To investigate the validity of the assumption that quiescent galaxies and active galaxies follow the same black hole mass ( M _ { BH } ) - stellar velocity dispersion ( \sigma _ { * } ) relation , as required for the calibration of M _ { BH } estimators for broad line AGNs , we determine and compare the M _ { BH } - \sigma _ { * } relations , respectively , for quiescent and active galaxies .
For the quiescent galaxy sample , composed of 72 dynamical M _ { BH } measurements , we update \sigma _ { * } for 28 galaxies using homogeneous H-band measurements that are corrected for galaxy rotation .
For active galaxies , we collect 25 reverberation-mapped AGNs and improve \sigma _ { * } measurement for two objects .
Combining the two samples , we determine the virial factor f , first by scaling the active galaxy sample to the M _ { BH } - \sigma _ { * } relation of quiescent galaxies , and second by simultaneously fitting the quiescent and active galaxy samples , as f = 5.1 _ { -1.1 } ^ { +1.5 } and f = 5.9 _ { -1.5 } ^ { +2.1 } , respectively .
The M _ { BH } - \sigma _ { * } relation of active galaxies appears to be shallower than that of quiescent galaxies .
However , the discrepancy is caused by a difference in the accessible M _ { BH } distribution at given \sigma _ { * } , primarily due to the difficulty of measuring reliable stellar velocity dispersion for the host galaxies of luminous AGNs .
Accounting for the selection effects , we find that active and quiescent galaxies are consistent with following intrinsically the same M _ { BH } - \sigma _ { * } relation .