That makes sense but aren't IQ tests supposed to be done at a young age for that very reason? The only legit IQ test I have ever had was when I was in 1st grade as did everyone at my school. NowI could train myself to do better on a IQ test just like the SATs GREs and LSAT.
How is culture going to impact 6 year olds? Also as a more practical matter how often was IQ testing being done around the turn of the century? How were they getting the mean?
I'll give you a personal example. I had my kid take an IQ exam at 3.5 y.o. for access to the prep school pre-k program. He wasn't in preschool at the time so I didn't have preschool teacher recommendations (which seems insane as I type it). He wasn't in preschool because they had to teach to the average kid and it was boring my child to tears.
On the language section, he tested in the 99th% but here's an example of something that he got wrong. He's supposed to identify a pictured item. The item in the picture is a violin, my son said "guitar". To a relatively young child, the picture of a guitar can easily resemble the picture of a guitar and if the child has more experience with one instrument over the other he'll select the one he's familiar with. Well, depending on your community, some kids might not have ever seen a violin.
Another example is the days of the week. My son didn't know them because we hadn't taught him yet. He was doing addition, subtraction, and reading at a 1st grade level at 3, we weren't worried about if he'd ever learn the days of the week, lol. But the IQ exam asked him to name them. He got that wrong. In the report, the tester pointed out that his IQ score, while still in the 99th% percentile, could have been higher if he'd known some of those things. Now for my kid, the difference between the 99.X% and the 99.Y is meaningless in terms of how he's going to be perceived.
But those aren't raw intelligence questions, they're acquired knowledge questions. In a culture, where the acquired knowledge at the tested age is low - such as in poor communities with bad schools - then those kids will test lower because of that.
There's another section on putting together blocks to match an abstract image. Well, in my house, we have plenty of Legos, puzzles, blocks, etc. These are things that align with the IQ exam but some kids might not have the same amount of exposure and so the block portion of the IQ exam is the first time they're exposed to the concept. THey're going to score lower than kids with greater exposure/experience.
It's an interesting conversation because the exam tests both pure reasoning and acquired knowledge. And there are so many places in society where we can't be sure that kids are being exposed to the foundational skills that the IQ exam is testing. They'll get lower IQ scores but that doesn't necessarily mean that they're unintelligent.
And they weren't doing much in the way of IQ exams at the turn of the century, not the way you're thinking about it. And even then, the scores were ratio scores, not percentile scores. The ratio score simply took your mental age divided by your chronological age. So an IQ of 150 equated to a 6 year old with the mental ability of a 9 year old ( 9/6 = 1.5. 1.5x100 = 150 IQ). And an IQ of 80 meant a 10 year old with the mental ability of a 8 year old (8/10 = .8. 0.8 x 100 = 80 IQ). Well, given the education standards of the time - how easy is it for a 10 year old to have only acquired the education of an 8 year old before leaving school. We switched to rarity scores precisely to minimize that impact. So now an IQ of 150 means that that you score better than 99.9% of other people in your age taking the same exam. But that means that an IQ of 70 means that something like 99% of the population would have outperformed you, there's nothing strange with 95% of the modern population in the U.S. knowing more than some dude from the 1800s.
So how do you responsibly speculate that backwards over multiple generations? You really can't. And how do you speculate test scores in countries with extremely different education standards? Again you can't. And how do you compare scores in communities where the parents are paying for nannies to individually treat and teach their kids to kids in communities where the local daycare barely teaches the basics? You can't.