> Bet you'd drive a car nearly as well as a old-time racing champion too,
> from all the practice you've got that they didn't have. I mean we've
> had decades of refining teachers's techniques of training kids to excel
> at answering short snappy quiz questions. You don't really suppose that
> we're actually a whole lot smarter today than they were a century ago,
> do you?
That's actually the idea. In fact, it's on *non-verbal,* problem-solving type tests (e.g., if I fold this piece of paper like this and then punch a hole in it here, where will the holes be when I unfold it) that show the largest IQ gains over time. These tests aim to strictly measure so-called "fluid" intelligence, i.e., tests that are not dependent upon formal knowledge and are independent of "culture." IQ tests that include verbal components and measure both fluid and so-called "crystallized" intelligence have lower rate gains, and, of course, purley verbal tests the lowest. (The rate difference is about 15 points per generation on tests of fluid intelligence like the Ravens and 9 points per generation on mixed tests like the WAIS and Stanford-Binet.) This suggests that people *are* getting smarter (on average, anyway), or, at least, that it isn't formal education that is making the difference. In other words, people don't *know* more (at least not much), but they are *performing* much better over time, on average. Some people have theorized the gains are due to better nutrition which, I assume, aids development of the noggin.
And keep in mind this isn't just the U.S. Increases in the 20th century exist in all countries for which there is data. Lately, the slowest gains have come from countries like Sweden and Norway. Of course, Norway also scores highest on the Human Development Index, suggesting IQ gains will top out at some point for all countries when they catch up. (Norway may even have stopped having gains at all: http://linkinghub.elsevier.com/retrieve/pii/S0160289604000522) Since the Flynn effect only concerns the rise in mean IQ scores, it has been suggested it is being driven from the bottom, i.e., people on the lower end of the scale performing better and driving up the average over time.
Flynn's data in an older paper can be found online here, although I wouldn't endorse the site (I haven't looked at it and don't know what it is, other than that it's taken the time to put up the comparative data from one of Flynn's papers): http://www.iqcomparisonsite.com/Locations.aspx