When a statistical object, such as a test, is analyzing a collection whose values are generated randomly, it becomes useful to collect the results from the test for repeated rerandomizations. This allows you to build distributions from such things as linear-regression slopes, t-statistics, p-values, and so on. The dot plot at right shows the results of collecting 100 p-values under conditions in which the null hypothesis is true.
You can also use sliders in statistical objects and then collect measures for different slider values.
1. Select the statistical object. (It should be one whose results are going to change either manually or automatically.)
2. Choose Test | Collect Results as Measures.
You will get a new collection with five cases (the default number) whose attributes correspond to statistical results computed by the object.
To have Fathom collect another set of measures whenever the statistical object changes (by rerandomizing its collection or changing a slider),
1. Check Recollect when source changes in the Collect Measures panel.
2. Change the number of measures to collect to 1.
Key Curriculum Press/Key College Publishing www.keypress.com/fathom |