IQ scores are falling and have been for decades, new study finds

It takes a lot more than one paper and a CNN article to overturn the evidence that is against this. I'm not talking about IQ scores falling, but rather the cause of it suggested here. Although the environment does have notable roles (for example the impact of lead), it's naive to come out and say that the environment is all that matters in driving these effects. Dysgenic Fertility (https://en.wikipedia.org/wiki/Dysgenics#Fertility_and_intell...) is different in different countries, and the pattern of change mentioned in the article is also different, with Norway often being a bit of an outlier.

>It's not that dumb people are having more kids than smart people, to put it crudely. It's something to do with the environment

This is false. The correlation between cognitive ability and number of offspring is negative. Here's a link to a systematic review on this topic: http://sci-hub.tw/https://www.sciencedirect.com/science/arti... that covers many datasets, where you can see the negative correlations on page 114 (8 in this pdf), which vary from -0.01 to -0.27 depending on the group in question. Although this does not rule out an environmental influence by itself, it's dishonest to simply dismiss it as the article has done.

> In a separate study that has not been released, he and his colleagues looked at existing research in an effort to demonstrate that staying in school longer directly equates to higher IQ scores. Keyword here is "an effort to demonstrate". That sounds like they really want the results to come out in their favor, and we know that is not a good sign. They can mess around with correlation and causation and p-values and subgroups and framing all they want, but whatever they come up with will contradict a lot of the existing literature in this area in order to push their agenda.

The simplest explanation would be that either the IQ tests are missing some aspect of human cognition that is getting more developed in spite of others, or that the weighting used to calculate the final IQ number based on the particular results might be wrong or not representative of the actual cognitive effort required for each of them.

Let's not forget that IQ tests are just statistical analyses, while we still lack a bottom-up explanation of intelligence.

This is not a new discovery. It's known as the Flynn effect. [1] It initially was the observation that IQ scores were constantly increasing over time. This was one of big correlations in support of intelligence being driven by environment. Our environmental conditions were improving, and so was our IQ.

But then sometime starting around the mid 90s it began to reverse in developed nations. And by reverse, I mean IQ's began declining - not approaching some sort of asymptotic 0 level of growth. So, if we normalize for age, an average person born in the 70s will have a higher IQ than an average person born in the 90s.

[1] - https://en.wikipedia.org/wiki/Flynn_effect

I have no PNAS*, but just for starters...

Firstly, this is such a massive and important finding, even cessation of the Flynn effect would be, that one has to wonder why we haven’t seen it published until now.

Secondly, one of the frustrating parts of being a statistician is that a 90% explanation of nearly every effect comes from some common-sense factor. So much so that it raises suspicions when a strong effect is shown along with a series of specific and fairly complicated ‘anti-explanations’. The article seems more interested in disproving the ‘idiocracy’ effect than finding some alternative explanation.

Literally for decades people have been predicting average IQ scores would drop due to social/genetic factors. It's evidence of a successful theory. Society should stop pandering to the lowest common denominator and allow gifted students to excel in separate environments

A few days ago naturalness.com had an article about this topic and refers to 3 scientific studies that show that the more toxins the mother was exposed to, the less IQ the child has. It is a linear function.

Dupe of https://news.ycombinator.com/item?id=17302830