A method is provided for evaluating the reliability of an electrical power measuring device for measuring high-frequency electrical power. To build an evaluation system, the measuring device, together with a reference electrical power measuring device, is arranged between a high-frequency power supply device and an artificial reproduction load, which includes an impedance conversion device and a reference load. Using this system, an uncertainty range of an electrical power measured value measured by the measuring device is calculated, according to a prescribed calculation formula, from the electrical power measured value, and a judgment is made as to whether or not the electrical power measured value measured by the measuring device is within the uncertainty range. If it is within the uncertainty range, the measuring device is evaluated as being reliable, while if it is not within the uncertainty range, the measuring device is evaluated as being unreliable.