The traditional isoscaling technique has been used to analyze all isotopically identified particles from 70Zn+70Zn and 64Zn+64Zn collisions at 35 MeV/u. Two additional techniques, using an energy-averaged ratio and using the data present in the tail of energy spectra, are compared to the traditional method and show similar results. Isoscaling fit parameters α and β are found both globally and for each individual series of constant Z and N. The data are then split up between emitted fragments and projectile-like fragments. Isoscaling values for the two different types of fragments are shown to be different, emphasizing the importance of experimentally distinguishing between projectile-like fragments and emitted fragments in order to achieve accurate isoscaling parameters.