9

Our current Area51 Statistics are mostly in the "Okay" category, which in my opinion is a sign that we're progressing at an acceptable rate. I've been collecting statistics for a while now and we're showing a steady improvement in many of the numbers.

However, two of our statistics have stayed the same or decreased for the most part. Our Answer ratio (roughly, the number of answers per question, but exactly how it is calculated is not entirely clear) has been around 1.5 for a long time, which is one of the lowest in Area51. StackExchange would prefer that we be getting 2.5 or more answers per question. While 1.5 technically is still "Okay", I can't find any sites with answer ratios that are significantly lower (the lowest I could find was 1.4), and this number is unlikely to improve on its own. The trend seems to be for just 1 answer per question, and at that point there's not much advantage of the StackExchange Q&A model over a wiki answers site.

Our % Answered has also been dropping a bit. In our early days we were well above 90%, but when we started seeding questions from lesser-known series this dropped to around 89-90%. We've recently dropped to 88% a couple of times. This is still far above the borderline for what's considered "Okay" but we are in the bottom 20% of all Area51 sites (most of which are in the 95-100% range). This number also doesn't show signs of improving. It may be that this site will always have a somewhat low % Answered simply because our scope includes a lot of very obscure things, but in my opinion we should at least try to improve this.

What can we do to improve these numbers, i.e. to attract more answers per question, and get people to answer our unanswered questions?

| |
10

I'll throw my two cents in here:

Don't be afraid to speculate!

Even if your answer isn't "supported by facts" (what facts?, this is anime for crying out loud!), or has a "definite citing source", don't be afraid to give it! If it's speculative by nature, but has good, intelligent arguments that are in your favor, don't be afraid to share it!

The whole point of this site is that multiple answers are posted, and the best one is voted to the top.

Can I have a second opinion?

If you see a question posted, but the answer there doesn't sit well with what you understood of the plot/situation/question, don't be afraid to give your own answer to it!

Leaving a comment is fine, but providing with a full answer to better explain your opinion is better! Who knows? Maybe yours will be voted to the top, if it's better.

Remember that not all is definite

This is anime, not math, there's more than meets the eye, and there's more than one side to any story. Many questions don't have absolute answers as told in the anime or manga, don't be afraid to give your answer, even if it has no definitive proof elsewhere (just make sure to say so!)


All in all

Have fun, this is a recreational site, we value the answerer's opinions as well as their knowledge, remember that there's no such thing as a real "anime expert".

| |
  • 3
    An addendum: Remember that for an answer to "count" in our stats, it either has to be accepted or +1'd. – Killua Mar 30 '13 at 23:21
10

Bounties

After adding the bounty to What is the purpose of characters shouting out the name of their attack? the question received 4 more answers not including my own. None of them really got to what I wanted, but at least we got more input on the topic from multiple people. If you see an interesting question that hasn't received enough love, try giving it a bounty. Even if there is already an accepted answer for the question, you could place a bounty asking for an answer with a source, an answer with more detail, etc.

| |
  • 1
    +1 excellent incentive! – iKlsR Apr 1 '13 at 11:44
8

Second Opinion!

  • Publicly share questions via our personal social accounts etc.

  • Ask small, simple general questions, the easier and more relatable (some popular Anime) the question is, the more likely someone knows the answer, this will also help boost our content

  • Work a tag you that you see is low on answers or occasionally skim through the Unanswered tab.

  • If a question is already answered, still chip in and add your view(s) or additional information to it.

  • Don't be afraid to post spoilers.. that's what this tag is for!

| |
5

I think all three of the previous answers are good advice. However, thinking that just following those good practices will solve the problem without making any other changes to how we do things around here seems too idealistic to me. All three of the existing answers are things that we should be doing, and to some degree we are doing them, but our statistics have not improved. This suggests that more drastic actions may be warranted.


I decided to look at what kinds of questions are getting answered, and how well they are getting answered. For this, I divided up the questions on our site into 3 categories. These are the following:

  1. General Questions: Questions not about any particular series, but about anime in general. ID requests for a particular series are included here also. The reasoning behind this is that answering such a question typically doesn't require expert-level knowledge of a particular series, but rather broad knowledge of a lot of series.

  2. Popular Series: Series with 10+ questions as of writing this. Specifically, these are Naruto, Bleach, One Piece, Dragon Ball, Fullmetal Alchemist, Pokemon, Fairy Tail, and Death Note. This was a convenient place to cut off, because above this point are mostly series that are popular enough that casual anime fans are likely to have watched them, while below are mostly less common series. There are some exceptions (Cowboy Bebop and Neon Genesis Evangelion come to mind) but most series with fewer than 10 questions are not as popular as those above.

  3. Other Series: Questions about any series other than those listed above.

Here are some statistics. I compiled these at roughly 2:00 UTC today, March 19th. There are 118 unanswered questions. Out of the current unanswered questions, here's how they're divided up.

General Questions: 5 unanswered (4.2% of all unanswered)

Popular Series: 7 unanswered (5.9% of all unanswered)

Other Series: 106 unanswered (89.8% of all unanswered)

Most of our unanswered questions are about particular series which tend to be less popular. Let's go one step further, and figure out how many total questions there are in each category. There are currently 919 total questions. The following statistics are my best estimates, but I may have miscounted a few questions, though this shouldn't change the numbers much.

General Questions: 133 total (14.5% of all questions, 96.2% answered and/or closed)

Popular Series: 309 total (33.6% of all questions, 97.7% answered and/or closed)

Other Series: 477 total (51.9% of all questions, 77.8% answered and/or closed)

It shouldn't be surprising that we do better on popular series and general questions than on less popular ones, but the degree to which this is true was surprising to me. It also seems to be true that the first two categories tend to have more answers per question. That is to say, with mainstream series and general questions, we're doing pretty well, but with less popular series we are having a lot of difficulty.


This suggests to me that the reason our answer stats are low is because we have a lot of questions on this site for which there aren't any experts on the series. To put it differently, the long tail on this site is particularly long, compared to what we are prepared to answer. There are thus 3 things which we can do:

  1. Get more experts: This would be the ideal solution, but I do not know of any way to accomplish it, except waiting and promoting our site. We've tried various things, but they don't seem to be working particularly quickly.
  2. Ask more questions in categories 1 and 2 and fewer in category 3: These will dilute the questions in category 3 in our statistics. However, to do so effectively, we'd need roughly 63% of our questions to be of those categories 1 and 2 to reach 90% answered, and 78% to reach 93%. Even if we got no new questions in category 3, those would require roughly 370 and 1250 questions of category 1 or 2, respectively. These do not seem like feasible numbers in the near future. Closing some particularly bad questions in category 3 will help some, but I don't think this is likely to solve the issue.
  3. Become experts in these series ourselves: To some degree, some users have already been doing this by doing extensive research on these questions, but this isn't really consistent with the Stack Exchange philosophy for Q&A.

None of these are good solutions, but they are the best options that I can think of. There are other options, which are even worse, such as restricting our scope to only popular series or providing and upvoting low quality useless answers, but at this point I think we would be actively harming the site for the sake of statistics, which is an incredibly moronic thing to do.

With that in mind, I personally will not be asking any more questions from category 3 unless I'm certain they will get answered, and I'd urge other users who are asking questions mostly for the purpose of seeding to do the same. This may hurt our questions per day statistic, but we're doing fine in that department in my opinion. We're around the median for questions per day, while we're way below the median in terms of answer stats. Also, we're near the bottom in terms of our total number of users. I suspect that our small user base is simply overwhelmed with 5+ questions per day, and by temporarily decreasing this number substantially we could improve the other stats.

Also, I am in no small part to blame for the problem, since I have 19 unanswered questions myself out of 72 total (only 73.6% answered). Some of those are questions for which I know the answer to; I'll answer them myself or give the answer to some other user. Others are probably answerable, but maybe too difficult or too niche. I'll gradually be going through these and either editing them to be easier or deleting them entirely. I may undelete them at some point in the future if we get more users with broader interests. I don't particularly want to delete these questions, since they provide lots of good things for new users with broad interests to answer, but for the moment it makes more sense to focus on more mainstream series.

As a note, I hope that having more questions about Naruto, Bleach, etc. will not be a turn-off to future experts. There is a difficulty in anime that many people who have very broad interests are not particularly interested in mainstream series. However, I think that we can simultaneously please the experts and keep our content more mainstream than it is right now as long as we focus on things which are still popular, but not extremely popular. That is to say, we should be looking through lists of popular anime (such as MAL's list) and targetting the top 100-or-so series, as well as currently airing popular seires. These are things that still have lots of fans, and consequently lots of experts, but are diverse enough to have things that interest all types of fans.

| |
  • I feel like there's no easy way to find questions about more obscure series. I bet there'd be questions of that type I could answer, but without going down a list of every anime I'm familiar with, I don't have a good way of finding them. I found out about this site too late to say "hey I think something like genre tags should be consistently used", but otherwise questions I might be interested in are drowned out by Naruto/Bleach, etc. – Mark S. Dec 23 '13 at 13:16

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .