janitor

janitor

Simple tools for examining and cleaning dirty data.

Clean Names

library(janitor)

# Clean column names
df <- clean_names(df)

# Options
df <- clean_names(df, case = "snake")      # default
df <- clean_names(df, case = "lower_camel")
df <- clean_names(df, case = "upper_camel")
df <- clean_names(df, case = "screaming_snake")
df <- clean_names(df, case = "title")

# Make names from vector
make_clean_names(c("First Name", "Last Name"))

Tabulations

# One-way tabulation
df %>% tabyl(category)

# Two-way tabulation
df %>% tabyl(category, group)

# Three-way tabulation
df %>% tabyl(category, group, year)

Tabyl Adornments

df %>%
  tabyl(category, group) %>%
  adorn_totals("row") %>%           # Add row totals
  adorn_totals("col") %>%           # Add column totals
  adorn_percentages("row") %>%      # Convert to row percentages
  adorn_pct_formatting() %>%        # Format percentages
  adorn_ns() %>%                    # Add counts in parentheses
  adorn_title("combined")           # Add title row

Remove Empty

# Remove empty rows and columns
df <- remove_empty(df, which = c("rows", "cols"))

# Remove empty rows only
df <- remove_empty(df, which = "rows")

# Remove empty columns only
df <- remove_empty(df, which = "cols")

# Remove constant columns
df <- remove_constant(df)

Duplicates

# Find duplicates
df %>% get_dupes()

# Find duplicates by specific columns
df %>% get_dupes(name, date)

# Count duplicates
df %>% get_dupes() %>% nrow()

Data Comparison

# Compare data frames
compare_df_cols(df1, df2)

# Check if same columns
compare_df_cols_same(df1, df2)

Excel Dates

# Convert Excel numeric dates
excel_numeric_to_date(44197)  # Returns Date

# Convert Excel datetime
excel_numeric_to_date(44197.5, include_time = TRUE)

# Convert to Excel date
convert_to_date("2021-01-01")
convert_to_datetime("2021-01-01 12:00:00")

Rounding

# Round half up (Excel-style)
round_half_up(2.5)  # Returns 3

# Round to fraction
round_to_fraction(0.37, denominator = 4)  # Returns 0.25