Purpose

    Factorial experiments permit the experimenter to evaluate the combined effect of two or more experimental variables when used simultaneously. Information obtained from factorial experiments is more complete than that obtained from a series of single-factor experiments, in the sense that factorial experiments permit the evaluation of interaction effects. An interaction effect is an effect attributable to the combination of variables above and beyond that which can be predicted from the variables considered singly.

    At the end of a factorial experiment, the experimenter has information which permits him to make decisions which have a broad range of applicability. In addition to information about how the experimental variables operate in relative isolation, the experimenter can predict what will happen when two or more variables are used in combination. Apart from the information about interactions, the estimates of the effects of the individual variables is, in a sense, a more practically useful one; these estimates are obtained by averaging over a relatively broad range of other relevant experimental variables. By contrast, in a single-factor experiment some relevant experimental variables may be held constant, while others may be randomized. In the case of a factorial experiment, the population to which inferences can be made is more inclusive than the corresponding population for a single-factor experiment.

    StatPlus is able to analyze following plans:

• Three Factors (A, B, C) With No Interactions

• Four Factors (A, B, C, D) With Partial Interactions

• Four Factors (A, B, C, D) With Partial Confounding Of AxBxC Interaction

• Greco-Latin Square With No Interactions

• Repeated Measures Latin Square (Random Assignment Of Groups To Rows)

• Fractional Replication Of A Three-Factor Factorial Experiment In Incomplete Blocks

• Repeated Measures Latin Square With Superimposing An Orthogonal Latin Square

• AxBxC (Same Squares Used For All Levels Of Factor C)

Preparations   

    Run Statistics→Analysis of Variance(ANOVA)→Latin and Greko-Latin Squares Analysis menu and select necessary plan.