|
| quote: | Originally posted by shaolin_Z
I fail to see these "gaping holes" in logic, assuming one is not blind to the context. |
Well, to begin with they are merely "reviewing" 173 previous studies. 80% of them suggest one result, and 20% of them suggest otherwise; therefore, the 80% are correct? Realistically, there are probably enough methodological differences between studies that this is not an apples-to-apples comparison anyway, but even if that weren't the case you cannot summarily infer that the more common result is the accurate one.
Even if their method was sound, and it isn't, they would have still only succeeded in demonstrating an apparent correlation. There is no grounds in the data from which to infer causation. There may be other reasons to infer that, for example, sitting on the couch all day will tend to cause a higher rate of obesity, but this study adds nothing to them.
Additionally, even if they had shown causation, they would merely be committing a complex cause fallacy: surely large amounts of time spent on the internet is neither necessary nor sufficient to cause, for example, a child to begin smoking. As a result, there is no grounds for their apparent conclusion that reducing media exposure would produce any substantial benefit -- and even if it would produce a substantial benefit, they fail to establish that there is not a more efficient way to achieve the same result by addressing other causal factors.
Furthermore, even if they had conclusively shown that media exposure contributes to health problems and that it has sufficient "causal potency" that reducing media exposure would produce a substantial health benefit, they are still too hasty in suggesting that parents "limit their children's exposure to media." Sending your children to school no doubt increases their odds of being bullied, but that does not suffice to establish that it would be wise to keep them home all the time. Their recommendation ignores the possibility that media exposure has benefits outweighing its potential costs. It also ignores the possibility that whatever other activities these particular children would partake in would pose a risk of harm as well...
They also lump media together indiscriminately -- probably some television programs have greater net utility than others, for example. But they conclude with a blanket recommendation that parents limit "media exposure." Their blanket recommendation also does not define what "limit" means: are they suggesting that reducing exposure is always going to produce a health benefit, regardless of the initial degree of exposure or the amount of reduction? At best, it's unclear.
Finally, even if they had used a valid method of inquiry; and shown that there was more than a mere correlation; and established that media exposure had sufficient causal potency; and demonstrated that the potential risks of media exposure outweighed the potential benefits in all circumstances; it would not be enough to show that their recommendation is wise (although at that point it might at least be justified on a purely theoretical level.)
To see why this is the case, consider the inherent futility of their recommendation. Since media exposure is not sufficient to produce any of these adverse health consequences, at least some other contributing factor would be necessary. The most likely candidate would seem to be "inattentive parenting." Now consider which parents would be most likely to heed this recommendation -- it's precisely those parents whose conduct is probably already sufficient to avoid the harm that the recommendation is intended to avoid. And since we know that media exposure alone is insufficient to cause this harm, and we can probably infer that at least some media exposure has some benefit, it may well be the case that the recommendation would produce a net loss of utility, even if it would have produced a net gain of utility if it had been heeded by all parents.
I don't hold this last point against them, but it's still lurking in the background.
My real problem here is not so much with their message (although I think it is horribly oversimplified at best). Rather, I object to the presentation of research based on unsound methods and questionable logic as if it were legitimate. Not only is it a waste of resources that could have been put towards legitimate inquiry, but when inevitable contradictory findings are reported elsewhere, it will undermine the public belief in science as a method of inquiry, producing dangerous externalities. Furthermore, it serves to lower the standards of what constitutes genuine scientific inquiry and denigrates the efforts of those who seek to actually animate and explain the data by examining the way in which complex phenomena interact piece-by-piece, rather than throwing broad categorical data into the statistical centrifuge and slapping a conclusory label on whatever reaches the point of "statistical significance" first.
|
.
.
