In my opinion I think that twins studies do provide an accurate measure on determining the extent to which our personalities are affected by genes; most personality traits are in the range of 35%-60% heritability. However that doesn't mean that 'shared enviroment' (for lack of a better term) doesn't influence personality at all. 'Shared enviroment' has a small, but siginificant impact on personality though for the majority of traits it never accounts for more than 20% in explaining personality traits. In his book, Human Diversity, Charles Murray listed a few traits and how much of the variance in those traits are explained by Shared enviroment, Non-shared enviroment and Genes. A significant number of traits either have 0% or less than 15% of the variance explained by shared enviroment; the ones listed that had shared enviroment account for more than 15% are as follows:
- Emotionally unstable Personality - 19% Shared Enviroment, 46% Non-shared Enviroment and 35% Genes
- Higher-level cognitive functions - 18% Shared Enviroment, 27% Non-shared Enviroment and 55% Genes.
- Language Functions - 22% Shared Enviroment, 32% Non-shared Enviroment and 46% Genes.
- Mild mental retardation - 22% Shared Enviroment, 45% Non-shared Enviroment and 33% Genes.
- Disorders due to tobacco use - 17% Shared Enviroment, 29% Non-shared Enviroment and 54% Genes.
- Recreation and leisure - 18% Shared Enviroment, 27% Non-shared Enviroment and 55% Genes.
- Disorders due to Alchohol use - 19% Shared Enviroment, 38% Non-shared Enviroment and 44% Genes.
- Religion and Spirituality - 21% Shared Enviroment, 43% Non-shared Enviroment and 36% Genes.
- Disorders due to Cannabiniod use - 22% Shared Enviroment, 25% Non-shared Enviroment and 54%.
- Educational Attainment - 25% Shared Enviroment, 26% Non-shared Enviroment and 50% Genes.
- Disorders due to multiple drug use - 26% Shared Enviroment, 29% Non-shared Enviroment and 46% Genes.
- Problems related to upbringing - 34% Shared Enviroment, 40% Non-shared Enviroment 27% Genes.
- Basic interpersonal interactions - 36% Shared Enviroment, 34% Non-shared Enviroment and 30% Genes.
So as we can see shared enviroment does matter, but it never matters more than non-shared enviroment (except in the case of basic human decency where it accounts for 36% of the variance). So the Fruedian idea that all of your adult problems can be traced back to how horribly your parents treated you is likely wrong (your young peers/friend/bullies are much more likely to have done things to you that still impact you in your adult life). Though this doesn't mean that parents don't matter; they do, they just aren't the main determinant in personality outcomes for the large majority of traits.
My second contentsion with this line of thinking would be, well, what the hell does shared enviroment even mean? Well this is usually defined as the effect of parenting on children and home enviroment. My problem with this is that home enviroment can be different even amongst sibling because parents do not treat each of their children equally, so the enviroment isn't exactly shared. Children that grow up in large families will often tell you how unfair and unequal parents can be in their treatment, parents can (and do) treat ugly children worse than attractive children which might cause some effect on personality. Most parents agree that all children are different and that you would definitely be mistaken if you were to treat each of your children exactly the same; instead you might wish to treat them in a way such that their positive traits are enhanced. The home enviroment is complex but it is not equal nor is it equally shared between siblings; I'm not sure exactly how you would take into account this complexity or how you'ld measure it without directly observing behaviour of parents, God only knows then to what extent parenting influences children.
Let's say you have a greater genetic propensity to be an asshole when compared to your siblings, this is likely to irratate your parents and will change the way in which your parents treat you relative to your siblings. In other words genetic propensities influence parenting and will thus cause a positive feedback loop whereby the genetic traits are exaggerated through parenting effects (i.e you're an asshole so your parents treat you like you're an asshole so you become even more of an asshole) so it looks like enviromental effects are smaller than they really are. Now I'm not advocating for a blank slate of human nature and I'm fairly certain a large amount of our behaviour can be accounted for by genes, but I do reject the deterministic (and somewhat nihilistic) view that parents don't matter