A theoretical model for predicting the electrical capacitance of various materials was developed by taking into account geometrical configurations of crucible and rod electrodes. The calculated results were in good agreement with the corresponding measurement data obtained at room temperature (20°C) for liquid materials with known relative permittivity. The measured capacitances of aqueous suspensions containing oxide powders with various grain sizes and relative permittivity values systematically decreased at room temperature with increases in their volume fractions regardless of the sizes of the dispersed solid phases and matched the results obtained from the proposed capacitance prediction model combined with Lichtenecker’s equation for calculating the relative permittivity of dual-phases mixtures. In addition, the validity of the proposed model for predicting capacitances of supercooled oxide melt suspensions was tested at elevated temperatures (above 1300°C). The observed decrease in capacitance for silicate melts with known crystallinities estimated from the corresponding phase diagrams was consistent with the data predicted by the proposed capacitance model combined with Nielsen’s equation instead of Lichtenecker’s equation due to the large differences in relative permittivity between the utilized oxide melts and the solid phases.