I realized today that my post about doing quantitative genetic back-of-the-envelopes was rather wordy. And, I have a hunch that those who "got it" already have an intuitive feel for what I'm talking about, so I thought perhaps the easiest way to get people to develop a better feel was to roll up the calculations into a small Javascript calculator. You can find it below the fold. The terms are defined in the earlier post, but I plugged in the default values (you can change them if you want). Obviously you enter in inches or centimeters depending on whether you're normal or Canadian. Finally, if you are from another dimension where heritabilities are outside of the 0 to 1 interval and heights are negative then you will encounter problems. Remember the garbage-in-garbage-out principle; I did this in a hurry so no catching your errors and throwing back intelligible warnings.
- Log in to post comments
Very Cool - Thanks. However, this pretty well ends the dream that any of my kids will be stars in the NBA.
...well, NBA quality kids are many standard deviations from the norm on many traits. so yeah, this is a prior that can crank down your expectations a bit ;-)
Most Canadians still use US units for personal height and weight. I don't know why, it's one of those things that didn't make the conversion. (At least, anybody in their twenties or older.)
I'm 79 inches in height. So, if I were to have a child with a woman who was 5'10'', the expected height for our hypothetical child would be 6'4'' if it were a son and 5'10'' for a daughter. So maybe they could have a decent shot of playing ball professionally. But hopefully they wouldn't inherit my lack of jumping skills.
dude, you're in 3 sig range!
This is a very useful application. Substitute any metric you like for height, use reasonable parameters for average and SD on that metric, and calculate what your kids, or somebody else's, will be.
Given the post on GNXP classic on IQ and UK universities, this can serve as an IQ predictor (e.g. make avg for both parents 100, SD=15, heritability 0.5). For two parents at 145 (+3 SD above population average) the expect IQ is 122.5 = +1.5 SD >population average. Only about 2.25% of children will be above 137.5 (+2.5 SD above pop average).
There must have been some height enhancers from my mother gene pool.
Either using the default means at the calculator or trying to locate the 1950 stats, I should not be 178 cm female, having father of 174 cm and mother of 168 cm. Interestingly, my half-brothers from the first marriage of my father are smaller than I am, and all under the height of my father (regression to mean?). The deceased first wife of my father was (my guesstimate according to photos) around 160 cm.
And it goes further - my husband (174 cm) and me (178 cm) have a son who is 192 cm. There was an acceleration in the mean height of Czech men from 174 cm in 1955 to 181 in 2001, but I think that the 192 cm is a bit too much (data in Czech, here: http://local.fsport.uniba.sk/a_veda_vyskum/CASOPIS/r2006/c3/tvs3-06.pdf Graf 4, red curve).
EW, using the calculations above there's a ~2-3% chance that you would be the height you are. that being said, that's an underestimate since the guassian approximate gets less accurate as you deviate from the mean.
There would be an additional factor if the expecting mother is (now) poor and from the tropics? If she gets to live in western Europe, wouldn't be her kids become taller anyway?
(I'm a white male 1.98, she's a 1.55 female and yes, I met her during field work... now were expecting a baby!).
There would be an additional factor if the expecting mother is (now) poor and from the tropics? If she gets to live in western Europe, wouldn't be her kids become taller anyway?
yes.
This is a very rough aproximation; since heritability is a property of a population....not of a pair if individuals.
Of course you could figure out the parents breeding values....but that would be a pain.. ;]
This is a very rough aproximation; since heritability is a property of a population....not of a pair if individuals.
well, that's why i added into the dispersion factor into the results. nevertheless, it is a good point which needs to be reiterated.
I put in my parents' heights (5'7" and 4"11") and it was accurate for my brother (a shade under 5'7") but wrong for me (said I would be a bit under 5'1" but I shot up all the way to 5'2-1/2").
My own kids are predicted to be around 5'1" (they are girls.). My pediatrician thinks that two of them will probably be around 4'11", based on their very petite bone structure. I'll check back in 15 years or so and see who's right.
It's accurate for me (a bit under 5'9", parents ca 5'7" and 5'4"). I do wonder about my son though - it predicts 5'10", but at 2 years old he's in the top 15% for height of his age group and I'd guess he looks set to be 6' or so.