# Standard deviation of difference of means

#### Ronguru

##### New Member
Hi,

I am sorry of this stupid question I would like to know if it possible to calculate the standard deviation of a difference of means, having the two primary standard deviation and the two primary sample size and the two primary means.

For example:
I have 24 men that at T0 weigh [mean (a)=77.5 Kg SD(b)=1.4 N=24]; after one year the same 24 men weigh [mean (a)=81.4 kg SD(a)= 3.9 N=24]. Hence on average every man increases his weight of 3.9 Kg; but what is the SD of this difference? Is it possible to calculate it with this data?

Thanks

#### Dragan

##### Super Moderator
Simply compute the difference between each of the 24 subjects weights and subsequently use the usual formula for the standard deviation on these (24) "difference scores."

#### rollnrocking

##### New Member
You need more data... As in, the actual data points. Without that, you cannot compute this.

#### Ronguru

##### New Member
I don't have each 24 subjects weights, I have only the means and the SD of the means :'(

Simply compute the difference between each of the 24 subjects weights and subsequently use the usual formula for the standard deviation on these (24) "difference scores."

#### shahnawaz

##### New Member
You cannot find the standard deviation of difference of means. You need a distribution for it. What you could have computed was the standard deviation of the difference of weights for all the men. But since you do not have data points, you can't.