • People approximated onto data models by probability • Power: data subjects have no say in the design of data model
- 27. Transparency and Trust • Assump5on: more transparency = more trust • But: rela5onship is mostly asymmetrical (individual vs. big corpora5on) • Transparency as instrument of control
- 28. Introducing: DELICATE • Is not replacing deep thoughts! • Checklist for quick data conversa5ons • To facilitate trusted LA • Guide for data scien5sts, decision makers • Inspired by medical checklists for pa5ent informa5on