Context : Aims : We analyze systematics in the asteroseismological mass determination methods in pulsating PG 1159 stars . Methods : We compare the seismic masses resulting from the comparison of the observed mean period spacings with the usually adopted asymptotic period spacings , \Delta \Pi _ { \ell } ^ { a } , and the average of the computed period spacings , \overline { \Delta \Pi _ { \ell } } . Computations are based on full PG1159 evolutionary models with stellar masses ranging from 0.530 to 0.741 M _ { \odot } that take into account the complete evolution of progenitor stars . Results : We conclude that asteroseismology is a precise and powerful technique that determines the masses to a high internal accuracy , but it depends on the adopted mass determination method . In particular , we find that in the case of pulsating PG 1159 stars characterized by short pulsation periods , like PG 2131+066 and PG 0122+200 , the employment of the asymptotic period spacings overestimates the stellar mass by about 0.06 M _ { \odot } as compared with inferences from the average of the period spacings . In this case , the discrepancy between asteroseismological and spectroscopical masses is markedly reduced when use is made of the mean period spacing \overline { \Delta \Pi _ { \ell } } instead of the asymptotic period spacing \Delta \Pi _ { \ell } ^ { a } . Conclusions :