Statebins.Rmd

---
title: "socviz chapter 7 section 3 Statebins"
author: "Yaqiong Li"
date: "Dec 8, 2018"
output: pdf_document
---

Instead of using *ggplot2* or *chloroplethr* packages to achieve chloropelth maps in the United States, *statebins* is another choice.

Install the book's package from github.
```{r}
# devtools::install_github("kjhealy/socviz")
```

Several arguements:
* state_data refers to the data frame
* state_col refers to a vector of state names
* value_col refers to the value to be shown
* statebins_continuous() refers to continuous variables

```{r message=FALSE}
library(statebins)
library(socviz) # this package has election as a data frame
library(tidyverse)

head(election)
```

```{r}
election %>% select(state, total_vote, r_points, pct_trump, party, census) %>% sample_n(5)
```

```{r}
statebins_continuous(state_data = election, 
                     state_col = "state", 
                     text_color = "white", 
                     value_col = "pct_trump", 
                     brewer_pal = "Reds", 
                     font_size = 3, 
                     legend_title = "Percent Trump")
```

```{r}
statebins_continuous(state_data = subset(election, st %nin% "DC"), 
                     state_col = "state",
                     text_color = "black",
                     value_col = "pct_clinton",
                     brewer_pal = "Blues",
                     font_size = 3,
                     legend_title = "Percent Clinton")
```

For categorical data, use statebins_manual(). Here a variable called _color_ is created to represent the party.
```{r}
election <- election %>% mutate(color = recode(party, 
                                               Republican = "red",
                                               Democrat = "blue"))
head(election)
```

```{r}
statebins_manual(state_data = election,
                 state_col = "st",
                 color_col = "color",
                 text_color = "black",
                 font_size = 3,
                 legend_title = "Winner",
                 labels = c("Trump", "Clinton"),
                 legend_position = "right")
```

Alternatively, use statebins() cut the data by the _breaks_ argument.
```{r}
statebins(state_data = election,
          state_col = "state",
          value_col = "pct_trump",
          text_color = "white",
          breaks = 4,
          labels = c("4-21", "21-37", "37-53", "53-70"),
          brewer_pal = "Reds",
          font_size = 3,
          legend_title = "Percent Trump")
```