Here is a long and valuable paper on the topic. From the abstract:
Consistent with positive selection of emigrants in terms of observable skill, emigration rates appear to be highest among individuals with earnings in the top half of the wage distribution.
There is much more along those lines. To be frank, I know this paper will not convince most of the skeptics. They will say, or perhaps think, "Yikes, what must the others be like?" But at the very least evidence should improve a debate. The next time you hear it argued that we receive "the dregs" of Mexico, send along this link.
The paper also finds that wages tend to rise in parts of Mexico where many people leave. You could argue this one of two ways. First, it might cause you to doubt David Card’s view that wage effects in the U.S. are small (although the U.S. is a much bigger economy and thus the labor shift should have a smaller impact here). Second, it raises our estimate of how much Mexico benefits from emigration.
Thanks to Eric Husman for the pointer. Here is another relevant paper on Mexican emigration, forthcoming in the Journal of Economic Literature. Full of facts, as they say.