The degree to which a data set is peaked. Like many other basic statistics, kurtosis can help you establish an initial understanding of your data. You can evaluate kurtosis visually via a graph (like a histogram) or mathematically through the kurtosis value statistic.
|
Baseline: the normal distribution Normally distributed data establishes the baseline for kurtosis: not too flat, not too sharply peaked. Data that followed a normal distribution perfectly would have a kurtosis value of 0. Because significant kurtosis indicates that the data are not normal, you may think of the statistic as a first check for normality. |
|
Sharply peaked data A distribution with a sharper than normal peak will have a positive kurtosis value. |
|
Flat peaked data A distribution with a flatter than normal peak will have a negative kurtosis value. |