Final Project

Agenda

• Updating your yaml
• Putting all code in the appendix
• Hiding messages and warnings from your package loadings
• Hiding interactive output
• Creating pretty tables
• Creating figure and table caption
• Putting intermediate figures at the end
• Including citations

Application Exercise

Copy starter files from:

https://github.com/sta-112-f23/final-project.git

Updating your yaml

• For the final project I want you to create .pdf files.

---
title: "The title of your document"
author: "Your name"
format: pdf 
---

Updating your yaml

• For the final project I want you to create .pdf files.

---
title: "The title of your document"
author: "Your name"
format: 
  pdf: 
    fontsize: 12pt 
---

• 12 pt font

Putting all code in the Appendix

• For these fancy reports, we want to see the R code, but we don’t want it interspersed throughout the document.
• You can “hide” all of your chunks with the execute: echo: false in your yaml
• THEN at the end of you document (in the Appendix) you can add

```{r ref.label=knitr::all_labels()}
#| echo: true
#| eval: false
```

Creating pretty tables

• We’ve been using the summary() function to print tables of our model output.
• These kind of look like “code” in our final reports

Creating pretty tables

• We can make these prettier with the gtsummary package
• Load the gtsummary package with library(gtsummary)
• Use the tbl_regression() function to output a pretty table

library(gtsummary) 
lm(wt ~ am, data = mtcars) |>
  tbl_regression() 

Creating Overall Table

tbl_summary(mtcars)

Creating Overall Table

tbl_summary(mtcars, include = c(am, wt, mpg))

Creating table captions

library(gtsummary) 
lm(wt ~ am, data = mtcars) |>
  tbl_regression() |>
  modify_caption(caption = "Predicting Car's weight from transmission type") 

Creating figure captions

• If you have an R chunck that produces a Figure, you can caption it using the fig.cap chunk option. For example:

```{r}
#| fig.cap: "Scatterplot of Miles per gallon by Displacement"
ggplot(mtcars, aes(x = mpg, y = disp)) +
  geom_point()
```

Putting extra figures at the end of your document

• Your prompt tells you to put all “intermediate” figures in the appendix
  • This includes figures that you used to check the assumptions of the models that were not the final model
• To do this:
  • Save your figures as R objects
  • Create code chunks at the beginning of the Appendix that run these objects (with captions)

Putting extra figures at the end of your document

• Save the plot to an R object, p

p <- ggplot(mtcars, aes(mpg, disp)) +
  geom_point()

• Put this R object, p in a chunk at the beginning of your Appendix

```{r, fig.cap = "Scatterplot of Miles per gallon by Displacement"}
p
```

Citations

• Your introduction should cite prior research in the area of your research question
• You will need to create a citations.bib file with citations, and then reference them using [@citation]

Citations

• Open the citations.bib file in your RStudio project

Citations

• This bibliography is using BibTex, which is a certain way citations are saved in Latex
• Your citations.bib is going to be a file with multiple BibTex entries
• You can get Bibtex entries from Google Scholar
  • Search for the article you’d like to cite
  • Under the article there will be ", click that
  • A window will pop up with several citation styles, click BibTex on the bottom

Citations

• This bibliography is using BibTex, which is a certain way citations are saved in Latex
• Your citations.bib is going to be a file with multiple BibTex entries
• You can get Bibtex entries from Google Scholar
  • Search for the article you’d like to cite
  • Under the article there will be ", click that
  • A window will pop up with several citation styles, click BibTex on the bottom
  • Copy the text in the window that pops up into your citations.bib file in RStudio

Citations

• The first argument of this citation is the reference key that you will use in your main document. Here it is roumie2017comparative

@article{roumie2017comparative,
  title={Comparative safety of sulfonylurea and metformin monotherapy on the risk of heart failure: a cohort study},
  author={Roumie, Christianne L and Min, Jea Young and D'Agostino McGowan, Lucy and Presley, Caroline and Grijalva, Carlos G and Hackstadt, Amber J and Hung, Adriana M and Greevy, Robert A and Elasy, Tom and Griffin, Marie R},
  journal={Journal of the American Heart Association},
  volume={6},
  number={4},
  pages={e005379},
  year={2017},
  publisher={Am Heart Assoc}
}

Citations

• Go back to your final report document
• When you want to cite this paper in the main text, use the key like this: [@roumie2017comparative]

Citations

• What if the thing you want to cite isn’t on Google Scholar?
• Generate a BibTex object here:

http://www.citationmachine.net/bibtex/cite-a-website

Citations

• The first line of your results should say “All analysis is completed in R” and then cite the R packages you used.
• Run this to get the citation:

print(citation("tidyverse"), bibtex = TRUE)

• Make sure there is a reference key, if not add one.

@Manual{,
    title = {tidyverse: Easily Install and Load the 'Tidyverse'},
    author = {Hadley Wickham},
    year = {2017},
    note = {R package version 1.2.1},
    url = {https://CRAN.R-project.org/package=tidyverse},
  }

Citations

• The first line of your results should say “All analysis is completed in R”[@rstats; @tidyverse] and then cite the R packages you used.
• Run this to get the citation:

print(citation("tidyverse"), bibtex = TRUE)

• Make sure there is a reference key, if not add one.

@Manual{tidyverse,
    title = {tidyverse: Easily Install and Load the 'Tidyverse'},
    author = {Hadley Wickham},
    year = {2017},
    note = {R package version 1.2.1},
    url = {https://CRAN.R-project.org/package=tidyverse},
  }

Citations

• The first line of your results should say “All analysis is completed in R”[@rstats; @tidyverse; @broom] and then cite the R packages you used.
• For more than one, separate with a ;
• Run this to get the citation:

print(citation("broom"), bibtex = TRUE)

• Make sure there is a reference key, if not add one.