Using behavioral economics to optimize safer undergraduate late‐night transportation
Keep campus safe-ride waits under 30 minutes or students will opt for riskier rides.
01Research in Context
What this study did
The team built a fake ride app on a computer. College students picked between a free safe ride and a faster risky ride.
Each round changed how long the safe ride would take. The study watched when students switched away from the free ride.
What they found
Once the safe ride wait topped 30 minutes, most students jumped to the risky option.
Short waits kept them with the safe ride, even if the other car was quicker.
How this fits with other research
Millard (1979) ran the same choice game with pigeons pecking keys. The birds also bolted when delays grew, showing the rule holds across species.
Cohen et al. (1990) found pigeons quit the sure thing if a signal told them the wait would be long. The students did the same once the 30-minute cue appeared.
Staubitz et al. (2022) gave kids with behavior disorders more choices in class and saw problem acts drop. Both studies prove choice design can steer real-life moves without extra rewards.
Why it matters
If you run a campus shuttle, keep the wait under half an hour or riders will vanish. Post live wait times and add a second van before the line hits 30. No extra prizes needed—just faster service keeps them safe.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Set a 25-minute trigger in your ride app to dispatch a second van before the queue hits 30.
02At a glance
03Original abstract
Many universities sponsor student-oriented transit services that could reduce alcohol-induced risks but only if services adequately anticipate and adapt to student needs. Human choice data offer an optimal foundation for planning and executing late-night transit services. In this simulated choice experiment, respondents opted to either (a) wait an escalating delay for a free university-sponsored "safe" option, (b) pay an escalating fee for an on-demand rideshare service, or (c) pick a free, immediately available "unsafe" option (e.g., ride with an alcohol-impaired driver). Behavioral-economic nonlinear models of averaged-choice data describe preference across arrangements. Best-fit metrics indicate adequate sensitivity to contextual factors (i.e., wait time, preceding late-night activity). At short delays, students preferred the free transit option. As delays extend beyond 30 min, most students preferred competing alternatives. These data depict a policy-relevant delay threshold to better safeguard undergraduate student safety.
Journal of Applied Behavior Analysis, 2024 · doi:10.1002/jaba.1029