Communicating your results/analysis

You can view this presentation at https://rpubs.com/alperyilmaz/quarto-slides

Communicate

Tidy workflow

Communicating your findings is part of Data analysis/Data science

Packages we have learned so far specialize in one part of the workflow. Quarto (also RMarkdown) helps communicating our results.

Quarto (aside RMarkdown)

How it works

• A compute engine (knitr for R or Jupyter Notebook) will compute and run your code chunks and embed their results in Markdown file
• Pandoc will convert markdown format into various formats

Quarto document is a plain text file which can be rendered into …

… HTML

Your turn

Let’s create our first Quarto document and then render it.

Provide title for your document (this is not filename!)

We have Untitled file, let’s save this file

Now, we are ready to Render the document

The rendered HTML document will be opened automatically in your browser

Editor

You can use Source (left) or Visual editor (right) when editing your Quarto document.

---
title: "ggplot2 demo"
author: "Norah Jones"
date: "5/22/2021"
format: 
  html:
    fig-width: 8
    fig-height: 4
    code-fold: true
---

## Air Quality

@fig-airquality further explores the impact of temperature on ozone level.

```{r}
#| label: fig-airquality
#| fig-cap: "Temperature and ozone level."
#| warning: false

library(ggplot2)

ggplot(airquality, aes(Temp, Ozone)) + 
  geom_point() + 
  geom_smooth(method = "loess"
)
```

• lines 1-10 (called YAML header) describe document properties
• line 14 contains citation to the image (which will be calculated as Figure 1 automatically)
• lines 16-27 contains the R code chunk and lines 17-19 contains chunk options

This document will be rendered as follows

Pandoc can convert markdown format into HTML, PDF or Word (actually many more). Let’s convert the same document into three different formats by changing the format: html into either format: pdf or format: docx

More code chunk options

In HTML format, you can show the tables (data frames, tibbles) as HTML tables (browsing next page) just like in RStudio.

Let’s add the following chunk to our document and then render it (also add df-print: paged option in document properties)

```{r}
#| echo: false
#| warning: false
#| cache: true
library(tidyverse)

mpg_avg <- mpg |>
  group_by(manufacturer, cyl) |>
  summarize(avg_cty=mean(cty), avg_hwy=mean(hwy))

mpg_avg
```

Let’s go over the chunk options: echo: false means that the code will not be shown in final document, warning: false will suppress any warning or messages, cache: true means that the calculation in that chunk will be cached and won’t be re-calculated (unless chunk is modified) in following renderings.

Packages for better tables

There are numerous R packages for generating beautiful tables to be embedded into HTML or PDF formats. {reacttable} and {gt} are very capable, professional looking tables can be easily prepared with these packages.

Especially {gt} package is very capable. Please visit its introduction site and gallery

gt table example

image-20230604164713201

gt table example 1

image-20230604165458528

Inline code

Again, let’s imagine, in our document we want to have statements like “the dataset contained 234 rows”. We’d face the same challenge if data is updated, we have to locate such statements scattered in our document and then update one by one. Actually, you can have inline R code have such calculations embedded by code.

The rendering of following line in Quarto document is shown in next slide.

There are `r nrow(mpg)` observations in **mpg** data. The table below shows average city and highway mileage 
per make and cyclinder number.

And much more

Quarto supports much more features; citations, formulas, cross references, callouts, publishing. Please review the contents of the following quarto document (please scroll down to view whole source). The published version is found here

---
title: "ggplot2 demo"
author: "alper"
date: "6/1/2023"
toc: true
number-sections: true
bibliography: references.bib
format: 
  html:
    fig-width: 8
    fig-height: 4
    code-fold: true
    code-link: true
    df-print: paged
  pdf:
    geometry: 
      - top=30mm
      - left=30mm
  docx: default
---

# Air Quality

@fig-airquality further explores the impact of temperature on ozone level.

```{r}
#| label: fig-airquality
#| fig-cap: "Temperature and ozone level."
#| echo: true
#| warning: false

library(tidyverse)
library(ggplot2)

ggplot(airquality, aes(Temp, Ozone)) + 
  geom_point() + 
  geom_smooth(method = "loess"
)
```


# mgp data

## mpg scatter plot

@fig-scatterplot shows correlation between highway and city mileages. 

```{r}
#| label: fig-scatterplot
#| fig-cap: "City and highway mileage for 38 popular models of cars."
#| fig-alt: "Scatterplot of city vs. highway mileage for cars, where points are colored by the number of cylinders. The plot displays a positive, linear, and strong relationship between city and highway mileage, and mileage increases as the number of cylinders decreases."
#| fig-width: 6
#| fig-height: 3.5

ggplot(mpg, aes(hwy, cty, color=cyl)) + 
  geom_point(alpha=0.5, size=2) +
  scale_color_viridis_c() +
  theme_minimal()
```

## the mpg data as table

There are `r nrow(mpg)` observations in our data.

```{r}
#| echo: false
#| warning: false
#| cache: true
mpg |>
  group_by(manufacturer, cyl) |>
  summarize(avg_cty=mean(cty), avg_hwy=mean(hwy))
```

Hypothetical `table_viewer` will show your tables in a much better way.

```{r}
#| eval: false
table_viewer(mpg)
```

# Formulas

you can have formulas in your document


$$
price = \hat{\beta}_0 + \hat{\beta}_1 \times area + \epsilon
$$

# Citations

We're going to do this analysis using literate programming [@Knuth1984].

# Layout

## Callouts

::: {.callout-note}
Here's some information
:::

Shiny

Shiny allows your users to interact with your content (i.e a web app) online without requiring any web development skills.

In a Shiny web app, web elements (text input, drop down menus, sliders, etc.) are used as input for your R code where the code is update upon user input.

Please visit the Gallery and interact with the demos.

Additional resources about Quarto

Please visit these sites for more information about Quarto and its capabilities

• Quarto guide
• Quarto chapter at R for Data Science book
• DataCamp Quarto Cheatsheet

About the exam

In the exam, you’re expected to submit/upload a Word or HTML document as an answer. The document will contain your code and your output.

So, please practice opening new Quarto document and then rendering it in HTML or Word.