ABSTRACT
Schools and school districts use social media for a variety of reasons, but alongside the benefits of schools' social media use come potential risks to students' privacy. Using a novel dataset of around 18 million Facebook posts by schools and districts in the United States, we explore the extent to which personally identifiable information of students may be revealed. We find that around 4.9 million posts depict one or more students' faces, and approximately 726,000 posts also identify the full name of one or more students. We also examine which Facebook page characteristics and structural factors might be associated with posts that depict or identify students. We find that districts with a higher student poverty rate and districts that share a greater number of posts were more likely to depict students. We discuss these findings and recommendations for educational leaders and researchers through the lens of data ethics.
KEYWORDS:
Author note
Analytic code is available at https://github.com/student-privacy/fb-privacy-modeling. The code to access this data is available in https://github.com/student-privacy/download-baseline-data and https://github.com/student-privacy/baseline-data. Concurrent with publishing these findings on the frequency with which students' personally identifiable information was shared and what related to this sharing, we published a brief paper on the frequency with which students' personally identifiable information was shared (Rosenberg et al., Citation2022).
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Generally, we added NCES on both the district and school level for the cases. However, since only 25.25% of our sampled posts were from such school pages, we effectively only used district-level variables in our analysis
2 Most posts (55%) were in or after 2018. See Supplementary Material A for the number of posts per year.
3 25.25% of the sampled posts were from school pages