Lying with Charts & Graphs

(silly WordPress mysteriously ate this essay yesterday. So, painfully reconstructed…)

Here is a nice chart that’s been making the rounds:

So, nice harmless little doodad attempting to benevolently terrify us into eating our veggies? Or Exhibit Z in What The Heck is Wrong With You People!?! Or something?  Let us count the issues that should be OBVIOUS to anyone with ANY KNOWLEDGE OF SCIENCE AT ALL:

1. Isn’t that an odd set of countries? There are over a hundred countries in the world, yet only these exact 12 are chosen, and no reason given why. Hmmm – could it be that these exact 12 happen to give the exact ‘right’ graph? Like, the data was cherry-picked to show exactly what the grapher wanted to show? Inquiring minds want to know: what does the graph look like if you picked, say, the top 100 most developed countries by per capita income, or any other selector not determined by the results you’d like to see?

2. Correlation doesn’t equal causation. There are famously charts showing stock market performance relative to the location of hemlines in fashionable women’s clothing – the shorter the dresses, the better the stock market does (or did – not sure this holds over the last couple decades). So, even if the data had not been so obviously cherry-picked, we’d need to wonder if it is really telling us what its makers want it to tell us. Wouldn’t we want to take wars, famines, and nuclear disasters, among other things, into account?

3. If this were the Brave New World, very few would die of cancer or heart disease – but only because everyone is summarily executed at the age of 60. So, are the clearly virtuous people on the right hand side of the chart just dying of something else before they get a chance to die of cancer or heart disease? In other words, in countries where all the obvious health problems that kill people young have been addresses – adequate calories, clean water, antibiotics, basic medical care – wouldn’t we expect to see a greater prevalence of causes of death that are not so easily addressed, like cancer and heart disease?

4. Are we measuring things the same in every country? Are causes of death accurately and consistently recorded in each country so that this sort of comparison is valid at all? For example, if I die of cancer as a result of getting AIDS, what is my cause of death?

So here, by the miracle of the interwebs, is a quick chart I put together:

This is an attempt to illustrate a couple of the problems mentioned above, not to answer them. In the “Life expectancy – % of table average” column, we note that 5 of the 6 ‘worse’ countries, where pouty children of all ages turn up their collective noses at raw veggies in appalling numbers, people live noticeably longer than the unweighted average life expectancy of the table as a whole. So, it seems you could expect to live a few more years eating hideous yet tasty and convenient prepackaged food and then die of heart disease or cancer than your more virtuous yet sorter-lived consumer of not so tasty endive and arugula.

Next, for kicks, I added a column for percentage of the unweighted average of the number of cigarettes smoked per year per capita, to highlight the countries with greater or lessor cigarette consumption.  4 of the 5 countries with the greatest life expectancies also consume significantly more cigarettes per capita than the average, and 3 of the 4 with the shortest life expectancies smoke markedly fewer. Soooo, maybe we should forget the kale and fennel, and light up a Camel if we want to live long?

Or just maybe prepackaged foods and cigarettes are relatively expensive? So that countries that are relatively wealthy, and might therefore reasonably be expected to have better health care and less life-shortening poverty, also have people who can afford to buy more Ding-Dongs and Marlboros? The country with the lowest life expectancy on the table – Laos – also has the lowest consumption percentage of packaged foods and the 2nd lowest use of cigarettes, and is relatively poor. Mere coincidence?

Who knows? But a couple dozen well-funded studies would be required to find out.

The issue is not really this one little piece of propaganda wrapped in a lab coat pretending to be Science! It’s that a college educated person brought this to my attention – person with strong political views, who consistently promotes items in various social media which attempt to advance the claim that Her Party is OBVIOUSLY the party of the Smart People, and that only stupid bigoted yahoos belong to the Other Party. Over the late election unpleasantness, anytime I ventured onto Facebook or Twitter, I was subjected to dozens of little charts and diagrams and maps making the We’re the Smart Party, They’re the Dumb Party claim, all by people who showed no evidence of being able to recognize something as patently phony as the graph presented here – in fact, it is more often than not the same people proposing *both* political and PC propaganda graphs.

Here, to put it bluntly, are people who by their own words and actions have proven themselves utterly incompetent to judge even the most trivial claims made in the name of science – yet they feel qualified to call other people stupid, AND are the people who tell pollsters they ‘believe in science’.

And they vote.

(If I get a chance, I’ll treat a few of the more insidious charts and graphs in the future.)


Author: Joseph Moore

Enough with the smarty-pants Dante quote. Just some opinionated blogger dude.

One thought on “Lying with Charts & Graphs”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s