Leaflet map 'Error in polygonData.default(data) : Don't know how to get path data from object of class data.frame'

Question

I'm trying to create a leaflet map of census data.

I have a csv file contains the number of deaths across Ireland. The data consists of the year($Year), the Sex ($Sex), the location($NAME_1) and the age at death ($Age.at.Death)

Likewise I have a dbf file that contains the location ($NAME_1) and the geometry needed for polygons to create the map.

All of the data merges correctly into DEATHS1, but as soon as I pass this to leaflet I get the error 'Error in polygonData.default(data) : Don't know how to get path data from object of class data.frame'

library(leaflet)
library(sf)
library(dplyr)

DEATHS <- read.csv("VSA07.csv")

Shape <- read_sf("gadm36_IRL_1.dbf")


MergedData = inner_join(DEATHS,Shape) #inner joins by NAME_1 column, works okay.

DEATHS1 <- filter(MergedData, Year == "2007", Age.at.Death == "All ages", Sex == "Both sexes")
# Want a map satisfying the above conditions only for the moment, also works as intended

DEATHS1


bins <- c(0, 500, 1000, 2000, 3000, 4000, Inf)
pal <- colorBin("YlOrRd", domain = DEATHS1$VALUE, bins = bins)


m <-leaflet() %>%
  addProviderTiles(providers$Stamen.TonerHybrid) %>% 
  addPolygons(data = DEATHS1)
m

The strange thing is that leaflet can plot a map based on MergedData but as soon as I apply any kind of filters through dplyr, ie DEATHS1, I get the error. So I don't think its anything to do with it being a data frame as MergedData is exactly that.

Any help is greatly appreciated.

A snapshot of the data DEATHS

     ï..Statistic Year      NAME_1        Sex      Age.at.Death   UNIT VALUE
1   Deaths Occurring 2007       State Both sexes      Under 1 year Number   230
2   Deaths Occurring 2007       State Both sexes       1 - 4 years Number    54

Shape

A tibble: 26 x 11
   GID_0 NAME_0  GID_1  NAME_1  VARNAME_1     NL_NAME_1 TYPE_1     ENGTYPE_1 CC_1  HASC_1                                    geometry
   <chr> <chr>   <chr>  <chr>   <chr>         <chr>     <chr>      <chr>     <chr> <chr>                           <MULTIPOLYGON [°]>
 1 IRL   Ireland IRL.1~ Carlow  Ceatharlach   NA        Administr~ County    NA    IE.CW  (((-6.592422 52.7081, -6.592915 52.7073, -~
 2 IRL   Ireland IRL.2~ Cavan   An Cabhán     NA        Administr~ County    NA    IE.CN  (((-7.275174 53.78388, -7.275377 53.78377,~
 3 IRL   Ireland IRL.3~ Clare   An Clár       NA        Administr~ County    NA    IE.CE  (((-9.731943 52.64875, -9.731943 52.64903,~

DEATHS1

       ï..Statistic Year    NAME_1        Sex Age.at.Death   UNIT VALUE GID_0  NAME_0    GID_1              VARNAME_1 NL_NAME_1            TYPE_1 ENGTYPE_1 CC_1 HASC_1                       geometry
1  Deaths Occurring 2007   Wicklow Both sexes     All ages Number   819   IRL Ireland IRL.26_1  Administrative County    County <NA>  IE.WW MULTIPOLYGON (((-6.109166 5...

Any help is much appreciated/

Answer 1

Issue is that when merging an sf file (the .dbf file) with the csv file data, R secretly converts the output to a tibble, thus leaflet loses the ability to identify the 'geometry' column without assistance.

To fix it you must explicitly tell leaflet where the geometry column is as shown below %>% addPolygons(data = MergedDataFilter$geometry)

m

m <-leaflet() %>%
  addProviderTiles(providers$Stamen.TonerHybrid) %>% 
  setView(lng = -7.4653, lat = 53.5345, zoom = 6.95) %>%
  addPolygons(data = DEATHS1$geometry,
              weight = 1,
              color = "#660000",
              fillColor = pal(DEATHS$VALUE))
#m

Answer 2

You need to convert the data.frame back to an object with sf class.

library(sf)

DEATHS1 <- sf::st_as_sf(DEATHS1)

m <-leaflet() %>%
  addProviderTiles(providers$Stamen.TonerHybrid) %>% 
  addPolygons(data = DEATHS1)
m

It should work.