Overweight and obesity are major risk factors for noncommunicable diseases. Their prevalence must be regularly assessed in order to best adapt public health policies and prevention programs. Epidemiological surveillance studies rely on either measured (gold standard) or self-reported data. The objective of this study was to estimate the discrepancies between self-reported and measured anthropometric data in the general French population, to identify the factors associated with these discrepancies, and to assess any changes over time. Analyses were conducted on a sample of 2,429 adults aged 18 to 74 years, included in the study on health, environment, biomonitoring, physical activity, and nutrition (Esteban 2014–2016), and for whom both self-reported and measured anthropometric data were available. Simple and multivariate linear regressions were performed to determine the associations between the body mass index (BMI) discrepancy and various sociodemographic variables. The results were compared with those previously obtained in the National Nutrition and Health Survey (ENNS 2006–2007). In general, in 2016, men and women underreported their weight and overreported their height, leading to an underestimation of BMI of approximately -0.41 for men and -0.79 for women. As in 2006, the difference in BMI between self-reported and measured data was significantly associated with measured BMI, with larger differences observed among overweight or obese individuals. In 2016, these differences also varied by marital status, which was not the case previously. The results highlight the existence of an underestimation of the prevalence of obesity when using self-reported weight and height data. The differences between self-reported and measured data and the associated factors vary over time, which justifies the implementation of effective and regular anthropometric measurements in epidemiological surveillance surveys.