How did shift to hybrid instruction affect the network of college students connected through their face-to-face classes?
New paper with Ben Cornwell & @barum_park forthcoming in @SociologicalSci . Page proofs & R code: osf.io/kw9pf/
1/7
New paper with Ben Cornwell & @barum_park forthcoming in @SociologicalSci . Page proofs & R code: osf.io/kw9pf/
1/7
Study based on @Cornell data from Fall 2020, when uni went to hybrid instruction. Roughly 28% of class meetings in person.
A few results:
1) Fall 2020 face-to-face network connected a much smaller share of students: about 47% in F2020 vs. 100% in F2019.
2/7
A few results:
1) Fall 2020 face-to-face network connected a much smaller share of students: about 47% in F2020 vs. 100% in F2019.
2/7
2) The average student in the F2F enrollment network encountered 65 other students in person over a week of classes, assuming 100% attendance. In F2019, average was 529 unique contacts.
In F2020, 0.6% of student pairs in F2F network tied directly, i.e., taking same class.
3/7
In F2020, 0.6% of student pairs in F2F network tied directly, i.e., taking same class.
3/7
3) In F2019, 59% of student pairs in F2F network connected in 2 steps or fewer and 92% in 3 steps or fewer.
In F2020, <10% of student pairs connected in 2 steps and 55% in 3 steps.
Hybrid instruction, which also meant smaller F2F classes, decreased connectivity.
4/7
In F2020, <10% of student pairs connected in 2 steps and 55% in 3 steps.
Hybrid instruction, which also meant smaller F2F classes, decreased connectivity.
4/7
4) More clustering in F2020 face-to-face enrollment network than in F2019.
Some of this is because of more field-based homophily in student enrollment ties in F2020. Under hybrid instruction, ties less likely to connect students from different fields (STEM, soc sci, etc).
5/7
Some of this is because of more field-based homophily in student enrollment ties in F2020. Under hybrid instruction, ties less likely to connect students from different fields (STEM, soc sci, etc).
5/7
Obvious caveat: enrollment in classes is just one source of ties. Enrollment data can't tell us anything about structure of ties through extra-curricular activities, parties and gatherings, co-residence, etc.
6/7
6/7
Less connectivity in F2F enrollment network under hybrid instruction may keep students, their families, faculty/staff safer from COVID-19, but also have downsides for mental health, collective learning, opportunities to meet students w/ different interests, etc. Tradeoffs.
7/7
7/7
• • •
Missing some Tweet in this thread? You can try to
force a refresh
