October
2004

Vol.
6, No. 10 


Quality
Quiz from Professor Cleary Congratulations!


Click here for a more complete video explanation of kurtosis. Kurtosis is a measure of the combined weight of the tails (on a distribution or histogram) in relation to the rest of the distribution. As the tails of a distribution become heavier, the kurtosis value will increase. As tails become lighter, this value decreases. A histogram with normal distribution has a kurtosis of 0. If the distribution is peaked (tall and skinny), it will have a kurtosis of greater than 0 and is said to be leptokurtic. If the distribution is flat, it will have a kurtosis value of less than zero (as Marge’s distribution does), and is known as a platykurtic distribution. Note: While some references to kurtosis do not subtract 3, both SQCpack and CHARTrunner do, in order to make the statistic 0 for a normal distribution. The figure below shows a thin and tall distribution with a positive (and relatively large) kurtosis value. The following represents the short, spreadout distribution with a negative kurtosis that Marge showed to her boss. It can be seen that the skewness is negative. This is a normal distribution, with a kurtosis of 0.0. The online video will address the chisquared value of 1.40. The next time you have an urge to dismiss someone who’s arguing with you, accuse them of holding onto a platykurtic point of view. 