Emissivity measurements are of great interest for both theoretical studies and technological applications. Emissivity is a property that specifies how much radiation a real body emits as compared to a blackbody. The emissivity determination of a sample should be an easy task: a simple comparison between the sample and blackbody radiation at the same temperature. Unfortunately, when measuring the emissivity, some practical problems arise due to the differences between the true emitted radiation and the detected quantity. To clarify this point, an analysis of different direct methods for emissivity measurement is presented. Furthermore, a method that includes multiple reflections is developed. The systematic errors associated with each method are computed theoretically as a function of wavelength, sample temperature, and emissivity, and the surrounding enclosure temperature and emissivity. In general, the error is very high for small sample enclosures, but it strongly decreases when the enclosure area increases. Although at short wavelengths all the analyzed methods produce similar errors, noticeable differences appear under other conditions, and methods considering more radiation terms do not always produce lower errors.