Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
Good news everyone! The Boards.ie Subscription service is live. See here: https://subscriptions.boards.ie/

Do you trust this set of statistics or not and why ?

  • 19-02-2011 06:14PM
    #1
    Closed Accounts Posts: 9,273 ✭✭✭


    I thought it might be interesting to have a dedicated thread to discuss the merits (or not) of this particular set of new statistics which purport to be the worlds first definitive WW2 Pow mortality rates :

    146567.JPG

    These have been posted by a member here as being reliable. Questioning of them has been the cause of puzzlement to the person who posted them :
    It is hard to understand from posts by yourself and others exactly what the agenda is here. Specifically I mean refusing to accept legitimate source material, refusal to accept facts as represented in statistic tables, even plain refusal to accept points that are backed by the views of respected historians.

    I had and continue to have strong doubts on this set of 'statistics'/Jpeg. So I thought it might be interesting to have a wider discussion dedicated to this specific set of data in the hopes of uncovering more about how they were arrived at and so on.

    They originate on wikipedia here:

    http://en.wikipedia.org/wiki/Other_Losses

    they are used half way down the page as a rebuttal against the author of a book.

    Even though the creator of the statistics is not a WW2 author or historian, he has not written on WW2 (he has written on the Rothschilds and is working on Kissingers papers) these figures are presented on wikipedia as conclusive, devoid of context or qualification or margin of errror. There is a surprising lack of information on them. There is no evidence at all as to how they were arrived at, what data was used and how, what was excluded from which archives and so on. It is simply a jpeg on wiki with a footnote which is :

    Niall Ferguson tabulated the total death rate for POWs in World War II as follows:[89]

    89=# ^ Ferguson 2004, p. 186

    Thoughts ?

    Reliability of Ferguson POW mortality statistics on wiki 12 votes

    Yes the Ferguson statistics are reliable and conclusive
    0%
    No the Ferguson statistics are unreliable and inconclusive
    33%
    jonniebgood1MajorMaxdarlettNelly100 4 votes
    Don't know - not sure
    66%
    marcsignaljam_mac_jamMahatma coatMorlarDonny5arnhem44getzgaelicred 8 votes


«1

Comments

  • Closed Accounts Posts: 1,827 ✭✭✭Donny5


    Don't know - not sure
    I wouldn't trust any statistics when the source is unknown. Where do the datasets come from? What was the methodology? What are the margins of error?


  • Registered Users, Registered Users 2 Posts: 4,072 ✭✭✭marcsignal


    Don't know - not sure
    Donny5 wrote: »
    I wouldn't trust any statistics when the source is unknown. Where do the datasets come from? What was the methodology? What are the margins of error?

    Very well put, if I may say so.

    One could, and should, feel the same way about survey results.
    Survey results can be notoriously unrealible, and are generally meaningless, in terms of reaching a consensus, unless they include the other vital information necessary to properly analyse them.

    Any survey result worth its' salt, must also include, at the very least:

    1) The list of questions asked (so as one can more accurately determine the context)

    2) Details on the demographic of the respondants (socio economic group, nationality, religion, level of education etc.)

    .


  • Closed Accounts Posts: 1,827 ✭✭✭Donny5


    Don't know - not sure
    Yes, that's very true. If the basic methadology isn't disclosed, you can instantly discount the survey as useless. If it is disclosed, you usually need to see the raw results before editorialisation for the data to be any worth.


  • Registered Users, Registered Users 2 Posts: 1,368 ✭✭✭arnhem44


    Don't know - not sure
    Wiki is known for the information of a lot of articles to be inaccurate,if there is no proof of where these statistics have arrived from then how can these statistics be taken seriously.If the Wiki article was to show people how these percentages were originally gathered then it may help to sway opinion but until then I'm with the lads above.


  • Registered Users, Registered Users 2 Posts: 3,578 ✭✭✭jonniebgood1


    No the Ferguson statistics are unreliable and inconclusive
    arnhem44 wrote: »
    Wiki is known for the information of a lot of articles to be inaccurate,if there is no proof of where these statistics have arrived from then how can these statistics be taken seriously.If the Wiki article was to show people how these percentages were originally gathered then it may help to sway opinion but until then I'm with the lads above.

    You are correct about Wiki- In this case however the original poster has deliberately misled people who read/ vote in this poll to try and get support for his agenda. he incorrectly says
    They originate on wikipedia here:

    http://en.wikipedia.org/wiki/Other_Losses
    .
    They actually originate in "Prisoner Taking and Prisoner Killing in the Age of Total War: Towards a Political Economy of Military Defeat", by Niall Ferguson (Professor of History at Harvard). This is also stated on the wiki article but the OP chose to not put in the source as he wishes people to vote in a particular way. Your query about how they are tabulated is valid and would be down to the professors methods of calculation and I suppose some trust in his standing as a historian.


  • Advertisement
  • Registered Users, Registered Users 2 Posts: 3,578 ✭✭✭jonniebgood1


    No the Ferguson statistics are unreliable and inconclusive
    Morlar wrote: »

    They originate on wikipedia here:

    http://en.wikipedia.org/wiki/Other_Losses

    they are used half way down the page as a rebuttal against the author of a book.

    Even though the creator of the statistics is not a WW2 author or historian, he has not written on WW2 (he has written on the Rothschilds and is working on Kissingers papers) these figures are presented on wikipedia as conclusive, devoid of context or qualification or margin of errror. There is a surprising lack of information on them. There is no evidence at all as to how they were arrived at, what data was used and how, what was excluded from which archives and so on. It is simply a jpeg on wiki with a footnote which is :

    Niall Ferguson tabulated the total death rate for POWs in World War II as follows:[89]

    89=# ^ Ferguson 2004, p. 186

    Thoughts ?

    Thoughts? hmmm.

    Seriously- could you not just ask the question straight without having to lie about the origin of the table. And also trying to put your own slant on them later on. This type of deception is unfair on people who view the poll from a neutral point of view. It also shows a lack of trust in your own assertions that you must do this.

    For your information I will post alternative information on the thread regarding Soviet POW's that is broadly in line with the extraordinarily high mortality rate in German run camps for Soviet POW's to back up Fergusons figures.


  • Closed Accounts Posts: 1,827 ✭✭✭Donny5


    Don't know - not sure
    For your information I will post alternative information on the thread regarding Soviet POW's that is broadly in line with the extraordinarily high mortality rate in German run camps for Soviet POW's to back up Fergusons figures.

    It doesn't matter whether you can post information that doesn't contradict Ferguson's firgures. If the origin of the data, methadology of collection and analysis and the margins of error are not availble, the statistics can not be trusted. It doesn't mean they're wrong, just not trustworthy.


  • Registered Users, Registered Users 2 Posts: 3,578 ✭✭✭jonniebgood1


    No the Ferguson statistics are unreliable and inconclusive
    Donny5 wrote: »
    It doesn't matter whether you can post information that doesn't contradict Ferguson's firgures. If the origin of the data, methadology of collection and analysis and the margins of error are not availble, the statistics can not be trusted. It doesn't mean they're wrong, just not trustworthy.

    If this level of cross checking of data is required it would make it very difficult for any publication of historical data to be of any use. I agree with being thorough on data but I also think that in reading history that if the author is respectable that their opinions can be trusted as accurate. Thats just my opinion though- I would read such tables as a type of summary of data rather than calculating the figures myself.

    The purpose of quoting alternative figures to reach a consensus on whether Fergusons figures are accurate. My point in this would be that is different sources reach similar conclusions then the conclusions should be accepted as accurate. ???


  • Moderators, Society & Culture Moderators Posts: 9,817 Mod ✭✭✭✭Manach


    My own initial reaction is to accept them as accurate ( though I'd acknowledge the valid points raised by previous posters about datasets and methologies etc.)
    - numerous books which mentioned the poor conditions and high death rates in PoW camps: Lastest one was T Dungan's "V-2" about slave PoW labour.
    - Neill Ferguson has been offered as a reliable source in an academic history module I've completed, so this would suggest him being a reliable interpreter of primary evidence.


  • Closed Accounts Posts: 2,494 ✭✭✭citizen_p


    first of all I'd view the germans held by eastren europeans and by russians as the same. except for partisans etc... in the balkans.

    too come up with anyfigure so exacting endless research would be need in various national archives.


  • Advertisement
  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    Mousey- wrote: »
    too come up with anyfigure so exacting endless research would be need in various national archives.

    That would be pretty much my take on it.

    *
    I do think it is noteworthy that one of the people voting 'Yes the Ferguson statistics are reliable and conclusive' is a person with a zero post count who joined today :

    * Total Posts: 0
    * Posts Per Day: 0
    * Find all posts by Nelly100
    * Find all threads started by Nelly100

    Total Thanks

    * Thanked 0 Times in 0 Posts
    * Find all thanked posts by Nelly100
    * Find all posts thanked by Nelly100

    General Information

    * Last Activity: Today 18:06
    * Join Date: 20-02-2011
    * Referrals: 0


  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    I believe these 2 posts here belong more in this thread than in the Double Standards one :
    to discount his [Ferguson's] opinion you need to provide information rather than just opinion. Anything other than this would quite simply suggest a pro-Nazi bias

    http://boards.ie/vbulletin/showpost.php?p=70770597&postcount=369

    & response :

    http://boards.ie/vbulletin/showpost.php?p=70771878&postcount=370


  • Registered Users, Registered Users 2 Posts: 5,932 ✭✭✭hinault


    I would go along with Donny's assertion.
    Unless the methodology pertaining to a survey is disclosed one cannot verify the validity of the statistics produced.

    Did Ferguson not supply a bibliography in his book detailing where the statistics were sourced from?
    If he did then the reference in the bibliography should disclose the sample size and the methodology used to conduct the survey.


  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    hinault wrote: »
    I would go along with Donny's assertion.
    Unless the methodology pertaining to a survey is disclosed one cannot verify the validity of the statistics produced.

    Did Ferguson not supply a bibliography in his book detailing where the statistics were sourced from?
    If he did then the reference in the bibliography should disclose the sample size and the methodology used to conduct the survey.

    I will be honest with you I don't even believe they originated in a book, not least that the book was the central subject for such a global comprehensive study (which would be required for the figures to be taken as credible). I now believe they were presented in a magazine article which no one on boards seems to have access to. All that is available so far (as has been presented thus far across multiple boards.ie threads) is exactly what is on wiki article - and that is the sum total of . . . . . a jpeg chart with absolutely no context or supplementary information whatsoever. There really is that little information on this fergie study at this point. If anyone can point to a thread posted up to this date on boards.ie where more information on the methodology, sources etc has been presented by jonniebegood (the poster asserting this is a credible chart and that to question it's cerdibility denotes a nazi bias) then I would be intersted to see. Otherwise at this point (feb 21 2011) based on what is presented so far here, in my view the case for these statistics remains severely lacking.


  • Banned (with Prison Access) Posts: 6,869 ✭✭✭Mahatma coat


    Don't know - not sure
    Couldnt find that book in any of the bookshops I went to Yesterday whilst trying to avoid the Vicious humidity, None, and I tried as many bookshops as possible.

    So on the face of it a .JPEG of unconfirmed origin and no Backrouind to how the figures were collected, make it useless as anything more than anecdotal reference.

    WE could all do that


    Total World Population of Jews 1938 & 1948

    America Europe Asia Africa Oceania Total

    1938 5,343,319 8,939,608 839,809 598,339 27,016 15,748,091

    1948 5,198,219 9,372,666 572,930 542,869 26,954 15,763,638

    Diff -145,100 +433,058 -266,879 -55,470 -9,938 +15,547


    See theres a Table which has been Published in a Book and everything, not only that but I can provide the backround documents that lead to these numbers.

    Somehow tho I doubt that many people will put much credence in them


  • Registered Users, Registered Users 2 Posts: 3,592 ✭✭✭Tristram


    Full Text (PDF) available to members at SAGE Journals Online

    Niall Ferguson, Prisoner Taking and Prisoner Killing in the Age of Total War: Towards a Political Economy of Military Defeat War in History April 2004 11: 148-192


  • Registered Users, Registered Users 2 Posts: 3,592 ✭✭✭Tristram


    Niall Ferguson. The War of the World: Twentieth-Century Conflict and the Descent of the West. New York: Penguin Books, 2007. lxxi + 808 pp. $18.00 (paper), ISBN 978-0-14-311239-6; $35.00 (cloth), ISBN 978-1-59420-100-4.

    Reviewed by Talbot Imlay, Published on H-German (April, 2008)

    http://www.h-net.org/reviews/showrev.php?id=14466


  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    Tristram wrote: »
    Niall Ferguson. The War of the World: Twentieth-Century Conflict and the Descent of the West. New York: Penguin Books, 2007. lxxi + 808 pp. $18.00 (paper), ISBN 978-0-14-311239-6; $35.00 (cloth), ISBN 978-1-59420-100-4.

    Reviewed by Talbot Imlay, Published on H-German (April, 2008)

    http://www.h-net.org/reviews/showrev.php?id=14466

    The one above :

    Full Text (PDF) available to members at SAGE Journals Online

    Niall Ferguson, Prisoner Taking and Prisoner Killing in the Age of Total War: Towards a Political Economy of Military Defeat War in History April 2004 11: 148-192

    seems to be the source of the wiki-jpeg/table of statistics. However it is not a book it is a Journal which might explain why no one here has so far claimed to have read it.


  • Registered Users, Registered Users 2 Posts: 3,578 ✭✭✭jonniebgood1


    No the Ferguson statistics are unreliable and inconclusive
    Morlar wrote: »
    The one above :

    Full Text (PDF) available to members at SAGE Journals Online

    Niall Ferguson, Prisoner Taking and Prisoner Killing in the Age of Total War: Towards a Political Economy of Military Defeat War in History April 2004 11: 148-192

    seems to be the source of the wiki-jpeg/table of statistics. However it is not a book it is a Journal which might explain why no one here has so far claimed to have read it.

    Maybe yourself or Mahatma will let us know if its reliable when you've analysed it further!


  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    Not withstanding that you have in no way discredited Fergusons ability or record in this regard I wonder/ point out:

    1. Do you accept that if Fergusons figures are verified by alternative sources that it verifies their accuracy? i.e. do you just refuse to accept statistics as an important and fundamental way of analysing POW treatment?

    2. They are mortality rates of POW in WWII: the context is the disregard for human life in POW camps in WWII. The calculation is based on the number of prisoners who died after being taken in charge of an enemy as opposed to being released back after hostilities end, be it immediately or in 1953 or 55. This is obvious points. Why don't you clarify the doubts you have about the figures rather than this ridiculous pandering behind context or supplementary information. Perhaps if you can do this honestly we could at least agree to disagree on the matter.

    3. The figures show a global figure- You are welcome to dispute any of them if you can substantiate your issue. I don't accept your view as holding much weight therefore I must insist on you backing any view up with a source which I can review. (similarily I would'nt expect you to take my word for these figures, hence I defer to figures of a highly rated historian).



    This is truly, truly moronic.

    I have outlined repeatedly the reasons for not finding a wiki-jpeg, which lacks context or supplementary information, to be credible as a definitive source of the global Pow mortality rates 1939-1955.

    1: If the figures are repeated it has no relevance as to whether or not I find the original study behind those figures, and the methods used and circumstances, depth and so on of that original study to be adequate. You will always find some idiot somewhere who mindlessly repeats a statistic with no interest in how it was arrived at.

    2 & 3 the information you supply in part 2 & 3 of your 'question', shows a lack of understanding of the word 'context'. When I mentioned the 'context of the study', I mentioned several other things also, I also asked for supplementary information. When I used the word 'context' in this case it would include, are the figures designed to be definitive ? How many researchers worked on this and for how long, who were they and what was their brief, which parameters were used ? Which archives were used, Which factors were counted and discounted and so on. That would be one part of the context and the general supplementary inforamtion which is still unknown, and would be required before making a reasoned conclusion on the accuracy of these figures.

    Regarding the parameters used in this ferguson study which you mention, can you confirm where you got this informtion from ?
    I am referring to where you say :

    The calculation is based on the number of prisoners who died after being taken in charge of an enemy as opposed to being released back after hostilities end, be it immediately or in 1953 or 55. This is obvious points.

    Is this based on your assumption ? Or is there anything more to it than that ?


  • Advertisement
  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    Maybe yourself or Mahatma will let us know if its reliable when you've analysed it further!

    I take that as your confirmation that you have not read this edition of this journal either. Therefore you base your belief on these definitive, global, pow mortality rates on a wiki-jpeg which has no context or supplementary information whatsoever. And you go on to claim that people who do not agree with your approach :
    to discount his [Ferguson's] opinion you need to provide information rather than just opinion. Anything other than this would quite simply suggest a pro-Nazi bias


  • Registered Users, Registered Users 2 Posts: 5,932 ✭✭✭hinault


    Posting the ISBN and a review of the book, brings us no closer to verifying the basis of Ferguson's statistics concerning POW casualty levels in the book.

    I am almost tempted to buy this book to see if Ferguson has an explicit reference in it to the source of the statistics supplied.


  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    hinault wrote: »
    Posting the ISBN and a review of the book, brings us no closer to verifying the basis of Ferguson's statistics concerning POW casualty levels in the book.

    I am almost tempted to buy this book to see if Ferguson has an explicit reference in it to the source of the statistics supplied.

    The ISBN refers to a book, the statistics are referenced in the book, (and possibly expanded upon in that book).

    However the book is not the source of the statistics.

    The article is the source. This was published in a journal and not in book form (to the best of my knowledge).

    So my point is buying the book you may (and this is a guess) find that it simply references the article as being the source.


    I think this post here puts the issue very well from my perspective:

    http://boards.ie/vbulletin/showpost.php?p=70779613&postcount=374
    Donny5 wrote: »
    You see, here is the problem. It's not domain specific, so you don't have to worry about pro- or anti-nazi biases, since the issue is universal. When you make an assertion, like Ferguson is credited with doing, then it is your responsibility to provide the sources and methods you used in generating that assertion. It's not taken as fact until discredited. It's taken as supposition until proven. That's how it works and that's the only way it can work.

    If you want to use those figures as evidence for an argument you are making, then it is on you to prove that the figures are correct. As it stands, the figures have no merit to me because the datasets and metadata aren't available. That means that any argument based on them becomes purely supposition.

    The basis under which I refuse to accept to his figures is that they are not supported by evidence.


  • Registered Users, Registered Users 2 Posts: 3,578 ✭✭✭jonniebgood1


    No the Ferguson statistics are unreliable and inconclusive
    And an alternative perspective from same thread, if I could be so bold as to seek balance
    These same high standards could be applied to most threads then? For example this thread was initially based upon http://www.bbc.co.uk/news/world-europe-12059475 and
    http://ec.europa.eu/justice/doc_cent..._crimes_en.pdf

    I do not know how the BBC or Dr. Carlos Closa Montero broke down the data they had to conclude as they did. I do not see the data/evidence that they chose not to include. As you will note- for me to seriously take this view would make any discussion of this as a subject impossible.

    This is I feel what is being attempted in the POW issue.
    As I am not likely to have access to the historians own notes and figures it is extremely unlikely that I could show "the datasets and metadata" in this regard. The same would be the case for most historiographical data represented on these forums in my opinion?

    I have attempted to contact the author to seek his clarification on how he put these figures together. He is currently on academic leave and thus I do not expect a reply. If I receive one I will post it. My point remains that as a respected historian his methods demand respect. ref: http://boards.ie/vbulletin/showpost.php?p=70772088&postcount=10


  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    I do not know how the BBC or Dr. Carlos Closa Montero broke down the data they had to conclude as they did. I do not see the data/evidence that they chose not to include. As you will note- for me to seriously take this view would make any discussion of this as a subject impossible.

    This is I feel what is being attempted in the POW issue.
    As I am not likely to have access to the historians own notes and figures it is extremely unlikely that I could show "the datasets and metadata" in this regard. The same would be the case for most historiographical data represented on these forums in my opinion?

    One person or more choosing NOT to accept the ferguson figures in their current format as being definitive, does not make discussion of this subject impossible.


  • Registered Users, Registered Users 2 Posts: 3,578 ✭✭✭jonniebgood1


    No the Ferguson statistics are unreliable and inconclusive
    Morlar wrote: »
    One person or more choosing NOT to accept the ferguson figures in their current format as being definitive, does not make discussion of this subject impossible.

    Indeed, nor does a person who choses for there own reasons to ignore a reality make that reality untrue. http://www.boards.ie/vbulletin/showpost.php?p=70778596&postcount=21


  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    Indeed, nor does a person who choses for there own reasons to ignore a reality make that reality untrue. http://www.boards.ie/vbulletin/showpost.php?p=70778596&postcount=21

    Can you clarify exactly which 'reality' I have ignored ?

    If by 'reality' you mean 'a set of statistics with no context or supplementary information', and by 'ignored' you mean 'not believe to be 100% credible' then you might have a point. Otherwise you don't.

    Btw that post you linked to had a question I asked to you which you have so far not answered :
    Regarding the parameters used in this ferguson study which you mention, can you confirm where you got this informtion from ?

    I am referring to where you say :

    The calculation is based on the number of prisoners who died after being taken in charge of an enemy as opposed to being released back after hostilities end, be it immediately or in 1953 or 55. This is obvious points.

    Is this based on your assumption ? Or is there anything more to it than that ?


  • Registered Users, Registered Users 2 Posts: 3,578 ✭✭✭jonniebgood1


    No the Ferguson statistics are unreliable and inconclusive
    Morlar wrote: »
    Can you clarify exactly which 'reality' I have ignored ?
    :

    The reality being the difference between the German treatment of Russian POW's in comparison to all other POW's in that conflict.
    Morlar wrote: »
    If by 'reality' you mean 'a set of statistics with no context or supplementary information', and by 'ignored' you mean 'not believe to be 100% credible' then you might have a point. Otherwise you don't.
    Not for the first time, wrong. See above for reality.
    Morlar wrote: »
    Btw that post you linked to had a question I asked to you which you have so far not answered :

    You should refer back to the table of figures. Theres not that much text in it but if you read it carefully you will see what Ferguson has compiled. What you call my 'assumption' was me trying to explain in simple terms to you what is described there as I was unsure if you genuinely didnt understand it (as opposed to not wanting to understand it)


  • Closed Accounts Posts: 9,273 ✭✭✭Morlar


    Don't know - not sure
    The reality being the difference between the German treatment of Russian POW's in comparison to all other POW's in that conflict.

    If your reality is determined from a jpeg of a chart that has no context or information on how it was arrived at, then it would not be surprising if you were often wrong.
    You should refer back to the table of figures.

    here is the question you have not yet answered :
    Regarding the parameters used in this ferguson study which you mention, can you confirm where you got this informtion from ?

    I am referring to where you say :


    The calculation is based on the number of prisoners who died after being taken in charge of an enemy as opposed to being released back after hostilities end, be it immediately or in 1953 or 55. This is obvious points.

    Is this based on your assumption ? Or is there anything more to it than that ?

    Would it be too much to ask for a clear answer to this ?


  • Advertisement
  • Registered Users, Registered Users 2 Posts: 3,578 ✭✭✭jonniebgood1


    No the Ferguson statistics are unreliable and inconclusive
    I thought repeating questions was something that you did not like
    Morlar wrote: »

    Would it be too much to ask for a clear answer to this ?

    http://www.boards.ie/vbulletin/showpost.php?p=70810816&postcount=29


Advertisement