The world in which we live: Hard data & clear charts

[iNow]

I like the way this summary is pulled together. Lots of different issues are summarized, and the focus is broad and data-based. It's called "40 charts that explain the world we live in," and it gives a good sense of where we are, where we used to be, and where we seem to be heading.

 

http://www.washingtonpost.com/blogs/wonkblog/wp/2014/01/15/40-charts-that-explain-the-world/

We searched for charts that would tell not just the story of how the world is -- but where it's going. Some of these charts are optimistic, like the ones showing huge gains in life expectancy in poorer nations. Some are more worryisome -- wait till you see the one on endangered species. But together they tell a story of a world that's changing faster than at arguably any other time in human history.

Here is one of my favorites:

 

 

 

 

What do you think? Anything surprise you, stand out, or cause you concern or joy?

[imatfaal]

I suppose that natural disasters being the fastest growing cause of death is a bit of a thorny issue. Should we be proud that preventable disease is on the decline, deaths through malnurishment and dirty water are declining, and that diseases of old age are coming at a later and later age - leaving only the unforseeable and unpreventable acts of mother nature to claim a larger percentage of an otherwise healthier population. Or should we be ashamed that some countries are bearing the brunt of our actions upon the climate and what used to be manageable and rare rains, monsoons, and typhoons are massively worsened due to deforestation, increased frequency, enlarged very poor quality coastal urban settlement.

 

There are some great charts there - and like all good representations of data they both answer questions and cause more questions to be asked.

[CharonY]

Could be confirmation bias or the fact that I have seen similar things floating around, but no real surprises (except the gender gap in Eastern and Middle Africa was lower, but the I am completely ignorant in that area). I expect that there will be a drop of the proportion of malaria in the mid-future, as there is some significant progress in understanding that parasite.

I think it also underlines the probably obvious point that prosperity has wide-ranging effects prosperity has.

Finally, as India and China modernize, they pretty much undergo the same environmental issues that Europe and other Western countries have faced. While they do invest more into alternative energy as the West did, the short timeline and population sizes makes this endeavor incredibly complicated. And obviously they will not give up their shot at prosperity for environmental reasons (as any other nation, really).

[swansont]

The presentation style is interesting, but I don't think it eliminates the problem of death rates being zero-sum. You can reduce infant mortality due to whatever mechanism, but that person will still die, just by something else. That death mechanism rate will necessarily increase at some point (though the dynamics of this would make analysis a tad complicated). So a rate increase is not necessarily due to some failure on society's part.

––

edit to add: but the graph does allow one to account for some improvements. If e.g. your treatment of heart disease doesn't prevent heart disease deaths but pushed that fatal heart attack from age 50 to age 70, you would see an improvement here in years of life saved, which wouldn't show up in a normal death rate analysis.

 

––
I am surprised death by violence is relatively small. For all of the objections we have to e.g. gun violence, the number of years lost to it is small in the big picture.

[iNow]

I think this one signifies well the nature of our changing world (increases in education and literacy will impact many downstream measures):

 

[michel123456]

I see that there is no "natural death", as people used to say some time ago.

Now death has a cause that looks "unnatural", I mean death has a cause that you can fight.

But at the last point, everybody has to die. The volume of thick square is a direct function of Earths population.

[iNow]

I only shared a link to what I've read. Credit is better directed at each of the researchers behind each of the charts, and perhaps the person who helpfully aggregated them at the link in the OP.

 

Also, to remind readers, there are 39 more charts than just the one about deaths I shared in my opening post. I only shared that one because I liked it, but the link is about much more than just death and what causes it.

[imatfaal]

Whilst death is eventually inevitable - ie you shift people from one block of that square prism to another. Death rates, even overall rates, are dropping; this statistic is usually represented in the deaths per year per 100,000 of a population (sector).

 

This was the chart that caught my eye

 

 

Yes the causes have changed - hugely, the largest killer in 1900 has almost been pushed off the list - but in addition the size of the total has diminished by almost 50%.

 

But what caught my eye was that suicide is now a significant cause of death in 2010

[michel123456]

I have to admit I don't understand clearly the above statistics:

 

_in 1900 there were 1100 deaths/100,000

 

_in 2010 there were 600 deaths/100,000

 

Do I read it correctly?

 

the last numbers mean 600 deaths/year/100,000. Does that mean it will take on the average 100,000/600= 166 years for all to die?

[imatfaal]

I have to admit I don't understand clearly the above statistics:

 

_in 1900 there were 1100 deaths/100,000

 

_in 2010 there were 600 deaths/100,000

 

Do I read it correctly?

 

the last numbers mean 600 deaths/year/100,000. Does that mean it will take on the average 100,000/600= 166 years for all to die?

I noticed that - and that's why I included the word (sector) after population in my post - I assume that the above is the deaths per hundred thousand of adults from one age to another. Another thought is that these are only the top 10 causes - the rest of the causes could be very long and change the total considerably

 

I will have a dig and find out

[iNow]

Here's the source and citations for that particular chart: http://www.nejm.org/doi/full/10.1056/NEJMp1113569

 

And here's a breakdown of the data in another way (only current mortality, though):

 

http://www.washingtonpost.com/blogs/wonkblog/wp/2012/08/01/how-we-die-in-one-more-chart/

 

Edited January 17, 2014 by iNow

[michel123456]

One must be very careful when reading statistics from diagrams like those. The image has a power that strict numbers do not have.

I know very well from my experience in urbanistic & geographic statistics, that when you make a diagram, you must make some choices, that you diagram must show something, sometimes must (or will)hide something else.

 

For example in the above:

_the 4 pies of same dimension represent in fact different dimensions: the second pie is about 3 times bigger than the first, the third 3 times bigger than the 2nd, and the fourth 3 times bigger thatn the third.

_the people from the second pie are the one who survived from the first, and so on.

_the ages have been chosen on what basis? the first pie has a gap of 24 years, the 2nd 19 years, the 3rd also 19 years, the 3rd unknown (I guess 100-65=35 years).

_the parts of the pie are not in the same order from the beginning till the end, it is very difficult to understand a progression for lets say homicide because it is not represented in all 4 charts.(is it in "other causes" in pie #4?, or is it null?)

_where is "warfare"? In "other causes"? hidden for people under 24, the brave U.S. soldiers that die in Irak?

_how many deaths in 2010? Is it the sum of the 3 pies (2,441,199)? If this is correct, the 1st pie represents 1,5% of deaths in the U.S. in 2010,and thus accidents at age 1-24 represents 0.6% of deaths. I am sure if one makes a pie at age 16-24 the most injuries are there.

_and so on.

Edited January 19, 2014 by michel123456

[imatfaal]

One must be very careful when reading statistics from diagrams like those. The image has a power that strict numbers do not have.

I know very well from my experience in urbanistic & geographic statistics, that when you make a diagram, you must make some choices, that you diagram must show something, sometimes must (or will)hide something else.

 

For example in the above:

_the 4 pies of same dimension represent in fact different dimensions: the second pie is about 3 times bigger than the first, the third 3 times bigger than the 2nd, and the fourth 3 times bigger thatn the third.

Good point - although the total magnitude is clearly stated.

 

_the people from the second pie are the one who survived from the first, and so on.

No this is all 2010 - no one is in more than one group. Those in pie chart2 have "survived" picehart 1 of previous years but not this set

_the ages have been chosen on what basis? the first pie has a gap of 24 years, the 2nd 19 years, the 3rd also 19 years, the 3rd unknown (I guess 100-65=35 years).

Yes. I see no reason for these division - seem arbitrary but perhaps there is some rhyme or reason

_the parts of the pie are not in the same order from the beginning till the end, it is very difficult to understand a progression for lets say homicide because it is not represented in all 4 charts.(is it in "other causes" in pie #4?, or is it null?)

In some charts homicide is a top 5 cause of death - in others it is not; I see no problem with this. The order changes because the causes change and I think they have tried to maintain differences and colouration where possible

_where is "warfare"? In "other causes"? hidden for people under 24, the brave U.S. soldiers that die in Irak?

456 servicemen died in 2010 due to enemy action - this is too small to show in a top 5 listing

_how many deaths in 2010? Is it the sum of the 3 pies (2,441,199)? If this is correct, the 1st pie represents 1,5% of deaths in the U.S. in 2010,and thus accidents at age 1-24 represents 0.6% of deaths. I am sure if one makes a pie at age 16-24 the most injuries are there.

_and so on.

http://www.cdc.gov/nchs/fastats/deaths.htm gives the total number of deaths as 2,468,435 which is close - and all these organisations use different metrics for their calculations.

[Dekan]

The above posts exhibit multi-coloured pie-charts and 3-D blocks. These are admirably sophisticated.

 

However, they do make the information quite incomprehensible to most viewers

 

Couldn't the essential facts be put in a simplified format - ie as lines on graphs?

[iNow]

Ensuring that the dumbest among us can understand everything is not always the primary objective, nor should it be.