equal environments assumption

The equal environments assumption in behavioral genetics is the assumption that environments for identical twins are not more similar than the environments of fraternal twins. One might say the assumption is violated more than many behavioral geneticists would like to admit and less than many sociologists would like to think. The point of this post, however, is just to share a video of an example of unmistakable violation:

Author: jeremy

I am the Ethel and John Lindgren Professor of Sociology and a Faculty Fellow in the Institute for Policy Research at Northwestern University.

8 thoughts on “equal environments assumption”

  1. I find that assumption ridiculous. If you’re the same height you experience everything and interact with people from the same height. Identical twins are the same height more often than fraternal twins. Assumption false. What am I missing?


  2. The problem is that usually the answer for why identical twins are more alike than fraternal twins in terms of something like, for example, height, is that they are genetically more similar. So in that respect consequences of genetic differences in height are indirect genetic effects.

    If one adopts that reasoning, however, it’s important to point out that what classical behavioral genetics models like to call variance explained by “shared environment” is actually variance explained by shared exogenous environment, that is, environmental characteristics that are not genetically influenced. The heritability estimate includes any indirect effects of genes of the sort suggested by your height example.

    What makes the video trippy is that the women talk about making all sorts of choices about their environments together specifically to be more similar.


    1. On the issue of interpreting proximate environmental causes that are predictable consequences of ultimate genetic causes as part of heritability (eg, if genes lead to height but socially height leads to status, therefore we would observe heritability of status), I’ll just remind this house that Jeremy had an interesting discussion of this on the Contexts podcast and I provided my own gloss.


  3. I think that it’s worth a comment to note that the equal environments assumption is restricted to the traits under study. Identical twins might dress or be dressed more similarly than fraternal twins dress or are dressed — or have more similar heights — but that does not pose a problem for studies based on the equal environments assumption if researchers test for something unrelated to dress habits or height, such as heritability of political views, unless there is some reason to expect views on, say, capital punishment to be biased in a specific direction due to wearing more similar clothes or interacting with people of similar height or some other known or suspected environmental factor that varies between identical and fraternal twins.


    1. Things like height and attractiveness affect how people are treated – by parents, friends, teachers, employers. Mall things that affect the outcomes we’re often interested in (intelligence, self-efficacy, etc.).

      The twins decide to do lots of things that will affect how they are treated and how successful they are, increasing the identical twin correlations. But some of that would happen even if they were separated.


      1. Hi Philip,

        I wouldn’t disagree with anything that you said. But I’d say that it’s better to go further than “affect” and consider the direction and size of bias caused by suspected or known violations of the equal environments assumption.

        It’s probably safe to assume that violations of the equal environments assumption tend to bias heritability estimates upwards, but I’m not sure that there’s a similar rule of thumb for assessing how much to discount heritability estimates from twins studies because of this presumed bias.


      2. One implication of the appearance mediation argument is that unrelated look-alikes should have personality associations about as strong as those of identical twins. In fact this is not true, personality traits have a much lower correlation for unrelated look-alikes than they do for identical twins. By implication, the personality correlations we see in identical twins are largely through direct genetic pathways, and not through socialization as a reaction to phenotype. Caveat, the sample size is pretty small, but even if it’s not ironclad, it nonetheless should make us a little less smug about saying “obviously genes affect appearance and appearance affects socialization and socialization affects personality so therefore these heritability studies that imply direct effects of genetics are nonsense.”

        Liked by 1 person

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.