To underline what is partly implicit here in Svend's
reply:
You must feed -reshape- one or more stubnames,
not a varlist.
Nick
[email protected]
Svend Juul
> Clive Nicholas wrote:
>
> Following the example given in -help reshape- when converting
> one's data from
> wide to long format, thus
>
> (wide form)
> i .............. x_ij ...............
> id sex inc80 inc81 inc82
> ------------------------------------------
> 1 0 5000 5500 6000
> 2 1 2000 2200 3300
> 3 0 3000 2000 1000
>
> (long form)
> i j x_ij
> id year sex inc
> -------------------------------
> 1 80 0 5000
> 1 81 0 5500
> 1 82 0 6000
> 2 80 1 2000
> 2 81 1 2200
> 2 82 1 3300
> 3 80 0 3000
> 3 81 0 2000
> 3 82 0 1000
>
> The help file clearly states that the above is achieved by
> . reshape long inc, i(id) j(year)
> But I can't reproduce this simple command with my own data,
> and I _don't_ understand why!
>
> . des
> Contains data from bepsnew2.dta
> obs: 3,615
> vars: 13 18 Feb 2007 02:54
> size: 65,070 (99.8% of memory free)
> --------------------------------------------------------------
> -----------------
> storage display value
> variable name type format label variable label
> --------------------------------------------------------------
> -----------------
> id int %9.0g
> rsex01 byte %8.0g rsex01 Female voter
> conle98 byte %8.0g votelc98==conservative
> lable98 byte %8.0g votelc98==labour
> ldmle98 byte %8.0g
> votelc98==liberal democrat
> othle98 byte %9.0g
> conle99 byte %8.0g votelc99==conservative
> lable99 byte %8.0g votelc99==labour
> ldmle99 byte %8.0g
> votelc99==liberal democrat
> othle99 byte %9.0g
> conle00 byte %8.0g votelc00==conservative
> lable00 byte %8.0g votelc00==labour
> ldmle00 byte %8.0g
> votelc00==liberal democrat
> othle00 byte %9.0g
> --------------------------------------------------------------
> -----------------
>
> All of the data here (recalculated from the 1997-2001 British
> Election Panel
> Study) - apart from the sex variable - are dummy variables
> arrayed in wide
> format over three waves, with quite a few missings on some of them.
> The year is clearly indicated at the end of each variable,
> apart from the
> -id- variable.
> This is a snapshot of the data I need to -reshape-: there are
> also variables
> indicating survey answers to questions asked on only two
> waves. Every time I run
> -reshape- to convert to long, I get this:
>
> . reshape long conle98-othle00, i(id) j(year)
> no xij variables found
> r(111);
> ---------------------------------------------------------------------
> If possible, avoid statements about inexplicability.
>
> You should:
> reshape long conle lable ldmle othle , i(id) j(year)
>
> Apart from that: I suspect that you will get a problem with
> the years 98,99,00;
> You might want to rename:
> rename conle00 conle100
> to get increasing years.
*
* 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/
