CityCyclingEdinburgh Forum » Cycling News

Another cycling death on Scotland's roads

(78 posts)

No tags yet.


123

  1. Greenroofer
    Member

    @kaputnik A better statistician than me may be able to chip in with an observation about uncertainty in proportion data (or to tell me that I'm barking up the wrong tree), but I have an inkling that we need to take into account the size of the population from which the KSI's are drawn. There's gazillions of people in cars. There's fewer people on bikes (2% modal share, from SRD above) and the relative size of the source population influences the uncertainty in the reported KSI figures.

    Of course it could be that in every case the population is sufficiently large that this is irrelevant for the purposes of the graphic. I'm just not enough of a statistician to know.

    Whatever the details, the message is stark and pretty unavoidable, and the graphic makes it clear.

    Posted 11 years ago #

  2. kaputnik
    Moderator

    There's probably about 40x more people in cars than on bicycles (I'm pretty sure I see less than 1 cyclist for every 40 cars on the roads though!?)

    I'm not sure how we define the "road travelling" population of Scotland though, even if it was only 1,000,000 people (discounting those not travelling and those on the train), 2% of 1,000,000 is still a pretty substantial population. If that 2% make only 1 bicycle journey per day over a working year of 40 weeks, 5 days a week, that's still some 4,000,000 journeys a year. I guess it's just something that's nigh-on-impossible to quantify accurately though.

    Posted 11 years ago #

  3. Instography
    Member

    This is population data so any question of sampling errors or statistical variability can be ignored. A dead cyclist is a dead cyclist. There's no probability that some of them might not be dead. There's no uncertainty to be taken into account.

    Between any two years you might consider large percentage changes as being an artefact of the small base - if there are only 10 cyclists in any year and one of them dies every year it'll look like at 10% death rate (which it is) but if in one year two die it'll look like at 100% increase but since there's nothing between 1 and 2 the increase is a partial artefact of the small base (even though the 100% increase is also true). What we see here is a continuous trend. Steady upward movement repeated year after year. This is no artefact.

    The politics of the deaths, for me at least, runs like this: I'm not interested in the base. At the behest of government (whether congestion, health, or whatever) or driven by economics, people are taking to their bikes. We can't be clear about the scale of the increase, I'll calculate the errors tomorrow, but it's there. Still, encouraged by government, people are cycling and increasing numbers of them are dying because of it. I don't care if the increase is in line with the increase in cycling, ahead of it or behind it. More people dying is not an acceptable price for the 'gains' from more people cycling.

    Posted 11 years ago #

  4. KarenJS
    Member

    Don't know much about statistics, but from what I do know it sounds like what it's showing is statistically significant and not an artefact of a small sample. I think basing it on all KSIs is correct and shows more impact. I'm always slightly uncomfortable with the focus on deaths anyway, so many lives are ruined through serious injury and trauma, it's almost like they're discounted and we only care about those who have died.
    Two thoughts regarding the graph:
    1) if all the queries can be answered and it's sufficiently robust is there somewhere to present this more formally? Even an msp (thinking Alison Johnstone) bringing it to parliament?
    2) would the press be likely to pick up on it given the obvious trend it shows in contrast to KBs earlier claims?
    I know very little about how either the parliament or press work, so may be barking up the wrong tree, but just seems like this is something that should be more widely publicised?

    Posted 11 years ago #

  5. chdot
    Admin

    "I'm always slightly uncomfortable with the focus on deaths anyway, so many lives are ruined through serious injury and trauma, it's almost like they're discounted and we only care about those who have died"

    Good point.

    I think one reason is that 'dead' is simple.

    SI much greater spread and - more to your point - recovery takes place at different rates and with different degrees of 'success' (a return to full health).

    'We' know, just from people on CCE, that even complete physical recovery - bath/wine/time - can leave residual fear/unease/reluctance to cycle.

    ALL of which discourage cycling by individuals and the wider public.

    Perhaps even more useful than KSIs as they relate to 'random' events would be serious analysis of things like individual roads, junctions, speed, time of day, HGVs etc.

    Will the creation of Police Scotland make that easier/more likely to happen?

    'Deaths' (and presumably SI) led to plans to spend billions on the A9.

    'We' may argue that this is disproportionate. 'We' may be right. Can 'we' prove it?

    Does enough information exist - and is it public?

    Posted 11 years ago #

  6. kaputnik
    Moderator

    I've hummed and hawwed about doing something similar for a while now. But I think with a little more work, and input from the statistically inclined posters of this parish, then this would make a pretty effective postcard to be mailed to every single MSP, with a short message on the back describing the trend and asking them to challenge the Transport Minister on his claims.

    I forget how much 2nd class stamps cost these days but it wouldn't be a massive whip-round to cover the postage costs.

    Printing costs on the likes of Vistaprint would be very low for a few hundred cards.

    Posted 11 years ago #

  7. PS
    Member

    Easy enough to hand deliver to Holyrood and save on the postage, I'd have thought?

    Posted 11 years ago #

  8. kaputnik
    Moderator

    Does it have a letterbox in the front door?

    Posted 11 years ago #

  9. Morningsider
    Member

    If you are trying to make an impact on politicians then don't worry about statistical detail, obviously make sure the numbers are correct - but don't worry about showing a graph with a trend line of increasing cyclist carnage and another showing a relatively static number of cyclists (TS figures for number of trips by bike show a drop, figures for total distance cycled show and very small increase as a proportion of total trips). I would say that is all the detail you need - anything more is liable to confuse them.

    I would try and hang this on the forthcoming draft Scottish budget myself (due out in the next week or so). You could add a third line to the graph showing the static/falling budget for cycling - might be worth getting the postcards to MSPs the day before Keith Brown makes his appearance before the ICI Committee (Transport) on the draft budget (possibly late September/October, gives a chance to work this up).

    Happy to help with any work on this.

    Posted 11 years ago #

  10. KarenJS
    Member

    As long as you go when the building is open I imagine you would be able to hand them into reception, assuming this is what the postmen/women do. I'd be happy to help but am away for the next couple of weeks if it were to be done sooner than later. Maybe at same time send the postcards to newspapers?

    Posted 11 years ago #

  11. kaputnik
    Moderator

    @Morningsider if you can point me in the direction of the cycling budget figures for those years in the graph then that would be swell.

    Posted 11 years ago #

  12. Instography
    Member

    @kaputnik
    The Transport Scotland Data for 2010 - 2011 suggests a modal share increase from 0.8% of all journeys to 1.3%. I would take this with a massive pinch of salt and statistical error, as I don't seriously think that cycling increased by 62% in Scotland in that time.

    Knowing where the data comes from you'd be right to have it appropriately seasoned. You would create a 0.5% increase in cycling by including an extra 75 cyclists in the sample. But since the selection of households and adults within households is random, the odds of doing that are pretty low. You need to keep finding cyclists and not finding non-cyclists. Like tossing a coin and getting a long run of heads. Theoretically possible but unlikely. Also, the inherent bias in household surveys runs against the types of people more likely to be cyclists (males, aged under 50, in employment). Still, you could discount about half of the increase as potentially sampling error. Personally, I never report data with decimal places at all - it implies a level of precision that just isn't there - so for me the 0.8% and 1.3% would both be 1%. No change.

    Can't think what more your graphic needs. It's Government data. It's been through the internal QA processes and published. There's no arguing with it. Agree with Morningsider but happy to help in any way, including stumping up for some printing / postage.

    Posted 11 years ago #

  13. Morningsider
    Member

    @Kaputnik - you have a PM.

    Posted 11 years ago #

  14. chdot
    Admin

    http://www.thecourier.co.uk/news/local/fife/fife-cycling-community-pays-tribute-to-gentleman-alistair-speed-1.128080

    Better class of on-line commenters in Fife.

    Posted 11 years ago #

  15. chdot
    Admin

    http://news.stv.tv/tayside/1322251-julie-watson-who-killed-alistair-speed-on-a91-in-fife-deleted-phone-call/

    Posted 10 years ago #

  16. Nelly
    Member

    She should go to jail for a long, long time. Probably won't though.

    Posted 10 years ago #

  17. le_soigneur
    Member

    Well she's been remanded pending sentencing on dangerous driving so the judge seems to have her number. This & this recent cases do not inspire confidence though

    Posted 10 years ago #

  18. steveo
    Member

    With out annoying video adverts.
    http://www.bbc.co.uk/news/uk-scotland-32999770

    Posted 10 years ago #

RSS feed for this topic

123

Reply

You must log in to post.


Video embedded using Easy Video Embed plugin