[snip interesting analysis]
CONCLUSION
Even considering only positive and negative correlation, we have 10 minor inverse effects versus only 6 minor positive effects, and 4 major inverse effects versus only 2 major positive effects.
A sample size of seven is obviously not sufficient for sophisticated analysis. However, finding NO pattern of correlation between parents and offspring indicates that no genetic inheritance of personality exists.
I've dabbled in this analysis (not as precisely as you have, I've not had many native children, and wasn't prepared to include the unknown forebear factor in immigrant Single Parent Famlies) and tentitively come to the conclusion that there's no correlation due to an apparent negativity to the correlation.
Although, with my smaller dataset I really couldn't prove the negative correlation either (nor could I have technically proven any observed positive one). It could have been some Regression Towards The Mean effect, statistical fluke or similar... The conceptual error bars were too large.
But you've gotten me thinking, with your results showing a tendency towards an inverse pattern as well:
a) Rebelliousness of youth? I know these are personality traits rather than preferences[1], but the
Ab Fab effect might apply.
b) Recessive and generation-skipping variations, possible including epigenetic factors[2]. And for that, second-generation births probably need to be studied.
In a world with a known Designer, of course, we might be able to say that the possibility of these things are a little unlikely, given the complexity needed to encode such heritability. Our own world probably only exhibits generational skips such as this as an emergent property. However, one could argue that it's an Ineffable Creator running our particular show, and while the Creator of the Dwarven world is perhaps a bit more communicative with us than one that might exist of our own world (and his base code can
technically be analysed without hitting the usual problems of deriving any Theory Of Everything, such as Heisenburg limits, observation horizons and the inability to divine the precise nature of the substrate of the universe and divorce this from the data laid upon it) I don't think I'd put it past a certain Toady-type person to have deliberately and with malice aforethought inserted such complications into the respective simulations...
[1] Which I've similarly failed to find correlations for, although I still have a lot of historical 40D records (and, indeed, save-games) that I could also analyse, I suppose...
[2] In phenotype terms, an example of this would be the oft-reported case that maternal-line grand-children of women who suffered famine are smaller babies than their contemporaries without the same heratige, the reason for which appears to be that it is the grandmother's epigenetic cellular environment that dictates the way the grandmother/father genetic cross mixture develops into the mother's DNA, and thus the grand-child themselves develops from an egg that is a direct product of the expression of genes activated (or otherwise) by the stressed biology of the grandmother. But
a study regarding grand-maternal smoking that I found doesn't appear to have any indication of similar bi-generational transmission, if i read it correctly.
BTW, Maggarg, click clickety-click, eh? [edit: Oh, it looked like you'd multi-posted, but not now.]
And Jude: Yeah, that's a good summary of what I'm still trying to post, except that, with Toady... who knows!