Interpreting results

[scojo95]

I know this may seem like a basic problem but this is my first class and I have yet to totally understand SD, and Skew. I understand what each of them are but don't know how to translate my data.

I've been given N, Mean, SD, and Skew
For example:
N=1783
Mean = 26.345
SD = 7.89
Skew = -.789

N is the number of respondents, mean is the average on a scale of 0-45 with 45 being high.

Thanks

[Mean Joe]

SD is the standard deviation. It measures how "spread out" your data is. SD is always >= 0. It can only =0 if all the data values are the same (so it is rarely 0). You can find a 95% confidence interval for the true average of your data using the SD, e.g. Mean +/- (SD/N). You can also calculate a coefficient of variation = SD/Mean. I'd say the SD is small (relative to Mean), so your data is not spread all over the place.

You should see very few observations outside the range of Mean +/- 3*SD.

Skewness is a measure of the asymmetry (I'm getting this from Wikipedia.org). You'll find that having normal (symmetric) data is amenable to doing many statistical analyses. So skew=0 would mean you have symmetric data. Since the skewness is negative, that means the left tail of your data's distribution is longer than the normal (so you can guess that there's more data below the average than would be expected).

I don't know of anything you can do with the skewness, like you can with the SD.

Hope this helps.

[abi khan]

I am getting a skewness of 0.202, kurtosis 1.568872 and Jarque bera 2.765432 with probability 0.250896......... what is the interpretation of these results,,,,,,,, i need simple explanation of the concepts to understand the basic idea of these terms ,,,,,,,,,,, immediate response is appreciated