Correction ultrasonographic measurement of carcass in rabbits using mixed linear Model with covariates
Datasets usually provide raw data for analysis. This raw data often comes in spreadsheet form, but can be any collection of data, on which analysis can be performed.
ABSTRACT The objective of this study was to evaluate the correction ultrasonographic measurement of area the Longissimus dorsi muscle in New Zealand rabbits by covariance analysis using mixed linear models. The analyzes were performed in randomized block design with 5 treatments (operators) and 6 blocks (animals), considering in the analysis: absence of covariates; rib eye length as covariate; rib eye depth as covariate; the two covariates together. As the animals are a random sample, the block effect was considered to be random. The covariates were considered as measures of fixed effect without error, independent of treatment and linear comportment. The decision criterion statistics CV%, R², and R ¯ ²showed a direct relationship between them and can be taken into consideration to evaluate the experimental accuracy in tests with carcass evaluation. The AIC, BIC, and AICC statistics are consistent with the interpretation of the decision criteria and indicate that the two covariates in model provides accurate results. The inclusion of covariates complements the local control to improve the accuracy of the experiment. The use of ultrasound measurements of depth and length corrects the mean area of the Longissimus dorsi muscle evaluated by different operators.