An article in Quality Engineering (“Quality Quandaries:

Two-Level Factorials Run as Split-Plot Experiments,”

Bisgaard et al., Vol. 8, No. 4, pp. 705–708, 1996) describes a 25 factorial experiment in a plasma process focused on making paper more susceptible to ink. Four of the factors (A–D) are difficult to change from run to run, so the experimenters set up the reactor at the eight sets of conditions specified by the low and high levels of those factors, and then processed the two paper types (factor E) together. The placement of the paper specimens in the reactor (right versus left) was randomized.

This produces a split-plot design with A–D as the whole-plot factors and factor E as the subplot factor. The data from this experiment are shown in Table P14.4. Analyze the data from this experiment and draw conclusions.