Initial ggplot set up

setwd("~/Desktop/University of Utah PhD /Course Work/Fall 2022 Semester/GEOG6000_Data Analysis/lab05b")

library(ggplot2)
library(dplyr) #for helping rename/shape data
library(ggpubr) #additional package for producing plots
library(plotly) #interactive figures
library(ggthemes) #adding additional themes

penguins = read.csv("../datafiles/penguins.csv")
orange = read.csv("../datafiles/orange.csv")
gapdata = read.csv("../datafiles/gapminderData5.csv")
VADeaths = read.csv("../datafiles/VADeaths.csv")

str(penguins)

## 'data.frame':    344 obs. of  8 variables:
##  $ species          : chr  "Adelie" "Adelie" "Adelie" "Adelie" ...
##  $ island           : chr  "Torgersen" "Torgersen" "Torgersen" "Torgersen" ...
##  $ bill_length_mm   : num  39.1 39.5 40.3 NA 36.7 39.3 38.9 39.2 34.1 42 ...
##  $ bill_depth_mm    : num  18.7 17.4 18 NA 19.3 20.6 17.8 19.6 18.1 20.2 ...
##  $ flipper_length_mm: int  181 186 195 NA 193 190 181 195 193 190 ...
##  $ body_mass_g      : int  3750 3800 3250 NA 3450 3650 3625 4675 3475 4250 ...
##  $ sex              : chr  "male" "female" "female" NA ...
##  $ year             : int  2007 2007 2007 2007 2007 2007 2007 2007 2007 2007 ...

str(orange)

## 'data.frame':    35 obs. of  3 variables:
##  $ Tree         : int  1 1 1 1 1 1 1 2 2 2 ...
##  $ age          : int  118 484 664 1004 1231 1372 1582 118 484 664 ...
##  $ circumference: int  30 58 87 115 120 142 145 33 69 111 ...

str(gapdata)

## 'data.frame':    1704 obs. of  6 variables:
##  $ country  : chr  "Afghanistan" "Afghanistan" "Afghanistan" "Afghanistan" ...
##  $ year     : int  1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 ...
##  $ pop      : num  8425333 9240934 10267083 11537966 13079460 ...
##  $ continent: chr  "Asia" "Asia" "Asia" "Asia" ...
##  $ lifeExp  : num  28.8 30.3 32 34 36.1 ...
##  $ gdpPercap: num  779 821 853 836 740 ...

str(VADeaths)

## 'data.frame':    5 obs. of  5 variables:
##  $ X           : chr  "50-54" "55-59" "60-64" "65-69" ...
##  $ Rural.Male  : num  11.7 18.1 26.9 41 66
##  $ Rural.Female: num  8.7 11.7 20.3 30.9 54.3
##  $ Urban.Male  : num  15.4 24.3 37 54.6 71.1
##  $ Urban.Female: num  8.4 13.6 19.3 35.1 50

Data Reshaping

##Reshaping data from wide to long format
colnames(VADeaths)[1] = "Age" ##Renaming the column header to age

library(tidyr) #package to reshape data


##First, names_to specifies the names of new columns for demographic class data. Second, names_sep specifies the character for separating the column names as values into the respective columns; in this case, the period (‘.’), which is a reserved expression and thus needs to be escaped with two back ticks (‘\’). Finally, values_to specifies the name of the new column for death rate data.

VADeaths2 = pivot_longer(data = VADeaths,
                          cols = Rural.Male:Urban.Female,
                          names_to = c("Residence", "Gender"),
                          names_sep = "\\.",
                          values_to = "DeathRate")


VADeaths2

## # A tibble: 20 × 4
##    Age   Residence Gender DeathRate
##    <chr> <chr>     <chr>      <dbl>
##  1 50-54 Rural     Male        11.7
##  2 50-54 Rural     Female       8.7
##  3 50-54 Urban     Male        15.4
##  4 50-54 Urban     Female       8.4
##  5 55-59 Rural     Male        18.1
##  6 55-59 Rural     Female      11.7
##  7 55-59 Urban     Male        24.3
##  8 55-59 Urban     Female      13.6
##  9 60-64 Rural     Male        26.9
## 10 60-64 Rural     Female      20.3
## 11 60-64 Urban     Male        37  
## 12 60-64 Urban     Female      19.3
## 13 65-69 Rural     Male        41  
## 14 65-69 Rural     Female      30.9
## 15 65-69 Urban     Male        54.6
## 16 65-69 Urban     Female      35.1
## 17 70-74 Rural     Male        66  
## 18 70-74 Rural     Female      54.3
## 19 70-74 Urban     Male        71.1
## 20 70-74 Urban     Female      50

qplots()

These are quick plots with similar syntax to base r.

Scatterplots

qplot(bill_depth_mm, bill_length_mm, data = penguins)

## Warning: Removed 2 rows containing missing values (geom_point).

qplot(bill_depth_mm, bill_length_mm, 
      data = penguins,
      col = species, ##color by species
      size = body_mass_g, ##symbol size by body mass
      alpha = I(0.7), ##make the symbols slightly transparent to see overlap
      xlab = "Bill Depth (mm)",
      ylab = "Bill Length (mm)",
      main = "Penguin Bill Measurements")

## Warning: Removed 2 rows containing missing values (geom_point).

Histograms and Boxplots

qplot(bill_length_mm,
      data = penguins,
      fill = 'coral2',
      binwidth = 1,
      geom = 'histogram')

## Warning: Removed 2 rows containing non-finite values (stat_bin).

qplot(species, bill_depth_mm,
      data = penguins,
      geom = 'boxplot',
      col = species,
      main = "Bill Depth (mm) by Species",
      xlab = "Species",
      ylab = "Bill Depth (mm)")

## Warning: Removed 2 rows containing non-finite values (stat_boxplot).

Line Plots

qplot(age, circumference,
      data = orange,
      geom = 'line')

#Reshaping the orange data set to make the lines factors instead of continuous data. This will break it up in to separate lines

orange$Tree = factor(orange$Tree)
orange$Tree

##  [1] 1 1 1 1 1 1 1 2 2 2 2 2 2 2 3 3 3 3 3 3 3 4 4 4 4 4 4 4 5 5 5 5 5 5 5
## Levels: 1 2 3 4 5

qplot(age, circumference,
      data = orange,
      geom = 'line',
      col = Tree,
      main = "Orange Tree Circumference by Age",
      ylab = "Circumference (cm)",
      xlab = "Age (years)")

Advanced Figures with ggplot()

From Simon’s Notes:

The ggplot() function offers more control over your plots. It works in quite a different way to the other plotting functions, that start with a base plot, then add other points, lines, etc to the figure. Instead, ggplot creates a plot object, which can be adjusted and added to as you proceed.

In order to understand how ggplot makes a figure, we need to establish what the fundamental parts are of every data graph. They are:

Aesthetics – these are the roles that the variables play in each graph. A variable may control where points appear, the color or shape of a point, the height of a bar and so on.
Geometries – these are the geometric objects which represent the data: points, lines, bars, …
Statistics – these are the functions which add some interpretation to the data, e.g. best fit line, location of median, etc
Scales – these are legends that show the relationship between variables and different symbols or colors (e.g. circular symbols represent females while squares represent males)
Facets – these are groups in your data which may be used to make multiple graphs, each for one of the groups. For example, faceting by gender would cause the graph to repeat for the two genders.

penguin.plot = ggplot(data = penguins,aes(x = bill_depth_mm, y = bill_length_mm))
penguin.plot ##The initial plot is just that, a plot. We need to tell it geometry

penguin.plot + geom_point() ## telling the plot to use point geometry. If we just type penguin.plot, it will just draw the first figure with no points.

## Warning: Removed 2 rows containing missing values (geom_point).

#Alternate setup?

penguin.plot2 = ggplot( data = penguins, aes(x = bill_depth_mm, y = bill_length_mm, color = species)) + ##Initial Plot set up; color by species
  geom_point() + ##Point Geometry
  labs(title = "Penguin Bill Measurments by Species",
       subtitle = "By Dave Leydet",
       caption = "GEOG6000 Demonstration - University of Utah") + ##Title/Subtitle Labels
  xlab("Bill Depth (mm)") +
  ylab("Bill Length (mm)") ##x and y labels
  

  
  
penguin.plot2

## Warning: Removed 2 rows containing missing values (geom_point).

Histograms

life.exp.plot = ggplot( data = gapdata, aes(x = lifeExp, fill = continent), position = 'identity') + ##Fill - use color to fill by continent. Position by stacking bars starting at zero (0)
  geom_histogram(binwidth = 1) + ##histogram with a bin width of 1
  labs(title = "Life Expectancy Histogram",
       subtitle = "By Dave Leydet") +
  xlab("Life Expectancy")

life.exp.plot

Facet (Splitting Figures by Variables)

life.exp.plot2 = ggplot( data = gapdata, aes(x = lifeExp)) + ##color is for the outline
  geom_histogram(binwidth = 2, color = 'lightblue', fill = 'white') + ##histogram with a bin width of 1. Put the color here so it doesnt show as a label
  labs(title = "Life Expectancy Histogram",
       subtitle = "By Dave Leydet") +
  xlab("Life Expectancy") +
  facet_wrap(~continent) ##Need to use a tilda (~) here to identify the variable

life.exp.plot2

##Use ```{r fig.width = 7, fig.height = 10} to adjust the size of the figure once is knit'd to html. Reminder - this is in the chunk header.


life.exp.plot3 = ggplot( data = gapdata, aes(x = lifeExp)) + ##color is for the outline
  geom_histogram(binwidth = 2, color = 'darkgoldenrod1', fill = 'darkgoldenrod1') + ##histogram with a bin width of 1. Put the color here so it doesnt show as a label
  labs(title = "Life Expectancy Histogram",
       subtitle = "By Dave Leydet") +
  xlab("Life Expectancy") +
  facet_wrap(year ~ continent) ##split by year and 

life.exp.plot3

Density Plots

life.exp.plot4 = ggplot( data = gapdata, aes(x = lifeExp)) + 
  labs(title = "GapMinder Life Expectancy",
       subtitle = "By Dave Leydet") +
  xlab("Life Expectancy") 

life.exp.plot4

life.exp.plot4 + 
  geom_density(aes(fill = continent), alpha = 0.4) +
  ylab("Density")

Boxplots

lcplot = ggplot( data = gapdata, aes(x = continent, y = lifeExp))
lcplot + geom_boxplot(aes(fill = continent))

Barcharts

lcplot2 = ggplot(gapdata, aes(x = continent, fill = continent))

lcplot2 + geom_bar()

#geom_bar(stat = 'identity') overrides the default behavior of r to create a bar graph based on the count of x observations instead of just y values. geom_col() accomplishes the same thing by default. 

dr.plot = ggplot(data = VADeaths2, aes(x = Age, y = DeathRate)) +
  geom_bar(stat = 'identity') +
  xlab("Age") +
  ylab("Death Rate") +
  labs(title = "Death Rate by Age Group")


dr.plot2 = ggplot(data = VADeaths2, aes(x = Age, y = DeathRate, fill = Residence)) +
  geom_bar(stat = 'identity') +
  xlab("Age") +
  ylab("Death Rate") +
  labs(title = "Death Rate by Age Group")

##Use ggarange from the ggpubr package to arrange the plots in an array
plot.comb = ggarrange(dr.plot, dr.plot2,
                     ncol = 2,
                     nrow = 1)

plot.comb

#Split out the residence to be side by side using position = 'dodge'

dr.plot3 = ggplot(data = VADeaths2, aes(x = Age, y = DeathRate, fill = Residence)) +
  geom_bar(stat = 'identity', position = 'dodge') +
  xlab("Age") +
  ylab("Death Rate") +
  labs(title = "Death Rate by Age Group")

dr.plot3

##Adding a facet wrap to break out the charts by gender ( ~ Gender syntax) and flipping the x and y axes (coord_flip syntax)

dr.plot4 = ggplot(data = VADeaths2, aes(x = Age, y = DeathRate, fill = Residence)) +
  geom_bar(stat = 'identity', position = 'dodge') +
  xlab("Age") +
  ylab("Death Rate") +
  labs(title = "Death Rate by Age Group") +
  facet_wrap( ~ Gender) +
  coord_flip()

dr.plot4

Scatterplots

#Basic scatter plot setup.
#Using the variable "+" syntax to build/change/modify the plot. Can be used for all plots.

scttr.plot1 = ggplot(data = gapdata, aes(x = gdpPercap, y = lifeExp))

scttr.plot1 + geom_point()

##Log transformation options

## Direct option

scttr.plot2 = ggplot(data = gapdata, aes(x = log10(gdpPercap), y = lifeExp)) +
  geom_point() + 
  labs(title = "Direct Transformation")

##Keeps the original values on the axis labels.
scttr.plot3 = scttr.plot1 +
  geom_point() + 
  scale_x_log10() +
  labs(title = "Scale_x_transformation")

##Create side by side ggplots
ggarrange(scttr.plot2, scttr.plot3,
         ncol = 2,
         nrow = 1)

##Add a trend line with geom_smooth() argument fitted using a spline or local regression (loess).

##Help with the theme and color by Abby

scttr.plot1 + geom_point(color = "white") + scale_x_log10() + geom_smooth(color = "purple") + theme_dark() + labs(title = "By Abby and David Leydet")

## `geom_smooth()` using method = 'gam' and formula 'y ~ s(x, bs = "cs")'

scttr.plot1 + 
  geom_point(aes(color = continent)) +
  scale_x_log10() +
  theme_bw()

##Adding just a single linear model line through the data

scttr.plot1 + 
  geom_point(aes(color = continent)) +
  scale_x_log10() +
  theme_bw() +
  geom_smooth(method = 'lm', color = "black")

## `geom_smooth()` using formula 'y ~ x'

##Create another scatter plot with the main object colored by continent and scaled. Then putting the smoothing lines through each group of countries by continent. 

scttr.plot4 = ggplot(data = gapdata, aes(x = gdpPercap, y = lifeExp, color = continent)) + scale_x_log10()

scttr.plot4 + 
  geom_point() +
  geom_smooth(method = 'lm') +
  theme_bw()

## `geom_smooth()` using formula 'y ~ x'

#Subtract out geom_point() to just observe the trend lines

scttr.plot4 +
  geom_smooth(method = 'lm') +
  theme_bw()

## `geom_smooth()` using formula 'y ~ x'

#facet wraps

scttr.plot4 +
  geom_point() +
  facet_wrap(~continent) +
  theme_cleveland()

##add trendlines to each facet

scttr.plot4 +
  geom_point() +
  facet_wrap(~continent) +
  theme_cleveland() +
  geom_smooth(method = 'lm', color = "black") ##need to color the trendlines black to see them

## `geom_smooth()` using formula 'y ~ x'

##Removing warnings by including ```{r warning=FALSE}
##facet wrap by year and continent

scttr.plot4 +
  geom_point() +
  facet_grid(year ~ continent) +
  theme_cleveland() +
  geom_smooth(method = 'lm', color = "black")

## `geom_smooth()` using formula 'y ~ x'

Line Plots

line.plot1 = ggplot(data = gapdata, aes(x = year, y = lifeExp))

line.plot1 + geom_line(aes(group = country, color = continent)) +
  facet_wrap(~continent) +
  theme_bw()

##Plotly version of the previous figure
##  geom_line(aes(group = country, color = continent) shows the data by country and allows the hover to display the data for each strand

line.plot2 = ggplot(data = gapdata, aes(x = year, y = lifeExp)) + 
  geom_line(aes(group = country, color = continent)) +
  facet_wrap(~continent) +
  theme_bw()

ggplotly(line.plot2) ##convert a ggplot to plotly

Subsetting Data for Lines

## United States Subset

usa = subset(gapdata, country == "United States")

usa.plot = ggplot(data = usa, aes(x = year, y = lifeExp)) +
  geom_line(color = "blue") +
  geom_point() +
  theme_light()

ggplotly(usa.plot)

slct.countries = c("Canada", "Rwanda", "Cambodia", "Mexico", "United States")

slct.country.plot = ggplot(subset(gapdata, country %in% slct.countries), aes(x = year, y = lifeExp, color = country)) +
  geom_line() +
  geom_point() +
  theme_bw() +
  labs(title = "Life Expectancy by Country") +
  xlab("Year") +
  ylab("Life Expectancy")

ggplotly(slct.country.plot)

Saving Figures

scttr.plot5 = ggplot(data = gapdata, aes(x = gdpPercap, y = lifeExp, color = continent)) + 
  scale_x_log10() +
  geom_point() +
  geom_smooth(method = 'lm') +
  theme_bw() +
  labs(title = "Life Expectancy by Country GDP") +
  xlab("GDP (log scaled)") +
  ylab("Life Expectancy") +
  guides(color = guide_legend(title = "Continent"))

scttr.plot5

## `geom_smooth()` using formula 'y ~ x'

##Saving it

##adjusting the size with height and width arguments
ggsave("lifeExp_plot.jpg", scttr.plot5,
       width = 7,
       height = 7)

## `geom_smooth()` using formula 'y ~ x'

#save as a .pdf
ggsave("lifeExp_plot.pdf", scttr.plot5)

## Saving 7 x 5 in image
## `geom_smooth()` using formula 'y ~ x'

Themes

scttr.plot1 +
  geom_point() +
  scale_x_log10() +
  theme_gdocs()

scttr.plot1 +
  geom_point() +
  scale_x_log10() +
  theme_fivethirtyeight()

Please check out the ggplot cheatsheets from RStudio!

ggplot practice

David Leydet

2022-09-27

Initial ggplot set up

Data Reshaping

qplots()

Scatterplots

Histograms and Boxplots

Line Plots

Advanced Figures with ggplot()

Histograms

Facet (Splitting Figures by Variables)

Density Plots

Boxplots

Barcharts

Scatterplots

Line Plots

Subsetting Data for Lines

Saving Figures

Themes