The normal distribution is said to be mesokurtic with a kurtosis of 3. That is the standard. A distribution with a kurtosis of more than 3 is said to be leptokurtic and one that has a kurtosis of less than 3 is said to be platykurtic.

Dear Kavitha,

Following on from Ette's answer, there are two definitions of kurtosis. In one of them the value of normally distributed data is 3 but in the commonly used definition it is zero. Rather than looking for limits in kurtosis it is better to divide it by its standard error and compare its absolute value with 2. The standard error is provided in SPSS using the explore function.

Thanks a lot for the advice.

Kurtosis for normal distribution is 3, but it's very common to use the excess of kurtosis that is Kurtosis minus 3. In this case, the normal distribution have an excess of 0 and coincides with skewness.

Dear Antonio,

May I check the kurtosis for normal distribution 3 or +/- 3?

Kurtosis is not negative, but "excess of kurtosis" could be negative. For normal distribution Kurtosis is 3 and "excess of kurtosis" is zero.

Noted. Thanks a lot for your advice.

In a normal distribution, the kurtosis is equal to three (or zero in some models).