how to add more rows for a unique id

[Chhavi Paul]

childid session id
1 UP/GBN/RU/3I/KHP/KHZ/CLC/1
2 UP/GBN/RU/3I/KHP/KHZ/CLC/1
3 UP/GBN/RU/3I/KHP/KHZ/CLC/1
4 UP/GBN/RU/3I/KHP/KHZ/CLC/1
5 UP/GBN/RU/3I/KHP/KHZ/CLC/1
1 UP/GBN/RU/3I/KSN/KNB/CLC/3
2 UP/GBN/RU/3I/KSN/KNB/CLC/3
3 UP/GBN/RU/3I/KSN/KNB/CLC/3

In case total number of rows should be same for both the sessions ids viz, UP/GBN/RU/3I/KHP/KHZ/CLC/1 and UP/GBN/RU/3I/KSN/KNB/CLC/3, lets say it should be 10. For first session id, 5 students are present and 10-5=5 are absent and I would to include five more rows with the same session id and blank childids.

Similarly, for second session id (UP/GBN/RU/3I/KSN/KNB/CLC/3), 3 are present and 10-3 =7 are absent and I want to include 7 more rows with the second session id and blank child ids.

What is the stata command for that, Please help?

[Carlo Lazzaro]

Cchavi:
see -expand-.

[William Lisowski]

* Example generated by -dataex-. For more info, type help dataex
clear
input byte childid str26 sessionid
1 "UP/GBN/RU/3I/KHP/KHZ/CLC/1"
2 "UP/GBN/RU/3I/KHP/KHZ/CLC/1"
3 "UP/GBN/RU/3I/KHP/KHZ/CLC/1"
4 "UP/GBN/RU/3I/KHP/KHZ/CLC/1"
5 "UP/GBN/RU/3I/KHP/KHZ/CLC/1"
1 "UP/GBN/RU/3I/KSN/KNB/CLC/3"
2 "UP/GBN/RU/3I/KSN/KNB/CLC/3"
3 "UP/GBN/RU/3I/KSN/KNB/CLC/3"
end
sort sessionid childid
generate copies = 1
by sessionid: replace copies = 11-_N if _n==_N
list, clean
expand copies, generate(new)
replace childid = . if new==1
sort sessionid childid
drop copies new
list, clean

. sort sessionid childid

. generate copies = 1

. by sessionid: replace copies = 11-_N if _n==_N
(2 real changes made)

. list, clean

       childid                    sessionid   copies  
  1.         1   UP/GBN/RU/3I/KHP/KHZ/CLC/1        1  
  2.         2   UP/GBN/RU/3I/KHP/KHZ/CLC/1        1  
  3.         3   UP/GBN/RU/3I/KHP/KHZ/CLC/1        1  
  4.         4   UP/GBN/RU/3I/KHP/KHZ/CLC/1        1  
  5.         5   UP/GBN/RU/3I/KHP/KHZ/CLC/1        6  
  6.         1   UP/GBN/RU/3I/KSN/KNB/CLC/3        1  
  7.         2   UP/GBN/RU/3I/KSN/KNB/CLC/3        1  
  8.         3   UP/GBN/RU/3I/KSN/KNB/CLC/3        8  

. expand copies, generate(new)
(12 observations created)

. replace childid = . if new==1
(12 real changes made, 12 to missing)

. sort sessionid childid

. drop copies new

. list, clean

       childid                    sessionid  
  1.         1   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
  2.         2   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
  3.         3   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
  4.         4   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
  5.         5   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
  6.         .   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
  7.         .   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
  8.         .   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
  9.         .   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
 10.         .   UP/GBN/RU/3I/KHP/KHZ/CLC/1  
 11.         1   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 12.         2   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 13.         3   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 14.         .   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 15.         .   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 16.         .   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 17.         .   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 18.         .   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 19.         .   UP/GBN/RU/3I/KSN/KNB/CLC/3  
 20.         .   UP/GBN/RU/3I/KSN/KNB/CLC/3  

.

Note that because you did follow the instructions in the FAQ (to which you were referred in your earlier post on this problem) for presenting your example data using the dataex command, this example code will very likely need modification to work with your actual data.

[Chhavi Paul]

I tried the above syntax, it is working but it is creating null values for my other important variables.

[Chhavi Paul]

Thanks @William Lisowski . I have been able to solve this situation.