Sibling height

Victor

This thread is moderately interesting, although the hypothesis seems to be shot down. http://www.boards.ie/vbulletin/showthread.php?t=2056901922

But I'm wondering, if you adjust for gender, is there a pattern is sibling adult height from eldest to youngest? One the one hand, as the parents get older, one would expect income and nutrition to improve, but on the other hand, those resources may be shared amongst more children. In very large families or those spread over a long time duration, the eldest children may serve as additional providers for the younger children or free a parent from some child care duties, thereby increasing income potential.

What of only children -v- small families -v- larger families? One may need to adjust for only children/smaller families due to parental casualties.

Thoughts?

cookie1977

Think of hair colour or inheritable diseases. There can be huge variability. My wife and I have dark brown hair and one of my sons has light brown and the other is ginger. Another example is with diseases (Apologies if you've an idea of science). In cystic fibrosis you need two copies of the defective gene to get the disease. You inherit 23 chromosomes from your mother and 23 from your father. If both are carriers of the disease but unaffected then their children have a 1 in 4 chance of getting the disease. On top of that there's chance too as in they just might be lucky to have multiple children who are carriers or all their children could have CF.

What I'm saying is that no one could give a definitive answer on this as it not only depends on the parents genetics but also (as you said) external environmental factors such as nutrition, lots of other genes etc...

Victor

I'm asking are there trends. I realise that there is a huge amount of potential variability - does that variability have a trend between siblings based on birth order.

To clarify, I would qualify siblings by being full siblings, not half-siblings. Also, on a macro level most siblings have similar environments (same nutrition, diseases, societal norms, education, etc.). Although on a micro level things will be different, e.g. they may be exposed to famine or life-threatening disease at the same time, but at different stages of their development.

cookie1977

I'm still not sure if I've got you completely but there's trends in that we're all getting taller:
Secular changes and predictors of adult height for 86 105 male and female members of the Thai Cohort Study born between 1940 and 1990.
http://www.ncbi.nlm.nih.gov/pubmed/20805198

Growing up before growing out: secular trends in height, weight and obesity in 5-6-year-old children born between 1970 and 2006.
http://www.ncbi.nlm.nih.gov/pubmed/23413314

The heights and weights of Irish children from the post-war era to the Celtic tiger.
http://www.ncbi.nlm.nih.gov/pubmed/19208691

If you want any of the papers just let me know.

Victor

If I take all first born children and put them in Cohort 1. Take all second born children and put them in Cohort 2. And so on. As you go on, the cohorts get smaller, but you should be able to establish a mean height (and standard deviation) for each. If your sample is big enough, cohort size shouldn't really matter.

Is there a difference in mean height and standard deviation between the cohorts? I've invented some data to demonstrate one part of the hypothesis.

An alternative to first born, second born, etc. would be to measure years since the first born was born.

Capt'n Midnight

you also have to correct for diet

over time people are getting taller

cookie1977

Victor wrote: »

If I take all first born children and put them in Cohort 1. Take all second born children and put them in Cohort 2. And so on. As you go on, the cohorts get smaller, but you should be able to establish a mean height (and standard deviation) for each. If your sample is big enough, cohort size shouldn't really matter.

Is there a difference in mean height and standard deviation between the cohorts? I've invented some data to demonstrate one part of the hypothesis.

An alternative to first born, second born, etc. would be to measure years since the first born was born.

I can find no analysis or data on this out there at present but I could speculate. If, as we see, that people in general are getting taller with each passing decade then would it not be reasonably safe to assume that each subsequent sibling (on average) would be taller (even minutely so) then their older brother or sister.

Victor

Capt'n Midnight wrote: »

you also have to correct for diet

To a degree that is part of what is being measured. In any case, siblings would have very similar diets, based on what is available to the parents.

cookie1977 wrote: »

I can find no analysis or data on this out there at present but I could speculate.

Thanks.

If, as we see, that people in general are getting taller with each passing decade then would it not be reasonably safe to assume that each subsequent sibling (on average) would be taller (even minutely so) then their older brother or sister.

I appreciate that and I suppose one would need to adjust for that.

However, are such increases purely intergenerational and not strictly time based?

Capt'n Midnight

Victor wrote: »

To a degree that is part of what is being measured. In any case, siblings would have very similar diets, based on what is available to the parents.

the diet would change over time

it's expected that most people would have more disposable income later in life

and there is a lot more variety in shops than when we were young

also look up effects of starvation in Holland in WWII it affected the ovaries of unborn children so there was a measurable effect on the grandchildren of those starved - so that will skew some averages over there - and it's not a genetic thing

darjeeling

There are a few studies looking at the effects of sibling number and birth order on height

Here's one on childhood height in families in Britain: link.
Children with one or more siblings have reduced height at age 10, with greater reductions as numbers of siblings goes up. Children with 4 siblings were over 3cm shorter than only children. Effects were greater for children later in the birth order, so the more older siblings you have the shorter you are in relation both to only children and to first-borns, 2nd-borns, ...

A study of all 18-year old males conscripted into the Swedish army [EDIT] born between [/EDIT] 1965 and 1978 (link) didn't find that children in larger families were shorter on average, but did find a progressive reduction in height with increasing birth order. The effects aren't huge, (within families, 5-th borns were on average 1.2cm shorter than first-borns), but they are significant.

A still larger study, again on Swedish men, apparently finds the same, according to the abstract (link). 2nd-borns were 0.4cm shorter on average than 1st-borns, and 3rd-borns 0.7cm shorter.

leanonme

darjeeling wrote: »

There are a few studies looking at the effects of sibling number and birth order on height

Here's one on childhood height in families in Britain: link.
Children with one or more siblings have reduced height at age 10, with greater reductions as numbers of siblings goes up. Children with 4 siblings were over 3cm shorter than only children. Effects were greater for children later in the birth order, so the more older siblings you have the shorter you are in relation both to only children and to first-borns, 2nd-borns, ...

A study of all 18-year old males conscripted into the Swedish army from 1965 to 1978 (link) didn't find that children in larger families were shorter on average, but did find a progressive reduction in height with increasing birth order. The effects aren't huge, (within families, 5-th borns were on average 1.2cm shorter than first-borns), but they are significant.

A still larger study, again on Swedish men, apparently finds the same, according to the abstract (link). 2nd-borns were 0.4cm shorter on average than 1st-borns, and 3rd-borns 0.7cm shorter.

Thank god I was not an only child so, I would have been even taller. I am the fourth born, and am 6ft two female.

In my family order the oldest is male is 6ft 3, the next is a girl 5ft 10, the next a boy 6ft 1, and then me a girl 6ft 2, so our high is all over the place, and we are all only two years apart, so no increased income as such, probably a decreased income and added stress from the time I was born.

I would say this is the same for most of the families around me there is no colloration between position born and high.

Victor

leanonme wrote: »

I would say this is the same for most of the families around me there is no colloration between position born and high.

There are very many factors at play, so while it may not be obvious within individual families, it may be more obvious across large populations. It is still a modest difference.

cookie1977

As stated in some of those papers, there are so many variable and bias. The UK study excluded almost 10% (parents of lesbians, absent fathers or mothers) of their total births in the Bristol area (which they state is predominantly white). The second swedish study is unable to take into account decisions by people "not" to have more children due to economic effects or decisions by the parents.

None of the studies take into account things like the age of the parents at time of birth from what I can see, which has to play a role nor was sex or ages of the children. I dont believe you can just arbitrarily ignore these variables and still come up with a significant observation.

There's insufficient data on the second swedish study to comment on other then again they've been unable to include the entire female sex

darjeeling

cookie1977 wrote: »

As stated in some of those papers, there are so many variable and bias. The UK study excluded almost 10% (parents of lesbians, absent fathers or mothers) of their total births in the Bristol area (which they state is predominantly white). The second swedish study is unable to take into account decisions by people "not" to have more children due to economic effects or decisions by the parents.

None of the studies take into account things like the age of the parents at time of birth from what I can see, which has to play a role nor was sex or ages of the children. I dont believe you can just arbitrarily ignore these variables and still come up with a significant observation.

There's insufficient data on the second swedish study to comment on other then again they've been unable to include the entire female sex

The studies are showing an effect of height reduction in children or adult males with older siblings. What the exact cause is, and whether the effect would be seen in all circumstances isn't clear. However, when the sample sizes are large enough it seems reasonable to assume that a similar pattern will be seen for similar societies.

The first Swedish one does actually control for parents' age and income too, from my reading.

cookie1977

Yes it does, sorry I should have specified which I was talking about. But surely you'd agree leaving out all women skews the studies a little, no (Swedish ones only)?

darjeeling

cookie1977 wrote: »

Yes it does, sorry I should have specified which I was talking about. But surely you'd agree leaving out all women skews the studies a little, no (Swedish ones only)?

It does indeed mean they can't be used to say anything about the heights of Swedish women, which may or may show the same trend in relation to birth order.

Victor

cookie1977 wrote: »

Yes it does, sorry I should have specified which I was talking about. But surely you'd agree leaving out all women skews the studies a little, no (Swedish ones only)?

They may have been dealing with a pre-prepared data set.

Sweden doesn't conscript women (women can / do volunteer), so obtaining the data would have been more difficult.

Stenth

"Children with many siblings are taller/shorter than average" is not the same thing as "Children are generally taller/shorter than their older siblings (adjusted for age)".

I can think of several factors that might be common to families with many children and which could also affect their height (say, if immigrants from certain regions tend to have larger families and also have different genetics than native families, or if there is a correlation between many children and bad nutrition, to take two random examples. I have no idea whether they are true.)

The advantage of Sweden's conscription data is that for a very long time, all males were tested in the same way at age 18, even the ones who, like me, were deemed unfit for army duty. Lots and lots of wonderful information to test for differences over time.

Victor

Stenth wrote: »

"Children with many siblings are taller/shorter than average" is not the same thing as "Children are generally taller/shorter than their older siblings (adjusted for age)".

Oh come on, my minefield is big enough already!

Yes, a third child in a family of 10 would probably struggle more than a third child in a family of 5.