Sure, and I regularly do (“Well, if situation X seems like it would produce anecdote Y, then all anecdote Y shows us is that situation X happened, not that contention Z is necessarily true—only if situation X shows us that contention Z is true”).
I would surmise that not all commentors are willing to be that forgiving.
And how else should I update after reading two self-selected, subjective assessments? This has a perfectly reasonable Bayesian interpretation.
EDIT: Also note that the grandparent was posted before AnnaSalamon actually fixed the problem at hand.
EDIT x2: And while I’m endlessly editing this comment, let me note that most of this drama could have been averted if someone had just posted the damn data instead of coming up with multiple, bad excuses. Lots of guilty parties, only a couple heroes (in my book, at least).
And how else should I update after reading two self-selected, subjective assessments?
Very little. I was explaining why your comment was downvoted so much. I said “technically inaccurate” as opposed to “wrong” because I am sympathetic to your point of view; it is almost no data. But it is a little bit of data.
Data is not the plural of anecdote.
(That quote is commonly used by Science and is technically inaccurate under Bayes, in case you were wondering.)
Can we be forgiving and assume that multiple anecdotes fail because they have a consistent bias related to how they are obtained?
Sure, and I regularly do (“Well, if situation X seems like it would produce anecdote Y, then all anecdote Y shows us is that situation X happened, not that contention Z is necessarily true—only if situation X shows us that contention Z is true”).
I would surmise that not all commentors are willing to be that forgiving.
And how else should I update after reading two self-selected, subjective assessments? This has a perfectly reasonable Bayesian interpretation.
EDIT: Also note that the grandparent was posted before AnnaSalamon actually fixed the problem at hand.
EDIT x2: And while I’m endlessly editing this comment, let me note that most of this drama could have been averted if someone had just posted the damn data instead of coming up with multiple, bad excuses. Lots of guilty parties, only a couple heroes (in my book, at least).
Very little. I was explaining why your comment was downvoted so much. I said “technically inaccurate” as opposed to “wrong” because I am sympathetic to your point of view; it is almost no data. But it is a little bit of data.