Problem with Overall Standard Deviation (SD)
By rmendoza123
@rmendoza123 (637)
Philippines
February 1, 2012 11:27pm CST
Guys i got a big problem regarding the computation of overall standard deviation pls help. here's the data
Question Mean Standard Deviation
1. 3.2044 0.50204
2. 2.9416 0.37898
3. 3.9781 0.52113
4. 3.8175 0.77856
5. 2.5912 0.52236
Overall 3.3066 0.26574
The question who did it arrived at overall SD of 0.26574? I really need an explanation or formula how did it arrived to that. thanks
1 response
@owlwings (43910)
• Cambridge, England
2 Feb 12
It seems to me that the answer is wrong for the data as given.
The rule for combining the means of a number of distributions is to take the mean of the means ( = sum of means/number or means).
The rule for combining the standard variations of a number of distributions (where no correlation exists between any pair of data sets) is to take the root mean square (the square root of the sum of the squares of the values). I calculate this to be: 1.2434878858 ( = v(s1²+s2²+s3²+s4²+s5²)
This page may help to explain it better: http://pballew.net/combdis.htm
@owlwings (43910)
• Cambridge, England
2 Feb 12
The "v" before the bracket is supposed to be the square root sign. MyLot's text entry box can't handle some Unicode symbols!
