The purpose of this study was to assess the accuracy, in older adults, of questions assessing hearing loss. Study participants (n=3,556), aged 48-92 years, were examined in a population-based study of age-related hearing loss in Beaver Dam, Wisconsin. Self-report data from the ten-question Hearing Handicap Inventory for the Elderly-screening Version (HHIE-S), and four additional questions were compared with hearing loss as measured by pure-tone air conduction audiometry. The single question, 'Do you feel you have a hearing loss?' was the most sensitive question (sensitivity=71 per cent); its overall and gender-specific prevalence estimates were within 3.2 per cent of prevalence measures derived audiometrically, although age-group specific estimates were not as accurate. Using an HHIE-S total score >8 resulted in low sensitivity (34 per cent) and inaccurate prevalence estimates. These results indicate that, for some applications, one simple question may be sufficient for prevalence surveys of hearing loss among older adults.