Unicode problem when switching from Stata 13 to Stata 14

[Frauke Hildebrand]

Hello,

I have developed my files (.dta) in Stata 13.

Some observations within variables contain letters as ä,ü,ö (german Umlaute)
I want to merge data with the help of a variable that contains city names and thus also ü,ä or ö.

After switching to stata 14 I am unable to merge such data.

I have tried some commands, but none of them really helped.

I would be very happy if you can help me with the problem.

Thank you

[Mike Lacy]

Please understand that saying "some commands" and "none of them really helped" does not help us help you, since we can't know what "some" you tried and what precisely happened. Per the FAQ, you'll be more likely to get help if you can do this.

My recollection is that v.13 and v.14 have different data file formats. So, in your more detailed description of the problem, knowing what the file formats are would be necessary. One place to start--perhaps you have done this--would be to open and save both your master and using files as v. 14 files, and then try the merge.

[Frauke Hildebrand]

Hello,

Thank you very much.

It finally works.

I used:

unicode encoding set latin1
unicode translate *dta

[Celia Skaarup]

I have the same problem with the danish letters æ, ø, å. My .dta files are a couple of years old, and contains the letters in varnames and labels, and I am unable to access/edit or use them. Which means many of my variables are useless.

Example: Bop�l_urbaniseringsgrad

I have tried

clear
cd "F:\Folder"
unicode encoding set latin1 //and Windows-1252
unicode analyze *
unicode translate *


And get an output saying multiple varnames and labels are changed, but when i reopen the file, nothing has happened - the varnames and labels are the same as before. Showing a �.

Hope someone can help!
/Celia