Objective: The Beck Anxiety Inventory (BAI) is an established measure of anxiety in adults. Given that many items on the BAI measure somatic symptoms, questions remain regarding whether this instrument accurately assesses anxiety in older adult populations, due to the tendency for increased somatic concern in older adults. The Geriatric Anxiety Inventory (GAI), an instrument specifically designed to measure anxiety in older adults, places greater emphasis on cognitive and affective symptoms. Factor structure of the two measures was compared in order to assess the associations between domains on the BAI and GAI. Method: GAI and BAI data from a non-clinical sample of 60 older adults were examined. A previously established 4-factor model (Subjective, Neurophysiological, Autonomic, and Panic) of the BAI was used. These domains were compared to a 4-factor model (Gastrointestinal, Central Nervous System Hyperarousal, Excessive Worry, and Decision Making) of the GAI. Results: Analyses indicated that each factor of the GAI was most highly correlated with the subjective domain of the BAI (Gastrointestinal r = .67, p < .001; Hyperarousal r = .60, p < .001; Worry r = .46, p < .001; Decision r = .59, p < .001). While additional significant correlations were found, the association between the BAI subjective domain and each factor of the GAI demonstrated the strongest relationship. Conclusion: Results suggest that the GAI most accurately captures the subjective experience of anxiety in older adults based upon the emphasis on measurement of cognitive and affective symptoms, as opposed to the emphasis placed upon somatic symptoms in the BAI.