DaSEH Day 5 Cheatsheet - Data Cleaning - Cheatsheet

Oct 14, 2026

DaSEH Day 5 Cheatsheet focuses on data cleaning fundamentals and practical checks for missing or abnormal values. It explains NA, NaN (“Not a Number” from 0/0), and Inf/-Inf from division by zero, and introduces tools such as naniar for exploring missingness and tidyverse stringr/tidyr/dplyr functions for detecting and transforming character data and handling missing rows. It summarizes key R functions like is.na(), is.nan(), is.infinite(), pct_complete(), gg_miss_var(), drop_na(), na_if(), replace_na(), case_when(), mutate(), separate(), unite(), str_detect(), str_replace(), str_replace_all(), and str_sub.

LanguageEnglish

CategoryPresentations

FormatPDF

Pages2

Reading Time1 min

AccessFree to download

Data Cleaning

Major concepts

Examples and core functions

Libraries and functions

DaSEH Day 5 Cheatsheet Data Cleaning Major concepts • Most important rule of data handling: Always look at your data. • NA: General missing data. • NaN: “Not a Number”; occurs when you calculate 0/0 . • Inf and-Inf: Infinity; occurs when you divide a positive or negative number by 0. • naniar: Useful for exploring missingness in large datasets. See the naniar package. • tidyverse (stringr): Useful for detecting and working with parts of character values, especially with filter() and str_detect() . Examples Example of case_when() with mutate() : Orange | > mutate (old = case_when ( age > 1000 ~ TRUE ,.default = FALSE )) Functions Library/Package Piece of code Example of usage What it does Base R Base R Base Rnaniar [is.na](is.na) () is.nan() is.infinite() pct_complete() [is.na](is.na) (x) any( [is.na](is.na) (x)) is.nan(x) is.infinite(x) pct_complete(x) Checks if x is NA. With any(), checks if any values are NA. Checks if x is NaN. Checks if x is Inf or-Inf. Reports the percentage of complete data in x. | Library/Package | Piece of code | Example of usage | What it does | | --- | --- | --- | --- | | naniar | gg_miss_var() | gg_miss_var(x) | Plots the percentage of missing data by variable in x. | | naniar | miss_var_summary() | miss_var_summary(x) | Reports the percentage of missing data by variable in x. | | naniar | miss_var_which() | miss_var_which(x) | Returns variable names that contain | | tidyverse (tidyr) | drop_na(df) | drop_na(df)
drop_na(variable) | missing values in x.
Drops rows with NA values. When a variable is specified, only drops rows where that variable is NA. | | tidyverse (dplyr) | na_if() | na_if(vector, value) | Converts specified values to NA. Use with mutate() . | | tidyverse (tidyr) | replace_na() | replace_na(vector, value) | Replaces NA values with a specified value. Use with mutate() . | | tidyverse (dplyr) | case_when() | case_when(test ~ outcome, .default = x) | Vectorizes multiple if_else() statements. If no cases match, NA is returned unless | | tidyverse (dplyr) | mutate() | df < -mutate(df, newcol = wt / 2 .2) | .default is specified. Use with mutate() . Adds a new column based on existing columns. | | tidyverse (tidyr) | separate() | df |> separate(x, c("A", "B")) | Separates one character column into multiple columns using a regular expression or numeric locations. | | tidyverse (tidyr) | unite() | df | > unite("z", x:y, remove = FALSE) | Unites multiple columns into one column. | | tidyverse (stringr) | str_detect() | df |> filter(str_detect(col_name,"string_pattern")) | Returns a logical vector indicating whether a string pattern was detected. | | tidyverse (stringr) | str_replace() | str_replace(string = vector,"replace_me", "with_me") | Replaces the first instance of one specified string with another. | | tidyverse (stringr) | str_replace_all() | str_replace_all(string = vector,"replace_me", "with_me") | Replaces all instances of one specified string with another. | | tidyverse (stringr) | str_sub() | str_sub(string = vector, start = 1, end = 3) | Subsets a string to the specified character positions. | * This cheatsheet format was adapted from Alex’s Lemonade Stand materials (source) . * Find more resources at [https://daseh.org](https://daseh.org).

FAQs

What are NA, NaN, and Inf/-Inf in data cleaning?

NA represents general missing data. NaN is “Not a Number,” commonly produced by operations like 0/0. Inf and -Inf indicate Infinity, typically occurring when a positive or negative number is divided by 0.

Which tools help explore missingness in large datasets?

The cheatsheet highlights the naniar package, including functions like gg_miss_var() and miss_var_summary() to visualize and report missingness by variable.

How can tidyverse functions be used to handle missing and transform data?

It covers drop_na() to drop rows with NA values, na_if() to convert specific values to NA, and replace_na() to substitute NA with a chosen value. It also includes transformations such as mutate(), separate(), and unite(), plus string operations like str_detect(), str_replace(), str_replace_all(), and str_sub().

Oct 14, 2026

Upload by I feel Different