Adopting the autocorrelation method in the ionospheric short-term forecasting, we put forward a simple and practical forecasting method―the sectional autocorrelation method, that is, for predictions of one hour to four hours ahead the autocorrelation coefficient of RDF with the "iteration" method is selected, for prediction of more than four hours ahead, the autocorrelation coefficient of f0F2 with the "at once" method is used. The prediction precisions have been quantitatively estimated based on the data from Chongqing and Guangzhou Ionosonde Stations. It is shown that the method is much improved for the predictions of one hour to four hours ahead. For the predictions of more than four hours ahead the prediction error reaches a saturation value, which is still lower than that of the "median" method. This new method could also be applied to the short-term forecasting of other ionospheric parameters.