This document contains code for Group 3. It is your task to modify the code in places where I have left ... as a placeholder for a variable name. Please feel free to consult with myself and Quyen for guidance.
## Libraries for interacting with data
library(readr) # reads in spreadsheet data
library(dplyr) # lets you wrangle data more effectively
library(tidyr) # lets us interact with data with additional functions
## Libraries for plotting
library(ggplot2) # plotting library
library(waffle) # square pie chart (waffle chart) library
## Libraries for spatial data
library(sf) # a nicer way to deal with spatial data
library(spData) # spatial datasetsNext, we’re going to import the class dataset.
CSPdf <- read_tsv("https://raw.githubusercontent.com/EA30POM/Fall2021/main/data/OTEC_CSP.tsv") # use the function read_tsv from readr to pull in a spreadsheet from this URL.
# Then store the output in an object called "CSPdf"
# CSPdf stands for Conservation Science Partners data frameWe can always pull up a viewer of our data by using these commands:
## Version 1:
CSPdf %>% # take CSPdf and feed it into the next command
head( ) %>% # look at the first 6 rows
View( ) # open the spreadsheet viewer
# akin to x --> f(x) --> g(x)
## Alternatively, version 2:
View( head( CSPdf ) ) # akin to g( f( x ) )What are the difference in sentiments towards elephants in countries where they are native vs. non-native?
We first need to subset the data to the articles about elephants. We also need to recode the data based on which countries are in the native range of elephants.
### 3.1: Subset data to articles focused on elephants
group3sub <- CSPdf %>%
filter(...) # How would you filter to the articles where the Taxon is "elephant"?
### 3.2: Recategorize countries into native vs. non-native range
# 3.2.1: See the set of countries in the data
table(group3sub$Country)
# 3.2.2: Recategorize data based on Country
group3sub1 <- group3sub %>%
mutate(ElephantRange = case_when(Country %in% c("India", "Thailand") ~"Native",
Country %in% c("United States","United Kingdom") ~ "Non-native"))
# Create a new column called ElephantRange that lists
# whether or not each article is in the native or non-native range
# of elephants. How would we add on the other countries
# to recast them as native or non-native?We are now ready to plot the data
### 3.3: Visualizing the distribution of sentiments in native vs. non-native range
g3 <- ggplot(data=group3sub1, aes(x=...,y=...)) # create a plot using the appropriate columns (hint: range of elephants and sentiment)
g3 # inspect intermediate output
g3 <- g3 + geom_boxplot() # add a boxplot
g3 # inspect intermediate output
g3 <- g3 + geom_hline(yintercept=0, linetype="dashed", col="red") # add a red dahsed line to demarcate where the positive vs. negative articles are located
g3 # inspect intermediate output
g3 <- g3 + labs(x="...",y="...") # change axis labels to something more informative
g3 <- g3 + coord_flip() # switch the x- and y-axis
g3 # display final plotIf we subset by HWI (Human Wildlife Interaction), what is the distribution of positive vs negative sentiments towards elephants?
It is unclear to me if the group is more keen on 1) ultimately visualizing the distribution of sentiment across countries only for those articles focusing on HWI (a map plot) or 2) displaying the distribution of sentiment values for HWI and non-HWI articles.
In case my first guess is correct, here is scaffolded code for this analysis.
### 3.4: Subset data to articles about HWI
group3sub1 <- CSPdf %>%
filter(...) # How would you filter to the articles where the Topic is "HWI"?
## 3.4.1: Summarize the values of sentiment by country
group3sum <- group3sub1 %>%
group_by(Country) %>% # group the data by Country
summarise(MedianSent=median(Sentiment)) # calculate the median sentiment in each countryThe next step is to combine these subset data with global GIS data.
### 3.5: Pull in global GIS data
worldDF <- world # store the dataset for the world in worldDF, world data frame
### 3.6: Join the subset data with the GIS data
speciesMap <- worldDF %>% # take the worldDF object and feed forward
left_join(group3sum, by=c("name_long"="Country")) # combine the group3sum data with worldDF by country,
# getting the data to match on the columns "name_long" (countries) and "Country"We’re ready to plot the data now.
### 3.7: Generating the map
# Your task: How do you color in countries based on Sentiment?
# Hint: you may want to change the ... in fill=...
# Perhaps using MedianSent makes sense
g3 <- ggplot(data=speciesMap, aes(fill=...)) # use the speciesMap object to create a map
g3 # inspect the current output
g3 <- g3 + geom_sf(aes(fill=...),color="gray47",size=0.05) # add on the boundaries of countries, change the ... for fill
g3 # inspect the current output
g3 <- g3 + labs(x="",y="",fill="...: ") # change the names of the variables as needed
g3 # inspect the current output
g3 <- g3 + theme_minimal() # change the display style
g3 # inspect the current output
g3 <- g3 + theme(legend.position = "top") # move the legend bar to the top
g3 # display the final plot