Archive for the ‘Statistics’ Category

Articles

Lies, damn lies, and terrorism

In Governmental stupidity,Justice,Politics,Real-World Stats,Statistics on Friday 26th February, 2010 by Guy Tagged: , , , ,

The Research Development and Statistics section of the Home Office regularly publishes extremely interesting reports, even though its website is a dog (last time I tried the mailing list signup didn’t work), and more substantively its independence is hard to gauge (I don’t know if, and if so why, they carry out distinct work from the Office for National Statistics, which is supposedly on a more independent footing than it was before).

I found my own way to be told of updates to its “What’s New” page. The latest report is the lithely titled “Operation of police powers under the Terrorism Act 2000 and subsequent legislation: Arrests, outcomes and stops & searches. Quarterly update to September 2009. Great Britain”, with an accompanying press release which I can’t read, although that is almost certainly my fault.

The most eye-catching figure is that a smidgen (444, to be exact) over 200,000 “stops and searches were made in Great Britain under [Section] 44 of the Terrorism Act 2000 in the year ending 30 September 2009” (page 15 of the report and PDF). That is a lot of terrorists stopped! Although a few of them might have been, say, innocent photographers just trying to, erm, take photographs. Interestingly, the rate at which these stopsandsearches took place decreased markedly in just one year, as can be seen clearly in Figure 2 of the report, reproduced here. The South Wales force, for example, managed to cut their number of stopsandsearches from 662 to, erm, 0, and the analogous figures from Hampshire went from 633 to, well, 0! I guess all those terrorists ran away, or maybe the police started to realise they were taking the piss. Now that Section 44 has been declared illegal by the European Court of Human Rights, one would hope they will stop altogether elsewhere, but the Home Office has so far refused to countenance this.

Just to show how ridiculous the number of stopsandsearches is (the vast, vast majority of which — always over 90% — are carried out by the Metropolitan Police), consider how many arrests and convictions follow from these stopsandsearches. According to the report, “For the year ending 30 September 2009 there were 965 arrests resulted from s44 stops
and searches in Great Britain, an arrest rate of 0.5%“. The number of terrorism-related charges, let alone convictions, is not given. I’m going to bet it’s quite small.

Consider the only other figure in the report (how convenient!), Figure 1, also reproduced here. It shows the proportion of charges following a terrorism-related arrest that are actually, you know, terrorism-related. It goes from almost 80% in the second quarter of 2008 to under 20% by the third quarter of 2009.

And don’t forget that’s just the proportion of those charged! Table 1.2, “Outcome of terrorism arrests”, shows that for the two years leading to the 30th of September 2009, there were 379 terrorism arrests [although I’m not sure how this squares with the figure for arrests following Section 44 stopsandsearches given above. I think maybe they are only counting Section 41 arrests and suchlike, looking at Table 1.1], of which 66 (about 17%) led to terrorism-related charges (including those determined to be terrorism-related as “assessed by the ACTCC”, whatever the hell that is), 64 (ditto%) led to non-terrorism related charges, 193 (over 50%) were released without charge, and finally 56 (about 15%) were, ahem, served with “alternative action”, which “includes cautions for non-terrorism offences, transfers to immigration authorities, transfers to the PSNI, summonses and those dealt with under mental health legislation”.

In case you didn’t see the bold bit above, let me repeat it here: over 50% of terrorism-related arrests led nowhere. And this doesn’t include the Section 43 or 44 arrests.

Truly pathetic.

Extra stuff:

The report (viewable here)

Update: Well, lookee here! The Guardian has stumbled on this data too and put it into a spreadsheet. Go get ’em, boy (or girl, if that’s your thing).

Articles

Graphs of the Economist: Permanent Edition

In Real-World Stats,Statistics,The Economist on Thursday 26th February, 2009 by Guy

Unsurprisingly I didn’t keep up with my extremely intermittent series of Graphs of the Economist. In the meantime, though, I took the trouble to set up an RSS feed that will deliver a dose of Economist graphingdom every. Single. Day. Right to your, erm, RSS feed reader.

Get it here: http://feed43.com/economistdailychart.xml

Articles

More trouble in the Middle East — with the y-axis

In Miraculous discoveries,Politics,Random fun stuff,Statistics on Tuesday 10th February, 2009 by Guy Tagged: , , ,

Who is going to get how many seats in the Israeli general election today? Let’s look at the exit polls, courtesy of Ha’aretz:

Exit poll results... sort of

Exit poll results... sort of

Wow, the Labor Party and Yisrael Beitenu are doing pretty badly. They are only predicted to gain one quarter of the seats that Likud and Kadima are projected to get! But, wait a second. 15 is one half of 30, not a quarter, right? So how did they get the graph to look like that? The bars match up perfectly with the y-axis; they have interpolated the numbers between 15 and 30 correctly. What’s going on??

Oh. I see. It’s a, *cough*, “non-linear scale”. Got it. And this is the country with one of the highest per-capita levels of PhD holdership in the world. Come on Ha’aretz, wake up!

[source: Ha’aretz home page, although they have corrected the graph for their special election coverage page].

Articles

Belloc on statistics

In Statistics on Tuesday 9th October, 2007 by Guy Tagged: , ,

Before the curse of statistics fell upon mankind we lived a happy, innocent life, full of merriment and go, and informed by fairly good judgment. We knew when the weather was cold and when it was dry; we knew what public opinion was; we knew what was good for us and what was bad for us, and all the rest. That state of affairs lasted for centuries. It was too good to last. The statistician was let loose. He came in the train of Discovery and the rest. He was part of Progress. He took up his authority in a world which could only count and was ceasing to think or to feel. He appealed to those what had learnt to read figure and to add up and to multiply and to spell, but who had learnt nothing else, who were even rapidly unlearning all things worth knowing.

— Hilaire Belloc, (Anglo-French poet and essayist), “The Silence of the Sea: And Other Essays”, 1940

I have seen the first sentence quoted in many places, but never the passage beyond it, and rarely the correct source, so I have taken it upon myself to rectify the situation. Consider the situation now-rectified.

Articles

China not so much rising as regressing to a new mean

In Politics,Real-World Stats,Statistics,The Economist on Tuesday 17th April, 2007 by Guy

Trust that leader of ‘papers, The Economist, to supply graphs that get to the heart of a matter faster than a million pundits’ articles (give or take):

Photo Sharing and Video Hosting at Photobucket [source: The Economist, 29/03/07 edition]

All that hogwash about how China will imminently rule us all, presumably just before we all get to live underwater, is just that: a pigbath, a swineshower. Utter nonsense, for those less deft with metaphors amongst you. GDP is not the all-important measure of wealth that many think it is, let alone happiness or power, but even if China did once again reach the world share of it that it had 400 years ago, the consequences would not be as dire for the “West” as they were for China under occupation by that same “West” only a century ago. I smell the rotten whiff of racism emanating from fears of lots of Orientals becoming as wealthy as Europeans, but I can’t find the corpse to prove it.

Chill out.

[I do wonder how the graph-creators decided on what the height of the bars should be, but even with the possibility of substantial errors in estimation, the message seems clear enough].

Articles

Rational decision-making in action

In Real-World Stats,Statistics on Sunday 18th June, 2006 by Guy

I always like some of that, so ogle an example of it here:

Dilbert cartoon

This is my life.