Background Advances in technology have made it possible to examine real-world driving using naturalistic data obtained from in-vehicle monitoring devices. These devices overcome the weaknesses of self-report methods and can provide comprehensive insights into driving exposure, habits and practices of older drivers. Aim The aim of this study is to compare self-reported and objectively measured driving exposure, habits and practices using a travel diary and an in-vehicle driver monitoring device in older drivers with bilateral cataract. Methods A cross-sectional study was undertaken. Forty seven participants aged 58–89 years old (mean = 74.1; S.D. = 7.73) were recruited from three eye clinics over a one year period. Data collection consisted of a cognitive test, a researcher-administered questionnaire, a travel diary and an in-vehicle monitoring device. Participants’ driving exposure and patterns were recorded for one week using in-vehicle monitoring devices. They also completed a travel diary each time they drove a motor vehicle as the driver. Paired t-tests were used to examine differences/agreement between the two instruments under different driving circumstances. Results The data from the older drivers’ travel diaries significantly underestimated the number of overall trips (p < 0.001), weekend trips (p = 0.002) and trips during peak hour (p = 0.004). The travel diaries also significantly overestimated overall driving duration (p < 0.001) and weekend driving duration (p = 0.003), compared to the data obtained from the in-vehicle monitoring devices. No significant differences were found between instruments for kilometres travelled under any of the driving circumstances. Conclusions The results of this study found that relying solely on self-reported travel diaries to assess driving outcomes may not be accurate, particularly for estimates of the number of trips made and duration of trips. The clear advantages of using in-vehicle monitoring devices over travel diaries to monitor driving habits and exposure among an older population are evident.