Technical Question Regarding Data File Access

[jasonp220]

Although Media Lab has the unique machine code field which could solve my issue, I'm also interested in saving time. I'm a grad student with no $$$ to pay assistants.

Since Media Lab can write to data files on network shares, I'm wondering when the program opens the data file, writes to it, and closes it. If the file access and write only happens at the end of the experiment, then it seems I could simply use a password to prevent the experiment from ending in order to avoid conflicts.

If you can know the technical details of the data file access, please share them with me. For now, I'll use the unique machine code.

Jason

[jarvis24]

Clever idea. Yes, the data are written at the end of the session. However, I would embed your password in an instruction item rather than by using the global password option in the Preferences. There are some data writing activities that occur before the latter but not the former.

Also, be aware that some item types do write data during the session--e.g., essay item data, fill in the blank data when requested by use of the file name parameter. There are also some initial files that are modified when the session starts such as currentsubjectinfo.txt and autostart.txt (if applicable). If you start each session one after the other, that should not be an issue.

Again, I think this is a clever solution but I would only recommend trying it on an experimental basis to see if it works for you and to be mindful that the content of your experiment could have an impact on its success (as mentioned above).

If you try it I would be curious to hear how it goes.

[jasonp220]

Thanks for the info, Jarvis.

I'll let you know how it goes.

Jason

[jasonp220]

Jarvis,

It turns out using the password option with an instruction line does the trick except for one odd outcome. The CSV data file by questionnaire puts all the data on one line and merges the last value of one subject with the first value of the next subject (typically the subject number).

For example, if subject 1's last entry is "ok" for an instruction then, then their will be a cell with the value "ok2". Any ideas why this happens?

I could fix this with a macro, but it seems easier just to use the "by variable" data file which looks fine. The SPSS (.sav) file looks fine as well; I haven't checked the .txt file because I'm not SPSS literate and am not sure how to make that work with the .SPS file yet.

Jason

[jarvis24]

Thanks for the update--that is an odd result. Will contemplate it but I can't imagine how that would happen as consequence of this particular strategy. Hmm.