The aim of this study was to estimate the overall effect of chilling on the Salmonella occurrence on pig carcasses during pork processing. Meta-analyses were conducted on the outcomes of research studies conducted in 32 abattoirs. A random-effects model was applied, and the between-abattoir variability (τ2=0.548) was significant (p<0.001). Results also showed that publication bias is unlikely. Thus, this meta-analysis allowed the generalisation that chilling reduces the Salmonella incidence by a factor of 1.62 (95% CI: 1.09-2.41). However, a multilevel meta-analysis showed that the 'number of sampling visits', within abattoir, significantly (p<0.05) affected the chilling effect size, accounting for 40%> of the between-abattoir variability. Thus, the beneficial effect of chilling only became apparent in a consistent way when at least two sampling visits took place in an abattoir.