Purpose <p>Total laryngectomy (TL) is a referral procedure useful in case of locally advanced laryngeal cancer but is difficult to train during residency. The objective of this work was to assess the dead porcine model for training in TL by studying its face, content, and construct validity.</p> Methods <p>Young residents (postgraduate year (PGY) 1 to 3), experienced residents (PGY-4 to 6) and senior surgeons (SS) trained TL on dead swine. Face and content validity were assessed by questioning the perception of realism and the interest of this model for teaching. Construct validity was assessed by comparing the mean Ottawa Surgical Competency Operating Room Evaluation (O-SCORE), obtained during the simulation sessions, and the mean modified Objective Structured Assessment of Technical Skills (OSATS) score, obtained by independent blind video analysis, between PGY1-3, PGY4-6 and SS groups.</p> Results <p>Twenty-three individuals trained TL on dead swine. The mean OSATS score were 11.3, 14.9 and 22 for young residents, experienced residents and SS respectively (<i>p</i> &lt; 0.05). The mean O-SCORE were 3 for young residents, 3.8 for experienced residents and 4.95 for SS (<i>p</i> &lt; 0.05). 89% of the participants agreed or strongly agreed that the steps resembled those of a TL in humans. SS strongly agreed that the model was a useful training tool, may improve performance and should be used to train students.</p> Conclusion <p>Dead swine is a valid option to train TL steps as it offers strong face, content, and construct validity.</p>