Watts-Lab / team_comm_tools

An open-source Python library that turns multiparty conversational data into social-science backed features.
https://teamcommtools.seas.upenn.edu/
MIT License
3 stars 5 forks source link

Get Dependent Variable Data for the Fracture Tasks #62

Closed xehu closed 3 months ago

xehu commented 1 year ago

@markwhiting Currently, I have a clean version of the Fracture data (cleaned up from https://github.com/Watts-Lab/team-process-map/blob/main/feature_engine/data/raw_data/fracture_data_raw.json.txt, into this csv: https://github.com/Watts-Lab/team-process-map/blob/main/feature_engine/data/raw_data/fracture_conversations.csv). However, the only dependent variable I have for Fracture is whether the team fractured or not; I do not have insight into how the team performed on the actual task.

Do you happen to still have any of the original data, where it would be possible to extract that dependent variable information! It would definitely make it easier to compare these tasks apples to apples!

xehu commented 1 year ago

Just documenting this as the current link to the Fracture old repo: https://github.com/StanfordHCI/predicting-team-viability/tree/master/inference/Fracture%20data/Fracture%20data

xehu commented 1 year ago

One potential thing we might consider doing: we can extract the final submissions from the text chat and evaluate them ourselves? Most of the submissions are within the text themselves, so we just have to (1) create some kind of evaluation rubric; and (2) use that to evaluate the tasks.

xehu commented 1 year ago

This leads to a question:

xehu commented 3 months ago

Closing as this is about a previous version of the project and no longer related to the toolkit