I didn't read through all your code, but perhaps, using -merge- can
accomplish your goal. See the exmaple below. Also it is not clear
how, starting with your initial data set, you get the 1991 Divorced
and 1992 Remarried for momid 2 in your final data set.
Scott
. l , noobs sepby(mom)
+---------------------------------------------------+
| momid year type1 y1 type2 y2 |
|---------------------------------------------------|
| 1 2000 Married 2000 . |
| 1 2001 . . |
| 1 2002 Separated 2001 Divorced 2001 |
| 1 2003 . . |
| 1 2004 . . |
|---------------------------------------------------|
| 2 1988 Married 1987 . |
| 2 1989 . . |
| 2 1990 . . |
| 2 1991 . . |
| 2 1992 . . |
| 2 1993 . . |
| 2 1994 . . |
| 2 1995 . . |
| 2 1996 Divorced 1993 . |
| 2 1997 . . |
| 2 1998 . . |
| 2 1999 . . |
| 2 2000 Remarried 1998 . |
+---------------------------------------------------+
. drop year
. rename y1 year
. sort mom year
. merge mom year using "C:\Documents and
Settings\scott.merryman\Desktop\foo.dta"
variables momid year do not uniquely identify observations in the
master data
. drop if year ==.
(13 observations deleted)
. drop _m
. sort mom year
. order mom year type1 y1 type2 y2
. l, noob sepby(mom)
+---------------------------------------------------+
| momid year type1 y1 type2 y2 |
|---------------------------------------------------|
| 1 2000 Married 2000 . |
| 1 2001 Separated . Divorced 2001 |
| 1 2002 Separated 2001 Divorced 2001 |
| 1 2003 . . |
| 1 2004 . . |
|---------------------------------------------------|
| 2 1987 Married . . |
| 2 1988 Married 1987 . |
| 2 1989 . . |
| 2 1990 . . |
| 2 1991 . . |
| 2 1992 . . |
| 2 1993 Divorced . . |
| 2 1994 . . |
| 2 1995 . . |
| 2 1996 Divorced 1993 . |
| 2 1997 . . |
| 2 1998 Remarried . . |
| 2 1999 . . |
| 2 2000 Remarried 1998 . |
+---------------------------------------------------+
----- Original Message -----
From: "Claire M. Kamp Dush" <[email protected]>
Date: Wednesday, August 23, 2006 12:52 pm
Subject: st: programming: stopping loops?
To: [email protected]
> Hello, I feel embarrassed to post this because I am sure the
> answer to this
> is obvious, but I have been puzzling over this issue for a few
> hours. I am
> trying to recode the family structure data in the NLSY 79 through
> 2004. I
> am trying to go back and recode the data for missing years based
> on reports
> of marital changes between interviews at follow-ups. For
> instance, if an
> individual was interviewed in 1991 and not in 1992, in 1993 they
> are asked
> to report up to 3 marital changes since the last time they were
> interviewed. My data is stacked, with each individual having 26
> lines of
> data, for years 1979 through 2004. The id variable is momid and
> the year
> variable is year. change1type, change2type, and change3type are
> measured
> each year where the respondent has data, and is a categorical
> variable with
> categories including married, divorced, separated, widowed, etc.
> changey1_
> , changey2_, and changey3_ are the years in which the each change
> is said
> to occur. Here is an example of what the data look like:
>
> momid year change1type changey1_ change2type
> changey2_1 2000 Married 2000
> 1 2001
> 1 2002 Separated 2001 Divorced
> 2001
> 1 2003
> 1 2004
> 2 1988 Married 1987
> 2 1989
> 2 1990
> 2 1991
> 2 1992
> 2 1993
> 2 1994
> 2 1995
> 2 1996 Divorced 1993
> 2 1997
> 2 1998
> 2 1999
> 2 2000 Remarried 1998
>
> My goal is to have my data look like the following:
>
> momid year change1type changey1_ change2type
> changey2_
> change1misstype change2misstype
> 1 2000 Married 2000
> Married
> 1 2001
> Separated Divorced
> 1 2002 Separated 2001 Divorced
> 2001
> 1 2003
> 1 2004
> 2 1987
> Married
> 2 1988 Married 1987
> 2 1989
> 2 1990
> 2 1991 Divorced 1991
> Divorced Remarried
> 2 1992 Remarried 1991
> 2 1993
> Divorced
> 2 1994
> 2 1995
> 2 1996 Divorced 1993
> 2 1997
> 2 1998
> Remarried
> 2 1999
> 2 2000 Remarried 1998
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
