When modeling analyte-additive interactions in capillary electrophoresis (CE), it is necessary to correct for all changes in the apparent electrophoretic mobility of an analyte that are not due to specific binding. Current models based on dynamic complexation have corrected for bulk viscosity changes in the background electrolyte (BGE) when additives are used, while assuming negligible changes in the dielectric constant and other physicochemical properties of the solution. In this report, a study of weak interactions between deoxyribonucleotides and hydroxypropyl-beta-cyclodextrin (HP-beta-CD) revealed significant nonideality in binding isotherms. Changes in the dielectric properties of the solution due to the addition of high concentrations of HP-beta-CD to the BGE was observed to alter the electrophoretic mobility of analytes. A relative dielectric correction factor was required to normalize analyte mobilities to a reference state of zero additive concentration. The use of both a relative dielectric factor and a viscosity correction factor was found to increase the accuracy of the model, reflected by a higher degree of correlation between predicted and measured analyte mobilities. This type of correction is particularly relevant when studying weak analyte binding interactions or when using high concentrations of additive in the BGE. This work is vital for accurate determination of weak binding constants and mobility values, as well as providing a deeper understanding of the fundamental parameters influencing a separation in CE.