In certain statistics simulations, scrambling the values of an attribute provides a way to enforce the null hypothesis in which there is no relationship between two attributes. When you scramble attribute values (see Use Scrambling to Test for Independence), you get a new collection that is a scrambled collection, with a Scramble panel for controlling the simulation.
The functioning of the three controls is explained below.
Key Curriculum Press/Key College Publishing www.keypress.com/fathom |