Skip to main content

Table 5 Spearman correlation coefficients for each wealth index variation calculated using polychoric PCA (a model using smartphone and data dummy variables only could not reach convergence using polychoric 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

Full model

HH expenditure

1

     

Base model

0.4886

1

    

+ Number of cells

0.4931

0.9917

1

   

+ smartphone dummy

0.4897

0.9937

0.9863

1

  

+ Number of smartphones

0.4945

0.9883

0.9976

0.9898

1

 

Full model

0.4944

0.9851

0.9952

0.9887

0.9985

1