A Change-Over Design for Testing a Treatment Factor at Four Equally Spaced Levels
Journal of the Royal Statistical Society: Series B (Statistical Methodology) · 1967 · Volume 29 · Issue 2
DOI: 10.1111/j.2517-6161.1967.tb00702.x
Abstract
A new design for testing a quantitative factor at four equally spaced levels is presented. This design is suitable for quick analysis when the presence of first residual effects can be adequately accounted for by a single (linear) degree of freedom. The calculation for testing this assumption and the analysis in the general case are given.
Enable JavaScript for citation tools and additional paper interactions.