We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

Crime statistics

Some time ago I referred to statistics on reported crimes in the UK, which prompted a rather heated discussion (that’s putting it mildly, ed) about the value of such numbers, given the obvious difficulties in knowing whether reporting of crimes gives an accurate picture of just how bad the situation really is. The British Crime Survey (BCS) which takes the public’s impressions of the impact of crime through interviews with thousands of people, can sometimes give a quite different picture.

This story shows that reports of armed violence are on the rise, and also contains data showing that experiences of crime have also gone up. A rather sobering set of numbers with which to start the New Year.

27 comments to Crime statistics

  • Jake

    Compare those number to what is happening in New York City.

    “Crime dropped 57 per cent under Mr Giuliani, and has fallen a further 20 per cent in the past four years under Michael Bloomberg, the media billionaire who began his second term yesterday’.

    The Telegraph via Ace of Spades

  • guy herbert

    Jake:

    The point is you can’t compare them, because the British statistics (at least) are continually rebased and redefined so that you can’t even compare them with themselves fairly, never mind anyone else’s.

    It would make collecting crime statistics a touch easier, no doubt, if the Government weren’t also engaged in a continuous revolution in the criminal law.

  • Johnathan

    guy, your last point is absolutely spot-on. With so many new criminal acts being created by this government, intepretation becomes difficult.

  • Jake

    I don’t live in NYC, but I visit the city. The change is that city in the last twelve years is miraculous. New Yorkers used to be afraid (and rightly so) they would be attacked at any time even in daylight. Now they are not afraid to walk alone at night in most neighborhoods. So New Yorker’s perception of crime is in line with the statistics.

    The point is that if a mayor wants to reduce crime he can.

  • Sandy P

    Investor’s Business Daily had an informative editorial on Saturday re crime rates in Australia Britain Canada and the US from 1996-2003.

  • The gold standard for tracking crime rates is the murder rate. Most murders get recorded and there is considerably less room for redefining the crime away.

    You can tell if an area is verging on breakdown in law and order when the statistics show an increasing murder rate but a leveling off of relatively minor crimes like theft and vandalism. This indicates that enforcement has become so ineffective that the ordinary citizens believe there is no point in reporting minor crimes to the authorities.

  • zmollusc

    Perhaps it is time, since the ban on handguns hasn’t stopped criminals from getting handguns, for Mr Bliar to introduce a mandatory negative level of gun ownership. Every law abiding citizen must apply and pay for a gun licence that entitles them to keep -1 gun. This would counterbalance the illegally held weapons and could be extended to -2 or even -3 guns so long as stringent background checks were met. It would also fit in nicely with the stupidity level of much of the rest of government policy.
    There are bound to be a few morons who will query how a negative gun can be obtained, so here is the most obvious answer: One chooses a weapon from the government catalogue ( say a 9mm semi-auto pistol, £16,000 including all gun taxes and vat ) , fills out the mountain of paperwork etc and then the Ministry of Firearms sends you an IOU for that weapon. You then place the IOU in your mandatory gun cabinet. Similar procedure for ammunition.

  • Verity

    Jake says: The point is that if a mayor wants to reduce crime he can.

    This is not the case, except in cities where the mayors have formal power, as in the US. Britain is centralised. British mayors have absolutely no power whatsoever over crime control or detection.

  • Julian Taylor

    Dunno about anyone else but I find A Day In The Life of Our Little Tony (now available in all sleazy good DVD and video stores) is guaranteed to make me want to commit extremely violent murder.

  • Verity

    Julian Taylor – I couldn’t watch it due to having good taste. I stabbed viciously at the Off button the minute I saw the silly fool about 10 seconds in, pretending to be David Beckham.

    Before that, walking with such self-regard and fiddling with his cufflinks as though he is the Prince of Wales.

    Cherie Blair should never wear tight trousers. I say this in a spirit of helpfulness.

    Who paid for this garbage? The taxpayer? Why?

    Why was this little hymn to himself made? To what advantage to the average Briton? Looks like he’s wearing body armour under his jacket.

  • Can any of these Crime Statistics be trusted ?

    Who trusts Home Office crime statistics ? Nobody, according to the Statistics Commission

    The Statistics Commission is an “independent non-departmental public body” (i.e. what used to be known as a quango)

    It has published a scathing interim report on the Crime Statistics:

    Crime Statistics User Perspectives interim report December 2005 (13 page .pdf)

    Presumably if any of the planned mergers of County Constabulary police forces will lead to another set of incomparable police recorded crime statistics.

  • Guy,

    Most of the new crimes enacted are not for violent acts (in fact, are any?). Looking at violent crime should at least overcome that problem, if not the government’s continual rebasing.

    I’m with Shannon on the murder rate being a clear indicator. Does ineffective policing causes a fall in the percentage of murders reported (people are more scared of gang bosses etc..) or a rise (criminals care less about covering their tracks)?

    I do wonder if our increasingly liberal courts might be downgrading more murders to manslaughter in cases of perceived social injustice. I don’t have any evidence of this (mayber someone reading this does), I just can’t help suspecting it.

    Verity,

    The local authority in Liverpool did buy more police officers out of their budget. They can also reduce crime with measures like proper street lighting (which Liverpool doesn’t have – something about light pollution I think). I do agree with your central premise that in Britain a mayor doesn’t have the power alone to seriously reduce crime.

  • Verity

    I read Shannon’s post, Mark, not as saying that people will stop reporting murder! – but that as murder becomes increasingly common, reporting of other crimes falls drastically and that signals a breakdown in law and order. People realise there’s no point in reporting crime any more. I thought this was a very interesting point. This seems to be happening all over Britain.

    I didn’t know cities could buy extra police officers out of their budgets, though. I thought it was all centralised and centrally allocated. So … not yet, then?

  • No, I didn’t think she said people will stop reporting murder. I was just musing out loud as to how a breakdown in law and order would affect the proportion of murders reported. Probably not by much either way, which I think was her first point.

    “I thought it was all centralised and centrally allocated. So … not yet, then?”

    I doubt you’ll have to wait too long, the force reorganisation will probably make it impossible.

  • Completely off the topic of this post–and completely off the topic of any recent posts here–are any of the Bush groupies at Samizdata paying attention as the U.S. descends into martial law? Read Marty Lederman at (Link).

    It is my opinion that no one here should give a rat’s ass about the technical legality of what Dubya is doing. Bear in mind, please, that not everything that is legal is right; and–particularly from a libertarian perspective–the oh-so-impeachable Dubya is wrong, wrong, WRONG.

  • Johnathan Pearce

    none, go and read what we said about the Patriot Act. No “Bush groupies” here.

  • michael farris

    I think Shannon’s comment was very valuable. But it could have been clearer. I think (please correct me if I’m wrong Shannon) the point is:

    Murder doesn’t have to reported (by a survivor or witness) to be counted. Dead bodies are hard to get rid of (and in the scenario she’s talking about the murderer doesn’t even try) and it’s comparitively easy to figure out how a person died even when the body is badly decomposed. People don’t kill themselves through multiple stab wounds or bullets (except usually to the head area) or strangle themselves. A large majority of people who are murdered will sooner or later show up on the statistics as homocide victims.

    Breaking and entering or strongarm crime on the other hand _does_ have to be reported to be counted. If you return home and find someone’s taken your stereo but figure the police are useless and don’t bother with telling them, then it won’t enter the statistics, ever.

    When the number of people being killed is large or rising and have-to-be-reported-to-be-counted crimes falls then you have a major social breakdown on your hands.
    I have no idea if this is happening in Britain. If it is, then (from what little I know of the way the present British government works) expect new restrictive definitions of homocide most likely restricting it to convicted cases (so that a person found with multiple stab or gunshot wounds in the back will be counted as ‘dead from unknown causes’).

  • guy herbert

    Most of the new crimes enacted are not for violent acts (in fact, are any?)

    Actually, yes; it is not that many genunely violent acts were not previously illegal, but creating new crimes in order to make some things more readily prosecuted is a common theme. Further, many of them redefine crimes of violence round the edges, moving things in and out of categories. When I referred to a continuous revolution in the criminal law, I was thinking in this case of public order and crimes against the persons and property. There have been vast quantities of regulatory offences created in the last decade or so, but they don’t usually figure in crime statistics.

    The Sexual Offences Act 2003 created numerous new “sex crimes”, many of which involve no physical contact, or which previously would not have been treated as sexual offences. It also moved a lot of boundaries around. Sexual and violent offences are often treated together. Together they are the centre of public (and therefore, political) interest in crime.

    With the exception of murder, and the skeleton of the Offences Against the Person Act, 1861 (which has some bizarre anomalies in it that have never been fixed, despite accelerating legislative fever) the whole treatment of violent crime has been in turmoil for a while. (In England–I know nothing about Scotland.) I don’t think the present Government’s ghetto bully posture: seeking to extract respec’ by the threat of the arbitrary application of power is helping.

    Shannon has a good point. Though it would be interesting to know whether missing persons statistics contain a large number of undetected murders. Not that I want to encourage the official investigation of that question, which would provide yet further excuse for tracking and surveillance. What are we to make of a situation, as in Britain, where murder is pretty stable if rising but recorded crime increases much faster?

  • John East

    I saw these claims on the Crimeinfo website.

    BCS crime figures have been going down since 1995, after a long upward trend since 1981.
    Police recorded figures went down from 1991 until 1997, but have been going up since then.
    In other words, the BCS indicates that crime rose under the Conservatives and dropped steeply under Labour, while recorded crime (police figures) fell under the Conservatives and rose under a Labour government.

    Is there any point in drawing any conclusions from crime statistics when it is obvious that some of them must be lies?

  • guy herbert

    It isn’t necessarily true that either of them is a lie, just because they appear contradictory when reduced to headlines. Social and economic statistics tend to be very difficult to gather and to understand.

    If you want to know how much crime there actually is, BCS is more likely to be of help. The (much lower) rate of recorded crime is more likely to be sensitive to policy, but also to statistical fiddling. In that respect it is much like other government performance figures, including the–joyous, for the more anarchistic among us–paradox that the more you strive to administer something by employing more officials (in this case police officers), the more of it they can discover to administer.

    More policing means higher crime figures. Create enough crimes and tolerate less borderline transgression and the proportion of the population that turns out to be criminal approaches 100%.

  • Verity

    Anyone who trusts any of the Soviet-style figures this government puts out is nuts or awfully naive. In fact, I don’t think awfully naive people trust them any more either.

  • John East

    Guy, I take your point, but you will have to excuse my cynicism. I just don’t buy the theory that we are seeing the facts from different points of view, or that the data is so complex, inter-related, and affected by so many other variables as to be open to opposite interpretations.
    However, if this is the case, then lies or no lies the data is obviously useless.

    Earlier in this thread Shannon made an excellent point, monitor murder rates as an indicator of law and order. If I might change this slightly to, “Monitor deaths associated with criminal acts.” Needless to say, those who would happily distort the figures, i.e. by charging murderers with manslaughter instead of murder, might then find it more difficult to pull the wool over our eyes.

    As for all of the other data, well its compilation might keep a few hundred bureaucrats out of the dole office, otherwise it seems pretty pointless to me.

  • Verity

    Yes, Shannon’s point was very astute and interesting.

  • stoatman

    Cross-posted in part from here: http://www.arrse.co.uk/cpgn2/Forums/viewtopic/t=16070/highlight=Crime+statistics.html

    2. The British Crime Survey.

    This is the government’s preferred measure, principally because it says what they want people to hear – that crime is down. The only thing that’s really certain about the BCS is that if it had shown consistent rises since 1997 and the recorded crime had seen a fall, the Govt would be saying that the BCS is just a survey, and is therefore unreliable.

    The reality is that the BCS is just a survey, and is unreliable.

    The media never go into details, but people are left with the assumption that the BCS is a huge survey of people in Britain. This is not the case, however:

    The 2000 BCS had a sample size of 19,411 people aged 16 and above, plus a further ethnic booster sample of 3,874 (source). That is 0.042% of the British population, and therefore cannot be considered as representative on an epidemiological basis: it’s about 1 in 23,800, i.e. roughly 1 person per mid-sized market town completed the survey.

    Also, the “ethnic booster sample” means that AT LEAST 16% of the survey respondants were ethnic minority (since if the initial sample is representative, 8% of those selected will also be ethnic minority (2001 census)). If we assume that the initial sample is ethnically representative, 1,533 of those asked in this initial sample will also have been minority ethnic. Thus the total proportion of the survey population who are minority ethnic is likely to be 23.2%, i.e. ethnic minorities are over-represented by almost 200% Representative? I think not.

    Of course, the figures gained from the survey are then multiplied up to the whole population of the UK.

    Other problems with surveys are also serious: we don’t know whether the questions asked have been the same from year to year, so they could have been engineered to produce a fall in crime (survey bias). Since only 1 in 23,800 completed it, the chances of completing it twice are slim, and the chances of knowing at least one other person who has completed it before are also slim, so there’s no basis for comparison.

  • stoatman

    Cross-posted in part from here: http://www.arrse.co.uk/cpgn2/Forums/viewtopic/t=16070/highlight=Crime+statistics.html

    2. The British Crime Survey.

    This is the government’s preferred measure, principally because it says what they want people to hear – that crime is down. The only thing that’s really certain about the BCS is that if it had shown consistent rises since 1997 and the recorded crime had seen a fall, the Govt would be saying that the BCS is just a survey, and is therefore unreliable.

    The reality is that the BCS is just a survey, and is unreliable.

    The media never go into details, but people are left with the assumption that the BCS is a huge survey of people in Britain. This is not the case, however:

    The 2000 BCS had a sample size of 19,411 people aged 16 and above, plus a further ethnic booster sample of 3,874 (source). That is 0.042% of the British population, and therefore cannot be considered as representative on an epidemiological basis: it’s about 1 in 23,800, i.e. roughly 1 person per mid-sized market town completed the survey.

    Also, the “ethnic booster sample” means that AT LEAST 16% of the survey respondants were ethnic minority (since if the initial sample is representative, 8% of those selected will also be ethnic minority (2001 census)). If we assume that the initial sample is ethnically representative, 1,533 of those asked in this initial sample will also have been minority ethnic. Thus the total proportion of the survey population who are minority ethnic is likely to be 23.2%, i.e. ethnic minorities are over-represented by almost 200% Representative? I think not.

    Of course, the figures gained from the survey are then multiplied up to the whole population of the UK.

    Other problems with surveys are also serious: we don’t know whether the questions asked have been the same from year to year, so they could have been engineered to produce a fall in crime (survey bias). Since only 1 in 23,800 completed it, the chances of completing it twice are slim, and the chances of knowing at least one other person who has completed it before are also slim, so there’s no basis for comparison.

  • guy herbert

    The reality is that the BCS is just a survey, and is unreliable.

    Rot. The sequel shows that, as usual with “just a survey” people, this opinion can be lumped with those of enthusiasts for quack medicine who reject scientific studies because “statistics don’t account for real people”. Statistics is the best weapon we have. Unfortunately it is hard to do and hard to understand.

    Sure there are lots of potentially valid criticisms of the BCS, but sample size isn’t one of them. It is perfectly standard and valid to boost elements of your raw data, provided you put the quota back together in proper proportions when you are examining the whole. Since the quota sample pa quota sample, which

    [...] we don’t know whether the questions asked have been the same from year to year [...]

    Well, it is perfectly possible to find out. The methodology is published, if not in quite so helpful a form as opinion pollsters produce it. (I do suspect that the Home Office is untrustworthy on the matter of consistent questioning for reasons of studied contempt for the consumers of its statistics and for the professional methods of commercial polls–the Civil Service always knows best–combined with a reflex secretiveness, rather than conspiracy to manipulate the figures.) If one were really concerned about this, a request to the Home Office under the FOI Act ought to produce a copies of the survey forms. But that would be horribly hard work.

  • guy herbert

    Sorry about gobbledegook. I tend to rewrite my posts back to front:

    “Since the quota sample pa quota sample, which”

    should have read,

    “Since social surveys have to work with a quota anyway to get decent useful results, a quota sample which has raw data boosted for ethnic minorities is fine. (One assumes it is needed to get better maginfication on some problems they may have which might otherwise get lost because they are otherwise such a small sub-sample.)”