First of all, thank you for your great work! The tool is very intuitive to use (by researcher and participants), very stable, and the documentation is great too.
I’d like to suggest a possible feature, or ask whether there’s already a workaround for it. Would it be possible to filter alters in interfaces based on their positions in earlier sociograms, that is, in which circles they were placed?
For example, it would be very useful to use the dyad census or per alter form only for alters who were placed in specific circles of an earlier concentric sociogram. At the moment, this can be partially solved by using ordinal bins instead of a concentric sociogram, but then the resulting closeness variable can’t be used as a layout variable in later narrative interfaces.
This kind of filtering would be a great feature, since concentric sociograms are such a helpful way filter alters in larger networks. Maybe there’s already a workaround that I’ve missed?
Thank you so much for the kind words about the tools. It means a lot to our small team to know the impact!
The feature you describe doesn’t presently exist, but we appreciate the suggestion. @Joshua and others might be able to chime in about ways you could tackle building this in yourself, but, depending on the needs of your study, there are a couple of alternatives that could be worth looking into. The first would be the use of the Tie-Strength Census interface as a way to generate edges with ordinal values attached to those edges and then filtering based on the ordinal values in subsequent screens (e.g., per alter form). Another could be to use variable toggling on the sociogram, which would only allow you to filter based on boolean values, but would enable you to generate and utilize the layout variable, etc. You may have already tried both, but wanted to offer these as potential workarounds depending on your needs.
I have added an issue to our roadmap to consider making it so that the concentric circles on the sociogram can be configured as “drop zones”, which automatically assign the value of an ordinal variable to a node placed within them.
Unfortunately I don’t have any idea when such a feature will end up being implemented, as it will fall into our (very large!) backlog of development work. As Kate suggested, you are very welcome to develop this feature yourself, either through directly contributing code (I would be happy to show you where to get started) or by contracting us/an external developer to implement the feature. Let me know if that is of interest to you.
In the meantime, Kate did a wonderful job explaining some other options.
Thanks for the fast and helpful replies! Toggling the nodes in the inner circle of the sociogram, then use the boolean value as a filter is a great idea and solves the problem with little extra effort.
I appreciate that the idea is now on the roadmap. Unfortunately, my developer skills aren’t good enough to contribute to the code myself.