Social-norms campaigns are intended to mitigate problem behaviors by conveying the message that problem behavior occurs with far less frequency than people think (e.g. teenage drinking). But for individuals who already abstain from the undesirable behavior, this can actually produce a boomerang effect (similar to the effect we discussed for the Crying Indian PSA) where people see that others are behaving in a certain way and actually do more of the undesired behavior. In other words, there is a tendency toward normalization.
In social messaging, therefore, it is important to build in something to convey the perception of what is commonly approved or disapproved of within the culture (i.e. not simply what is normal but what society thinks should be normal). This is called the injunctive norm.
A team of psychologists tested the injunctive norm in a study on energy use in 290 households in San Marcos, California. They gave half of those households a series of messages containing information on their home's energy and how that energy use compared to their neighbors along with information on how to conserve energy. The other half they sent the same information plus an injunctive message: a smiley face if the household consumed less than average and a frown if it consumed more. Researchers hypothesized that the emoticon would prevent households who consumed less from consuming more when they saw they were below average (the undesirable boomerang effect) and lead to greater overall energy savings. They were right.
With just the descriptive message about energy use relative to their neighbors, short-term energy use (the 3-week period following the final message) fell by 1.22kWh/day (so it did have a positive effect on energy conservation). But the households who received the injunctive message (the emoticon that signaled how their energy use was perceived) reduced their energy consumption by 1.72kWh/day. An emoticon reduced energy use by an additional 40 percent and saved 0.50 additional kWh/day. Maybe it would be best if all behavior came with smiley or frown faces...
Schultz, P.W., J.M. Nolan, R.B. Cialdini, N.J. Goldstein, V. Griskevicius. 2007. The constructive, destructive, and reconstructive power of social norms. Psychological Science 18(5): 429-434.
"...problem behavior occurs with far less frequency than people thing ..."
Do you mean "people think"?!
Thanks for the blog. Keep up the great work!
Thanks for the catch, Ian! Corrected it...
Interesting... though I wonder how much of the effect can be explained by the uber-simple visual summary that the emoticon provided?
I can't think of an easy way to control well for that. Most very simple visual signals have a good/bad connotation. A partial control might be the emoticons vs something like an up or down arrow (for higher or lower)... though down=bad, though maybe less strongly than frown=bad.
Apologies if they actually did this sort of thing. I'm away from journal access at the moment.
Nice job. This is one of the clearest overviews of that research that I have read. - Ryan