On Post-Hoc Blocking

Publisher Website
Google Scholar
Abstract
Post-hoc blocking and the analysis of covariance (ANCOVA) are two alternative procedures that employ a concomitant variable as a means of increasing statistical power relative to the completely randomized design. The assumptions underlying the ANCOVA are generally considered to be too restrictive for most applications and the use of this statistical technique is often discouraged. It will be argued that the advantages attributed to the block design are not always valid and that there are circumstances when the ANCOVA would be preferred to post-hoc blocking.