SPSS in Chinese
Locked
 
|
Dear all, I've created a poll in Chinese. The data collection (via surveymonkey) worked without issues. While loading the Excel data set in SPSS, the Chinese characters are not transferred correctly. It appears only ?? - both in the labels and the free text. In the EXCEL data sets all is displayed properly in Chinese. Can you give me a hint? I tried … SET LOCALE='SChinese'. SET DEFOLANG = SChinese. SET UNICODE YES. … but no success. With GET DATA there is no appropriate option – only for GET SAS or GET STATA. Regards Heino PS I'm using Windows 19 and SPSS. PPS Sorry to say the IBM (SPSS) helpdesk is rather lousy or subterraneous = unterirdisch – as we say in German. --- |
|
|
I can't see anything obviously missing
-- unless the poll is in traditional Chinese. You ran these command before
trying to read the Excel file into SPSS Statistics?
Rick Oliver Senior Information Developer IBM Business Analytics (SPSS) E-mail: [hidden email] From: "Haase, Heino, Springer DE" <[hidden email]> To: [hidden email] Date: 03/16/2012 08:32 AM Subject: SPSS in Chinese Sent by: "SPSSX(r) Discussion" <[hidden email]> Dear all, I've created a poll in Chinese. The data collection (via surveymonkey) worked without issues. While loading the Excel data set in SPSS, the Chinese characters are not transferred correctly. It appears only ?? - both in the labels and the free text. In the EXCEL data sets all is displayed properly in Chinese. Can you give me a hint? I tried … SET LOCALE='SChinese'. SET DEFOLANG = SChinese. SET UNICODE YES. … but no success. With GET DATA there is no appropriate option – only for GET SAS or GET STATA. Regards Heino PS I'm using Windows 19 and SPSS. PPS Sorry to say the IBM (SPSS) helpdesk is rather lousy or subterraneous = unterirdisch – as we say in German. --- Heino Haase Springer Market Intelligence & Web Analytics --- Tiergartenstr. 17 | 69121 Heidelberg | Germany tel +49 (0)6221 / 487 - 8273 fax +49 (0)6221 / 487 - 68273 [hidden email] --- http://www.springer.com |
|
|
In reply to this post by Haase, Heino, Springer DE
Excel, at least all 21st century versions,
keep data in Unicode. If Statistics is in Unicode mode, the characters
should automatically work. In code page mode, they would be converted
according to the Statistics locale, and some could become ? if the locale
does not match.
Did SM create the ss directly, or did you read that data into Excel? If the former, it's possible that the Excel file is not as expected, and resaving it from Excel might help. Presumably the text looks right in Excel. Another possibility, only relevant for code page mode, is that the character set is Traditional Chinese (Taiwan), not Simplified. In that case, you would need to be in Tchinese code page. But the locale settings need to match the proper locale names if you are in code page mode. Schinese is not valid. Use SHOW LOCALE UNICODE to confirm that the settings are what you expect. The olang setting is irrelevant for character conversion. HTH, Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] new phone: 720-342-5621 From: "Haase, Heino, Springer DE" <[hidden email]> To: [hidden email] Date: 03/16/2012 07:35 AM Subject: [SPSSX-L] SPSS in Chinese Sent by: "SPSSX(r) Discussion" <[hidden email]> Dear all, I've created a poll in Chinese. The data collection (via surveymonkey) worked without issues. While loading the Excel data set in SPSS, the Chinese characters are not transferred correctly. It appears only ?? - both in the labels and the free text. In the EXCEL data sets all is displayed properly in Chinese. Can you give me a hint? I tried … SET LOCALE='SChinese'. SET DEFOLANG = SChinese. SET UNICODE YES. … but no success. With GET DATA there is no appropriate option – only for GET SAS or GET STATA. Regards Heino PS I'm using Windows 19 and SPSS. PPS Sorry to say the IBM (SPSS) helpdesk is rather lousy or subterraneous = unterirdisch – as we say in German. --- Heino Haase Springer Market Intelligence & Web Analytics --- Tiergartenstr. 17 | 69121 Heidelberg | Germany tel +49 (0)6221 / 487 - 8273 fax +49 (0)6221 / 487 - 68273 [hidden email] --- http://www.springer.com |
|
|
Jon Peck is correct. Although it theoretically
shouldn't matter in Unicode mode, try this:
set locale zh_cn. Rick Oliver Senior Information Developer IBM Business Analytics (SPSS) E-mail: [hidden email] Phone: 312.893.4922 | T/L: 206-4922 From: Jon K Peck/Chicago/IBM@IBMUS To: [hidden email] Date: 03/16/2012 08:52 AM Subject: Re: SPSS in Chinese Sent by: "SPSSX(r) Discussion" <[hidden email]> Excel, at least all 21st century versions, keep data in Unicode. If Statistics is in Unicode mode, the characters should automatically work. In code page mode, they would be converted according to the Statistics locale, and some could become ? if the locale does not match. Did SM create the ss directly, or did you read that data into Excel? If the former, it's possible that the Excel file is not as expected, and resaving it from Excel might help. Presumably the text looks right in Excel. Another possibility, only relevant for code page mode, is that the character set is Traditional Chinese (Taiwan), not Simplified. In that case, you would need to be in Tchinese code page. But the locale settings need to match the proper locale names if you are in code page mode. Schinese is not valid. Use SHOW LOCALE UNICODE to confirm that the settings are what you expect. The olang setting is irrelevant for character conversion. HTH, Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] new phone: 720-342-5621 From: "Haase, Heino, Springer DE" <[hidden email]> To: [hidden email] Date: 03/16/2012 07:35 AM Subject: [SPSSX-L] SPSS in Chinese Sent by: "SPSSX(r) Discussion" <[hidden email]> Dear all, I've created a poll in Chinese. The data collection (via surveymonkey) worked without issues. While loading the Excel data set in SPSS, the Chinese characters are not transferred correctly. It appears only ?? - both in the labels and the free text. In the EXCEL data sets all is displayed properly in Chinese. Can you give me a hint? I tried … SET LOCALE='SChinese'. SET DEFOLANG = SChinese. SET UNICODE YES. … but no success. With GET DATA there is no appropriate option – only for GET SAS or GET STATA. Regards Heino PS I'm using Windows 19 and SPSS. PPS Sorry to say the IBM (SPSS) helpdesk is rather lousy or subterraneous = unterirdisch – as we say in German. --- Heino Haase Springer Market Intelligence & Web Analytics --- Tiergartenstr. 17 | 69121 Heidelberg | Germany tel +49 (0)6221 / 487 - 8273 fax +49 (0)6221 / 487 - 68273 [hidden email] --- http://www.springer.com |
|
|
In reply to this post by Haase, Heino, Springer DE
Hi,
As far as I understand you have to use the command set locale="Chinese-PRC". the command set locale = SChinese. does not work in SPSS 19 or 20 (at least not on my copies). The working locales are defined in the loclmap.xml file in the Program Files\IBM\SPSS\Statistics\20(or19)\ folder. In our experience (I am a visiting professor at a Chinese research institute in Beijing at the moment) the whole thing also seems related to whether one is using XP or Windows 7. Windows 7 is generally more well behaved than XP and SPSS 20 seems to be more well behaved than its predecessors. hth, Jon |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |