Add the Person Identifier (NRPID) to the 1994-95 Migrant Follow-Up data

******************************************************************************
Attach NRPID to the 1994 Migrant Follow-Up Individual-Level Data File 
   1. Select ONLY 1994 Migrants in the PERSONID Data File
   2. Restructure MIGRANT94 into "Child" File,
      as 3 Migrants are in Migrant Follow-Up TWICE
   3. Match SMIGRANT94B to the MINDIV94 Data File

Input data: /nangrong/personid.X01
            /nangrong/1994/mindiv94.03
*****************************************************************************;

libname in1 xport '/nangrong/personid.X01';
libname in2 xport '/nangrong/1994/mindiv94.03';

***************************************************************************
*  Attach NRPID to the 1994 Migrant Follow-Up Individual-Level Data File  *
***************************************************************************;

* 1. Select ONLY 1994 Migrants in the PERSONID Data File *
-----------------------------------------------------------;
data migrant94;
  set in1.personid(keep=MID MCEP8 MIDB MCEP8B NRPID);

  if (MID ne .);

*** Rename Variables ***; 

rename
MID=MIDA
MCEP8=MCEP8A
;

run;

* 2. Restructure MIGRANT94 into "Child" File,
     as 3 Migrants are in Migrant Follow-Up TWICE *
----------------------------------------------------;
data migrant94b;
  set migrant94;
  length MCEP8 $ 3;

  keep MID MCEP8 NRPID;

  array m1 {2} MIDA MIDB;
  array m2 {2} MCEP8A MCEP8B;

  do i=1 to 2;
   MID=m1{i};
   MCEP8=m2{i};
   if MID ne . then output;
  end;

run;

*** Sort MIGRANT94B by MID MCEP8 ***;

proc sort data=migrant94b out=smigrant94b nodupkey;
  by MID MCEP8;
run;

* 3. Match SMIGRANT94B to the MINDIV94 Data File *
---------------------------------------------------;
data mindiv94_nrpid notin_mindiv94 notin_migrant94b;
  merge smigrant94b(in=a)
        in5.mindiv94(in=b);
  by MID MCEP8;

  if a=1 and b=1 then output mindiv94_nrpid;
  if a=1 and b=0 then output notin_mindiv94;
  if a=0 and b=1 then output notin_migrant94b;

run;

*** Check for Duplicates on NRPID in MINDIV94_NRPID (SHOULD HAVE 3!) ***;

proc sort data=mindiv94_nrpid out=smindiv94_nrpid nodupkey;
  by NRPID;
run;

206        ***************************************************************************
207        *  Attach NRPID to the 1994 Migrant Follow-Up Individual-Level Data File  *
208        ***************************************************************************;
209        
210        * 1. Select ONLY 1994 Migrants in the PERSONID Data File *
211        -----------------------------------------------------------;
212        data migrant94;
213          set in1.personid(keep=MID MCEP8 MIDB MCEP8B NRPID);
214        
215          if (MID ne .);
216        
217        *** Rename Variables ***;
218        
219        rename
220        MID=MIDA
221        MCEP8=MCEP8A
222        ;
223        
224        run;

NOTE: There were 57416 observations read from the data set IN1.PERSONID.
NOTE: The data set WORK.MIGRANT94 has 3791 observations and 5 variables.
NOTE: DATA statement used:
      real time           0.78 seconds
      cpu time            0.77 seconds   

225        
226        * 2. Restructure MIGRANT94 into "Child" File,
                as 3 Migrants are in Migrant Follow-Up TWICE *
227        ----------------------------------------------------;
228        data migrant94b;
229          set migrant94;
230          length MCEP8 $ 3;
231        
232          keep MID MCEP8 NRPID;
233        
234          array m1 {2} MIDA MIDB;
235          array m2 {2} MCEP8A MCEP8B;
236        
237          do i=1 to 2;
238           MID=m1{i};
239           MCEP8=m2{i};
240           if MID ne . then output;
241          end;
242        
243        run;

NOTE: There were 3791 observations read from the data set WORK.MIGRANT94.
NOTE: The data set WORK.MIGRANT94B has 3794 observations and 3 variables.
NOTE: DATA statement used:
      real time           0.10 seconds
      cpu time            0.06 seconds
      

244        
245        *** Sort MIGRANT94B by MID MCEP8 ***;
246        
247        proc sort data=migrant94b out=smigrant94b nodupkey;
248          by MID MCEP8;
249        run;

NOTE: 0 observations with duplicate key values were deleted.
NOTE: There were 3794 observations read from the data set WORK.MIGRANT94B.
NOTE: The data set WORK.SMIGRANT94B has 3794 observations and 3 variables.
NOTE: PROCEDURE SORT used:
      real time           0.10 seconds
      cpu time            0.05 seconds     
250        
251        * 3. Match SMIGRANT94B to the MINDIV94 Data File *
252        ---------------------------------------------------;
253        data mindiv94_nrpid notin_mindiv94 notin_migrant94b;
254          merge smigrant94b(in=a)
255                in5.mindiv94(in=b);
256          by MID MCEP8;
257        
258          if a=1 and b=1 then output mindiv94_nrpid;
259          if a=1 and b=0 then output notin_mindiv94;
260          if a=0 and b=1 then output notin_migrant94b;
261        
262        run;

NOTE: There were 3794 observations read from the data set WORK.SMIGRANT94B.
NOTE: There were 3794 observations read from the data set IN5.MINDIV94.
NOTE: The data set WORK.MINDIV94_NRPID has 3794 observations and 316 variables.
NOTE: The data set WORK.NOTIN_MINDIV94 has 0 observations and 316 variables.
NOTE: The data set WORK.NOTIN_MIGRANT94B has 0 observations and 316 variables.
NOTE: DATA statement used:
      real time           2.83 seconds
      cpu time            2.69 seconds
      

263        
264        *** Check for Duplicates on NRPID in MINDIV94_NRPID (SHOULD HAVE 3!) ***;
265        
266        proc sort data=mindiv94_nrpid out=smindiv94_nrpid nodupkey;
267          by NRPID;
268        run;

NOTE: 3 observations with duplicate key values were deleted.
NOTE: There were 3794 observations read from the data set WORK.MINDIV94_NRPID.
NOTE: The data set WORK.SMINDIV94_NRPID has 3791 observations and 316 variables.
NOTE: PROCEDURE SORT used:
      real time           2.03 seconds
      cpu time            1.24 seconds