Psych-DS Update: Upcoming publicity, R tools, and opportunities

[Melissa Kline Struhl]

TL;DR

(1) We are going to begin encouraging adoption - please help us spread the word!!!

(2) The validator is live, and ready for feedback

(3) Next up is scoping the R package - help us figure out what this tool needs!

Hi Everyone!

This email is primarily a sneak peak for a major Psych-DS milestone coming next week: we are going to start a first round of publicity for Psych-DS and encourage people to start using it widely! 

We've made a lot of progress on some of the big technical hurdles standing in the way of researchers using the standard in practice, and looking at the timing of our remaining grant funding and some upcoming collaborations and workshop opportunities, we’ve decided that this is the time to start building momentum for adoption!

What's New

Validator is Live: Thanks to Brian Leonard, we now have a working validator, available both in the browser and as a command-line tool. (There's a longer update on the other materials we've been working on at the bottom of this email.) 

Next Tool in Development: We’re planning an R package featuring a local Shiny app that will help researchers structure their data and build high-quality data dictionaries. This builds on work several of you have contributed to, and we'd LOVE your feedback on the initial direction. In the tradition of Psych-DS, you can get started by commenting directly on this google doc. 

Tutorial/Workshop: Returning to our roots at SIPS, I will be leading a workshop to help people convert datasets to Psych-DS during the remote conference (May 21-22). It's at a fantastic time for our European collaborators, slightly less so for the Americas [12:30pm in Budapest, 6:30am EDT on Thursday, May 22]. We'll use this opportunity to develop reusable training resources and gather more community feedback. 

Spreading the Word

Next week, I’ll be sending an email to the broader community of people interested in Psych-DS, as well as posting on a few listservs and Bluesky, announcing that Psych-DS is ready for use. 

If you know of other venues where we should be advertising, please share them! No rush—we'll be continuing outreach past this initial push next week, and any suggestions are welcome.

Similarly, if you know of people I should talk to or venues/opportunities for Psych-DS tutorials or discussions, I’d love to hear about them.

Opportunities to get involved

If you're interested in getting (re)involved with developing Psych-DS, please take a look at the technical update below - if you see something you'd like to work on, either send me an email (mek...@mit.edu) or open a github issue on the relevant repo to start the conversation!

In particular, if you have a moment this week, please check out the validator tool and let us know if anything is confusing or needs improvement.

We’d also love your thoughts on our candidate design for the R tool—you can add feedback directly to the Google document.

I deeply appreciate the support and patience from all of you, and I’m excited to finally get Psych-DS in front of more people. Thank you for being part of this!

Best,

Melissa Kline Struhl

*********

Over the last year+, Brian has taken the lead on creating a number of Psych-DS resources; most of these are still in progress and open for contributions.  Here is a brief overview of some of what we have now: 

A working in-browser and CLI validator! You can read more about the technical details of how this works by starting at our GitHub landing page. The short version is that we've been able to take the big Psych-DS specifications document and re-implement it as a schema model which can then be used by validators across tools and contexts. 

Documentation, including a Getting Started Guide for people who create Psych-DS datasets by hand 

--> We've now tested this version of the validator/documentation with grad students at MIT, but are eager for much more feedback on the dataset building + validation process and especially the error messages you may receive as you go through the process. The validator webpage has a feedback form that can be used to automatically create github issues without an account - this is a quick and easy way to drop us a line if you encounter bugs or messages that don't make sense.

A gallery of example datasets [caveat - we're working on updating these for next week's release, the posted versions have not been run through the current validator.]  If you need some valid datasets in the meantime, check the test data included in the validator codebase. 

A couple other activities of note: 

* We've implemented Psych-DS for datasets downloaded from Children Helping Science, which is used by researchers at 150+ universities to collect developmental psychology data.  This gives this community access to already-built Psych-DS datasets, and we hope it will be a model for future tools as well!

* We are working with the NIH-funded BBQS consortium to help standardize behavioral datasets in the neurosciences, especially where behavioral data is aligned with imaging, neurophys, or other kinds of data streams. This is very nicely parallel to our planned work to build converters for the BIDS data standard.