3-starwars_opinions_final.Rmd

---
title: "Star Wars Opinion Data"
author: "Angela Zoss"
date: "3/20/19"
output: html_document
---

## Setup your environment

```{r}

# Load required libraries

library(tidyverse)

```

## Load your data

```{r}

# data from https://fivethirtyeight.com/features/americas-favorite-star-wars-movies-and-least-favorite-characters/
# note: CSV has two rows of headers, so I have manually created a list of headers 
# and am adding that after loading just the data rows

# also extracting just the opinion columns and reshaping them to long data format

new_names <- read_csv('data/StarWarsNames.csv') %>% dplyr::select(NewNames)

starwars_opins <- read_csv('data/StarWars.csv', skip=2, col_names=FALSE) %>% setNames(unlist(new_names))
  
starwars_opins_tidy <- starwars_opins %>% dplyr::select(RespondentID, starts_with("Opinion")) %>% gather("Character","Opinion",-RespondentID) %>% mutate(Character=sub("Opinion","",Character)) %>% na.omit()

# finally creating three separate datasets - Han Solo, Jar Jar, and combined

solo <- starwars_opins_tidy %>% dplyr::filter(Character=="Solo")
jarjar <- starwars_opins_tidy %>% dplyr::filter(Character == "JarJar")
combined <- bind_rows(solo, jarjar)

```

## Plot the opinions for Han Solo

```{r}

# hint: there is a geom that will take a categorical variable and count the data points in
# each category

ggplot(solo) +
  geom_bar(aes(Opinion))

```

## Plot the opinions for Jar Jar Binks

```{r}

ggplot(jarjar) +
  geom_bar(aes(Opinion))

```

## Edit the axes of each of these charts to make them comparable

```{r}

# remember, editing an axis usually requires a scale

ggplot(solo) +
  geom_bar(aes(Opinion)) +
  scale_y_continuous(limits=c(0,610))

ggplot(jarjar) +
  geom_bar(aes(Opinion)) +
  scale_y_continuous(limits=c(0,610))

```

## Advanced: can you think of another way to standardize the axes?

```{r}

ggplot(combined) +
  geom_bar(aes(Opinion)) +
  facet_wrap(~Character)

```

## Reorder the opinion levels so they are in a logical order

```{r}

opinion.levels <- c("Unfamiliar (N/A)","Very unfavorably","Somewhat unfavorably",
                    "Neither favorably nor unfavorably (neutral)",
                    "Somewhat favorably","Very favorably")

combined.f <- combined %>% 
         mutate(Opinion=factor(Opinion, opinion.levels))

ggplot(combined.f) +
  geom_bar(aes(Opinion)) +
  facet_wrap(~Character)

```

## Flip the axes so the opinion levels are on the y-axis

```{r}

ggplot(combined.f) +
  geom_bar(aes(Opinion)) +
  facet_wrap(~Character) +
  coord_flip()

```

## Use a different color for each opinion level

```{r}

ggplot(combined.f) +
  geom_bar(aes(Opinion, fill=Opinion), show.legend = FALSE) +
  facet_wrap(~Character) +
  coord_flip()

```

## Customize the color palette

```{r}

# hint: editing colors requires a change of scale
# can google "R colors" for a list of all named colors

opinion.colors <- c("grey50","firebrick4","firebrick1","grey85","dodgerblue1","dodgerblue4")

ggplot(combined.f) +
  geom_bar(aes(Opinion, fill=Opinion), show.legend = FALSE) +
  facet_wrap(~Character) +
  coord_flip() +
  scale_fill_manual(values = opinion.colors) +
  theme_minimal()

```