Quote:
Originally Posted by GoSharks
No, @arjaybe just said that replicating real world conditions "would be an untenable restriction to impose on experimenters"...which is an even bigger laugher.  |
Except that real world conditions would result in a meaningless uncontrolled muddle AKA an even bigger than bigger laugher.
For a real world study, there needs to be a preliminary study of real world use. Consider that a majority of posters on this thread think the great majority of tablet users turn down the light intensity at night, but a minority, including myself, think most do not. Since majority doesn't rule on factual questions, the researcher would have to study this first.
After the preliminary study was used as the basis for another sleep study, variables such as ambient room lighting and reading time would no longer be controlled. That's because, in the real world, people who read on different devices are different in their reading habits in other ways. There would then be no basis to say which variable caused any difference in sleepiness.
Most likely, average ambient room lighting used with tablets is different from than that used with eInk is different from that used for paper books. So you then, in a real world study, couldn't say whether the effect was due to the device or the ambient light.
Another real world difference is that people who use tablets switch back and forth between reading, web surfing, gaming, etc. Book and eInk readers rarely surf on their device. They just read*. And they may use their device (considering here that a paper book is also a sort of device) for a shorter period of time because it is more limited. On the other hand, paper book readers may watch more television. Or less. I don't know. I do know that there is a heck of a lot of difference, in the real world, in how people use screens depending on their chosen device.
You can have real world, and you can have controlled experimentation. You can't have them both.
One corollary of this is that every study has strengths and weaknesses. Controls and real-world are both virtues, but the more you control, the less the experiment is real world.
This issue is not limited to sleep medicine. Real world medical practice, vs. controlling as many variables as possible, is, or should be, considered, and even agonized over, in design of most drug studies.
Some experiments are better than others. One reason the study under discussion is better than most is the careful control of variables so that the only difference between the two groups is the reading device. Is that real world? No. Does it tell us something real about the effects of different reading technology? Absolutely.
_____________________
* I use the Kindle Keyboard web interface a lot. But this is atypical.