Applying specific postcodes to UK regions in choropleth

Slavek · January 17, 2023, 3:14pm

Hi, I have this sample where I created a map showing one of the values based on a post code:

region.scores <- data.frame(
  stringsAsFactors = FALSE,
  NAME_3 = c("BB7","BB1","BB12","BB6",
             "BB9","BB3","BB10","PR7","DG1","CA11","CA4","DG2",
             "CA13","DG11","ST7","CW9","CW4","CW6","ST5","CW3",
             "SY13","WA16","PR4","PR2","PR7","PR1","FY8","PR3",
             "PR5","PR25","FY6","FY5","PR6","FY4","PR26","PR9",
             "L40","PR8","LA1","FY3","WN6","ST17","ST18",
             "ST16","ST15","WS15","WS12","ST19","WS11","TF10",
             "ST20","ST21","ST14","ST12","ST15"),
  RetentionA = c(0.717647058823529,
                 0.761194029850746,0.69811320754717,0.764705882352941,
                 0.606060606060606,0.628571428571429,0.827586206896552,
                 0.722222222222222,0.822222222222222,0.693877551020408,
                 0.714285714285714,0.631578947368421,0.695652173913043,
                 0.736842105263158,0.830769230769231,0.868421052631579,
                 0.740740740740741,0.789473684210526,0.730769230769231,
                 0.705882352941177,0.769230769230769,0.8125,0.819672131147541,
                 0.732673267326733,0.741379310344828,0.673076923076923,
                 0.689655172413793,0.84,0.833333333333333,0.682926829268293,
                 0.787878787878788,0.810810810810811,0.857142857142857,
                 0.64,0.666666666666667,0.695652173913043,0.8,0.5625,
                 0.733333333333333,0.823529411764706,0.75,
                 0.811764705882353,0.761194029850746,0.844827586206897,
                 0.736842105263158,0.6,0.769230769230769,0.576923076923077,
                 0.708333333333333,0.92,0.7,0.785714285714286,
                 0.764705882352941,0.684210526315789,0.733333333333333),
  RetentionB = c(0.465648854961832,
                 0.467889908256881,0.493333333333333,0.5,0.384615384615385,
                 0.448979591836735,0.571428571428571,0.393939393939394,
                 0.536231884057971,0.5,0.540540540540541,0.342857142857143,
                 0.470588235294118,0.451612903225806,0.650602409638554,
                 0.55,0.526315789473684,0.405405405405405,
                 0.558823529411765,0.444444444444444,0.416666666666667,
                 0.565217391304348,0.581395348837209,0.471337579617834,
                 0.425742574257426,0.406976744186047,0.512820512820513,0.56,
                 0.533333333333333,0.451612903225806,0.448275862068966,0.6,0.5,
                 0.4,0.315789473684211,0.432432432432432,0.375,
                 0.28125,0.354838709677419,0.482758620689655,
                 0.428571428571429,0.547619047619048,0.5,0.521276595744681,
                 0.518518518518518,0.288461538461538,0.408163265306122,
                 0.319148936170213,0.485714285714286,0.657142857142857,0.4375,
                 0.733333333333333,0.541666666666667,0.481481481481481,
                 0.523809523809524),
  Dealer = c("Blackburn","Blackburn",
             "Blackburn","Blackburn","Blackburn","Blackburn",
             "Blackburn","Blackburn","Carlisle","Carlisle","Carlisle",
             "Carlisle","Carlisle","Carlisle","Crewe","Crewe","Crewe",
             "Crewe","Crewe","Crewe","Crewe","Crewe","Preston",
             "Preston","Preston","Preston","Preston","Preston",
             "Preston","Preston","Preston","Preston","Preston",
             "Preston","Preston","Preston","Preston","Preston",
             "Preston","Preston","Preston","Stafford","Stafford",
             "Stafford","Stafford","Stafford","Stafford","Stafford",
             "Stafford","Stafford","Stafford","Stafford","Stafford",
             "Stoke","Stoke"),
  Cor1 = c(53.874,53.756,53.803,53.792,
           53.837,53.698,53.799,53.645,55.073,54.664,54.847,
           55.065,54.66,55.114,53.088,53.259,53.207,53.162,
           53.015,53,52.952,53.303,53.754,53.778,53.645,53.757,
           53.751,53.869,53.731,53.694,53.868,53.874,53.667,
           53.791,53.687,53.66,53.602,53.626,54.046,53.821,
           53.574,52.789,52.814,52.813,52.9,52.758,52.706,
           52.713,52.689,52.769,52.798,52.868,52.903,52.948,52.9),
  Cor2 = c(-2.386,-2.462,-2.29,-2.417,
           -2.214,-2.465,-2.219,-2.652,-3.58,-2.786,-2.833,
           -3.657,-3.366,-3.336,-2.265,-2.501,-2.347,-2.666,
           -2.237,-2.406,-2.689,-2.371,-2.833,-2.708,-2.652,
           -2.701,-3.002,-2.716,-2.656,-2.693,-2.981,-3.021,
           -2.609,-3.028,-2.74,-2.968,-2.841,-3.01,-2.799,-3.021,
           -2.669,-2.099,-2.081,-2.118,-2.148,-1.917,-1.999,
           -2.147,-2.017,-2.385,-2.263,-2.259,-1.868,-2.171,
           -2.148),
  PostCode2digits = c("BB","BB","BB","BB","BB",
                      "BB","BB","PR","DG","CA","CA","DG","CA","DG","ST",
                      "CW","CW","CW","ST","CW","SY","WA","PR","PR",
                      "PR","PR","FY","PR","PR","PR","FY","FY","PR","FY",
                      "PR","PR","L4","PR","LA","FY","WN","ST","ST",
                      "ST","ST","WS","WS","ST","WS","TF","ST","ST","ST",
                      "ST","ST"),
  Region = c("North West","North West",
             "North West","North West","North West","North West",
             "North West","North West","Scotland","North West",
             "North West","Scotland","North West","Scotland",
             "West Midlands","North West","North West","North West",
             "West Midlands","North West","Yorkshire and the Humber",
             "North West","North West","North West","North West",
             "North West","North West","North West","North West",
             "North West","North West","North West","North West",
             "North West","North West","North West","North West","North West",
             "North West","North West","North West",
             "West Midlands","West Midlands","West Midlands","West Midlands",
             "West Midlands","West Midlands","West Midlands",
             "West Midlands","West Midlands","West Midlands",
             "West Midlands","West Midlands","West Midlands","West Midlands"),
  City = c("Blackburn","Blackburn",
           "Blackburn","Blackburn","Blackburn","Blackburn",
           "Blackburn","Preston","Dumfries","Carlisle","Carlisle",
           "Dumfries","Carlisle","Dumfries","Stoke on Trent","Crewe",
           "Crewe","Crewe","Stoke on Trent","Crewe","Sheffield",
           "Warrington","Preston","Preston","Preston","Preston",
           "Blackpool","Preston","Preston","Preston",
           "Blackpool","Blackpool","Preston","Blackpool","Preston",
           "Preston","Liverpool","Preston","Lancaster","Blackpool",
           "Wigan","Stoke on Trent","Stoke on Trent",
           "Stoke on Trent","Stoke on Trent","Walsall","Walsall",
           "Stoke on Trent","Walsall","Telford","Stoke on Trent",
           "Stoke on Trent","Stoke on Trent","Stoke on Trent",
           "Stoke on Trent")
)

region.scores

library(tidyverse)
library(sf)

#Download UK postcode polygon Shapefile
download.file(
  "http://www.opendoorlogistics.com/wp-content/uploads/Data/UK-postcode-boundaries-Jan-2015.zip",
  "postal_shapefile"
)
unzip("postal_shapefile")

uk <- read_sf('Distribution/Districts.shp')

uk.all <- left_join(uk, region.scores, by = c('name'='NAME_3'))
uk.all

uk.all %>% filter(!is.na(RetentionA))

ggplot() +
  geom_sf(data = uk.all, alpha = 0.3) +
  geom_sf(data = uk.all, aes(fill = RetentionA)) +
  scale_fill_gradient(low = "red", high = "green") +
  labs(title = "Retention Scores by region",
       fill = "Option A") +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5, size = 20),
        axis.title = element_blank(),
        plot.background = element_blank(),
        legend.position = c(0.10, 0.20),
        plot.margin = margin(5, 0, 5, 0))

rs_sf <- st_as_sf(region.scores, coords=c('Cor2','Cor1'), crs=st_crs(uk.all))

ggshopA <- ggplot()+
  geom_sf(data=uk.all,mapping=aes(fill=RetentionA)) +
  geom_sf(data=rs_sf,mapping=aes(shape=Dealer))+
  scale_fill_gradient(low = "red", high = "green") +
  coord_sf(xlim=c(-4,-2),ylim=c(52.6,55.15))

ggshopA

Is it possible to apply colours to bigger areas (UK Regions)?
I guess I should calculate averages for all four regions (North West, Scotland, West Midlands and Yorkshire and the Humber but how can I change a map to show colours for each region? Basically, the entire Scotland should be red as the average is the lowest, the entire Yorkshire and the Humber should be green as the average is the highest.

Is it easy to change the existing code?

nirgrahamuk · January 17, 2023, 4:53pm

provide a fill and colour mapping to region; or some variable which has the same value across a given region; unlike RetentionA which varies within region.

Slavek · January 17, 2023, 5:24pm

ok

stage1 <- region.scores %>%
  group_by(Region) %>%
  summarise_at(.vars = vars(ends_with(match = "A")),.funs = list(Sc = ~mean(.,na.rm=TRUE), Count = ~sum(!is.na(.))))
stage1

but then? How can I link it with the map?

nirgrahamuk · January 17, 2023, 5:26pm

your geometery is on uk.all so you need your data joined to that. 'provide a fill and colour mapping to region' means to use aes(fill=, colour=) on the variable of choice

Slavek · January 18, 2023, 11:09am

A variable containing post codes is linked with the UK map and I don't know how to link regions.

nirgrahamuk · January 18, 2023, 11:49am

NAME_3 on your region scores looks like at least the start of postcodes to me

system · March 1, 2023, 11:50am

This topic was automatically closed 42 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.