Necessarily, they cannot study the things that make us, us and them, them. "We" do not share a genome, and "we" are not raised in the same house by the same parents; and ditto for "they" who do not share an entire genome or a common set of parents, which would contrast with our own. Now we're getting into sociology, anthropology, and so on.
There's nothing wrong with the behavior genetics approach, but it misses all of those group-defined traits that give us our cultural identity and our sense of belonging to something larger than ourselves. The things that give greater meaning to our lives, beyond noticing how smart, attractive, or extraverted we may be. So, we ought to supplement behavior genetics with group-ier social sciences to get the full picture of how we develop and turn out the way we do.
One of the major weaknesses of using behavior genetics to study the effects of divorce on children is that the main thing that marriage and family are about is group-level stuff. It's interesting to study how marital strife does and does not affect a child's individual characteristics -- mood, behavior problems, drug use, and so on -- but that's not where we should look for the effects of divorce.
Fundamentally, divorce represents a permanent rupture in the social connectedness among nuclear family members, not only between husband and wife but critically between the children and the absent parent ("non-custodial," "non-residential"). Your parents are not getting back together. And if you were raised by a single parent, without the other having been there for long, your parent is not ever going to bring the absent parent into the home to give you both parents.
Divorce typically results in a change of residence, since one parent (the custodial parent) will not be able to afford payments and maintenance of a house that had been a joint investment of two parents, while also raising the kids. Thus divorce also robs children of the "sense of place" of the home that they had attached themselves to so far during development. Houses, yards, sidewalks, blocks, and neighborhoods are not interchangeable, any more than parenting adults are to the kid. A stepmother is not the child's true mother, and the new place of residence will feel more alien -- more "not meant for us" -- than the marital home.
Moving one level up, the change in location will usually sever the children's social and cultural ties to the group of kids their age, whether at school or in the neighborhood broadly. They have to start all over again with new kids, and again -- kids are not interchangeable. Children of divorce have permanently lost their old friends who they'd grown close to, and now must do the best they can with a different set of kids -- and often not as able to get as close to them as they are able to get to each other, since they have grown up together, while the child of divorce is late to the party.
Finally, not only have they lost the connection to their home, but of their community environment in general. The neighbors' houses, the parks, the libraries, the malls, and the churches will all be different. So will the secret hiding places, the writings carved into the sidewalk, and all of those places where some meaningful experience took place. All those places, anchoring all those experiences, give you a sense of belonging to a particular, special world -- and you must now leave all of them behind for the rest of your formative years.
Whether children are resilient is not the question. Of course nature has programmed us to adapt to changing circumstances -- to roll with the punches. But again, resilience, adjustment, etc., are individual traits: however well the child of divorce adjusts to his new set of parents, new home, new social circle, and new community, he has lost much of his sense of belonging to a particular nested set of groups, and he won't ever get those back. (And to make things worse, resilience is only a tendency back toward normality, not a full recovery.)
We can ignore the predictable glib rejoinders about how any set of parents and peers will do, how a house is nothing more than a memory-free building on a featureless lot of dirt, and how all the streets and parks and malls in this country are all so similar that the kids will never be able to tell the difference. Of course they're not -- particular pieces of social and cultural life are not fungible stuff like dollars and cents.
Divorce is not the only way that children may be uprooted and transplanted willy-nilly into an alien place, but it sure is one of the more common and most reliable ways. And no matter how much the autistics may wish to insist on the fungibility of home, peer group, and community, none will say so for the parents. The children used to have both parents, now they only have one; if a new one moves in, everybody is always aware of their alien status within the family, and behave accordingly. The vastly higher rates of child abuse coming from step-parents rather than parents provides a dark example of what happens when we play down how particular the social-cultural arrangement should be once the ball gets rolling.
Little disruptions here and there are inevitable, and our minds are designed to handle that in the way that we can handle taking a hard fall or getting slugged in the gut. But ripping the children right out from their existing network of support and belonging is far too severe for them to bounce right back from. And treating the world as though it worked that simply is the height of arrogance and heartlessness on the part of the adults.