3,014 followers
RT @Lester_Domes: @OstlundOllie @apsmunro @DrNickTwit Cornfield 1976 https://t.co/vcrhRtCTRq gave a Bayesian account of randomization as an…
3,565 followers
RT @Lester_Domes: @OstlundOllie @apsmunro @DrNickTwit Cornfield 1976 https://t.co/vcrhRtCTRq gave a Bayesian account of randomization as an…
4,210 followers
@OstlundOllie @apsmunro @DrNickTwit Cornfield 1976 https://t.co/vcrhRtCTRq gave a Bayesian account of randomization as an act of rendering exchangeable priors (rexing) for potential outcomes across treatment groups. That exchangeability is then voided by o
6,977 followers
RT @stephensenn: @PavlosMsaouel has reminded me that @Lester_Domes has a paper highlighting similar issues https://t.co/Lmhdf49z7P. I would…
574 followers
RT statsepi https://t.co/VhaCD4jWOv "RT stephensenn: PavlosMsaouel has reminded me that Lester_Domes has a paper highlighting similar issues https://t.co/co9IwKzzOq. I would…"
28,052 followers
RT @stephensenn: @PavlosMsaouel has reminded me that @Lester_Domes has a paper highlighting similar issues https://t.co/Lmhdf49z7P. I would…
3,014 followers
RT @stephensenn: @PavlosMsaouel has reminded me that @Lester_Domes has a paper highlighting similar issues https://t.co/Lmhdf49z7P. I would…
12,257 followers
@PavlosMsaouel has reminded me that @Lester_Domes has a paper highlighting similar issues https://t.co/Lmhdf49z7P. I would also value the opinion of @omaclaren whose preprint is cited https://t.co/SfZ1Hwfm6l
172 followers
RT @stephensenn: @snoble @yudapearl As far as I am aware (but would be interested to see evidence to the contrary) the theory cannot distin…
772 followers
RT @stephensenn: @snoble @yudapearl As far as I am aware (but would be interested to see evidence to the contrary) the theory cannot distin…
772 followers
901 followers
RT @stephensenn: @snoble @yudapearl As far as I am aware (but would be interested to see evidence to the contrary) the theory cannot distin…
12,257 followers
@snoble @yudapearl As far as I am aware (but would be interested to see evidence to the contrary) the theory cannot distinguish between different kinds of RCT and has nothing to say about standard errors. However this paper https://t.co/e0hxNmCv1B may be r
12,257 followers
@hmatejx @soboleffspaces @yudapearl I suspect that this https://t.co/Lmhdf49z7P by @Lester_Domes and Mansournia is relevant.
4,210 followers
@stephensenn @analisereal @yudapearl @PavlosMsaouel @ESteyerberg @LinkedIn @MariaGlymour @NP_Jewell @XiaoLiMeng1 Careful with terms: "You can get rid of the confounding by declaring hall random"? Does that mean I can get rid of confounding in comparing 2 h
3,014 followers
RT @Lester_Domes: @PavlosMsaouel @drjgauthier @yudapearl @f2harrell @kdpsinghlab @AleksiReito For even more details see the new overview ht…
4,210 followers
@PavlosMsaouel @drjgauthier @yudapearl @f2harrell @kdpsinghlab @AleksiReito For even more details see the new overview https://t.co/KXYY7UCPX7 but note conflicts in use of 'confounding' in RCTs (random confounding), reviewed here https://t.co/MZDHxqnKVx, a
957 followers
RT @_MiguelHernan: “Bad luck” means that some prognostic factors happened to be more common in the intervention group. In observational st…
143 followers
RT @_MiguelHernan: “Bad luck” means that some prognostic factors happened to be more common in the intervention group. In observational st…
1,265 followers
RT @_MiguelHernan: “Bad luck” means that some prognostic factors happened to be more common in the intervention group. In observational st…
1,995 followers
RT @_MiguelHernan: “Bad luck” means that some prognostic factors happened to be more common in the intervention group. In observational st…
4,249 followers
RT @_MiguelHernan: “Bad luck” means that some prognostic factors happened to be more common in the intervention group. In observational st…
2,018 followers
RT @_MiguelHernan: “Bad luck” means that some prognostic factors happened to be more common in the intervention group. In observational st…
562 followers
RT @_MiguelHernan: “Bad luck” means that some prognostic factors happened to be more common in the intervention group. In observational st…
43,147 followers
“Bad luck” means that some prognostic factors happened to be more common in the intervention group. In observational studies, we call this imbalance confounding. In randomized trials, the imbalance occurs by chance so we can call it "random confounding."
28,052 followers
@Research_Tim @tmorris_mrc @stephensenn This bit seems to get at the part I was referring to: https://t.co/qMbykSlyFc
249 followers
RT @_MiguelHernan: @statsepi @PWGTennant @ken_rothman @statsmethods 3) ... which is why some people refer to this form of bias as "random c…
263 followers
RT @_MiguelHernan: @statsepi @PWGTennant @ken_rothman @statsmethods 3) ... which is why some people refer to this form of bias as "random c…
379 followers
RT @_MiguelHernan: @statsepi @PWGTennant @ken_rothman @statsmethods 3) ... which is why some people refer to this form of bias as "random c…
43,147 followers
@statsepi @PWGTennant @ken_rothman @statsmethods 3) ... which is why some people refer to this form of bias as "random confounding". See https://t.co/g1Cih7XkrV by Greenland & Mansournia
60,492 followers
16,105 followers
179 followers
43,147 followers
@PWGTennant @statsepi @EeroTeppo @ken_rothman @f2harrell @fledglingStat @kippwjohnson @thebyrdlab @stephensenn If interested in "random confounding", check out this paper by Greenland & Mansournia (also Ch 10 of our CI book) https://t.co/g1Cih7XkrV