@Cruor99     About     Archive     Feed     Discussion     Faq

Programming and stuff

Brexit, Salt, Petitions and Open Data!

Today is the 25th of June, two days after the British people bravely voted to leave the European Union. I will not inject any of my personal feelings or leanings as to the referendum in this post, but I will note that I was extremely pleased with the result.

The reason I write this blogpost is because I noticed something came up on Twitter after a certain petition became popular:

Petition for second referendum

What is interesting, and what I want to explore, is what has appeared in the discussion around it.

The UK government is phenomenal in its use of Open Data and as such, every single petition has a very detailed json data set

In this, we can find the following result: {“name”:”United Kingdom”,”code”:”GB”,”signature_count”:361698} This has caused somewhat of a stir in pro-brexit camps, as it seems to point to only 361 thousand britons having signed this petition, among 2.3 million! What an outrage Sargon of Akkad tweet But is that truly so?

I found this an interesting result, so I wrote a little tool to parse the json and give it to me in a readable format.

There, I discovered something significant! The “signatures_by_country” attribute only amounted to 450331 signatures! This is odd I thought to myself. Then I noticed that there is a map (and indeed an attribute!) for the different constituencies. So I did another look on that. This showed that the number of votes by Constituency amounted to 2194084 votes!

So if we add up the unspecified United Kingdom votes, 365,483, with the specified Constituency votes, 2,194,084, we arrive at the following number: 2,559,567 This is much closer to what I posted in my screnshot, done half an hour after I first collected my data - and the petition surely gaining more internet popularity as newspapers mention it.

So what does this mean?

This seems to be something people tend to do, in a time of great happenings. People are prone to rapidly jump to conclusions on self-gathered data from sources they are unfamiliar with. This has happened at many times before, and I have even done it myself! I rapidly jumped on the 350k voter I am no stranger to this feeling. The UK referendum has also shown that polling in and off itself is extremely inaccurate, and getting factual data on peoples feelings is difficult.

Why I wrote this

My goal with this blog post was to show the importance of taking a breather, when looking at data that seems incredible.

Taking that extra moment to look at the data may be the difference of breaking news, and making a very public mistake.

If the 350k myth had spread and gotten a foothold, it might have created even larger resentment between the British people and the rest of europe - in a time where it needs to come together for what comes next.

As a final fun fact, did you know that 2015 people from the Vatican City have signed this petition? This is also interesting, because it only has 800 people living there, with only 450 with citizenship

Maybe online petitions isn’t such an accurate thing, when it comes to self-declaring where you come from?