Generating and - saving - random values from responses.xls files

[acattani]

Hi Empirisoft,

Please help!
I believe I am not able for Medialab to read the excel file responses.xls. It shows the word <vig1> but not the actual sentences that should be generated with the random strings in the excel file.
Please take a look at the files I have enclosed (this is really my first attempt ..). I know that in a previous thread this problem came up, but I can't see what is wrong here ..

[jarvis24]

Acattani,

It gets tricky if you are trying to SAVE randomly generated values in an XLS file because they'll be different every time you open the document. However, this is only a problem if you want o see what values were generated *after* a session when you open the .xls file. Is that the case, or is your issue with generating and presenting the random stimuli in the first place? If it's the latter, try entering in some simulated responses where appropriate (i.e., where they are currently now blank). When you do that, do you see results in the new variable cells you are creating? Does that make sense?

Blair

[acattani]

Thank for your response, but I am not sure I understood. My actual problem is in generating and presenting the random stimuli in the first place when running the study. But, rather than showing the full vignette it only shows 'ok' (in place of the full random string included in <vig1>) and on the next slide it shows '<vig2>' (in place of the full random string included in <vig2>). So I cannot simulate any response for the vignettes ... Hope it makes sense.

[jarvis24]

If I understand correctly--in your responses.xls file, you have a calculated variable called "vig1" which you are inserting into a questionnaire variable wording by using <vig1>? If so, this makes sense. I think the trouble might be arising from the fact that you have named the questionnaire item "vig1" also. Consequently, the response for that instruction item is getting inserted which is "ok" -- the data value for an instruction item.

A fix here I think would be to simply rename the variable name in your .que file from "vig1" to "vig" or something like that. So there is no confusion that the inserted <vig1> value is coming from a calculated cell in responses.xls and not from a response to an item within the MediaLab .que file.

[acattani]

Thank you - i made some changes in the variable name and made other changes .. so atlas it now work! Thank you!
One thing at a time .. I am now starting to check what values were generated *after* a session when a open the participant .xls file.

[jarvis24]

Glad you found the resolution--and the individual responses.xls files for each subject (they can come in very handy!!)