National Microbiological Baseline Study in Broiler Chicken
December 2012 – December 2013

Statistical analysis

Although the number of allocated samples to each abattoir and CMA was based on production volume and population density, respectively, the prevalence estimates presented in this report are not yet considered fully weighted and thus called "unweighted". The unweighted prevalence is the proportion of positive samples to a pathogen or indicator organism in the sampled population and products. The prevalence estimates are all representative on a national scale, regardless of sample type, with the exception of farm and flock prevalence that are also generated on a provincial/regional basis. To maintain confidentiality of data sources, the province of Manitoba and Saskatchewan were grouped under the Midwest region while Newfoundland, New Brunswick, Nova Scotia and Prince-Edward-Island were grouped under the Maritimes. All statistical analyses were performed using SAS software version 9.2 (SAS Institute Inc., Cary, NC, USA) and the significance level was set at p < 0.05. Comparison of prevalence values among product types within a stage of the supply chain and between seasons was done by performing a Chi-Square test. As the log-transformed concentration data for both pathogens was not normally distributed using the Shapiro-Wilks W statistic (p < 0.0001), the Kruskal-Wallis one-way analysis of variance was conducted to determine whether the median concentrations values differ among multiple product types followed with the Wilcoxon-Mann-Witney test for the comparison between individual groups.

Date modified: