Heat flow and power compensation calorimetry measures the power generation of a reaction via an energy balance over an appropriately designed isothermal reactor. However, the measurement of the power generated by a reaction is a relative measurement, and calibrations are used to eliminate the contribution of a number of unknown factors. In this work the effect of the error in the measurement of temperature of electric power used in the calibrations and the heat transfer coefficient and baseline is assessed. It has been shown that the error in all aforementioned quantities reflects on the baseline and it can have a very serious impact on the accuracy of the measurement. The influence of the fluctuation of ambient temperature has been evaluated and a means of a correction that reduces its impact has been implemented. The temperature of dosed material is affected by the heat loses if reaction is performed at high temperature and low dosing rate. An experimental methodology is presented that can provide means of assessment of the actual temperature of the dosed material. Depending on the reacting system, the heat of evaporation could be included in the baseline, especially if non-condensable gases are produced during the course of the reaction. (c) 2006 Published by Elsevier B. V.