Coronavirus data: what story are you reading?
Let’s talk about COVID-19 data, how it’s presented, and what you should be looking for.
If we want to compare data between countries, the first thing we have to note is the huge variations in population. What does that mean for our data? It means that we need to standardize before we compare. Which means we don’t want to look at totals, we want to look at rates.
So, how do totals and rates affect the way the data looks?
Okay, first we need data. The data I’m using here is pulled from Worldometer, an independent, non-affiliated, and well-reputed information source. Their COVID-19 data section is here: worldometers.info/coronavirus. Keep in mind the numbers are updated very frequently.
Here is some data on totals case reports by country. What’s the first thing you think when you look at this pie chart?
Probably something like, “America is doomed.”
But what’s the context here? We have already noted that these are total case reports for countries with very different populations. So it’s not unreasonable to think that the U.S., as the 3rd most populated country in the world, could be reporting such a large number of cases.
Now let’s pull out the top 4 countries from the above pie chart for example’s sake. Here is the data for the 4 countries with the most reported cases:
Now let’s see what happens when the numbers are standardized. When we standardize, we turn Apples vs. Oranges into Fruit vs. Fruit. To do that, we can ask – what is the number of cases per 1 million people for each of the 4 countries above? That is, if we scooped up a randomly chosen group of 1 million people from each of these countries, how many people would be infected in each scoop?
Okay – presenting things this way seriously changes up the story! Looks like all 4 countries are on equal ground here. Behold, the power of standardization – the power of making sure all of your players are on the same field. We don’t say a basketball player is a better athlete than a soccer player because they score more points per game. The games are fundamentally different - the points are on different scales.
Beyond standardization, you have to ask yourself what question the data is answering. Yes, the raw U.S. totals look astounding, but just like our athlete example, they’re not comparable to the other totals unless everything is standardized first. The only question you can answer by comparing raw totals is, “How many cases are in each country?” Which I suppose isn’t a completely useless question. It tells us………..nope. It tells us nothing. It is actually a useless question.
Standardizing the numbers for comparison lets us ask whether the overall population of each of these countries is similarly affected by the virus. And that is a useful question.
We cannot afford not to think critically about the numbers in the world’s current situation. If we want to make the most informed decisions for our families, communities, and countries, we have to know what the numbers are really saying, and we have to ask the right questions.
Worldometers.info is a really great place to find numbers that can answer your COVID-19 questions. They have all kinds of visuals like the first pie chart I used above. Just make sure you really understand the stories the numbers are telling you.
Stay safe and healthy,
kdoh
