Statisitcal claims- how to approach with caution.

From the Financial times conversation with Sarah O’Connor and Tim Harford, we as a public need to be careful about how we approach statistical claims.

 This is made broad from by Tim Harford who states that people should observe their feelings, he goes on to examine how motivated reasoning is key to the way that we interpret statistics, and any other factual claims and the decision to share them.

Tim Harford also goes on to discuss that people should make a clear distinction between the concepts of correlation and causation, he also provides an example of this which is “You’ll hear a headline like people who play violent video games are more likely to engage in violent behaviour.”

 He stresses that people shouldn’t react hastily and should break down the terms and ask what the headline actually wants to reveal. He also highlights that headlines often won’t make it clear if the issue in question is a causal claim or not.

One of the key pieces of advice given by Tim Harford that I particularly agree with is his point that people should look for details that have been left out of the statistics given, extending to getting a grip on the backstory of the factual claim or other information.

 I find it to be sound advice and good way for people to find out the root of how to be certain on how to tell the fact from the fiction.

A further point that I agree with Harford on is to be curious and always look at the source of the statistics and be sure in your own mind that you are confident with the quality of the information.

However I do have one point of reservation from this conversation which is that all of this advice is helpful, but in other ways it isn’t as it all depends as to which side you are looking at the information from, in which case these ways of advice could cause more friction between opposing groups; leading on to further confusion.

My approach to statistical claims is that I read the information and try to think to myself, first and foremost does is make sense, and moreover does it make sense. If not I do a cursory search for further background on the given subject.

One instance where I felt I had been misled was  reading the polls during the 2016 US presidential election which were all pointing to Hilary Clinton being odds on to win, but I read it wrong and was shocked at the end result.

The new information on understanding statistical claims has changed my view on a current story, that story being the ongoing struggle to renegotiate the terms of the Brexit withdrawal agreement to eliminate or find a workaround the proposed “backstop”.

With this new information, I would seek to research further into claims, and find out other ways which could neutralize the backstop or find a workable solution.