Design of a Central Composite (Response Surface) Experiment - Display Design Tab

Design Considerations

Select the Display design tab of the Design of a Central Composite (Response Surface) Experiment dialog box to access options to specify the appearance of the design, to review the design, and to change any factor names or levels.

Summary: Display design

Click the Summary: Display design button to display a spreadsheet specifying the settings for each experimental run. The way in which the design is displayed is affected by the respective settings of the options on this tab. To save the design, select the spreadsheet and use the Save or Save As options on the File menu. Note that if you request to display Text labels (see below) in your spreadsheet, and then save it as a data file, Statistica saves the text labels along with the "underlying" factor minima and maxima. To use this spreadsheet in further analyses, you must specify it as an input spreadsheet. For more information, see the Input Spreadsheet Overview.

Note: Partial replications. If you want to replicate selected points from the design (e.g., so that you can later obtain an estimate of pure error; see the Introductory Overview), first add rows to the design (use option Insert - Add Cases), and then copy and paste the respective runs into those new rows.

Anchoring of star points

Note that the star points are always "anchored" against the respective center point values. If you specify minimum, center point, and maximum factor values in the spreadsheet with the Change factor names, values, etc. option (see below), then the star point values are computed as:

center ± alpha * 1/2 * (max-min)

where center, min, and max stand for the respective center point, minimum, and maximum factor values.

Change factor names, values, etc

Click the Change factor names, values, etc. button to display the Summary for Variables (Factors) spreadsheet, which contains 1) the current factor names, 2) minimum, center, and maximum values, 3) the corresponding text labels (for minimum, center, and maximum values), and 4) the text labels for the high and low star point values (i.e., Alpha, see the Introductory Overview). Type in the desired text and numeric values, and then click the OK button; the values that you entered will then be used in the spreadsheet displaying the design (option Summary: Display design, see above). See also, General User Entry Spreadsheet.

Denote factors

Use the options in the Denote factors group box to determine how the factors are labeled in the spreadsheet.

By numbers

Select the By numbers option button to denote the factors by numbers; that is, to label consecutive factors as F1, F2, ..., Fn.

By letters

Select the By letters option button to denote the factors by letters; that is, to label consecutive factors as A, B, C, and so on.

By names

Select the By names option button to denote the factors by their names, as specified via the button Change factor names, values, etc., on the respective results dialog.

Order of runs

The choice of options in the Order of runs group box affects the order of runs displayed in the spreadsheet. Remember that for standard experiments, it is common practice to randomize the runs in order to avoid any systematic biases due to fluctuations in the production process across runs.

Standard order

Select the Standard order option button to display the design in standard order.

Random

Select the Random option button to randomize the order of runs.

by blocks

Select the by blocks option button to sort by blocks after randomization, resulting in randomization within blocks (if the current design is not blocked, then sorting by blocks is not available).

Seed

The positive integer value in the Seed box is used to seed the random number generator. Note that the random number generator is reseeded every time you enter the Design of a Central Composite (Response Surface) Experiment dialog box. Thus, when you generate the same designs more than once, the default random seed in the Seed box will be different every time. If you want to generate the identical order in consecutive analyses, enter into this box the same number as that shown in the initial analysis. In practice, since randomization is supposed to accomplish unbiased assignment of runs to experimental conditions, you usually do not want to change this value.

Show (in Spreadsheet). Use the options in the Show (in Spreadsheet) group box to specify the way the levels of factors are displayed in the spreadsheet.

Numbers (+1, -1). Select the Numbers (+1, -1) option button to change the display to show positive Alpha values and negative Alpha values: +1's, 0's, -1's.

Mini/maxima. By default (the Mini/maxima option is the default setting) the levels of the factors are denoted by the respective factor lows and highs.

Text labels

Select the Text labels option button to show the respective text labels for the low star point, low, center, high, and high star point factor settings (note that the values for the factor settings as well as the respective text labels can be changed via the Change factor names, values, etc. option, see above). If you select the Text labels option button, when you click the Summary: Display design button (see above), the spreadsheet will show the respective text labels for the factor settings; however, if you save the design as a Statistica data file (via the Save or Save As options on the File menu), then Statistica saves the text labels along with the "underlying" numeric values for the respective factor settings.