Abstract: This paper studies significant differences in functional connectivity between schizophrenia patients and healthy controls by different methods. Firstly, three different brain functional networks are constructed by Pearson correlation coefficient, partial correlation coefficient and wavelet correlation coefficient, and then the paper uses risk difference to find out those significant differences. Finally, a comparative analysis is made. The result shows that common and abnormal connectivity exists in different networks where functional connectivity between the precuneus and posterior cingulate cortex weakens significantly among schizophrenia patients compared with controls. And more notably, the sign of some statistic appears in some significant functional connectivity is completely contrary. That is to say, as opposed to controls this functional connectivity in patients is both strengthened and weakened simultaneously. It shows that statistical test makes the results more persuasive as searching for abnormal functional connectivity in brain networks in terms of statistics.