The objective of this collaborative study was to evaluate the proposed method for determining the total nitrogen in soy sauce by the Kjeldahl method submitted to the Codex Alimentarius Commission for endorsement in accordance with the protocol for the design, conduct, and interpretation of method-performance studies. The digestive conditions of the proposed method are the addition of 10 mL of H₂SO₄, 10 g (8 g by using a block digester) of K₂SO₄, and 1 mL of 20% CuSO₄·5H₂O and 80 min boiling period after the liquid is cleared by a heating device. Seventeen laboratories participated, analyzing five soy sauce samples as blind duplicates. Since the volume sampling method used in the JAS (Japanese Agricultural Standard) method showed lower accuracy of data because of the density of soy sauce, the method of sampling by weight was adopted as the proposed method. The total amount of outlier data was within acceptable limits for method-performance studies (≤ 22.2%). Lysine and ammonium sulfate recoveries for all laboratories were ≥ 98% and ≥ 99% respectively. The RSD_r (repeatability relative standard deviation) values ranged from 0.4 to 1.3%, and the RSD_R (reproducibility relative standard deviation) values were from 0.8 to 1.9%. HORRAT (RSD_R/predicted RSD_R) for the reproducibility showed 0.2 to 0.4, indicating acceptable precision of the method and excellent analytical performance.