A method for evaluating a measurement result of a thermal analysis. A program-controlled computer unit is used to calculate at least one probability of the agreement of the measurement result with at least one dataset previously stored in the computer unit, wherein this calculation is based on a comparison of effect data previously extracted from a measurement curve of the thermal analysis with corresponding stored effect data of the dataset. The evaluation can advantageously include, an automatic recognition and classification of measurement curves and can be carried out in particular more efficiently, more economically and more quickly than previously, with at the same time a high quality of evaluation.