Are neighbour methods preferable to analysis of variance for completely systematic designs? ‘Silly designs are silly!’
R. A. BAILEY, J.-M. AZAÏS, H. MONOD
Biometrika · 1995 · Volume 82 · Issue 3
Abstract
SUMMARY It is shown that the usual analysis of a complete-block design is weakly valid under a range of assumptions but that more sophisticated analyses of systematic block designs may not be.
Enable JavaScript for citation tools and additional paper interactions.