Table 4 Spearman correlation coefficients for each wealth index variation calculated using PCA
From: Is the lack of smartphone data skewing wealth indices in low-income settings?
HH expenditure | Base model | + Number of cells | + Smartphone dummy | + Number of smartphones | + Smartphone and data dummies | Full model | |
|---|---|---|---|---|---|---|---|
HH expenditure | 1 | ||||||
Base model | 0.377 | 1 | |||||
+ Number of cells | 0.4159 | 0.9901 | 1 | ||||
+ smartphone dummy | 0.3981 | 0.9963 | 0.9955 | 1 | |||
+ Number of smartphones | 0.451 | 0.9554 | 0.9835 | 0.9753 | 1 | ||
+Smartphone & data dummies | 0.4215 | 0.9829 | 0.9936 | 0.9945 | 0.9893 | 1 | |
Full model | 0.4651 | 0.9299 | 0.9661 | 0.9551 | 0.9958 | 0.9765 | 1 |