What is the difference between the add and replace methods?

The difference between the add and replace methods when working with fragments is as follows:

• add: Adds a new fragment on top of existing ones. Previous fragments remain in the back stack (if addToBackStack was called), and their lifecycle is paused (they transition to the PAUSED state).
• replace: Removes the previous fragment from the container and adds a new one in its place. If the previous fragment was in the back stack, it will be recreated upon return.

Example of using add:

supportFragmentManager.beginTransaction()
    .add(R.id.fragment_container, NewFragment())
    .addToBackStack(null)
    .commit()

Example of using replace:

supportFragmentManager.beginTransaction()
    .replace(R.id.fragment_container, NewFragment())
    .addToBackStack(null)
    .commit()

The main differences can be summarized in a table:

Characteristic	add	replace
Visibility	The new fragment overlays the previous	The new fragment replaces the previous
Previous state	Paused (PAUSED)	Removed from the container, recreated upon return
Back stack	The previous fragment remains in the back stack	The previous fragment is removed from the container but can be in the back stack
Performance	May be less performant with many fragments	Usually more performant with frequent fragment changes