Perceptions Are Not Reality: Things are NOT as Bad as they Seem

Ipsos’ latest Perils of Perception survey highlights how wrong the online public across 38 countries are about key global issues and features of the population in their country.

On many subjects – murder rates, terrorist deaths, teenage pregnancy, diabetes and how healthy people feel – things are NOT as bad as they seem!

Some of the key patterns are:

  • Only 7% of people think the murder rate is lower in their country than it was in 2000 – but it is significantly down in most countries, and, across the countries overall, it’s down 29%.
  • Only 19% think deaths from terrorist attacks are lower in the last 15 years than they were in the 15 years before that – when they are significantly down across most of these countries, and overall they are around half the level they were.
  • People hugely overestimate the proportion of prisoners in their countries who are immigrants: the average guess is 28% when it’s actually only 15%.
  • Teen pregnancy is overestimated across the world, often by a staggering amount.  Overall, the average guess is that 20% of teenage girls give birth each year when the reality is 2%.  And some countries guess that around half of teenage girls give birth each year, when the highest actual figure in any country is 6.7%.
  • Six in ten people across the countries are unsure or believe that there is a link between some vaccines and autism in healthy children, despite the claim being widely discredited – only 42% think it is false.
  • Russia is seen as the booziest nation in the world, when they actually only rank 7th. Very few correctly pick out Belgium as the highest alcohol drinking nation in the study.
  • But the USA is correctly seen as having the sweetest tooth, a clear winner, picked well ahead of any other country.
  • People generally overestimate how connected by technology we are, with the average guess across the countries that 75% have a Facebook account when only 46% actually do.

Looking across all 38 countries included, many are wrong…

  1. Murder rate: only 7% of people think the murder rate is lower in their country, but it is significantly down in most countries – and across all countries as a whole, it’s down 29%.  For example, 85% in South Africa the murder rate is higher, when it’s actually down 29%.  And only 8% think it’s lower in Italy, when it’s actually down 39%.
  2. Terrorism: only 19% across the countries as a whole think deaths from terrorist attacks are lower in the last 15 years than they were in the 15 years before that – when across most of these countries, they are significantly down, and overall the number of deaths from terrorism across all these countries has halved.  For example, 60% in Turkey think deaths from terrorism are higher in the last 15 years, when they are around half the level of the previous 15 years.  Some countries are correct, however, with 65% in France correctly believing that deaths from terror attacks are higher.
  3. Foreign born prisoners: most countries greatly overestimate the proportion of prisoners in their country that are immigrants, with the average guess at 28% when it’s actually only 15%.  And some countries are much more wrong.  For example, in the Netherlands the average guess is that 51% of prisoners are immigrants when it is actually 19%. There is a similar overestimation in South Africa, France and the USA. However, in Saudi Arabia the proportion of foreign born prisoners is greatly underestimated. The average guess was 26%, but the actual figure is 72%.
  4. Teenage pregnancy: all countries overestimate teenage births, some by a staggering amount. For example, in Brazil, the average guess is that 48% of 15-19-year-old women and girls give birth each year when it is actually 6.7%. The other Latin American countries are also massively out, particularly Colombia, Mexico and Peru. South Africa were similarly incorrect with an average guess of 44%, where the actual figure is 4.4%. Even those who are closest still overestimate the extent of the issue: for example, Denmark and Norway both guess 8% but the actual figures are 0.4% and 0.6% respectively.
  5. Vaccines: despite the claim being widely discredited, under half of the population in most countries correctly say that vaccines do NOT cause autism in healthy children. Some have very high levels of belief in the claim. For example, in Montenegro and India, 44% of people believed that the statement “some vaccines cause autism in healthy children” was true.
  6. Diabetes: every country in our study overestimates the extent of diabetes in their country. In India and Brazil, the average guess is 47%, but the actual diabetes figures are 9% and 10% respectively. The average guess for the prevalence of diabetes across all countries is 34%, whereas actual average figure is 8%.
  7. Good health: nearly all countries think people report their own health as much worse than they actually say in surveys. In South Korea, the average guess is that 39% of people rate their own health as good or very good, but when asked themselves, 80% of South Koreans report good or very good health. A similar pattern was found in New Zealand and Malaysia. The only country to greatly overestimate self-reported health was Japan whose average guess was 47%, but when asked themselves only 35% reported good or very good health.
  8. Suicide: there is a real mix in overestimating and underestimating suicide among both women and men aged 15-24, with some guessing significantly over and some significantly under the actual share that suicides make up of deaths among these groups.  For example, the percentage of deaths by suicide among young women in Hong Kong is an incredibly high 50%, but in our study Hong Kong’s average guess was only 14%. At the other end of the spectrum, South Africa’s average guess for suicides among young men was 27%, whereas their actual official suicide figure is 1.1%. Looking across all countries, the average guess is that 20% of deaths of both young men and women are through suicide, when the actual figure is slightly higher for young men, at 20% compared with 17% for young women. 
  9. Smartphone ownership:  almost every country overestimates smartphone ownership, with some incredibly high estimates in some countries. Topping the chart is Indonesia whose average guess at smartphone ownership is 85%, when the actual figure is only 21%. This may be because our study is conducted online, and in lower internet penetration countries these populations will be more middle class and connected, and therefore drawing from their personal experience, thinking the rest of the country is more like them than they really are.  But even in countries with high internet penetration, overestimation is still significant: for example, in Germany the average guess for smartphone ownership was 86% whilst the actual figure is 69%.
  10. Facebook membership: similarly, every country overestimates the proportion of those with Facebook accounts, again, often by very large amounts. For example, our online sample in India think that 64% of all Indians aged 13+ have a Facebook account when actually only 8% do.  But even more developed nations such as Germany again are way out: they think that 72% of Germans aged 13+ have a Facebook account when actually only 34% do. 

And on our more “festive” questions, about our spiritual beliefs, sugar and alcohol consumption, there are many global misperceptions…

  1. Alcohol: Russia is seen as the booziest nation in the world – but they actually only rank 7th from the 38 countries.   Very few people pick out the correct answer of Belgium as the highest consumer of alcohol per person – including in Belgium: only 5% of Belgians rank themselves as the booziest country.  The USA on the other hand is seen as much more heavy drinking than it actually is: they are the second most likely country to be picked out as a top drinker, but are actually only 13th in our list of 38.
  2. Sugar: the USA is correctly seen as having the sweetest tooth, with 58% picking them out as the highest sugar consuming nation, way ahead of anyone else.  Other nations such as France and China are wrongly seen as high consumers of sugar. France came up as joint 3rd most mentioned across participants in our study, but their actual global ranking is 16th, and China came up in our top 5 mentions, but globally they are 24th. The USA are most likely to think of themselves as the most sugar consuming nation with 69% of people in the USA putting themselves top.
  3. Belief in Heaven, Hell and God: across the study as a whole, we’re not bad at guessing how many people believe in Heaven, Hell or God – but this hides some very wrong views in individual countries.  For example some countries significantly overestimate belief in Heaven: Japan guessed that 42% of people believe in Heaven when the actual figure is just 19%. In South Africa, the pattern is the opposite; their average guess is that 67% believe in Heaven, but actual survey results show belief in Heaven is 84%. Guesses on how many people believe in Hell follow a similar pattern of big errors in both directions. For example, people in Spain think that 43% of Spaniards believe in Hell, when actually only 19% say they do.  Belief in God was also split: for example, Swedes think nearly twice as many people believe in God than actually report they do (37% versus 22%).

Looking across the seven key questions where we get people to estimate factual realities, there are clear patterns in which countries have a more accurate view of their countries.  To capture this, we’ve calculated the Ipsos “Misperceptions Index”, as shown in the table below. 

South Africa receive the dubious honour of being the most inaccurate in their perceptions on these issues, with Brazil and the Philippines also high up the list.

Sweden are the most accurate, followed by Norway, with Denmark in third.





South Africa

Least accurate


























Saudi Arabia





















South Korea






New Zealand






Hong Kong



























Great Britain

























Most accurate

Bobby Duffy, Managing Director of Ipsos Social Research Institute, London, said:

“Across all 38 countries in the study, each population gets a lot wrong. We are often most incorrect on factors that are widely discussed in the media, such as deaths from terrorism, murder rates, immigration and teenage pregnancy.
There are multiple reasons for these errors – from our struggle with maths and proportions, to media and political coverage of issues, to social psychology explanations of our mental shortcuts or biases.
But in particular, we know from previous studies that this is partly because we overestimate what we worry about: the more we see coverage of an issue, the more prevalent we think it is, especially if that coverage is frightening or threatening. Our brains process negative information differently - it sticks with us and affects how we see realities. We’re more worried than we should be about how our countries are and how they’re changing.
Some of the patterns are also worrying for our own decisions: our uncertainty about the link between vaccines and autism in healthy children, despite this being widely discredited, can affect our behaviour and therefore health outcomes in nations.
We also have the wrong image of other countries in many instances: Russia and America’s image as hard drinkers probably come from cultural cues we see widely in entertainment - while Belgians get off lightly as they don’t feature so much!  But there is some truth in these national images: the USA is also correctly nailed for its sweet-tooth!
It is also clear from our “Misperceptions Index” that the countries who tend to do worst have relatively low internet penetrations: given this is an online survey, this will reflect the fact that this more middle-class and connected population think the rest of their countries are more like them than they really are.”

These are the findings of the Ipsos Perils of Perception Survey 2017. 29,133 interviews were conducted between 28th September – 19th October 2017. The survey is conducted in 38 countries around the world, via the Ipsos Online Panel system.