Lowering the freezing point of water is proportional to the molality of the salt. The coefficient of propotionality is a cryoscopic constant with an isotonic coefficient. Instead of the concentration for electrolytes, activity is introduced. According to the law of average force, the ion activity coefficient does not depend on the specific type of ions, but depends on the ionic strength of the solution. The ionic strength depends on the charge. As you can see, there is no dependence on the radius of the ion.

Lowering the freezing point of water is proportional to the molality of the salt. The coefficient of propotionality is a cryoscopic constant with an isotonic coefficient. Instead of the concentration for electrolytes, activity is introduced. According to the law of average force, the ion activity coefficient does not depend on the specific type of ions, but depends on the ionic strength of the solution. The ionic strength depends on the charge. As you can see, there is no dependence on the radius of the ion.

I support the Yuri views, and I'd like to add that these laws are valid for totally dissociated salts (or electrolytes), meaning that solutions should be enough diluted. The simple theory can be read in Wikipedia:

https://en.wikipedia.org/wiki/Boiling-point_elevation

Interesting, you are right to a point, but dilute solutions of ions will not lower the boiling point of water. The water structure will however to a certain point affect the level of increase in the boiling point. For lowering boiling points of water, we would be looking at ethanol for example (famous case as an azeotrope) and this certainly has to do with the interactions of water and ethanol in the solution.

Thermodynamics (ebullioscopic constant) predicts a rise in the boiling point for ions as correctly stated in the other answers, but thermodynamics is a "mean field" approach ignoring individual molecules. The scientific community has for years (if not for ages) been trying to nail thermodynamics down to molecular scale and is getting closer, but calculations are often still (slightly) off, not to mention the computer power necessary to do so.

Thus, the effects that you mention are related, but the easier way for prediction is certainly the cryoscopic or ebullioscopic constant as mentioned by Jean-François and Yuri.

Thank for your answers!

I meant to write earlier that the boiling point elevation should be lower for ions which have a smaller stokes radius compared to ions which have a larger stokes radius. Seeing that the boiling point elevation formulas don't depend on the type of ion, presumably the effect is for most experiments so minimal, that it can be ignored.

The presently available theories of aqueous ionic solutions are too crude to deal with problems of the sort you address and limited to very dilute solutions.Ionic size definitely affects colliigative properties and phase equilibria at higher concentrations since it is a first order measure of the specific local interactions of an ion with hydration waters in addition to the ubiquitous screened Coulomb interactions crudely treated via Debye

Hueckel like theories and the ionic strength concept.Unfortunately computer simulations of dilute to moderate molality salt solutions are also

not feasible due to the long range Coulomb interaction and the complex nature of water and ion water interactions.For the time being one has to resort on fits of experimental data etc.