Questionnaire Extremism and National Character

By Neuroskeptic | August 11, 2012 11:32 am

“Personality differences” between people from different countries may just be a reflection of cultural differences in the use of ‘extreme’ language to describe people.

That’s according to a very important paper just out from an international team led by Estonia’s René Mõttus.

There’s a write up of the study here. In a nutshell, they took 3,000 people from 22 places and asked them to rate the personality of 30 fictional people based on brief descriptions (which were the same, but translated into the local language). Ratings were on a 1 to 5 scale.

It turned out that some populations handed out more of the extreme 1 or 5 responses. Hong Kong, South Korea and Germany tended to give middle of the road 2, 3 and 4 ratings, while Poland, Burkina Faso and people from Changchun in China were much more fond of 1s and 5s.

The characters they were rating were the same in all cases, remember.

Crucially, when the participants rated themselves on the same personality traits, they tended to follow the same pattern. Koreans rated themselves to have more moderate personality traits, compared to Burkinabés who described themselves in stronger tones.

Whether this is a cultural difference or a linguistic one is perhaps debatable; it might be a sign that it is not easy to translate English-language personality words into certain languages without changing how ‘strong’ they sound. However, either way, it’s a serious problem for psychologists interested in cross-cultural studies.

I’ve long suspected that something like this might lie behind the very large differences in reported rates of mental illness across countries. Studies have found that about 3 times as many people in the USA report symptoms of mental illness compared to people in Spain, yet the suicide rate is almost the same, which is odd because mental illness is strongly associated with suicide.

One explanation would be that some cultures are more likely to report ‘higher than normal’ levels of distress, anxiety – a bit like how some make more extreme judgements of personality.

So it would be very interesting to check this by comparing the results of this paper to the international mental illness studies. Unfortunately, the countries sampled don’t overlap enough to do this yet (as far as I can see).

ResearchBlogging.orgMõttus R, et al (2012). The Effect of Response Style on Self-Reported Conscientiousness Across 20 Countries. Personality and Social Psychology Bulletin PMID: 22745332

  • petrossa

    “yet the suicide rate is almost the same”

    Should read: “yet the REPORTED suicide rate is almost the same”

    Especially in strongly christian religious regions suicide tends to be covered up so they can be buried in hallowed ground.

    The big problem with all national statistics, they reflect prevalent religious and or political tendencies rather then reality.

    The most obvious example is crime. No politician wants high crime rates. So by adjusting what counts as crime you can regulate crime rates.

    Same for unemployment, what constitutes unemployment? Make it unemployed from 1 month to 6 months out of work and the statistics show a significant drop.

    Actually this is all across this paper. The same kind of assumption of accuracy of information. Unfortunately reality isn't in the numbers.

  • Nitpicker

    I'm afraid I haven't had a chance to read the paper so bear with me if this was addressed. But couldn't this be a more general trait rather than being related to linguistics? What I mean is, the tendency to rate extremes rather than middle-ground may also show up in much simpler tasks, such as confidence ratings on perceptual decisions or any sort of questionnaire that requires ratings.

  • Neuroskeptic

    Good point and we know suicide misclassifications do happen but they can't obviously explain the data.

    E.g. Spain and Italy are Catholic countries where suicide is fairly taboo. So maybe their suicide rate is in reality higher than it seems.

    But that would actually make the correlation worse, because they have low reported mental illness rates.

    Brazil, Columbia and Mexico on the other hand have extremely low suicide rates despite high-ish “mental illness rates” so in those cases, if the true suicide rate is higher, it would improve the correlations.

    But they are the only countries where that picture seems plausible.

    (all data from here).

  • Neuroskeptic

    P.S. By “the correlation” I mean the hypothetical correlation, which doesn't exist, between national suicide rates & national estimates of mental illness.

  • pyl

    It could not be otherwise since things such as levels of distress, anxiety, etc., are imaginary. Only if talking about real things we coul pretend objectivity. The consecuence is that as well as 'people', a survey with professionals (judging about their patients) would give the same results.

  • petrossa

    “Because they have low REPORTED mental illness rates”

    Again you fall into the same trap. Health care is expensive. As long is you are not likely to kill someone you'll be just treated by a GP and you are not rated as mentally ill.

    It's patently absurd to assume that mental illness across western europe varies greatly.

    It's not like oriental communities where inbreeding is common.

    Overall this paper and papers like it all suffer from the assumption pitfall. That statistical data is clean.

    For example, as an entrepeneur in the netherlands we have to fill out yearly a 6 doublefaced A4's densely packed complex questions about our business for the central bureau of statistics.
    This a law. If you don't you get fined.

    So what do you do, either you spend precious time filling tedious questions or you just write down whatever seems somewhat appropriate.

    And since practically everyone does that no statistical cleaning is going to clean that mess up.

    Another example. As a physician you 'helped' a terminal patient die.What are you going to write on his death certificate? Cardiac arrest ofcourse. It's true. The heart stopped beating.

    So heart attacks are a major cause of death.

    Again you can't clean that up via statical analyses if it's widely practiced since it's not an outlier.


    All across the board national databases are severely polluted, whichever the data.

  • Neuroskeptic

    Petrossa: That's my point – mental illness reported rates are messy. So are suicide rates, but I think suicide rates are rather less so.

  • petrossa

    Which makes any conclusion drawn from it as valuable as a questimate, NeuroSceptic.

    We here in the computerized world have great faith in numbers, but not too long ago (going back maybe 20 years) even large parts of France was all handwritten records in small tight-knit communities where any entry could find its way into the official records depending on family bonds or the price you were willing to pay.

    Greece, well needs no mention. That's very obvious.

    Spain's hinterland still is mainly in such a state.

    Overall what happens out of sight in the rural communities is hard to get a grasp on.

    And let us not begin about the Eastern block countries.

    So be careful with papers using 'official' numbers. No official number is correct bar none.

  • E

    Interesting, puts the lie to the German national character as a bunch of officious, regimented and stiff necked busy bodies which makes what happend in 1932 and the rise of national socialism all the more inexplicable.

    Why is it so absurd to assume that levels of mental illness vary widely across western europe as indeed they do across the world. Happiness surveys (surely also linked to mental illness) show a direct correlation with per capita income up to a certain level at which point they level off.

    On one such survey Denmark and Switzerland were rated top while the UK came in at No.41 and Zimbabwe second to last at 177. Interestingly USA was rated at 23 while Spain came in at 46 which is perhaps why the suicide rates are roughly the same. Bhutan the only country with a ministry for happiness came in at No8 but expect that to change now they can get Jeremy Clarkson on Sky telling them about the latest super car they can't afford and would have nowhere to drive even if they could.

  • Nitpicker

    @E: I assume your comment about the German national character was in jest but I don't believe you can take any of these findings to generalize to different times. These things are not set in stone and probably change on very short time scales. Considering that “what happened in 1932” is 80 years ago, German character certainly must have changed quite dramatically as a result of the war, the post-war period and 50 years of being a divided nation. Other countries (like China or India) probably changed a lot more even.

  • E

    My comment about the German national character was meant to be ironic but still the stereotype persists. The German national character is still seen today as officious and judgemental by some and what happened in Germany in 1932 could never happen here because we are so, well, so more reasonable and then there is that celebrated British sense of fair play.

    No that could never happen here could it? (irony alert). But then a study like this comes along that seemingly indicates that far from being rigid, reactionary and judgemental the Germans are in fact rather laid back about things.

    As for national characters changing, yes of course they do but Germany in the 1920's and early 1930's was a refined and cultured society that gave us Hegel and Schopenhauer. Even Nietzche, Hitlers favourite philosopher, had to be extensively misinterpreted to fit the Nazi ideal. And it was Heinrich Heine who said.

    “Wherever books are burnt so are people”

    and the Nazi's burn't a lot of books. and people.

  • petrossa


    Ignoring the godwin, you assume a lot.
    First you assume reported happiness is a valid scientific qualifier, then you assume that there is a direct relation between reported mental illness and reported happiness.

    That is a whole lot of if's and but's in a small phrase.

    Reminds my of another nonsense correlation my economics teacher was fond of, graphing shoesize to beer consumption

  • E

    I am not making any comparison with or to Nazis or Hitler merely commenting that a cultured and peaceable people like the German nation can nonetheless be lured into a totalitarian system of government and using the findings of the report cited by Neurosceptic to demolish the stereotype that the Germany was then or is now the sort of nation that is more likely to support a system of Government like National Socialism. So I hardly think Godwins law applies in this case.

    I am not sure what you mean by “a valid scientific qualifier” but what is so outrageous by drawing a correlation between “happiness” on the one hand and mental illness on the other. People with a mental illness are not generally considered by themselves or others to be “happy” . Clinical depression is to a large extent defined by the absence of happiness and before you say it someone in the throes of a manic psychosis can hardly be described as “happy” not in the conventional sense of the word anyway. Not sad either but not happy.

    So perhaps not such a nonsense correlation afte rall?

  • Nitpicker

    @E: totally agreed about the notion that “it can't happen here” (I believe there is actually a dystopic novel or movie with that title that has precisely this topic). It can happen anywhere regardless of what the national character on tests like this shows. And it can probably happen a lot faster than we'd like to believe. This is what makes it so frightening.



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar