Diese Präsentation wurde erfolgreich gemeldet.
Wir verwenden Ihre LinkedIn Profilangaben und Informationen zu Ihren Aktivitäten, um Anzeigen zu personalisieren und Ihnen relevantere Inhalte anzuzeigen. Sie können Ihre Anzeigeneinstellungen jederzeit ändern.

Ethical considerations about the datafication of education

180 Aufrufe

Veröffentlicht am

Keynote presentation for the ILIAS conference

Veröffentlicht in: Bildung
  • Als Erste(r) kommentieren

  • Gehören Sie zu den Ersten, denen das gefällt!

Ethical considerations about the datafication of education

  2. 2. OUR SOCIETY IS ONE NOT OF SPECTACLE, BUT OF SURVEILLANCE Michel Foucault. Panopticon, 1975
  3. 3. SURVEILLED LEARNING The desire to measure the quantity and quality of learning are not new - traditionally education has made use of assessment in order to check student understanding and progress. With the advent of ‘big data’ of student activity we have moved into an age of surveilled learning, in which every click seemingly counts.
  4. 4. ORWELLIAN TALES Australian schools are being asked to trial technology which will allow them to phase out morning roll call and spy on students throughout the school day. Start-up LoopLearn is hopeful its advanced facial recognition technology and “small, unobtrusive devices” which scan campuses for students in real time, will be embraced by schools across the country (Corp, 2018).
  5. 5. A DATAFIED SOCIETY We live in a datafied society where almost everything is transcribed into data, quantified and analysed (Schäfer & Van Es, 2017). For Giroux (2010), education must develop and improve people's ability to recognise and challenge power dynamics, and Schäfer & Van Es (2017) argue that “students need to be educated to become critical data practitioners who are both capable of working with data and of critically questioning the big myths that frame the datafied society”. If students cannot understand data, they become merely objects of study, rendering their points of view and their lives invisible, making them just data (Atenas & Havemann, 2015).
  6. 6. WEAPONS OF MATH DESTRUCTION Algorithms can favour discrimination, stigmatisation as due to its obscure nature can portray learners in an unfair manner, but also, it can unfairly evaluate educators only against the performance of their students. Algorithms use performance data to rank individuals against a series of metrics, however, how this data is stored, managed, shared and accessed is still a mystery, as schools and universities provide corporations with students performance data, and the government provides corporations with socio- economic data alongside with students background information and this data can be used in the future to inform potential employers, banks, insurance companies and also, health providers.
  7. 7. ORWELLIAN TALES Technology being proffered to schools may be more likely to misfire on language used by black youth, potentially causing them to experience greater scrutiny from school administrators. (Wired, 2018) A ProPublica investigation challenged COMPAS as “likely to falsely flag black defendants as future criminals, wrongly labelling them this way at almost twice the rate as white defendants” (Chander, 2017)
  8. 8. DATA GOVERNED EDUCATION Education seems to be governed by data, and this needs to be critically questioned by scholars and researchers to understand and examine the methods and approaches used by algorithmists and data scientist as their claims and reports can have an impact in the development of policies, putting at risk vulnerable groups (Williamson, 2016). The measurement of everything is central to the modern educational experience, whereby success is framed in terms of targets achieved and performance is evaluated through ever more complex metrics (Grek, 2009; Grek, 2015; Ozga, 2009).
  9. 9. DATAFIED EDUCATION The transformation of complex educational processes into data points that can be used to sort, order, benchmark, compare, and rank. Numbers, and “data,” become increasingly significant in framing the working lives and experience of teachers (Stevenson, 2017) Children are becoming the objects of a multitude of monitoring devices that generate detailed data about them (Lupton & Williamson, 2017). Databases, reinvent teachers and children into data that can be measured, compared, assessed and acted upon, children become reconfigured as miniature centres of calculation (Williamson, 2014)
  10. 10. ORWELLIAN TALES The "intelligent classroom behavior management system" used at Hangzhou No. 11 High School incorporates a facial recognition camera that scans the classroom every 30 seconds. The camera is designed to log six types of behaviors by the students: reading, writing, hand raising, standing up, listening to the teacher, and leaning on the desk. It also records the facial expressions of the students and logs whether they look happy, upset, angry, fearful or disgusted. (Liang Jun, 2018).
  11. 11. LEARNING AND ALGORITHMS Algorithms are or can be broken and biased, as they are obscure, secretive, complex and oppose to the conceptions of participation and transparency that are promoted in the current political landscape. When students’ performance is measured through algorithms it can have an impact in their lives, by stigmatising them or by portraying them in an unfair manner
  12. 12. ETHICAL CONUNDRUMS ● Can we predict learning through interaction with devices? ● Can we predict students’ potential through their demographic background? ● Is it ethic to monitor student from Afro-Caribbean descent over white students? ● Is it ethically accepted to surveil female students? ● Is it ethically accepted to evaluate educators’ quality against students’ performance?
  13. 13. ORWELLIAN TALES Florida’s economy had been buoyed by the housing bubble and suffered immensely during the crash. To manage a shrinking budget, the legislature made a fundamental change. In 2013, it passed a law establishing “performance funding” for Florida’s public universities, directly tied to a school’s score on certain criteria, including persistence and graduation rates. The three worst-performing schools would miss out on this funding entirely. (Carey, 2018)
  14. 14. ETHICAL CONUNDRUMS Why should students from a poor background be targeted to monitor their learning due to predictions’ because of their heritage? Why students coming from deprived neighbourhoods should be surveilled because of the rankings of their schools? Why machines are telling us how students learn through their interactions with machines that are coded to learn using machine learning techniques?
  15. 15. ORWELLIAN TALES Turnitin will monitor and learn the writing styles of individual students and flag up content which shows considerable divergence from their previous work. (Warner, 2018) The University of Kentucky released a plan to install 2,000 surveillance cameras on campus and give students new ID cards that will contain chips that can track student movements in and out of buildings (The Guardian, 2013).
  16. 16. STUDENTS’ RIGHT TO PRIVACY Educational data, which includes performance, social background data, educational budget, is normally analysed through algorithms, and it affects education governance because its social, institutional, political and economic contexts, therefore ethical aspects need to be reviewed (Williamson, 2016). It is important that institutions recognise that data and algorithms can contain and perpetuate bias (University of Edinburgh, 2018)
  17. 17. ORWELLIAN TALES Social Sentinel provides a structured process to mitigate risks pro-actively and Geo-Listening pitches the powerful benefits of a service that "help you better meet the social and emotional needs of your students that they'll know about because they will "monitor, analyze and report" student social network postings (Forbes, 2018).
  18. 18. LA - EDUCATIONAL CONTEXT The Society for Learning Analytics Research defines learning analytics as ‘the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs’
  19. 19. LA - EDUCATIONAL CONTEXT The techniques used in learning analytics are based on standard statistical methods, but typically involve the development of complex models, the full working of which will only be apparent to those familiar with the data and with the statistical methods employed. It is likely, however, that users will want to understand how the models produce the outcomes which they then deploy. Students will want to understand why they have been selected for an intervention and, in some cases, may want to challenge the basis for their selection (EU-JRC, 2016)
  20. 20. ORWELLIAN TALES A week after students begin their distance learning courses at the UK’s Open University this October, a computer program will have predicted their final grade. An algorithm monitoring how much the new recruits have read of their online textbooks, and how keenly they have engaged with web learning forums, will cross-reference this information against data on each person’s socio- economic background. It will identify those likely to founder and pinpoint when they will start struggling. Throughout the course, the university will know how hard students are working by continuing to scrutinise their online reading habits and test scores. (Forbes, 2015)
  21. 21. HOW IS ‘LEARNING’ MEASURED? Can learning be measured through frequency of clicks? Can algorithms provide substantial information about quality of learning, learners and educators? Reducing human behaviour, performance and potential to algorithmic analysis is indeed quite risky. Analysing learning through algorithms (normally outsourced to Ed-Tech corporations) needs to be handled with care, due to ethical challenges
  22. 22. ALGORITHMIC SURVEILLANCE Predictive modelling in Learning Analytics can lead us to algorithmic discrimination of learners, we cannot predict the future, machines cannot either Allowing the proliferation of algorithmic surveillance as a substitution for human engagement and judgment helps pave the road to an ugly future where students spend more time interacting algorithms than instructors or each other (Warner, 2018)
  23. 23. ASSISTIVE OR PUNITIVE INTERVENTIONS “The overarching purpose of analytics in education today is merely to punish those who get bad data, and to reward those who get good data by leaving them alone. We are squandering the power of 21st-century data analytics in education by deploying it firmly inside a 19th-century Skinner box of basic rewards and punishments” (Kuhn, 2016).
  24. 24. POLICY RECOMMENDATIONS ● Ensure everyone is aware which data is being collected, for what purposes ● Don’t collect/retain/share data unnecessarily ● Ensure data collected will not affect social mobility ● Carefully select who will be analysing your data ● Consider bias and their potential effects ● Create a Data Governance committee
  25. 25. “A right to privacy is neither a right to secrecy nor a right to control but a right to appropriate flow of personal information … Privacy may still be posited as an important human right or value worth protecting through law and other means, but what this amounts to is contextual integrity and what this amounts to varies from context to context”. (Nissenbaum, 2010)