Date: Fri, 30 Sep 2005 16:11:11 -0500
Reply-To: "Peck, Jon" <firstname.lastname@example.org>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: "Peck, Jon" <email@example.com>
Subject: Re: String data
Content-Type: text/plain; charset="us-ascii"
SPSS is fully capable of handling Chinese and Japanese characters, although the details depend a bit on which SPSS and Windows versions you are using. Newer is better.
The key issue is to set the locale that SPSS is running in to the locale that is appropriate for the data. Since SPSS 12.0.1 or 12.0.2, you can issue the command
to set the SPSS locale. However, by default it is set to the user locale of the os. Typically these match, so this command is not necessary, but if you are trying to process Japanese characters on a Windows system running in an English locale, you would need to make this change.
For Japanese, you would say
For Chinese, it is a little more complicated, since you have to pick traditional (Taiwan) or simplified (PRC). The locale names for those are
'Chinese_Taiwan.950' or just Chinese
set locale="Chinese_People's Republic of China.936". (Watch out for the single quote in the name.
It is a good idea to do a SHOW LOCALE to make sure you got what was intended.
If you do not have os support installed for these locales, they will fail, but this can be added on to your Windows installation, depending again on the Windows version via the Control Panel Regional Settings or Regional Options. For Windows 9.x systems, you mostly have to have the appropriate national language version of Windows. Microsoft Office carries some of its own language support, so it may work even when Windows is not configured for this.
Once you have done this, SPSS will process the characters correctly, but the default fonts will need to be changed in order to be able to see them (sometimes useful!). When you change the font, you also need to change the "script" setting (in the lower right of the font dialog) to match.
For Japanese, Windows provides MS Mincho and MS Gothic.
For traditional Chinese, MingLIU,
For Simplified Chinese, Simsun
When SPSS is installed, it configures itself automatically to use fonts appropriate for the current Windows locales.
If you wish for total harmony, you can change the output language of SPSS to match the data via the Edit/Options/General Language control. The user interface remains in its original language, but Viewer output will switch. (Certain combinations will not work, but an English user interface will work with any available output language).
In case you ever want to get back to your original locale, you should do a SHOW LOCALE command before making the changes so you know what to switch back to.
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of Arthur Aguirre
Sent: Friday, September 30, 2005 11:12 AM
Subject: [SPSSX-L] String data
At times I deal with text data that are in Chinese and Japanese
characters. These commonly are lost in SPSS but readable in Excel. Is
there some format that I can use so that characters are retained in the
SPSS database and these can be exported to/imported from Excel with no
loss of information?
Any info or places to read up on the subject would be useful.
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email