Reply to Did your 99th percentile kid stay tall?

Forum Index » General Parenting Discussion

Reply to "Did your 99th percentile kid stay tall? "

Subject:

Emoticons





More smilies

Text Color: Font: Close Marks

[quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous]The few 99th percentile kids I know from childhood did stay tall- girls around 5'11" and boys 6'4" or taller.  All had tall moms and tall dads.

I have one short kid (DD16 and 5'1") and one tall kid (DS14 and 6'1"); both have followed [b]their percentile projection.[/b]  FWIW, I'm 5'3" and DH is 6'1". [/quote]

Not sure if people cannot understand or are choosing to ignore - percentile projection does not exist. The curves are not a projection. [/quote]

I wrote this and no, I clearly don't understand, so please explain it in clear English.  I am trying to learn. What exactly are the curves?

However, I'll restate it- DS was born at the 75th percentile and at some point, rose to sit between the 90th and 95th percentiles.  His height has remained at that percentile so it doesn't really matter whether it's a projection or not a projection to me.  [/quote]

Thanks for the question - if we wanted to project the growth of a 10 year old at the 50th percentile for height, the best way to do that would be to look at all 10 year olds who are at the 50th percentile, and then follow their height over time until they reached age 20. Ideally, you would also get a confidence interval as well. But that is not how the curves are created.

The CDC took a set number of kids and measured their height and weight and head circumference from birth to age 20 at specific time points. Each kid was measured at the same time point in their lives - 14 days, 1 year, 18 months, etc. The points were then put into a table. Using the range and the number of entries, the CDC determined the 5th, 10th, … , 95th percentile value at each time point.

They then graphed all the 5th percentile values and used either geometric or algebraic means to fit a curve to the data. One individual kid could be represented by data points of 80th percentile at age 2, 90th percentile at age 6, and 75th percentile at age 16. That kid has data points in several curves.

The curve is made up of data points that belong to anyone that happened to be at the nth percentile value at some point in their childhood. This is why the curve isn’t predictive. We don’t know how many kids stay on the curve that they started on, because the cdc didn’t calculate that.

If the CDC had wanted to create curves to project growth, they would have taken the kids at age 2 at the 5th percentile and graphed their weight and height over time until they reached age 20. And they would have repeated this for every age division and every percentile. We would have had a mishmash of snaking lines. We have no idea how many kids start the curve at the 50th percentile and finish there, not because it’s not possible to figure it out, but because it’s not provided. My guess is that the margin of error is so great and the confidence interval so wide, that it’s unhelpful. [/quote]

Options

	Disable HTML in this message
	Disable BB Code in this message
	Disable smilies in this message

Review message

Search Recent Topics Hottest Topics