Analysis of systematic errors in the calculation of renormalization constants of the topological susceptibility on the lattice.

A Ginsparg-Wilson based calibration of the topological charge is used to calculate the renormalization constants which appear in the field-theoretical determination of the topological susceptibility on the lattice. A systematic comparison is made with calculations based on cooling. The two methods agree within present statistical errors (3%\u20134%). We also discuss the independence of the multiplicative renormalization constant Z from the background topological charge used to determine it