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

#delimit;

* Slovakia LIS 1996 : Example CAMSIS matching ;
* (here to household file);


use $sk96h;

* Current occ unit (isco88 two digit)=d14; 
* valid d14 in range > 999 and < 10000. 
* Employment status differences not applicable to Slovakia;

gen isco88=-999;
replace isco88=d14 if (d14 > 999 & d14 < 10000);


* Key indexing variables is isco88;

tabulate d3 if (isco88 > 0) , 
        summarize(isco88) mean standard freq ; 

sort isco88;
save cstemp, replace;
clear;
infile isco88 mcam fcam siops isei 
      using $mydata\camsis\csskis96.dat;
summarize;
sort isco88;
merge isco88 using cstemp;
gen occinfo=_merge;
tab occinfo; 
drop if (occinfo==1);
* (this removes any CAMSIS isco88 values not represented in LIS sample);
tab occinfo d3; 


* Current values are within gender; 
* also can make a cross-gender CAMSIS variable : ;

gen cgentcs=-999;
replace cgentcs=mcam if (d3==1);
replace cgentcs=fcam if (d3==2);

summarize isco88 cgentcs  ;
summarize isco88 cgentcs  if (d14 > 999 & d14 < 10000) ;


***************************************;


* Quick assessment: compare values by an education recode;
gen educ=d10; 
recode educ 1=1 2 3 4=2 5=3 *=-9;
label define edu  1 "Primary or less" 2 "Intermediate" 
    3 "University" ; 
label values educ edu;
label variable educ "education categories";
tab educ  ;
sort d3;
by d3:
  summarize educ 
      cgentcs  if (d14 > 999 & d14 < 10000); 
by d3: 
 oneway cgentcs educ  if (d14 > 999 & d14 < 10000 & educ > 0), tabulate;



*****************************************;