spssforum.com

Dear all.

I have made a survey on limesurvey. I can download my results as an Syntex and data file for SPSS.
Now I would like to import the data in to SPSS. So first I have opened the Syntex file in SPSS. Second I changed the exact location of the data file in the fourth row and clicked Run >> Run all.

These are the errors a get (I only copied 1 error #100 but every questionvalue he gives this error):

Does somebody know what I do wrong?

Kind regards,

Mari Hilgerdenaar
The Netherlands

>Error. Command name: GET DATA
>(2269) Failure opening file: survey_41972_SPSS_data_file.dat
>Execution of this command stops.
CACHE.
EXECUTE.

>Error # 105. Command name: EXECUTE
>This command is not valid before a working file has been defined.
>Execution of this command stops.
*Define Variable Properties.
VARIABLE LABELS V1 "id".

>Error # 100. Command name: VARIABLE LABELS
>This command is not permitted before the beginning of file definition
>commands.
>Execution of this command stops.
VARIABLE LABELS V2 "submitdate".

>Error # 100. Command name: VARIABLE LABELS
>This command is not permitted before the beginning of file definition
>commands.
>Execution of this command stops.
VARIABLE LABELS V3 "startlanguage".

>Error # 100. Command name: VARIABLE LABELS
>This command is not permitted before the beginning of file definition
>commands.
>Execution of this command stops.
VARIABLE LABELS V4 "A1: Hoe frequent heeft u contact met de Belastingdienst?[Dagelijks]".

My first guess is that the file path of the file survey_41972_SPSS_data_file.dat has not been correctly defined. For instance, if the file is in c:\temp, you should put "c:\temp\" in front of the file name by editing their syntax file.

My second guess is that you are using other software like Excel to view the same .dat file while trying to import. Close that Excel program.

Hi Penguin_Knight!

I thought I already tried to put the exact location in front, but now it works fine, thanks a lot!

But I have a second problem. This is about the variable view:
Normaly every question is a variable with a couple of values

Question 1: How are you A=Good B=Bad C=Not Good, Not Bad
Variable 1 is question 1
Values A=1 B=2 C=3

In my case I have only variables and no values! So:
Question 1A = variable 1
Question 1B = variable 2
Question 1C = Variable 3

So I can only see 1's and 0's in my data view.

Do you also have an idea how I can solve this second problem? I hope so

Kind regards!

Mari Hilgerdenaar

This is likely because the original online questionnaire used a wrong type of question format. In your scenario, the questionnaire designer should have used radio button type of response, where user can only pick one answer. Instead, he/she used a multiple response format, which will return one yes/no (representing checked/unchecked) for each single option.

Assume your respondents didn't check multiple boxes, there are ways to savage the data. For example, you can use this syntax:

Hi Penguin_Knight!

Thanks for the reply. I am affraid you're right. (I designed the survey myself and select the multiple question option, and then a put al rule in it that no more than one answer is alowed.)

Are there other option to take all the variables of one question together. For example:
Analyze >> Multiple Response >> Define Variable Sets

So I make a variable set of every question and than I compare them.

Do you think this is the best way?

Kind regard!

If frequency table and cross tabulation are all you'd do then multiple response set is fine. If you'd like to do more like correlation test, then it's not suitable because the variables created from multiple response set are unavailable to procedures outside multiple response set frequencies and multiple response set crosstab.