We revisit the feasibility of a cosmological test with the geometric distortion focusing on an ambiguous factor of the evolution of bias. Starting from defining estimators for the spatial two-point correlation function and the power spectrum, in a rigorous manner, we derive useful formulas for the two-point clustering statistics which take the light-cone effect and the redshift-space distortions into account. Then we investigate how the predicted correlation functions are affected by the redshift-space distortions and the bias assuming quasar samples which roughly match the 2dF survey. Effect of the bias is crucial, in particular, on anisotropic component of the clustering statistics in redshift-space. By surveying behavior of the predicted correlation functions to parameters of a phenomenological model for the evolution of bias, it is shown that the correlation functions on the light-cone is sensitive to a mean amplitude of the bias and rather insensitive to the speed of its redshift-evolution. This feature is useful for an analysis of the geometric test.