Objective: To test the assumptions concerning the compensation of middle ear pressure (MEP) in using peak Ytm and the tympanogram asymmetry phenomenon, and to address the variability and reliability of 226 Hz tympanometry under negative MEP. Design: (1) Repeated measures in normal ears with experimentally-induced negative MEP; (2) Retrospective analysis of clinical data from ears with negative MEP. Study sample: Experimental data: 57 adult ears. Clinical data: 200 adult ears. Results: Mean peak Ytm significantly increased (˜ 0.10 mmhos) under negative MEPs (> − 190 daPa). TW increased, up to ˜50 daPa, for MEP ≤ − 130 daPa. Vea decreased when MEP ≤ − 190 daPa. With increasing negative MEP, peak Ytm and tympanogram gradient tended to decrease. The within-subject variation of all measures and difference between normal and negative MEP were small. Conclusions: The tympanometry procedure increases peak Ytm in ears with low to moderate negative MEP, and decreases gradient and Vea with high negative MEP. This is referred to as hypercorrection of the admittance and should be accounted for in clinical norms. Results imply that tympanogram asymmetry is due to divergent effects of air pressure on the middle ear. The variability of tympanometry does not increase and reliability not decrease under negative MEP.