The first randomized evaluation of the 1 Laptop Per Child initiative in Peru a couple months back showed NO effects on :
– Test scores in reading and math
– Improvements in enrollment or attendance
– Changes in time spent on homework or motivation
BUT … some improvements were noted in cognitive ability. So, what can we glean from this? That laptops don’t do much, that kids will take a computer and do what they will with it, learning be damned? Probably not, but the bigger concern should not be the intervention, but rather the scale of the intervention, especially considering there was no rigorous assessment of this initiative prior to the Peruvian government shelling out $180 million to purchase 900,000 laptops. Not a drop in the bucket for a developing nation.
In Nepal, similar findings were reported but their Department of Education thankfully spent a whole lot less opting to do a couple small pilot tests first. When I read evaluation reports like this, and then see what Mayor Bloomberg for example is doing in New York with the banning of soda sales in 16 ounce + cups, and now talking about looking into popcorn and other such food bans, I ponder, why can’t we just step back, run some randomized trials in a couple communities, test the effects of these proposals before rolling them out in the most densely populated city in the U.S.? Is that too much to ask?
I understand the quick trigger pull on this, especially when it comes to something like a computer. But the question is not is a computer bad for a kid, or even, does it have a negligible effect, but rather, we (the government) have $180M, what’s the best use of these funds to get kid test score and attendance levels up? Same in New York, would be great if folks tempered their consumption of bad food. But that shouldn’t be the question everyone is asking, rather, how much are we (tax payers) spending on this, and can this money be better spent on another initiative seeking to curb obesity. Test the puppy out and once you have some auspicious results at a small level, then roll it out, bad idea?