Abstract: Aiming at the problem of reasonable selection of cycle slip detection and repair methods in GNSS data processing, this paper evaluates and analyzes the performance of three common cycle slip detection and repair methods. A new method for evaluating the reliability of cycle slip detection results based on mathematical statistics is given. The new method is used to evaluate the reliability of cycle slip detection results obtained by polynomial fitting method and the TurboEdit method on the same set of dual-frequency real data, and the reliability of the cycle slip detection results obtained by the polynomial fitting method and the triple-frequency geometric-free combination method on the same set of three frequency measured data is analyzed.Based on the evaluation results of the two groups,it can be seen that the detection effect of the TurboEdit method is better than that of the polynomial fitting method for the dual-frequency real data while for the triple-frequency real data,the detection effect of the triple-frequency geometric free combination method is better than the polynomial fitting method;The polynomial fitting method can detect and repair cycle slips on dual-frequency and triple-frequency observation data, so it has a wider applicability.