README.md

In BingoLaHaye/DCmapR: Easy mapping of DC Wards and Precincts

How to DCmapR

Edward ‘Bingo’ LaHaye 2019-04-09

Introduction

Hi and welcome to my first package, DCmapR. This package was first created in response to the 2nd annual hackathon at The Catholic University of America School of Engineering. Me and my team ran into the problem of plotting our points among the different wards and precincts of DC, and so this package was born from that struggle. This package will help you do a couple of things:

1. Get the shape files for the wards and precincts in DC

2. Get the data.frame coordinates and labels for all of the wards and precincts

3. Calculate the centroids for both wards and precincts so that you can easily label them

4. Create detailed and beautiful graphs of Washington, DC!

These objectives will be accomplished through each of their own functions that I will walk you through.

get_Ward

This is a pretty basic function that gives you either the shape data or the dataframe data. To select the dataframe output just put dataframe = TRUE as the argument, or you can leave it blank.

library(DCmapR)
# For example lets get the dataframe for wards
WardsDF <- get_Ward(dataframe = TRUE)
# And you can easily plot this using ggplot2
mapOG <- ggplot() +
  geom_polygon(data = WardsDF, aes(x = long, y = lat, group = group, fill = factor(Ward)),
               col  = "black", alpha = 0.3, size = 1) + 
  scale_fill_discrete(name = "Ward") + 
  coord_quickmap()
mapOG

And if you want to you can use the shape file in the same way for plotting, or you can also extract the data from it using data.frame on it. You can plot the SF as well it just can be a bit difficult in the naming conventions, so i recommend sticking with the dataframe.

WardsSF <- get_Ward(dataframe = FALSE)
WardsSFData <- data.frame(WardsSF)
# but you can combine this for some nice maps :)
library(dplyr)
biggerset <- WardsDF %>%
  full_join(WardsSFData, by = c("Ward" = "WARD"))
#You can then use this to make a nice little cloropleth map
plot1 <- ggplot() +
  geom_polygon(data = biggerset, aes(x = long, y = lat, group = group, 
               fill = as.numeric(as.character(POP_2010))), 
               col  = "black", alpha = 0.6, size = 1) +
  scale_fill_continuous(name = "Population \nin 2010") +
  theme(legend.title = element_text(size = 10)) + 
  coord_quickmap()
plot2 <- ggplot() +
  geom_polygon(data = biggerset %>% group_by(Ward), aes(x = long, y = lat, group = group, 
               fill = as.numeric(as.character(UNEMPLOYME))), 
               col  = "black", alpha = 0.6, size = 1) +
  scale_fill_gradient(name = "Number of \nUnemployed", low = "blue", high = "red") +
  theme(legend.title = element_text(size = 10)) +
  coord_quickmap() 
plot1
plot2

Now as you can see from this, we can make nice looking graphs but we need labels for them! And my solution comes with the next function: get_centroid

get_centroid

get_centroid came about to help with labeling and getting those labels into a pleasing spot for the wards. I wanted to get into the center for each shape so I used gCentroid from rgeos to get them. You can get the centroids of either the Wards or Precincts depending if you select TRUE or FALSE for one of the options. Selecting neither will get you nothing. Now lets get these labels on there!

Wardlabs <- get_centroid(Ward = TRUE)
# And then you just add it in!
plot1 + 
  geom_text(data = Wardlabs, aes(x, y, label = Ward), size = 5)

plot2 + 
  geom_text(data = Wardlabs, aes(x, y, label = Ward), size = 5)

Now there is a bit more information for the maps! The next step is to dive deeper into these wards and look into the precincts with in them.

get_Precinct

This function was designed to get a better perspective on the wards themselves and how their parts added up to the whole of DC. You can use this function to give more detail to the map as a whole, or filter through to get more detail on a singular ward. get_precinct contains only one parameter dataframe = TRUE, which you can change to false for the shapefile.

PrecinctDF <- get_Precinct(dataframe = TRUE)
#Take the first map we made and layer onto it
mapOG + 
  geom_polygon(data = PrecinctDF, aes(x = long, y = lat, group = group), inherit.aes = FALSE, fill = NA, colour = 'black') + 
  ggtitle("DC with Wards and Precincts")

Future Features

Along the development I came accross some roadblocks in zooming in with the precincts and wards as they don’t have labels for which Ward they belong in. So right now we are working on catagorizing all of the precincts so that they word properly with wards if you so wish to zoom in on each ward.

#Trim to ward 8, I'm not sure why its polygon number is 1
PrecinctSF <- get_Precinct(dataframe = FALSE)
Ward1SF <- WardsSF[1,]
Just1 <- PrecinctSF[Ward1SF, ]
ggplot() + 
  geom_polygon(data = Ward1SF, aes(x = long, y =  lat, group = group)) + 
  coord_quickmap()

#Use over to get the overlapping datapoints
PrecinctSF <- get_Precinct(dataframe = FALSE)
ggplot() + 
  geom_polygon(data = Ward1SF, aes(x = long, y =  lat, group = group), fill = "white") + 
  geom_polygon(data = Just1, aes(x= long, y = lat, group = group), fill = NA, col = "black") +
  coord_quickmap() + 
  ggtitle("Ward 8 with Precincts")

A refined precincts dataset will be in the next version of DCmapR. As you can see it gets pretty close, and with some tweaking could turn into a nice chloropleth map.

Conclusion

Overall I hope you enjoy using this package as I enjoyed developing it. All info was acquired from (via), so go get exploring that data as well if you are interested.

BingoLaHaye/DCmapR documentation built on May 29, 2019, 7:20 a.m.