Power Analysis for Lack-of-Fit Tests in Designed Experiments

An article by R. Dennis Cook provides a method for determining power and sample size for lack-of-fit tests in unreplicated designs. The 328.6 KB resource, last updated in 2026, explores the effect of different replication patterns on statistical power. It offers guidelines for augmenting factorial, screening, and response surface experiments.

Use Cases

• Calculate required sample size for lack-of-fit tests based on the described bounds on the null model.
• Plan the distribution of replicate points in factorial designs based on the analysis of replication patterns.
• Evaluate the power of center-point replication strategies in continuous-factor experiments as discussed in the article.
• Augment unreplicated base designs like response surface constructions using the provided guidelines.

Strengths

• Method is described as easy-to-implement and based on standard lack-of-fit tests.
• The approach is not restricted to a single class of design, though it emphasizes factorial and response surface designs.
• The resource is openly licensed under CC-BY-4.0.

Limitations

• The actual data content (e.g., column names, sample data) is unavailable for inspection.
• The description is conceptual; data quality and structure require manual inspection after download.
• The 328.6 KB size suggests a limited scope, likely containing the article and supplementary files rather than a large dataset.

Provenance

Source

R. Dennis Cook

Collection Method

Likely contains methodological exposition and analysis.

Freshness

Last updated 2026-05-18 13:59:24.