A Note on the Analysis of the Two-Period Crossover Design When the Period-Treatment Interaction is Significant

✍ Scribed by A. P. Grieve

• Publisher: John Wiley and Sons
• Year: 1987
• Tongue: English
• Weight: 250 KB
• Volume: 29
• Category: Article
• ISSN: 0323-3847
• DOI: 10.1002/bimj.4710290703

✦ Synopsis

In the analysis of a two-period cro88over study GBIZZLE (1965) suggeata that, if a preliminary k t for a period by treatment inbraction (residual effeot, carryover effect) fsignificant at the 10 o/o level, the direct effects of the treatments should be compared by performix$ a t -h t on the data from the first period only. In this note it is shown that under Grizzle's model the comparieon of the direct treatment effects is equivalent to a Behrens-Fisher problem. Depending on one's viewpoint-Bayesian, fiduciei, or ssmpling theory4ifferent solution8 are p i b l e . The eolutions are illustrated using three well-known data seta.