Working With cancensus

library(griddy)
library(dplyr)

This article is a website-only template for Canadian Census workflows: it requires live CensusMapper access, so it is rendered on the pkgdown site when the API key is set and skipped otherwise. It does not ship with the installed package.

library(cancensus)

## Census data is currently stored temporarily.
## 
##  In order to speed up performance, reduce API quota usage, and reduce unnecessary network calls, please set up a persistent cache directory via `set_cancensus_cache_path('<local cache path>', install = TRUE)`.
##  This will add your cache directory as environment varianble to your .Renviron to be used across sessions and projects.

library(sf)

## Warning: package 'sf' was built under R version 4.5.2

## Linking to GEOS 3.13.0, GDAL 3.8.5, PROJ 9.5.1; sf_use_s2() is TRUE

library(sfdep)

regions <- list(CSD = "5915022") # Vancouver CSD
vectors <- c(
  income_2016 = "v_CA16_2397",
  income_2021 = "v_CA21_906"
)

panel_2016 <- get_census(
  dataset = "CA16",
  regions = regions,
  vectors = vectors["income_2016"],
  level = "CT",
  geo_format = "sf"
) |>
  transmute(GeoUID, year = 2016L, median_income = income_2016, geometry)

## Census data is currently stored temporarily.
## 
##  In order to speed up performance, reduce API quota usage, and reduce unnecessary network calls, please set up a persistent cache directory via `set_cancensus_cache_path('<local cache path>', install = TRUE)`.
##  This will add your cache directory as environment varianble to your .Renviron to be used across sessions and projects.

## Census data is currently stored temporarily.
## 
##  In order to speed up performance, reduce API quota usage, and reduce unnecessary network calls, please set up a persistent cache directory via `set_cancensus_cache_path('<local cache path>', install = TRUE)`.
##  This will add your cache directory as environment varianble to your .Renviron to be used across sessions and projects.

## Querying CensusMapper API...

## Census data is currently stored temporarily.
## 
##  In order to speed up performance, reduce API quota usage, and reduce unnecessary network calls, please set up a persistent cache directory via `set_cancensus_cache_path('<local cache path>', install = TRUE)`.
##  This will add your cache directory as environment varianble to your .Renviron to be used across sessions and projects.

## Querying CensusMapper API...

panel_2021 <- get_census(
  dataset = "CA21",
  regions = regions,
  vectors = vectors["income_2021"],
  level = "CT",
  geo_format = "sf"
) |>
  transmute(GeoUID, year = 2021L, median_income = income_2021, geometry)

## Census data is currently stored temporarily.
## 
##  In order to speed up performance, reduce API quota usage, and reduce unnecessary network calls, please set up a persistent cache directory via `set_cancensus_cache_path('<local cache path>', install = TRUE)`.
##  This will add your cache directory as environment varianble to your .Renviron to be used across sessions and projects.
## 
## 
## Querying CensusMapper API...

## Census data is currently stored temporarily.
## 
##  In order to speed up performance, reduce API quota usage, and reduce unnecessary network calls, please set up a persistent cache directory via `set_cancensus_cache_path('<local cache path>', install = TRUE)`.
##  This will add your cache directory as environment varianble to your .Renviron to be used across sessions and projects.

## Querying CensusMapper API...

common_ids <- intersect(panel_2016$GeoUID, panel_2021$GeoUID)
panel <- bind_rows(panel_2016, panel_2021) |>
  filter(GeoUID %in% common_ids)
geography <- panel |>
  filter(year == 2021L) |>
  arrange(GeoUID) |>
  mutate(
    nb = st_contiguity(geometry, queen = TRUE),
    wt = st_weights(nb, allow_zero = TRUE)
  )

## Warning: There were 2 warnings in `stopifnot()`.
## The first warning was:
## ℹ In argument: `nb = st_contiguity(geometry, queen = TRUE)`.
## Caused by warning in `spdep::poly2nb()`:
## ! some observations have no neighbours;
## if this seems unexpected, try increasing the snap argument.
## ℹ Run `dplyr::last_dplyr_warnings()` to see the 1 remaining warning.

classes <- classify_dynamics(panel, GeoUID, year, median_income, k = 5)
classic <- markov_dynamics(classes, GeoUID, year, class)
spatial <- spatial_markov(panel, GeoUID, year, median_income, geometry = geography, k = 5)

plot_transition_matrix(classic)

plot_spatial_markov(spatial)

Canadian census geographies can change between cycles. A production workflow should harmonize geography before treating units as a longitudinal panel.