Using the mock galaxy catalogues created from the N-body simulations, various biasing prescriptions for modelling the relative distribution between the galaxies and the underlying dark matter are statistically tested by using scale-scale correlation. We found that the scale-scale correlation is capable of breaking the model degeneracy indicated by the low-order clustering statistics, and could be taken as an effective discriminant among a variety of biasing models. Particularly, comparing with the APM bright galaxy catalogue, we infer that the two parameter Lagrangian biasing model gives the best fit to the observed clustering features.