3 ms·
Dismissing a study based on it being single is idiotic. The study was based on 71 ,527 individuals, it is statistically significant. http://www.ncbi.nlm.nih.g
by ye 13y ago
Dismissing a study based on it being single is idiotic.
The study was based on 71 ,527 individuals, it is statistically significant.
http://www.ncbi.nlm.nih.gov/pubmed/24217719 http://www.ncbi.nlm.nih.gov/pubmed/24217719
Let me guess, you're obese?
- gizmo686 13y agoA difference of 10^-100% is statistically significant with a large enough sample size. The question of actual significance is how relevant is the measured difference. Statistical significance asks how likely it is that the difference actually exists.
- king_jester 13y ago> Dismissing a study based on it being single is idiotic. I am not dismissing this study. It is interesting and has merit as a study. > The study was based on 71 ,527 individuals, it is statistically significant. I am not disputing the study or its methodology, only that drawing a final conclusion about weight and heart health from a single study is generally a bad idea. Meta analysis of studies on being overweight and obese (EDIT: by BMI category) have shown different conclusions on the impact to health. This is covered under that linked news story paragraph titled 'Mixed Results from Previous Studies'. > Let me guess, you're obese? Not at all, I'm not even overweight. (EDIT: By BMI, I don't meet those categories. BMI is a horrible metric).