How Fat are High School Wrestlers?

[Jtm]

A look at the data for 2016-17.


https://www.levelchanger.com/blog/2017/7/30/how-fat-are-high-school-wrestlers

[Bob H.]

Seems mighty high to me.  Maybe the numbers are different on JV vs varsity.

[a guest]

If an entire population is getting larger, I would assume that a subsection of that population is getting larger too.

[AMorris]

I've got to say, as an AP Stats teacher, that a classic example of misleading graphs is to not scale your vertical axis at zero.

You can make any difference look dramatic if you start at some arbitrary value.

A look at the data for 2016-17.


https://www.levelchanger.com/blog/2017/7/30/how-fat-are-high-school-wrestlers

[Jtm]

The axis are set to show the data min/max values, nothing arbitrary about that.  Misleading in what way?

[AMorris]

I didn't mean to imply that it was intentional, sorry. Thanks for providing the data, and your product is great.

Graphs really need to start at zero, otherwise any differences look significant. This is a classic issue that gets discussed in almost every intro stats class. Here's an example of Quaker Oats showing what looks like a dramatic reduction in cholesterol when people eat oatmeal, compared with the real difference when the vertical axis starts at zero. Quite a difference. They were forced to make an adjustment.

http://ksrowell.com/blog-visualizing-data/2012/01/03/the-oatmeal-graph/

The axis are set to show the data min/max values, nothing arbitrary about that.  Misleading in what way?