Goldacre says all AUA members work really hard*
Ben Goldacre presenting at AUA2016
Goldacre says all AUA member work really hard*
Dr Chris Ince MAUA
Secretary, SOAS, University of London
The Opening Plenary at the AUA’s 2016 Annual Conference in Leeds was given by Ben Goldacre (www.badscience.net) on the subject of randomisation and how scientists “misuse evidence to bamboozle the public.”
Now the first thing I must write about, before the content of his amazing talk, is his amazing hair. I have the feeling Goldacre is a Samson-like figure and shorn of his locks he would look and be someone else less impressive. I haven’t had the chance to test this though, certainly not in a controlled way, so this may not (although it may) be true; which brings me on to the topic of the talk…
The first part was on how evidence and statistics are used, often in a (deliberately) misleading way. Some of the examples highlighted drew loud laughs from the audience as we had a whistle-stop tour of the bad use of graphs in newspapers, poor extrapolation of data and the difference between causation and correlation. The last was used to ‘prove’ there was a direct link between the number of films Nicholas Cage has appeared in and the number of people drowning in swimming pools. As the phrase goes: lies, damned lies and statistics.
The second section focused on the importance of randomised controlled trials and on how these could be used outside medicine. He spoke about how these could be used to provide evidence to support the implementation of Government policy and a number of questions following the talk covered how these might be used in the HE sector.
So, thinking of these two points I decided to undertake a piece of data analysis and commission a related trial.
Chris’s ‘expert’ data analysis and trial
I decided that AUA members might be very interested in how their work habits were affected by other activities and whether any specific factors impacted on their effectiveness. After all, we are all passionate advocates of the value-for-money agenda.
During the period between 2007/08 and 2013/14 AUA membership has fluctuated between 3,300 and 3,900 or thereabouts. This is, you will be delighted to know, approximately the same as the number of lawyers in Idaho over this period. I knew this already and didn’t have to look it up… possibly.
Next, there is an excellent website that Goldacre referred to (tylervigen.com/spuriouscorrelations) where you can compare data sets. Luckily for me it has the number of lawyers in Idaho – who would have guessed? This shows the following graph for a comparison between the number of lawyers (so let’s just say AUA members) and the number of physical copies of video games sold in the UK.
The graph clearly shows that there is an inverse relationship between the two. In other words, the number of members of AUA is inversely proportional to the number of physical copies of video games sold in the UK. What this means is that as we have more AUA members, more people are working hard and fewer games are being bought and therefore played, i.e. being an AUA member means you are hardworking!
To test this, I am now playing computer games that I bought on my work credit card all day long at work. Anyone else want to join me in this trial?
* some of the points in this article, and indeed the title, may not be factually correct.
Dr Chris Ince MAUA
Secretary, SOAS, University of London
Originally published in newslink 84 (16 June 2016)
Sources and ABA:
0 comments on “Goldacre says all AUA members work really hard*”