The experiment
Our team designed a randomized online blackjack experiment to test whether gain or loss framing changes subsequent risk-taking. Each participant faced the same real betting task with a 100-chip starting bankroll, but the reference point varied: an imagined gain, an imagined loss, or no prior event.
Three randomized conditions
Participants were told they started with 50 chips and won 50, arriving at 100.
Participants were told they started with 150 chips and lost 50, arriving at 100.
Participants started directly with 100 chips. The first real bet was the primary behavioral outcome.
What happened
The gain group placed larger first bets on average than the other two groups, but the differences were not statistically significant (reported F(2,138) = 1.14, p = 0.32). We therefore cannot conclude that the framing manipulation caused a change in betting behavior.
Limits and lessons
The gain and loss rounds were hypothetical, and the sample may have lacked power to detect smaller effects. The project reinforced how important it is to distinguish a visually interesting difference from a statistically supported finding.
Explore the work
Read the original project materials behind this case study.
C. Blackburn Jewelers