3. Results
3.1. Quality of selected models
In meta-analysis, heterogeneity in the data set is a crucial point
as it can affect the statistical model that the meta-analyst decides
to apply to the meta-analysis database. One of the main results of
the present study was that for all models, the part of the variability
attributable to interstudy variance was less than 50%, except for the
models analyzing the effect of the three pre-slaughter factors on L*.
This apparent high heterogeneity related to L* could be explained
by the difference in the scale of measurement in each study. Sensitivity
analysis showed that the use of equal study weights and a
uniform SD distribution did not significantly influence the posterior
estimation of means, SD and 95% confidence interval (CI) of the different
model parameters. These findings increased robustness of
our results.