The Screening Phase

This chapter provides guide for developers to properly organize a factorial experimental design through proper randomization, blocking, and replication. The main interest in the statistical analysis of a screening experiment is to detect which factors show a significant contribution to the explanation of the response variable and also something about how the factors interact. For each factor, it is possible to evaluate if the mean response varies between the factor levels, taking into account the effect of other factors on the response. The appropriate procedure to use for this purpose is the analysis of variance (ANOVA). In the ANOVA table, p‐values are examined to determine whether any factors or interactions, or even the blocks, are statistically significant. Remember that the p‐value is a probability that measures the evidence against the null hypothesis.