Amazon Killed The Book Reviewer Star

[fjtorres]

Originally Posted by HarryT

That depends on the specific situation. I would submit to you that it would perhaps be unwise to have papers submitted to scientific journals peer-reviewed by a crowd of random strangers, for example.

Agreed...
...90%...

Unfortunately, the peer review process has had its share of embarassments over the years and the trend, alas, is towards more, not less malfeasance.

http://scienceblogs.com/insolence/20...rticle_ret.php

But then, we are generally talking apricots and oranges; scientific papers are reviewed as to accuracy whereas literary works are reviewed for compliance to accepted form.

[Fbone]

Originally Posted by kennyc

I'm not looking for examples, you claimed

"Many titles have almost all (>90%) fake reviews. "

I don't believe it unless you have facts to support that statement.

And first define what "Many" means.

Many as in more than a few. The 90% refers to the percentage of reviews per indie title. Not the number of samples. Sorry if that was confusing.

And by fake I mean not written by the author, friends, family, for hire or profit.

As for proof, each person will have their own experiences and reach their own conclusions. For me (your quote of mine) is how I see it. Today there are 497 free titles added to Amazon and I looked at over 200 of them most with average of 4-5 stars. I don't look at erotica or spiritual/Christian titles. I do this daily hoping to find the next Hugh Howey or Shayne Parkinson. I guess I'm a sucker for titles that receive 34 out of 36 5-stars but am almost always disappointed when I see all the "Harriet Klausners" and similar reviewing. I read samples first if there are no or few reviews.

As for actual studies, the one I remember was that somewhere around 85% of Amazon top reviewers/respondents were compensated and 88% gave positive reviews. I think this is it.
http://paidcontent.org/2011/06/24/41...azons-reviews/

I wish the system was better. People are buying books and short stories based on reviews and become disappointed with their purchase. This does not help the indie writer.

[kennyc]

Originally Posted by Fbone

Many as in more than a few. The 90% refers to the percentage of reviews per indie title. Not the number of samples. Sorry if that was confusing.

And by fake I mean not written by the author, friends, family, for hire or profit.

As for proof, each person will have their own experiences and reach their own conclusions. For me (your quote of mine) is how I see it. Today there are 497 free titles added to Amazon and I looked at over 200 of them most with average of 4-5 stars. I don't look at erotica or spiritual/Christian titles. I do this daily hoping to find the next Hugh Howey or Shayne Parkinson. I guess I'm a sucker for titles that receive 34 out of 36 5-stars but am almost always disappointed when I see all the "Harriet Klausners" and similar reviewing. I read samples first if there are no or few reviews.

As for actual studies, the one I remember was that somewhere around 85% of Amazon top reviewers/respondents were compensated and 88% gave positive reviews. I think this is it.
http://paidcontent.org/2011/06/24/41...azons-reviews/

I wish the system was better. People are buying books and short stories based on reviews and become disappointed with their purchase. This does not help the indie writer.

I certainly believe it's your opinion, I just don't believe it a fact and I believe that by stating it the way you did you are attempting to make it more than it is.

More than a few doesn't mean anything btw. If you could actually provide evidence that 60, 70, 80% of the reviews are fake then you might have something to contribute but just throwing around opinions and possibilities does more harm than good.

The validity of the O.P. study specifically indicated that it did not apply to reviews with one or very few reviews. The point is that the masses are basically correct (how could they not be, that's the point) moreso than the "professional" reviewers who tend to be biased. This was the point of Sir Francis Galton's work as explained in the article.

Are there fake/paid reviews on Amazon? yes.
Are there spam campaigns of one-star reviews? yes.

Are they rampant? No
Do they make Amazon reviews unreliable? No.

Do we all want the system to be better? Yes.

I'd still rather rely on reviews by a variety of readers than have to rely strictly on the NYT Books Review. I'd prefer hearing from readers like myself and then in conjunction with the blurbs and sample(s) make up my own mind whether to read/purchase it.

[Steven Lyle Jordan]

It's one thing to say it; it's another thing to get the public to accept it, which is what really counts.

[kennyc]

Originally Posted by Steven Lyle Jordan

It's one thing to say it; it's another thing to get the public to accept it, which is what really counts.

Not sure what you mean by that Steve? You mean as in sales vs. a good review?
Or do you mean the public accepting the word of the reviewer?

[ProfCrash]

Originally Posted by HarryT

That depends on the specific situation. I would submit to you that it would perhaps be unwise to have papers submitted to scientific journals peer-reviewed by a crowd of random strangers, for example.

Based on some of the reviews that I received, I am not sure that is the case. "This paper is excellent and the conclusions are very important. Please Revise and Resubmit after removing the game theory elements." When the papers conclusions are based on a game theoretic model.

Not very helpful.

[din155]

I tend to agree with the article especially in case of large number of reviews. There will always be fake reviews and spams but law of average will take over with large number of data points.

din

[Joykins]

Originally Posted by kennyc

As well as the spammed 1 star reviews?

I personally haven't seen those (other than complaints about kindle prices). Where may they be found?

[Joykins]

Originally Posted by kennyc

Are there fake/paid reviews on Amazon? yes.
Are there spam campaigns of one-star reviews? yes.

Are they rampant? No
Do they make Amazon reviews unreliable? No.

The fake reviews start to feel rampant and make the system feel unreliable when you are looking at a bunch of indie and self-pub books. This is not a bestseller / major publishing house problem. It's a problem for ME as I look for niche titles and get suckered into borrowing or buying dogs like this.

(That "most helpful" review? It's *mine*. I took the trouble to review it in some detail because the darn thing has *20* 5-star reviews. I wanted to explain *why* it wasn't worth the time invested in reading it).

[LuvReadin]

Originally Posted by Fbone

Here's a random example where the same person wrote multiple reviews under different names for this title. Whether intentional or not it still affects the ratings.
http://www.amazon.com/Chop-Darby-Sta...7185703&sr=1-1

Just curious - how do you know it's the same person?

[geronimo]

Originally Posted by HarryT

That depends on the specific situation. I would submit to you that it would perhaps be unwise to have papers submitted to scientific journals peer-reviewed by a crowd of random strangers, for example.

It would not be only unwise, but completely impossible -
"peer-reviewed by a crowd of random strangers": classical oxymoron ...

[Steven Lyle Jordan]

Originally Posted by kennyc

Not sure what you mean by that Steve? You mean as in sales vs. a good review?
Or do you mean the public accepting the word of the reviewer?

I mean the public accepting the validity of reviews by the unknowns and strangers you usually get online. I'm not sure they put as much credence in the worth of online reviewers as they do from famous, established reviewers, and most especially when the reviews are either 1- or 5-stars; more often the value of the online review is automatically questioned, whereas the value of such a glowing (or stinking) review from an established source is less likely to be challenged.

I mean, if no one knows me from Adam, why should they care if I give a 5-star review of Mars Girl? They'll just assume I'm in Garrity's posse, and move on.

[kennyc]

Originally Posted by Steven Lyle Jordan

I mean the public accepting the validity of reviews by the unknowns and strangers you usually get online. I'm not sure they put as much credence in the worth of online reviewers as they do from famous, established reviewers, and most especially when the reviews are either 1- or 5-stars; more often the value of the online review is automatically questioned, whereas the value of such a glowing (or stinking) review from an established source is less likely to be challenged.

I mean, if no one knows me from Adam, why should they care if I give a 5-star review of Mars Girl? They'll just assume I'm in Garrity's posse, and move on.

Thanks. That makes sense.

[Ninjalawyer]

Originally Posted by HarryT

That depends on the specific situation. I would submit to you that it would perhaps be unwise to have papers submitted to scientific journals peer-reviewed by a crowd of random strangers, for example.

Oh, I totally agree that crowdsourcing has more or less juice depending on the situation. However, I would note that science can definitely benefit from it in some circumstances. For instances, there's an article here Science News about a computer science problem that spontaneously became a crowdsourced research project. Crowdsourced peer review probably doesn't make sense in most cases because so few people will have relevant knowledge, but I think the peer review process could benefit by making the whole process transparent.

As far as Amazon reviews, the validity of the review certainly increases where there are more reviews. However, the problem of paid reviews is probably much lower in the case of a crowdsourced review system because you can't buy off everyone.

[piper28]

Originally Posted by Fbone

As for proof, each person will have their own experiences and reach their own conclusions. For me (your quote of mine) is how I see it. Today there are 497 free titles added to Amazon and I looked at over 200 of them most with average of 4-5 stars. I don't look at erotica or spiritual/Christian titles. I do this daily hoping to find the next Hugh Howey or Shayne Parkinson. I guess I'm a sucker for titles that receive 34 out of 36 5-stars but am almost always disappointed when I see all the "Harriet Klausners" and similar reviewing. I read samples first if there are no or few reviews.

I would guess without trying to lookup the original research that they're looking at things that have far more reviews than 36. I'm thinking that the number probably has to be in the couple hundred before there's any real statistical significance behind it. A lot of the indie type books probably just don't get enough reviews that this would be valid for them.