Deflategate - Standard Deviation

[deleted-332635]

Hello -

My daughter's science project looks at whether football pressure influences accuracy of throwing. She had each of 10 subjects throw a "normal" or "deflated" football 10 times each (randomly handed to the subject) toward a designated target. She measured the distance between the target and where the football landed.

For each subject she now has an average distance from the target for the "normal" football and an average distance for the "deflated" football. FYI - for half of the subjects the distance for the normal football was > for the deflated and the opposite was true for the other half.

Her question to me was whether there was a good way to look at whether variability decreased with a deflated football. Example Data:
Normal Football: 1,1,2,4,6,6,6,8,10 (Average 5.2)
Deflated Football: 5,5,5,5,6,6,6,7,7,7 (Average 5.9)

In the example above, the average distance between football/target was lower for the "normal" football than for the "deflated" football but the throws with the deflated football were more consistent (for lack of a better term). Is SD the best way to look at this? Would you compare SD of the Normal vs. Deflated football for each subject (see ATTACHED)?

Appreciate any help with this one!

[williamcolocho]

Great project!

Yes standard deviation will give you a measurement on the variability of the data.
A better parameter is standard deviation divided by the mean: (Coefficient of variation)
http://www.usablestats.com/lessons/sdcv

I would suggest a plot with all the data (scatter plot) markers of different shape or color for inflated vs deflated data.


o
o o x
x x
o x x x

X axis can subject # (1 to 10)

Would love to see the results.

[deleted-332635]

Great idea! Thank you for the information and will post data once compiled

[deleted-345986]

On a scale from one to ten how hard would you say this project was.