Shall we talk about Gay Marriage?
Or rather, shall we talk about statistics?
Or rather, shall we talk about what people want you to know, about manipulation, about foggyness, about fudge, and - yes -about lies?
I'm not the first; and I am not going to be the authority on it. You can spend decades at university learning about statistics; statistical manipulation; interpretation; flaws; averages - norms etc; deviations; demographics; gaining degree and then masters and then doctorate and then teaching the things and everything that you can do to avoid work.
My last study of statistics was before I went to university to study something other. But university introduced me to 'university perennials': people who can't or won't work. I've met many over the years (few) since I left; and they have different reasons behind continually returning or getting drawn back to university.
Many of those reasons are good, true and honourable. Some are not. I know of people who have done a Masters - and even gone on to do a PhD purely on the basis that they couldn't think of anything else to do. Some have genuinely looked for work and tried hard and simply found nothing. Some have looked for a job in a very limited field, and after a few knock-backs given up. Some have done less well, come to the end of their course and merely scoured prosepectuses for the next course that they could do. Some have tried, really tried, even going for jobs far beneath their skills because - whether in the time-lapse between applying for a degree and completing in the job market has changed; or whether when applying for a degree as a young student you really don't have your eyes set on the ultimate goal and (quite rightly) select a course which you enjoy and in which you will thrive - there are no jobs available when they are released int the world, and there is a glimmer of hope if you have that MA or other letters. There are some who have their eye on their eventual goal: employment by the university and lecturing at the university so that they can pursue their personal passion of research.
But there are those who become lazy, indoctrinated, isolated, insulated. They love a lifestyle where they can get out of bed at 9 (or perhaps later) where they can work when they want, to the plan they want, with occasional deadlines that they can ask to be extended. They don't want to think about appraisals, and targets and clients or customers or taxes or pensions or anything like that. They are cocooned in an environmental establishment where the careers service is a very optional extra-curricular activity.
From my life at university, it is different to some course. Most courses it is difficult to 'break out' of that system. Yes, there is an expectation that end of the course you will go into some gainful employment, but few direct routes. I did law - there, there is an expectation that you will go on to do a post graduate diploma and either follow the route to becoming a solicitor or barrister. There is more than enough careers advice telling you how to do that; considerably less so if you decide that is not your cup of tea. But there isn't the help and assistance actually achieving the ultimate goal: securing that training contract or pupilage. That's down to your own hard work and perseverance. In other courses, unless during your entire course you look around for jobs for which you have skills, hone your skills appropriately to those jobs, and - obviously - are ultimately successful with your application; I can wholly understand how simply easy it is to stay within the cosy education system.
/rant
Anyhow, this takes me a hundred, million miles away from where I was supposed to be arguing. Yes, statistics is a mammouth, huge and diverse subject. But it should be mandatory that someone in every single media outlet be qualified on statistics. Just as they all have a solicitor - or usually a team of solicitors - to ensure that they comply with the press code, libel laws, and data protection and such like; so there should be a statistician there to advise as to what precautionary caveats should be included in any report as to things such as sample size, demogarphic, margin of error, even error on error.
For example, this week I saw there was a media report that "the public overwhelmingly supports gay marriage". And what was the basis for this report?
It was the fact that the hashtag #gaymarriage was trending very highly on Twitter.
So, shall we deconstruct this...
Demographic
Well, it's Twitter. I shall include my own considerations of margin of error here:
The people I follow on Twitter are predominantly under 35; predominantly liberal; predominantly pro-active and ready-to-adopt/adapt (certainly in terms of media); generally more affluent; and predominantly based in or around London.
Now, I know that my sample size (around 100) is very VERY limited and my further arguments will actually work against it being taken seriously; however it is something I have observed quite widely. I live in the heart of England (though if you ask me, it is North in a strict North-South divide). If I ask people round here if they use Twitter, hardly anyone does. Quite a few people have set up accounts but rarely - if ever - use them. I have to say, I was slightly slow on the uptake myself for that very reason (despite having my account since 2007): it was only when I realised that my London friends used it as their predominant form of communication that I became a much more frequent user. But amongst my many many Northern friends, I think two of them use Twitter frequently.
Again, small sample size, but of the few people I know over 35, they don't use Twitter. My mam and dad have just gotten around to Facebook, which they like because of photo albums. They ain't gonna tweet any time soon. In fact, unless you're someone famous with something (you think you need) to say or someone with something to promote, you're not going to use Twitter. But I've discussed a lot of this already in a different post as to why people use Twitter.
Anyhow, that's a bit of an idea as to the people who use Twitter. And it was reported in the 'liberal' press - those that are greater users of Twitter.
Bias
This isn't statistics - this is human psychology. Everyone has probably heard the phrase "You do a good thing, they'll tell one person about it; do a bad thing and they tell ten people." I'm not so sure that this is actually true, and having seen some on the internet memes that go around, people are very willing to share good news - humanity as a species is a naturally optimistic one, and despite the vast doom and gloom of the news we probably look forward to the '...and finally' story at the end. Yes we enjoy a good tut and scowl and 'isn't this wrong'; but we love watching a dog skateboard all the more. And it is - very weirdly, counter-culturally the pessimistic side which we seem to emphasize about ourselves. Perhaps it is like those school reports where we were told 'could do better' - actually, we had done very well, but as an English nation we don't want to acknowledge that. But if we have done well, do we not slack off? Achieve less? I suppose that's another subject.
The point I am really trying to make is that we are much more eager to speak out (albeit digitally) where we feel there is an oppression occurring; a minority that needs saving. Those that support the status quo are either absolutely indifferent, not wishing to swim against the current for disturbing the waters; or radicalists whose views will be heard; but could be dismissed. It is those with something to fight for that have something to fight for - those for whom the change matters, makes a difference. Those for whom there is no change, whom still can marry regardless, hav less of a podium to speak from so have to find some other arugment to build themself up with.
We naturally shout up when something is wrong. When it is right, we don't rightly acknowledge when it is...because it's turned out just exactly how it should have turned out.
So there is natural human nature which means the single voices shouts above a reasonable contented murmur.
Sample Size
So, we've already done demographic; what about the sample size? The report was that it was trending high. But trending where? Was it UK only? Was it worldwide?
So, what do you need to trend anyway?
Well, I've just done a quick search. Top trending at this very second in the UK is #NTAs. The National Television Awards. For the UK actually quite a huge thing. And I've just totted up the uses of the hashtags.
500
Of which many were 'media' uses: media sites which tweet this things with a referral to their websites.
500.
Twitter claims to have 500 million accounts worldwide. In the UK it is vague: it has between 7 and 15 million accounts. Now, an account is not the same thing as an active user; and if we assume that 75% of accounts are dormant or fake, that's still a lot of 'active users' most distinctly not tweeting about one of the hottest topics in the UK.
And when we look at actual tweets - which, obviously differ again because many 'active users' will choose not to tweet about a subject; whilst some active users will tweet many times on the same subject.
So, worldwide there are around 200 million tweets per day. In the UK it is around 50 million per day.. So 500 tweets out of 50 millions... well, that's around 1 in 100,000 tweets. Today. And that's a good trend. A 'reasonably decent' trend is in at number ten - #garybarlow. With 45 tweets.
45 Tweets will get you into the top tweets?
And this is when the media pick up on it? If an account were to have 45 followers; and each of them retweet a comment; or hashtag a comment once then it gets into the top ten, it is highly easy to see how rarely the idea of the hashtag has actually been adopted, and how useless they actually are - unless you are a newspaper journalist scrabbling around to find a statistic to confirm your story... which finally brings me to:
Journalism
Everyone has a very biased opinion to put across, journalists particularly who seem to think that their opinion is more worthwhile, more acceptable, more reasearched and therefore more reasonable than anyone else's. And this is th thing about statistics: there is always someone out there who is willing to give you the statistic that you want to corroborate your point of view. All you need to do is find it.
And you may find it in many different sources.
I have to say, the BBC is fantastic both to a good and bad extent. Whilst they are one those who reprint press releases from those with a vested interests; quoting from biased reports funded by institutions with even more of a vested interest; so they also have the Magazine, and the Monitor with a regular feature from Michael Blastland analysing very carefully how unreliable almost every statistic is; and their own letter writers and commentators commenting on how biased or unreliable reports actually are.
So it leaves me with a note of caution - never believe anything.
No comments:
Post a Comment