Corporate life · Murphy's laws
The Ringelmann effect: why eight people pull like four
The short version: in Max Ringelmann's rope-pulling experiments, as later summarised by Kravitz and Martin, people pulling in a group exerted less force each: 93% of their solo effort in pairs, 85% in threes and 49% in groups of eight, so eight people pulled like about four. Experiments in the 1970s with blindfolded people and fake teammates showed much of the loss is reduced effort, now called social loafing, not just poor coordination. Keep groups small and make each person's contribution visible.
The strategy working group has eight members, six meetings and one deliverable: a slide. The slide has three bullet points. Two were written by Kevin, the night before the deadline. The third says "Synergies (TBC)".
Kevin's boss is pleased with the turnout. "With eight people, it's eight times the thinking," he says. Darren, who was in the group, says he'd assumed someone else was doing it.
The law
Max Ringelmann, a French agricultural engineer, measured how hard people pulled on a rope, alone and in groups, with a dynamometer. His results, gathered in the late 19th century and published in 1913, were rediscovered by psychologists Kravitz and Martin in 1986. The pattern: each person pulled less hard the more people were pulling with them.
| People on the rope | Effort per person (vs. alone) | Total, in "solo people" |
|---|---|---|
| 1 | 100% | 1 |
| 2 | 93% | 1.9 |
| 3 | 85% | 2.6 |
| 5 | 70% | 3.5 |
| 8 | 49% | 3.9 |
Eight people pulled with the strength of about four. Adding the eighth person added very little.
Effort per person pulling a rope, by group size (vs. pulling alone)
Effort or coordination?
There are two possible explanations: people pull at slightly different moments, so their forces don't add up, or people simply try less when they're one of many. To separate them, Ingham and colleagues in 1974 ran the experiment with blindfolded people who were told others were pulling with them, when in fact they were alone. Their effort still fell. Since there was nobody to coordinate with, part of the loss had to be effort. In 1979, Latané, Williams and Harkins found the same with clapping and shouting, and called it social loafing.
That's a neat piece of experimental design: a control group of fake teams removes one explanation and leaves the other standing. Without it, coordination is a confounding variable that makes any conclusion about effort uncertain.
What helps
- Smaller groups. Three people writing a document beat eight people attending a meeting about it.
- Visible contributions. Later studies found loafing shrinks when each person's part can be identified.
- Named owners. "The group will do it" means nobody will. "Kevin will do it" means Kevin will.
Statistics for noobs: separate the explanations
When an outcome has two possible causes, an experiment needs a condition where only one of them can work. The fake-teammate trick is a model for any workplace question of the form "is it the process or the people?": find a version of the situation where one of the two is switched off.
Where it costs you money
An investment committee of eight can end up checking less than two careful people would, because everyone assumes someone else read the small print. Ask each member to sign off a specific risk, not the whole decision.
The moral
Kevin suggested splitting the working group into pairs, each owning one bullet point. His boss liked the idea so much that he set up a new working group to plan the split. It has nine members.
What this can't tell you
- The figures come from rope pulling. Knowledge work isn't a rope; the size of the effect varies a lot between tasks.
- Old data, retold. Ringelmann's numbers are known mostly through later summaries, so treat the exact percentages as approximate.
- Groups can also help. When contributions are visible and the task is meaningful, people sometimes work harder together.
The working group, Kevin's boss and Darren are an illustration. Ringelmann's rope-pulling figures are as reported by Kravitz and Martin in the Journal of Personality and Social Psychology (1986); the fake-teammate experiment is Ingham, Levinger, Graves and Peckham (1974); social loafing is from Latané, Williams and Harkins (1979). Educational content, not business or financial advice.