The C factor is an important artificially controllable factor of soil erosion prediction model. It plays a vital role in reducing soil erosion and controlling water and soil loss. In this paper, the maize, a typical crop on sloping farmland of the Loess Plateau, was studied using an artificial rainfall simulation experiment. Besides, five change features of maize on near-surface conditions in different growth periods were analyzed and the C value was calculated according to the sediment yield of maize in different growth periods. The results showed that vegetation coverage, plant height and crusted thickness increased gradually with the prolongation of maize growth period and that the surface roughness decreased first and then increased with the prolongation of maize growth period. Besides, the sediment yield decreased gradually with the continuous growth of maize and sediment reduction benefit increased with the extension of maize growth period. Based on the previous C-value model calculated by vegetation coverage, the vegetation coverage was taken as the key factor and the plant height, soil crust and surface roughness as the adjusting factors to build a local C-value model. On the basis of that, a better C-value model for maize slope farmland (model R²=0.94, RMSE=0.017, MAE=0.014, NSE=0.992) was obtained. Based on the research results, a C-value calculation formula was established according to change characteristics of near-surface conditions. This helped improve the accuracy of C-value estimation and its applicability to the Loess Plateau, which could provide a scientific basis for improving the accuracy of soil erosion prediction model on the Loess Plateau.