Table 3 The number of cities, at different onset risk levels, under two scenarios (i.e. with and without Wuhan lockdown measure) on the dates corresponding to Fig. 2eā€“j.

From: An extended Weight Kernel Density Estimation model forecasts COVID-19 onset risk and identifies spatiotemporal variations of lockdown effects in China

No. of cities

Onset risk

Wilcoxon signed-rank test, P (H0)

0ā€“0.2

0.2ā€“0.4

0.4ā€“0.6

0.6ā€“0.8

>0.8

25 January 2020 Lockdown

183

43

48

56

17

1.180ā€‰Ć—ā€‰10āˆ’58

Non-lockdown

124

58

26

98

41

30 January 2020 Lockdown

126

40

41

25

115

2.009ā€‰Ć—ā€‰10āˆ’37

Non-lockdown

101

35

23

28

160

5 February 2020 Lockdown

178

67

51

26

25

1.076ā€‰Ć—ā€‰10āˆ’57

Non-lockdown

126

49

39

86

47

  1. The rightmost column shows the P values of the Wilcoxon signed-rank test on paired onset risk values of a city under two scenarios on the same date. The null hypothesis of this test, H0, is that the median difference between the onset risk values under lockdown and non-lockdown scenarios is zero. At a normal significance level, H0 could be rejected, thus showing that the onset risk under the non-lockdown scenario is statistically significantly higher. Nā€‰=ā€‰347 samples on every date were use in each Wilcoxon signed-rank test.
Back to article page