Design & Analysis of Experiments with Two- and Three-Level Factors - Design Experiment Tab

Designs for Factors at 2 and 3 Levels

Select the Design experiment tab of the Design & Analysis of Experiments with Two- and Three-Level Factors dialog to access options to select the mixed two- and three-level factorial design. STATISTICA generates full and fractional factorial designs as enumerated by Connor and Young (see McLean and Anderson, 1984) for the National Bureau of Standards of the U.S. Department of Commerce. Note that the program first generates a standard design; the Design of an Experiment with Two- and Three-Level Factors dialog provides options for enhancing the basic design, e.g., by adding replications, etc.

Number of factors (2-level/3-level)

In the Number of factors (2-level/3-level) box is a list of the standard designs as tabulated by Connor and Young. To select a design, click on it, and then click OK; alternatively, you can double-click on the respective design.

Generate a full factorial design

If you select the Generate a full factorial design check box, then STATISTICA generates a full factorial design in which all factor main effects and interactions can be estimated, and are orthogonal. Connor and Young (see McLean and Anderson, 1984) construct the fractional factorial designs by combining factorial 2-level and 3-level designs. Note that the main effects for these mixed level fractional factorial designs are not necessarily orthogonal (see McLean and Anderson, 1984, page 127).