Conventions and tips

General remarks

• Minimum aim: computational reproducibility of your code
• Better: Other people understand your data & code and can use it

Please note:

• There are no universal guidelines for coding in Stata
• What follows is a compilation of different guidelines & own experience
• Most of the time, there is no “best solution”
• No guarantee for completeness

These principles will also help yourself, because:
No one is perfect! Coding errors happen.

But with a good strategy, you can

• Reduce errors while coding
• Enable others to cross-check (“four eyes see more than two”)
• Fix errors more easily

Some key principles

• Code and data should be consistent
  → Agree on standards such as naming conventions within your team
• Code and data should be as self-explanatory as possible & sufficiently commented/documented
  → Choose meaningful names for objects, comment workflow/decisions from the beginning
• Code should be as simple & short as possible
  → Only keep what is needed, structure your code