Wednesday, 26 October 2011

Irritating twaddle

So, I'm somewhat annoyed after having to read two bullshit, inaccurate stories in the media yesterday.

The first is all the stuff about the fall in the number of young people applying to university. Now, aside from the fact that we haven't reached the deadline for applications yet, the main annoyance is the way that not a single news story mentioned the real fall in application numbers.

It's very simple and it's not rocket science. If we compare the figures from this year to the same time in 2009 (we have to ignore 2010 due to the distorting effect of the pre fee increase bulge where people skipped taking a gap year in order to apply before the fees went up) then we'll see that the fall in applications is 2.4%. This isn't surprising given that the number of 18 year olds in the UK (e.g. the people who apply to uni) has decreased by 2.3%. That's right, the number of young people applying to uni this year is almost exactly the same proportion of young people who applied last year. This goes to show what nonsense all this speculation about what caused this "fall" is.

The other story which has got my temper up is one reporting a study by a group in America which shows that drinking fizzy drinks causes violence. This is based entirely on the fact that they found that people who drank lots of fizzy drinks were more likely to be violent. The problem with this is that it's bullshit.

As any 14 year old GCSE student should be able to tell you, correlation does not imply causation. Let me give you an example. If you look at the global population you will see that those who have microwaves are more likely to suffer heart disease. Does this mean microwaves cause heart disease? Of course it doesn't. The reason people with microwaves are more likely to suffer from heart disease is simply because ownership of a microwave occurs mostly in the developed world which has a higher rate of heart disease anyway due to the kind of lifestyle that you see in developed countries.

The only way you could tell if fizzy drinks cause violence is to take a large group of people, two thousand or so, and vary the amount of fizzy drinks they consume in order to see if they become more or less violent as a result. But this "study" hasn't done that and therefore their so-called results are nonsense. And yet the media is treating this as a bona fide story rather than the straight-in-the-bin twaddle it should really be treated as.

So, well done to the BBC, the i, the Independent and goodness knows how many others. You've successfully managed to show a level of statistical stupidity that would have made my 14 year old self face palm and which contributes nothing to informing the public. Well fucking done to you.

P.S. This is why I try not to read newspapers anymore - they only get my temper up.

P.P.S. For some accurate and interesting analysis of the student numbers figures, go read Just One More Ten Pence Piece here.

1 comment:

  1. Absolutely right - the press has completely missed the point on the UCAS story and has been seduced by the "down 9%" headline. However, it's come from somewhere, and the bulk of the decline is in the over 25 age groups. These are mostly ELQ students (like I've been for the past 5 years with the OU), who have no access to loans under the new arrangements.

    My considered thoughts are here:

    I also feel your pain about the endless and seemingly willful confusion of correlation and causation in the media! No-one should be allowed to write such stories without a basic grounding in statistics ... grrr, etc.



I'm indebted to Birkdale Focus for the following choice of words:

I am happy to address most contributions, even the drunken ones if they are coherent, but I am not going to engage with negative sniping from those who do not have the guts to add their names or a consistent on-line identity to their comments. Such postings will not be published.

Anonymous comments with a constructive contribution to make to the discussion, even if it is critical will continue to be posted. Libellous comments or remarks I think may be libellous will not be published.

I will also not tolerate personation so please do not add comments in the name of real people unless you are that person. If you do not like these rules then start your own blog.

Oh, and if you persist in repeating yourself despite the fact I have addressed your point I may get bored and reject your comment.

The views expressed in comments are those of the poster, not me.