I have run an experiment using jsPsych hosted on Pavlovia recruiting via Prolific. Almost all of the data was collected and stored fine. However, 5 data sets for participants that completed the study (had the correct completion code for Prolific) just did not save and no credits were spent for these 5.
I have now spent additional funds to replace them, but I am wondering why this might have happened.
Could it possibly be something to do with the fact that I only assigned enough credits to complete data collection, but a total of 8 credits have been ‘reserved’. Could it be that Pavlovia stopped storing the data once it ran out of ‘assigned’ credits because some of them had been ‘reserved’? If not this, can you think of any other reason why a handful of data files did not save correctly even though the rest of them worked just fine?
As I said, I have already paid additional participants to replace these lost data, but it would be good to know where the problem came from so that I can avoid it in the future.
Have you tried downloading the data from the project page? There have been some recent issues related to the connection between the run server and the gitlab server.
As I said, 5 credits were not used and 5 data files were not saved. Whatever happened, the credits were not spent either. Nonetheless, all participants successfully completed the experiment and provided the correct code having been automatically redirected back to Prolific.
I’m also having an issue accessing my experiment data. I tested the study with one pilot participant through Prolific, as well as with myself, and in both cases no data were saved in the Pavlovia file.
The study first runs on Pavlovia and then continues on LimeSurvey. I was able to retrieve the LimeSurvey data without any problem, which suggests that the participant completed the full study. However, no Pavlovia data file was generated for this participant.
What is confusing is that the experiment works perfectly in pilot mode. When I test it locally, the data are saved correctly and I can download the files. The issue only appears when running the study online. When I tested the online version using the browser DevTools, I received several error messages, including a “500 Internal Server Error.” I am attaching a screenshot of the console so you can see exactly what happened.
Would you be able to take a look at the error messages and let me know what might be causing the data-saving failure, and what I should adjust to prevent this from happening again?
It’s really hard to know what has gone wrong in jsPsych scripts because of
people writing their own code (and I’m afraid I don’t have time to inspect the code for individual studies). One common issue that we’ve seen when people write their own code, in any framework, is that they forget to await the call that pushes the data to the server. It’s really important to include an await call or your redirect will occur too quickly for the data to have saved.
In @alexisb case I see you’ve got something in the manually asking the participants to look out for warning signs from the browser but I’m not sure these warning signs always occur.
So I think the first port of call is to revisit your data saving code.
