The statistics of chubby smart babies

A new analysis has found a correlation between birth weight and school achievement… and the upshot (from the NYTimes upshot blog) is that “heavy babies do better in school.” My babies, being very very average in weight (about 7.5 lbs each), are projected to also be very average students.

heavy babies

But here’s the kicker. These are totally correlational data. That means that birth weight may not be destiny. And check out that BIG gray part around the middle lime green line… that’s the variability of this effect! (Specifically it is the interquartile range — 50% of babies fall in this range.) What’s more, the cool infographic at NYTimes allow you to fiddle around and check out this data for mothers of a particular education level, ethnicity, and age. So when I clicked on “college graduate” and “Asian, non-Hispanic” and “30-35,” here’s what the graph looks like.

heavy babies2

A few things to note: First, in this category of babies, even the bottom 25% of average birth weight babies is at 70th percentile in educational attainment… higher than the 57th percentile of all babies born at 10 lbs! Second, note that there is virtually no variation accounted for by birth weight… 6 lb-ers and 10 lb-ers are all super achieving.

So the moral of this story is: if you are a well educated Asian mom who gave birth in her early 30s, don’t worry about low birth weight and just keep doing your tiger mama thing. HA.


One thought on “The statistics of chubby smart babies

  1. Annie says:

    Great job dissecting the data Ji! The graph with Asian babies pretty much makes the case.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: