The Big Mistake

What it is

Cialdini treats the error in Influence, as a failure of public communication rather than a failure of the audience. His example comes from his own research programme. Visitors to Petrified Forest National Park in Arizona were taking pieces of petrified wood, and signs were posted to stop them. The 2006 paper in Social Influence, by Robert Cialdini and five co-authors, reports that the signs carried either a descriptive norm, how many others engage in the behaviour, or an injunctive norm, how many others disapprove of it. Its abstract states that descriptive messages were most likely to increase theft and injunctive messages most likely to reduce it.

The figures Cialdini prints in Influence come from alternating two signs in high-theft areas. With no sign, theft ran at 2.95 per cent of visitors. A sign picturing three thieves, a picture of theft as common, was followed by theft at 7.92 per cent, which he describes as nearly tripling it. A sign picturing a single thief was followed by 1.67 per cent. These percentages are taken here from the book and were not checked against the paper's full text.

The label itself belongs to the book. The underlying distinction, between what people do and what they approve of, is older and is covered under Descriptive and Injunctive Norms.

In effect

The mechanism is the one that makes Social Proof work for a seller, run in the wrong direction. A hotel card saying that most guests reuse their towels uses a count to tell guests what is normal, and the count raised reuse in the 2008 field experiment by Noah Goldstein, Cialdini and Vladas Griskevicius. A park sign saying that many visitors take wood uses a count in the same way, and the thing it makes normal is theft.

A second field result shows the same logic acting on people who were already behaving well. In the 2007 energy study by P. Wesley Schultz and colleagues, a message reporting average neighbourhood use pulled heavy users down and pushed light users up, the second being the boomerang effect, and adding a line conveying approval removed it. The average works as a target as well as a description.

Cialdini's defence is that persuasive communications should avoid employing information that can normalise undesirable conduct. He adds that park officials missed a positive option, which was to harness true social proof on behalf of park goals by labelling honourable guests as the great majority.

The closing chapters add a refinement and a self-correction. He writes that his earlier advice to steer clear of social-proof evidence when it is not fully present is mistaken. The replacement is future social proof, which describes a trend instead of a current majority. In the water-conservation study he took part in, students told that a minority of their peers conserved water used the most water of any group at a secretly metered sink, and students told that the minority was growing over two years reversed that. The book does not print percentages for it in the text read here.

What it does not say

It does not say that reporting a common behaviour always backfires. In hotels, descriptive-norm cards raised towel reuse, and Bohner and Schlüter's 2014 study of 928 observations in two German hotels found norm messages beat a no-message baseline, though they did not beat the standard environmental appeal. The Big Mistake concerns messages whose subject is an unwanted behaviour.

It is also a finding reported mainly by its originator. The Petrified Forest work and the water study are Cialdini's own, the percentages are taken from his book, and the vault has not read the 2006 paper beyond its abstract. The water study's key figures are not available here, so its direction is reported and its size is not.

Nothing here supports a rule that bad news should never be published. The paper's abstract offers recommendations for using normative messages where undesirable behaviour is already widespread, and the claim that survives is narrower: the form of the message decides what it makes normal.

---