Normal distribution on SPSS

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Normal distribution on SPSS
Message posted by stephanie on January 18, 2001 at 12:00 AM (ET)

An SPSS book tells me that I can check to see if data are normally distributed by dividing skewness by the standard error of skewness. If the result lies between -2 and +2 it can be said to be normally distributed. In practice, however, the figure given tells me the data is normally distributed, but a histogram doesn't appear so. Which do I believe?? (I want a method that doesn't rely on looking at a picture and making a subjective decision of normality)

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Normal distribution on SPSS
Message posted by Phil on January 21, 2001 at 12:00 AM (ET)

If you don't want to look at a histogram or "normal probability plot" (another visual method), then I suggest using the Chi-Square Goodness-of-Fit test. I am not sure if SPSS has that pre-programmed for the normal, or if you would have to use brute force.

I know many quality control software packages have it built in.

Good luck.

Re: Normal distribution on SPSS
Message posted by martin on January 22, 2001 at 12:00 AM (ET)

The "1-sample-Kolmogrof-Smirnov-Test" is the correct test. You'll find it at "nonparametric tests"

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.