Abstract: In analyzing γ-emitters in high level liquid waste with γ spectrometry, the coincidence summing effect is an important factor affecting measurement accuracy for nuclides with multiple γ lines. This paper is to study the influence of the coincidence summing effect on the measurement of more than 10 major γ-emitters in high level liquid waste, and perform coincidence summing correction. Using the LabSOCS of Genie 2000, the influence of the coincidence summing effect was studied with the consideration of various detectors, sample volumes, and sample densities. In addition, experimental verification was conducted by combining γ spectrometry and liquid scintillation counting, with γ-emitters ⁶⁰Co, ¹³⁷Cs/¹³⁷Ba^m, ¹⁵²Eu, ²³⁷Np/²³³Pa and ²⁴¹Am. It is shown that for detectors of the same series, the larger the crystal size, the greater the influence of the coincidence summing effect. The coincidence summing effect decreases with increasing sample volume and does not change significantly with sample density. With regard to the preferred full energy peak which has higher intensity and less interference, ¹³⁴Cs and ¹⁵⁴Eu exhibit the highest coincidence summing interference, ²³⁷Np and ²³³Pa the smallest, between which are ⁶⁰Co, ¹⁰⁶Rh, ¹²⁵Sb, ¹⁵²Eu, and ²³⁹Np. The γ-emitters ¹³⁷Cs, ¹⁵⁵Eu, ²⁴¹Am, and ²⁴³Am are free of coincidence summing interference. Three types of counting efficiencies have been discussed: ε is defined as CR/(A•I), where CR is the count rate measured directly, A is the radioactivity, and I is the intensity of γ-ray; ε′ is defined as ε/COI, where COI is the true coincidence summing correction factor; and ε″ is the counting efficiency obtained from LabSOCS calculation. Both ε′ and ε″ are free of the coincidence summing interference, but ε is not. The usual efficiency curve is free of the coincidence summing interference, and when using it, consideration should be given to whether coincidence summing correction is necessary depending on the measurement condition. If nuclides with multiple γ lines are measured and the distance between the source and detector is very close, it is necessary to perform coincidence summing correction; otherwise, the usual efficiency curve cannot be used for activity calculation. Since the efficiency curve in the range of 200-2 000 keV is approximately a straight line on a double logarithmic coordinate, this line can be determined using common γ-emitters such as ⁶⁰Co, ¹³⁷Cs, and ¹⁵²Eu, and then used to measure other γ-emitters in high level liquid waste that can emit γ-rays with energy ranging from 200 to 2 000 keV, such as ¹⁰⁶Rh, ¹²⁵Sb, and ¹³⁴Cs.