Predictably Irrational Chapter 11 & 12 – The Context Of Our Character

Chapters 11 and 12 of “Predictably Irrational” by Dan Ariely are all about dishonesty. He discusses the decline in ethics which is relevant at a time of corporate scandals and financial scams. He examines and measures the likelihood of people cheating for financial gain and ways to keep them honest. The level of integrity his experiments reveal is particularly interesting. He also finds how people differentiate between cheating, being more likely to cheat if non-monetary items are involved.

Ariely begins by painting a picture of the disturbing level of dishonesty in society. He doesn’t just refer to the theft and robberies and that make the news but also white collar crime. All robberies in the United States costs an estimated $525 million per year but employee theft costs an estimated $600 billion. Tax fraud costs an estimated $350 billion dollars. And this is only the dishonesty we can put a number on. There are also politicians who accept donations and gifts from lobbyists in return for favours. There are the doctors receiving money to promote a certain brand of medicine even if it isn’t the best for the patient. Surveys of lawyers found most believe the law profession suffers from a lack of ethics. Even geologists are accused of falsifying and exaggerating data. All of which is to say nothing of the financial industry.

So Ariely did an experiment to measure levels of dishonesty and find ways to limit it. A group of Harvard undergraduates and MBA students (future captains of industry) would be given 50 standard multiple choice questions. They would write their answers on a worksheet before transferring it over to a score sheet at the end of the test. They would then be paid 10 cent per right answer. Ariely took a second group and did the same thing, except this time their score was pre-marked with the right answers. Therefore they could choose the correct answer before handing their test up. Would they cheat? To further test the hypothesis, Ariely took a third group which was given the answers but told didn’t have to hand up their worksheet, therefore there was no way of telling if they had got the right answer or cheated. Ariely decided to push students integrity to the limit by taking a fourth group who didn’t have to hand any answer sheet up but instead simply pulled as many coins as they felt they earned from a jar and left. Did the best and brightest of America’s elite cheat?

The first group, who were the control and had no opportunity to cheat, got an average of 32.6 questions out of 50 right. The second group (who had to hand up their worksheet and score sheet but were given the answers) got an average of 36.2 answers right. The third group who got to keep their worksheet (and therefore any proof they got any questions wrong) got 35.9. The final group (who did hand any answers up but helped themselves to the money) got an average of 36.1.

There are three interesting points from this. First of all, it shows that if given an opportunity people will cheat (not much of a surprise there). However, what is more important is how little they cheat. Those students who had the answers could have answered correctly for every question without punishment, yet they only slightly increased their score. The final group could have stuffed their pockets with cash yet despite the much higher incentive to cheat, they were as (dis) honest as the other group. Finally people seem to either cheat or not, there were no levels or degrees and the risk of getting caught had no effect on the level of cheating.

Ariely did another experiment but with a novel twist. This time there were three groups, the control group and two groups given the opportunity to cheat. What was particularly interesting was that of the cheat groups, the first was asked to list 10 books and the second was asked to list the 10 Commandments. The control group answered 3.1 questions correctly, the cheat group got 4 questions correct, but the cheat group who were asked to recall the 10 Commandments got 3 questions right, in other words they didn’t cheat at all.

I know my religious readers will get quite excited about this and use it to argue that religion should be brought back into public life. But let me explain why this is jumping the gun. Hardly any of the students could recall all 10 commandments and those who could remember only a handful were just as honest. Ariely shows  that it was not the use of religion that influenced people but rather the use of a moral code and a reminder to be honest. He repeated the test but instead of asking the third group to recite the 10 Commandments, they had to sign a form stating “I understand that this study falls under the MIT honor system”. The control group got three answers correct, the cheat group claimed 5 correct answers and the honour group who had the opportunity to cheat, claimed 3 correct answers. This especially interesting as MIT doesn’t have an honour code. This shows that the mere reminder of a duty to be honest, makes people act more honest.

There is a second interesting point about cheating that Ariely discusses in Chapter 12, that is that people are less likely to cheat when it involves money but more likely if it involves goods. For example, Ariely left six packs of Coke cans in communal fridges around MIT. Within 72 hours, every single one was gone. He then repeated the test but instead of six packs, he left a tray with six one dollar bills (in other words the monetary value of a can of coke). After 72 hours, not a single dollar was gone.

Ariely ran further experiments similar to those above, involving a control group and two cheat groups. As usual the control group couldn’t cheat, the second group could cheat and get paid based on their results. The third group was the same except, instead of getting paid in cash, they were given tokens and then walk to another desk and receive cash (this was an attempt to separate monetary and non-monetary cheating). The control group got 3.5 questions correct and the cheat group claimed they got 6.2 questions correct. However, the token cheat group claimed a whopping 9.4 correct answers. Even if something is one remove from cash, the likelihood of cheating skyrockets.

All of this is important in understanding crime and why white collar crime is such a great problem. By fiddling with expense account, you’re not dealing in money or people, but rather numbers on a page. It seems like a victimless crime. Likewise people who don’t pay all their taxes, aren’t dealing in cash or physically taking it out of people’s pockets, so it doesn’t seem like stealing. They don’t see the underfunded schools or hospitals or the victims of their theft. This is why muggings capture the public’s attention, despite being much more minor than white collar crime which is a silent crime. Ariely shows the value of a moral code to keep people honest. What is interesting about his studies, is not just that people cheat but how little they actually do.


Filed under Books, Economics

2 responses to “Predictably Irrational Chapter 11 & 12 – The Context Of Our Character

  1. Nice recap. An important open question, then, is why do people cheat as “little” as they do? One possible explanation is that they don’t want to get caught and believe that there are statistical methods for detecting cheating as well as the usual other methods (cameras, witnesses, getting caught with the cheat sheet, talking to other students, etc.). In other words, they cheat enough to achieve a meaningful advantage that is, at the same time, not preemptive enough to point to their cheating. Since my grade on this test is pass/fail, I don’t have to ace this test, so I’ll cheat just enough to pass. It’s rather like the student who fails the state comprehensive achievement test in the 4th grade and the eighth grade, then, passes it with flying colors in the tenth grade without corroborating evidence of academic achievement or change of attitude toward it. Corroborating evidence (such as a large surfeit of erasures, as determined by statistical methods such as hypothesis testing) is often cited, but, none of this is accepted as conclusive, when it should be. So, the very savvy cheat cheats enough to reach his objective and no more. I’m not surprised at this result.

    As a former direct participant in financial markets, I could supply anecdote upon anecdote of cheating that isn’t considered cheating. Rather, it’s considered as “obtaining a competitive advantage”. Arielly’s work and that of others in this area is crucial in determining the psychological basis for economic theories that involve assumptions about how incentives affect behavior. My father, a lifelong consumer finance lender to high risk individuals told me in 2009 (at the age of 94) that the financial meltdown was due to, “in two words: loose credit”. This judgment amounts to the claim that lenders and market under priced risk, which we know to be the case. In fact, as I recall, there is a great deal of recent research activity into this phenomenon (that people habitually, always, under price, under estimate, and undervalue risk). We do that even as we know we do it. We must figure out how to manage this destructive tendency.

    Btw, Prof. Arielly is offering a course to begin in March, 2012 in this subject that is open to everyone; it’s a MOOC via Coursera. My experience with Coursera’s offerings is that they are well produced, well managed and the professors are first rate.

    • The way I read these results is that people either cheat or they don’t, there doesn’t seem to be any level of proportionality. Once they have gotten over the mental block in their mind, the level of risk doesn’t seem to affect them. However, I still agree with you that they internalise the risk and don’t give themselves 100% for fear of raising suspicion. They may also rationalise their actions by claiming they aren’t really cheating but giving themselves the answers to questions they should have got right or really knew the answer to.

      By a funny coincidence, I came across Coursera yesterday and was surprised to find Ariley’s course. I am in my final semester of university so I wouldn’t have the time for it, but perhaps over the summer it would be something to look into.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s