Incomplete data output by JATOS - columns missing (Data output is complete within OpenSesame itself)

krmkuik

Whenever I want to upload my results data from JATOS into a txt and then excel file, some crucial columns seem to be missing. Whenever I test my experiment within OpenSesame, these columns with crucial information do show up in de results, but when I do it online and use the JATOS results import, it isn’t complete. I’m doing a Lexical Decision Task. In the block loops I included an item column, with all the words/non-words. Next to it a column with the correct_response (z or m) and next to that the item_type (summer, winter, neutral words or non-words).

So, I need this information, but this is completely missing when I downloaded the results from JATOS. I've tried to attach the txt and excel files, but it doesn't seem to be working here. I did upload a shorter first draft of my experiment to the try-out server called "Lexical Decision Task Kuik". I hope you can see how in the JATOS data version there seems to be no specification of “item” and “item_type”. It does contain “practice_item” with the practice words I had, but not the necessary data from the actual “item”s and “item_type”s.

Btw: In my longer and complete version I have a trial block of 8 items, and 4 blocks of 45 items, 180 words total. In between these blocks is a feedback/break slide. This version has no feedback or breaks, so it’s only 2 blocks: the trial block and a long actual block of 180 words. I set it at 0,10 x so it’s only 18 words long if you test it.

Does anyone happen to know why the data output is different in JATOS than it is in OpenSesame? Is there a way that the columns “item” and “item_type” are logged by JATOS?

sebastiaan

Hi,

Thanks for reposting your question here 😊

I see what you mean. The problem seems to be that some of the variables (such as item and item_type) are only defined later on in the experiment, after the logger has already been executed once, and therefore OSWeb doesn't pick up on them. This is clearly undesirable behavior, and it also deviates from how OpenSesame works on the desktop, so we'll address that soon.

For now, simply adding the missing columns to the trial_block_loop (without any content) should be a workaround. Does that work for you?

Cheers!

Sebastiaan

krmkuik

Yes no problem!

Wow, I didn't expect to solution to be this fast/easy, but you're right, this does solve the problem!

Thank you so much for the fast replies.