The best stats you've ever seen - Hans Rosling

About 10 years ago, I took on the task to teach Global Development to Swedish undergraduate students. That was after having spent about 20 years together with African institutions studying hunger in Africa, so I was sort of expected to know a little about the world. I started in our Medical University, Karolinska Institute, an undergraduate course called Global Health.

But when you get that opportunity, you get a little nervous. I thought these students coming to us actually have the highest grade you can get in the Swedish college system, so I thought maybe they know everything I'm going to teach them about. I did a pre-test when they came, and one of the questions from which I learned a lot was this one: which country has the highest child mortality of these five pairs? I put them together so that in each pair of countries, one has twice the child mortality of the other.

This means that it's much bigger. The difference is greater than the uncertainty of the data. I won't put you at a test here, but it's Turkey which is highest. There are Poland, Russia, Pakistan, and South Africa. These were the results of the Swedish students: I did it, so I got a confidence interval which was pretty narrow, and I got happy, of course; I got 1.8 right answers out of five possible. That means that there was a place for a professor of international health and for my course.

But one late night, when I was compiling the report, I really realized my discovery. I have shown that Swedish top students know statistically significantly less about the world than the chimpanzees. Because the chimpanzee would score half right if I gave him two bananas with Sri Lanka and Turkey, they would be right in half of the cases. But the students are not there. The problem for me was not ignorance; it was preconceived ideas.

I did also an unethical study of the professors of the Karolinska Institute that hands out the Nobel Prize in medicine, and they are on par with the chimpanzee there. So this is where I realized that there was really a need to communicate, because the data of what's happening in the world and the child health of every country is very well-aware. So we did this software which displays it like this: every bubble here is a country.

This country over here is China and this is India. The size of the bubble is the population. On this axis here, I put fertility rate because my students, what they said when they looked upon the world and I asked them what do you really think about the world, well, I first discovered that the textbook was Tintin mainly. They said the world is still 'we' and 'them'; we is Western World and them is the Third World.

What do you mean with Western World? I said, well, that's long life in a small family, and Third World is short life in a large family. So this is what I could display here. I put fertility rate here, the number of children per woman: 1, 2, 3, 4, up to about eight children per woman. We have very good data since 1962-1960 about the size of families in all countries.

The arrow margin is narrow here. I put life expectancy at birth from 30 years in some countries up to about 70 years. In 1962, there was really a group of countries here that were industrialized countries, and they had small families and long lives. These were the developing countries; they had large families, and they had relatively short lives.

Now, what has happened since 1962? We want to see the change. Are the students right? Is it still two types of countries, or have these developing countries got smaller families and they live here? Or have they got longer lives and live up there? Let's see, we stop the world, and this is all UN statistic that has been available. Here we go, can you see there? It's China; they're moving against better health. They're improving there. All the green Latin American countries, they are moving towards smaller families.

The yellow ones here are the Arabic countries, and they get larger families but no longer life. But not larger families. The Africans are the green down here; they still remain here. This is India. Indonesia is moving on pretty fast. And in the '80s, here you have Bangladesh, still among the African countries there. But now Bangladesh, it's a miracle that happens; in the '80s, the imams start to promote family planning, and they move up into that corner.

And in the '90s, we have the terrible HIV epidemic that takes down the life expectancy of the African countries, and all the rest of the world moves up into the corner where we have long lives and small families. And we have a completely new world. Let me make a comparison directly between the United States of America and Vietnam. In 1964, America had small families and long life. Vietnam had large families and short lives.

And this is what happens: the data during the war indicates that even with all the death, there was an improvement of life expectancy by the end of the year. The family planning begins in Vietnam, and they went for smaller families. The United States up there is getting longer life while keeping family size. And in the '80s, now they give up communist planning and they go for a market economy, and it moves faster even than social life.

Today, we have in Vietnam the same life expectancy and the same family size here in Vietnam in 2003 as in the United States in 1974 by the end of the war. I think we all, if we don't look in the data, we underestimate the tremendous change in Asia, which was in social change before we saw the economic change.

So let's move over to another way here in which we could display the distribution in the world of the income. This is the world distribution of income of people: $1, $10, or $100 per day. There's no gap between rich and poor any longer—this is a myth. There's a little hump here, but there are people all the way.

And if we look where the income ends up, this is 100% of the world's annual income. The richest 20% take out of that about 74%, and the poorest 20% take about 2%. This shows that the concept of developing countries is extremely doubtful. We sort of think about aid like these people here giving aid to these people here, but in the middle, we have most of the world population, and they have now 24% of the income.

We heard it in other forms. Who are these? Where are the different countries? I can show you Africa. This is Africa, 10% of world population, most in poverty. This is OECD, the rich country, the country club of the UN, and they are over here on this side, and there's quite an overlap between Africa and OECD.

This is Latin America; it has everything on this Earth, from the poorest to the richest in Latin America. And on top of that, we can put Eastern Europe, we can put East Asia, and we put South Asia. And how did it look like if we go back in time to about 1970? Then there was more of a hump, and we had most who lived in absolute poverty were Asians. The problem in the world was the poverty in Asia.

If I now let the world move forward, you will see that while population increases, there are hundreds of millions in Asia getting out of poverty, and some others get into poverty, and this is the pattern we have today. The best projection from the World Bank is that this will happen and we will not have a divided world. We will have most people in the middle.

Of course, it's a logarithmic scale here, but our concept of economy is growth with percent; we look upon it as a possibility of percental increase. If I change this and I take GDP per capita instead of family income, and I turn these individual data into regional data of gross domestic product, and I take the regions down here, the size of the bubble is still the population, and you have the OECD there and you have Sub-Saharan Africa there.

We take off the Arab states, they're coming both from Africa and Asia, and we put them separately. We can expand this axis, and I can give it a new dimension here by adding the social values there—child survival. Now I have money on that axis and I have the possibility of children to survive there.

In some countries, 99.7% of children survive to 5 years of age; others only 70. And here it seems that there is a gap between OECD, Latin America, Eastern Europe, East Asia, Arab states, South Asia, and Sub-Saharan Africa. The linearity is very strong between child survival and money, but let me split Sub-Saharan Africa.

Health is there, and better health is up there. I can go here, and I can split Sub-Saharan Africa into its countries. When it bursts, the size of each country bubble is the size of the population: Sierra Leone down there, Malawi up there. Malawi was the first country to get away with trade barriers, and they could sell their sugar; they could sell their textiles on equal terms as the people in Europe and North America.

There's a huge difference between Africa and Ghana is here in the middle. In Sierra Leone, humanitarian aid here, in Uganda, development aid here; time to invest there; you can go for holiday; it's a tremendous variation within Africa, which we very often make that it's equal everywhere.

I can split South Asia here. India is the big bubble in the middle but huge differences between Afghanistan and Sri Lanka. And I can speak about Arab states: how are they same climate, same culture, same religion—huge difference even between neighbors: Yemen, Civil War; United Arab Emirates, money which was quite equal and well used—not as the myth is, and that includes all the children of the foreign workers who are in the country.

Data is often better than you think. Many people say data is bad—there's an uncertainty margin—but we can see the difference here: Cambodia, Singapore; the differences are much bigger than the weakness of the data. Eastern Europe had a Soviet economy for a long time, but they come out after 10 years very, very differently.

And there is Latin America today; we don't have to go to Cuba to find a healthy country in Latin America. Chile will have a lower child mortality than Cuba within some few years from now. And here we have high-income countries in OECD, and we get the whole pattern here of the world, which is more or less like this.

If we look at it, how does the world look in 1960? It starts to move; 1960, this is when he brought health to China, and then he died, and then Dening came and brought money to China and brought them into the mainstream again. We have seen how countries move in different directions like this.

So it's sort of difficult to get an example country that shows the pattern of the world but I would like to bring you back to about here, at 1960, and I would like to compare South Korea, which is this one, with Brazil, which is this one. The label went away for me here. I would like to compare Uganda, which is there, and I can run it forward like this.

You can see how South Korea is making a very, very fast advancement, whereas Brazil is much slower. And if we move back again here and we put on trails on them like this, you can see again that the speed of development is very, very different, and the countries are moving more or less in the same rate as money and health.

But it seems you can move much faster if you are healthy first than if you are wealthy first. And to show that, you can put on the way of the United Arab Emirates. They came from here, a mineral country; they caught all the oil; they got all the money. But health cannot be bought at the supermarket; you have to invest in health, you have to get kids into schooling, you have to train health staff—you have to educate the population.

Sheikh Zayed did that in a fairly good way, and in spite of falling oil prices, he brought this country up here. So we got a much more mainstream appearance of the world, where all countries tend to use their money better than they used in the past.

Now, this is more or less, if you look at the average data of the countries, they are like this. Now, that's dangerous to use average data because there's such a lot of difference within countries. So if I go and look here, we can see that Uganda today is where South Korea was in 1960.

If I split Uganda, there's quite a difference within Uganda. These are the quintiles of Uganda; the richest 20% of Ugandans are there, the poorest are down there. If I split South Africa, it's like this. If I go down and look at Nigeria, where there was such a terrible famine, it’s like this. The 20% poorest of Nigeria is out here, and the 20% richest of South Africa is there.

And yet we tend to discuss what solutions there should be in Africa. Everything in this world exists in Africa, and you can't discuss universal access to HIV for that quintile up here with the same strategy as down here. The improvement of the world must be highly contextualized, and it's not relevant to have it on a regional level; we must be much more detailed.

We find that students get very excited when they can use this, and even more, policymakers and the corporate sectors would like to see how the world is changing. Now why doesn't this take place? Why are we not using the data we have? We have data in the United Nations, in the National Statistical Agencies, and in universities and other non-governmental organizations because the data is hidden down in the databases.

The public is there, and the internet is there, but we have still not used it effectively. All that information we saw changing in the world does not include publicly funded statistics. There are some web pages like this, you know, but they take some nourishment down from the databases. But people put prices on them, stupid passwords, and boring statistics, and this won't work.

So what is needed? We have the databases; it's not a new database you need. We have wonderful design tools, and more and more are added up here. So we started a nonprofit venture, which we called Linking Data to Design. We call it Gapminder, from London Underground where they warn you to mind the gap. So we thought Gapminder was appropriate, and we started to write software which could link the data like this.

It wasn't that difficult; it took some person's years, and we have produced animations you can take a data set and put it there. We are liberating UN data; some few UN organizations, some countries accept that their databases can go out into the world. But what we really need is, of course, a search function— a search function where we can copy the data up to a searchable format and get it out into the world.

What do we hear when we go around? I've done anthropology on the main statistical units; everyone says it's impossible. This can't be done—our information is so peculiar and detailed that it cannot be searched as others can be searched. We cannot give the data free to the students, free to the entrepreneurs of the world. But this is what we would like to see, isn't it? The publicly funded data is down here, and we would like flowers to grow out on the net.

One of the crucial points is to make them searchable, and then people can use the different design tools to animate it there. I have pretty good news for you. I have good news that the present new head of UN statistic, he doesn't say it's impossible; he only says we can't do it, and that's a quite clever guy.

So we can see a lot happening in data in the coming years. We will be able to look at income distributions in completely new ways. This is the income distribution of China in 1970; this is the income distribution of the United States in 1970—almost no overlap, almost no overlap.

And what has happened? What has happened is this: that China is growing; it's not so equal any longer, and it's appearing here overlooking the United States almost like a ghost, isn't it? It's pretty scary, but I think it's very important to have all this information. We really need to see it.

Instead of looking at this, I would like to end up by showing the internet users per 1,000. In this software, we access about 500 variables from all the countries quite easily. It takes some time to change for this, but on the axes, you can quite easily get any variable you would like to have.

The thing would be to get up the databases free, to get them searchable, and with a second click to get them into the graphic formats where you can instantly understand them. Now the statisticians don't like it because they say that this will not show the reality. We have to have statistical analytical methods, but this is hypothesis-generating.

I end now with the world there; the internet is coming, the number of Internet users is going up like this. This is the GDP per capita, and it's a new technology coming in. But amazingly how well it fits to the economy of the countries—that's why the $100 computer will be so important.

But it's a nice tendency; it is as if the world is flattening off, isn't it? These countries are lifting more than the economy, and it will be very interesting to follow this over the year as I would like you to be able to do with all the publicly funded data. Thank you very much.