In epidemiology studies, researchers may choose to use summary scores of hearing or visionbased on self-report or clinical measures. Self-report and clinical measures may yield differentclassifications. Of the possible clinical measures, there is no consensus regarding which puretonethreshold average (PTA) or visual acuity (VA) measures are optimal. We aimed todetermine how well different PTAs and VA measures predicted self-reported measures ofsensory function. A cross-sectional analysis of 30,097 Canadians aged 45-85 years participatingin wave 1 of the Canadian Longitudinal Study on Aging in 2012-2015 was performed. Wecalculated the area under the receiver operating characteristic curves (AUC) for 9 different PTAsand 6 different VA measures. In the analysis of the PTAs, the classifiers used as comparatorsincluded self-reports of hearing and hearing aid use. In the analysis of the VA measures,comparators were self-reports of vision, and corrective lens use. The top-ranked PTA was thebinaural mid-frequency PTA (i.e., the average of hearing thresholds at 1000, 2000, 3000 and4000 Hz in both ears). The top-ranked VA measure was the average pinhole-corrected VA inboth eyes. These measures are not commonly used, but should be considered for epidemiologicalresearch.