Analysis of a Screening Experiment with Two-Level Factors - Quick Tab

Analyzing the Results of a 2(k-p) Experiment

Select the Quick tab in the Analysis of a Screening Experiment with Two-Level Factors dialog box to access the options described here. Note that these results are for the currently specified model, and often the spreadsheets or graphs generated on this tab are dependent on options on other tabs as indicated. You can specify a new model on the Model tab.

Summary: Effect estimates

Click this button to produce a spreadsheet with the ANOVA effect estimates and coefficients for the coded model. The effects that are computed depend on the current model, as specified via the options in the Include in model group box on the Model tab. Additional options for confidence intervals, Alpha highlighting, and sorting effects are available on the ANOVA/Effects tab. All estimates in this spreadsheet pertain to the coded factor settings, that is, to the factor settings scaled to the ±1 range. To see the results for the original (untransformed) factor settings, click the Regression coefficients button on the ANOVA/Effects tab. For a description of the statistics given in this spreadsheet, see Main Effects and Interactions for Experiments with Two-Level Factors.

ANOVA table

Click the ANOVA table button to produce an ANOVA table for the current model, as specified in the Include in model group box and based on the selected error term in the ANOVA error term group box, both on the Model tab. Note that if you select the estimate of Pure error for the error term (if it is available), the ANOVA table also includes a Lack of fit test (see Introductory Overview). This is a test of the residual variance, after controlling for all effects in the model, against the estimate of pure error. If significant, then there is indication of additional significant effects, or differences between means of the design that cannot be accounted for by the parameters currently in the model. For example, there may be higher-order interactions between the factors in the design.

Pareto chart of effects

Click this button to produce a Pareto chart of the ANOVA effect estimates, or, optionally, the standardized effect estimates (if the Plot standardized effects check box is selected on the ANOVA/Effects tab). The Pareto chart shows the effect estimates sorted by their absolute size. If you plot the standardized effects, a vertical line is also shown to indicate the minimum magnitude of Statistically significant effects, given the current model and choice of error term, and using the criterion of Statistical significance selected in the Alpha (highlighting) box on the ANOVA/Effects tab. The Pareto chart is very useful for reviewing a large number of factors and for presenting the results of an experiment to an audience that is not familiar with standard Statistical terminology.

Observed marginal means

Use the options in the Observed marginal means group box to compute the marginal means for the design (e.g., the means for Factor 1 by Factor 2, collapsed across Factor 3 and Factor 4). If the Show text labels instead of factor values check box is selected on the Design tab or Means tab, the factor levels in this spreadsheet and/or plot are labeled with their text labels.

Display

Click the Display button to display the Compute marginal means for dialog box, in which you specify for which factors to display marginal means. Specify the factors and click the OK button to compute the marginal means for the design and display them in a spreadsheet. If the Show text labels instead of factor values check box is selected on the Design tab, the factor levels in this spreadsheet are labeled with their text labels. See Marginal Means for Screening Experiments for a description of that spreadsheet.

Means plot

Click the Means Plot button to produce a plot of weighted or unweighted marginal means. The computation of the marginal means, standard errors, and confidence intervals follows the procedures outlined in Marginal Means for Screening Experiments. After you click the Means Plot button, the Compute marginal means for dialog box is displayed, in which you specify for which factors to display marginal means. After selecting the factors, the Specify the arrangement of the factors in the plot dialog box is displayed, in which you select the assignment of factors. Note that the orientation and layout of the x-upper value labels can be adjusted via the Graph Options dialog box. The computation of the marginal means, standard errors, and confidence intervals follows the procedures outlined in the note on Marginal Means.

Display/plot weighted means

Select the Display/plot weighted means check box if you want to compute weighted means for the marginal means plot or spreadsheet.  

Predicted (estimated) means

Use the options in the Predicted (estimated) means group box to produce the so-called square or cube plots for the predicted (estimated) means (see also the Introductory Overview).

Square plot of predicted means. Click this button to display the Factors for square plot dialog box, in which you select the two lists of factors for the square plot. Click the OK button in this dialog box to plot the predicted means for the low and high settings for two factors. If the Show text labels instead of factor levels is selected on the Design tab or the Means tab, this plot will label the factor levels with the respective text labels. The predicted means are computed based on the current model as specified in the Include in model group box on the Model tab. If there are more factors in the current design (including blocks and curvature check) than what is to be plotted in the square plot (2 factors), Statistica computes the predicted means based on the means for all other factors in the design, including block factors (i.e., the recoded new variables to compute the block effects; see Main Effects and Interactions; if the design has equal N per experimental condition, the block effects will be ignored because the means for the recoded new variables will be 0). The coefficient for curvature (if requested) will always be ignored.

Cube plot of predicted means

Click this button to display the Factors for cube plot dialog box, in which you select the factors for the cube plot. Click the OK button in this dialog box to plot the predicted means for the low and high settings for three factors. If the Show text labels instead of factor levels is selected on the Design tab or the Means tab, this plot will label the factor levels with the respective text labels. The predicted means are computed based on the current model as specified in the Include in model group box on the Model tab. If there are more factors in the current design (including blocks and curvature check) than what is to be plotted in the cube plot (3 factors), Statistica computes the predicted means based on the means for all other factors in the design, including block factors (i.e., the recoded new variables to compute the block effects; see Main Effects and Interactions; if the design has equal N per experimental condition, the block effects will be ignored because the means for the recoded new variables will be 0). The coefficient for curvature (if requested) will always be ignored.