Investigating Disparities in Public Infrastructure Performance: A Case Study of Drinking Water Sites in New Brunswick

There are systemic inequalities in the performance of public infrastructure in New Brunswick, resulting in vulnerable populations who may be disproportionately impacted by lower levels of service. By incorporating customer demographic characteristics into decision-making processes, infrastructure owners and policy-makers can begin to address these inequalities. This study explores the relative performance of 68 drinking water sites (DWS) in New Brunswick from 2006 to 2021 using correlation and univariate analysis. The results are used to determine if there is a significant relationship between customer demographics and performance, or if some demographic groups receive a statistically significant lower level of service compared to others. In this study, DWS performance is measured using three performance indicators: (i) the average number of days spent under a boil water advisory (BWA), (ii) the percentage of samples that detect E.coli and Total Coliforms in the distribution system, and (iii) the percentage of samples that exceed Health Canada’s aesthetic objective for manganese concentration in the distribution system. The correlation between system performance and demographics is calculated using the Spearman’s rank correlation coefficient. Furthermore, DWSs are stratified into three demographic subgroups based on the surrounding community’s income, language, and population size to determine if there is a significant difference in DWS performance across demographics. Mann–Whitney U tests are performed to evaluate the statistical significance of these differences. The study concludes there are several demographic measures that have a weak to moderate correlation with DWS performance, but with no clear trends across all three performance measures. Additionally, the study finds a significant (α = 0.05) difference in the performance of DWSs when compared across income and population size.