Model-free isoconversion methods are the most reliable methods for the calculation of activation energies of thermally activated reactions. A large number of these isoconversion methods have been proposed in the literature. A classification of these methods is proposed. Type A methods such as Friedman methods make no mathematical approximations, and Type B methods, such as the generalised Kissinger equation, apply a range of approximations for the temperature integral. The accuracy of these methods is investigated, by deriving expressions for the main sources of error which includes the inaccuracy in reaction rate measurement, approximations for the temperature integral and inaccuracies in determination of temperature for equivalent fraction transformed. Both highly inaccurate and highly accurate Type B methods are identified. In cases where some uncertainty over baselines of the thermal analysis data exists or where accuracy of determination of transformation rates is limited, type B methods will often be more accurate than type A methods. (C) 2003 Elsevier Science B.V. All rights reserved.