[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
Nick Winter <nw53@cornell.edu> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: RE: length of a variable name |

Date |
Wed, 16 Jul 2003 13:09:39 -0400 |

I have had the exact same issue interfacing between Stata and SPSS--the dataset I was using had the full question text as the variable labels, which was truncated on transfer. And I experienced similar embarrassment, given my (very loudly and frequently expressed) attitude about SPSS.

Another issue has to do with SPSS's user-defined missing values. Stata now supports multiple missing values, but there is no clean way to turn the various user-missings for a variable in SPSS into Stata's .a, .b, etc.

One approach, which I've started working on but not finished, is to read the SPSS data dictionary direclty using -file-, and then use -note- to attach the full labels to the variables, and re-map the missing values. But that's ugly, messy, and difficult.

So I endorse the request for longer variable labels, at least in SE.

Nick Winter

From: Lee Sieswerda [mailto:Lee.Sieswerda@tbdhu.com]

--------------------------------------------------------------------Sent: Tuesday, July 15, 2003 6:34 PM To: Statalist (statalist@hsphsun2.harvard.edu) Subject: st: length of a variable name According to -help limits-, the maximum length of a variable name is 80 characters (I use Intercooled Stata 8, but the limit is the same in Stata/SE). One would think that this would be enough. However, I have a dataset in SPSS format wherein the variable labels are the full text of a survey question. This often exceeds the 80 character limit. The longest variable label in the dataset is 210 characters long. I am working on this dataset with a number of other people, all of whom use SPSS, which has a much higher limit. I much prefer Stata, so I always convert the SPSS dataset to Stata format to work on it. However, the 80-character variable label limit means that when I convert back to SPSS format and send the altered dataset back to my colleagues, the variable labels are all truncated at 80 characters (embarassing, as I have been proselytizing my benighted SPSS-using colleagues for some time). I could work around this by re-labeling after I convert back to SPSS format, but with a couple of hundred variables, this is a pain in the neck. I wonder how much support there would be for increasing the variable label limit to, say, 244 characters? Best, Lee Sieswerda * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

Nicholas Winter

Assistant Professor

Department of Government

Cornell University

*

* For searches and help try:

* http://www.stata.com/support/faqs/res/findit.html

* http://www.stata.com/support/statalist/faq

* http://www.ats.ucla.edu/stat/stata/

**References**:**st: RE: length of a variable name***From:*Lee Sieswerda <Lee.Sieswerda@tbdhu.com>

- Prev by Date:
**st: RE: dialog box programming** - Next by Date:
**st: From: Kim Price <kprice@hsph.harvard.edu>** - Previous by thread:
**st: RE: length of a variable name** - Next by thread:
**st: Matrix** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |