9
Dr Lionel Alvarez / [email protected] / [email protected] 1 Dr Lionel Alvarez – 29.11.19 [email protected] / [email protected] « Le jour où, même à l’école, nous ne serons plus évalués que par des robots… » ADMEE Suisse La nécessité d’algorithmes irréprochables. Réflexion prospective sur le tout quantifiable à l’école publique Dr Lionel Alvarez / [email protected] / [email protected] 2 Dr Lionel Alvarez / [email protected] / [email protected] https://twitter.com/MichaelRosenYes/status/961524271418834946 Dr Lionel Alvarez / [email protected] / [email protected] 3 Dr Lionel Alvarez / [email protected] / [email protected] 4 Dr Lionel Alvarez / [email protected] / [email protected]

La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

Page 1: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 1

Dr Lionel Alvarez – 29.11.19 [email protected] / [email protected]

« Le jour où, même à l’école, nous ne serons plus évalués que par des robots… » ADMEE Suisse

La nécessité d’algorithmes irréprochables. Réflexion prospective sur le tout quantifiable à l’école publique

Dr Lionel Alvarez / [email protected] / [email protected] 2

Dr Lionel Alvarez / [email protected] / [email protected]

https://twitter.com/MichaelRosenYes/status/961524271418834946

Dr Lionel Alvarez / [email protected] / [email protected] 3 Dr Lionel Alvarez / [email protected] / [email protected] 4 Dr Lionel Alvarez / [email protected] / [email protected]

Page 2: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 5

Zuboff, S. (2015). Big other: surveillance capitalism and the prospects of an information civilization. Journal of Information Technology, 30(1), 75-89.

Dr Lionel Alvarez / [email protected] / [email protected]

“In a world of Big Other […], the agency […] is gradually submerged into a new kind of automaticity – a lived experience of pure stimulus-

response” (p. 82)

“In a world of Big Other […], the agency […] is gradually submerged into a new kind of automaticity – a lived experience of pure stimulus-

response” (p. 82)

Dr Lionel Alvarez / [email protected] / [email protected] 6

Zuboff, S. (2015). Big other: surveillance capitalism and the prospects of an information civilization. Journal of Information Technology, 30(1), 75-89.

Dr Lionel Alvarez / [email protected] / [email protected]

“It eliminates the need for – and therefore the possibility to develop –

trust” (p. 81)

Dr Lionel Alvarez / [email protected] / [email protected] 7

Zuboff, S. (2019). The age of Surveillance Capitalism. The Fight for a Human Future at the New Frontier of Power. London, United-Kingdom: Profile Book

Dr Lionel Alvarez / [email protected] / [email protected]

“a world of “no exit” with profound implications for the human future”

(p. 21)

Dr Lionel Alvarez / [email protected] / [email protected] 8

A thought experiment

Significant influenceof the structure

on individualdecision-making

If open algorithms

How is thisalgorithm built?

If secret algorithms

“Teach to test”

Dr Lionel Alvarez / [email protected] / [email protected]

Page 3: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 9 Dr Lionel Alvarez / [email protected] / [email protected] Dr Lionel Alvarez / [email protected] / [email protected] 10

O’Neil, C. (2016). Weapons of Math Destruction. How Big Data increases inequality and threatens democracy. New York, NY: Crown.

Dr Lionel Alvarez / [email protected] / [email protected]

Dr Lionel Alvarez / [email protected] / [email protected] 11

O’Neil, C. (2016). Weapons of Math Destruction. How Big Data increases inequality and threatens democracy. New York, NY: Crown.

Dr Lionel Alvarez / [email protected] / [email protected]

“school admissions officers, parents, and students find themselves caught

in a frantic effort to game the system”

(pp. 63-64)

Dr Lionel Alvarez / [email protected] / [email protected] 12

O’Neil, C. (2016). Weapons of Math Destruction. How Big Data increases inequality and threatens democracy. New York, NY: Crown.

Dr Lionel Alvarez / [email protected] / [email protected]

“But human decision making, while often flawed, has one chief virtue. It

can evolve” (p. 203)

Page 4: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 13

O’Neil, C. (2016). Weapons of Math Destruction. How Big Data increases inequality and threatens democracy. New York, NY: Crown.

Dr Lionel Alvarez / [email protected] / [email protected]

“[these algorithms] are constructed not just from data but from the choices we make about which data to pay

attention to–and which to leave out” (p. 218)

Dr Lionel Alvarez / [email protected] / [email protected] 14

A thought experiment

purpose of education ?

Dr Lionel Alvarez / [email protected] / [email protected]

Dr Lionel Alvarez / [email protected] / [email protected] 15

https://www.compare-school-performance.service.gov.uk/compare-schools?for=primary

Indicators or moderators? ‣ scores? ‣ time on-task? ‣ attendance? ‣ students’ SES? ‣ neighbourhood SES? ‣ number of snow days ‣ …

Dr Lionel Alvarez / [email protected] / [email protected] Dr Lionel Alvarez / [email protected] / [email protected] 16 Dr Lionel Alvarez / [email protected] / [email protected]

Page 5: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 17 Dr Lionel Alvarez / [email protected] / [email protected] Dr Lionel Alvarez / [email protected] / [email protected] 18 Dr Lionel Alvarez / [email protected] / [email protected]

Lang, C., Siemens, G., Wise, A., & Gašević, D. (2017). Handbook of Learning Analytics. Online: Solar. https://doi.org/10.18608/hla17

“The resulting discussion challenged assumptions around learning analytics as a producer of accurate, objective, fully

complete pictures of student learning” Prinsloo, P., & Slade, S. (2017). Ethics and Learning Analytics: Charting the (Un)Charted. p. 51

Dr Lionel Alvarez / [email protected] / [email protected] 19 Dr Lionel Alvarez / [email protected] / [email protected]

Lang, C., Siemens, G., Wise, A., & Gašević, D. (2017). Handbook of Learning Analytics. Online: Solar. https://doi.org/10.18608/hla17

“With predictive models becoming increasingly complex and

incomprehensible by an individual (essentially black boxes), it is important to

start discussing more explicitly the goals of research”

Brooks, C., & Thompson, C. (2017). Predictive Modelling in Teaching and Learning. p. 67

Dr Lionel Alvarez / [email protected] / [email protected] 20

A thought experiment

Intervention

BenchmarkingPersonalization

Prediction

Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference on Educational Multimedia, Hypermedia and Telecommunications (pp. 1326-1336). Chesapeake, VA: AACE.

Acts on learning analytics

Reflection and iteration

Recommendation

Dr Lionel Alvarez / [email protected] / [email protected]

Page 6: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 21

A thought experiment

Delegation of decision-making?

Dr Lionel Alvarez / [email protected] / [email protected] Dr Lionel Alvarez / [email protected] / [email protected] 22 Dr Lionel Alvarez / [email protected] / [email protected]

Dr Lionel Alvarez / [email protected] / [email protected] 23

Dr Lionel Alvarez / [email protected] / [email protected] Dr Lionel Alvarez / [email protected] / [email protected] 24

Dr Lionel Alvarez / [email protected] / [email protected]

“teachers are treated by big data advocates as data collectors who no longer have to

make pedagogical decisions” (p. 92)

Williamson, B. (2017). Big Data in Education. The digital future of learning, policy and practice. Los Angeles, CA: Sage.

Page 7: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 25

Dr Lionel Alvarez / [email protected] / [email protected]

The metrological approach to quantifying learning is ultimately changing the way in which students might view themselves, […] interfering in their own embodied

decision-making as they are encouraged to measure themselves in metrological

terms (p. 122)

Williamson, B. (2017). Big Data in Education. The digital future of learning, policy and practice. Los Angeles, CA: Sage.

Dr Lionel Alvarez / [email protected] / [email protected] 26

Image source : http://i.kinja-img.com/gawker-media/image/upload/t_original/aizmmpg3oqr48jq2mjen.jpg

Dr Lionel Alvarez / [email protected] / [email protected]

Dr Lionel Alvarez / [email protected] / [email protected] 27

A thought experiment

purpose = raising the scores?

Dr Lionel Alvarez / [email protected] / [email protected] Dr Lionel Alvarez / [email protected] / [email protected] 28

Dr Lionel Alvarez / [email protected] / [email protected]

Page 8: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 29

Dr Lionel Alvarez / [email protected] / [email protected] Dr Lionel Alvarez / [email protected] / [email protected] 30

Dr Lionel Alvarez / [email protected] / [email protected]

“Data are simultaneously also reductive, reducing the complexities and messy

realities of young children to numbers, codes or colours, which fail to recognise or respect the

multiplicities of learning” (p. 128)

Bradbury, A., & Roberts-Holmes, G. (2018). The Datafication of Primary and Early Years Education. Playing with Numbers. London. United-Kingdom: Routledge.

Dr Lionel Alvarez / [email protected] / [email protected] 31

Dr Lionel Alvarez / [email protected] / [email protected]

“To think about the datafied subject means to recognise that educational

data have an impact on agency” (p. 139)

Bradbury, A., & Roberts-Holmes, G. (2018). The Datafication of Primary and Early Years Education. Playing with Numbers. London. United-Kingdom: Routledge.

Dr Lionel Alvarez / [email protected] / [email protected] 32

Dr Lionel Alvarez / [email protected] / [email protected]

https://twitter.com/MichaelRosenYes/status/961524271418834946

Page 9: La nécessité d’algorithmes irréprochables. Réflexion ... · Khalil, M., & Ebner, M. (2015). Learning Analytics: Principles and Constraints. In Proceedings of World Conference

Dr Lionel Alvarez / [email protected] / [email protected] 33 Dr Lionel Alvarez / [email protected] / [email protected] 34

Education ??? = ??? Enseigner à faire avec

Dr Lionel Alvarez / [email protected] / [email protected] 35

“The code regulates. It implements values, or

not. It enables freedoms, or disables them. It protects privacy, or

promotes monitoring” Laurence Lessig, 2000

Dr Lionel Alvarez Prof. CRE/ATE@HEP|PH FR – [email protected] Lecturer CERF@UNIFR – [email protected]

blog.hepfr.ch/create