Aoife Moloney venit, vidit, dixit 2024-07-01 22:21:37: > I disagree a little on the results to be meaningless just because the tone > or phrasing is positive, particularly with the topic of AI because it is > far too easy to list all the negatives associated with AI, and doing that > doesn't move the conversation forward. Structuring that question in this > ranking way allows for a better conversation on how to use AI in whatever > area comes out as the best or one of the better fitting spots for AI in > Fedora based on how people ranked them. We want to focus on the potential > benefits (or positives) that we could leverage from AI while understanding > the preferences of our community with this initial survey. The point is that "rank 3" for one person has no connection to "rank 3" for another person, yet you will aggregate results based on statistics which use the numerical value of the rank. This is really meaningless from the point of test design. I'm not sayting that ranking itself is a meaningless concept. There are a lot of theories about ranking/ordering versus point scales etc. Even the concept of having to rank among a fixed set of choices can make sense if there is a good reason for that fixed set of outcomes, for example in voting between candidates. There are voting systems which use exactly that system and award "points" based on rank, in order to determine the total outcome. This really only makes sense when the set of outcomes is predetermined. But your goal is to find out the "usefulness" of a set of questions. And you will use "rank 3" to assign a "usefulness level", that is you will use it as a "metric" on a scale. Yet you force everyone to assign a "usefulness level" of 3 to all questions on average ( (1+2+3+4+5)/5 = 3 ). Or, to put it differently: Think of everyone using their own scale/ruler, as is very clear in the answers to this thread. You align those scales by labeling them with a scale from 1 to 5 when indeed each user labelled their rulers quite differently, say from -10 to -6, 1 to 5, 6 to 10 on some imaginative "usefulness scale". The least you can do is a semi-absolute scale "from least useful to most useful" (relying on a common understanding of the terms, as usual), or an open choice of "useful options" between 0 and 5. Michael -- _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
