R: Introduction to the Tidyverse
1. data wrangling
library(gapminder) /*加载数据集*/
library(dplyr)
gapminder /*check the dataset*/The filter verb
# Single condition
gapminder %>%
filter(year == 1957)
# Multiple conditions
gapminder %>%
filter(year == 2002, country == "China")* note: %>% take whatever is before it, and feed it in the next steps
The arrange verb
arrange sorts the observations in a dataset, in ascending or descending order based on one of its variables
# Sort in ascending order of lifeExp
gapminder %>%
arrange(lifeExp)
# Sort in descending order of lifeExp
gapminder %>%
arrange(desc(lifeExp))The mutate verb
gapminder %>%
mutate(lifeExpMonths = 12 * lifeExp)Combine the verbs
# Filter, mutate, and arrange the gapminder dataset
gapminder %>%
filter(year == 2007) %>%
mutate(lifeExpMonths = 12 * lifeExp) %>%
arrange(desc(lifeExpMonths))2. data visualization
1) assign the variable
创建筛选后的新变量
# Create gapminder_1952
gapminder_1952 <- gapminder %>%
filter(year == 1952)
2) create a scatter plot
ggplot(gapminder_1952, aes(x = pop, y = gdpPercap)) +
geom_point()3) log scales
图像变形，log转换
ggplot(gapminder_1952, aes(x = pop, y = lifeExp)) +
geom_point() +
scale_x_log10()
4) additional aesthetics (size, color)
为小类添加颜色， 标识点的大小
ggplot(gapminder_1952, aes(x = pop, y = lifeExp, size = gdpPercap, color = continent)) +
geom_point() +
scale_x_log10()
5) faceting
facet can visualize subgroup
ggplot(gapminder_1952, aes(x = pop, y = lifeExp)) +
geom_point() +
facet_wrap(~ continent) +
scale_x_log10()按照大洲进行可视化，如图：
Take another example,
ggplot(gapminder, aes(x = gdpPercap, y = lifeExp, color = continent, size = pop)) +
geom_point() +
scale_x_log10() +
facet_wrap(~year)如图
3. grouping and summarizing
1) summarize()
e.g. calculate the median life expectancy into a column
gapminder %>%
filter(year == 1957) %>%
summarize(medianLifeExp = median(lifeExp))e.g. two orders in summarize(）
gapminder %>%
filter(year == 1957) %>%
summarize(medianLifeExp = median(lifeExp), maxGdpPercap = max(gdpPercap))2) The group_by verb
可以用于分组总结
gapminder %>%
filter(year == 1957) %>%
group_by(continent) %>%
summarize(medianLifeEXP = median(lifeExp), maxGdpPercap = max(gdpPercap))3) visualizing over time
# Summarize medianGdpPercap within each continent within each year: by_year_continent
by_year_continent <- gapminder %>%
group_by(continent, year) %>%
summarize(medianGdpPercap = median(gdpPercap))
# Plot the change in medianGdpPercap in each continent over time
ggplot(by_year_continent, aes(x = year, y = medianGdpPercap, color = continent)) +
geom_point() +
expand_limits(y = 0)note:
expand_limits(y=0): y-axis starts at zero
如图
4. types of visualization
4.1 line plot
geom_line()
# Summarize the median gdpPercap by year, then save it as by_year
by_year <- gapminder %>%
group_by(year) %>%
summarize(medianGdpPercap=median(gdpPercap))
# Create a line plot showing the change in medianGdpPercap over time
ggplot(by_year, aes(x=year, y=medianGdpPercap)) +
geom_line() +
expand_limits(y=0)
4.2 bar plot
geom_col()
# Create a bar plot showing medianGdp by continent
ggplot(by_continent, aes(x=continent, medianGdpPercap)) +
geom_col()
4.3 histograms
geom_histogram()
notes: in the bracket, we can add "bins=*number" or "binwidth=*number", etc.
geom_histogram()
gapminder_1952 <- gapminder %>%
filter(year == 1952)
# Create a histogram of population (pop), with x on a log scale
ggplot(gapminder_1952, aes(x=pop)) +
geom_histogram() +
scale_x_log10()
4.4 boxplots
geom_boxplot()
gapminder_1952 <- gapminder %>%
filter(year == 1952)
# Create a boxplot comparing gdpPercap among continents
ggplot(gapminder_1952, aes(x=continent, y=gdpPercap)) +
geom_boxplot() +
scale_y_log10()
4.5 other details
1) add a title to the graph
+ ggtitle(" *title ")