In the SPSS syntax output a variable is missing and therefore all variables after the missing one are named wrong. Did anyone encounter this also? It only is in one survey, the others are OK. But it is consistent: every time I output the syntaxt I receive this:

GET DATA
/TYPE=TXT
/FILE='survey_742376_SPSS_data_file.dat'
/DELCASE=LINE
/DELIMITERS=","
/QUALIFIER="'"
/ARRANGEMENT=DELIMITED
/FIRSTCASE=1
/IMPORTCASE=ALL
/VARIABLES=
V1 A100
V2 A100
V3 A100
V4 A100
V5 A100
V6 A100
V7 A100
V8 A100
V9 A100
V10 A7
V11 A16
V12 DATETIME23.2 (after this V13 is missing!)
V14 A94
V15 DATETIME23.2
V16 DATETIME23.2
V17 A15
V18 A2
V19 A3
V20 A1
V21 A1
V22 A370
V23 A2
V24 A2
V25 F1.0
V26 F13.12

Also: I asked for year of birth, in the datafile it is outputted as: '1957.0000000000' (Question type Numerical input, integer only, maximum 4 characters, between 1940 and 2000)
and in the syntax this is formatted to as A3, while it should have been F4 (I think). Can anyone give me a direction where I should try to find (and solve) this strange behaviour?

Thanks in advance,

Tammo ter Hark

We narrowed down this to an issue of not outputting the lastpage on which the respondent stopped the survey.

In a survey with my settings this is V13, but when you add extra atributes to the token tabel, this will get another variable value.

I am not sure if this is the cause of the SPSS malfunction (I guess I would have seen many more complaints if it is the case with all users), but I would very much like to solve this.

So please:
* add the lastpage to the SPSS output (both to syntax as well as data)
* review the numbering of variables in the output of SPSS

And another wish:
* please ask for a delimiter, since the stadard delimiters (a comma for a field) will break things when people can answers with texts, inserting comma's at will...

tammo wrote: So please:
* add the lastpage to the SPSS output (both to syntax as well as data)
* review the numbering of variables in the output of SPSS

www.limesurvey.org/en/community-services/feature-requests

And always remind : LimeSurvey core developpement are on our free time.

Of course, I am very much aware of that. And I am also grateful for that.

But bugs and improvements can be reported, can't they? The fact that V13 is missing seems like a bug. And the other things I mentioned could improve LimeSurvey.

Should I have stated it differently or in another place?

Bug reports are always welcome, as well as improvements.

But if you feel it is a bug, then the best way is to report it in the bugtracker, and everything that will improve Limesurvey at the link "feature requests" that Dennis postet.

I will do so. I first wanted to report it here and hoped to hear if this was either my fault or lack of experience or that it was really a bug. I guess it is really a bug and will report that.

Thanks,

Tammo

holch wrote: Bug reports are always welcome, as well as improvements.

But if you feel it is a bug, then the best way is to report it in the bugtracker, and everything that will improve Limesurvey at the link "feature requests" that Dennis postet.

Bug report and Feauture request use same service. Then bug can be moved to FR and FR to bug

Thanks, I submitted an error report!

On the same time my colleague is working on a solution, which we have working locally now for the lastpage issue.

In testing it we found several other strange things in the way variables are formatted in the SPSS syntax file. Working on it now. Jan (my colleague) will write a pull request to the GIT.

Feature request / bug report:
bugs.limesurvey.org/view.php?id=9398

And pull request:
github.com/LimeSurvey/LimeSurvey/pull/257

This is the change:
github.com/LimeSurvey/LimeSurvey/pull/257/files

I copied the two lines from $fieldname=='id' to $fieldname=='lastpage'.

Please support this issue if you want to see the lastpage variable in SPSS output.

bugs.limesurvey.org/view.php?id=9398