Important facilities constructed during last decades of 20th century are near completion of design life. For extending their service life or to evaluate these for new demands (loads), assessment of strength of concrete in existing structure becomes necessary, a task generally performed with nondestructive tests (NDT); ultrasonic pulse velocity (USPV) and rebound hammer being most commonly executed. Compressive strength is estimated using empirical expressions relating NDT to partially destructive tests (PDT) such as core test. For the development of structure-specific expressions, results of adequate number (depending on variability and desired confidence level) of PDT are essential but these might not be available due to operational constraints. Correlation expressions from literature could be used in such cases but having been developed for different ingredients, curing regimes, and environmental exposure conditions, there would be associated uncertainties. A practical method for the estimation of these uncertainties is not readily available in the literature. This article proposes the statistical approach of resampling for quantifying the uncertainty of indirect strength estimates using expressions from literature. Parametric (probability distribution) and nonparametric (bootstrap) tools are employed and demonstrated with a case study from India. Both parametric and nonparametric approaches could capture across-member variability whereas overall uncertainty incorporation, as well as repeatability, was better in nonparametric approach. Parametric approach is traditionally used and well accepted by practitioners in contrast to nonparametric methods, which have certain advantages. The detailed methodology enumerated in the article would be very useful for practitioners across the world.