DANL 310 Lecture 06

p <- ggplot(data = gapminder,
            mapping = aes(x = year,
                          y = gdpPercap))

p + geom_line(color="gray70", aes(group = country)) +
    geom_smooth(size = 1.1, method = "loess", se = FALSE) +
    scale_y_log10(labels=scales::dollar) +
    facet_wrap(~ continent, ncol = 5) +
    labs(x = "Year", y = "GDP per capita",
         title = "GDP per capita on Five Continents") + 
    theme(axis.text.x = element_text(angle = 45),
          axis.title.x = element_text(margin = margin(t = 20)))

• The facet_wrap() function is best used when you want a series of small multiples based on a single categorical variable.

p <- ggplot(data = gss_sm,
            mapping = aes(x = age, y = childs))

p + geom_point(alpha = 0.2) +
    geom_smooth() +
    facet_grid(sex ~ race)

• The facet_grid() function is best used when you cross-classify some data by two categorical variables.

  • e.g., the relationship between the age and the number of children by sex and race

p <- ggplot(data = gss_sm,
            mapping = aes(x = bigregion))
p + geom_bar()

• The count statistic is the one geom_bar() uses by default.

p <- ggplot(data = gss_sm,
            mapping = aes(x = bigregion))
p + geom_bar(mapping = aes(y = ..prop..))

• If we want a chart of relative frequencies rather than counts, we will need to get the prop statistic instead.

• Our call to statistic from the aes() function generically looks like this: <mapping> = <..statistic..>.

p <- ggplot(data = gss_sm,
            mapping = aes(x = bigregion))
p + geom_bar(mapping = aes(y = ..prop.., group = 1)) 

• We need to tell ggplot to ignore the x-categories when calculating denominator of the proportion, and use the total number observations instead.

  • To do so we specify group = 1 inside the aes() call.

table(gss_sm$religion)

## 
## Protestant   Catholic     Jewish       None      Other 
##       1371        649         51        619        159

p <- ggplot(data = gss_sm,
            mapping = aes(x = religion, color = religion))
p + geom_bar()

• If we map religion to color, only the border lines of the bars will be assigned colors, and the insides will remain gray.

p <- ggplot(data = gss_sm,
            mapping = aes(x = religion, fill = religion))
p + geom_bar() + guides( fill = "none" )

• If the gray bars look boring and we want to fill them with color instead, we can map the religion variable to fill in addition to mapping it to x.

• The default legend is about the color variable, which is redundant.

  • If we set guides(fill = "none"), the legend is removed.

p <- ggplot(data = gss_sm,
            mapping = aes(x = bigregion, fill = religion))
p + geom_bar()

• A more appropriate use of the fill aesthetic with geom_bar() is to cross-classify two categorical variables.

  • The default output of such geom_bar() is a stacked bar chart, with counts on the y-axis.

p <- ggplot(data = gss_sm,
            mapping = aes(x = bigregion, fill = religion))
p + geom_bar(position = "fill")

• An alternative choice is to set the position argument to "fill".
  • It is to compare proportions across groups.

p <- ggplot(data = gss_sm,
            mapping = aes(x = bigregion, fill = religion))
p + geom_bar(position = "dodge",
             mapping = aes(y = ..prop..))

• We can use position="dodge" to make the bars within each region of the country appear side by side.

p <- ggplot(data = gss_sm,
            mapping = aes(x = bigregion, fill = religion))
p + geom_bar(position = "dodge",
             mapping = aes(y = ..prop.., group = religion))

• In this case our grouping variable is religion, so we might try mapping that to the group aesthetic.

p <- ggplot(data = gss_sm,
            mapping = aes(x = religion))
p + geom_bar(position = "dodge",
             mapping = aes(y = ..prop.., group = bigregion)) +
    facet_wrap(~ bigregion, ncol = 1)

• Instead, we can ask ggplot to give us a proportional bar chart of religious affiliation, and then facet that by region.
  • The proportions are calculated within each panel, which is the breakdown we wanted.

?midwest
glimpse(midwest)
skim(midwest)
view(midwest)

• The ggplot package comes with a dataset, midwest, containing information on counties in several midwestern states of the USA.

p <- ggplot(data = midwest,
            mapping = aes(x = area))
p + geom_histogram()

• By default, the geom_histogram() function will choose a bin size for us based on a rule of thumb.

p <- ggplot(data = midwest,
            mapping = aes(x = area))
p + geom_histogram(bins = 10)

• When drawing histograms it is worth experimenting with bins and also optionally the origin of the x-axis.

oh_wi <- c("OH", "WI")

p <- ggplot(data = subset(midwest, subset = state %in% oh_wi),
            mapping = aes(x = percollege, fill = state))
p + geom_histogram(alpha = 0.4, bins = 20)

• We can facet histograms by some variable of interest, or as here we can compare them in the same plot using the fill mapping.
  • We subset the data here to pick out just two states.
  • The %in% operator is a convenient way to filter on more than one term in a variable when using subset()

p <- ggplot(data = midwest,
            mapping = aes(x = area))
p + geom_density()

• When working with a continuous variable, an alternative to binning the data and making a histogram is using the geom_density() function.
  • It calculates a kernel density estimate of the underlying distribution.
  • The area under the whole curve is usually 1.

p <- ggplot(data = midwest,
            mapping = aes(x = area, fill = state, color = state))
p + geom_density(alpha = 0.3)

• We can use color (for the lines) and fill (for the body of the density curve).
  • These figures often look quite nice.
  • When there are several filled areas on the plot, as in this case, the overlap can become hard to read.

oh_wi <- c("OH", "WI")
p <- ggplot(data = subset(midwest, subset = state %in% oh_wi),
            mapping = aes(x = area, fill = state, color = state))
p + geom_density(alpha = 0.3, mapping = (aes(y = ..scaled..)))

• For geom_density(), the stat_density() function can return its default ..density.. statistic, or ..scaled.., which will give a proportional density estimate.

• It can also return a statistic called ..count.., which is the density times the number of points. This can be used in stacked density plots.

p <- ggplot(data = titanic,
            mapping = aes(x = fate, y = percent, fill = sex))

p + geom_bar(position = "dodge", stat = "identity") + theme(legend.position = "top")

• We can tell geom_bar() not to do any work on the variable before plotting it.
  • stat = "identity" means “don’t do any summary calculations”.

p <- ggplot(data = oecd_sum,
            mapping = aes(x = year, y = diff, fill = hi_lo))
p + geom_col() + guides( fill = "none" ) + 
  labs(x = NULL, y = "Difference in Years",
       title = "The US Life Expectancy Gap",
       subtitle = "Difference between US and OECD
                   average life expectancies, 1960-2015",
       caption = "Data: OECD. After a chart by Christopher Ingraham,
                  Washington Post, December 27th 2017.")

• For convenience ggplot also provides a related geom, geom_col(), which has exactly the same effect as geom_bar() but its default stat is stat = "identity".

• The position argument in geom_bar() and geom_col() can also take the value of "identity".

  • position = "identity" means “just plot the values as given”.