"Taming" data Principles - summary

1. naming

• no more 20 char long
• lower case char, replace space w/ underscores (snake_case instead of camelCase)
• variable name: underscore instead of spaces

2. observational units

• observations in the left-hand columns

3. dates

• if only year — numerical
• if year and month — convert to Date (year-month-01) — associate all observations from the same month to have a day of 01
• if year, month, day — convert to Date (year-month-day)

4. other class variables

• ordinal — ordered
• categorical w/ a fixed and known set of levels — factor
• categorical w/ unknown/very large number of possible levels — characters
• “yes/no” character, binary variables — TRUE/FALSE logical

5. tidy format

• save all data frames in tidy format

1. each variable forms a column
2. each observation forms a row
3. each type of observational unit forms a table

• if converting alerts the dataset too much, make the code to convert easily accessible (i.e. Examples in the documentation)