Publication date: 07/30/2020

A screening design often provides no degrees of freedom for error. So classical tests for effects are not available. In such cases, Effect Screening is particularly useful.

For these designs, most inferences about effect sizes assume that the estimates for non-intercept parameters are uncorrelated and have equal variances. These assumptions hold for the models associated with many classical experimental designs. However, there are situations where these assumptions do not hold. In both of these situations, the Effect Screening platform guides you in determining which effects are significant.

The Effect Screening platform uses the principle of effect sparsity (Box and Meyer 1986). This principle asserts that relatively few of the effects that you study in a screening design are active. Most are inactive, meaning that their true effects are zero and that their estimates are random error.

The following Effect Screening options are available:

Scaled Estimates

Gives parameter estimates corresponding to factors that are scaled to have a mean of zero and a range of two. See Scaled Estimates and the Coding of Continuous Terms.

Normal Plot

Identifies parameter estimates that deviate from normality, helping you determine which effects are active. See Normal Plot Report.

Bayes Plot

Computes posterior probabilities for all model terms using a Bayesian approach. See Bayes Plot Report.

Pareto Plot

Plots the absolute values of the orthogonalized and standardized parameter estimates, relating these to the sum of their absolute values. See Pareto Plot Report.

Want more information? Have questions? Get answers in the JMP User Community (community.jmp.com).