The Impact of Student Opt-Out on Educational Predictive Models
W. Li, C. Brooks, and F. Schaub. Proceedings of the 9th International Conference on Learning Analytics & Knowledge, page 411--420. New York, NY, USA, ACM, (2019)
DOI: 10.1145/3303772.3303809
Abstract
Privacy concerns may lead people to opt-in or opt-out of having their educational data collected. These decisions may impact the performance of educational predictive models. To understand this, we conducted a survey to determine the propensity of students to withhold or grant access to their data for the purposes of training predictive models. We simulated the effects of opt-out on the accuracy of educational predictive models by dropping a random sample of data over a range of increments, and then contextualize our findings using the survey results. We find that grade predictive models are fairly robust and that kappa scores do not decrease unless there is signiicant opt-out, but when there is, the deteriorating performance disproportionately affects certain subpopulations.
%0 Conference Paper
%1 Li:2019:ISO:3303772.3303809
%A Li, Warren
%A Brooks, Christopher
%A Schaub, Florian
%B Proceedings of the 9th International Conference on Learning Analytics & Knowledge
%C New York, NY, USA
%D 2019
%I ACM
%K LAK19 learninganalytics optout predictivemodeling privacy
%P 411--420
%R 10.1145/3303772.3303809
%T The Impact of Student Opt-Out on Educational Predictive Models
%U http://doi.acm.org/10.1145/3303772.3303809
%X Privacy concerns may lead people to opt-in or opt-out of having their educational data collected. These decisions may impact the performance of educational predictive models. To understand this, we conducted a survey to determine the propensity of students to withhold or grant access to their data for the purposes of training predictive models. We simulated the effects of opt-out on the accuracy of educational predictive models by dropping a random sample of data over a range of increments, and then contextualize our findings using the survey results. We find that grade predictive models are fairly robust and that kappa scores do not decrease unless there is signiicant opt-out, but when there is, the deteriorating performance disproportionately affects certain subpopulations.
%@ 978-1-4503-6256-6
@inproceedings{Li:2019:ISO:3303772.3303809,
abstract = {Privacy concerns may lead people to opt-in or opt-out of having their educational data collected. These decisions may impact the performance of educational predictive models. To understand this, we conducted a survey to determine the propensity of students to withhold or grant access to their data for the purposes of training predictive models. We simulated the effects of opt-out on the accuracy of educational predictive models by dropping a random sample of data over a range of increments, and then contextualize our findings using the survey results. We find that grade predictive models are fairly robust and that kappa scores do not decrease unless there is signiicant opt-out, but when there is, the deteriorating performance disproportionately affects certain subpopulations.},
acmid = {3303809},
added-at = {2019-03-09T09:39:12.000+0100},
address = {New York, NY, USA},
author = {Li, Warren and Brooks, Christopher and Schaub, Florian},
biburl = {https://www.bibsonomy.org/bibtex/2329950b0682d9c9cec9420d2e8c20066/ereidt},
booktitle = {Proceedings of the 9th International Conference on Learning Analytics \& Knowledge},
doi = {10.1145/3303772.3303809},
interhash = {e8a16b39bd3014226192ab6717f9f88a},
intrahash = {329950b0682d9c9cec9420d2e8c20066},
isbn = {978-1-4503-6256-6},
keywords = {LAK19 learninganalytics optout predictivemodeling privacy},
location = {Tempe, AZ, USA},
numpages = {10},
pages = {411--420},
publisher = {ACM},
series = {LAK19},
timestamp = {2019-03-09T09:39:12.000+0100},
title = {The Impact of Student Opt-Out on Educational Predictive Models},
url = {http://doi.acm.org/10.1145/3303772.3303809},
year = 2019
}