You are here:Home>Anderson-Darling Normality Test

Anderson-Darling Normality Test

The Anderson-Darling test is used to test if a sample of data came from a population with a specific distribution. It is a modification of the Kolmogorov-Smirnov (K-S) test and gives more weight to the tails than does the K-S test. The K-S test is distribution free in the sense that the critical values do not depend on the specific distribution being tested. The Anderson-Darling test makes use of the specific distribution in calculating critical values. This has the advantage of allowing a more sensitive test and the disadvantage that critical values must be calculated for each distribution.

If you need further help, please contact support.

By |Comments Off on Anderson-Darling Normality Test

About the Author:

Scott Baird has been president of ProcessModel for more than 15 years. His focus has been to teach others how to improve processes dramatically. He has been successful in transferring these skills to over 200 companies, including ESPN, NASA, GE, Nationwide, Cendant, SSA and many more. Specialties: Group facilitation for process improvement, process design and simulation, simulation modeling, business management and training others to see opportunities. Scott loves to teach process improvement and has often been heard to say, “Of all the things I do, training others to improve processes is my favorite.” Scott is a father of four and a grandfather of eight. He is an avid woodworker, designing and creating presentation boxes. In his spare time, he volunteers in a college preparation program.