Myself, Coding, Ranting, and Madness

Relative and Absolute

Facts — often misleadingly called statistics1 — are wonderful things. People have a wonderful tendency to throw them about without too much thought to what they really mean.

The one which got me today was a list of registered interests for the Conservative party with relation to the Health and Social Care Bill — something that I have, so far, avoided saying too much about. The list is long, and appears to have been researched2 rather well. But, it has some real problems

First of all, lats make something clear. In a vote to alter the way the private sector interacts with the NHS, there will always be some conflicts of interests. For a start, the Secretary of State for Health is rather likely to have some affiliation to the NHS. At some point. But, the bit which really got me was:

In addition to the list below research by Dr Éoin Clarke - has revealed 333 donations from private healthcare sources totalling £8.3 million has been gifted to the Tories.

My reaction basically boiled down to “What does that actually mean?”. I'll happily accept the figure to be true but, by itself, it means nothing at all. Being an engineer, I asked Wolfram Alpha about this point, and got this wonderful little chart, which accurately represents what I saw.

So, some context for that number:

What I haven't included here are donations to individuals; I got into something heading argument ways whilst I was researching this posts, mainly because I think my point was entirely missed — the assumption of debating the political fairness of the donations over the fact that the information was displayed in a way to make it entirely meaningless.

This kind of thing keeps coming — take, for example, the exchanges at Prime Ministers Questions a few weeks back over the number of front line police officers. In the true spirit of intellectual debate, both sides were sticking solidly to one point without as a rebuttal for anything the other had to say, specifically that the ‘The number of front line police officers had dropped’ and that ‘The proportion of front line police had increased’. Both, I am lead to believe, are true facts10, and yet both were implied to be wrong by the other, and the clear feeling that accepting both as true would be impossible came across.

As for the Health Bill...£8.3 million is a lot of money to a student. I don't really know how much it is worth to a parliamentary party — I specifically have no idea if it's enough to swing this vote in any visible way. Much like the bill itself, I have a large amount of information, and very little understanding or context. A wonderful example is, according to some rough estimates aggregated from a PMQs and newspapers over the last few weeks, well over 100% of the medical community has voiced an opinion one way or the other on the Bill. Take from that what you will.

The other thing that was doing the rounds recently was shock and outrage at the idea of privatised road construction. The details were (and are) rather scarce, but the it does all rather sound like someone is trying to run a “Private Finance Initiative” whilst trying hard to avoid those words; they've been getting a rather bad press lately, due to the delayed and over budget construction of Queen Elizabeth Hospital in Birmingham. The basic premise is something like this: big projects like this often go wildly over budget, something the government can do without. So, in an attempt to keep as much of the liability off of the tax payer's budget sheets, someone else builds it, and we slowly buy it off of them.

This idea was dreamed up back in the earlier part of the 90s by Labour, and used to fund such projects as the channel tunnel — thus allowing it to not destroy our economy. Possibly, applying it to the Health Service is a bad idea, but for roads, it makes lot of sense: our lives won't be made shorter if a new road keeps getting delayed, and it seems fair enough to me that some tolls are collected by the builders whilst it's slowly being bought out by the Government.

In fact, if you design the contracts right, you can even make it seem like quite a good idea — don't commit yourself to more than a token purchase per year, and buy them up as you have room in the budget. The unprofitable ones will go bankrupt and end up in the state's hands anyway, and the rest are being useful to people whilst generating taxable income. Hell, you could buy it up by just giving them their taxes back each year...

1. 1 ↑ A statistic has to do with a measure over a sample. Although the argument can be made that an entire population is only a sample, it is somewhat silly. To illustrate this point, consider “Only 11% of people agree” and “Police officers are being paid 11% more”. The former is a statistic in that you haven't asked everyone (probably); the latter, as much as it may sound like a statistic, is some objective fact that doesn't vary.
2. 2 ↑ By the Daily Mail, originally, for what I can tell
3. 3 ↑ http://www.guardian.co.uk/politics/2011/feb/23/labour-trade-union-donations-miliband
4. 4 ↑ They appear to be a business brokerage. As much as I'd rather my government was not funded as such, I don't really see the relevance — unless there's some other side to their business, or the contracts for private work would not be set up through the Health Secretary.
5. 5 ↑ I couldn't find figures for over the same time period, I'm afraid
6. 6 ↑ http://eoin-clarke.blogspot.co.uk/2012/02/333-donations-from-private-healthcare.html
7. 7 ↑ http://www.telegraph.co.uk/news/politics/7377307/Britains-biggest-union-takes-over-Labour-after-11m-donations.html
8. 8 ↑ Plus/minus £10M or so. I could only find annoying 3D graphs to pull the data off. Thankfully, Conservatives were at the bottom, so I didn't have to estimate two data points per year...
9. 9 ↑ http://www.electoralcommission.org.uk/party-finance/party-finance-analysis/party-funding
10. 10 ↑ Untrue facts are, I admit, something of an oxymoron. Except that most people will accept a ‘fact’ said in an emphatic enough voice