0
$\begingroup$

What happens when Z score for Skewness is not within the range of -1.96 to 1.96 and Kurtosis is within the range of -1.96 to 1.96

Z-Score for Skewness is 2.58; Kurtosis -1.26;

I should consider this data as not normally distributed right?

I did my self-study : https://www.youtube.com/watch?v=IiedOyglLn0

$\endgroup$
  • 2
    $\begingroup$ It would depend on why you need to know whether it's normal and what the consequences of violation of normality are. $\endgroup$ – Peter Flom - Reinstate Monica Sep 4 '16 at 11:59
  • 1
    $\begingroup$ What was the actual skewness value? $\endgroup$ – Glen_b -Reinstate Monica Sep 4 '16 at 19:43
  • 1
    $\begingroup$ Thank you so much for your help. I really appreciate it. The reason is because I would like to run Pearson Correlation. Skewness Value is 0.497; SE=0.192 ; Kurtosis = -0.481, SE=0.381 $\endgroup$ – MengZhen Lim Sep 5 '16 at 17:53
  • 1
    $\begingroup$ With skewness and kurtosis that close to 0, you'll be fine with the Pearson correlation and the usual inferences from it. See, for example, link.springer.com/article/10.3758%2Fs13428-016-0702-8 $\endgroup$ – Anthony Sep 29 '16 at 15:51
0
$\begingroup$

Consider also whether a transformation might bring the variable closer to normality. But normality is overrated.

$\endgroup$
  • $\begingroup$ I will try or otherwise I will just run Spearman’s test insteade of Pearson's correlation. Thank you so much for your help. I am very sorry, I would like to clarify what do you means by normality is overrated ? The range ? -1.96 to 1.96 ? $\endgroup$ – MengZhen Lim Sep 5 '16 at 17:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.