Does SPSS report only the excess kurtosis i.e., calculated value in excess of 3 or the exact value?
thanks in advance
SPSS, Excel and SAS use a correction factor (-3) to set the kurtosis measure to zero. The mesokurtic curve Is the standard normal distribution used to judge the degree of kurtosis in leptokurtic and platykurtic distibutions. Zero kurtosis does not indicate an absence of the phenomena 'kurtosis' any more than zero on a centigrade or Fahrenheit scale indicate an absence of the phenomena 'temperature'.
Note that the BMDP and Minitab computer programs use a different formula again in their kurtosis measure.
I do agree with Professor Muayyad that SPSS reports the exact value for Kurtosis.
SPSS reports exact value of Skewness and Kurtosis, while it depends on the cut off value that you decide for normality of the data among recommended ranges.
"Kurtosis is a measure of the extent to which observations cluster around the mean. For a normal distribution, the value of the kurtosis statistic is 0."
Retrieved from
https://www.ibm.com/support/knowledgecenter/en/SSWLVY_1.0.1/com.ibm.spss.analyticcatalyst.help/analytic_catalyst/kurtosis.html
For a normal distribution, the value of the kurtosis statistic is 3, not 0 as reported by Ka Chun Lee.
SPSS, Excel and SAS use a correction factor (-3) to set the kurtosis measure to zero. The mesokurtic curve Is the standard normal distribution used to judge the degree of kurtosis in leptokurtic and platykurtic distibutions. Zero kurtosis does not indicate an absence of the phenomena 'kurtosis' any more than zero on a centigrade or Fahrenheit scale indicate an absence of the phenomena 'temperature'.
Note that the BMDP and Minitab computer programs use a different formula again in their kurtosis measure.
I agree with Ian Crawford, because IBM writes at its website: "A measure of the extent to which observations cluster around a central point. For a normal distribution, the value of the kurtosis statistic is zero. "
That means, SPSS calculates the excess kurtosis.
Kurtosis of normal distribution is zero. So indeed SPSS reports excess kurtosis =kurtosis-3. It is uprising that some people in Researchgate report misleading/wrong answers i.e. without checking facts.
According to Kim (2013), " For some practical reasons, most statistical packages such as SPSS provide ‘excess’ kurtosis obtained by subtracting 3 from the kurtosis (proper). "
Source:
Kim, Hae-Young. "Statistical notes for clinical researchers: assessing normal distribution (2) using skewness and kurtosis." Restorative dentistry & endodontics 38.1 (2013): 52-54. Article Statistical notes for clinical researchers: Assessing normal...
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS