3 ms·
I like that you can run your own versions of these studies, e.g. https://github.com/mozilla-rally/rally-study-01 https://github.com/mozilla-rally/rally-study-01
by quadrifoliate 5y ago
I like that you can run your own versions of these studies, e.g. https://github.com/mozilla-rally/rally-study-01 https://github.com/mozilla-rally/rally-study-01 to collect your own data. I think it would be a novel idea to make this the primary focus of the project, enabling users to see and understand their own data before donating it to research projects.
For example, I know that the study above is called "Your Time Online and 'Doomscrolling'", but I don't have much of an idea on how they plan to quantify doomscrolling via attention or audio events. I managed to get a reasonable idea from https://rally.mozilla.org/current-studies/your-time-online-and-doomscrolling/ https://rally.mozilla.org/current-studies/your-time-online-a... and https://github.com/mozilla-rally/rally-study-01/blob/master/DATA_SCHEMA.md https://github.com/mozilla-rally/rally-study-01/blob/master/... though, so it's possible I just need to put in more effort as a user :)
- 9wzYQbTYsAIc 5y ago> I don't have much of an idea on how they plan to quantify doomscrolling via attention or audio events. It seems that what they’ve done is told us exactly what data they will collect but have left it unanswered as to what specific social science methodologies they will use to analyze the data. It is very likely that their intention is to not set that expectation with the data crowdsourcers so that the researchers have the flexibility to adjust their methodological approaches in an iterative fashion.