MAD and Standard Deviation

[mfbarbmff]

Hey there is a question in the mock that compares MAD and Standard Deviation. Is MAD always smaller or equal to the standard deviation?

Is MAD always divided by the sample size? Is the sample Standard Deviation always divided by n-1? Even for small samples sizes like 5?

Thanks!

 

[atomicrooster]

I believe it can only be used to calculate the population so you divide by the # in the population. Also, I think the MAD is always less then the Standard Deviation because in-order to calculate the Standard Deviation you take the sum of the squared errors and divide by n or n-1 to find the variance and then take the square root. Where as MAD is literally just the average of the sum of the errors.

Anyone else?

 

[Spanishesk]

MAD is always less than or equal to STD Dev....
from wikipedia:http://en.wikipedia.org/wiki/Absolute_deviation. Scroll down a bit.

As far as standard deviation, you divide by N when taking the population standard dev, and divide by n-1 when taking the std dev of the sample