The Curious Case of Zimbabwe
A Codeacademy project…and probably some stuff to talk about with my students in AP Stats next year. I was supposed to maybe write this more like a data analysis/report. Next time.
In looking over the data of six countries (all data from the World Bank and World Health Organization via Codeacademy), their GDPs, and their life expectancies, there were some predictable findings and some interesting findings. The six countries in the dataset were Chile, China, Germany, Mexico, the USA, and Zimbabwe. Most people would probably assume that countries with higher GDPs, ostensibly more developed countries, would have a higher life expectancy. A bar graph of the six countries and their life expectancies for the last year in the dataset, 2015, is below. If you haven’t already looked, it might be interesting to cover the names with your left hand and guess which country is which.
Did you guess correctly? Further down there will be some graphs that look at GDP, but for now it is clear that of the first five countries on the chart, regardless of their GDP, their life expectancies are all within five years of each other. It is perhaps notable that America, a very rich country (for some people), does not have the highest life expectancy. It’s not even in second place.
Okay, but what’s more interesting is to see life expectancy (an idea based on time) and its relation to time:
According to this graph, as time goes on you get more time! Simple!
With all of the medical advancements over the years, people are living longer. By now you’ve probably developed the somewhat obvious hypothesis that these countries in the dataset are getting wealthier. You are not wrong. There were a few clues of course. I mentioned GDP just a few sentences ago and, for those who pay attention to the world, China is in the dataset and China has been growing wealthy for a long, long time. But that’s all inconsequential, in my opinion, to what this chart shows us: Zimbabwe!
Look at that gain in life expectancy! Granted, those other countries are starting at a baseline of 70 years old (it would be cool to do adjust for magnitude and determine how hard it is to increase an already high life expectancy…), but that doesn’t take away from Zimbabwe’s achievement, if it is an achievement. Is Zimbabwe getting wealthier? If not, what happened (More money and the things that come with it? A statistical anomaly? Nothing?)? First, let’s take a look at the general relationship between GDP and life expectancy for all of these countries.
Be certain to pay attention to that number at the bottom, far-right corner of each graph so that you can truly understand the magnitude of difference between the respective GDPs. For example, each tick on the graph of the United States of America graph represents a number that is NOT one trillion dollars but no smaller than ten(!) trillion dollars. By contrast, Zimbabwe’s highest recorded GDP (in this dataset) is ten billion dollars. Yes, that means that the lowest USA GDP was 1000 times higher than Zimbabwe’s highest GDP.
Another noteworthy aspect of these graphs is the median line. It is a median calculated from the dataset, i.e. it is not the median age life expectancy of the world. Still, the only countries that appear to be outperforming or underperforming the median by at least a year are Germany and Zimbabwe.
Leaving Germany aside, let’s focus on Zimbabwe. How did the country gain 20 years on its life expectancy so rapidly only to still be that far below these other countries? Is it a fluke? Can the data answer this question?
The first thing we want to know: is this actually uncommon? Did countries with a similar GDP in the year 2000 undergo a similar transformation? It would be pointless to investigate the significance of the other countries in this dataset with higher GDPs (although in terms of GDP per capita some of these countries qualify) and less robust gains in life expectancy given that their life expectancy was already above 70 when this data was recorded, a mark that still eludes Zimbabwe. So, we had to dig for some more data.
The second dataset I used (also from Codeacademy) had data for 158 countries but the GDP was given as a per capita measure. It makes things a bit more confusing when making comparisons. For instance: China’s population is roughly five times that of the USA’s, its overall GDP is roughly equal to the USA’s, but despite that per capita GDP gap China’s life expectancy is just about the same as the USA’s.
Anyway, I looked at countries with a GDP as low as Zimbabwe (under $1000 USD) in the year 2000. I also looked at countries that had a life expectancy under 60 years of age. There were 81 countries that matched the first query and 49 countries that matched the second one. It was too much data to visualize effectively, but from a quick glance, it appeared that all of the countries increased their life expectancy though none increased as much as Zimbabwe.
It seemed the thing to do was to find the countries with the highest positive net change and then look at a distribution of the change in life expectancy over this 15-year period.
According to the data, Zimbabwe still has the highest net change in life expectancy but there are several other countries with significant gains. Are they outliers?
The boxplot shows that Zimbabwe and perhaps five other countries (those dots to the right of the last box) from the table are outliers. What do these countries have in common? They are all African nations, four of them are sub-Saharan, and three of them border each other. Over the 15-year period their GDPs generally increased (according to the World Bank) but the growth rate, generally speaking, did not.
So, the countries were getting a bit wealthier and medical technology and availability improved. What about the rate of growth for their life expectancies?
The spikes on the graph are interesting (not too interesting because these aren’t algebraic functions with fluid, continuous rates of growth; they’re just scatterplots with lines and the spikes could be smoothed by changing the scale). Every country other than Botswana has at least one sharp jump followed by a dip. It would be interesting to know what causes these jumps; life expectancy is calculated with a function but year-to-year fluctuations could cause a jump. Another interesting aspect of the jumps is that Zimbabwe spiked from 2014 to 2015 which suggests the true gain in life expectancy might be slightly smaller than what the available data says.
We still don’t know what caused this gain, the biggest in both datasets. If it wasn’t robust growth what was it?
It seems like it might have been the exact opposite. According to this opinion piece in New Zimbabwe, the cost of the regime of former president-cum-dictator Robert Mugabe was in the billions. The writer has a caveat about the statistics but it’s brief. In the article, there are many estimates for losses suffered that might be due to Mugabe or might be the result of a variety of factors. I’m not suggesting Mugabe was great for the economy, but severe droughts and epidemics were outside of his control (his response however…). Other Southern African countries with different political situations also suffered economic downturns during the same and earlier time periods. Still, there’s no doubt about the harm that the handling of health crises like AIDS, international relations crises like land seizures (not that the land shouldn’t have been redistributed), and other Mugabe policies did to the welfare of the country.
The article, written in 2017, also mentions that life expectancy had been cut in half because of death rates tripling and fertility rates remaining the same. Not sure what data this is referencing. Even in the year 2000, it couldn’t be said that 45 years old was half of Zimbabwe’s previous life expectancy. Again, this doesn’t mean that the article’s thesis is wrong (i.e. Mugabe cost Zimbabwe resources and lives) but the stats seem a bit off. At any rate, we know that Zimbabwe suffered heavy losses under Mugabe.
And yet…Mugabe was still there in 2015. I can accept that the country's gain in life expectancy was a recovery from a particularly bad time but it’s confusing that it happened under the same man who helped to cause it. Instead of answering my question I only have more questions.
Having reached the limits of what this data can tell me I’m turning to the books, specifically A History of Zimbabwe by Professor Alois Mlambo. I’ll follow up here when I finish it.
