Study finds honesty varies significantly between countries

November 16, 2015
Study finds honesty varies significantly between countries

Research from the University of East Anglia (UEA) has found that people's honesty varies significantly between countries.

It also suggests that is less important to a country's current economic growth than during earlier periods in history.

The study examined whether people from different countries were more or less honest and how this related to a country's economic development. More than 1500 participants from 15 countries took part in an online survey involving two incentivised experiments, designed to measure honest behaviour.

Firstly, they were asked to flip a coin and state whether it landed on 'heads' or 'tails'. They knew if they reported that it landed on heads, they would be rewarded with $3 or $5. If the proportion reporting heads was more than 50 per cent in a given country, this indicated that people were being dishonest.

The same participants were then asked to complete a music quiz where they were again rewarded financially if they answered all questions correctly. They were asked not to search for the answers on the internet, and had to tick a box confirming they had answered on their own before moving on to the next question. Three of the questions were deliberately difficult so it would be highly likely that participants would need to look up the answer—getting more than one of these questions right indicated cheating.

Data from the tests was compared to estimate whether people from particular countries were more likely to tell the truth.

The countries studied—Brazil, China, Greece, Japan, Russia, Switzerland, Turkey, the United States, Argentina, Denmark, the United Kingdom, India, Portugal, South Africa, and South Korea—were chosen to provide a mix of regions, levels of development and levels of social trust.

The study's author Dr David Hugh-Jones, of UEA's School of Economics, found evidence for dishonesty in all the countries, but that levels varied significantly across them. For example, estimated dishonesty in the ranged from 3.4 per cent in the UK to 70 per cent in China. In the quiz, respondents in Japan were the most honest, followed by the UK, while those in Turkey were the least truthful.

Participants were also asked to predict the average honesty of those from other countries by guessing how many respondents out of 100 from a particular country would report heads in the coin flip test. However, participants' beliefs about other countries' honesty did not reflect reality.

Dr Hugh-Jones, a senior lecturer in economics, will present his findings today at the London Experimental Workshop conference, hosted by Middlesex University London. "Differences in honesty were found between countries, but this did not necessarily correspond to what people expected," he said. "Beliefs about honesty seem to be driven by psychological features, such as self-projection. Surprisingly, people were more pessimistic about the honesty of people in their own country than of people in other countries. One explanation for this could be that people are more exposed to news stories about dishonesty taking place in their own country than in others."

In the coin flip test, the four least honest countries were China, Japan, South Korea and India. However, Asian countries were not significantly more dishonest than others in the quiz, where Japan had the lowest level of dishonesty. Dr Hugh-Jones said the difference between Asian and other countries in the coin flip may be explained by cultural views specific to this type of test, such as attitudes to gambling, rather than differences in honesty as such.

People expected Greece to be the least honest country but in the coin flip it was one of the most honest, while in the quiz it ranked in the middle. Of the respondents who expected less honesty in their own country, Greece and China were the most pessimistic. Another finding was that less honest respondents also expected others to be less honest, as, unexpectedly, did those from more honest countries.

Dr Hugh-Jones said there was increasing interest in the cultural and behavioural roots of . He found that while the honesty of countries related to their economic growth—poor countries were less honest than rich ones—this relationship was stronger for growth that took place before 1950.

"I suggest that the relationship between honesty and economic growth has been weaker over the past 60 years and there is little evidence for a link between current growth and honesty," said Dr Hugh-Jones. "One explanation is that when institutions and technology are underdeveloped, honesty is important as a substitute for formal contract enforcement. Countries that develop cultures putting a high value on honesty are able to reap economic gains. Later, this itself improves institutions and technology, making contracts easier to monitor and enforce, so that a culture of honesty is no longer necessary for further growth."

Dr Hugh-Jones added: "People's beliefs about the honesty of their fellow citizens, and those in other countries, may or may not be accurate, and these beliefs can affect how they interact. For example, a country's willingness to support debt bailouts may be affected by stereotypes about people in the needing help. So it is important to understand how these beliefs are formed."

Explore further: In the job hunt, people do lie, but honesty pays off, study finds

More information: 'Honesty and beliefs about honesty in 15 countries' www.uea.ac.uk/documents/3154295/7054672/Honesty+paper/41fecf09-235e-45c1-afc2-b872ea0ac882

Related Stories

The psychology of the job hunt

November 16, 2010

In addition to the traditional application form and interview, job seekers are increasingly asked to take a pre-employment test -- an assessment that employers use to screen candidates. But how easy is it to fake the "right" ...

How honest are you at work?

November 15, 2012

(Medical Xpress)—A new study has revealed we are basically honest. The research by the University of Oxford and the University of Bonn suggests that it pains us to tell lies, particularly when we are in our own homes.

Power can corrupt even the honest

October 1, 2014

When appointing a new leader, selectors base their choice on several factors and typically look for leaders with desirable characteristics such as honesty and trustworthiness. However once leaders are in power, can we trust ...

Recommended for you

18 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

antialias_physorg
4.4 / 5 (7) Nov 16, 2015
Countries that develop cultures putting a high value on honesty are able to reap economic gains

That makes sense to me. If you can expect to trust people you need to waste less of your resources on securing yourself against dishonesty in any one particular deal. In effect it means you can get away with lower "insurance premiums".

"I suggest that the relationship between honesty and economic growth has been weaker over the past 60 years and there is little evidence for a link between current growth and honesty,"

The world has become much more a global village in those past 60 years. Before that the effect of honesty/dishonesty of a people was mostly confined to within that country. Today (dis)honest people can do dealings abroad as easily as they can do at home. And when a dishonest country can deal with a honest one then it's big-con time.
BartV
1.3 / 5 (14) Nov 16, 2015
In the US, they should have broken down the group to include D's and R's. And I think they would have found D's results to be similar to China's.

antialias_physorg
4.7 / 5 (12) Nov 16, 2015
In the US, they should have broken down the group to include D's and R's. And I think they would have found D's results to be similar to China's.


If the D's numbers are like Chinas then the US is in deep trouble (because R's are three times worse accprding to a study by George Mason University: http://www.mediai...w-study/
Eikka
2.1 / 5 (7) Nov 16, 2015
$3 or $5 to a Brit is a lot less than the same amount to a Chinese, which makes for a greater incentive to cheat

antialias_physorg
4.3 / 5 (6) Nov 16, 2015
$3 or $5 to a Brit is a lot less than the same amount to a Chinese, which makes for a greater incentive to cheat

If you read the study (linked at the bottom of the artice) you will find that the author accounted for this via testing whether there was a difference when high or low incentive was offered (there was none)
jljenkins
3.7 / 5 (6) Nov 16, 2015
Age wasn't an independent variable??? That has to be generating a lot of error variance.

@BartV What kind of mindless parrot has to make the same kind of parochial cut-and-paste trolling remark on every damned subject? Right. A stupid American.
jim_xanara
1 / 5 (3) Nov 16, 2015
What about population density? Do you think that if you did that with a random passer-by in New York or London, compared to a rural village you wouldn't find less honesty in the big cities? China and India. Right. Asia is the first variable that springs to mind. Not. And they say as much. Now we're asking some hard questions. Were the researches incompetent to not covary the obvious covariates mentioned, or did they mit them to make an a priori hypothesis? I'll bet every coin in my pockets that if you covaried population density, without including the other things mentioned, you'd have to accept the null hypothesis, that there is no difference in any of the populations.

That's not even addressing dilusion. Is Netanyahu lying when he revises the Holocaust, or deluded? What about Ben Carson on Egyptian pyramids? As they mention, tossing a coin is a POOR test as it brings delusions beliefs about luck into play.
Eikka
3.4 / 5 (5) Nov 17, 2015
If you read the study (linked at the bottom of the artice) you will find that the author accounted for this via testing whether there was a difference when high or low incentive was offered (there was none)


That's likely because they didn't account for purchasing power at all. Both the $3 and $5 reward levels can be considered high if you're doing the test on a person who normally earns the equivalent of 50 cents an hour.

However, it is not clear whether and how lying behaviour responds to incentive
size: this topic has been much debated in the literature. To check whether incentive sizes would affect my results, I randomized the size of payments offered. Half the subjects were assigned to the HIGH payment treatment. They were offered $5 for reporting heads in the coin flip and $5 for getting all quiz questions right. The remaining subjects were assigned to the LOW payment treatment, where both payments were $3.


I don't find this sufficient for a control.
Eikka
3.4 / 5 (5) Nov 17, 2015
For example, the average salary in China is 8,844 USD per year, which makes it about $4,60 per hour. Meanwhile the average hourly wage in the US is $25.20

So offering a Chinese person $3 is like offering an American person $16

That's not a trivial amount of money. I bet many Americans would tell a fib for $16.
antialias_physorg
3.7 / 5 (3) Nov 17, 2015
Age wasn't an independent variable??? That has to be generating a lot of error variance.
If age is not correlated with nationality it doesn't affect the outcome.
Jeffhans1
5 / 5 (1) Nov 17, 2015
Did the author consider that the coin flips may not have been statistically even as expected? Large variations have been observed possibly due to technique or type of coin used. Unless the study verified that there were actual lies occurring, it may just be noise. Knowledge of random trivia is also something that can be thrown off by a few lucky people who get questions they know.
my2cts
5 / 5 (1) Nov 18, 2015
Is Netanyahu lying when he revises the Holocaust, or deluded?

Can you back up this casual remark ? If not you are slipping in some form of holocaust denial here. It wouldn't be your first revolting statement here.
If I were a mediator here you would have been banned a long time ago.

antialias_physorg
5 / 5 (2) Nov 18, 2015
Large variations have been observed possibly due to technique or type of coin used.

How would you explain variations that arise for people of one nationality? You think Chinese are better at flipping coins so that what they want comes up than UK citizens? If that were so then there would be all kinds of betting going on in the world and UK citizens would consistently lose while Chinese nationals would consistently win. That is not observed.

Unless the study verified that there were actual lies occurring, it may just be noise

As with all probability experiments you can calculate the likelyhood of a given outcome being a statistical fluke (that's why statistical significance is used as a threshold in the first place)

Knowledge of random trivia is also something that can be thrown off by a few lucky people who get questions they know.

That is why your sample size must be large (read: your study must have statistical power)
nitin_gupt_a
not rated yet Nov 18, 2015
Seems more like noise, its avg 100 people per country and that too were divided among mix of regions, levels of development and levels of social trust. ... that is too small a sample size to judge or conclude anything, that too with the guess type of questions.

Were they noticing the real outcome of the coin, to make sure that the respondents are lying? If not then the margin of error is too high to to say anything meaningful from the outcome.
In quiz type it may happen that everyone is searching about it on internet and if the music belongs to their country its easy to know about it if not then its difficult. Also just three questions, if randomly guessed makes it equal chances to get two rights or two wrongs

so, if you repeat this type of experiments you will get random output each time. Did they repeat these experiments few times and got the same result?

And their own result shows that, Japan became most honest in one of the quiz and in other, one of the least honest.
antialias_physorg
5 / 5 (1) Nov 18, 2015
Seems more like noise, its avg 100 people per country and that too were divided among mix of regions, levels of development and levels of social trust.

Not neccesarily. If your statistical sample is small then you need smaller standard deviation/larger expectation value difference to reject the null hypothesis. That's all part of the math of probability distributions. Just a gut feeling that it may be a fluke or it may be correct isn't enough. You need to show that the math is wrong.
With a typical p value of 0.05 (or less) for these kinds of experiments you do get about 1 in 20 false positives IF all 20 experiments are juuuust at the border of statistical significance. That is the accepted criterium for reporting significance in the more 'soft' sciences (psychology, medicine, .. ). In the hard sciences that's a lot more stringent as experiments happen in a muchh more controlled environment (c.f. Higgs boson discovery).
antialias_physorg
5 / 5 (1) Nov 18, 2015
In the end - if there is no inherent problem with the method used (and there isn't one that I can see here) - it comes down to this: Do you think 1/20 failure rate (as a HIGH bound) is an acceptable scientific MO?
Because if you accept this then you don't get to pick and choose which studies you think are a fluke and which aren't - just because the results aren't to your liking.
Ryan1981
not rated yet Nov 19, 2015
If the proportion reporting heads was more than 50 per cent in a given country, this indicated that people were being dishonest


The first thing I was thinking (I might be wrong so correct me if you like)

Though the law of large numbers states the percentage to even out to 50/50 this does not mean it will always be 50%. Some people might just have been really lucky! Some people might have gotten a coin that is not perfect and tends to give heads more than tails. It is always tricky to base your results on "random" events. I would propose a scenario where you allow the person to get to a target as fast as they can by both being allowed to use the dishonest way or the honest way or something.
antialias_physorg
5 / 5 (2) Nov 19, 2015
Though the law of large numbers states the percentage to even out to 50/50 this does not mean it will always be 50%.

That's why studies like this (or more precisely: almost all scientific studies in all scientific fields) use p vales to show significance. I.e. stuff is only deemed significant if the measured value is x times the standard deviation off the ideal average.

Example: 1000 coin tosses. 490-510 'heads' is not unlikely. Such an outcome supports 'coin is fair' (i.e. the result is not significant for a 'biased coin' hypothesis). Getting only 0-20 heads on 1000 throws is very unlikely . Such a coin is very likely biased.

The more standard deviation off it is the less likely that it's just 'blind luck'. The accepted value for such studies is 5% likely that the observed deviation is a pure chance event and not caused by an actual effect (in this case: more/less dishonesty than average in the persons being tested).

Statistics are a huge part of science.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.