The problem with those tests is that they did not screen for cultural/educational bias. I would expect people who are literate with better educations to do better than those without.
Well, I think it's inescapable that educated literate people do better at most jobs. So if the IQ test preferentially selects educated literate people, so much more power to the IQ test.
But the tests, particularly in that period were not at all useful for determining people's intelligence.
Who wants to measure people's intelligence? -- whatever that is. There is no agreed-upon definition of what intelligence is -- nor of what you can measure -- nor even of what you want to measure. The whole exercise is a waste of time.
But if you are looking for a practical test that measures people's potential to do well in (and to find fulfilment in) especially in the intellectual professions, then the IQ test is the test for you.
Abandon the notion that the IQ test is a bad test, because it purports to measure intelligence. Trying to measure (or even define) intelligence is the ultimate mug's game.
Instead, regard the IQ test as a cheap practical convenient way of selecting the top ten out of a hundred job candidates -- being a selection method that yields results that are better than drawing names out a hat. Regard the IQ test, not as a measure of the person's intelligence, but as a measure of the person's ability to do well in jobs.
The scenario you propose, already pre-screens in some way, but limiting your pool to "qualified candidates."
Well, in any practical job selection process, you would pre-screen the candidates. Obviously, many candidates de-select themselves. Don't waste resources on them.
We also really don't know what "g" measures. Even the most rudimentary testing done today usually breaks IQ into verbal and non-verbal. Which is just a start. It's a very complex area right now, and except for limited applications, I don't see a great deal of use for it.
If you want the IQ test to measure intelligence, I agree that it is of little use. In fact, it is worse than no use -- it does harm -- it is divisive and it enables people to label people.
But, again, as a way of selecting from candidates for a job, it is highly effective especially in the intellectual professions, but also in perhaps-surprising fields, like tool-makers.
If we would save the IQ test for
that purpose, and don't worry about pure academic nonsense like whether it measures "g", the IQ test could be very useful indeed. Join me in re-naming it the test-for-who-is-likely-to-do-well-in-this-job-that-requires-thinking.