You know where you are comparing the Expected with the observed interval readings to see whether or not there is a skewness in standard deviations.. Where are the expected values derived from? Are there something to do with distribution of standard curve? E.g. Interval: (-3,-2) (-2,-1) (0,-1) (0,1) (1,2) (2,3) Expected: 0.2 1.4 3.4 3.4 1.4 0.2 Observed: 0 0 6 2 2 0 There is an obvious symmetry to the expected values but from where are they derived?
In a normal distribution, you expect roughly 68% values to lie between (mu - sigma) and (mu +sigma). In the standard normal case, mu =0, sigma =1. Therefore you should have 68% values in (-1,1). Due to symmetry, 34% should lie within (-1,0) and another 34% in (0,1). In your case you would have in all 10 values. So 3.4 each in (-1,0) and (0,1). Then look at (mu-2 sigma) to (mu +2 sigma). Here you should have 96% of all the values. Hence the next 14% from (-2,-1) and (1,2) and so on...
