Simpson's paradox: even though two variables might be correlated in subgroups of a population they can be negatively correlated or uncorrelated in the population.
--- Frank C. Wimberly 140 Calle Ojo Feliz, Santa Fe, NM 87505 505 670-9918 Santa Fe, NM On Wed, Aug 18, 2021, 8:51 PM Roger Critchlow <r...@elf.org> wrote: > Eric - > > No problem with the modification, I was just making noise at that point: > death panels, statistical police, knowledge is oppression, and so on. > > I see, Morris did exactly what you were suggesting and then went on to > itemize every quibble that might come up in the conclusion. > > I might even remember Simpson's Paradox after this, at least for a few > days. > > -- rec -- > > -- rec -- > > On Wed, Aug 18, 2021 at 8:27 PM Frank Wimberly <wimber...@gmail.com> > wrote: > >> So that's why I'm feeling uneasy as I approach 80. 😐 >> >> >> >> >> --- >> Frank C. Wimberly >> 140 Calle Ojo Feliz, >> Santa Fe, NM 87505 >> >> 505 670-9918 >> Santa Fe, NM >> >> On Wed, Aug 18, 2021, 5:38 PM David Eric Smith <desm...@santafe.edu> >> wrote: >> >>> May I offer a tiny rewording to one of your sentences, Roger? >>> >>> Given that right now I am in a months-long redraft of a paper, which >>> feels like having my teeth drilled, trying to say some things about the >>> difference between what one can do with summary statistics and what one >>> does with population-wide regressions. I am in a situation where I can’t >>> use terms (this community doesn’t use them), can’t cite to where they are >>> explained (not their literature; they will not go follow a cite), and can’t >>> explain them in-line (too long; people don’t want to read all that). >>> Complain complain complain. >>> >>> Interesting thing is that the particular “data” we are getting are, of >>> course, summary statistics. So: >>> >>> You said: >>> >>> Next the statistics police will be telling us how to use the data! >>> >>> >>> How about: >>> >>> Next the statistics police will be telling us how to use the statistics! >>> >>> >>> Eric >>> >>> >>> On Aug 19, 2021, at 7:00 AM, Roger Critchlow <r...@elf.org> wrote: >>> >>> The Israeli data turned up in Andrew Gelman's blog today >>> >>> >>> https://statmodeling.stat.columbia.edu/2021/08/18/coronavirus-and-simpsons-paradox-oldsters-are-more-likely-to-be-vaccinated-and-more-likely-to-have-severe-infections-so-you-need-to-adjust-for-age-when-comparing-vaccinated-and-unvaccinated-peopl/ >>> >>> The argument broken out in his correspondent's blog: >>> >>> >>> https://www.covid-datascience.com/post/israeli-data-how-can-efficacy-vs-severe-disease-be-strong-when-60-of-hospitalized-are-vaccinated >>> <https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fwww.covid-datascience.com%2fpost%2fisraeli-data-how-can-efficacy-vs-severe-disease-be-strong-when-60-of-hospitalized-are-vaccinated&c=E,1,r88QWsw9ZFie03L99WBI-GfoUio7QJL9q0f6LgfquHy6yusuS3afGu9rl-P2qKIe3neZBMCmbb-dV2vLzn84I9vqbj9PkeulwJ9dRSdlThEZ27Vn6yCOrCAc&typo=1> >>> >>> Next the statistics police will be telling us how to use the data! >>> >>> Oh, and someone noticed that a lot of the unvaccinated can't afford to >>> take two days off for vaccination side effects, no medical leave, you could >>> try offering her 2 days and a few hours pay to get vaccinated. >>> >>> -- rec -- >>> >>> >>> On Wed, Aug 18, 2021 at 4:58 PM Marcus Daniels <mar...@snoutfarm.com> >>> wrote: >>> >>>> Another hypothesis is that they tend to be less engaged in the economy >>>> and their communities and so are thus less susceptible; smaller social >>>> networks and lower spatial density. And further they attach themselves to >>>> these crazy ideas because no one ever has the opportunity to push back >>>> except people that go way out of their way like Glen. >>>> >>>> -----Original Message----- >>>> From: Friam <friam-boun...@redfish.com> On Behalf Of David Eric Smith >>>> Sent: Wednesday, August 18, 2021 1:09 PM >>>> To: The Friday Morning Applied Complexity Coffee Group < >>>> friam@redfish.com> >>>> Subject: Re: [FRIAM] vax v unvax >>>> >>>> This is one that it would be nice to see broken down by age and other >>>> conditions. >>>> >>>> If unvaxxed hospitalizations are a true “cross section of America” >>>> (like jury duty), whereas the vaxxed ones are mainly old or sick with >>>> something else, that would be an important variable for deriving a risk >>>> profile by category. >>>> >>>> Time, time, time, to chase things down…. >>>> >>>> Eric >>>> >>>> > On Aug 19, 2021, at 4:53 AM, uǝlƃ ☤>$ <geprope...@gmail.com> wrote: >>>> > >>>> > >>>> > Soooo ... our house cleaner is anti-vax. My friends tell me we should >>>> fire her. I've resisted confronting her over it. But hospitals being >>>> overrun influenced me to press her a little bit. I claimed that the >>>> majority of hospitalizations are the unvaxxed. She claimed that down near >>>> her (I think closer to Centralia, WA), it was flipped, that the majority of >>>> hospitalizations were vaccinated. We had the usual conversation where she >>>> ended with "you believe your data and I believe my data" blather. So I >>>> decided to see if I could find news articles claiming the majority of >>>> hospitalized covid cases are the vaccinated. And I found this: >>>> > >>>> > Higher Death, Hospitalization Rates Among Vaccinated Individuals: UK >>>> > COVID-19 Data >>>> > >>>> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fwww.visiontimes.co >>>> > m%2f2021%2f07%2f04%2fdeaths-hospital-vaccinated-uk.html&c=E,1,S5wT7eG_ >>>> > LxnTLuUE1RbuVEziaTP99eeGAnZxv_dtTYhpTiIkicxQt2mvokLXVbixkXhDcXvoLjNg1n >>>> > GUqSLe66RnTqmVs_5EGMgv3Z6cPUYQEg,,&typo=1 >>>> > >>>> > Most COVID-19 Patients at Israel Hospital Fully Vaccinated, Doctor >>>> Calls Mandates ‘Diabolic’ >>>> > >>>> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fwww.visiontimes.co >>>> > m%2f2021%2f08%2f08%2fisrael-hospital-vaccinated.html&c=E,1,FDPrPMNiU5e >>>> > 621cpzQhoViPiTlcpkww8iHTLVFR9vwhMrJ_5m1j77NIXnZYqEweYmhJcGmbR5PUWpuOvQ >>>> > F1ZPV23_7Fw9AKvzzPGnjmS-w,,&typo=1 >>>> > >>>> > They're both from the "Vision Times", which looks to me like a Falun >>>> Gong outlet, popular with Trumpists, anti-CCP, etc. So, I don't really want >>>> to do any debunking or verification. I *want* to write it off as garbage. >>>> But their source: >>>> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fassets.publishing.service.gov.uk%2fgovernment%2fuploads%2fsystem%2fuploads%2fattachment_data%2ffile%2f1001359%2fVariants_of_Concern_VOC_Technical_Briefing_16.pdf&c=E,1,Z0AX7_CvU6zmg1ca9s0IY3_E3dshyhV2o2XdB9IZiLRuF4FqVWdK_4ijYuyJIk-wyRPEoaaRJUlchbHLmor1XP3Y-0_ioCec73m-PpoXtafAhLUS702P8MUqhA,,&typo=1 >>>> seems legit. And the numbers in the article seem to match Table 4. >>>> > >>>> > Plugging in the numbers from the latest briefing: >>>> > >>>> https://www.gov.uk/government/publications/investigation-of-novel-sars >>>> > -cov-2-variant-variant-of-concern-20201201 >>>> > >>>> > It really does seem to show that the death rate among vaxxed is >>>> higher than that among unvaxxed. >>>> > >>>> > The LifeSiteNews site is obviously troll nonsense. But my concern is >>>> mostly about that tech report from the UK and Herzog Hospital. Are these >>>> counter examples to the NYT unvax/vax data? Are they small enough to be >>>> within error? >>>> > >>>> > -- >>>> > ☤>$ uǝlƃ >>>> > >>>> > - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . >>>> > FRIAM Applied Complexity Group listserv Zoom Fridays 9:30a-12p Mtn >>>> > GMT-6 bit.ly/virtualfriam >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fbit.ly%2fvirtualfriam&c=E,1,fqQLi64iLssMYVmLhhr5W8t0KdMj_LE-IXRGm1vT8wMK2gPQywI6g3jvt0TsGwGqKRw2xXOg23Kouyrwk3s4pQeq7aJKphQnq4nB-Bu-GNStrNrMtw,,&typo=1> >>>> un/subscribe >>>> > >>>> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailm >>>> > an%2flistinfo%2ffriam_redfish.com >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2f2ffriam_redfish.com&c=E,1,MKGWfqQ2Rq5b4hn9iSRvaahs4WgjbW9jAOdeU4ZqoXzT9toPBd-7CCO4Ts4HQA671NbIIzIkOILwQx2tI_FgCroihk_CnjcQHa4fNIGzmgDIVm5fvekhsNfMgKnJ&typo=1> >>>> &c=E,1,5m3q9gwLjB5XgPfPMCUS8tSOyRmgs8 >>>> > oigic3gckRviSyeVj6o_a-VaEwwwPhcJt7wS65-CWA2pK-6wFh36mmNd1jOvXxdQYJTlzg >>>> > ky_HdveQg2ry8kGqzQ,,&typo=1 FRIAM-COMIC >>>> > >>>> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspo >>>> > t.com >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ft.com&c=E,1,-7XZAMRZP6TA8dSxx_53ATuhaKiiofnaexLhMNZ74_ziSZfpFtS3N_L93yVoiW9xuICVr2DylBCCoHCtdKgUwMx7Ht-W9A4Q0-4GYJ6oO7X41yHzWBVf3ujc-1w,&typo=1> >>>> %2f&c=E,1,g2WREPPELfNBefvRDG6LLsnR1UKZKBLNBalnm3PnxQvK_DKQefTBvLl >>>> > Zy9Ci6bQW70bCMXoDcDNcdxjOxvGNQiXt9fPCIhZ6E-r5HJQdjHEGI71jMQ,,&typo=1 >>>> > archives: http://friam.471366.n2.nabble.com/ >>>> >>>> >>>> - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . >>>> FRIAM Applied Complexity Group listserv >>>> Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fbit.ly%2fvirtualfriam&c=E,1,U5kTUt08kTFVFr_888krkhUOcdy-42InyaM9C85ohWzrIXEid56y63LZjtetYNvmhAm6NuUlEDq_i_9CQVEIv6WfSFf0XHzYusFI9lGg-Wzk-S4CnUo,&typo=1> >>>> un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,_MvX6cZ971lVWmf_de9GgiJAFkDss-FARY-tj1cUg5IgW3Lr5tGMlbTK602JPW7veof2cHnTK7rbZsSrfcAngBcoYxYgozpm_WnA0fSpLbrGgt7L&typo=1> >>>> FRIAM-COMIC http://friam-comic.blogspot.com/ >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,x51SIhs6yOhLAlN0oCuiZtxyx-zhAy2UJ5D1j024NXZS4gRqLlh8K40ROmgX2llQOaQkAGHT6X2iJmQn8ETSnhgE7PGRZtDth_pWgRgfT0YjZA4qJunvWrA,&typo=1> >>>> archives: http://friam.471366.n2.nabble.com/ >>>> - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . >>>> FRIAM Applied Complexity Group listserv >>>> Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fbit.ly%2fvirtualfriam&c=E,1,uyl0FI_o7_e9Ju7j_8fD9os9U9paR5bSSm63hmZe73ZVJdCLTheSxmXLnLsTgxU107cQyt36h7BhL39_4xX1xo-BoxSda0gDezCq5FD2o7I9LcctiOHH0w,,&typo=1> >>>> un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,HWeRupD9_Ml0WZpZw9-2f6n1_EJa35yT6Q5iwHXL08RmdHQIrVQMY_nqSLH7feeRgeTiCf2uNzu-RowMcjBWtSq8wn7dbkTb_ZsSsqgW&typo=1> >>>> FRIAM-COMIC http://friam-comic.blogspot.com/ >>>> <https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,R4_BP72WMF0a2IUAK0ZAkYXdpkJkbU5P7LEet-BiTPNGJdcHFryn2JrmsuI9mo5MfJxSMJLS1SNspeiuhuVsNpo8gve8sTc6lX7Nz4d-LvUob8PO2_E,&typo=1> >>>> archives: http://friam.471366.n2.nabble.com/ >>>> >>> - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . >>> FRIAM Applied Complexity Group listserv >>> Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam >>> un/subscribe >>> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,w-EqIn_ZNn9-26Q4yHRetL2lNbh91wH0BKI2YaaVBduUP7nWJ0zPpQ7BPVXEC6aXwtbZxx5IrPlIghnt5u2U9cTZdp3kgXIPjTGwCraw&typo=1 >>> FRIAM-COMIC >>> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,vu1KvMoJoBKEg8Oz03u4RRIbFfTls7eBtSCyOWq15FSEJeEVMSr_NmB67amb3wgexA9iyGry_EPje0JS097nxVbyjJLBCGiC1KbG0gbilO8,&typo=1 >>> archives: http://friam.471366.n2.nabble.com/ >>> >>> >>> - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . >>> FRIAM Applied Complexity Group listserv >>> Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam >>> un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com >>> FRIAM-COMIC http://friam-comic.blogspot.com/ >>> archives: http://friam.471366.n2.nabble.com/ >>> >> - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . >> FRIAM Applied Complexity Group listserv >> Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam >> un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com >> FRIAM-COMIC http://friam-comic.blogspot.com/ >> archives: http://friam.471366.n2.nabble.com/ >> > - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . > FRIAM Applied Complexity Group listserv > Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam > un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com > FRIAM-COMIC http://friam-comic.blogspot.com/ > archives: http://friam.471366.n2.nabble.com/ >
- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . FRIAM Applied Complexity Group listserv Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com FRIAM-COMIC http://friam-comic.blogspot.com/ archives: http://friam.471366.n2.nabble.com/
