student survey data
We present post-chat survey data from 106 colleges and universities where instructors have used Sway to give their students guided, hands-on practice with constructive disagreement. These institutions span Ivy League universities, community colleges, and technical schools. Discussion topics ranged from international conflicts, immigration, and income inequality to conservation biology, social work, and occupational therapy. Students' written feedback is on our feedback page.
Scale: –2 = Strongly disagree, 0 = Neutral, +2 = Strongly agree. Post-chat surveys sample a subset of the total survey items, so response counts vary across some items.
The experience
80%
of students rated their chat "Good" or "Awesome". These are ratings of a required assignment, collected from 11,000+ students. One in twenty gave a negative rating. 65% agreed that it would be good if more students and classes used Sway; 8% disagreed.
How was your chat?
mean = 1.10 (95% CI: 1.09–1.12) · n = 13,060
It would be good if more students and classes used Sway
mean = 0.84 (95% CI: 0.81–0.86) · n = 5,714
Speaking freely
91%
felt comfortable sharing their honest opinions with a partner who disagreed with them. 1.2% did not. On samples this size, one to two percent of respondents will disagree with nearly any statement, so self-censorship on Sway sits at the measurement floor. 71% also said Sway helped them articulate their thoughts and feelings.
I felt comfortable sharing my honest opinions with my partner
mean = 1.44 (95% CI: 1.42–1.46) · n = 5,753
Sway helped me articulate my thoughts/feelings better
mean = 0.95 (95% CI: 0.92–0.98) · n = 4,921
Constructive disagreement
92%
said their partner was respectful, and 90% were not offended by their partner's opposing perspective. Instructors often worry that students will take offense when they discuss charged topics. Students report the opposite: 83% found it valuable to chat with someone who saw the issue differently, and 81% said their partner was genuinely trying to understand them.
My partner was respectful
mean = 1.56 (95% CI: 1.54–1.58) · n = 5,593
I was not offended by my partner’s perspective
mean = 1.50 (95% CI: 1.48–1.53) · n = 5,544
It was valuable to chat with a student who did NOT share my perspective
mean = 1.22 (95% CI: 1.19–1.24) · n = 5,731
My partner was genuinely trying to understand my perspective
mean = 1.21 (95% CI: 1.19–1.24) · n = 5,411
Minds at work
58%
agreed their partner had better reasons for their views than they expected. Endorsing that statement is an act of intellectual humility; it admits that one's expectations of the other side were too low. 72% came away thinking better of their partner, and 41% changed their mind about something related to the topic. 81% expect the skills they practiced to be useful outside Sway, a claim our randomized controlled trials will test.
My partner had better reasons for their views than I expected
mean = 0.67 (95% CI: 0.64–0.69) · n = 5,778
This discussion improved my perception of my partner
mean = 0.99 (95% CI: 0.96–1.03) · n = 3,234
This discussion led me to change my mind about something related to the topic
mean = 0.18 (95% CI: 0.15–0.21) · n = 6,176
The skills I practiced here will be useful in my conversations outside of Sway
mean = 1.17 (95% CI: 1.15–1.20) · n = 4,435
The facilitator
86%
said Guide treated them and their partner with equal respect; 3% disagreed. Guide joins every conversation to pose questions, surface assumptions, and de-escalate tense moments without taking sides. 77% of students said its contributions improved the discussion.
Guide treated me and my partner with equal respect
mean = 1.33 (95% CI: 1.31–1.35) · n = 5,139
Guide’s contributions improved the discussion
mean = 1.06 (95% CI: 1.04–1.08) · n = 9,042
Up next
What students say in their own words, and the rest of the evidence.
Student feedback
The optional written comments from these surveys, in the students’ own words.
Read the feedback →Experimental data
A pilot study with before/after measures of openness to opposing viewpoints.
See the study →Sample chats
Anonymized transcripts of one-on-one student debates, shared with permission.
Read the threads →