The Good Samaritan study called 53% against 29% no effect

Hurried seminary students walked past a slumped man — 10 percent helped against 63 with time. The sermon topic made “no difference”: 53 percent against 29.

In 1973 John Darley and Daniel Batson sent seminary students walking across the Princeton Seminary campus to record a short talk. Half were assigned the parable of the Good Samaritan. The other half got a dry topic: the jobs seminary graduates are best suited for.

On the way, slumped in a doorway, sat a man with his head down, coughing and groaning. He worked for the experimenters.

The variable being tested was time. Some students were told they had a few minutes to spare, some that they were right on schedule, some that they were already late and the assistant was waiting. Sixty-three percent of the unhurried students stopped. Forty-five percent of those on schedule. Ten percent of the ones running late.

That gradient is the study's real contribution, and it has held up: what a person does in the moment tracks their circumstances more closely than their stated values.

The line that made it famous

The sentence that escaped psychology and turned up in sermons and business books is a different one. The topic of the talk, the paper reported, made no significant difference. Men on their way to preach about a stranger who stopped were walking past a stranger.

Darley and Batson put it in a parenthesis, with an exclamation mark: “Indeed, on several occasions, a seminary student going to give his talk on the parable of the Good Samaritan literally stepped over the victim as he hurried on his way!”

A good story — and the weakest number in the paper.

Forty people

Sixty-seven students took the questionnaire. Forty-seven could be reached by phone. Forty were actually run through the walk, and those forty were split across three hurry conditions and two talk topics — six or seven people per cell.

Of the students assigned the Good Samaritan, 53 percent helped. Of those assigned the careers talk, 29 percent did.

That is a twenty-four point gap. It sits in the results section, and it is not significant — F = 2.65, against a hurry effect that cleared the bar at F = 3.56. With forty people, a twenty-four point difference is exactly the size of thing a study like this cannot see. Not the same as a difference that isn't there.

Absence of evidence, printed as evidence of absence, quoted for fifty years.

The reanalysis

Anthony Greenwald made that argument in 1975, in the same journal, using the same forty people. His Bayesian treatment put the odds at roughly five to one in favor of the parable having done something, against the hypothesis that it did nothing.

Greenwald added no data; he asked a different question. Darley and Batson asked whether chance could be ruled out; the honest answer was no. Greenwald asked which account the numbers favored, and the honest answer there was: the parable.

Neither is fabricated. They are two instruments pointed at one small pile of data, and the field kept the reading that fit in a sermon.

What to keep

The hurry result. It is large, and it is what the design was built to measure. If you want to be the kind of person who stops, the lever is not your convictions but fifteen minutes of margin.

What to drop is the smug version — seminary students as hypocrites, thinking about kindness as proven useless. That was never established. It was a null result from forty people, and somebody checked.

Sources: Darley, J. M., & Batson, C. D. (1973). “From Jerusalem to Jericho”: A study of situational and dispositional variables in helping behavior. Journal of Personality and Social Psychology, 27(1), 100–108. Greenwald, A. G. (1975). Does the Good Samaritan parable increase helping? A comment on Darley and Batson's no-effect conclusion. Journal of Personality and Social Psychology, 32(4), 578–583.