Reshaping: Wide to Long

• Wide means:

  • multiple measurement columns across each row
  • Each row includes a column for
    • revenue
    • profit
    • profit margin

• Ling means:

  • only one measurement per row
  • likely multiple categories
  • The column financialCategory
    • tells me what type of measurement each value is

• melt(){reshape2}

install.packages("reshape2")
library(reshape2)

• format to longData:

longData <- melt(your original data frame, a vector of your category variables)

• wide-to-long code

  • code assumes that all the other columns except fy and company are measurements -- items you might want to plot.

    companiesLong <- melt(companiesData, c("fy", "company"))
    

• lists all the column in the data frame

  • assigning them to either id.vars or measure.vars,
  • changes the new column names from the default "variable" and "value":

companiesLong <- melt(companiesData, id.vars=c("fy", "company"),   measure.vars=c("revenue", "profit", "margin"), variable.name="financialCategory", value.name="amount")

Reference:

• 4 data wrangling tasks in R for advanced beginners | Computerworld