Acoustic admittance testing was done on the ears of two groups of children. The first group was comprised of children undergoing myringotomy and tube surgery for treatment of chronic or recurrent otitis media, and the second group was comprised of hospital outpatients who were unscreened with respect to a history of middle ear disease and who were more representative of children in the general population. The admittance measures were then analyzed with respect to middle ear status (effusion versus no effusion) as determined by the surgeon (surgery group) and by validated otoscopy (outpatient group). Based on the data from the surgery group, the best admittance criteria for the identification of middle ear effusion were determined and then tested with respect to the data from subjects in the outpatient group. Sensitivity and specificity of several different criteria were similar. The role of prevalence of disease and the value of admittance measures such as gradient and acoustic reflex are discussed in the context of diagnostic testing in clinical or research protocols as well as in the context of immittance screening programs.