Yes, but you'd have to take care of that through the client side (programming your experiment), and saving the data in the batch session. If you explain some more details of how you've coded your study and which information you want to store to continue later, I can help
I have a full list of tasks that need to be completed. If someone's computer crashes or they have to stop I would like them to go back to JATOS and start where they left off. This would require a check to see what data was recorded for this worker and then to start at the end of the list.
I had a feeling it had something to do with the batch. I am assuming it is possible to have a worker write to the batch session and then be able to read from the batch session for that particular worker.
Ah, ok, that's a relatively straightforward use case. As I understand it, you don't necessarily want people to be able to start again half way through a task that they just started, it would be ok if they restart the task.
If you haven't done so already, I would start by separating tasks into different components.
Then, you'll need a first "landing and sorting" component. You won't be able to code this with jspsych, not at least with existing plugins. In that component, you'd need to:
Get the participant's ID (more on this below)
Figure out if they already started any of the tasks
Assign them to the correct component/task
When each component/task finishes, first update the batch session to include the last completed component by each participant, then submit the data to JATOS.
On the participant ID: There are multiple ways to do this, depending on how you recruit participants. If you recruit via prolific/mturk, you can use the IDs from the marketplaces. If you recruit sending personal multiple links, you can use the JATOS ID. If not, you could ask them to create and remember an ID (can be unreliable, but certainly an option).
Let me know if you need any more details or pointers on any of these points. I attached an example for this kind of landing and sorting component. it's overly complicated because it was a longitudinal study with multiple randomization dimensions, but it might give you a starting point. (it's an .html file but I changed the extension to txt to upload it here)
Thank you very much for sharing. I already started working with the batch session data, and I think that will work. I already have what you call a "landing" page and all tasks are separate components. I will got through your attached code to see how you do it.
Comments
Hi,
Yes, but you'd have to take care of that through the client side (programming your experiment), and saving the data in the batch session. If you explain some more details of how you've coded your study and which information you want to store to continue later, I can help
Best
Elisa
Hello Elisa,
Everything is written with jsPsych.
I have a full list of tasks that need to be completed. If someone's computer crashes or they have to stop I would like them to go back to JATOS and start where they left off. This would require a check to see what data was recorded for this worker and then to start at the end of the list.
I had a feeling it had something to do with the batch. I am assuming it is possible to have a worker write to the batch session and then be able to read from the batch session for that particular worker.
Jason
Hi Jason,
Ah, ok, that's a relatively straightforward use case. As I understand it, you don't necessarily want people to be able to start again half way through a task that they just started, it would be ok if they restart the task.
If you haven't done so already, I would start by separating tasks into different components.
Then, you'll need a first "landing and sorting" component. You won't be able to code this with jspsych, not at least with existing plugins. In that component, you'd need to:
On the participant ID: There are multiple ways to do this, depending on how you recruit participants. If you recruit via prolific/mturk, you can use the IDs from the marketplaces. If you recruit sending personal multiple links, you can use the JATOS ID. If not, you could ask them to create and remember an ID (can be unreliable, but certainly an option).
Let me know if you need any more details or pointers on any of these points. I attached an example for this kind of landing and sorting component. it's overly complicated because it was a longitudinal study with multiple randomization dimensions, but it might give you a starting point. (it's an .html file but I changed the extension to txt to upload it here)
hope this helps
Elisa
Hello Elisa,
Thank you very much for sharing. I already started working with the batch session data, and I think that will work. I already have what you call a "landing" page and all tasks are separate components. I will got through your attached code to see how you do it.
Thank you,
Jason