If you want comments made in some non-standard way (e.g., certain categories of comments to go only under your top-level “meta” comment) then there should be some indication of this in the post itself.
Is it worth giving two scores, one for “how well it worked while I was doing it” and one for “how well it worked, taking into account whether I gave up using it”? My impression is that it’s quite common for an anti-akrasia technique to work well right up to the point where one becomes akratic about actually using it.
The advantage would be not so much giving you twice as much data to work with, as giving respondents the ability to express their actual experiences. If someone tried something and it worked really well but they soon gave up on it, they may be reluctant to give it either a very high or a very low score, but you might want to treat it as one or other of those.
Easier to compile in some sense, perhaps, but much much less amenable to discussion, and also much less failproof. For example, the 2nd akrasia tactics review stopped getting its responses compiled after a short while, but at least people could still read the comments.
(Hm, maybe it would make sense to pull in some of those?)
META: Put discussion about the format etc in response to this comment
If you want comments made in some non-standard way (e.g., certain categories of comments to go only under your top-level “meta” comment) then there should be some indication of this in the post itself.
Good call. Sorry for the messy experience in response to your question! I didn’t think of it until afterwards.
Is it worth giving two scores, one for “how well it worked while I was doing it” and one for “how well it worked, taking into account whether I gave up using it”? My impression is that it’s quite common for an anti-akrasia technique to work well right up to the point where one becomes akratic about actually using it.
I think that both of those could make sense, but I’m not sure how I’d go about aggregating the scores from that. I would probably use the second one.
The advantage would be not so much giving you twice as much data to work with, as giving respondents the ability to express their actual experiences. If someone tried something and it worked really well but they soon gave up on it, they may be reluctant to give it either a very high or a very low score, but you might want to treat it as one or other of those.
Would it be easier for you to compile responses if we used a survey format (either LW’s built-in feature) or Google Forms?
Easier to compile in some sense, perhaps, but much much less amenable to discussion, and also much less failproof. For example, the 2nd akrasia tactics review stopped getting its responses compiled after a short while, but at least people could still read the comments.
(Hm, maybe it would make sense to pull in some of those?)