Well, I'm certainly using it—I've offered a bit over 150 bounties (>90% of them on other people's questions). Other users have offered a substantial number of bounties as well. I'd say it's somewhat effective, but better questions are more effective.
However, most of my bounties have been to reward good answers rather than to draw attention. A number ...
That's a good, if perhaps risky, use of your hard-earned reputation: it presumes the bounty will actually result in the kinds of answers you seek. As alternatives (or supplements) to this strategy you might consider (a) posting your own answers; (b) rewarding existing answers of the type you wish to encourage; (c) promoting selected questions or answers in ...
See Etiquette for answering old questions addressed in comments?. I don't think anything changes just because a bounty's at stake. (If someone's really keen to win that bounty they needn't give anything away in comments before they're ready to write an answer.)
You're mixing questions.
Are repeated bounties against any rules? Apparently not. Your reputation is yours, to gamble away if you think it's worth the risk.
Is offering them regarded as poor behaviour? I haven't detected any feeling against repeated bounties, but then they seem to be rare.
Would a second bounty work? An optimist would argue that you ...
(before looking at the linked question, as a general principle)
One advantage of featuring an answer you take the trouble to award a bonus to is that more users get to see a very good answer or answers, benefit from it, and upvote it (or all of the useful answers) as well. It's often hard to get reputation here - especially for newer users - so good answers ...
The first clustering algorithm you will learn about is k-means. That is nice and good, but unfortunately people will sometimes think that k-means is the One Tool to solve all their clustering problems and neglect finding out about drawbacks and alternatives to k-means.
In such cases, I find both Anony-Mousse's and David Robinson's answers to the question ...
Bounties operate as follows:
You cannot start a bounty if you have less than 75 reputation. You cannot make them for more reputation than you have (you can't start a 100 point bounty with 90 reputation but you can start a 50 point bounty)
Reputation cost for bounties is paid immediately, as soon as you start them. Presumably the user had 106 reputation and ...
We have a question on Anscombe-like datasets with the same box and whiskers plot (mean/std/median/MAD/min/max), which currently has five upvotes and no answers.
I like Anscombe's quartet a lot, since it wonderfully illustrates how summary univariate or multivariate statistics lose (you can also say "compress") information. Frank Harrell's example of very ...
Moderators can't do anything about what happened to the bounty, sorry. In fact there are very few circumstances in which we can do anything at all with bounties.
I definitely agree that the "will be auto-awarded" message is somewhat misleading about what the situation is, because it either seems to suggest that there's nothing to do and it will definitely ...
Unfortunately, the answer is no. You must assign the bounty during the bounty + grace period; otherwise it vanishes. If you don't award a bounty and there is an answer with the highest score (minimum 2 upvotes) gets half of the bounty. Bounties cannot be canceled, prolonged, etc.
Our soon-to-be-infamous question At each step of a limiting infinite process, put 10 balls in a urn and remove one at random. How many balls are left? is spawning its share of answers I wish I could show everyone (along with the, ahem, other kind). In particular:
whuber's novel-length treatment
Yes, neither one needs bounty rep. But both ...
You might do better to ask this question on meta.SE (or perhaps have it migrated); the bounty policy is system-wide. There is nothing we (including CV's moderators) can do about it here. On meta.SE you might get some attention from the developers, who could address this, if they so desired.
For what it's worth, I sympathize with your situation. This ...
Getting Information out of Blackbox Models - RandomForest / XGBoost
Although I closed the thread as a duplicate (and I continue to think that was correct), the answer is really good and has not gotten the votes it deserves, IMO.
(Actually, it isn't clear that it's possible to add a bounty to a closed thread. That makes @amoeba's argument for ...
That seems reasonable, but it is beyond our control; it would have to be adopted by the developers. You could ask on meta.SE (but also note that they keep tabs on site metas to stay aware of requests like this).
Note however, that you can always add a bounty after the existing bounty has expired / been awarded. The rules are that you may be required to ...
Is there an intuitive interpretation of $A^TA$ for a data matrix $A$? should be definitively awarded as it provides short but clear and pretty detailed answer for the very basic and very important question.
[Bounty started 7 April 2016, and has since been awarded]
I think that lacerbi's answer deserves some recognition because its bibliography is incredibly good. The articles cited here are incredibly helpful to me and I had not previously encountered them.
[A bounty has been started on this answer, 5 Feb 2016. Now awarded.]
I do not know if it is a memorable answer by the overall standards of CV, but the accpeted one here is the one I found most useful of all questions I have asked:
Example for a prior, that unlike Jeffreys, leads to a posterior that is not invariant
[Bounty started 9 Feb 2016. Now awarded.]
Thank you very much for your largess!
I constantly go back to some posts that have helped me understand certain concepts. A good number of them are yours...
Right off the top of my head:
@gung's post on the linearity of polynomial regression with absolutely self-explanatory plots is fabulous.
@amoeba got a prior 50-pt bounty for one of the all-time ...
The basics are as follows:
If you place the bounty you can award it as you see fit.
If nobody posts an acceptable answer and so you don't choose to award the bounty, typically the bounty simply lapses unawarded (the reputation already having been lost).
Under particular circumstances, an unawarded bounty may be - fully or partly - awarded automatically. ...
I am new here, but a long-time user of stack overflow and other exchange sites.
I have used a bounty once, to see how it worked. I have wanted to use a bounty many, many times.
Here's a problem I see with bounties (that is, a problem that makes them not useful for the only purpose I want to use them):
I only ever want to put a bounty on something because ...
It's possible to extract the number of bounties awarded (regardless of reputation transferred) with this query. The top 10 questions by number of bounties + 1 for an accepted answer are:
Examples where method of moments can beat maximum likelihood in small samples?
Conditional inference trees vs traditional decision trees
Negative binomial regression ...
Are you familiar with the Advanced Search options? They help a lot filtering questions (and answers).
For example, if you want to search for questions with accepted answer type in the search field box:
hasaccepted:1 (click me).
To search for questions with accepted answers and which were not closed or put on hold type:
hasaccepted:1 closed:0 and so on ...
Alecos Papadopoulos wrote this really nice answer to the question "Testing for autocorrelation: Ljung-Box versus Breusch-Godfrey". It's one of the most solid posts I have seen so far on CV, providing a well-researched canonical answer.