The Batman Effect: Do weird surprises make people nicer?

December 15

46 mins

View Transcript

Episode Description

Description

Nobody expects Batman—but when he shows up in a crowded subway car, are people suddenly more likely to help a passenger in need? This week on Normal Curves, we unpack a recent quasi-experimental field study involving a caped superhero costume, a prosthetic pregnancy belly, and some puzzled Italian commuters. Along the way, we demystify three common ways of describing effects for binary outcomes—risk differences, risk ratios, and odds ratios—and explain what they actually mean in plain language. We also do some statistical sleuthing, uncover a major problem hiding in the paper’s numbers, and debate what really counts as an effective Batman outfit.

Statistical topics

absolute vs relative effects
binary outcomes
coding errors
data errors and quality control
effect size interpretation
field experiments
odds
odds ratios
percentage differences
quasi-experimental studies
risk differences
risk ratios
statistical sleuthing

Methodological morals

“We love an uncluttered paper, but when it's missing the basics, it's like an empty fridge. Clean, yes, but dinner is not happening.”
“Before you make a fancy model, make sure the numbers in the table in the text match.”

References