As of lately, the demand for developing artificial sensors with improved capabilities for the detection of explosives, toxics or drugs has increased. Ideally, sensor devices should provide high sensitivity and give a response that is specific to a given target molecule without being influenced by possible interfering molecules in the atmosphere. These properties strongly depend on the structure of the chemical compound used as a sensitive material. It is thus crucial to select the right compound and this step would be facilitated with the aid of predictive tools. The present investigations have been focused on a family of functionalized polysiloxane polymers deposited on a QCM device, producing only weak interactions compatible with reversible sensors. The quartz frequency variation at equilibrium has been linked to the partition coefficient that was evaluated using a thermodynamic description of the adsorption process. We have shown that the relative responses of two polymers can be directly determined from the Gibbs free enthalpy of mixing as determined from NMR measurements performed on neat liquid mixtures. An equivalence of this term-including both enthalpy and entropy contributions-to the energy interaction term calculated using Hansen solubility coefficients, has been demonstrated previously. These results constitute a basis for the development of a numerical program for calculating equilibrium sensor responses. For small molecules, the adsorption kinetics can be easily accounted for by a Fick diffusion coefficient estimated from the Van der Waals volume.