motivation.md

layout	title	subtitle
slides	R for reproducible scientific analysis	Why Use R?

Why R?

• Powerful statistical analysis
• and powerful visualisation
• integrated elegantly

What We'll Accomplish

• Get to know R and RStudio
• Analyze a meaningful data set
• Extract insights and deliver them visually
• Leave ready to learn more R independently

R loves ingesting data

gapminder <- read.csv(
  "data/gapminder-FiveYearData.csv",
  header=TRUE,
  sep=',')

Data w/ column names

head(gapminder, 1) # Show me the first row

country year pop continent lifeExp gdpPercap 1 Afghanistan 1952 8425333 Asia 28.801 779.4453

Quickly graph ...

ggplot(
  data=gapminder,
  aes(x=lifeExp, y=gdpPercap)
) + geom_point()

... to see what we have

Let's graph more factors

ggplot(
  data=gapminder,
  aes(x=year, y=lifeExp, by=country, colour=continent)
) + geom_line()
  + geom_point()

Pretty!

dyplr gives us ...

library(dplyr)
cors <- gapminder %>%
  group_by(year) %>%
  summarise(
    gdpPercap.lifeExp = cor(gdpPercap, lifeExp),
    gdpPercap.pop = cor(gdpPercap, pop),
    pop.lifeExp = cor(pop, lifeExp))

... pairwise correlations

head(cors, 1)
Source: local data frame [1 x 4]
year gdpPercap.lifeExp gdpPercap.pop  pop.lifeExp
1 1952         0.2780236   -0.02526041 -0.002724782

Restructuring the table ...

library(tidyr)

Error in library(tidyr): there is no package called 'tidyr'

tidy.cors <- cors %>% gather(
  variables, correlation,
  gdpPercap.lifeExp, gdpPercap.pop,
  pop.lifeExp)

Error in function_list[[k]](value): could not find function "gather"

... a subtle art ...

head(tidy.cors, 1)

Source: local data frame [1 x 3]
  year         variables   correlation
1 1952 gdpPercap.lifeExp   0.2780236

... produces great results