Renaming person identity numbers in survey data

[Gregory Whittaker]

Hello, I have survey data with lengthy person identity numbers, for example:
1000600101
1000600101
1000600107
1000600107
1000600107
1000600110
1000600110
etc.
I would like to create a new column where the identity numbers (which have been sorted already) are reduced to the following:
1
1
2
2
2
3
3
etc.
I would appreciate any advise on how to renumber the identity column taking into account duplicates (all identity numbers have at least one duplicate, some up to 8).
Thank you.

[Carlo Lazzaro]

Gregory:
welcome to this forum,
You may want to try:

egen new_id=group(id)

[Nick Cox]

To expand on @Carlo Lazzaro's answer.

This is also an FAQ. See https://www.stata.com/support/faqs/d...p-identifiers/

The solution below should work even if your identifier is string.

clear 
input double bad_id 
1000600101
1000600101
1000600107
1000600107
1000600107
1000600110
1000600110
end 

format bad_id %10.0f 

egen better_id = group(bad_id)

list , sepby(bad_id)
     +-----------------------+
     |     bad_id   better~d |
     |-----------------------|
  1. | 1000600101          1 |
  2. | 1000600101          1 |
     |-----------------------|
  3. | 1000600107          2 |
  4. | 1000600107          2 |
  5. | 1000600107          2 |
     |-----------------------|
  6. | 1000600110          3 |
  7. | 1000600110          3 |
     +-----------------------+

[Gregory Whittaker]

Many thanks Carlo and Nick