In this paper, different methodologies to characterize the flotation rate distribution at laboratory batch scale are presented. Conventional model fitting is compared with a selectively chosen set of one-parameter models to describe the flotation rate distribution F(k). In addition, a numerical inversion of the Laplace Transform, which is applicable to first-order systems, is shown. A wide range of time-recovery curves and probability density functions (PDF) for F(k) are presented based on 85 flotation tests. The Single Rate Constant, Rectangular, Exponential, Rayleigh, Unilateral Raised Cosine and chi(2) models are applied to fit the time-recovery curves. The best model per test is chosen and compared to the model fitting obtained from the Gamma distribution, showing a good agreement between these two approaches. The one-parameter PDFs are compared with F(k) estimates obtained from a regularized Inverse Laplace Transform. In this approach, no assumptions on the F(k) shape are made. The flotation rate distributions estimated from the Laplace Inversion show similar shapes to those obtained from the Gamma and the optimal one-parameter approach. In addition, the F(k) estimates present positive skews in all the evaluated tests. The use of the Laplace Transform Inversion along with conventional model fitting is a powerful tool to validate the F(k) estimates at laboratory scale.