To use this function, choose .
Conducts the Freeman-Tukey transformation to stabilize variance for Poisson data (counts). For example, you can transform attribute data so that the assumptions of the regression or ANOVA model are met.
For Count, specify a column or stored constant that contains only nonnegative integers.
An editor tracks the number of typographical errors per issue over the course of the year (52 issues). For each error frequency she finds (0-4), the editor records the number of issues, which results in a Poisson distribution. To stabilize the variance before conducting regression, the editor transforms the counts. The editor enters the expression FTC(C1) and stores the results in C3.
|Number of errors||Issues||Transform counts|
Bisgaard, Soren and Howard T. Fuller (1994) Quality Quandaries, Analysis of Factorial Experiments with Defects or Defectives as the Response, Quality Engineering, 7(2), 429–443.