GroceryStoreHamiltonData-vignette

Load packages

The sf package is used to handle vector spatial data.

library(GroceryStoreHamiltonData)
library(sf)
#> Linking to GEOS 3.12.1, GDAL 3.8.4, PROJ 9.3.1; sf_use_s2() is TRUE

grocery_DA data

Load and plot the grocery_DA vector data. This dataset contains the geometry of all the census tracts in Hamilton and includes two variables: the population count in 2021 and the grocery store counts in each census tract.

data("grocery_DA")
plot(grocery_DA[ ,"v_CA21_1..Population..2021"],
     main = "2021 Population Count")

plot(grocery_DA[ ,"Freq"],
     main = "Grocery Store Count")

An OLS simple linear regression of these two variables

lm1 <- lm(Freq ~ v_CA21_1..Population..2021, data = grocery_DA)
summary(lm1)
#> 
#> Call:
#> lm(formula = Freq ~ v_CA21_1..Population..2021, data = grocery_DA)
#> 
#> Residuals:
#>     Min      1Q  Median      3Q     Max 
#> -0.4473 -0.0986 -0.0917 -0.0872  3.8979 
#> 
#> Coefficients:
#>                             Estimate Std. Error t value Pr(>|t|)   
#> (Intercept)                6.691e-02  2.135e-02   3.134  0.00178 **
#> v_CA21_1..Population..2021 4.985e-05  2.669e-05   1.868  0.06211 . 
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> Residual standard error: 0.3836 on 889 degrees of freedom
#> Multiple R-squared:  0.003909,   Adjusted R-squared:  0.002789 
#> F-statistic: 3.489 on 1 and 889 DF,  p-value: 0.06211

Based on this simple regression (p-value = 0.06 > 0.05), the population count is not a significant predictor of the grocery store count in Hamilton.