Table 2

Univariable and multivariable associations for presence of retailer and retailer density by sociodemographic indicators

Model 1Model 2Model 3Model 4
Univariable logistic regression for presence of retailerUnivariable linear regression for log of retailer densityMultivariable logistic regression for presence of retailerMultivariable linear regression for log of retailer density
VariableEstimate95% CIP valueEstimate95% CIP valueCohens fEstimate95% CIP valueEstimate95% CIP valueCohens f
Median income, €10000.8660.834 to 0.9<0.0010.9070.896 to 0.918<0.0010.4960.5420.481 to 0.608<0.0010.9560.929 to 0.9840.0020.506
% in lowest income category1.0711.041 to 1.103<0.0011.0591.051 to 1.068<0.0010.4220.8840.836 to 0.933<0.0011.0281.014 to 1.042<0.0010.068
% unemployed1.1861.142 to 1.234<0.0011.0371.027 to 1.047<0.0010.2141.0631.01 to 1.120.0200.9760.964 to 0.987<0.0010.117
% with higher education1.0201.007 to 1.0330.0020.9740.971 to 0.978<0.0010.4331.1391.102 to 1.178<0.0010.9820.976 to 0.989<0.0010.144
Population density, 1000 s per square kilometre2.9031.843 to 4.909<0.0011.0331.004 to 1.0630.0230.068
  • Estimates are exponentiated and can be interpreted as multiplicative differences in OR and density, given 1 unit change in predictor.