3-class-network.Rmd

---
title: "3a. Class Network"
subtitle: "Analytics Sandbox"
author: "K. Bret Staudt Willet | Florida State University"
date: "February 1, 2023"
output:
  xaringan::moon_reader:
    lib_dir: libs
    seal: true
    css: [default, 'css/custom.css', 'css/custom-fonts.css']
    nature:
      titleSlideClass: [left, bottom]
      highlightStyle: github
      highlightLines: true
      countIncrementalSlides: false
---

```{r setup, include=FALSE}
options(htmltools.dir.version = FALSE)

library(knitr)
library(tidyverse)
library(xaringan)
library(fontawesome)
library(igraph)
library(tidygraph)
library(ggraph)
library(ggthemes)
library(cowplot)
library(kableExtra)
```

class: inverse, center, middle

# `r fa("fas fa-diagram-project", fill = "#fff")` <br><br> **Part 3a:** <br> Class Network

---

# `r fa("fas fa-diagram-project", fill = "#fff")` Class Network data

```{r}
edgelist1 <- 
  read_csv("data/edgelist.csv", 
           show_col_types = FALSE) %>%
  mutate(group = week,
         week = substr(week, 1, 1)
         ) %>% 
  group_by(from, to) %>% 
  mutate(weight = n())
glimpse(edgelist1)
```

---

# `r fa("fas fa-diagram-project", fill = "#fff")` Class Network data

```{r}
head(edgelist1, 10)
```

---

class: inverse, center, middle

# `r fa("fas fa-otter", fill = "#fff")` <br><br> Try it Out! 

---

# `r fa("fas fa-otter", fill = "#fff")` Try it Out!

What do you think this code will do?

```{r, message=FALSE, warning=FALSE}
network_graph1 <-
  tidygraph::as_tbl_graph(edgelist1) %>%
  mutate(popularity = centrality_degree(mode = 'in'))
network_graph1
```

---

# `r fa("fas fa-otter", fill = "#fff")` Try it Out!

What do you think this code will do?

```{r, message=FALSE, warning=FALSE}
network_graph1 %>%
  ggraph(layout = 'kk')
```

---

# `r fa("fas fa-otter", fill = "#fff")` Try it Out!

What do you think this code will do?

```{r, message=FALSE, warning=FALSE}
network_graph1 %>%
  ggraph(layout = 'kk') +
  geom_edge_arc()
```

---

# `r fa("fas fa-otter", fill = "#fff")` Try it Out!

What do you think this code will do?

```{r, message=FALSE, warning=FALSE}
network_graph1 %>%
  ggraph(layout = 'kk') +
  geom_edge_arc() +
  geom_node_point()
```

---

# `r fa("fas fa-otter", fill = "#fff")` Try it Out!

What do you think this code will do?

```{r, message=FALSE, warning=FALSE}
network_graph1 %>%
  ggraph(layout = 'kk') +
  geom_edge_arc() +
  geom_node_point(alpha = .4, aes(size = popularity)) +
  scale_size(range = c(1,10))
```

---

# `r fa("fas fa-otter", fill = "#fff")` Try it Out!

What do you think this code will do?

```{r, message=FALSE, warning=FALSE}
network_graph1 %>%
  ggraph(layout = 'kk') +
  geom_edge_arc(alpha = .2, 
                width = .5, 
                strength = .5,
                color = 'steelblue') +
  geom_node_point(alpha = .4, aes(size = popularity)) +
  scale_size(range = c(1,10))
```

---

# `r fa("fas fa-otter", fill = "#fff")` Try it Out!

What do you think this code will do?

```{r, message=FALSE, warning=FALSE}
sociogram1 <-
  network_graph1 %>%
  ggraph(layout = 'kk') +
  geom_edge_arc(alpha = .2, 
                width = .5, 
                strength = .5,
                color = 'steelblue'
  ) +
  geom_node_point(alpha = .4, aes(size = popularity)) +
  scale_size(range = c(1,10)) +
  theme_wsj() + 
  scale_colour_wsj("colors6") +
  theme(axis.line=element_blank(),
        axis.text.x=element_blank(), axis.text.y=element_blank(),
        axis.ticks.x =element_blank(), axis.ticks.y =element_blank(),
        axis.title.x=element_blank(), axis.title.y=element_blank(),
        panel.background=element_blank(), panel.border=element_blank(),
        panel.grid.major=element_blank(), panel.grid.minor=element_blank())
sociogram1
```

---

# `r fa("fas fa-otter", fill = "#fff")` Picture it!

```{r, include=FALSE, eval=FALSE}
ggsave(sociogram1, file="output/3-class-network.png", width=8, height=4.5)
```

```{r, out.width="100%", echo=FALSE, fig.align="center"}
include_graphics("output/3-class-network.png")
```

---

# `r fa("fas fa-otter", fill = "#fff")` Look closer!

**Online Class Discussion - Social Network Analysis**

There are quite a few descriptive measures of networks:

- **Order:** number of nodes/vertices (students, in this case)
- **Size:** number of edges/connections (responses, in this case)
- **Reciprocity:** mutuality
- **Transitivity:** clustering
- **Diameter:** similar to degrees of separation
- **Density:** out of all possible connections, percentage that have been made
- **Node degree:** number of connections
- **Sentiment score:** how positive or negative in aggregate
- Character count, Word count, Length of threads

---

# `r fa("fas fa-otter", fill = "#fff")` Look closer!

**Order:** number of nodes/vertices (students, in this case)

```{r}
library(igraph)
gorder(network_graph1)
```

<hr>

**Size:** number of edges/connections (responses, in this case)

```{r}
gsize(network_graph1)
```

---

# `r fa("fas fa-otter", fill = "#fff")` Look closer!

**Reciprocity:** mutuality

```{r}
reciprocity(network_graph1)
```

<hr>

**Transitivity:** clustering

```{r}
transitivity(network_graph1)
```

---

# `r fa("fas fa-otter", fill = "#fff")` Look closer!

**Diameter:** similar to degrees of separation

```{r}
diameter(network_graph1)
```

<hr>

**Density:** out of all possible connections, percentage that have been made

```{r}
edge_density(network_graph1)
```

---

# `r fa("fas fa-otter", fill = "#fff")` Look closer!

**Node degree:** number of connections

```{r}
mean(degree(network_graph1))
```

<hr>

```{r}
degree(network_graph1) %>% mean()
```

<hr>

```{r}
median(degree(network_graph1))
```

---

class: inverse, center, middle

# `r fa("fas fa-otter", fill = "#fff")` <br> <br> **Try it differently!** <br><br> Interact with a Shiny App

https://fsuksu.shinyapps.io/fsuksu/

---

class: inverse, center, middle

# `r fa("fas fa-code", fill = "#fff")` <br><br> Try on your own!

---

# `r fa("fas fa-code", fill = "#fff")` Try on your own!

- Download a copy of this repository.
- Use the saved data in the "data" folder to play around a bit more, changing different parameters.
- Reflect: 
  - What other comparisons might you make?
  - How else might you analyze these data?

---

class: inverse, center, middle

# `r fa("fas fa-list", fill = "#fff")` <br><br> Appendix: <br> Helpful Resources <br> and Troubleshooting

---

# Resources

```{r child="notes/resources.Rmd"} 
```

---

# Troubleshooting

```{r child="notes/troubleshooting.Rmd"} 
```