A Response Surface Methodology approach to correlate the solubility data of solids in supercritical fluids directly to the pressure and temperature is described. Published data for 2,6-dimethylnaphthalene, 2,3-dimethylnaphthalene, phenanthrene and benzoic acid in both supercritical carbon dioxide and supercritical ethylene were correlated by means of quadratic models, with R-2 statistics greater than 0.99 at more than a 99.9% confidence level. With the developed quadratic models, the solubility for the above-mentioned solids in both supercritical fluids was predicted with average absolute percentage errors less than 5%. So, this approach to estimate solubility provides a possible guide for saving experimental effort and then giving simple models that could be used for industrial design purposes.