** 1) Creation of dummy indicators by a priori expectations. 
* all exact title-by-status / title / major group / submajor group / etc combinations :. 
compute dibst=(hbst=wbst). 
compute diocc=(hocc=wocc). 
compute digp1=(h1gp=w1gp). 
compute digp2=(h2gp=w2gp). 
*etc. 

* all self-employed husbands (empst=1) with wives employee company secretaries (wbst=2225). 
compute compsec=(wbst=2225 & hempst=1). 
* all couples where one is self employed (empst=1) and other is family assistant (empst=2). 
compute fambusi=((hempst=1 & wempst=2) | (hempst=2 & wempst=1)). 
* all teachers (any occ from 441 to 459) :. 
compute teach=(hocc ge 441 & hocc le 459 & wocc ge 441 & wocc le 459). 
* farming : husband farmers occ 110, wife farmer occ 110 or agricultural worker occ 160) :. 
compute farm=(hocc=110 & (wocc=110 | wocc=160)). 
* Catering : husband publican occ 550 and wife barmaid occ 930) :. 
compute pub=(hocc=550 & wocc=930). 

** 2) Specific combinations identified through examining models. 
compute specpsd=0. 
if (hocc=345 & wocc=456) specpsd=1. 
if (hocc=123 & wocc=789) specpsd=1. 
. 
etc . 

** Save out to a data file :. 
sav out="rev1pseuds1.sav". 

** Apply chosen specification to a given model :. 
compute m1prob=(diocc=1 | farm=1 | pub=1 | fambusi=1 | specpsd=1). 
temp. 
select if (m1prob=0). 
correspondence... 

 
 
Return to Handling diagonals in SPSS CA