Comments on: BBC - Hans Rosling - The Joy of Stats

BBC - Hans Rosling - The Joy of Stats

East Manitoba Regional Junior Kabaddi Champion '94 — Thu, 02 Dec 2010 10:41:22 -0800

Hans Rosling [previously, previously] compares the health and wealth of 200 countries over 200 years in 4 minutes using the best infographic ever. Interactive Flash version here.

By: Space Coyote

Space Coyote — Thu, 02 Dec 2010 10:48:43 -0800

I love the extra dimension of the size of a circle that I see in a lot of plots these days.

By: bonehead

bonehead — Thu, 02 Dec 2010 10:49:26 -0800

Pretty neat.

By: chavenet

chavenet — Thu, 02 Dec 2010 10:49:49 -0800

Pretty neat, indeed.

By: kavasa

kavasa — Thu, 02 Dec 2010 10:50:28 -0800

What I have yust seen was quite fun, and also optimistic. Yay!

By: shakespeherian

shakespeherian — Thu, 02 Dec 2010 10:55:50 -0800

This is completely awesome.

By: ao4047

ao4047 — Thu, 02 Dec 2010 10:58:34 -0800

And now: I start the world! His passion is so invigorating.

By: sexyrobot

sexyrobot — Thu, 02 Dec 2010 11:07:32 -0800

i'd be very interested in seeing this done for the ancient world...(in adjusted dollars, of course) particularly the transition into the dark ages...

By: flod

flod — Thu, 02 Dec 2010 11:07:45 -0800

Wow, amazing. I'm struck by how clearly the Great Chinese Famine is shown, between 1958 and 1961. Just a huge drop in that big red ball, followed by an equally huge rebound. I want this guy to explain the whole world for me.

By: East Manitoba Regional Junior Kabaddi Champion '94

East Manitoba Regional Junior Kabaddi Champion '94 — Thu, 02 Dec 2010 11:14:21 -0800

His enthusiasm is infectious, like Spanish Flu!

By: The 10th Regiment of Foot

The 10th Regiment of Foot — Thu, 02 Dec 2010 11:22:18 -0800

Can somebody explain Russia's 1933 life expectancy please? It's a data fluke right? I mean I knew Stalin was bad and all but I wasn't aware that he killed everybody over 15 in a single year!

By: East Manitoba Regional Junior Kabaddi Champion '94

East Manitoba Regional Junior Kabaddi Champion '94 — Thu, 02 Dec 2010 11:28:41 -0800

Holomodor. According to this study the life expectancy for those born in 1933 sharply fell to 10.8 years for females and to 7.3 years for males and remained abnormally low for 1934 but, as commonly expected for the post-crisis peaked in 1935–36. There is another dot for Ukraine that is not affected by the famine. I think that's a statistical glitch.

By: The Bellman

The Bellman — Thu, 02 Dec 2010 11:57:23 -0800

According to this study the life expectancy for those born in 1933 sharply fell to 10.8 years for females and to 7.3 years for males. . . Probably a glitch, but it does remind me of the blogger who dryly noted, when Mugabe was "re-elected" leader of Zimbabwe in 2005, that life expectancy in his country had fallen during his 25-year reign from 61 years to 34 years, "making his nation one of the most rapidly-youthful in the world". I have no idea if the statistic is true (Wikipedia has it at pretty close to that), but I've always loved the phrase.

By: epersonae

epersonae — Thu, 02 Dec 2010 12:01:08 -0800

East Manitoba -- I figured that was part of the overall bounce down there at the end of the 1910s, along with WWI. It was especially noticeable tracking just the US in slow-mo. The big backward shift in per capita GDP in Russia in the 90s is pretty startling, too. What happened in China in the 1850s? (Or is that a statistical glitch?)

By: ennui.bz

ennui.bz — Thu, 02 Dec 2010 12:02:44 -0800

According to this study the life expectancy for those born in 1933 sharply fell to 10.8 years for females and to 7.3 years for males. . . Probably a glitch... or lots of dead babies.

By: epersonae

epersonae — Thu, 02 Dec 2010 12:03:53 -0800

East Manitoba: I meant your first comment about Spanish Flu.

By: The Bellman

The Bellman — Thu, 02 Dec 2010 12:15:16 -0800

Probably a glitch... or lots of dead babies. I'm not a statistician, but I would think that for life expectancy to fall, in one year, from 33 years to 11 years the number of dead babies would have to be . . . Passover-esque. Then again: Ireland in 1849: Life expectancy 20. Ireland in 1850: Life expectancy 35. And the famine wasn't even over yet, I don't think.

By: ennui.bz

ennui.bz — Thu, 02 Dec 2010 12:18:52 -0800

it not clear how the "income" axis is adjusted, if at all, for inflation and/or variable purchasing power. but the odd thing is that even though he explains how china splits apart demographically, i'm betting that the proportion of population in shanghai that is really at the same standard of living as italy is quite small: highly polarized demographics get smeared over the larger geographic region. i bet if you controlled for this you'd see the group of smaller dots marching towards the right hand corner (elites in china and india marching with europe) while the large dots stick in the middle with elevated lifespan thanks to the green revolution. now throw in peak oil and watch the green revolution turn brown as petroleum based fertilizer and mechanization gets expensive, food gets expensive and those large dots in the middle get dragged down. not quite Hans's happy view of the future....

By: ennui.bz

ennui.bz — Thu, 02 Dec 2010 12:22:06 -0800

I'm not a statistician, but I would think that for life expectancy to fall, in one year, from 33 years to 11 years the number of dead babies would have to be . . . Passover-esque. but this is the decade after WWI and the russian revolution likely decimated the adult population in the former russian empire. thus, there were likely proportionally more children in the population, a post-war baby boom which got wiped out by stalin's genocidal tendencies. lots of dead babies and children.

By: East Manitoba Regional Junior Kabaddi Champion '94

East Manitoba Regional Junior Kabaddi Champion '94 — Thu, 02 Dec 2010 12:23:09 -0800

ennui.bz, they use 2005 International dollars, i.e., the purchasing power of the US Dollar in 2005.

By: Chuckles

Chuckles — Thu, 02 Dec 2010 12:25:53 -0800

The big backward shift in per capita GDP in Russia in the 90s is pretty startling, too. Yes.. Part of what's so startling about it is that there is no sign of the actual collapse of the USSR, only the aftermath. That isn't how most analysts see it. The primary explanation for the collapse is the fall in oil prices in the '80s, but that just isn't reflected. Contrast that with Saudi Arabia, where you can clearly see the price change.

By: Baby_Balrog

Baby_Balrog — Thu, 02 Dec 2010 12:34:10 -0800

The big backward shift in per capita GDP in Russia in the 90s is pretty startling, too. If you watch very closely, you'll see the entire right half of the graph take a sudden twitch backward at the very end of the "replay" timeline.

By: -jf-

-jf- — Thu, 02 Dec 2010 12:48:01 -0800

Does the BBC tend to put programs online? Or, What are the chances of being able to see the whole program in the US any time soon?

By: I_pity_the_fool

I_pity_the_fool — Thu, 02 Dec 2010 12:50:03 -0800

What happened in China in the 1850s? (Or is that a statistical glitch?) The Taiping rebellion maybe?

By: maudlin

maudlin — Thu, 02 Dec 2010 12:51:03 -0800

Have you downloaded Gapminder? You should really download Gapminder.

By: heathkit

heathkit — Thu, 02 Dec 2010 12:57:08 -0800

Ah, I've seen this kind of visualization before! It's called a motion chart in the google visualization API, you can play around with it by following the link. That was always my favorite visualization widget because it seemed you could pack so much information into it, but I had no idea how to make it intelligible. I think an enthusiastic person standing behind it explaining everything to you is probably a good way to go.

By: Panjandrum

Panjandrum — Thu, 02 Dec 2010 13:40:47 -0800

ennui.bz, if you go the interactive web version linked above it does state that they are adjusting for inflation. This PDF gives a more thorough explanation of their methods for calculating income.

By: Tubalcain

Tubalcain — Thu, 02 Dec 2010 14:05:09 -0800

Look at Rwanda in 1992. Then check again in 1993. The life expectancy drops from 49 to 26.

By: wanderingstan

wanderingstan — Thu, 02 Dec 2010 14:31:32 -0800

heathkit - Motion Charts are in the Google API because Google acquired the GapMinder software and team that Hans Rosling had been leading. Rumor is that the Google founders were in the audience when he gave his first TED talk, and immediately went up to him afterwards.

By: symbioid

symbioid — Thu, 02 Dec 2010 14:35:04 -0800

Didn't he give a google talk, too? I feel like that's where I first saw him. Do they do those anymore?

By: Megami

Megami — Thu, 02 Dec 2010 15:00:23 -0800

It's great, but I always have a problem with the way 'average' life expectancy is presented. As though it was rare in 1810 for someone older than 40 to be kicking around. But it wasn't. It was just that there were a lot of children dead before they hit five, so the average was very skewed. It is the same when you talk to many people about 'classical' civlizations - they have heard somewhere that the 'average' Roman lifespan was something like 20, then start going on about how by the time you were 30 you were considered old because it was so rare, which is not the case at all.

By: Ritchie

Ritchie — Thu, 02 Dec 2010 15:56:37 -0800

I'd like to see fertility rates represented somewhere in there.

By: unliteral

unliteral — Thu, 02 Dec 2010 21:37:21 -0800

Here's Children per woman.

By: dhartung

dhartung — Thu, 02 Dec 2010 21:48:00 -0800

even though he explains how china splits apart demographically, i'm betting that the proportion of population in shanghai that is really at the same standard of living as italy is quite small Perhaps, but China is developing a middle class -- and a paradox

By: Ritchie

Ritchie — Fri, 03 Dec 2010 05:02:30 -0800

Nice find unilateral - looks like fertility correlates negatively with prosperity.

By: Catfry

Catfry — Fri, 03 Dec 2010 07:37:24 -0800

It's great, but I always have a problem with the way 'average' life expectancy is presented. As though it was rare in 1810 for someone older than 40 to be kicking around. But it wasn't. It was just that there were a lot of children dead before they hit five, so the average was very skewed. For the purposes of illustrating how 'good' or 'bad' 'things' are I think infant deaths are pretty useful, so the statistic is still appropriate in this context.

By: Defenestrator

Defenestrator — Fri, 10 Dec 2010 13:52:28 -0800

Damnit! I came here to post this. This is so cool. I love the Internet.