Research 101: How to work as a modern quantitative social scientist?

This handbook provides a comprehensive guide for installing, configuring, and using Positron, Quarto, and Zotero for writing quantitative academic research papers.

1 Installation

This section will guide you through the installation and setup of Positron, Quarto, and Zotero for writing quantitative academic research papers.

1.1 Install R & Positron & Quarto

First, we need to download and install R, Positron and Quarto.

1. Follow the instructions on CRAN mirror to download and install R.

2. Follow the instructions on Positron Download to download Positron.

3. Run SETUP and install Positron. Check all of the boxes when asked to.

Check all of the boxes

4. Follow the instructions on Quarto Get Started to download and install Quarto.

1.2 Set up System Environment Variables

Next, we need to set up the system environment variables for proxys.

1. Open the Windows Start Menu and search for Env.

2. Click on Edit the system environment variables.

3. In the System Properties window, click on the Environment Variables button.

4. In the Environment Variables window, under the System variables section, click on the New button.

5. In the New System Variable window, enter HTTP_PROXY as the Variable name and http://127.0.0.1:xxxx as the Variable value. Click OK. (Replace xxxx with the port number you are using for your proxy.)

6. Repeat step 5 to create another system variable named HTTPS_PROXY with the same value http://127.0.0.1:xxxx.

7. Repeat step 5 to create another system variable named NO_PROXY with the value localhost,127.0.0.1.

8. Click OK to close the Environment Variables window. Click OK to close the System Properties window.

1.3 Join Github Eduction

To use Github Copilot, we need to join the Github Education program.

1. Open your web browser and go to GitHub Education.

2. Click on the Join GitHub Education button.

3. Follow the instructions to verify your student status and complete the application process.

1.4 Update Positron and Set Up Github Copilot Chat

Now, we need to update Positron and set up Github Copilot Chat.

1. Open Positron.

2. Click on the Manage Icon on the bottom left corner.

3. Click on Settings to open the Settings tab.

4. In the Settings tab, search for channel, and switch the releases to dailies.

5. Click on the Manage Icon on the bottom left corner. Click on Check for Updates to update Positron to the latest version. After the update is complete, restart Positron.

6. Repeat step 2 & 3 to open the Settings tab again.

7. In the Settings tab, search for assistant. Check on the box Enable Positron Assistant.

8. Click on the Chat Icon on the left sidebar to open the Chat tab.

9. In the Chat tab, click on add model provider button.

10. In the Configure Language Model Provider window, select GitHub Copilot. Click on Sign in.

11. In the GitHub login page, enter your GitHub credentials to sign in.

12. After signing in, you will be redirected back to Positron. You can now use the LLMs provided by GitHub Copilot to assist with your coding.

1.5 Install R Package Manager

Next, we need to install the R Package Manager to manage R packages.

1. Open Positron. Click on the Extensions Icon on the left sidebar.

2. Search for Positron R Package Manager. Click on Install to install the extension.

3. Once the extension is installed and activated, you can find the R Package Manager 📖 icon on the left sidebar.

4. Drag the 📖 icon to the secondary sidebar on the right.

1.6 Install Zotero

Finally, we need to download and install Zotero for managing references.

1. Download the Zotero installer for your operating system from the Zotero Download page.

2. Run the installer and follow the on-screen instructions to complete the installation.

3. Open Zotero, go to Edit - Preferences, create and log into an account to sync your references. Click New Collection in the top left corner to create a new collection in My Library, then click the magic wand icon on the right to add any literature by DOI to that collection.

4. Download and extract the Zotero Addons plugin.

5. Open Zotero, click on Tools in the top menu bar, select Add-ons, click the gear icon, select Install Add-on From File, and choose the .xpi file downloaded in step 4.

6. Select Tools - Add-on Market, install the Better BibTeX for Zotero plugin for better BibTeX support.

7. In the library, right-click on the title bar and check Citation Key to display citation names.

8. Select Edit - Preferences - Better BibTeX, in the Citation key format, keep auth.lower + year.

9. Right-click on the literature library you want to export, select Export Collection, check Keep updated, name the file references.bib and save it to your paper project folder.

2 Usage

This section will guide you through the usage of Positron, Quarto, and Zotero for writing quantitative academic research papers.

2.1 Fetch ORCID Template

Now, every time before you start writing a new paper, we could install the ORCID template for better formatting.

1. Open Positron, create a qmd file as your transcript. Click on TERMINAL on the bottom sidebar.

2. In the terminal, type the following command and press Enter: quarto add kv9898/orcid.

3. Enter Y when asked to trust the authors and to continue.

4. You can now work on an ORCID template. Copy and paste the example syntax from here: ORCID Template.

5. Click Preview on the top bar to try compile the document. Follow the instructions on the terminal to install any missing packages.

2.2 YAML header

You can make changes to the default YAML header to fit your needs. Refer to the comments on the code in the template for more information.

---
title: Orcid Template #Change article title
subtitle: with affiliation #Change article subtitle
date: last-modified #Set article date to last modified time, or manually set to specific date, e.g., '2024-06-01'
authors:
  - name: Dianyi Yang #Change author name
    email: dianyi.yang@politics.ox.ac.uk #Change author email
    orcid: 0009-0004-4652-3429 #Change author ORCID
    affiliations: #Change author affiliations
      - ref: ox #Select affiliation, must match id in affiliations section below
      - ref: sec #Select affiliation, must match id in affiliations section below
    corresponding: true #Set as corresponding author
    equal-contributor: true #Set as equal contributor
  - name: Yang Dianyi #Can add multiple authors
    orcid: 0009-0004-4652-3429
    affiliations: 
      - ref: ox
  - name: Someone else
    affiliations:
      - ref: ox
      - ref: sec
    orcid: 0009-0009-9108-5436
    equal-contributor: true

affiliations: #Set affiliations
  - id: ox #Change affiliation code
    name: Department of Politics and International Relations, University of Oxford
  - id: sec #Can set multiple affiliations
    name: Second Institute
format: #Choose compilation format
  #native: default #Compile to native HTML format
  #docx: default #Compile to Word format
  orcid-pdf: #Compile to PDF format
    keep-tex: true #Choose whether to keep .tex file, usually required for journal submission
    number-sections: true #Choose whether to number sections
    cover: false #Choose whether to include cover, if true, value is cover.pdf
    toc: false #Choose whether to include table of contents
    toc-newpage: false #Choose whether table of contents starts on new page
bibliography: references.bib #Set bibliography file
csl: extra/apa.csl #Set citation format
abstract: This is an abstract. #Modify abstract content
keywords: a, b, c #Modify keywords
acknowledgements: This is an acknowledgement. #Modify acknowledgements content
---

2.3 An Example Regression Analysis

Here is an example regression analysis using the ORCID template and the mtcars dataset to show you some basic syntax when you are writing a quantitative academic research paper.

First, let’s examine the relationship between car weight and fuel efficiency using a scatter plot.

Insert an r code cell by clicking the Insert Code Cell button on the top bar, and copy and paste the following code into the code cell:

# Assign a label to the scatter plot for cross-referencing
# Labels for figures should have the prefix "fig-"
#| label: fig-scatter
# Assign a caption to the scatter plot
#| fig-cap: "Relationship between Car Weight and Fuel Efficiency"
# Set echo, warning, and message to false for the plot output only
#| echo: false
#| warning: false
#| message: false
# Install the necessary packages
install.packages(c("ggplot2", "dplyr"))
# If the packages are already installed, you can skip the above line
library(ggplot2)
library(dplyr)
# Create descriptive statistics plot
ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point(aes(color = factor(cyl)), size = 3) +
  geom_smooth(method = "lm", se = TRUE, color = "blue") +
  labs(
    x = "Weight (1000 lbs)",
    y = "Miles per Gallon",
    color = "Cylinders"
  ) +
  theme_minimal()

Figure 1: Relationship between Car Weight and Fuel Efficiency

Next, we’ll perform a regression analysis. Equations written in syntax should be covered by double dollar signs $$ ... $$ and be assigned a label using {#label} for cross-referencing.

$$
mpg = \beta_0 + \beta_1 \cdot weight + \beta_2 \cdot cylinders + \epsilon
$$ {#eq-regression}  #Labels for equations should have the prefix "eq-"

\[ mpg = \beta_0 + \beta_1 \cdot weight + \beta_2 \cdot cylinders + \epsilon \tag{1}\]

We can also use & to align equations. For example, if we want our equations to be aligned at the equal signs and quote them as one group, we can write:

$$
\begin{aligned}
1+2&=3\\
3+4&=7\\
11&=5+6
\end{aligned}
$$ {#eq-group}

Note that here we use \begin{aligned} ... \end{aligned} to align a group of equations, and we use \\ to indicate line breaks, and & to indicate the alignment point.

The three equations are now quoted as one group and aligned at the equal signs:

\[ \begin{aligned} 1+2&=3\\ 3+4&=7\\ 11&=5+6 \end{aligned} \tag{2}\]

Finally, let’s create a regression table to summarize our regression results. We will use the modelsummary package to create a well-formatted regression table.

# Assign a label to the regression table for cross-referencing
# Labels for tables should have the prefix "tbl-"
#| label: tbl-regression
# Assign a caption to the regression table
#| tbl-cap: "Regression Results: Fuel Efficiency Model"
# Set echo, warning, and message to false for the plot output only
#| echo: false
#| warning: false
#| message: false
library(modelsummary)
# Fit regression models
model1 <- lm(mpg ~ wt, data = mtcars)
model2 <- lm(mpg ~ wt + cyl, data = mtcars)
model3 <- lm(mpg ~ wt + cyl + hp, data = mtcars)
# Create regression table
modelsummary(
  list("Model 1" = model1, "Model 2" = model2, "Model 3" = model3),
  stars = TRUE,
  gof_map = c("nobs", "r.squared", "adj.r.squared")
)

Table 1: Regression Results: Fuel Efficiency Model

	Model 1	Model 2	Model 3
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
(Intercept)	37.285***	39.686***	38.752***
	(1.878)	(1.715)	(1.787)
wt	-5.344***	-3.191***	-3.167***
	(0.559)	(0.757)	(0.741)
cyl		-1.508**	-0.942+
		(0.415)	(0.551)
hp			-0.018
			(0.012)
Num.Obs.	32	32	32
R2	0.753	0.830	0.843
R2 Adj.	0.745	0.819	0.826

The equations, figures, and tables are automatically numbered in a correct sequence, and could be cross-referenced in the text.

To create proper cross-references in Quarto, use the following syntax: @label at where you want to make cross references, where label is the label you assigned to the figure, table, or equation. For instance, @eq-regression refers to the regression equation Equation 1, @eq-group refers to the group of equations Equation 2, @fig-scatter refers to the scatter plot Figure 1, and @tbl-regression refers to the regression table Table 1. You could refer to Quarto Cross-References Documentation for more information on cross-reference.

2.4 An Example Literature Review

Here is an example of how to use Quarto for literature citations.

First, make sure you have added the references.bib file to your project according to the previous instructions in Section 1.6, and correctly set the bibliography field in the YAML header.

Second, make sure you have downloaded and installed the corresponding citation format *.csl file, and correctly set the csl field in the YAML header. Citation formats can be downloaded from the CSL repository.

Now, you can use @citation_key in the text to cite literature. We can place citations at the beginning of sentences. For example:

@lake2011 Lake (2011) argues that the “isms” debate in international relations research is harmful.

Or, we can also place citations at the end of sentences:

The “isms” debate in international relations research is harmful [@lake2011] (Lake, 2011).

To cite multiple articles, use comma-separated citation keys, for example:

Here are two good articles [@lake2011, @hale2020] (Hale, 2020; Lake, 2011).

Finally, add the following content at the end of the document to generate a reference list:

# References {.unnumbered} #Add unnumbered references section

::: {#refs} #Place references here, not at the end of document
:::

References

Hale, T. (2020). Transnational Actors and Transnational Governance in Global Environmental Politics. Annual Review of Political Science, 23(1), 203–220. https://doi.org/10.1146/annurev-polisci-050718-032644

Lake, D. A. (2011). Why “isms” Are Evil: Theory, Epistemology, and Academic Sects as Impediments to Understanding and Progress1: Why “isms” are Evil. International Studies Quarterly, 55(2), 465–480. https://doi.org/10.1111/j.1468-2478.2011.00661.x