Phase-locking is a physical phenomenon that refers to a system response which is synchronized with a specific phase of the periodic stimulus. The auditory neural phase-locking plays an important role in revealing the basic neural mechanism of auditory cognition and improving auditory perception. In the existing auditory researches, psychophysical and amplitude spectral methods are mainly adopted. However, those two methods cannot differentiate the envelope-related auditory response from the temporal-fine-structure-related auditory response, and cannot reveal the neural phase-locking mechanism directly either. In this study, a phase locking value (PLV), based on sample entropy, bootstrapping and discrete Fourier transform, is proposed for analyzing the temporal-fine-structure-related frequency following response (FFRT). The proposed PLV is applied to computing neural and physical data. Two electroencephalography experiments are carried out. Results show that the sample entropy of FFRT's PLV is significantly greater than that of FFRE's PLV, and the two PLVs are orthogonal and independent. Thus, the PLVs of FFRE and FFRT reveal the auditory phase-locking mechanisms effectively. In addition, the response to fundamental frequency is mainly attributed to the envelope-related phase locking. And human auditory capability of phase locking to the envelope of the unresolved frequency is superior to the capability of phase-locking to the envelope of the resolved frequency. Moreover, in the case of missing fundamental frequency, the distortion product is the mixture of FFRE in various auditory neural paths. Also, FFRE concentrates at the low harmonic frequencies, while FFRT concentrates at the mid and high order harmonic frequencies. Therefore, the auditory neural phase-locking is related to the frequency resolution of sound. In conclusion, the proposed method overcomes some disadvantages of existing FFR analyses, making it beneficial to exploring auditory neural mechanisms.