**announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

In your example, one variable is problematic. In that case something like forval i = 1/50 { use state`i' destring var1, replace save new_state`i' } should fix the problem. In particular, 1. Whenever -var1- is numeric, -destring- won't complain. 2. If -destring- finds non-numeric characters in a string variable it will stop the loop. In the last case, you will need to reverse the process: forval i = 1/50 { use state`i' tostring var1, replace save new_state`i' } 3. If more variables are problematic, you need to add them to the -destring- or -tostring- statements. (-tostring- is a command, not a function, in Stata.) 4. I'm recommending new filenames for changed datasets as a matter of general caution. If you have to iterate, you can always overwrite new_state1, etc., so long as your originals are safe somewhere. Nick njcoxstata@gmail.com On 8 May 2013 01:59, Erika Kociolek <ekociole@gmail.com> wrote: > Hi there, > > I am working with a large number of datasets that have some variables > in common. I'll use the two datasets below as an example to illustrate > my question: > > State1.dta > > var1 var2 > . 1 > 1 . > 4 . > 13 2 > > > State2.dta > > var1 var3 > . > 1' 2 > 3 3 > 4 4 > > As you can see above, the files State1 and State2 have var1 in common. > I would like to append the State1 and State2 files together to create > a master file containing data for both states (some fields will be > blank because those datapoints may not have been collected for a given > state). > > Due to some issues with the source datasets, in some of the State > files, var1 is a string (as is the case with State2) and in others, > var1 is numeric (as is the case with State1). Since the variable types > are different, I cannot append one file to the other. > > My question: is there is any way to define that certain variables be > imported as a particular type of variable (such as a string variable), > instead of having to import files and then generate new string > variables using a function like "tostring"? I'd like to be able to get > the variables in all .dta files defined consistently so it is easy to > append them to a master datafile. > > Thanks! > Erika > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/faqs/resources/statalist-faq/ > * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/

- Prev by Date:
**Re: st: Importing Data and Defining Variable Types** - Next by Date:
**Re: st: Problems imputting data** - Previous by thread:
**st: Stata journal paper on interaction terms in count models** - Next by thread:
**st: New version of -haif- on SSC** - Index(es):