tag:blogger.com,1999:blog-4761750297977694004.post4177778233574143345..comments2023-12-19T11:11:30.124+00:00Comments on The Spirit Level Delusion: Chopping and changingChristopher Snowdonhttp://www.blogger.com/profile/15963753745009712865noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-4761750297977694004.post-27775848953437742712011-05-04T09:51:21.086+01:002011-05-04T09:51:21.086+01:00Lies, damned lies and statistics...Lies, damned lies and statistics...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-4761750297977694004.post-50934449993562472802011-03-28T22:03:16.188+01:002011-03-28T22:03:16.188+01:001) No they're not.
2) No there isn't.
3)...1) No they're not.<br /><br />2) No there isn't.<br /><br />3) No they're not.<br /><br />Don't be so stupid.Christopher Snowdonhttps://www.blogger.com/profile/15963753745009712865noreply@blogger.comtag:blogger.com,1999:blog-4761750297977694004.post-60814292604982772052011-03-28T20:26:55.823+01:002011-03-28T20:26:55.823+01:001)Interesting you should choose an unfree society ...1)Interesting you should choose an unfree society like Singapore and Hong Kong, are not democratic - are quasi-fascist city-states.<br /><br />2)Whilst admit there is a few problems <br />- correlation does not always mean causation, this may be true in some cases; but there is too much evidence to show to not believe a more equal society is a better one.<br /><br />3) Further more, I notice how you ethnicity as a possible cause of crime. "...effectively disregard other variables such as absolute income, culture, history, ethnicity, geography, law, politics and climate" I am assuming you mean by ethnicity, you mean race! Which only leaves me to conclude you believe people of other races are more likely to cause crime - or is that how it appears?<br /><br />So I conclude you support fascistic states, and racist policies.<br /><br />Or is that correlation not mean causation?Thom A.noreply@blogger.comtag:blogger.com,1999:blog-4761750297977694004.post-33049133968297799092011-03-20T20:10:23.604+00:002011-03-20T20:10:23.604+00:00So you're portraying a study which has two co-...So you're portraying a study which has two co-authors who are affiliated with the World Bank as being the official position of the World Bank? Not a good start.<br /><br /><i>To make your date more reliable you should increase the sample size. </i><br /><br />Assuming you mean 'data' rather than 'date', you've missed the whole point of The Spirit Level if you haven't noticed that limiting the sample size is the whole point of their book. If they included poorer countries, the whole thing would fall apart so quickly that even the most gullible punter would see through it. <br /><br />Poorer countries are often very unequal. Crime is associated with poverty in poorer countries. The Spirit Level looks at rich countries and so do I. Inequality is not associated with crime amongst the countries featured in The Spirit Level. If you feel Wilkinson and Pickett should "increase the sample size" then tell them directly.Christopher Snowdonhttps://www.blogger.com/profile/15963753745009712865noreply@blogger.comtag:blogger.com,1999:blog-4761750297977694004.post-13882876760851490562011-03-20T18:19:04.785+00:002011-03-20T18:19:04.785+00:00"The main conclusion of this paper is that in..."The main conclusion of this paper is that income inequality, measured by the Gini index, has a significant and positive effect on the incidence of crime"<br /><br />"when poverty falls more rapidly, either because income growth rises or the distribution of income improves, then crimes rates tend to fall"<br /><br />From the World Bank at http://siteresources.worldbank.org/DEC/Resources/Crime&Inequality.pdf<br /><br />Which is a sophisticated, rational, econometrics based analysis and reaches the OPPOSITE conclusion to yours.<br /><br />Both Inequality and Poverty are correlated with crime.<br /><br />I believe the World Bank - who by the way, have been criticized for causing poverty and inequality in the past.<br /><br />To make your date more reliable you should increase the sample size. Also, given most of your countries are European some data tend to cluster and the small spread increases relative errors. Suggest that you either (1) include errors - I suspect it will render some charts inconclusive (2) Include more countries (3) include each state in the USA to avoid counting the USA as a single outlying data point - Do any of these and you will come to the same conclusions as the World Bank....unless of course you are motivated politically rather than an objective researcherKonkernoreply@blogger.comtag:blogger.com,1999:blog-4761750297977694004.post-60183757687073800342011-01-28T16:37:17.135+00:002011-01-28T16:37:17.135+00:00Nice analysis. This strikes me as providing a goo...Nice analysis. This strikes me as providing a good general rule: If someone changes their statistical methods for analyzing two (or more) comparisons that are similar then they should justify the change; if they seem to just be hoping that the reader does not notice, it is a safe bet that they are lying. (lying = speaking/writing/etc. in such a way that people are likely to believe something that you know to not be true)<br /><br />Interesting example, baseball. It is also quite popular in Japan and Cuba. Probably the bigger outlier sports (in terms of US participation) are American "football" and basketball. Basketball, in particular, is arguably tied to wealth inequality: When a large part of the population lacks access to most of society's open spaces and facilities, they will gravitate to a sport that just requires a few hundred square feed of pavement, a metal post, and a ball. Even street versions of baseball and futbol, though they require even less equipment, require more safe open space than many of the urban poor have access to in unequal societies.Carl V Phillipshttps://www.blogger.com/profile/01919902852457771666noreply@blogger.com