The 'Science' article is attached, along with a letter from one of the
authors.
It claims " that the authors of the (Lancet) report have destroyed the
information on where and how they sampled!"
JOHN BIBBY
> -----Original Message-----
> From: Spagat M [mailto:[log in to unmask]]
> Sent: 20 October 2006 16:45
> To: John Bibby; [log in to unmask]; [log in to unmask]
> Subject: RE: Iraq again
>
>
> Dear Mr. Bibby,
>
> Thank you for your interest. Very good question. I'll try to dash off
> a reply with very little time at my disposal. Please excuse some
> roughness.
>
> First let's distinguish two concepts of distance.
>
> 1. How far you have to walk to get somewhere ("normal distance").
> 2. How many times you have to change streets to get somewhere ("network
> distance")
>
> We'll say crossroads to a main road are at network distance 1,
> crossroads to crossroads at distance 2, etc. It's true that some
> houses at network distance 2 will be closer to the main road in terms of
> normal distance than some houses at network distance 1. But the average
> normal distance of houses at network distance 1 will always be lower
> than the average distance of the households at network distance 2. Why?
> Start from a cross road to a cross road to a main street. To travel to
> the main street you must first go to a cross road to the main street.
> Only then can proceed to the main street.
>
> This is all just to say that network distance and normal distance work
> similarly. Both concepts reinforce the main street bias point. The
> Lancet team sampled households that were closer than average to the main
> streets both in terms of normal distance and in terms of network
> distance. But increasing either network distance or normal distance
> will decrease the risk of violence.
>
> Let's put it a different way. They may have sampled some households
> fairly far away from main streets. But any time they did this there
> would have been other neighborhoods still farther away, both in the
> usual and the network senses, that could not have been sampled.
>
> You also wonder if it's possible to check how often they sampled far
> from main streets. Sadly the answer would appear to be know. I'm
> guessing that you saw only our press release and not the related article
> in Science (attached). This article reveals that the authors of the
> report have destroyed the information on where and how they sampled!
> How about that?
>
> I could make a few big points.
>
> In correspondence connected with the Science article one of the authors
> of the Lancet paper defined main roads as "major roads and commercial
> avenues". This would pretty much seem to rule out sampling from
> anyplace rural or without well-paved roads.
>
> Finally, they say that once the street was chosen they numbered the all
> the houses and selected one at random. They didn't have a list of
> households to begin with or they wouldn't be doing this craziness in the
> first place. So they would have had to build one with very limited
> time. (Remember that they conducted 40 interviews per day) So even in
> urban areas it seems unlikely that they would strayed very far from main
> roads.
>
> How this was clear.
>
> Mike Spagat
>
>
>
>
>
> Professor Michael Spagat
> Department of Economics
> Royal Holloway College
> University of London
> Egham
> Surrey
> TW20 0EX
> United Kingdom
> +44 1784 414001 (W)
> +44 1784 439534 (F)
> [log in to unmask]
> http://personal.rhul.ac.uk/uhte/014
>
>
>
> -----Original Message-----
> From: John Bibby [mailto:[log in to unmask]]
> Sent: 20 October 2006 14:12
> To: [log in to unmask]; [log in to unmask];
> [log in to unmask]
> Subject: Iraq again
>
>
> Dear Gentle-people
>
> Interesetd in your report on Iraq - do you have any comments on the
> following?
>
> ==========
> The BBC mentioned a study yesterday that refuted the 600,000 figure on
> the grounds that "houses near main roads were over-represented" (my
> paraphrase).
>
> >From the Lancet article it does appear that households on SHORT street
> near main roads will have been over-represented. (However, households on
> main roads appear not to have been sampled at all.
>
> If "distance from main road" was collected, as it easily could have
> been, it would be possible to attempt some control for this factor.
>
> Has anybody been in touch with the Johns Hopkins people about our
> discussion? (I bet they are fed up with it by now!)
>
> ========
> I was concerned by Triesman's explanation of why the UK government does
> not interest itself with the accuracy of such figures. (see Ted
> Harding's earlier email.)
>
> JOHN BIBBY
>
> ========== From NZ!: http://www.scoop.co.nz/stories/WO0610/S00436.htm
>
> Lancet Iraq Study Flawed: Death Toll Too High
> Friday, 20 October 2006, 10:36 am
> Press Release:
> Lancet Study Fundamentally Flawed: Death Toll Too High
> October 19, 2006 - 1 page -
> For immediate release:
>
> Researchers at Oxford University and Royal Holloway, University of
> London have found serious flaws in the survey of Iraqi deaths published
> last week in the Lancet.
>
> Sean Gourley and Professor Neil Johnson of the physics department at
> Oxford University and Professor Michael Spagat of the economics
> department of Royal Holloway, University of London contend that the
> study's methodology is fundamentally flawed and will result in an
> over-estimation of the death toll in Iraq.
>
> -> The study suffers from "main street bias" by only surveying houses
> -> that
> are located on cross streets next to main roads or on the main road
> itself. However many Iraqi households do not satisfy this strict
> criterion and had no chance of being surveyed.
>
> -> Main street bias inflates casualty estimates since conflict events
> -> such
> as car bombs, drive-by shootings artillery strikes on insurgent
> positions, and market place explosions gravitate toward the same
> neighborhood types that the researchers surveyed.
>
> -> This obvious selection bias would not matter if you were conducting a
> simple survey on immunisation rates for which the methodology was
> designed.
>
> -> In short, the closer you are to a main road, the more likely you are
> -> to
> die in violent activity. So if researchers only count people living
> close to a main road then it comes as no surprise they will over count
> the dead.
>
> During email discussions between the Oxford-Royal Holloway team and the
> Johns Hopkins team conducted through a reporter for Science, for an
> article to be published October 20, it became clear that the authors of
> the study had not implemented a clear, well-defined and justifiable
> methodology. The Oxford-Royal Holloway team therefore believes that the
> scientific community should now re-analyze this study in depth.
>
>
> The team can be reached for comment at;
>
> Gourley: s.gourley1 @ physics.ox.ac.uk mobile:+44 (0) 7733113558
> [log in to unmask] [log in to unmask]
> [log in to unmask]
>
> --
Thank you and Best Regards
JOHN BIBBY aa42/MatheMagic
1 Straylands Grove, York YO31 1EB (01904-330 334) NB: New phone 2005
NB: Visit *** www.tarquinbooks.com *** for a special gift - put 'aa42'
after your name when ordering.
All statements are on behalf of aa42.com Limited, a company wholly owned by
John Bibby and Shirley Bibby. See www.aa42.com/mathemagic and
www.mathemagic.org
> -----Original Message-----
> From: email list for Radical Statistics
> [mailto:[log in to unmask]]On Behalf Of John Bibby
> Sent: 20 October 2006 14:11
> To: [log in to unmask]
> Subject: Iraq again
>
>
> The BBC mentioned a study yesterday that refuted the 600,000 figure on the
> grounds that "houses near main roads were over-represented" (my
> paraphrase).
>
> >From the Lancet article it does appear that households on SHORT
> street near
> main roads will have been over-represented. (However, households on main
> roads appear not to have been sampled at all.
>
> If "distance from main road" was collected, as it easily could
> have been, it
> would be possible to attempt some control for this factor.
>
> Has anybody been in touch with the Johns Hopkins people about our
> discussion? (I bet they are fed up with it by now!)
>
> ========
> I was concerned by Triesman's explanation of why the UK
> government does not
> interest itself with the accuracy of such figures. (see Ted Harding's
> earlier email.)
>
> JOHN BIBBY
>
> ========== From NZ!: http://www.scoop.co.nz/stories/WO0610/S00436.htm
>
> Lancet Iraq Study Flawed: Death Toll Too High
> Friday, 20 October 2006, 10:36 am
> Press Release:
> Lancet Study Fundamentally Flawed: Death Toll Too High
> October 19, 2006 – 1 page –
> For immediate release:
>
> Researchers at Oxford University and Royal Holloway, University of London
> have found serious flaws in the survey of Iraqi deaths published last week
> in the Lancet.
>
> Sean Gourley and Professor Neil Johnson of the physics department
> at Oxford
> University and Professor Michael Spagat of the economics
> department of Royal
> Holloway, University of London contend that the study’s methodology is
> fundamentally flawed and will result in an over-estimation of the
> death toll
> in Iraq.
>
> -> The study suffers from "main street bias" by only surveying houses that
> are located on cross streets next to main roads or on the main
> road itself.
> However many Iraqi households do not satisfy this strict criterion and had
> no chance of being surveyed.
>
> -> Main street bias inflates casualty estimates since conflict events such
> as car bombs, drive-by shootings artillery strikes on insurgent positions,
> and market place explosions gravitate toward the same neighborhood types
> that the researchers surveyed.
>
> -> This obvious selection bias would not matter if you were conducting a
> simple survey on immunisation rates for which the methodology was
> designed.
>
> -> In short, the closer you are to a main road, the more likely you are to
> die in violent activity. So if researchers only count people
> living close to
> a main road then it comes as no surprise they will over count the dead.
>
> During email discussions between the Oxford-Royal Holloway team and the
> Johns Hopkins team conducted through a reporter for Science, for
> an article
> to be published October 20, it became clear that the authors of the study
> had not implemented a clear, well-defined and justifiable methodology. The
> Oxford-Royal Holloway team therefore believes that the scientific
> community
> should now re-analyze this study in depth.
>
>
> The team can be reached for comment at;
>
> Gourley: s.gourley1 @ physics.ox.ac.uk mobile:+44 (0) 7733113558
> Johnson: n.johnson @ physics.ox.ac.uk
> Spagat: M.Spagat @ rhul.ac.uk
>
--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.408 / Virus Database: 268.13.5/483 - Release Date: 18/10/2006
******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
Disclaimer: The messages sent to this list are the views of the sender and cannot be assumed to be representative of the range of views held by subscribers to the Radical Statistics Group. To find out more about Radical Statistics and its aims and activities and read current and past issues of our newsletter you are invited to visit our web site www.radstats.org.uk.
*******************************************************
|