If you are not experienced with R, it is strongly advised that you read-up on and more importantly test out R and RStudio before attempting analyse road crash data with R.
To read up on R, we recommend reading Chapter 1 Getting Started with Data in R of the online book Statistical Inference via Data Science, which can be found here: https://moderndive.netlify.app/1-getting-started.html
Reading sections 1.1 to 1.3 of that book and trying a few of the examples are considered essential prerequisites, unless you are already experienced with R.
Optionally, if you want a more interactive learning environment, you can try getting started with online resources, such as those found at education.rstudio.com/learn.
And for more information on how R can be used for transport research, the Transportation chapter of Geocomputation with R is a good place to start: https://r.geocompx.org/transport.html
Your computer should also have the necessary software installed.
To ensure your computer is ready for the course, you should have a recent (3.6.0 or later) version of R or RStudio installed. You should have installed packages stats19, tidyverse and a few others shown below. To check you have the necessary packages installed, try running the following line of code:
source("https://git.io/JeaZH")
That does some basic checks. For more comprehensive checkes, and to get used to typing in R code, you can also test your setup by typing and executing the following lines in the RStudio console (this will install the packages you need if they are not already installed):
install.packages("remotes")
c(
pkgs ="pct", # package for getting travel data in the UK
"sf", # spatial data package
"stats19", # downloads and formats open stats19 crash data
"stplanr", # for working with origin-destination and route data
"tidyverse", # a package for user friendly data science
"tmap" # for making maps
)::install_cran(pkgs)
remotes# remotes::install_github("ITSLeeds/pct")
To test your computer is ready to work with road crash data in R, try running the following commands from RStudio (which should result in the map below):
library(stats19)
library(tidyverse)
library(tmap) # installed alongside mapview
get_stats19(year = 2022, type = "ac")
crashes = crashes %>%
crashes_iow = filter(local_authority_district == "Isle of Wight") %>%
format_sf()
# basic plot
plot(crashes_iow)
You should see results like those shown in the map here: https://github.com/ropensci/stats19/issues/105
If you cannot create that map by running the code above before the course, get in touch with us, e.g. by writing a comment under that github issue page (Note: You will need a github account).
Perhaps the most important pre-requisite is time. You’ll need to find time to work-through these materials, either in one go (see suggested agenda below) or in chunks of perhaps 1 hour per week over a 2 month period. I think ~8 hours is a good amount of time to spend on this course but it can be done in small pieces, e.g.:
For the more structured 2 day course for R beginners, a preliminary agenda is as follows:
09:00-09:30 Arrival and set-up
09:30-11:00 Introduction to the course and software
Break
11:15-12:30 Working with temporal data
Lunch
13:30-15:00 Working with spatial data
Spatial data in R
Context: spatial ecosystem (see section 1.4 of Geocomputation with R - package ecosystem)
Exercises: Section 6 of the handout
Further reading: Section 3.2 to 3.2.2 of handouts
Break
15:15-15:30 Talk on Road Safety 1
15:30-16:15 Practical - Applying the methods to stats19 data - a taster
16:15-16:30 Talk on Road Safety 2
09:30-11:00 Point pattern analysis
11:15-12:30 Road network data
Lunch
13:30-15:00 Analysing crash data on road network
Break
15:15-15:30: Talk on Road Safety 3
15:30-16:30 Applying the methods to your own data