3 ms·

Can I remind everyone "Correlation does not imply causation"?

• eksith 14y ago
  Yes, but after the 8th time I left my TV remote in the fridge after many sleepless nights when before, I could remember how many inches away the keyboard is from the monitor... There may be something to this.
• dhimes 14y ago
  I don't understand why you are being downvoted- this is an essential question to answer. For instance, Alzheimer's patients show sleep disorder- specifically, they tend to not sleep as much as non-Alzheimer's patients. There may be some evidence that this starts before the symptoms of dementia are diagnosed. If it turns out that memory issues cause poor sleep, that could potentially be a big break in early diagnosis.
• danteembermage 14y ago
  And in this case there is an obvious omitted factor (which is important to consider when levying this allegation) age. Perhaps being old makes you sleep worse and have trouble retaining memories, but sleep has nothing to do with memories at all (although we have lots of reasons to suspect that it does). They would also have found arthritis, type II diabetes, prostate cancer, and a host of other things reduced memory if their design was as described by the article. Given that they have fMRI data it would be fairly simple to run a correlation between quality of sleep and retention within an age cohort and see what you get. Given that they got funding to do the MRIs I'm suspicious that in the actual study they did, I'm not sure you need scans to know the elders often don't sleep well. But if that works, perhaps it was weight causing poor sleep and memory retention. Or alcohol abuse. I'm not saying this research can't be done but it's a lot more trustworthy if you either 1. randomize subjects and do a treatment or 2. have a much larger dataset and statistically control for these suspicious factors.