Towards eliminating stimulus artifacts, alternating polarity stimuli have been widely adopted in eliciting the auditory brainstem response. However, considering the difference in the physiologic basis of the positive and negative polarity stimuli on the auditory system, it is unclear whether alternating polarity stimuli would adversely affect the auditory brainstem response characteristics. This research proposes a new polarity method for stimulus artifacts elimination, Sum polarity, that separately utilized the rarefaction and condensation stimuli and then summed the two evoked responses. We compared the waveform morphology and latencies of the auditory brainstem responses evoked by familiar stimuli (including click, tone-burst, and chirp) with different polarity methods in normal-hearing subjects to investigate the new method’s effectiveness. The experimental results showed that alternating polarity of the click and chirp had little effect on the auditory brainstem response. In contrast, alternating polarity affected the waveform morphology and latencies of the auditory brainstem responses to the low-frequency tone-burst, with the effect decreasing as the stimulus frequency increased. These results demonstrated the performance of any polarity method is related to the characteristics of the stimulus signal itself, and no polarity method is optimal for all types of stimuli. Based on the analysis of experimental results, a fixed polarity and alternating polarity were recommended for the click and chirp auditory brainstem responses, respectively. Furthermore, considering the apparent latency differences between the responses to opposite polarity stimuli, the Sum polarity was suggested for the tone-burst auditory brainstem responses. Moreover, this work verified the feasibility of the Sum polarity, which offers another choice for eliminating stimulus artifacts in an evoked potential acquisition.