Mekelle University, Department of Animal, Rangeland and Wildlife Sciences, PO Box 231, Tigray, Ethiopia
Kebede, K., Mekelle University, Department of Animal, Rangeland and Wildlife Sciences, PO Box 231, Tigray, Ethiopia; Gebretsadik, G., Mekelle University, Department of Animal, Rangeland and Wildlife Sciences, PO Box 231, Tigray, Ethiopia
Most statistical approaches in experiments of feeding-trial are based on variance analysis (ANOVA). However, most of the time, the assumption that data are independent is violated since several measures are performed on the same subject (repeated measures). As a result, standard regression and analysis of variance methods may produce invalid results of repeated measures data because they require mathematical assumptions that do not hold with these data. In addition, the presence of intra- and inter-observers variability can potentially obscure significant differences. The linear mixed models (LMM) is an extended multivariate linear regression method of analysis that have been proposed to circumvent these limitations, by adding random effects aimed at modelling the variability due to peculiarity of the observed subjects and thus leading to more efficient parameters estimates. In this study, repeated records of body-weight gains on lambs were considered for analysis. Up to nine 'repeated records' of body-weight gains per lamb, measured between 10th and 90th day of age post the time of initial body weight recording, were available. The objectives of the study were twofold: (a) to compare a two-step model, i.e. fixed and random regression (FR and RR) models, for evaluating body-weight gain of lambs. It should be clarified, whether RR models deliver better model fitting in contrast to FR models and (b) to test and quantify the difference in body-weight gains of animals in different feeding-groups and at different ages (days-on-test). Results showed that a linear regression on age modelled changes in variation of body weight adequately. As compared to FR models, RR models delivered better estimates of -2logL, AICC and BIC. Furthermore, significance results in LSQ-Means were found for the comparison of the four feeding-groups and difference in body-weight gains at age t=15.