« Big money in online poker | Main | Bad oil advice from the NY Times »

June 26, 2005

Comments

intelligence (as measured by an IQ-type test) in childhood predicts substantial differences in adult morbidity and mortality

Could go the other way though, too.

illness etc etc in childhood predicts substantial differences in intelligence (as measured by an IQ-type test)

Didn't and won't read the article to see if they causality stats.

Like Varangy I was somewhat sceptical of this, so I did some digging of my own.

For some reason, the .pdf link would not load for me but not to be put off I did find this...

http://news.bbc.co.uk/1/hi/health/1260794.stm

including -

"Scientists studied IQ tests taken by 2,792 11-year-old children living in Aberdeen in 1932.

They then followed up to see who was still alive in 1997.

It was found that both men and women with higher IQ scores as children were much more likely to be alive 60 years later.

The researchers took an average IQ in the group as 100.

Those still living at the age of 76 had an average IQ score of 102 at the age of 11, while those who had died by 1997 had an average score of 97.7. "


Now to quibble just a bit, a study of 2,792 11-year-old children living in Aberdeen in 1932is not "Large epidemiological studies of almost an entire population in Scotland".

Another quibble, neither your link nor the Beeb article mentions the number of surviving participants from the original sample. From the point of view of statistics this is a fundamental in defining the dependability of the conclusions drawn. For example, if the number of survivors was 10, the conclusions would be weak to say the least. If there were 1,500 survivors, then the conclusion would be more dependable and a meaningful level of confidence could be calculated.

What a pity I can't get hold of that .pdf... or any of Gustafsson.

Hi Half Sigma
One obvious connection between IQ and longevity is via genetic diseases that cause both low IQ and reduce lifespans. There are many of those. It would just take a few people with such diseases to skew the results.

There is research that shows poor people have shorter life expectancy than wealthy people. There also is a correlation between IQ test scores and wealth.

These correlations are small, but what causes what? Do poor people work in more physically demanding or more risky jobs, which shorten their life expectancy? Do wealthy people get better healthy care? Do wealthy people take better care of themselves, eat healthier foods, manage their weight better? Is any of this because wealthy people are smarter or because they are more wealthy?

I am not surprised at all that “smart” people are on average more wealthy and in better health. My issue with all this research is that I think IQ tests only measure one narrow aspect of what it means to be “smart”. That is why IQ test scores are only weakly correlated with success in later life.

mikeca, "success" is a pretty loaded word and it means different things to different people. And "weakly" isn't exactly what I'd say about the correlation between IQ scores and "success." I'd say it's "significantly" correlated, using the word significant to indicate that although the correlation is not anywhere close to 1.0, it's big compared to any other factors correlated with success.

And for some reason, there seems to be a lack of curioisity regarding why one person with a high IQ is a success while another is a failure. How do the two individuals differ? What are the mysterious other factors?

At the national level, IQ and life expectancy correlate at a statistically significant .85, the strongest correlation I've found of all the variables I've looked at (more than wealth, crime, even doctors per capita).

The comments to this entry are closed.