** user=**** 
** password=**** 
** package=stata
** project=lis

#delimit;

* US LIS 1997 : STATA Example CAMSIS matching to US 97 data;

use $us97p;
* Current occ unit (isco88's)=pocc; 
* (incomplete) Employment status info from ptypewk only;



* Derive employment status variable;
gen status=ptypewk;
recode status 1 8=5 2 3 4=4 5=1 6=3 7=2 *=9;
label define stat 
   1 "Self-Employed, not incorporated"
   2 "Family Worker"
   3 "Self-Employed, incorporated"
   4 "Government Employee"
   5 "Other Employee" 
   9 "Unknown";
label values status stat;
label variable status "Employment status";
tab status psex;




* Key indexing variable is occbst;
gen occbst=-999;
replace occbst=pocc*10 + status if (pocc > 0 );
tabulate status if (pocc > 0 ), 
       summarize(occbst) mean standard freq ;



* Merge files with CAMSIS scores using indexing variable;
sort occbst;
save cstemp, replace;
use occbst hscs wscs htcs wtcs siops isei 
      using $mydata\camsis\csusis97;
summarize;
sort occbst;
merge occbst using cstemp;
gen occinfo=_merge;
tab occinfo; 
drop if (occinfo==1);
* (this removes those occbst values which are not represented in the LES sample);
tab occinfo psex; 



* Current values are within gender; 
* also can make a cross-gender CAMSIS variable : ;
gen cgentcs=-999;
replace cgentcs=htcs if (psex==1);
replace cgentcs=wtcs if (psex==2);
gen cgenscs=-999;
replace cgenscs=hscs if (psex==1);
replace cgenscs=wscs if (psex==2);



* Review variable distributions;
summarize occbst cgentcs cgenscs siops isei ;
summarize occbst cgentcs cgenscs siops isei if pocc > 0;
summarize occbst cgentcs cgenscs siops isei if (pocc > 0 & siops > 0); 
* note that the last summary shows us that a handful of cases have occs 
*  and camsis scores but have missing siops scores by the ISMF translation;




* Quick assessment: compare values by an education recode;
di "Education";
* some recode of education ;
gen educ=peduc; 
recode educ 1=1 2 3 4=2 5 6 7 8=3;
label define edu  1 "Elementary" 2 "Intermediate" 3 "Degree or above"; 
label values educ edu;
label variable educ "education categories";
tab educ ;
sort psex;
by psex:
  summarize educ 
      cgentcs cgenscs siops isei if pocc > 0; 
by psex: 
 oneway cgentcs educ  if (pocc > 0), tabulate;
by psex: 
 oneway cgenscs educ  if (pocc > 0), tabulate;
by psex: 
 oneway siops educ  if (pocc > 0 & siops > 0), tabulate;
by psex: 
 oneway isei educ  if (pocc > 0 & isei > 0), tabulate;




/* For interest here are the Eta-squared statistics with education from above:

Men: 
CAMSIS title only : 0.
CAMSIS title by status : 0.
SIOPS : 0.
ISEI: 0.
Women: 
CAMSIS title only : 0.
CAMSIS title by status : 0.
SIOPS : 0.
ISEI: 0.
 
*/