A Genetic Basis For Political Ideology

My own intuition - for which I have no proof - is that genetics does likely play a factor in political ideology. A NY Times article, which has gotten a lot of play in the blogosphere, claims to report on a study that shows exactly that.

On school prayer, for example, the identical twins' opinions correlated at a rate of 0.66, a measure of how often they agreed. The correlation rate for fraternal twins was 0.46. This translated into a 41 percent contribution from inheritance.

As found in previous studies, attitudes about issues like school prayer, property taxes and the draft were among the most influenced by inheritance, the researchers found. Others like modern art and divorce were less so. And in the twins' overall score, derived from 28 questions, genes accounted for 53 percent of the differences.

However, the methodology is described as,

Calculating how often identical twins agree on an issue and subtracting the rate at which fraternal twins agree on the same item provides a rough measure of genes' influence on that attitude. A shared family environment for twins reared together is assumed.

Twin studies are some of the most powerful studies that have been performed to analyze how large a role genetics plays in such outcomes as alcoholism, homosexuality, and schizophrenia. However, these studies usually control for environment by studying identical twins reared in different homes. As both Julian Sanchez and Brian Hawkins note, based on the article, the identical twins studied grew up in the same household. If that is the case, i.e., if the NY Times description of the study is correct, then the conclusions are far weaker than claimed. It merely shows that genetic similarity leads to similar reactions to environment, not that genetics is a determinant independent of environment.

Share this

I just thought of a

I just thought of a confounding factor for twin studies that I haven’t heard discussed before: the effect of genetics on environment. If people treat others differently based on factors determined primarily by genetics, such as appearance, doesn’t that mean that identical twins, whether raised together or separated at birth, will tend to grow up in environments more similar than they would if they were fraternal twins?

I think you're right. It comes down to the difficulty of having a 100% true control. It may be that all we can get is gradations of controls.

I would imagine there are

I would imagine there are many people that are not related to me that I atleast agree with 40% of the time. How do they account for that in the study?

A correlation isn't the chance that two people happen to agree on something. Suppose that 60% of the population favors the death penalty. Knowing nothing else about someone, you can say with 60% certainty that he supports the death penalty. That's the baseline.

Now suppose you want to find the correlation for support of the death penalty for identical twins. Maybe we find that if one twin supports the death penalty, there's an 80% chance that the other will (conversely, if one opposes it, there's a 70% chance that the other will).

Since knowing the position of someone's twin on the death penalty gives us a better chance of guessing his position, that means that the correlation is positive. A correlation of 1.0 means that twins always agree on this issue, but a correlation of zero doesn't mean that they always disagree (that's -1.0). It just means that knowing one twin's position doesn't increase your chances of guessing the other twin's position above the baseline.

I'm going to go out on a limb here and say that the correlation might be .5 (since you cut your chances of guessing wrong in half). But I may very well be wrong on that, so you're still advised to study before your next statistics test.

I would imagine there are

I would imagine there are many people that are not related to me that I atleast agree with 40% of the time. How do they account for that in the study?

I just thought of a

I just thought of a confounding factor for twin studies that I haven't heard discussed before: the effect of genetics on environment. If people treat others differently based on factors determined primarily by genetics, such as appearance, doesn't that mean that identical twins, whether raised together or separated at birth, will tend to grow up in environments more similar than they would if they were fraternal twins?

And it's not just interactions with other people that are could be affected. For example, if genetic differences between fraternal twins make one more sensitive to heat, then he might spend less time outside than his brother, which could have an impact on his personality, especially if the lack of exposure to sunlight leads to vitamin D deficiency and consequently to depression.

I can't pretend to know what the magnitude of this effect might be, but I would be very surprised if it were entirely insignificant.