February 22, 2010

Why are online comments so negative while online product reviews are so positive?

Anyone reading this knows that comments left at blogs, discussion forums, and online newspaper websites tend to be a lot more negative than what you hear out of people's mouths in real life. The standard answer is that it's because of anonymity -- when people don't have a reputation to protect, they incur lower costs by being evil, so they'll indulge more in their evil side online.

But that cannot be right because anonymous web-surfers who leave comments and ratings on "products" at "retail" websites are overwhelmingly positive, often effusive. (I use quotes because I count YouTube as a retailer of video clips.) There was a WSJ article last year about this. The gist is that reviews of books at Amazon, of sellers at Ebay, of restaurants at foodie sites, etc., are so skewed toward the positive side that the ratings system doesn't seem to work -- the average product gets 4.3 out of 5 stars online. Where are all of the negative comments, whether nasty drive-bys or withering diatribes, that we would have expected from the behavior of anonymous online commenters at blogs, forums, and newspaper sites?

I'll put up the answer tomorrow or the next day. It's not an immediately intuitive answer -- you would have to have read and internalized some key insights from Adam Smith's Theory of Moral Sentiments or be clever enough to re-discover his views on your own. It is a pretty simple and real answer, though. In the meantime, what possible reasons can you think of to resolve the paradox between negative comments vs. positive reviews? The people interviewed for the WSJ article threw out some ideas, but they don't really work.


  1. In self-interested competition people accent differences in comments: negative to the extreme of playing the devil’s advocate (taking a position they might not agree with for the sake of argument). Sympathy is positive and people will cooperate to enhance a liked product.

  2. People who leave comments on products have to admit that they bought it. People who bought a poor product would rather not admit that they made a bad decision, while those who made a "good decision" (i.e., they love the product) are more likely to review it.

    On blogs, since there is no such product, it may be a simple case of not liking the idea via Not Invented Here syndrome.

    Interested to see your thoughts tomorrow.

  3. Here's how it works for me:

    On the blogs I visit, I err on the side of the critical. You think you know your stuff? Okay, then let's see how well it stands up. If it's too hard for you, then be humble and realize that blogging isn't your thing. Also, the blogs I visit are usually written by intelligent people, for intelligent people. You won't stand out if you lavish a blogger with praise, while if your ideas are better than his, you're golden.

    If I'm reviewing a book on Amazon.com, it's usually one I like. Also, I find the rating system to be arbitrary for my favorite books. The Aeneid's second half isn't that great. If the second half of a movie sucked, no one would watch it again. In the case of the Aeneid, the first half is so amazing that it still warrants five stars, and there are enough notable scenes in the second half to carry it along.

    I'm not inferring universal laws from these, these are just my personal experiences.

  4. For those who want to peek at the answer, agnostic explained here when Robin Hanson raised the question.

  5. I haven't read Smith. I feel that the answer is pretty obvious - people like to "think positively". So few are inclined to admit that they bought this real POS - rather, they will note all the good things about it.

    Comments are entirely different matter. One's opinion on anything is always the right one and the most important one. That needs to be signalled. Easiest thing to do so is to criticize another.

  6. hey agnostic, I'm writing a paper for linguistics, and I'd like to trace the use of politically correct terminology using your methodology of examining frequency of NYT usage over the years. I can get to a Lexis-Nexis search of the NYT for all times, but how do you create that graph?


You MUST enter a nickname with the "Name/URL" option if you're not signed in. We can't follow who is saying what if everyone is "Anonymous."