First of all, Lesswrong is the site where I always put my lasts hope.
I’m having lots of troubles trying to find the truth about IQ test taken by different ethnic groups
It seems there are lots of studies claiming differences in IQs. On the other side there is a lot of people saying the contrary (culturally biased tests,..)
What do we do as rationalists? I’m really confused. I have read tons of articles from both sides yet nothing is clear to me.
Speaking for myself, my position is “I don’t know”.
Ignoring the specific question, there are many situations in my life where (a) I am curious about something, (b) I don’t trust the existing research, and (c) it is not high enough priority for me to try doing the research myself. In such case, thinking “I don’t know” seems like a reasonable reaction. What else should I think?
In absence of solid research, people often return to armchair reasoning, inventing clever arguments why in absence of evidence we should stick with “default” opinion X, and put the whole burden of proof on people who say Y. Problem is, in the next room, people use similar armchair reasoning to argue that we should stick with the “default” opinion Y, and put the whole burden of proof on people who say X. I could easily provide “a priori” arguments for either position here, which is why I consider neither of them convincing.
Here are your options:
decide that you feel better about believing that the different ethnic groups have the same average IQs;
decide that you feel better about believing that the different ethnic groups have different average IQs;
do the research, which can take you a few years and probably more money than you have; or
accept that you do not know the answer, and learn to live with that feeling.
More importantly, the question is what purpose do you actually need the answer for.
Are you trying to find intelligent people? Then try to measure or estimate their IQs as individuals.
Are you deciding which country to live in? There are probably many other factors to consider, so use those.
Are you trying to win an internet debate? Consider better ways to spend your time.
Are you researching a transhumanist technology for creating superhumans? You will have to research the specific alleles and their interactions; knowing about averages of large populations is not going to help much.
EDIT:
I think there may be one situation where knowing the answer actually would be practical: If you want to start a project in a foreign country, and the success of the project depends on the country having enough high-IQ people. For example, you might be a philanthropist billionaire trying to build a university (or, more meaningfully, the whole educational structure, starting perhaps from pre-school education, and ending with a university) in the middle of a country that according to some sources is full of people stupid for genetical reasons (i.e. not just bad nutrition, etc.), so your project of achieving university-level education for local people may be doomed to fail.
I guess you’d just have to take the risk, and possibly find the answer to the question as a side effect.
One more thing to consider is that IQ is caused partially genetically, and partially non-genetically, e.g. diseases or lack of nutrition decrease IQ. So if you e.g. examine people from a sick and starving population, of course they are likely to have below-average IQ. But that doesn’t say anything about what IQ their descendants will have if the food and health problem gets fixed.
Intelligence is a polygenic trait, i.e. a trait influenced by multiple genes. There is an observed regression to the mean, that is although smart parents are likely to have smart children, and dumb parents are likely to have dumb children, the children of either are usually closer to the average than their parents.
In other words, if you would inhabit an island exclusively by Mensa members, the children born on this island would probably almost all have above-average intelligence; but many of them would not reach the Mensa level. Or the opposite experiment… well, this one was actually done in real life… 40-50 years ago when communists ruled Cambodia, they killed almost all literate people in the country in the attempt to create an agrarian utopia (spoiler: didn’t work as advertised), but Cambodia didn’t literally become a nation of retards.
It is difficult to find exactly which genes contribute to IQ. There are more than 50 suspects, but the experiments suffer from low sample sizes, so many of them are probably false positives.
It seems that first-born children have higher IQ than their siblings. (The official story seems to be that it’s because they get more parental attention and resources. To me it seems more likely that children born later simply receive higher mutational load from older parents. But maybe it’s both.)
It was suspected that breastfeeding increases IQ. Then it turned out this correlation was caused indirectly by mother’s IQ; i.e. smart mothers are more likely to breastfeed their children, and smart mothers are likely to have smart children, which creates a correlation between smart children and breastfeeding even if breastfeeding has no real effect on IQ. Later, there was an experiment showing that actually children with some genes may benefit from breastfeeding while children with other genes may not… but the experiment is unrealiable because of small sample size. (If you start researching this topic seriously, you are going to hear “small sample size” depressingly often.) If true, this wouldn’t be completely suprising, because similar effects are known; for example people with phenylketonuria are more likely to become retarded unless they get a special diet in which case there is no impact on IQ.
tl;dr—it’s complicated; people who pretend it isn’t are either stupid or lying, don’t be one of them
Christian’s question is spot on. What he doesn’t say is the reason he’s asking. What you’re describing isn’t a belief, it’s a somewhat vague cluster of beliefs. different beliefs in the cluster can have different credence levels, and treating them as a unit means it’s unanswerable how accurate you are.
Decompose your question to specific falsifiable statements. You should believe whatever lets you most accurately predict the future conditional on your choices.
So; what choices are you facing where beliefs on this topic pay rent? Or, if you prefer, what predictions are you testing with the belief?
[Excuse me. Not native english speaker]
First of all, Lesswrong is the site where I always put my lasts hope.
I’m having lots of troubles trying to find the truth about IQ test taken by different ethnic groups It seems there are lots of studies claiming differences in IQs. On the other side there is a lot of people saying the contrary (culturally biased tests,..)
What do we do as rationalists? I’m really confused. I have read tons of articles from both sides yet nothing is clear to me.
Speaking for myself, my position is “I don’t know”.
Ignoring the specific question, there are many situations in my life where (a) I am curious about something, (b) I don’t trust the existing research, and (c) it is not high enough priority for me to try doing the research myself. In such case, thinking “I don’t know” seems like a reasonable reaction. What else should I think?
In absence of solid research, people often return to armchair reasoning, inventing clever arguments why in absence of evidence we should stick with “default” opinion X, and put the whole burden of proof on people who say Y. Problem is, in the next room, people use similar armchair reasoning to argue that we should stick with the “default” opinion Y, and put the whole burden of proof on people who say X. I could easily provide “a priori” arguments for either position here, which is why I consider neither of them convincing.
Here are your options:
decide that you feel better about believing that the different ethnic groups have the same average IQs;
decide that you feel better about believing that the different ethnic groups have different average IQs;
do the research, which can take you a few years and probably more money than you have; or
accept that you do not know the answer, and learn to live with that feeling.
More importantly, the question is what purpose do you actually need the answer for.
Are you trying to find intelligent people? Then try to measure or estimate their IQs as individuals.
Are you deciding which country to live in? There are probably many other factors to consider, so use those.
Are you trying to win an internet debate? Consider better ways to spend your time.
Are you researching a transhumanist technology for creating superhumans? You will have to research the specific alleles and their interactions; knowing about averages of large populations is not going to help much.
EDIT:
I think there may be one situation where knowing the answer actually would be practical: If you want to start a project in a foreign country, and the success of the project depends on the country having enough high-IQ people. For example, you might be a philanthropist billionaire trying to build a university (or, more meaningfully, the whole educational structure, starting perhaps from pre-school education, and ending with a university) in the middle of a country that according to some sources is full of people stupid for genetical reasons (i.e. not just bad nutrition, etc.), so your project of achieving university-level education for local people may be doomed to fail.
I guess you’d just have to take the risk, and possibly find the answer to the question as a side effect.
One more thing to consider is that IQ is caused partially genetically, and partially non-genetically, e.g. diseases or lack of nutrition decrease IQ. So if you e.g. examine people from a sick and starving population, of course they are likely to have below-average IQ. But that doesn’t say anything about what IQ their descendants will have if the food and health problem gets fixed.
Intelligence is a polygenic trait, i.e. a trait influenced by multiple genes. There is an observed regression to the mean, that is although smart parents are likely to have smart children, and dumb parents are likely to have dumb children, the children of either are usually closer to the average than their parents.
In other words, if you would inhabit an island exclusively by Mensa members, the children born on this island would probably almost all have above-average intelligence; but many of them would not reach the Mensa level. Or the opposite experiment… well, this one was actually done in real life… 40-50 years ago when communists ruled Cambodia, they killed almost all literate people in the country in the attempt to create an agrarian utopia (spoiler: didn’t work as advertised), but Cambodia didn’t literally become a nation of retards.
It is difficult to find exactly which genes contribute to IQ. There are more than 50 suspects, but the experiments suffer from low sample sizes, so many of them are probably false positives.
It seems that first-born children have higher IQ than their siblings. (The official story seems to be that it’s because they get more parental attention and resources. To me it seems more likely that children born later simply receive higher mutational load from older parents. But maybe it’s both.)
It was suspected that breastfeeding increases IQ. Then it turned out this correlation was caused indirectly by mother’s IQ; i.e. smart mothers are more likely to breastfeed their children, and smart mothers are likely to have smart children, which creates a correlation between smart children and breastfeeding even if breastfeeding has no real effect on IQ. Later, there was an experiment showing that actually children with some genes may benefit from breastfeeding while children with other genes may not… but the experiment is unrealiable because of small sample size. (If you start researching this topic seriously, you are going to hear “small sample size” depressingly often.) If true, this wouldn’t be completely suprising, because similar effects are known; for example people with phenylketonuria are more likely to become retarded unless they get a special diet in which case there is no impact on IQ.
tl;dr—it’s complicated; people who pretend it isn’t are either stupid or lying, don’t be one of them
“What do we do” depends largely on the action that you are thinking about. What kind of decision do you want to make that’s effected by the knowledge?
Sorry, I didn’t expressed correctly. What I’m asking is “what should I believe”?
Christian’s question is spot on. What he doesn’t say is the reason he’s asking. What you’re describing isn’t a belief, it’s a somewhat vague cluster of beliefs. different beliefs in the cluster can have different credence levels, and treating them as a unit means it’s unanswerable how accurate you are.
Decompose your question to specific falsifiable statements. You should believe whatever lets you most accurately predict the future conditional on your choices.
So; what choices are you facing where beliefs on this topic pay rent? Or, if you prefer, what predictions are you testing with the belief?
Why should you believe any specific conclusion on this matter rather than remain in doubt?
Iq tests tell something, usually that clusters with intelligence. But there are many ways for that to go wrong.