Keep accented characters with ODBC

[Kevin Vasquez]

Hello dear,

I'm new using STATA and I need help with the following issue. I am using Stata / SE 15.1 and I obtain the information from Oracle using an ODBC connection. When I imported the data that has accented characters and the letter ñ, they are not recognized by STATA; then I change the columns to alphanumeric type, the problem is solved. However, I need to define labels for the dataset and this is only possible with number type columns. The same happens when I export the data with the value of the labels in TXT format delimited .

Please help me to solve the problem.

Sincerelly.

Thank you

[Joseph Coveney]

I would hazard a guess that there is something in the ODBC driver settings that you can tweak to fix this. I don't know about Oracle, but for Microsoft's SQL Server's ODBC driver, there are settings for "Perform translations for character data" and "Use regional settings when outputting currency, numbers, dates and times". The former I think takes care of converting the various non-ASCII character encoding and the latter will affect using commas versus dots/periods in decimal numeric values.

In WIndows 10 (mentioned in the companion post), you'd open the old Control Panel menu, and type odbc in the search box, which will bring up Set up ODBC data sources for both 32- and 64-bit drivers. You'll need administrator privileges to alter the driver settings.