4 ms·

For those interested, the study from Ms. Packham: https://apackham.github.io/mywebsite/opioidpaper_webcopy.pdf https://apackham.github.io/mywebsite/opioidpaper_

• rhaway84773 4y ago
  That’s interesting. I read the approach details. Nothing in there indicates why the results should reflect causation versus correlation. Before reading that I thought that maybe the SEP groups were wrong to call the publishing of the paper a mistake, but now having read it and seeing that it presents correlational data as causational data I understand why those groups were criticizing the research itself. I’m not an expert in this field so I would appreciate if others can also read the actual paper and explain to me why the author is correct to claim causation instead of correlation.
  • psychphysic 4y ago
    I think most of the lay public have the wrong end of the stick. Proving causation is the cherry on the top of research. Most of the time correlation and mechanism is all that can be found in a timely manner. This data is really quite good. Correlation between smoking and poor health was shown decades before causation was considered proven. It'd be ridiculous to tell someone don't worry keep smoking, all the data we have this far shows only correlation! Right up until the 50s, when in the 20s it became apparent smokers died young. What we see here is people moving thresholds depending on their bias. This is the root cause of confirmation bias.