car crash into home
© AP/REX/Shutterstock
In 1969, a car came off the road and landed in the front room of Mr and Mrs Striffolina’s New York home
IN 1965, a pair of psychologists from the University of Washington handed a questionnaire to 50 carefully selected motorists in the Seattle area. It focused on driving skills, but Caroline Preston and Stanley Harris weren't trying to find out how good the drivers were. They already had a pretty clear handle on that. They wanted to know how good the drivers thought they were.

The questionnaire was straightforward. It asked the drivers to rate their abilities from 0 to 9, with 0 being "very poor" and 9 being "expert". Preston and Harris probably expected the drivers to rank themselves nearer to zero than to 9. To their surprise, they found the exact opposite.

Given who these drivers were, that was very, very odd.

Back in the 1960s, traffic fatalities were a growing problem in the US. Around 36,000 people died in 1960, 39,000 in 1962 and 46,000 in 1964. Road crashes were the leading cause of death in children and young adults - and were costing a fortune.

A good deal of research into their causes was being done, mainly on vehicle design and traffic engineering. But a few researchers were becoming interested in the psychology and behaviour of drivers. That is what attracted Preston to the problem. She may have been seeking to discover some psychological trait that could be used to reduce the accident rate, but instead she inadvertently began a revolution in our understanding of the human mind that continues to unfold more than half a century later.

The 50 drivers were not run-of-the-mill motorists. All were chosen because they had recently been involved in an accident. And not just "involved" - they were behind the wheel at the time.

Their crashes were not trivial prangs or scrapes. They were serious enough to land them in hospital. "Multiple injuries... were the rule", the psychologists noted. In 28 cases, the driver had written off their own car, six had written off somebody else's vehicle and three had actually killed somebody. Preston and Harris often had to wait for the drivers to regain consciousness before asking them any questions.


Once cleared to get to work, the psychologists interviewed the subjects about their driving history and their accident, ran personality and attitude tests, and made discreet enquiries about their drinking habits. Then they asked about driving ability.

Nothing really stood out, except for the drivers' self-belief. Despite being in hospital with self-inflicted injuries, all 50 judged their driving ability to be above average - "much closer to the expert than to the poor driver area of the continuum".

Not only that, when asked how they were driving at the time of the crash, most said they were doing just fine, thank you. A few admitted to "less than usual driving efficiency" and 21 to drinking some alcohol before their crash, but 34 said their driving was "normal", "good", "100%" or even "extra good". And they largely denied any responsibility for the accident. Only 15 admitted that they were directly to blame, even though the police reports confirmed that 34 of them were.

When Preston and Harris dug into the drivers' pasts they found a trove of guilty secrets. Six had been banned from driving at some point; 11 had failed a driving test; 29 had citations for two or more traffic violations; and 11 had been prosecuted for negligent, reckless or drunk driving.

"The omens... had been gathering for some time," Preston and Harris observed drily.

Clearly these drivers were not among the elite. Many were incompetent, and had ample evidence of this available to them. And yet most still thought they were quite something. What planet were they on?

It turned out they were on exactly the same planet as everyone else.

As part of the same study, Preston and Harris recruited 50 control drivers who were similar in every way except for not having had an accident. If they anticipated finding differences, they were largely disappointed. On almost every score, the two groups were indistinguishable, including the 0-to-9 scale.

For the apparently safe drivers to have rated themselves among the elite might have been expected. But for all 100 subjects to do so - including 50 who clearly were not - defied statistical plausibility. Not everyone can be better than average. What was going on?

The research was published in the Journal of Applied Psychology (vol 49, p 284), and then ignored. That was probably no surprise. Preston and Harris had set out to discover what made drivers accident-prone, but found almost nothing. In other words, they had produced the kind of null result that often gets forgotten about.

On top of that, it went against the grain of psychology at the time. For most of the 20th century, having a realistic view of the world and oneself was considered to be the bedrock of good psychological health. For example, Abraham Maslow, the acknowledged expert on mental well-being, wrote in 1950 that "healthy individuals find it possible to accept themselves and their own nature without chagrin or complaint". What Preston and Harris discovered would have been considered delusional.

Preston soon moved on to other questions, becoming a distinguished gerontologist. She remained at the University of Washington for the rest of her long career and retired in 1984. Harris seems to have disappeared off the scene altogether.

But even as their research was fading into obscurity, other psychologists started to discover similar anomalies. Business leaders and management students were found to hold unrealistically positive views of their own competence. College professors frequently believed they were better-than-average teachers. It was the same wherever they looked - a phenomenon that became known as the better-than-average effect.

Unrealisitic positivity is now considered a fundamental feature of human nature. Almost everybody inflates their self-worth and downplays or ignores their faults. People often take credit for their successes while blaming failure on external forces, and expect the future to be rosier than their past. Ironically, people also claim that they are less likely than average to hold an inflated opinion of themselves.

Far from being delusional, these beliefs are seen as a hallmark of good mental health. They enable us to be happy and contented, care for others and engage in productive work. People who don't hold them are called " depressive realists".

The better-than-average effect remains the quintessential example of this tendency. "Arguably, it is the most well-used and best-validated index of self-enhancement," says Constantine Sedikides at the University of Southampton, UK, a leading researcher in what has become one of the most productive and surprising areas of psychology.

Preston and Harris have never received the recognition they deserve for discovering the better-than-average effect. Their jaw-dropping research is sometimes cited but remains little more than a cult classic. But we shouldn't be too concerned for them: they probably felt pretty good about themselves regardless.

This article appeared in print under the headline "Crash Test Dummies"