There's a common perception that an article packed full of stats is more credible than one without, but statistics can be just as subjective as anything else. We've put together a quick checklist that will help you to scrutinise those numbers.
There's a common perception that an article packed full of stats is more credible than one without, because fancy percentages must equal better research right? In reality, statistics can be just as subjective as any other source.
Here's a quick checklist that will help you to make sure that you aren't being fooled by the numbers.
1. Which statistic do you want to fact check?
This sounds like a no-brainer, but just because something is a number doesn't mean that it's a statistic.
It's important to identify exactly which claim you want fact check, and which numbers are being used to back it up.
2. Where does it come from?
You need to find the primary source of the statistic—that is, the first place it appeared. Check that the statistic in your claim matches the one in this primary source.
If possible, try to check this source against others covering similar subjects, to see how it compares.
3. How is the statistic being used?
Try to understand what the original data source is really trying to say. It's common to see statistics being used misleadingly, so when in doubt try to ask yourself, "what conclusions can I actually draw from this data?"
Knowing how to do basic calculations like mean, mode and average will help you to interpret the data critically.
4. How reliable is the source?
With stats, context is everything. Ask yourself who collected the data, how they did it, and why.
What methodology did they use, and does it give an accurate reflection of real-life conditions? Was it published by a global institution or a small company, and did they collect the data for research purposes, or to inform corporate strategy?
Keep in mind that fact checking statistics can be a messy business. The main thing is to gather as much information about the context of the claim as you can, and use this to judge how credible it is.
Remember that at its most basic, a statistical claim is using numbers to try and tell a particular story. Which other stories can explain the numbers that you see?
This article's header image is by M.B.M. from Unsplash.