Tidy data is long and narrow.

Each row is an observation or case.

Each column is a variable.

Examples:

1. gapmider
2. babynames

Being tidy does not mean neat.

All columns need to contain data from the variable in that column only.

All rows contain data from that specific observation in that row only.

Each column in a tidy dataset contains a variable.

Each variable is either categorical or numeric.

Categorical variables are often stored as factors in R.

Codebooks were common when variable names were short due to computer memory restrictions.

Now days variable names can be as long as you need. So codebooks are less important.

> ??gapmider

> ?babynames

mutate()

spread()

gather()

These functions have been updated to the new functions.

See the tidyr website.

pivot_longer()

pivot_wider()

1. finance.yahoo.com What is the current value of Ford? Google? Apple? Is the data in a tidy format? How to download the data? Is the downloaded .csv file tidy?
2. SF Open Data Pick a topic of interest. I looked at Transportation and searched for Parking data. Is the data in a tidy format? Click on View Data. Is the data in a tidy format?

Basically all data that is available through and API is in a tidy format.

1. Tidyverse style guide
2. Advanced R style guilde
3. Google's R style guide

Take a look at the variable names in the R datasets.

1. Is the gapmider dataset tidy? What do you think of the variable names?
2. Is the babynames dataset tidy? What do you think of the variable names? Note that n is used as one of the variable names. It represents counts not sample size. My suggestion, don't use n as a variable name, it can be confusing.