Author Information

L. A. Goonewardene
L. Z. Florence

Abstract

Many growth experiments, in which weights are taken at different times on the same animals, involve the comparison of factorial main effects and interactions but exclude time (period) as an effect. The objective of this paper is to show that more information can be obtained by analysing the data as a repeated measures design. As an example, feedlot cattle being prepared for market are often on growth implants and provided different diets depending on the stage of growth and maturity. Growth promoting implants, either single or double, may be slow or fast acting. During the growing period, a diet with less grain and medium energy is fed but during the finisher period the grain component is increased. Responses to implant and diet may be dependent on the length of time between measurements. Any model designed to analyze the responses within time, will be limited as it will not include all treatment x time interactions, which can be very important. A repeated measures or split plot in time can detect these treatment x time interactions, but criteria such as the sphericity of the covariance matrix should be satisfied, so that the within subject effects can be correctly tested. The paper describes four statistical models appropriate for such data using SASR/STAT software.

Keywords

repeated split-plot variance treatment period interaction

Creative Commons License


This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS
 
Apr 26th, 5:40 PM

THE NEED FOR THE ANALYSIS OF TREATMENT x PERIOD INTERACTION IN ANIMAL EXPERIMENTS

Many growth experiments, in which weights are taken at different times on the same animals, involve the comparison of factorial main effects and interactions but exclude time (period) as an effect. The objective of this paper is to show that more information can be obtained by analysing the data as a repeated measures design. As an example, feedlot cattle being prepared for market are often on growth implants and provided different diets depending on the stage of growth and maturity. Growth promoting implants, either single or double, may be slow or fast acting. During the growing period, a diet with less grain and medium energy is fed but during the finisher period the grain component is increased. Responses to implant and diet may be dependent on the length of time between measurements. Any model designed to analyze the responses within time, will be limited as it will not include all treatment x time interactions, which can be very important. A repeated measures or split plot in time can detect these treatment x time interactions, but criteria such as the sphericity of the covariance matrix should be satisfied, so that the within subject effects can be correctly tested. The paper describes four statistical models appropriate for such data using SASR/STAT software.