Background: The domains for moral emotions are still open to debate. Rozin et al.’s (1999) CAD triad hypothesis assumed selective, corresponding ties between three moral codes (community [C], autonomy [A], and divinity [D]) and three other-critical emotions (contempt [C], anger [A], and disgust [D]). Objective: To identify domains for violations of the big three moral codes in the CAD Scale and to examine the robustness of its 3-factor structure among a Japanese population. Methods: We used the data for a group of Japanese parents (n = 260) to whom the CAD Scale was distributed on the web. Each domain for the CAD hypothesis was analysed separately. Correlations between CAD Scale items and the total scores of the three domains, and alpha coefficients of the items belonging to each domain were calculated. Items with the highest correlation with another domain were deleted. A one-factor exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) of the items for each domain for the 1-factor model were performed. Because 1-factor models did not reach satisfactory levels, we examined parcelled CFAs. Results: All items showed the highest correlation with the domain assumed by the CAD hypothesis except for items 8, 10, 15, 22, and 41. In the EFAs, all factor loadings showed higher than 0.33, and alpha coefficients of the C, A, and D domains were 0.937, 0.881, and 0.949, respectively. The CFAs of the 1-factor model for the C, A, and D domains did not show an acceptable fit. Parcelled models for each domain showed a perfect fit to the data in each domain of the CAD scale (χ² = 0.000, CFI = 1.000, RMSEA = 0.000). Conclusion: These results support the 3-factor structure of the CAD Scale.