Using Mass Balance, Factor Analysis, and Multiple Imputation to Assess Health Effects of Water Quality

This dissertation evaluates the performance of multiple imputation method in filling in missing microbial data, and utilizes the chemical mass balance model and the exploratory factor analysis for the identification of sources of fecal contamination.;The multiple imputation method was applied on surface water measurements collected on the Chicago River from 2007 to 2009. The method was used to fill in missing values and the original dataset was compared to the imputed dataset. Descriptive statistics show that the imputed dataset has a similar distribution as the original dataset. In order to further evaluate the performance of the imputation, a portion of the original dataset was deleted, and the missing values were filled in using multiple imputation method. Results show that the imputed dataset can provide inferential parameter estimates, and that multiple imputation can fill in missing microbial data without distorting the distribution of the original dataset.;The chemical mass balance model and exploratory factor analysis were then utilized to identify sources of fecal contamination in the river system. Sources identified included physicochemical and densities of other microbes. Results from both methods suggested that microbial sources may vary at different locations on the river system. Identified sources were then used in predicting of the risk of Acute Gastrointestinal (AGI) illness among water users. However, no association between pollutant sources and health risk was identified..