This hasn't been a huge problem, but now that I've seen it a few times, I wanted to get this out there as more of cautionary tale than a dire warning.

With a subject as popular as pets, there is always a danger that this Q&A becomes more "support group" and story telling than actually providing factual, reliable information.

I'm not suggesting a wide-spread problem, but please consider carefully these widely accepted answers:

I don't mean to pick on these posts specifically (I don't suspect the information is wrong), but there's a line to cross from offering trouble-shooting advice (something to try based on your experience) versus cherry-picking an instance of personal experience to answer "Is it okay to do {x}?" — all while the community embraces hearsay as support for a broader claim.

I'm not suggesting that every answer be backed by university studies or quantitative scientific measures, but you should be wary about encouraging and accepting someone's personal experiences as canon for this site. At the very least, consider carefully if

  • (a) Are you up-voting these answers as a form of peer review (having shared those experience yourself); or
  • (b) if you're just up-voting the entertaining story… or what you merely believe to be correct.

Answer-by-anecdote can become a real problem on this site if you don't watch for it. It's one of the old forum problems that Stack Exchange is designed to avoid. So consider carefully if these answers are really something you want to embrace as canon. And in the most egregious examples, you may want to consider leaving a thoughtful comment letting the folks who share these stories understand how this site works:

Thank you sharing that story with us, but we're hoping for a bit more than in interesting anecdote. These type of stories don't always apply to the broader case, and it doesn't really answer the question in an objective way. Do you have anything more to back this up?

Again, this is just something to think about as you consider your answers and how to chose to vote. I don't want this site to become the pet-equivalent of my grandfather smoked, like, 30 cigarettes a day and lived until 97.


  • As someone new to SE (I came here FOR THE PETS BETA), seeing this in meta as well as meta.pets.stackexchange.com/questions/400/… is pretty frustrating. At this point I don't know what kind of answer is "best" here.
    – Zaralynda
    Nov 6, 2013 at 21:52
  • @Zaralynda Please see : What kind of answers are we looking for?
    – user34
    Nov 6, 2013 at 22:03
  • 1
    Score! I didn't vote (up OR down) on any of the answers of the cited questions, but regarding said questions, I wonder if "Is it harmful for my cat to eat bugs" needs to be regional and/or be more specific listing the bugs in the area.
    – JoshDM
    Nov 6, 2013 at 22:08
  • Of note, (I believe it was) @MonicaCellio hit me up early on the site when I gave an anecdotal answer to a question, so I edited in justification and cited my source. That just so happened to be my answer to "Why do Dogs Eat Grass", which was cited on the podcast. :-D
    – JoshDM
    Nov 6, 2013 at 22:10
  • 1
    @Zaralynda Robert and tim Post are community managers, always be guided by them. I have enjoyed your contributions. have a browse at Stack Overflow, Physics, Psychology & Neuroscience to see what other sites are doing. The idea is to improve the knowledge base already available on the internet, either by more searchable content and/or quality
    – user87
    Nov 7, 2013 at 4:40
  • 1
    @JoshDM - In the cats eating bugs answer I did point out that checking the local venomous critters and watching out for them was a good idea.
    – Kate Paulk
    Nov 8, 2013 at 18:58
  • @KatePaulk - that is the correct way to do it from an answer point of view, but it should be an aspect of the question and details/data should be provided by the question poster to further localize the question.
    – JoshDM
    Nov 8, 2013 at 19:03
  • @JoshDM - that was what I'd thought, but apparently it's not enough for some, including the OP here. I updated my answer with some of the evidence I picked up from a quick search and a little more of my own experience (which includes a brother-in-law who's a vet and would have stopped us doing anything likely to harm the cats)
    – Kate Paulk
    Nov 8, 2013 at 19:17

4 Answers 4


On some Stack Exchange sites, questions are pretty matter-of-fact and answers can be tested or otherwise subjected to rigor. But other sites, like The Workplace, Writers, UX, and this one (just to pick a few that I participate on), cover topics that are more subjective. Good Subjective, Bad Subjective only gets us so far, since it focuses on questions rather than answers.

We don't want to say "no experience-based answers", but we also don't want answers that just say "works for me" without any support. Maybe it works for you because it works for everybody, or maybe it works for you because of some special situation that we can't see.

Some sites have a back-it-up principle, and I think it's useful to think along those lines for Pets. Yes, share your anecdote as part of a supported answer, but tell us how you know it's ok. Maybe you had a conversation with your vet, or read a journal article saying this, or have seen this problem come up regularly in your ten years of involvement in gerbil rescue -- whatever it is, show us why your answer has some credibility to it. Remember, we're all pretty much anonymous people on the internet, so it's better to help us see what you see than to just hope people take people's word for it.

On the other hand, anecdotes that are short enough work great in comments.

  • 3
    A bit of caution when offering anecdotes in comments, being a comment doesn't make it less of an answer. Early on this was happening to a very large extent and it tends to give new users an incorrect 'discussion-ish' feel to the site. If it's just something to add, that's fine - just make sure it couldn't be easily posted as an answer too (however anecdotal).
    – user105
    Nov 7, 2013 at 4:57

I've had the same observations that Robert did, and I agree with his assessment that it's not yet a major problem, but it is a problem that we face. We're not in danger of falling apart at the seams due to unfounded anecdotal drivel causing people to try hoodoo on cats.

The most problematic anecdotal answers that I've seen have been third person, or third party answers, similar to the following:

My sister's cat had this problem, and her husband took care of it by (answer here)

This can be bad, for the following reasons:

  • You're answering based off of a possibly vague recollection of a problem that someone else had - are you certain you remember everything correctly?
  • Did your sister tell you everything? Did she per chance leave out something that would be very important to include in an answer?
  • Since you didn't really experience this yourself, you're leaving mostly a sympathy note; if you're engaged to expand your answer, your first inclination is to say "I'll ask my sister and get back to you."

Really, it's better to just send your sister a link to the question and ask her if she's interested in providing an answer. These 'answers' originally appeared mostly in the form of comments, which we quickly removed during the private beta while guiding folks to write answers that could actually be vetted.

Answers based on your original experience are much stronger and can stand on their own. Consider instead:

My cat (a Tabby) had this problem, and this is what I did to solve it. While you've got a Calico, I have no reason to believe you should do anything differently for him to warm up to a different brand of food. Here are the steps I took, and the results you should be looking for when you reproduce them.

That is a strong answer on its own, and can become even stronger when someone leaves a comment to an authoritative source that validates or suggests changes to the answer, which can later be included within the answer. While limited, you've done some science, which can yield an answer that can be vetted. I believe the majority of answers that this site receives are going to be based on personal experience and we wouldn't dream of excluding them.

What we want is a happy medium. We want writing that is entertaining and easy to read, which means a little bit of story telling. But, we also want this tempered with authoritative references, just not to the degree that we're simply duplicating reference material.

For the most part, I think folks have found the metaphorical happy place - this is just something to keep in mind.

  • 1
    Oh, great point about third-party answers. If you did this and it worked, then you are here to answer followup questions and provide more details (and we want to be able to benefit from your experience). But your sister's husband isn't so accessible. Nov 7, 2013 at 16:33
  • There is a "ridiculous" aspect to this answer, core to the anonymity of the internet. There is absolutely nothing stopping someone from changing "My sister's cat had this problem, and her husband took care of it by (answer here)" to "My cat had this problem, and I took care of it by (answer here)" to generate false credibility and perceived strength.
    – JoshDM
    Nov 8, 2013 at 19:07
  • 1
    @JoshDM My cat's husband wrote this, so ...
    – user105
    Nov 9, 2013 at 2:21

This is a general issue in fishkeeping. You can very easily find hard data like "what's the average pH of the waters this species is native to", but if you need to know "will this species do well in my tank with a pH of X", you usually have to rely on the collective wisdom of other fishkeepers. That's often enough (or at least close enough), but it can be hard to tell if two people say the same thing because they've both independently experienced it, or if they just heard the same baseless myth when they were ten. And people misinterpret what they have experienced all the time.

Personally, I'm comfortable giving an answer when it's one of three things:

  • Ideally, where I have some direct experience and can at least share what patterns I've seen.
  • Information I've read in the literature or learned from people I trust, as long as it seems reasonable based on my own knowledge.
  • The "common knowledge" about a popular animal that I may not personally have kept, where the general husbandry is pretty well figured out.

And to my mind, an anecdote is only really trustworthy as a one-off example of the first kind of answer, or to clarify a case where an answer that's otherwise good might not actually work.

  • I really like your comments about the fish. I think that explains it well. There's a lot of info that can be googled, but whether it is good info or from a reliable source is another thing. I like your answers on beta
    – user87
    Nov 7, 2013 at 16:12
  • I do think we have to be careful of what is considered common knowledge however. I'm sure you know what I mean
    – user87
    Nov 7, 2013 at 16:13
  • re "common knowledge" -- Most definitely. I'm using it here in the sense of the best-practices that aquarists have learned about something, which is a very different thing than what "everybody knows". I think distinguishing between the two is going to be a big part of the value of Pets.
    – toxotes
    Nov 7, 2013 at 17:57

Do we want to allow answers based upon personal experience and anecdotes?

I don't see how we could not allow them.

One of the cornerstone philosophies of the stackexchange platform (as I understand it) is to build a community of experts.

What, exactly, should we be experts on?

If we say answers cannot be based entirely upon personal experiences and knowledge, then we're basically saying we should be a community of experts at using google to find references.

While answers based upon being good at google certainly have a place here, they shouldn't be the only content we allow.

That being said... there are limits to what sort of anecdotal answers we should encourage.

Personal experience can only go so far. So long as the answer clearly defines the limits of that personal experience, its okay, but if an answer asserts that "this is what happened to me, so this is what everyone can expect", then I'd strongly discourage upvoting.

Using my answer from the ferret question (obviously I'm a bit biased on this one), I think it is perfectly acceptable to say "ferrets can play well with other pets", so long as the answer is qualified as mine is, that not all ferrets behave the same way.

If the answer says "I've had success with this", it is an acceptable answer only if it also highlights any risks, or explains the possibility of people having divergent experiences.

To me, that's a far cry from "my grandfather smoked 30 packs of cigarettes a day and was fine, so therefore cigarettes are safe."

  • 3
    Hi Beofett, your example doesn't cross the line. Objectivity in answers, or showing both sides of a situation and how to tell when one scenario is applicable vs another scenario, is a better answer than something merely one-sided with a single, one-off example to support a claim: "Dude! I had this cat this one time that ate a bug and totally died" You provided a lot of information to back up your claims, and you presented multiple situations. It's clear from reading your post you know what you're talking about, even though it was only your own experiences. Hope this helps!
    – jmort253
    Nov 7, 2013 at 5:14

You must log in to answer this question.

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