Hi all!
My Stroop Task experiment is running and I want to be able to read my participants’ results (answers, reaction time, etc) but it’s displayed in an Excel spreadsheet which is extremely difficult to read. I’m planning on having around 60 participants so I don’t know how I’d be able to extract this information without it taking a hundred years. Is there another way to view/decipher my results??
Sorry I’m completely clueless with this kind of thing, I just created my experiment following YouTube videos.
Here’s an example of my results. Thanks!
Alicia_stroopExperiment_2024-07-01_11h27.24.506.csv (26.1 KB)
Hi,
For each block you seem to have used a different copy of your trial routine, which means that your score (e.g. response2.corr) and reaction time (e.g. response2.rt) columns aren’t aligned. You could start off by creating columns for them which average across all blocks.
Personally I would probably want to calculate the median reaction time for the correct responses only in each block. You have a column called congruent which shows whether the block is congruent or not. However, you don’t seem to have a column to show whether the text was in English or French. You might need to add that manually based on whether the loop is trials, trials_2 etc.
I tend to process files in SPSS, and I know others use R.
Your blocks in the example are EC, EI, FC, FI, EI, FI. Did all participants see these six blocks in the same order? Is there a reason why there are two incongruent blocks for each language?
If you need help with this, and have all your data, then it would probably only take an hour of consultancy time for it to be processed into score and median RT for correct responses for each block for each participant.
Hello
It is probably not necessary to make different copies of your trial routine, but rather to reuse the trial routine so that the columns are aligned.
But I suppose it is too late to fix this as your experiment is already running. Try to analyse some test data and check that the data file has the structure and data you expect and need next time.
Best wishes Jens
Thank you so much for the detailed response!
Participants see the 6 blocks in that same order, yes. The second EI is a bilingual trial in English and French, so for example the word would be ‘noir’ written in the colour ‘yellow’ and participants would have to press ‘y’ for yellow. The second FI is French and English so the word BLACK written in the colour ‘yellow’ but participants would have to identify yellow in French, and press ‘j’ for ‘jaune’.
As I’m such a novice with this program it’ll take me some fiddling about to figure out how to implement your advice but really appreciate the help
Thank you so much for the help Jens.
It’s not actually too late because I’ve not had any real participants do the experiment yet, but how can I reuse a trial routine if I have different conditions/using different languages?
Hello
That is good to hear. Simply insert the routine a second, third aso. time and use a different Excel-file for the second or third time. See toy-example below.
BTW, you can turn off the on-/offset-times of routines and components when the experiment is running as intended and you don’t need this information for further analysis or participant selection.
ReUseRoutine.psyexp (10.5 KB)
ReUseRoutine1.xlsx (9.6 KB)
ReUseRoutine2.xlsx (9.6 KB)
Best wishes Jens
Also, add a column to your spreadsheets called Language with the language condition
1 Like
So I created a new experiment - I inserted the same routine multiple times using different Excel files, and it seems to run fine on a local browser but when I go to run it on Pavlovia I get an error message saying
" Unfortunately we encountered the following error:
- when starting the experiment
- when opening a session for experiment: dmuradkontos/stroopexperiment
- the experiment status is PILOTING but the pilot token is missing from the URL
Try to run the experiment again. If the error persists, contact the experiment designer."
While my old experiment is shown to be on the PSYCHOJS platform on Pavlovia, this one just says UNKNOWN
I’m at my wits end now because I’ve been struggling with this for hours…
Good morning
Have you tried setting the experiment to Running or Inactive? Did the experiment sync without errors, did you clear the cache? After testing the experiment locally, I usually delete the lib folder and the JS script files before resycing with Pavlovia. I am not sure if the last step is really necessary. The experiment won’t run until the platform is recognised.
Best wishes Jens
Coming back to this comment after having created a new experiment (I’ve now used the same copy of my trial routine with different excel spreadsheets and I also added a language column) does this excel sheet with my new results look any different / easier to digest than the original? Thanks so much
Thank you!!
My experiment simply wouldn’t work and I feel like I exhausted all my options so I just made a new ‘new’ experiment which, after a few hours of work and screaming into the void, I believe is now running. I implemented what you said earlier about inserting the same routine with different Excel files, as well as turning off the onset/offset times for instructions but I just realised that you might have meant turning it off for all routines?? I think it’s too late now because I don’t want to mess up what appears to be running…
Hello
Turning off the onset- offset-information of the routines and components does not affect your experiment. It only affects what is reported in the result file.
Best wishes jens
Hi dmurkon,
I totally drowned in overwhelm when I first tried to understand the excel and realize, that only the right lower quadrant of the entire excel holds valuable information. My solution was to let PP register onset… as it does and afterwards just select the rows of the needed variables. Beware of them also being presented first and then the reaction is recorded. In my case I could drop 80 variables and 169 rows. Of these 169 rows I needed one variable with the stimulus, which I inserted as new variable into the remaining 200 rows. That means some lines of R-code tailor, what you need. If you use loops they might be only presented once, in my case the difference between 169 and 200. I hope that calms frustration a bit. :)
Thanks for detailed explanation.