Many researchers have incorporated the effects of rock fragments in soil erosion studies to improve the prediction of erosion by raindrop impact and overland flow. According to the Box-Simanton approach, the effect of rock fragments should be incorporated into the crop factor of USLE, C, whereas the Poesen-McCormack and Sepaskhah et al. approaches include the rock fragments in the soil erodibility factor K. For this study we investigated which approach was most suitable for our research basins. This study is based on a comparison of observed sediment concentration data out of a representative agricultural watershed in the south ofIranwith the output of ANSWERS model. Preliminary results reveal that there is no meaningful statistical difference between the Poesen-McCormack and Box-Simanton approaches. Nevertheless, when the runoff coefficient exceeded 0.3, the Poesen-McCormack approach was more accurate, but under high antecedent soil moisture conditions, the Box-Simanton approach gave more accurate results. Finally, a comparison of the Sepaskhah et al. approach with other methods showed that, in general, the Sepaskhah et al. method is more practical and reliable than the other approaches.