UPDATE: Diederik Stapel, who led this study, has been accused of fabricating data and has been suspended from his post. It is not clear which of his papers are at stake, but until further details emerge, it would probably be best to take this paper and post with a pinch of salt.
UPDATE 2: This paper has now been officially retracted. As is this post.
In February 2010, cleaners working at Dutch railway stations went on strike for several weeks. Their stations quickly fell to dirtiness and disarray, but most people didn’t mind; public support for the strike was high. But two scientists – Diederik Stapel and Siegwart Lindenberg from Tilburg University – were particularly delighted. In the growing chaos of the stations, they saw an opportunity to test an intriguing concept – that disorderly environments promote stereotypes and discrimination.
Their big idea is that stereotypes, being a set of simplified categories and judgements, can help people to cope with chaos. They are “a mental cleaning device in the face of disorder”. When our surroundings are full of chaos – be it dirt or uncertainty – we react by seeking order, structure and predictability. Stereotypes, for all their problems, satisfy that need.
To test that, the duo went to Utrecht station after it hadn’t been cleaned for a few days and asked 40 travellers to fill in a questionnaire. Their task was to say how much Dutch, Muslim and homosexual people conform to different personality traits. When the cleaners returned to work, and the station had reverted to its usual spick self, Stapel and Lindenberg repeated their experiment.
They found that the volunteers held more strongly to stereotyped views when they sat in a dirty station, compared to a clean one. For example, they were more likely to rate Muslims as being ‘loyal’ and ‘aggressive’, gay people as ‘sweet’ and ‘feminine’, and Dutch people as ‘tolerant’ and ‘stingy’. The moods of the different volunteers didn’t differ between the two days, and didn’t affect their behaviour.
Of course, the volunteers could have lied on the questionnaires. But unbeknownst to them, there was a secret element to the study, designed to reveal their true opinions. Stapel and Lindenberg casually invited them to sit on a row of six chairs while they filled in their form. The chairs were empty except for the first one, which was taken by either a white or a black associate (deemed to be equally friendly, attractive and approachable in pre-tests).
On average, the volunteers (all white) sat three chairs away from the black person in a dirty station, and two chairs away from him in a clean station. In both situations, they sat two chairs away from the white person. In the disordered environment, they were more likely to distance themselves from people with a different ethnic background.
This is not the first study to link chaotic environments with bad behaviour. In 2008, I wrote about the work of another Dutch scientist called Kees Keizer, who showed that litter, graffiti and discarded shopping trolleys can increase the likelihood of more littering, trespassing and even theft. It was dramatic confirmation of the ‘Broken Windows Theory’, which suggests that signs of petty crimes, like broken windows, can trigger yet more criminal behaviour. Disorder breeds disorder. Stapel and Lindenberg have extended these results to stereotypes and discrimination.
Their railway station experiment was just the first in a series of five. Next, Stapel and Lindenberg went to a wealthy Dutch neighbourhood and subtly altered the environment, misplacing some pavement tiles, parking a car on the pavement and abandoning a bicycle in the street. They gave 47 volunteers five euros each to fill in the same questionnaire from before; afterwards, they could donate some of that money to a charity called Money for Minorities. A day later, the duo ran the experiment again but with the tiles, car and bicycle arranged in neat and orderly places.
The results mirrored those of the station experiment. When the environment was unkempt, the volunteers expressed more stereotyped views on the questionnaire and they gave less to the charity – €1.70 compared to €2.35 when everything was neat.
Are these differences really down to a need for order and structure? To find out, Stapel and Lindenberg headed back to their laboratory for three more experiments.
They found that people who saw messy pictures, such as bookcases with chaotic stacks, were more likely to cite stereotypes than those who saw orderly pictures (a neatly stacked bookcase) or neutral ones (a chair). The 47 volunteers in this experiment also filled in a different questionnaire designed to measure their need for structure – it asked them how far they agree with statements like “I don’t like situations that are uncertain” or “I need structure”. Those who had the strongest need for structure also made the most stereotyped judgements. Indeed, after adjusting for this need, the link between disorder and stereotypes disappeared.
This worked even if the volunteers weren’t aware that they had seen signs of disorder! In a fourth experiment, Stapel and Lindenberg recruited 58 volunteers and flashed different words at the side of their field of vision. They couldn’t consciously read the words, but they registered them nonetheless. People who saw words like ‘chaos’, ‘anarchy’ and ‘mess’ expressed a stronger desire for structure, and more stereotyped views, than those who saw orderly words like ‘structure’, ‘clarity’ and ‘neat’, or neutral ones like ‘chair’, ‘table’ or ‘ball’.
In a final experiment, Stapel and Lindenberg showed that even abstract signs of disorder could trigger stereotypes. They showed 66 volunteers a sheet of paper with circles, squares and triangles, either neatly arranged or randomly strewn. And as predicted, those that saw the chaotic symbols were more likely to crave structure and show stereotypes.
Together, the five experiments make a compelling package. The results were consistent across five different ways of triggering perceptions of disorder and different measures to judge the volunteers’ reactions. In the duo’s own words:
“The message for policy-makers is clear: One way to fight unwanted stereotyping and discrimination is to diagnose environmental disorder early and to intervene immediately by cleaning up and creating physical order. Signs of disorder such as broken windows, graffiti, and scattered litter will not only increase antisocial behaviour, they will also automatically lead to stereotyping and discrimination. Investing in repair and renovation, and preventing neighborhoods [from falling] into disarray, may be relatively inexpensive and effective ways reduce stereotyping and discrimination.”
Reference: Stapel & Lindenberg. 2011. Coping with Chaos: How Disordered Contexts Promote Stereotyping and Discrimination. Science http://dx.doi.org/10.1126/science.1201068
More on stereotypes:
- No love for outsiders – oxytocin boosts favouritism towards our own ethnic or cultural group
- Fake CVs reveal discrimination against Muslims in French job market
- Women make safer financial decisions when faced with sexual stereotypes
- 15-minute writing exercise closes the gender gap in university-level physics
- Williams syndrome children show no racial stereotypes or social fear