Should SE implement a way to poll the playerbase?

[HyoMinPark]

That would generally depend on the scope of the questions. There would be samurai related questions you could provide solid feedback for.

What elements do you dislike about the job or would make you consider playing the job more? for example.

Answers could provide a variety of feedback. From I just don't like katana or the aesthetics. Right down the job feels clunky to play or is too shallow.

The trick then would be finding solutions to that feedback without impacting the things people said they liked about the job..

Perhaps, but these seem to be more in regards to general feelings about the job versus something specific like SAM 6.1 changes or just job changes in general, which I think require a bit more experience and insight on. When it comes to job design, I think the best people to speak on that are those who actively play the jobs and play them at a decent level. Not someone like me who may not play it or randomly takes it into 24-mans every once in a while.

Honestly you'd need to ask everyone. Regardless of "qualifications." However you could split the results data based on answers to a specific question... such as how frequently do you play job x. Then segregated answers those who play it alot. Those who play moderately. And those who have never played or play it very rarely.

When it comes to specific feedback, though, I think a narrower scope is better. Asking everyone creates a lot of extraneous data that needs to be filtered out; at which point, said data is more or less irrelevant to the scope of the survey.

If the survey is something specific like “what are your thoughts on the 6.1 changes to SAM”, data from those who don’t play SAM or don’t care about them isn’t really needed. You want to see what the people who play SAM think about the changes, and how they feel. Everyone else is irrelevant.