Change the rating system?

[PashaRu]

I definitely see your point. What made the previous posts controversial for me was the idea of moving to a 5-star system.

I agree. In my initial post, I said that I would prefer a 4-star rating scale with half steps, and @Ryan correctly observed that this wouldn't skew past ratings or render them inaccurate.

[gali]

Thanks everyone for your input on the subject.

The reason I started this thread is simply because I feel that the current rating system is imprecise. To me, that is the overarching reason to adjust it. Any "complications" are minor by comparison. There have been comments referring to some "difficulties" associated with changing the system, and I think the subject has become overcomplicated. If the 4-star system were retained and half-steps added, I don't see how this could present a problem. Past 4-star reviews would remain as such, the highest rating possible.

I think that rating a book is not exactly the same as recommending a book, as one would or wouldn't recommend a restaurant. (Incidentally, restaurant ratings on sites such as Yelp.com are 1-5 stars.) The recommendation is given in the review, and should include specifics as to whom the book is recommended. Personally, I find a more precise, accurate rating to be more helpful, not less. And as a reviewer, it is often difficult for me to choose a rating because, as I mentioned, the current scale seems imprecise. A 2 sometimes seems too low, but a 3 seems too high. A 3 seems too low, but a 4 is too high. I find myself "rounding up" or "rounding down," and the rating doesn't accurately reflect my opinion. If half-steps were added, it would be much easier (for me) and I would feel more comfortable with my ratings. From the initial comments on this thread from several reviewers, I suspect they feel the same way. In fact, I often see reviewers state that they would give a book X.5 stars if they could.

Here's how I see it:

1 - This book is terrible. Avoid it at all costs.
1.5 - Don't waste your time. There may be something to work with here, but it needs a lot of work.
2 - Die-hard fans of the genre might like this, but you'll have to overlook some major flaws.
2.5 - This has a few things going for it. There are a few diamonds mixed in with the coal, and you might enjoy it.
3 - This is a good book! It's well done, and you'll enjoy it.
3.5 - Great book! Almost perfect, highly recommended.
4 - Fantastic!!! This is one-in-a-thousand!! It will change your life!!!

I can't see myself using 1.5 very often, but it would be WONDERFUL to be able to rate a book 2.5 or 3.5.

As for Bookshelves: I never look at Bookshelves to see where a book is placed. Never. I rely on the review and rating. I suspect most readers of reviews are like me. My Bookshelves? The books are arranged in alphabetical order, the way the system did it. (I moved a book to the #1 position once, and it automatically showed up as "my favorite book," which I did not want. I haven't touched anything since.) So don't look at the position of books on my shelves as an indicator of my favorites. Again, I suspect many other users are like me. And if users don't manage their Bookshelves regularly (I certainly don't), this could be a highly inaccurate barometer for rating/recommending books.

As I mentioned, I see no downside to a more precise rating system. I think it would only improve the site.

You state your case well.

[rssllue]

Agreed.

[Ryan]

Thanks everyone for your input on the subject.

The reason I started this thread is simply because I feel that the current rating system is imprecise. To me, that is the overarching reason to adjust it. Any "complications" are minor by comparison. There have been comments referring to some "difficulties" associated with changing the system, and I think the subject has become overcomplicated. If the 4-star system were retained and half-steps added, I don't see how this could present a problem. Past 4-star reviews would remain as such, the highest rating possible.

I think that rating a book is not exactly the same as recommending a book, as one would or wouldn't recommend a restaurant. (Incidentally, restaurant ratings on sites such as Yelp.com are 1-5 stars.) The recommendation is given in the review, and should include specifics as to whom the book is recommended. Personally, I find a more precise, accurate rating to be more helpful, not less. And as a reviewer, it is often difficult for me to choose a rating because, as I mentioned, the current scale seems imprecise. A 2 sometimes seems too low, but a 3 seems too high. A 3 seems too low, but a 4 is too high. I find myself "rounding up" or "rounding down," and the rating doesn't accurately reflect my opinion. If half-steps were added, it would be much easier (for me) and I would feel more comfortable with my ratings. From the initial comments on this thread from several reviewers, I suspect they feel the same way. In fact, I often see reviewers state that they would give a book X.5 stars if they could.

Here's how I see it:

1 - This book is terrible. Avoid it at all costs.
1.5 - Don't waste your time. There may be something to work with here, but it needs a lot of work.
2 - Die-hard fans of the genre might like this, but you'll have to overlook some major flaws.
2.5 - This has a few things going for it. There are a few diamonds mixed in with the coal, and you might enjoy it.
3 - This is a good book! It's well done, and you'll enjoy it.
3.5 - Great book! Almost perfect, highly recommended.
4 - Fantastic!!! This is one-in-a-thousand!! It will change your life!!!

I can't see myself using 1.5 very often, but it would be WONDERFUL to be able to rate a book 2.5 or 3.5.

...

As I mentioned, I see no downside to a more precise rating system. I think it would only improve the site.

You certainly do give a great explanation of the importance of giving accurate ratings, which right now are not comfortable for us and probably others too. The previous ratings can remain as they are precisely because the half-points were not available before, whereas in future they can be and I see no reason why the benefit of scoring more precisely is creating this much equivocation. Frankly, I think @bookowlie states it quite simply: people are used to the current system, even though that system could obviously use some improvement. Certainly, it worked well before ... but now it could work even better, which is something we should all desire. The need for the change is implicit in the very fact that users are expressing the need to change previous reviews to meet the new more precise system, meaning that their scores may not have been what they desired before. This cannot, and need not, be done, but it will certainly mean that in the future we don't have to award as imprecisely. Scott sees the half-points as vague, whereas I seem them as a clarification.

[PashaRu]

You certainly do give a great explanation of the importance of giving accurate ratings, which right now are not comfortable for us and probably others too. The previous ratings can remain as they are precisely because the half-points were not available before, whereas in future they can be and I see no reason why the benefit of scoring more precisely is creating this much equivocation. Frankly, I think @bookowlie states it quite simply: people are used to the current system, even though that system could obviously use some improvement. Certainly, it worked well before ... but now it could work even better, which is something we should all desire. The need for the change is implicit in the very fact that users are expressing the need to change previous reviews to meet the new more precise system, meaning that their scores may not have been what they desired before. This cannot, and need not, be done, but it will certainly mean that in the future we don't have to award as imprecisely. Scott sees the half-points as vague, whereas I seem them as a clarification.

Well said, and I agree.

I want to clarify, or rather correct, something I said previously. I mentioned that Bookshelves (because of the reasons I gave) is a "highly inaccurate barometer" for rating/recommending books. I did not choose my words carefully. The books are sorted according to rating, so there is something to be gleaned from noticing a book's position. But for the reasons I gave, I don't think Bookshelves adds or clarifies anything, and could be misleading if a book in the "top position" is simply there because it is in alphabetical order. In other words, I don't think it reliably or consistently adds an "extra star" to the current 4-star system.

[TrishaAnn92]

Thanks everyone for your input on the subject.

The reason I started this thread is simply because I feel that the current rating system is imprecise. To me, that is the overarching reason to adjust it. Any "complications" are minor by comparison. There have been comments referring to some "difficulties" associated with changing the system, and I think the subject has become overcomplicated. If the 4-star system were retained and half-steps added, I don't see how this could present a problem. Past 4-star reviews would remain as such, the highest rating possible.

I think that rating a book is not exactly the same as recommending a book, as one would or wouldn't recommend a restaurant. (Incidentally, restaurant ratings on sites such as Yelp.com are 1-5 stars.) The recommendation is given in the review, and should include specifics as to whom the book is recommended. Personally, I find a more precise, accurate rating to be more helpful, not less. And as a reviewer, it is often difficult for me to choose a rating because, as I mentioned, the current scale seems imprecise. A 2 sometimes seems too low, but a 3 seems too high. A 3 seems too low, but a 4 is too high. I find myself "rounding up" or "rounding down," and the rating doesn't accurately reflect my opinion. If half-steps were added, it would be much easier (for me) and I would feel more comfortable with my ratings. From the initial comments on this thread from several reviewers, I suspect they feel the same way. In fact, I often see reviewers state that they would give a book X.5 stars if they could.

Here's how I see it:

1 - This book is terrible. Avoid it at all costs.
1.5 - Don't waste your time. There may be something to work with here, but it needs a lot of work.
2 - Die-hard fans of the genre might like this, but you'll have to overlook some major flaws.
2.5 - This has a few things going for it. There are a few diamonds mixed in with the coal, and you might enjoy it.
3 - This is a good book! It's well done, and you'll enjoy it.
3.5 - Great book! Almost perfect, highly recommended.
4 - Fantastic!!! This is one-in-a-thousand!! It will change your life!!!

I can't see myself using 1.5 very often, but it would be WONDERFUL to be able to rate a book 2.5 or 3.5.

As for Bookshelves: I never look at Bookshelves to see where a book is placed. Never. I rely on the review and rating. I suspect most readers of reviews are like me. My Bookshelves? The books are arranged in alphabetical order, the way the system did it. (I moved a book to the #1 position once, and it automatically showed up as "my favorite book," which I did not want. I haven't touched anything since.) So don't look at the position of books on my shelves as an indicator of my favorites. Again, I suspect many other users are like me. And if users don't manage their Bookshelves regularly (I certainly don't), this could be a highly inaccurate barometer for rating/recommending books.

As I mentioned, I see no downside to a more precise rating system. I think it would only improve the site.

I see you point, I will keep my perch on the fence for now. I can see it from both sides. And when you state your case so well I tend to agree more so with the points that you present.

[gali]

I opened a poll per your request.

[PashaRu]

I opened a poll per your request.

Thank you!

[rachel_bruhn]

Here's how I see it:

1 - This book is terrible. Avoid it at all costs.
1.5 - Don't waste your time. There may be something to work with here, but it needs a lot of work.
2 - Die-hard fans of the genre might like this, but you'll have to overlook some major flaws.
2.5 - This has a few things going for it. There are a few diamonds mixed in with the coal, and you might enjoy it.
3 - This is a good book! It's well done, and you'll enjoy it.
3.5 - Great book! Almost perfect, highly recommended.
4 - Fantastic!!! This is one-in-a-thousand!! It will change your life!!!

I can't see myself using 1.5 very often, but it would be WONDERFUL to be able to rate a book 2.5 or 3.5.

As for Bookshelves: I never look at Bookshelves to see where a book is placed. Never. I rely on the review and rating. I suspect most readers of reviews are like me. My Bookshelves? The books are arranged in alphabetical order, the way the system did it. (I moved a book to the #1 position once, and it automatically showed up as "my favorite book," which I did not want. I haven't touched anything since.) So don't look at the position of books on my shelves as an indicator of my favorites. Again, I suspect many other users are like me. And if users don't manage their Bookshelves regularly (I certainly don't), this could be a highly inaccurate barometer for rating/recommending books.

I agree with the proposed explanations you have provided for each rating. I too find myself torn on a rating and often do have to "round" up or down. The rating is not supposed to be MY rating of the book, but a (mostly) impartial rating of the book as a whole.

I also do not look at someone's bookshelves to see how they rated a book. I have organized my books on my bookshelves, but even there I hate being limited to whole number ratings. I also find that my "currently reading" is something that I actually finished a while ago, so no, I don't update my bookshelves regularly. I was also under the impression that our ratings here can be used on Amazon and other sites, so people seeing the rating may not even know to access "bookshelves" to see how each person rates a book.