In this chapter, we will discuss the basics of wrangling an individual data set.
library("tidyverse")
── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr 1.1.4.9000 ✔ readr 2.1.5
✔ forcats 1.0.0 ✔ stringr 1.5.1
✔ ggplot2 3.5.2 ✔ tibble 3.3.0
✔ lubridate 1.9.4 ✔ tidyr 1.3.1
✔ purrr 1.0.4
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag() masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
23.1 Mutate
The mutate() function allows you to create new variables in a data.frame. Typically this involves changing units, performing calculations with other variables, and refactoring factors.
# Mutate can create variablesToothGrowth |>mutate(length = len) |>head()
# Mutate numericToothGrowth |>mutate(`Length (cm)`= len /10# `len` is in the data.frame ) |>summary() # `len` is in mm while `Length (cm)` is in cm
len supp dose Length (cm)
Min. : 4.20 OJ:30 Min. :0.500 Min. :0.420
1st Qu.:13.07 VC:30 1st Qu.:0.500 1st Qu.:1.308
Median :19.25 Median :1.000 Median :1.925
Mean :18.81 Mean :1.167 Mean :1.881
3rd Qu.:25.27 3rd Qu.:2.000 3rd Qu.:2.527
Max. :33.90 Max. :2.000 Max. :3.390
# Overwrite existing variableToothGrowth |>mutate(len = len /10# `len` is in the data.frame ) |>summary() # `len` is in cm
len supp dose
Min. :0.420 OJ:30 Min. :0.500
1st Qu.:1.308 VC:30 1st Qu.:0.500
Median :1.925 Median :1.000
Mean :1.881 Mean :1.167
3rd Qu.:2.527 3rd Qu.:2.000
Max. :3.390 Max. :2.000
# Use object outside data.framemm_per_cm <-10ToothGrowth |>mutate(len = len / mm_per_cm ) |>summary()
len supp dose
Min. :0.420 OJ:30 Min. :0.500
1st Qu.:1.308 VC:30 1st Qu.:0.500
Median :1.925 Median :1.000
Mean :1.881 Mean :1.167
3rd Qu.:2.527 3rd Qu.:2.000
Max. :3.390 Max. :2.000
# Use updated variableToothGrowth |>mutate(# Scale len between 0 and 1len = len -min(len),len = len /max(len) # reused `len` here ) |>summary() # len now ranges between 0 and 1
len supp dose
Min. :0.0000 OJ:30 Min. :0.500
1st Qu.:0.2988 VC:30 1st Qu.:0.500
Median :0.5067 Median :1.000
Mean :0.4920 Mean :1.167
3rd Qu.:0.7096 3rd Qu.:2.000
Max. :1.0000 Max. :2.000
23.1.2 Categorical
We commonly need to convert between factor and character representations of categorical variables.
# Convert between character and factord <- ToothGrowth |>select(supp) |># only keep `supp` columnmutate(supp_ch =as.character(supp), # convert to charactersupp_fa =as.factor(supp_ch) # convert to factor ) d |>summary() # summary() only informative for factor
# table() is always informatived$supp_ch |>table()
OJ VC
30 30
d$supp_fa |>table()
OJ VC
30 30
At this point, the main distinction between character variables and factor variables is that you can change the order of factor variables while character variables will always be in alphabetical order.
# Change order of factor variable# Cannot change order of character variablesd |>mutate(supp_fa =factor(supp_fa, levels =c("VC", # put VC first"OJ")) # then OJ ) |>summary()
# Recode character or factor levelsd |>mutate(supp_ch =fct_recode(supp_ch, `Ascorbic Acid`="VC",`Orange Juice`="OJ" ),supp_fa =fct_recode(supp_fa,`Ascorbic Acid`="VC",`Orange Juice`="OJ" ) ) |>summary()
supp supp_ch supp_fa
OJ:30 Orange Juice :30 Orange Juice :30
VC:30 Ascorbic Acid:30 Ascorbic Acid:30
23.1.3 Both
Here we will show you how to utilize the mutate() function to perform a number of calculations.
# Diamondsd <- diamonds |># Rather than precalculating depth # we will calculate depth in a scriptselect(-depth) |>mutate(# Calculate depthdepth =2* z / (x+y), # see ?diamonds for formuladepth =100* depth, # make depth a percent# Calculate $/weightprice_per_carat = price / carat,# Reorder cutcut =factor(cut, levels =c("Ideal","Premium","Very Good","Good","Fair" )) )# View calculated variablesd |>select(price_per_carat, depth, cut) |>summary()
price_per_carat depth cut
Min. : 1051 Min. : 0.00 Ideal :21551
1st Qu.: 2478 1st Qu.: 61.04 Premium :13791
Median : 3495 Median : 61.84 Very Good:12082
Mean : 4008 Mean : 61.74 Good : 4906
3rd Qu.: 4950 3rd Qu.: 62.53 Fair : 1610
Max. :17829 Max. :619.28
NA's :7
# View observations with NA depthd |>filter(is.na(depth)) |># see filter() belowselect(depth, x, y, z) # NaN stands for `Not a number`
# A tibble: 7 × 4
depth x y z
<dbl> <dbl> <dbl> <dbl>
1 NaN 0 0 0
2 NaN 0 0 0
3 NaN 0 0 0
4 NaN 0 0 0
5 NaN 0 0 0
6 NaN 0 0 0
7 NaN 0 0 0
23.2 Rename
We can use the rename function to rename an existing variable in the data set. If we want to use names that are not valid R object names, we need to enclose the name using backticks.
# We didn't actually change the object so the names in the ToothGrowth data set has not changed.ToothGrowth |>names()
[1] "len" "supp" "dose"
# Save the renamed data setd <- ToothGrowth |>rename(`Dose (mg/day)`= dose,Supplement = supp,`Length (mm)`= len ) # Saved data set has the new namesd |>names()
