SCOPING: Experimenter-created accounts and "accountless studies"

[mekline]

TL;DR

For a variety of workflow reasons, people would like it if families didn't have to make their own accounts, but these could be created either after or before a participant does the CHS-based study. This is usually in the context of handing off from other parts of a study where the family has already provided information, or where signup prior to the study is impractical or undesirable. (Maybe a booth at a fair is an example?)

For tractability and participant account control, this probably needs to be a one-way handoff: an "anonymous" account object with either/both of (1) partially filled details, children, demographic forms etc.. and (2) sessions objects can be created by the researcher, and is then passed to the family who is immediately prompted to change their password, after which the family has full control that can't be taken back by the researcher.

Imagined family experience: I follow links to take a CHS study and/or fill out forms on a non-CHS platform. Later, the researcher sends me an email that says "click here to claim your CHS account". Or, at the end of a CHS study, I'm told that I now need to make an account to let the researcher see my data. (Or make a choice whether or not to do this?)

I am first taken to a password-change page, and then to my account page, where I can see all the info that's been filled in as well as previous studies I participated in.

Big questions:

• Do we allow researchers to access data from sessions that belong to unclaimed accounts?
• Do session objects persist if the family finishes the CHS study without immediately claiming the account?

Narrative

An example workflow, from Emily Chen, with an fMRI study with desired CHS followup study:

...our lab primarily uses fMRI as a method to study infant cognition. We also use behavioral methods and have been using Lookit as our primary platform to collect behavioral data asynchronously from families. These studies are often decoupled from our fMRI experiments, in the sense that we will collect data from families online on a behavioral task, and then we invite different families to come into the lab for our fMRI experiments, even if we’re doing the same task. However, we want to start up a new study that would require infants to do a behavioral task the day before they come in for a fMRI study. Lookit would be a great platform to run this behavioral task on, as it allows families to do the behavioral study on their own time the day before rather than us having to find a time to meet them on Zoom. The sticking point is the added time it would take for them to sign up for the platform, since to get oriented for the fMRI part of the study, they must come in person to the lab for a 1-hour orientation session, which can occasionally last for up to 1.5 hours, and during that time, they sign and fill out 5-6 forms. Asking them to also sign up for another platform would likely add another 10-15 minutes (explaining the platform, having them walk through each of the steps of the signup, etc.) which we aren’t able to add to the orientation and would be an additional time burden for the family. As such, we were thinking that if we could sign them up for Lookit after they have agreed to the study and signed the consent forms, they could then log in on their own time to familiarize themselves with it, and we would not have to ask them to do another task during the orientation or ask more from them after the orientation.

Now, to the parts that we think might take the longest time: adding information for their child (and sometimes, they may see that they can fill out for multiple children and then also spend time doing that) and filling out the demographic information. We get most of the demographic information asked for on Lookit from them through our own forms for the MRI study, so it’s possible that we could create the account for them from that information, perhaps even while they are signing other forms at the orientation. Besides the time it takes to sign up, you can imagine that it may also be frustrating for parents to have to input the same information twice (once for Lookit and once for our own database for fMRI studies). I want to emphasize that this inquiry is coming only from the fact that we are running a behavioral study with a fMRI study on the same participants, not because we think the overall Lookit signup itself is the challenge!

Feature Request fields

**With this `details` wrapper, you can just use this ticket to record your findings. When you're ready to turn this into an implementation request, delete the scoping label and this wrapper, leaving only the filled out information.**

Acceptance Criteria

A precise description of how/when this issue will be considered closed, e.g.:

• Tests for new feature
• QA on new feature
• Sign off from Kim

Implementation Notes

Leave important notes/suggestions here for implementation (important files, suggested approaches, etc.)