The lung dose derived from radon is not attributed to the radon gas itself, but instead to its short-lived progeny. However, in many epidemiological studies as well as in case control studies of the radon risk, the excess number of cancers are related to the radon gas exposure, and not to the radon progeny exposure. A justification for such an approach has resorted to the assumption that there is self-compensation between the radiation doses from the unattached and attached fractions. In the present study, we used the Jacobi model to calculate the radon progeny concentrations in a room by varying the attachment rate and then calculated the resulting lung dose. It was found that self-compensation was not fully realised, and the effective dose can vary by a factor up to approximately 2 for the same radon gas concentration. In conclusion, the radon gas concentration alone does not provide adequate information on the effective dose.