A non-official template to write your PhD. dissertation at the University of British Columbia using R and rmarkdown
This repository provides a template for writing a PhD dissertation in R Markdown, and rendering those files into a PDF formatted according to the requirements of the Faculty of Graduate and Postdoctoral Studies of the University of British Columbia. Note that this is not an official template and thus, does not guarantee a successful submission. It follows the 2021 requirements to convert R Markdown files into a PDF formatted ready for submission at UBC. The faculty of G+PS has a relative flexible format so, as a reference, the current template will generate a PDF similar to Palacios-Abrantes, 2021 (You can also see the dissertation_template.pdf file). This project has drawn directly on code and ideas from Dan Ovando's gauchodown, with the modifications needed to deal with UBC's G+PS requirements. However, unlike gouchodown, this is not a bookdown-like package but a simple repository that you download and modify with your information.
Currently, the repository only renders a fully edited PDF as required by G+PS and I have no plans to expand this in the future. The word version is useful for revisions and keeping track changes, but it does not render a final formatted version.
This repository is inteded for those with some experience using R, R-projects and rmarkdown with child chunks, but no LaTex knowledge is required. I provide very basic information on bookdown and rmarkdown here, if you are new to working with these packages, please read over the documentation available in bookdown book. There are also numerous rmarkdown tuttorials online like those offered by the UBC IOF-based Mozilla study group Deep Sea Data Squad or the UCSB-based eco-data-science. See also the Related projects section at the bottom of this page.
You have the following two options for using this repository as a template for your UBC dissertation;
- A) If you have a GitHub account;
forkthe repo to your own github and work from there - B) If you do not have a GitHub account; there is a green button in the upper right part of this page that says "Code". Click it and download the ZIP file. Once you un-zip it you should see a
UBCdown.Rproj. Click on it and use it as a normal R project. To know more about R projects check the R for data science chapter.
From here on, the instructions below are the same regardless of the method you chose.
Using ubcdown has some prerequisites, such as Pandoc and LaTeX. To compile PDF documents using R, you need to have Pandoc, LaTeX and several related packages installed. If you have a recent version of RStudio, then you already have Pandoc and don't need to do anything more about that. Next is LaTeX. By far the easiest way to install LaTeX on any platform is with the tinytex package. Finally, there are some other packages that we will use in some sections of the dissertation, these are knirt , kableExtra , dplyr, tibble
install.packages(c('tinytex', 'rmarkdown', "knitr", "kableExtra","png","grid","dplyr","tibble"))
tinytex::install_tinytex()
# after restarting RStudio, confirm that you have LaTeX with
tinytex:::is_tinytex()
If you are getting an error message “destination /usr/local/bin not writable” you can see the following tidytex issue.
Note that this template was created using R version 3.5.2 (2018-12-20) -- "Eggshell Igloo" and tested in R version 4.0.4 (2021-02-15) -- "Lost Library Book"
To use gauchodown from RStudio:
-
Ensure that you have already installed LaTeX and the fonts described above, and are using the latest version of RStudio. You can use
gauchodownwithout RStudio. For example, you can write the Rmd files in your favourite text editor (e.g. Atom, Notepad++). But RStudio is probably the easiest tool for writing both R code and text in your dissertation. -
Install the
bookdownpackage:
install.packages("bookdown")
At this point you already forked this repository to your own github or download the zip file to your computer and got yourself a nice looking repo with a main_script.Rmd file. This is the main script of your dissertation (see below). In addition you will see the following structure (se below for detailed instructions of each section):
-
Images; This folder is intended to have any image you will use outside your main chapters (e.g., figures of the introduction)
-
Reference; this is where you will store your bibliography and the reference guideline. Note that, as of March of 2021, G+PS does not require an specific reference style.
-
reference_format.csl; This is the bibliography guiding file. The default is harvard style. If you want to change the format, Zotero provides a list of referencing styles you can download. Just make sure you keep the file name
reference_format.csl -
reference_list.bib; This file contains your bibliography. I just exported a bibliography from my reference-library program (e.g., Zotero, Mendeley) as a
.bibfile. Alternatively, you can copy the reference format to your clipboard and pasted it here.
-
-
Sections; Here you will find each of the sections of the dissertation according to UBC GP+s. Note that some of them (e.g., glossary) are optional.
-
Data; This repository is intended to store any data needed for rendering the dissertation. I strongly suggest you not to have any heavy data files in here. Github will allow you to have up to 2Mb with the Git Large Files plug-in, but anything larger than that should be outside this project.
You need to edit the individual chapter R Markdown files to write your dissertation but you can write in the Rmd files without RStudio on our favourite text editor. However, I suggest you to come back to RStudio to create the PDF and work on the R code in the documents.
While writing, you should git commit your work frequently, after every major activity on your dissertation. For example, every few paragraphs or section of text, and after major step of analysis development. You should git push at the end of each work session before you leave your computer or change task. For gentle novice-friendly guide to getting starting with using Git with R and RStudio, see http://happygitwithr.com/.
The main script of your dissertation is the main_script.Rmd file. This file calls all other sections and builds the final PDF. Note that there are some few parts of this script that need modification (e.g., Chapter titles). The "general" dissertation sections (e.g., abstract, acknowledgements, introduction), that is, those that are not data chapters, can be found in the Sections folder. Each Data chapter is called with the knitr::knit_child function (see below).
Figures and tables numbering and labelling can be a real pain at the end of the dissertation, specially if you remove/add one last figure/table and need to re-number all. Also, figure labels can easily become so huge they occupy a whole page so we can include a long label for the text and a short one for the table of contents.
For numbering figures and tables, we can use the following LaTex reference syntax for in-text figures (Figure \@ref(fig:figureTitle) and for in-text tables (Table \@ref(table:tableTitle). You just need to call the figure in a chunk (see example below) or using LaTex.
For labelling, if you are calling the Figure from a chunk you just need to include in the chunk options fig.cap = "Your long figure caption" for a long figure caption and fig.scap = "For a short figure caption. Note also you need to include out.extra='' for some reason (I don't really know why). Check Overleaf Figures and Table to do labelling in LaTex. See the 08_Introducction.Rmdin the Sections folder for an example.
You can certainly use the same project to house all of the data and code for each of your chapters (and if your analysis runs fast enough you could of course simply do all of your analysis and writing for a chapter in the .Rmd for that chapter). In my experience, I had one R-project per chapter which I then knitted together for the final dissertation. Here is an explanation, copied directly from Dan Ovando of gauchodown, as I couldn't to better:
My dissertation had three chapters. For each chapter, I created a separate RStudio project and folder on my computer, call it "~/PhD/zissou" (I nickname all my projects). Inside that folder I stored the data, code, and paper .Rmd for the zissou chapter. When I wanted to actually knit the dissertation, rather than copy-and-pasting all the required results or data from zissou over to my dissertation folder, I simply used knit_child (and some voodoo in the chunk options).
Note that the first time you render to PDF, R will install all of the LaTex dependencies and, thus, might take a while. Don't worry, this will only be the first time! Looks something like this:
tlmgr.pl: package repository https://ctan.math.illinois.edu/systems/texlive/tlnet (not verified: gpg unavailable)
[1/1, ??:??/??:??] install: ragged2e [3k]
running mktexlsr ...
done running mktexlsr.
To render your dissertation into a PDF, open main_script.Rmd in RStudio and then click the "knit" button. To change the output formats between PDF and Word, look at the output: field and comment-out the formats you don't want. The PDF file of your dissertation will be deposited in the main directory.
Note:, the final product will not be formatted, the cover will not show up, neither any LaTex command (e.g., tables with the kable package or the section titles up to the Introduction). The good news is that things like figures numbering and references will work. I suggest you only use the Word version for getting revisions from your committee.
If you feel like it, check out this issue and help rendering to a formatted Word!
To render your dissertation to Word, go to the top of the main_script.Rmd, comment line 3 bookdown::pdf_book: and comment out line 4 bookdown::word_document2: (or just change pdf_book to word_document2). Additionally, you will need to set toc: yes on line 5 and change the sections titles up to the Introduction by commenting the \section*{Title} and commenting out the # Tilte. This will render the classic rmarkdown table of contents and titles.
The following components are the ones you should edit to customize your dissertation:
This is the main configuration file for your dissertation. It provides the main structure of your dissertation and have some regions that require modifications. The main sections where you need to modify the file are identified with the label <!-- Modify me! --> right above the section. These include the Title of each data chapter and the title of each Appendix.
The basic structure has Four data chapters. If you have more than four data chapters, you can copy and paste the format below after chapter four, (note that for each new chapter you need to change the Title, the figure, and the table numbers):
# Chapter Five: Insert Title Chapter Here
<!-- Set figure and table numbering according to the chapter -->
\renewcommand{\thefigure}{5.\arabic{figure}}
\setcounter{figure}{0}
\renewcommand{\thetable}{5.\arabic{table}}
\setcounter{table}{0}
\renewcommand{\theequation}{5.\arabic{equation}}
\setcounter{equation}{0}
{r chapter_five, child = '~chapter_5_path/chapter_5_file.Rmd', eval = T}
Additionally, if you have less than four data chapters you can just comment out the last chapter and set the child option to eval = F. Alternatively you can just erase the chapter info in the main script.
The basic structure has one appendix for each data chapter. So, for each chapter you also need to create the appendix section as detailed above. Below is the form.
\section*{Appendix D - Supplementary information for "Name of Chapter Five"}
\addcontentsline{toc}{section}{Appendix D}
<!-- Set Figure, table and equations numbering according to the chapter -->
\renewcommand{\thefigure}{A5.\arabic{figure}}
\setcounter{figure}{0}
\renewcommand{\thetable}{A5.\arabic{table}}
\setcounter{table}{0}
\renewcommand{\theequation}{A5.\arabic{equation}}
\setcounter{equation}{0}
{r appendix_d, child = '~/path_to_appendix_5/appendix_5.Rmd', eval = T}
According to UBC's G+PS instructions as of March of 2021, there are some global sections that are optional. The current structure has all of the sections. If you want to remove any global section you simply need to comment out the \section and addcontentsline sections (these give the title in the body and the table of content), and set the child eval option to F. I recommend you not to delete a section you don't want in case you change your mind latter and want to include it. Note that you do not need to worry about page numbering or table of contents! Below is an example of the Dedication section commented out.
<!-- Dedication -->
<!-- \section*{Dedication} -->
<!-- \addcontentsline{toc}{section}{Dedication} -->
These are the Rmd files for each global chapter in your dissertation. Write your dissertation in these. If you're writing in RStudio, you may find the wordcount addin useful for getting word counts and readability statistics in R markdown documents. Check out the Section's ReadMe for specifics of each section.
This project has drawn directly on code and ideas from Dan Ovando's gauchodown.
Other relevant projects include:
-
https://github.com/SeungkiKwak/Kwak_S_PhD_thesis
-
Ed Berry's blog post 'Writing your thesis with bookdown', Posted on September 25, 2017
-
Rosanna van Hespen's (@rosannavhespen) five blog posts on 'Writing your thesis with R Markdown'
-
thesisdowndss by Mine Cetinkaya-Rundel at Duke University
-
beaverdown by Zhian Kamvar at Oregon State University
If you would like to contribute to this project, please start by reading our Guide to Contributing. Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.