RE: [tips] The joy of stats
Rick Froman writes: I nominate this thread as the most inappropriately (or ironically) named thread in the history of TIPS... In fact, my stats classes are a laugh riot compared to this dreary sequence of ad hominem attacks. Rick: I appreciate your exasperation with the way this thread has gone, but as I figure among this “dreary sequence” I would point out that Chris commended (for the third time) *The Spirit Level* in this thread, and that induced me to provide the criticisms of the book that I cited. As these centre around its controversial use of statistics (and the authors’ propensity towards a “correlations demonstrate cause” viewpoint), that has some relevance to the use and (possible) misuse of statistics. And I hope you will find that I do not figure in the “sequence” of ad hominem attacks that have disfigured the later stages of the thread. Allen Esterson Former lecturer, Science Department Southwark College, London allenester...@compuserve.com http://www.esterson.org RE: [tips] The joy of stats Rick Froman Fri, 17 Dec 2010 11:59:07 -0800 I nominate this thread as the most inappropriately (or ironically) named thread in the history of TIPS given the extent to which each subsequent post has clearly sucked all of the joy out of the list. In fact, my stats classes are a laugh riot compared to this dreary sequence of ad hominem attacks. Are there any other nominees before we vote? Rick Dr. Rick Froman, Chair Division of Humanities and Social Sciences Box 3055 x7295 rfro...@jbu.edu http://tinyurl.com/DrFroman --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7323 or send a blank email to leave-7323-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re:[tips] The joy of stats
Chris Green writes: Again, Allen overwhelms with far more bulk than (I feel) it is appropriate to respond to in a forum such as this. So who decides what is appropriate as a response to another TIPSters contentions? I could have quoted just a couple of the academic critics of Wilkinson Pickett’s thesis, but that would have left open the possibility that these were maverick opinions. I could have simple cited the articles, but three of them were lengthy critiques, and which TIPSters have the time to plough through them unless they have a good reason for doing so? That is why I extracted their conclusions, and showed that a wide range of relevant academics found WP’s thesis wanting. And why Chris, who told us previously that he was “enjoying” reading *The Spirit Level*, a book heavy with figures, statistics and diagrams that need thoughtful examination, should find a total of some 700 words in criticism of the book a chore is a puzzle to me. I will say, in response to Kay's critique, that one should look at the whole book. It is written for a popular audience, and so does not include correlations coefficients in the main text, but they are included in appendix. A fair point, though it doesn’t deal with WP’s over-confident contentions based on very-scattered scatter diagrams the lines through which are heavily dependent on a few outliers. Nor does it deal with the fact that the reviewers I cited are all working (or have worked) in the fields of economics and/or sociology, so their criticisms are not simply based on the book itself (as is made clear in the lengthier reviews). The burden of proof is on the critic to provide an equally compelling case for some other variable. I note that, keeping in mind Mill’s dictum that “He who knows only his own side of the case, knows little of that”, Chris did not answer my request for critical responses to the book that he has sought out. In fact some of the academics I cited do reference articles (some authored or co-authored by themselves) that make a case for different factors. For instance Goldthorpe cites such articles in his references, and Leigh cites two articles he has co-authored that run counter to one aspect of WP’s contentions. (The same is true for other scholarly critiques that I did not cite.) an equally compelling case What is evident from the reviews I cited is that WP’s thesis is far from compelling for quite a few academics who have considerable knowledge of the relevant subject matter. Allen Esterson Former lecturer, Science Department Southwark College, London allenester...@compuserve.com http://www.esterson.org -- From: Christopher Green chri...@yorku.ca Subject:Re: The joy of stats Date: Thu, 16 Dec 2010 08:23:56 -0500 Again, Allen overwhelms with far more bulk than (I feel) it is appropriate to respond to in a forum such as this. I will say, in response to Kay's critique, that one should look at the whole book. It is written for a popular audience, and so does not include correlations coefficients in the main text, but they are included in appendix. As for it seeming a bit utopian, yes it is, in the sense that there is essentially zero chance the the populations of the US and UK, especially, will ever accept these recommendations. Inequality has (sadly) become an entrenched part of their national ideologies - the pain that signals to them they are getting maximum gain from their economies. That people refuse to do what is good for them doesn't mean, however, that it wouldn't sill be good for them. For myself, I find the evidence provided in the book quite compelling. As with all correlational research, it is possible that there is something else, correlated with equality, going on. The authors are aware of this issue and address it extensively. The burden of proof is on the critic to provide an equally compelling case for some other variable. Chris --- Christopher D Green Department of Psychology York University Toronto, ON M6C 1G4 Canada - From: Allen Esterson allenester...@compuserve.com Subject:Re: The joy of stats Date: Thu, 16 Dec 2010 06:51:06 -0500 (EST) On 15 December Chris Green cited evidence he thinks is supportive of the conclusions of the book I was pushing a few weeks ago: _The Spirit Level: Why Equality is Better for Everyone_. As I wrote in a post under the thread on “objectivity” recently, while one can never completely remove one’s own biases and preconceptions in assessing a thesis, one of the basic principles is that one should seek out writings critical of the thesis before drawing any firm conclusions. (John Stuart Mill puts the same notion rather more strongly in his essay “On Liberty”: “He who knows only his own side of the case, knows little of that.”) This is the third time in recently weeks that Chris has “pushed” *The Spirit Level*. So I ask of Chris, which
Re:[tips] The joy of stats
Michael Smith writes: One funny thing in Allen's post was: ...who regards himself as “about as anti-inequality an economist as you’ll find”)... Well I guess that settles that. This is proof positive that God exists. We have at last found a truly objective (unbiased) individual who is, miraculously, a government worker !! Of course nowhere does Andrew Leigh remotely suggest or imply that his contribution is truly objective or unbiased when he prefaced his critical comment with “I’m about as anti-inequality an economist as you’ll find”. Michael has completely misconstrued the point of his doing so. Leigh knows that a standard response to criticism of a thesis such as that of Wilkinson and Pickett is that they come from the political Right, and are ideologically biased, so what do you expect? And indeed that is precisely what we find from Wilkinson – he castigates critiques as coming from “right wing institutes” that he describes as “professional wreckers of ideas”: http://www.guardian.co.uk/books/2010/aug/14/the-spirit-level-equality-thinktanks As for Leigh being “a government worker”, that is a rather strange way of describing a Labor member of the Australian House of Representatives. Michael’s implied notion (Leigh is “miraculously, a government worker!!”) that this fact goes some way (at least) to explaining his critical comments on Wilkinson Pickett’s thesis defies logical explanation. I think Chris made a good point that it is a popular book and so perhaps focuses on readability. Perhaps you should have read the lengthier academic reviews. Then you would have seen that the criticisms were not simply based on the book. And for it's critics to point out that it doesn't present detailed statistical analyses, is I think, ludicrous. It isn't after all, a journal article. So a comment by one critic has metamorphosed into a generalised “critics”. And to reiterate my comment immediately above, the lengthier reviews were not simply directed at the book but were on WP’s thesis more generally, as the authors have expounded it elsewhere. But the most hillarious one is from one critics response that Allen presents that includes: The evidence presented in the book is mostly a series of scatter diagrams, with a regression line drawn through them. Michael conveniently leaves out from the quotation: “If you remove the bold lines from the diagram, the pattern of points mostly looks random, and the data dominated by a few outliers. Allen Esterson Former lecturer, Science Department Southwark College, London allenester...@compuserve.com http://www.esterson.org --- From: Michael Smith tipsl...@gmail.com Subject:Re: The joy of stats Date: Thu, 16 Dec 2010 09:35:47 -0600 I think I'll try to match Allen's lengthy response. I haven't read it (The Spirit Level) either (although I think I have it somewhere and have been planning to). I think Chris made a good point that it is a popular book and so perhaps focuses on readability. This, however, doesn't mean the analysis is poor. And for it's critics to point out that it doesn't present detailed statistical analyses, is I think, ludicrous. It isn't after all, a journal article. But the best part of Allen's response is the funny parts. One funny thing in Allen's post was: ...who regards himself as “about as anti-inequality an economist as you’ll find”)... Well I guess that settles that. This is proof positive that God exists. We have at last found a truly objective (unbiased) individual who is, miraculously, a government worker !! But the most hillarious one is from one critics response that Allen presents that includes: The evidence presented in the book is mostly a series of scatter diagrams, with a regression line drawn through them. This is hillarious !!! The reason being, of course, is that the statement is a good description of all the results in sociology. I'm still laughing at that one. Thanks Allen. --Mike --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7269 or send a blank email to leave-7269-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
On Fri, 17 Dec 2010 03:49:01 -0800, Michael Smith wrote: Again Palij you miss the points. It's pretty funny how you do that so consistently, but not surprising. You don't confuse me by hiding your 'logic' behind being long winded...it's just really boring. Maybe instead of giving your interpretation of wikipedia, just provide a title and the link that should cut down your posts by probably 60%. lol Smith, you are a clown. You lack scholarship, you are inarticulate, you are a coward when confronted with your own errors, you confuse your opinions with the facts, and lack intellectual integrity. You laugh because that is all you have. You make statements that lack validity and are overgeneralizations and when called on them cannot respond with anything else. Laugh, clown, laugh. -Mike Palij New York University m...@nyu.edu --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7271 or send a blank email to leave-7271-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
I'm really getting tired of the personal arguing going on among a very small number of individuals on this list (two of whom appear in the message below this), and would like to request that the children take their petty arguments off list. Original message Date: Fri, 17 Dec 2010 07:14:26 -0500 From: Mike Palij m...@nyu.edu Subject: Re: [tips] The joy of stats To: Teaching in the Psychological Sciences (TIPS) tips@fsulist.frostburg.edu Cc: Mike Palij m...@nyu.edu On Fri, 17 Dec 2010 03:49:01 -0800, Michael Smith wrote: Again Palij you miss the points. It's pretty funny how you do that so consistently, but not surprising. You don't confuse me by hiding your 'logic' behind being long winded...it's just really boring. Maybe instead of giving your interpretation of wikipedia, just provide a title and the link that should cut down your posts by probably 60%. lol Smith, you are a clown. You lack scholarship, you are inarticulate, you are a coward when confronted with your own errors, you confuse your opinions with the facts, and lack intellectual integrity. You laugh because that is all you have. You make statements that lack validity and are overgeneralizations and when called on them cannot respond with anything else. Laugh, clown, laugh. -Mike Palij New York University m...@nyu.edu --- You are currently subscribed to tips as: drb...@rcn.com. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13185.d5346723901d967ccc167929e2ee94adn=Tl=tipso=7271 or send a blank email to leave-7271-13185.d5346723901d967ccc167929e2ee9...@fsulist.frostburg.edu . Robert W. Wildblood, PhD Adjunct Psychology Faculty Germanna Community College drb...@rcn.com --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7282 or send a blank email to leave-7282-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
Allen Esterson wrote: I note that, keeping in mind Mill’s dictum that “He who knows only his own side of the case, knows little of that”, This from the man who is criticizing a book he hasn't yet read. Chris -- Christopher D. Green Department of Psychology York University Toronto, ON M3J 1P3 Canada 416-736-2100 ex. 66164 chri...@yorku.ca http://www.yorku.ca/christo/ == --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7296 or send a blank email to leave-7296-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
I agree. I wouldn't normally respond to Mike P's personal innuendos and comments in kind (which are usually, if not always, initiated by him to multiple posters on TIPS), but I thought I would this time in order to highlight its inappropriateness and the degree to which Mike P is willing to engage in this kind of behavior. However, I have little hope that his behavior will change. --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7299 or send a blank email to leave-7299-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
RE: [tips] The joy of stats
I nominate this thread as the most inappropriately (or ironically) named thread in the history of TIPS given the extent to which each subsequent post has clearly sucked all of the joy out of the list. In fact, my stats classes are a laugh riot compared to this dreary sequence of ad hominem attacks. Are there any other nominees before we vote? Rick Dr. Rick Froman, Chair Division of Humanities and Social Sciences Box 3055 x7295 rfro...@jbu.edu http://tinyurl.com/DrFroman Proverbs 14:15 A simple man believes anything, but a prudent man gives thought to his steps. --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7300 or send a blank email to leave-7300-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re:[tips] The joy of stats
Chris Green writes of my posts in response to his repeated commendations of *The Spirit Level*: This from the man who is criticizing a book he hasn't yet read. Chris: Please check my posts and quote where I have given my personal view of the book. All I have done is quote the criticisms of a wide range of academics with expertise in the subject matter, with one exception. That relates to a section I came across in their book in which Wilkinson and Pickett contend that differences in certain psychological traits between citizens of the United States and Japan, such a personal modesty, results from the differences in economic inequality/equality in the two countries. And I added that “I fully acknowledge that this is just one brief item”, while noting that “it does at least exemplify one criticism widely levelled at the book: that the authors do not control for other possible relevant factors.” Allen Esterson Former lecturer, Science Department Southwark College, London allenester...@compuserve.com http://www.esterson.org --- From: Christopher D. Green chri...@yorku.ca Subject:Re: The joy of stats Date: Fri, 17 Dec 2010 12:48:56 -0500 Allen Esterson wrote: I note that, keeping in mind Mill’s dictum that “He who knows only his own side of the case, knows little of that”, This from the man who is criticizing a book he hasn't yet read. Chris -- Christopher D. Green Department of Psychology York University Toronto, ON M3J 1P3 Canada --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7322 or send a blank email to leave-7322-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re:[tips] The joy of stats
On 15 December Chris Green cited evidence he thinks is supportive of the conclusions of the book I was pushing a few weeks ago: _The Spirit Level: Why Equality is Better for Everyone_. As I wrote in a post under the thread on “objectivity” recently, while one can never completely remove one’s own biases and preconceptions in assessing a thesis, one of the basic principles is that one should seek out writings critical of the thesis before drawing any firm conclusions. (John Stuart Mill puts the same notion rather more strongly in his essay “On Liberty”: “He who knows only his own side of the case, knows little of that.”) This is the third time in recently weeks that Chris has “pushed” *The Spirit Level*. So I ask of Chris, which of the several informed critiques of the “Spirit Level” that have been published, some of which are easily found on the internet, have you read? There are plenty of informed critical responses out there, not by any means all from the political Right. For instance, Andrew Leigh (former professor of economics at the Australian National University and currently a Labor member of the Australian House of Representatives who regards himself as “about as anti-inequality an economist as you’ll find”) writes: “John Kay’s view in the FT http://www.ft.com/cms/s/2/77b1bd26-14db-11de-8cd1-779fd2ac.html#axzz17MJEMHNx comes closest to my own: ‘a larger source of irritation is the authors’ apparent belief that the application of regression methods to economic and social statistics is as novel to social science as it apparently is to medicine. The evidence presented in the book is mostly a series of scatter diagrams, with a regression line drawn through them. No data is provided on the estimated equations, or on relevant statistical tests. If you remove the bold lines from the diagram, the pattern of points mostly looks random, and the data dominated by a few outliers.’ http://previousleigh.wordpress.com/2009/12/14/look-at-the-changes-not-at-the-levels/ Here is David Runciman (Senior Lecturer in Political Theory at Trinity College Cambridge), in the well-left-of-centre London Review of Books: “Why then, given all this – the concise argument, the weight of the evidence, the unmistakable practical purpose of the authors – does the book still feel oddly utopian? Part of the problem, I think, is that the argument is not as straightforward as its authors would like. Despite their obvious sense of conviction, and maybe even because of it, they fudge the central issue at crucial moments, whereas at others, perhaps in order to compensate, they overstate their case, which only makes things worse.” http://www.lrb.co.uk/v31/n20/david-runciman/how-messy-it-all-is And from Christian Bjørnskov (Associate Professor of Economics, University of Aarhus, Sweden): “To ordinary readers without firm statistical training, this approach appears to represent careful and 'painstakingly marshaled' evidence, as The Economist put it. However, to readers with a background in economics or political science, the evidence in the book is wanting. When seeing strong conclusions drawn from scatter plots and other simple figures, for example, one has to ask three questions: 1) are the relations driven by outlier observations; 2) are the findings robust to controlling for other relevant factors; and 3) are the relations likely to be causal? Surprisingly often, Wilkinson and Pickett’s claims fail to address one or more of these questions. “The bottom line is that this is a well-written, stimulating polemic. It nevertheless suffers from the same problems as one-trick ponies: if the one trick does not impress you, the show is a failure. Wilkinson and Pickett’s trick simply does not hold up to empirical scrutiny.” http://nonicoclolasos.files.wordpress.com/2010/08/pdr-bjornskov-review-file.pdf Finally, John Goldthorpe, Emeritus Professor of Sociology, Nuffield College, Oxford, in European Social Review, 2009: “Questions might be raised about every link in this proposed causal chain originating in income inequality…[…] Wilkinson and Pickett have no time for nicely balanced judgements. They believe that the evidence they present shows beyond doubt that more equal societies ‘do better’, and they are also confident that they have the right explanation for why this is so... Their case is by no means so securely established as they try to make out... it has been called into question by other leading figures in the field – a fact that WP might have more fully acknowledged... WP’s inadequate, one-dimensional understanding of social stratification leads to major problems in their account of how the contextual effect is produced. http://esr.oxfordjournals.org/content/early/2009/10/22/esr.jcp046.abstract Allen Esterson Former lecturer, Science Department Southwark College, London allenester...@compuserve.com http://www.esterson.org
Re:[tips] The joy of stats
A postscript to my post on *The Spirit Level*: I don’t pretend to have done more than glance through the *The Spirit Level*. To properly read a book brimful of statistics I have to be (a) in the right frame of mind (b) have little else occupying my time, neither of which is currently the situation. In any case, I simply don’t have the expertise to judge whether the authors are making their (statistical) case reasonably dispassionately or fairly. But one short section did stand out for me. Wilkinson and Pickett (pp. 44-45) make a comparison between Japan and the USA as respectively “the most equal [and] almost the most unequal of the rich market economies”. Their theme here is the strong contrast between certain psychological traits of citizens in the two countries, such as personal modesty, which they assert to be the result of the differences in economic equality/inequality. (They note, supposedly in support of their argument, that the Japanese pattern of behaviour is also found in Taiwan and China.) But nowhere in the section is there any suggestion that at least part of these differences may relate to deeply embedded cultural traditions. I fully acknowledge that this is just one brief item, but it does at least exemplify one criticism widely levelled at the book: that the authors do not control for other possible relevant factors. Allen Esterson Former lecturer, Science Department Southwark College, London allenester...@compuserve.com http://www.esterson.org --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7240 or send a blank email to leave-7240-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
Again, Allen overwhelms with far more bulk than (I feel) it is appropriate to respond to in a forum such as this. I will say, in response to Kay's critique, that one should look at the whole book. It is written for a popular audience, and so does not include correlations coefficients in the main text, but they are included in appendix. As for it seeming a bit utopian, yes it is, in the sense that there is essentially zero chance the the populations of the US and UK, especially, will ever accept these recommendations. Inequality has (sadly) become an entrenched part of their national ideologies - the pain that signals to them they are getting maximum gain from their economies. That people refuse to do what is good for them doesn't mean, however, that it wouldn't sill be good for them. For myself, I find the evidence provided in the book quite compelling. As with all correlational research, it is possible that there is something else, correlated with equality, going on. The authors are aware of this issue and address it extensively. The burden of proof is on the critic to provide an equally compelling case for some other variable. Chris --- Christopher D Green Department of Psychology York University Toronto, ON M6C 1G4 Canada chri...@yorku.ca On Dec 16, 2010, at 6:51 AM, Allen Esterson allenester...@compuserve.com wrote: On 15 December Chris Green cited evidence he thinks is supportive of the conclusions of the book I was pushing a few weeks ago: _The Spirit Level: Why Equality is Better for Everyone_. As I wrote in a post under the thread on “objectivity” recently, while one can never completely remove one’s own biases and preconceptions in assessing a thesis, one of the basic principles is that one should seek out writings critical of the thesis before drawing any firm conclusions. (John Stuart Mill puts the same notion rather more strongly in his essay “On Liberty”: “He who knows only his own side of the case, knows little of that.”) This is the third time in recently weeks that Chris has “pushed” *The Spirit Level*. So I ask of Chris, which of the several informed critiques of the “Spirit Level” that have been published, some of which are easily found on the internet, have you read? There are plenty of informed critical responses out there, not by any means all from the political Right. For instance, Andrew Leigh (former professor of economics at the Australian National University and currently a Labor member of the Australian House of Representatives who regards himself as “about as anti-inequality an economist as you’ll find”) writes: “John Kay’s view in the FT http://www.ft.com/cms/s/2/77b1bd26-14db-11de-8cd1-779fd2ac.html#axzz17MJEMHNx comes closest to my own: ‘a larger source of irritation is the authors’ apparent belief that the application of regression methods to economic and social statistics is as novel to social science as it apparently is to medicine. The evidence presented in the book is mostly a series of scatter diagrams, with a regression line drawn through them. No data is provided on the estimated equations, or on relevant statistical tests. If you remove the bold lines from the diagram, the pattern of points mostly looks random, and the data dominated by a few outliers.’ http://previousleigh.wordpress.com/2009/12/14/look-at-the-changes-not-at-the-levels/ Here is David Runciman (Senior Lecturer in Political Theory at Trinity College Cambridge), in the well-left-of-centre London Review of Books: “Why then, given all this – the concise argument, the weight of the evidence, the unmistakable practical purpose of the authors – does the book still feel oddly utopian? Part of the problem, I think, is that the argument is not as straightforward as its authors would like. Despite their obvious sense of conviction, and maybe even because of it, they fudge the central issue at crucial moments, whereas at others, perhaps in order to compensate, they overstate their case, which only makes things worse.” http://www.lrb.co.uk/v31/n20/david-runciman/how-messy-it-all-is And from Christian Bjørnskov (Associate Professor of Economics, University of Aarhus, Sweden): “To ordinary readers without firm statistical training, this approach appears to represent careful and 'painstakingly marshaled' evidence, as The Economist put it. However, to readers with a background in economics or political science, the evidence in the book is wanting. When seeing strong conclusions drawn from scatter plots and other simple figures, for example, one has to ask three questions: 1) are the relations driven by outlier observations; 2) are the findings robust to controlling for other relevant factors; and 3) are the relations likely to be causal? Surprisingly often, Wilkinson and Pickett’s claims fail to address one or more of these questions. “The bottom line is that this is a
Re: [tips] The joy of stats
I think I'll try to match Allen's lengthy response. I haven't read it (The Spirit Level) either (although I think I have it somewhere and have been planning to). I think Chris made a good point that it is a popular book and so perhaps focuses on readability. This, however, doesn't mean the analysis is poor. And for it's critics to point out that it doesn't present detailed statistical analyses, is I think, ludicrous. It isn't after all, a journal article. But the best part of Allen's response is the funny parts. One funny thing in Allen's post was: ...who regards himself as “about as anti-inequality an economist as you’ll find”)... Well I guess that settles that. This is proof positive that God exists. We have at last found a truly objective (unbiased) individual who is, miraculously, a government worker !! But the most hillarious one is from one critics response that Allen presents that includes: The evidence presented in the book is mostly a series of scatter diagrams, with a regression line drawn through them. This is hillarious !!! The reason being, of course, is that the statement is a good description of all the results in sociology. I'm still laughing at that one. Thanks Allen. --Mike --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7248 or send a blank email to leave-7248-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
On Thu, 16 Dec 2010 07:35:53 -0800, Michael Smith allegedly wrote: But the most hillarious one is from one critics response that Allen presents that includes: The evidence presented in the book is mostly a series of scatter diagrams, with a regression line drawn through them. This is hillarious !!! The reason being, of course, is that the statement is a good description of all the results in sociology. I'm still laughing at that one. As someone who is familiar with the research methods literature in sociology, I am puzzled about (a) why you are laughing (outside of your being prone to laugh at things for no apparent reason) and (b) why you think regression lines and scatterplots describe all of the results in sociology. Hubert Blalock, a quantitative sociologist, is just one sociologist who has used sophisticated statistical analysis in sociological research in addition to writing/editing influential textbooks in statistical analysis for sociologists. Consider: His 1985 Causal Models in the Social Sciences; on books.google.com: http://books.google.com/books?id=Qvpa5n1OhRYCprintsec=frontcoverdq=blalock+%22causal+models+in+the+social+sciences%22source=blots=3b6T8vfz7Csig=zZCkVBwPvQLBAwRgd9UBdKZHIh8hl=enei=1D8KTfrVAoiosAPvsKGLCwsa=Xoi=book_resultct=resultresnum=1ved=0CBMQ6AEwAA#v=onepageqf=false or http://tinyurl.com/blalock1985 Note: this books continues to be used and is available on Amazon as are other books by Blalock. I would also suggest looking at his 2007 book Causal Models in Experimental designs; see: http://books.google.com/books?id=Mawj7EyctSACprintsec=frontcoverdq=blalock+%22causal+models+in+the+social+sciences%22source=blots=kMikWSiPJesig=pGDiycgAMpF1yQPPLyLuf5dkd-Mhl=enei=1D8KTfrVAoiosAPvsKGLCwsa=Xoi=book_resultct=resultresnum=6ved=0CDUQ6AEwBQ#v=onepageq=blalock%20%22causal%20models%20in%20the%20social%20sciences%22f=false or http://tinyurl.com/blalock2007 Examination of the serial Sociological Methods Research (SMR) will show that many different types of sophisticated forms of analyses are used and promoted by well-known experts in their field like Kenneth Bollen. SMR is published by Sage and the current issue (in volume #39) is available here: http://smr.sagepub.com/content/current There is also the Journal of Mathematical Sociology; see: http://www.tandf.co.uk/journals/GMAS For some background on mathematical sociology, there is a Wikipedia entry (yadda-yadda) that provides some history and background; see: http://en.wikipedia.org/wiki/Mathematical_sociology So, what's the basis for the claim that sociologists only use scatterplots and regression in their research? That's claim is almost as hilarious as saying that all psychologists are clinicians. -Mike Palij New York University m...@nyu.edu --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7252 or send a blank email to leave-7252-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
Hi It is not just sociology that suffers from excessive ideological promotion / criticism of non-experimental research. Psychology as well has many issues (theories, models, whatever, ...) that will only or primarily be examined non-experimentally. It behooves us as scholars / researchers / educators to document not only the limitations of such approaches but as well the potential. Consider Streissguth's work on maternal consumption of alcohol during pregnancy. http://www.sciencedirect.com/science?_ob=ArticleURL_udi=B6WYC-46X4819-35_user=1068128_coverDate=01%2F31%2F1989_rdoc=1_fmt=high_orig=search_origin=search_sort=d_docanchor=view=c_searchStrId=1581697283_rerunOrigin=google_acct=C51257_version=1_urlVersion=0_userid=1068128md5=f357fe5bd837dfee51da7b30f2d041f5searchtype=a OR http://tinyurl.com/2aromas and http://www.aamr.org/ehi/media/Streissguth%20PP%203%209%2007%20(2).pdf Essentially the research shows an effect of maternal alcohol consumption on child's intellectual abilities even controlling for numerous correlated (or potentially correlated) variables. Surely, the greater the number of competing hypotheses that are discredited by empirical research (i.e., statistical control) the greater our confidence in a causal relationship. Ideally, such non-experimental results are further confirmed by experimental studies (e.g., with pig fetuses). The idea that we can immediately arrive at a definitive study to settle some issue is perhaps one of the worst things that we can impart to students. Sound understanding only comes from prolonged and often difficult work ... no matter what ideologists might think (or pretend to think in order to undermine approaches with which they disagree, often intensely). Take care Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax j.cl...@uwinnipeg.ca Mike Palij m...@nyu.edu 16-Dec-10 11:00 AM On Thu, 16 Dec 2010 07:35:53 -0800, Michael Smith allegedly wrote: But the most hillarious one is from one critics response that Allen presents that includes: The evidence presented in the book is mostly a series of scatter diagrams, with a regression line drawn through them. This is hillarious !!! The reason being, of course, is that the statement is a good description of all the results in sociology. I'm still laughing at that one. As someone who is familiar with the research methods literature in sociology, I am puzzled about (a) why you are laughing (outside of your being prone to laugh at things for no apparent reason) and (b) why you think regression lines and scatterplots describe all of the results in sociology. Hubert Blalock, a quantitative sociologist, is just one sociologist who has used sophisticated statistical analysis in sociological research in addition to writing/editing influential textbooks in statistical analysis for sociologists. Consider: His 1985 Causal Models in the Social Sciences; on books.google.com: http://books.google.com/books?id=Qvpa5n1OhRYCprintsec=frontcoverdq=blalock+%22causal+models+in+the+social+sciences%22source=blots=3b6T8vfz7Csig=zZCkVBwPvQLBAwRgd9UBdKZHIh8hl=enei=1D8KTfrVAoiosAPvsKGLCwsa=Xoi=book_resultct=resultresnum=1ved=0CBMQ6AEwAA#v=onepageqf=false or http://tinyurl.com/blalock1985 Note: this books continues to be used and is available on Amazon as are other books by Blalock. I would also suggest looking at his 2007 book Causal Models in Experimental designs; see: http://books.google.com/books?id=Mawj7EyctSACprintsec=frontcoverdq=blalock+%22causal+models+in+the+social+sciences%22source=blots=kMikWSiPJesig=pGDiycgAMpF1yQPPLyLuf5dkd-Mhl=enei=1D8KTfrVAoiosAPvsKGLCwsa=Xoi=book_resultct=resultresnum=6ved=0CDUQ6AEwBQ#v=onepageq=blalock%20%22causal%20models%20in%20the%20social%20sciences%22f=false or http://tinyurl.com/blalock2007 Examination of the serial Sociological Methods Research (SMR) will show that many different types of sophisticated forms of analyses are used and promoted by well-known experts in their field like Kenneth Bollen. SMR is published by Sage and the current issue (in volume #39) is available here: http://smr.sagepub.com/content/current There is also the Journal of Mathematical Sociology; see: http://www.tandf.co.uk/journals/GMAS For some background on mathematical sociology, there is a Wikipedia entry (yadda-yadda) that provides some history and background; see: http://en.wikipedia.org/wiki/Mathematical_sociology So, what's the basis for the claim that sociologists only use scatterplots and regression in their research? That's claim is almost as hilarious as saying that all psychologists are clinicians. -Mike Palij New York University m...@nyu.edu --- You are currently subscribed to tips as: j.cl...@uwinnipeg.ca. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13251.645f86b5cec4da0a56ffea7a891720c9n=Tl=tipso=7252 or send a blank email to leave-7252-13251.645f86b5cec4da0a56ffea7a89172...@fsulist.frostburg.edu ---
Re: [tips] The joy of stats
Mike P: As someone who is familiar with the research methods literature in sociology. Congratulations. I think most here are familiar with regression. I am puzzled about (a) why you are laughing (outside of your being prone to laugh at things for no apparent reason) ...better than being a pedantic bore. (b) why you think regression lines and scatterplots describe all of the results in sociology. I think you're just permanently puzzled. If you think that a couple of names or a journal establishes your point of the great complexity of analysis used throughout sociology you're just wrong (again) and that despite your liberal use of Wikipedia --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7261 or send a blank email to leave-7261-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
On Thu, 16 Dec 2010 18:11:45 -0800, Michael Smith wrote: Mike P: As someone who is familiar with the research methods literature in sociology. Congratulations. I think most here are familiar with regression. Non sequitur. You self-servingly edited out most of post which showed that sociologists have used statistical techniques well beyond regression. If you think that that is the only type of analysis they do why don't you explain how structural equation modeling (SEM) can be reduced to OLS regression? I'd pay a dollar to see you do that. [Mike P. wrote:] I am puzzled about (a) why you are laughing (outside of your being prone to laugh at things for no apparent reason) ...better than being a pedantic bore. Maybe, but you then come off as being an intellectual coward becuase you say foolish things like I'm laughing at this and when confronted with evidence that you are wrong, don't even have the cahonies to admit that you're wrong. (b) why you think regression lines and scatterplots describe all of the results in sociology. I think you're just permanently puzzled. Again, non sequitur. If you cannot provide a meaningful answer, may I suggest you just STFU? If you think that a couple of names or a journal establishes your point of the great complexity of analysis used throughout sociology you're just wrong (again) and that despite your liberal use of Wikipedia Boy, where did you learn to misrepresent other people's arguments? Grade school? Sunday school? Grad school? Or are you a self-educated kind of guy? You're the one that said all of sociology was regression lines and scatterplots -- I called you on it and proved you wrong by showing that sociologists in fact use many techniques. Listen, I think a lot of people get it by now that you're not really serious when you make these kinds of foolish statements. I don't know what you take seriously and where you manifest this seriousness and that's okay. Just realize that you really do look foolish when you say things like: I'm still laughing at that one. -Mike Palij New York University m...@nyu.edu --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7262 or send a blank email to leave-7262-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re:[tips] The joy of stats
Stephen Black drew attention to: A remarkable graphical display of economic progress at http://www.youtube.com/watch?v=jbkSRLYSojo As the clip indicates, this is from a BBC 4 TV documentary, which I saw last week. I have no comment on the topic of the clip itself, but here's something on the presentation by Hans Rosling: Putting it simply, he opens the documentary by saying that, sociologically speaking, all was dark until statistics shone a light on everything. He proceeds to demonstrate with numerous examples the importance of statistics as in the clip, and of course statistics frequently give invaluable information, notably in Rosling's own field, world health studies. And as a presentation, it's great fun, illustrating his theme that statistics do not have to be boring. But just one caveat: he never gives the slightest indication that statistics can be a means of (deliberately or otherwise – and I think it is often otherwise) deceiving people. Evidently he has not heard of (or chooses to ignore in his celebration of statistics) the saying lies, damned lies, and statistics. An example occurred in the documentary itself. David McCandless, a data designer, was shown providing diagrammatic representation of various statistics involving very high numbers that he said brought home the reality of the information. One of his comparisons was UK NHS costs against those of the UK Trident programme, showing virtually equal areas: NHS £102 billion Total cost of Trident £104 billion. First one has to consider whether the figures are accurate. I checked out on the internet and found that Greenpeace (an organisation that has never knowingly understated their claims) gives £97 billion as the total cost of the programme (far more than the Government estimates, for what that is worth), but this is over a 30 year period. In other words, what McCandless's diagrams should have compared were the annual costs of the NHS and Trident, i.e., £102 billion against around £3 billion. (Greenpeace actually gives as a *current* estimate of annual cost of Trident around £2 billion per annum.) So the comparative diagrams that McCandless presents give a grossly misleading picture, making it seem that as much is spent on Trident as on the NHS. (Maybe that item should have been given under the heading of How to lie with statistics. :-) ) Another statistical point, though not specifically relevant to the documentary: The way life expectancies are presented often gives the impression that relatively few people lived past the age of life expectancy for the place and era. What is rarely explained is that in the past infant mortality was very high, and death from diseases in childhood also relatively high. This means that individuals who survived into adulthood had a good chance of living well beyond the statistical life expectancy age. So (making figures up for the sake of argument), in an era when the life expectancy was around 35, for those who reached adulthood most of them might well live to the age of 55 or 60. Allen Esterson Former lecturer, Science Department Southwark College, London allenester...@compuserve.com http://www.esterson.org -- From: sbl...@ubishops.ca Subject:The joy of stats Date: Tue, 14 Dec 2010 14:25:25 -0500 A remarkable graphical display of economic progress at http://www.youtube.com/watch?v=jbkSRLYSojo Pretty neat, eh? More on this in David Brooks' NY Times column at http://tinyurl.com/2vyv76d Stephen Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University Sherbrooke, Quebec, Canada e-mail: sblack at ubishops.ca --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7201 or send a blank email to leave-7201-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
Stephen Black drew attention to: A remarkable graphical display of economic progress at http://www.youtube.com/watch?v=jbkSRLYSojo Rosling fails to point out one of the most significant aspects of the data: almost all of the increases life expectancy come between incomes of $400 and $4000. Only relatively small increases occur between $4000 and $40,000 (which is, as Jim earlier noted, TEN TIMES as much of an increase in income). This is in line with the conclusions of the book I was pushing a few weeks ago: _The Spirit Level: Why Equality is Better for Everyone_. More important than just life expectancy, the same thing is roughly true for any measure of the quality of life. We appear to have reached something of a ceiling in the ability of money to improve the quality of our lives. Sure, it's great to have your own monster house or yacht or airplane, but the increase in happiness it produces, on a dollar for dollar basis, is minuscule compared to what the first few thousand dollars of income does for one's life. What does seem to produce big increases in quality of life measures, even for those at the top of the income scale, is living in more equal societies. I know -- these are nearly unsayable words in today's America (but then again, evolution by natural selection is almost unsayable as well). Chris -- Christopher D. Green Department of Psychology York University Toronto, ON M3J 1P3 Canada 416-736-2100 ex. 66164 chri...@yorku.ca http://www.yorku.ca/christo/ == --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7206 or send a blank email to leave-7206-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
RE: [tips] The joy of stats
Also, when we talk about life expectancy increasing, we're usually talking average, and most of the increase in the last hundred years has come about as a result of decreased infant mortality. If you make it to 65 or so, you'll probably live 3-4 years longer than a 65-year-old in 1900. Infant mortality, of course, does relate to the average wealth of a country (but not perfectly, by any stretch). I'm just adding to Chris's comment. m -- Marc Carter, PhD Associate Professor and Chair Department of Psychology College of Arts Sciences Baker University -- From: Christopher D. Green [mailto:chri...@yorku.ca] Sent: Wednesday, December 15, 2010 8:17 AM To: Teaching in the Psychological Sciences (TIPS) Subject: Re: [tips] The joy of stats Stephen Black drew attention to: A remarkable graphical display of economic progress at http://www.youtube.com/watch?v=jbkSRLYSojo Rosling fails to point out one of the most significant aspects of the data: almost all of the increases life expectancy come between incomes of $400 and $4000. Only relatively small increases occur between $4000 and $40,000 (which is, as Jim earlier noted, TEN TIMES as much of an increase in income). This is in line with the conclusions of the book I was pushing a few weeks ago: _The Spirit Level: Why Equality is Better for Everyone_. More important than just life expectancy, the same thing is roughly true for any measure of the quality of life. We appear to have reached something of a ceiling in the ability of money to improve the quality of our lives. Sure, it's great to have your own monster house or yacht or airplane, but the increase in happiness it produces, on a dollar for dollar basis, is minuscule compared to what the first few thousand dollars of income does for one's life. What does seem to produce big increases in quality of life measures, even for those at the top of the income scale, is living in more equal societies. I know -- these are nearly unsayable words in today's America (but then again, evolution by natural selection is almost unsayable as well). Chris -- Christopher D. Green Department of Psychology York University Toronto, ON M3J 1P3 Canada 416-736-2100 ex. 66164 chri...@yorku.camailto:chri...@yorku.ca http://www.yorku.ca/christo/ == --- You are currently subscribed to tips as: marc.car...@bakeru.edumailto:marc.car...@bakeru.edu. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13029.76c7c563b32ad9d8d09c72a2d17c90e1n=Tl=tipso=7206 (It may be necessary to cut and paste the above URL if the line is broken) or send a blank email to leave-7206-13029.76c7c563b32ad9d8d09c72a2d17c9...@fsulist.frostburg.edumailto:leave-7206-13029.76c7c563b32ad9d8d09c72a2d17c9...@fsulist.frostburg.edu The information contained in this e-mail and any attachments thereto (e-mail) is sent by Baker University (BU) and is intended to be confidential and for the use of only the individual or entity named above. The information may be protected by federal and state privacy and disclosures acts or other legal rules. If the reader of this message is not the intended recipient, you are notified that retention, dissemination, distribution or copying of this e-mail is strictly prohibited. If you have received this e-mail in error please immediately notify Baker University by email reply and immediately and permanently delete this e-mail message and any attachments thereto. Thank you. --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7207 or send a blank email to leave-7207-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
On 15 Dec 2010 at 5:34, Allen Esterson wrote: But just one caveat: he [Rosling] never gives the slightest indication that statistics can be a means of (deliberately or otherwise - and I think it is often otherwise) deceiving people. Evidently he has not heard of (or chooses to ignore in his celebration of statistics) the saying lies, damned lies, and statistics. Allen gives two choice examples, one relating to the cost of the NHS (Obamacare for Brits) vs the cost of British nuclear missile defense, and the other to a misinterpretation in considering life expectancy stats. Here's another example which just popped up, this one a bit closer to home (metaphorically), relating to the value of a university education: http://tinyurl.com/27llo97 Stephen Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University Sherbrooke, Quebec, Canada e-mail: sblack at ubishops.ca - --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7231 or send a blank email to leave-7231-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
As my students would say, really cool! I was looking (eyeball only) to see if the line of best fit remained about the same through the years ... despite some fluctuations, the basic positive (sort of linear) correlation remained. Now, if we achieved what he suggested at the end - greater wealth and more equality - the correlation would automatically weaken, the old restriction of range issue, as everyone piles into the upper right quadrant. == John W. Kulig, Ph.D. Professor of Psychology Director, Psychology Honors Plymouth State University Plymouth NH 03264 == - Original Message - From: sbl...@ubishops.ca To: Teaching in the Psychological Sciences (TIPS) tips@fsulist.frostburg.edu Sent: Tuesday, December 14, 2010 2:25:25 PM Subject: [tips] The joy of stats A remarkable graphical display of economic progress at http://www.youtube.com/watch?v=jbkSRLYSojo Pretty neat, eh? More on this in David Brooks' NY Times column at http://tinyurl.com/2vyv76d Stephen Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University Sherbrooke, Quebec, Canada e-mail: sblack at ubishops.ca - --- You are currently subscribed to tips as: ku...@mail.plymouth.edu. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13338.f659d005276678c0696b7f6beda66454n=Tl=tipso=7189 or send a blank email to leave-7189-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7190 or send a blank email to leave-7190-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
Hi Rosling has been doing a great job not only of presenting results, but of making massive amounts of statistical data widely available in accessible form. I've used his website at http://www.gapminder.org/ for a class exercise in my culture and psychology course the past few years. The BBC graph Stephen pointed to appears to have pushed the data back to the early 1800s for many more countries. Older graphs had very early data just for a couple of Scandinavian countries as I recall. Take care Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax j.cl...@uwinnipeg.ca sbl...@ubishops.ca 14-Dec-10 1:25 PM A remarkable graphical display of economic progress at http://www.youtube.com/watch?v=jbkSRLYSojo Pretty neat, eh? More on this in David Brooks' NY Times column at http://tinyurl.com/2vyv76d Stephen Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University Sherbrooke, Quebec, Canada e-mail: sblack at ubishops.ca - --- You are currently subscribed to tips as: j.cl...@uwinnipeg.ca. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13251.645f86b5cec4da0a56ffea7a891720c9n=Tl=tipso=7189 or send a blank email to leave-7189-13251.645f86b5cec4da0a56ffea7a89172...@fsulist.frostburg.edu --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7191 or send a blank email to leave-7191-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
Hi Just a small point ... the GDP axis is NOT linear ... it is a logarithmic scale. Note the 3 values given at equal intervals are 400, 4000, and 4, the logs of which are 2.6, 3.6, and 4.6. It is this transformation that makes the relationship appear linear. In fact there is a marked curvilinearity such that the larger the GDP the greater the increase needed to produce a change equivalent to that observed for lower GDPs. This same curvilinearity is observed in the relationship between GDP (or other measures of wealth) and Happiness or Subjective Well Being. One implication of this is that the gap between high and low GDP countries is MUCH larger than it might seem. Only $3600 increase needed to get from step 1 to step 2, but $36,000 increase needed to get to step 3. That is, there is a lot more inequity in the world than the graph suggests, all of which needs to be overcome before we're all crowded into the upper right quadrant. Take care Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax j.cl...@uwinnipeg.ca John Kulig ku...@mail.plymouth.edu 14-Dec-10 1:59 PM As my students would say, really cool! I was looking (eyeball only) to see if the line of best fit remained about the same through the years ... despite some fluctuations, the basic positive (sort of linear) correlation remained. Now, if we achieved what he suggested at the end - greater wealth and more equality - the correlation would automatically weaken, the old restriction of range issue, as everyone piles into the upper right quadrant. == John W. Kulig, Ph.D. Professor of Psychology Director, Psychology Honors Plymouth State University Plymouth NH 03264 == - Original Message - From: sbl...@ubishops.ca To: Teaching in the Psychological Sciences (TIPS) tips@fsulist.frostburg.edu Sent: Tuesday, December 14, 2010 2:25:25 PM Subject: [tips] The joy of stats A remarkable graphical display of economic progress at http://www.youtube.com/watch?v=jbkSRLYSojo Pretty neat, eh? More on this in David Brooks' NY Times column at http://tinyurl.com/2vyv76d Stephen Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University Sherbrooke, Quebec, Canada e-mail: sblack at ubishops.ca - --- You are currently subscribed to tips as: ku...@mail.plymouth.edu. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13338.f659d005276678c0696b7f6beda66454n=Tl=tipso=7189 or send a blank email to leave-7189-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu --- You are currently subscribed to tips as: j.cl...@uwinnipeg.ca. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13251.645f86b5cec4da0a56ffea7a891720c9n=Tl=tipso=7190 or send a blank email to leave-7190-13251.645f86b5cec4da0a56ffea7a89172...@fsulist.frostburg.edu --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7192 or send a blank email to leave-7192-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
Re: [tips] The joy of stats
Yes, thanks Jim, didn't notice that ... == John W. Kulig, Ph.D. Professor of Psychology Director, Psychology Honors Plymouth State University Plymouth NH 03264 == - Original Message - From: Jim Clark j.cl...@uwinnipeg.ca To: Teaching in the Psychological Sciences (TIPS) tips@fsulist.frostburg.edu Sent: Tuesday, December 14, 2010 3:28:22 PM Subject: Re: [tips] The joy of stats Hi Just a small point ... the GDP axis is NOT linear ... it is a logarithmic scale. Note the 3 values given at equal intervals are 400, 4000, and 4, the logs of which are 2.6, 3.6, and 4.6. It is this transformation that makes the relationship appear linear. In fact there is a marked curvilinearity such that the larger the GDP the greater the increase needed to produce a change equivalent to that observed for lower GDPs. This same curvilinearity is observed in the relationship between GDP (or other measures of wealth) and Happiness or Subjective Well Being. One implication of this is that the gap between high and low GDP countries is MUCH larger than it might seem. Only $3600 increase needed to get from step 1 to step 2, but $36,000 increase needed to get to step 3. That is, there is a lot more inequity in the world than the graph suggests, all of which needs to be overcome before we're all crowded into the upper right quadrant. Take care Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax j.cl...@uwinnipeg.ca John Kulig ku...@mail.plymouth.edu 14-Dec-10 1:59 PM As my students would say, really cool! I was looking (eyeball only) to see if the line of best fit remained about the same through the years ... despite some fluctuations, the basic positive (sort of linear) correlation remained. Now, if we achieved what he suggested at the end - greater wealth and more equality - the correlation would automatically weaken, the old restriction of range issue, as everyone piles into the upper right quadrant. == John W. Kulig, Ph.D. Professor of Psychology Director, Psychology Honors Plymouth State University Plymouth NH 03264 == - Original Message - From: sbl...@ubishops.ca To: Teaching in the Psychological Sciences (TIPS) tips@fsulist.frostburg.edu Sent: Tuesday, December 14, 2010 2:25:25 PM Subject: [tips] The joy of stats A remarkable graphical display of economic progress at http://www.youtube.com/watch?v=jbkSRLYSojo Pretty neat, eh? More on this in David Brooks' NY Times column at http://tinyurl.com/2vyv76d Stephen Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University Sherbrooke, Quebec, Canada e-mail: sblack at ubishops.ca - --- You are currently subscribed to tips as: ku...@mail.plymouth.edu. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13338.f659d005276678c0696b7f6beda66454n=Tl=tipso=7189 or send a blank email to leave-7189-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu --- You are currently subscribed to tips as: j.cl...@uwinnipeg.ca. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13251.645f86b5cec4da0a56ffea7a891720c9n=Tl=tipso=7190 or send a blank email to leave-7190-13251.645f86b5cec4da0a56ffea7a89172...@fsulist.frostburg.edu --- You are currently subscribed to tips as: ku...@mail.plymouth.edu. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13338.f659d005276678c0696b7f6beda66454n=Tl=tipso=7192 or send a blank email to leave-7192-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5n=Tl=tipso=7193 or send a blank email to leave-7193-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
