Uploaded February 2020 | Updated September 2026, 1 week ago
The Mendelian randomization studies reviewed in this video...
Gage et al (2017). "Assessing causality in associations between cannabis use and schizophrenia risk: a two-sample Mendelian randomization study." @ ncbi.nlm.nih.gov/pmc/articles/PMC5341491
Vaucher et al (2017). "Cannabis use and risk of schizophrenia: a Mendelian randomization study." @ ncbi.nlm.nih.gov/pmc/articles/PMC5984096
Pasman et al (2018). "GWAS of lifetime cannabis use reveals new risk loci, genetic overlap with psychiatric traits, and a causal influence of schizophrenia." @ ncbi.nlm.nih.gov/pmc/articles/PMC6386176
The sensitivity analyses reviewed @ 4:03 are from Pasman et al (2018), Supplementary Figure 9(a)(b)(c) @ static-content.springer.com/esm/art%3A10.1038%2Fs41593-018-0206-1/MediaObjects/41593_2018_206_MOESM1_ESM.pdf#page=17
My description of P values in this video is influenced by Sander Greenland and others [1,2] who advance an interpretation some call 'compatibilism', [3] which does a good job of both avoiding common misconceptions and being intuitive and simple to understand. Over the last couple decades their views have been gaining the upper hand, [4] eroding over-reliance on P values and statistical significance, ushering in a far more ‘relaxed’ non-alpha-centric interpretation of P values. I strove in this video to convey that interpretation, which includes not 'accepting' or 'rejecting' a test hypothesis simply because its p value is over or under 0.05.
[1] link.springer.com/article/10.1007%2Fs10654-016-0149-3
[2] tandfonline.com/doi/full/10.1080/00031305.2018.1529625
[3] lesslikely.com/statistics/s-values
[4] nature.com/articles/d41586-019-00857-9
The Mendelian randomization studies reviewed in this video...
Gage et al (2017). "Assessing causality in associations between cannabis use and schizophrenia risk: a two-sample Mendelian randomization study." @ ncbi.nlm.nih.gov/pmc/articles/PMC5341491
Vaucher et al (2017). "Cannabis use and risk of schizophrenia: a Mendelian randomization study." @ ncbi.nlm.nih.gov/pmc/articles/PMC5984096
Pasman et al (2018). "GWAS of lifetime cannabis use reveals new risk loci, genetic overlap with psychiatric traits, and a causal influence of schizophrenia." @ ncbi.nlm.nih.gov/pmc/articles/PMC6386176
The sensitivity analyses reviewed @ 4:03 are from Pasman et al (2018), Supplementary Figure 9(a)(b)(c) @ static-content.springer.com/esm/art%3A10.1038%2Fs41593-018-0206-1/MediaObjects/41593_2018_206_MOESM1_ESM.pdf#page=17
My description of P values in this video is influenced by Sander Greenland and others [1,2] who advance an interpretation some call 'compatibilism', [3] which does a good job of both avoiding common misconceptions and being intuitive and simple to understand. Over the last couple decades their views have been gaining the upper hand, [4] eroding over-reliance on P values and statistical significance, ushering in a far more ‘relaxed’ non-alpha-centric interpretation of P values. I strove in this video to convey that interpretation, which includes not 'accepting' or 'rejecting' a test hypothesis simply because its p value is over or under 0.05.
[1] link.springer.com/article/10.1007%2Fs10654-016-0149-3
[2] tandfonline.com/doi/full/10.1080/00031305.2018.1529625
[3] lesslikely.com/statistics/s-values
[4] nature.com/articles/d41586-019-00857-9
