SciFi and Fantasy Book Club discussion
Members' Chat
>
An analysis: my ratings vs. Goodreads ratings

So instead of:
5 = it was amazing
4 = I really liked it
3 = I liked it
2 = it was ok
1 = I didn't like it
We get something like:
5 = amazing
4 = good
3 = ok
2 = bad
1 = terrible "
Regardless of what scale you use, you’re rating with your gut. Whether or not you like a book is a very subjective thing. The second rating scale you listed, which is the one you associate with “rating the book with their gut”, is the less-biased scale. It has two levels for expressing bad and two levels for expressing good.
Goodreads’ scale gives you only one rating for bad and three levels for good, making it impossible to distinguish between the truly horrible books and the merely bad books. I haven’t actually given a book a rating of 1 star yet since joining Goodreads a little over a year ago. But that doesn’t mean terrible books don’t exist, beyond the merely “bad” books that I’ve rated two stars thus far, and that I won’t want to rate them appropriately when I encounter them.
However, I can understand why the Goodreads scale seems perfectly acceptable to many people. Since most people are more likely to read books they believe they’ll like, they like books more often than they hate them. So some people may find it more valuable to be able to differentiate between three levels of good at the sacrifice of only being able to specify one level of bad. I do enjoy having more variations (of both good and bad) on the sites that allow half-star ratings. But I like having an equal number of variations for both good and bad.
[Edited to remove one of the erroneous “X wrote” tags that I forgot to erase, making it unclear who I was responding to.]

4 = good
3 = ok
2 = bad
1 = terrible "
This is how I rate books. 1 star means so bad I just couldn't finish it. 2 is really bad but I managed to complete it. I try to be ..."
Agreed, good list and how I would see rating books.
"
Honeslty, from what I've gleaned from various thread discussing ratings, I think that's how most people use the star ratings.
I think those of use who have tried to adjust our ratings to goodreads scale are actually in the minority.

James wrot..."
The other rating site I use is MAL(for anime) they use 1-10. My average rating on that site is a 7.
Basically 10 is reserved for the best of the best, stuff that I find pretty much perfect with zero flaws.
9s are for stuff that's extremely good and I really love. Probably the show of the season
9-10s in books I give 5 stars on GR.
8s are very strong show, something I will still recommend to a large number of friends.
8s and better 7s are generally a 4 on GR.
7s pretty darn enjoyable, something I like and enjoy but not great/memorable I will probably not remember this show by next year.
6s I still liked it but filled with tons of flaws
5s meh
1-4 various forms of dislike.
Just flipping though my list the differences between 8s and 7s are actually fairly stark.

1 Stars - Are books that I hated and don't really want to read again.
2 Stars - These are books that are disappointing in some way, but are not so terrible that I don't want to read them again.
3 Stars - This is my neutral ground. These are books that I have read and enjoyed. There's nothing new or exciting, but they're not disappointing either.
4 Stars - These are books that I've really enjoyed, but are not quite my favorite.
5 Stars - These are books that I loved reading and ended up being my favorites.

That's why books where the writing isn't atrocious but the plot stinks still get ripped apart by people who do book snarks and why people give high ratings to books with stilted prose just because they liked the plot and/or characters.
Unfortunately you often also get 'atrocious plot' books with high ratings. I'm puzzled as to why with some books. (Especially those where nothing ever happens. How, as a writer, can you even be satisfied enough with your story to publish it if nothing really happens? Ugh.)
Just another example, I could never give a book a high rating if the narration was misogynic. Not sorry. Even if the plot was any good. That's personal taste there.
At the end of the day the GR rating system calculates the average of all ratings so does it really matter?
Yes, to people who follow your reviews. I do follow some people just because I'm interested in how they rate books. Even if they sometimes don't read the same books I would want to read.

I also noted the higher ratings range. I reflected on this a while back. It really depends on the sample, how many people have rated the book. If it's a lower number, say in the low hundreds, then surely more of those people will be fans (of course, many variables to consider, etc) and will tend to rate higher as they already prefer the author. Not necessarily as a rule, but I'm going to generalise there. If a book has a higher sample, say in the thousands, tens of thousands, that will be a more general rating range and be more true to the actual rating of the book.
James wrote: "Yes. I check reviews, when looking for books. I look for why they don't like it."
Me too. I definitely read the comments of the naysayers to find out where the book went wrong. I find them truer and more helpful.

1 Stars - Are books that I hated and don't really want to read again.
2 Stars - These are books that are disappointing in some way, but are not so terrible that I don't want to read them again.
3 Stars - This is my neutral ground. These are books that I have read and enjoyed. There's nothing new or exciting, but they're not disappointing either.
4 Stars - These are books that I've really enjoyed, but are not quite my favorite.
5 Stars - These are books that I loved reading and ended up being my favorites. "
This sums up how I use it, too.

I use all the ratings though I rarely write reviews here. And I can give four stars to books I lambast, and one star to classic well-written books, so I have an average rating of 2.60.
I don't read that much in a good average, I think it may be a book I won't rant at if it hasn't got lot of one or two stars on the first page. I also read the ratings of some people that I know can be harsh reviewers.

In other words, reviews are all that matter. Ratings are just random Star Pictures.

Even if we all used the same scale, the ratings would still be “utterly meaningless” if you’re looking for consistency. Everybody has their own criteria, whether conscious or unconscious, that determines whether or not they like a book. Star ratings have always been meaningless without an actual review to back them up; this is nothing new.
Some people are more sensitive to some types of flaws while being oblivious to other types of flaws. One person might give a book 5 stars because they liked the characters and the plot while another person might give the same book 1 star because they found so many grammatical and spelling errors that they couldn’t focus on the story. Some people care more about characters, some people care more about plot, some people care more about world-building.
Background and personal beliefs affect things too. One person might give a book a low rating because their science background helped them recognize a major world-building flaw, while another person might be oblivious to that flaw and give the book a high rating. I’ve seen religious people give low ratings to books that I really liked because something in the book violated their beliefs.
Everybody has their own criteria for what they want in a book. To understand the reasons behind the ratings, you have to read the actual reasons if they’re provided. Blindly going by star ratings has never been a good practice -- not on shopping sites, not on Netflix, not on Goodreads.

Actually, no.
Individual ratings would be unclear, true. But the aggregate would, at least, be talking about similar things. If they give 2 Stars, they are saying that it was (in their own opinion), at least "okay".
That is not "utterly meaningless".
But if some people use 2 stars to mean "okay" and others to mean "bad", then the results are "utterly meaningless".
Yes, everyone has their own criteria, but if a majority of people think a book is "bad", then it's probably not going to blow my socks off. There is no way to determine that, however, with the current reality.

How do you know the aggregate is “talking about similar things” based on an average a star rating?
Let’s say 50 people have rated a book, all using the GR-specified rating scale, and the average is 2.28 stars, broken out as follows:
* 20 people
** 6 people
*** 14 people
**** 10 people
You might pass this book up as soon as you see the 2.28 average star rating, because who wants to read a book that’s just “ok” when you could read an “amazing” book? But maybe there was a sex scene in the book and the book was accidentally selected by a religious book club, so 10 of those 1 star ratings are from people in the book club who were offended by that. Maybe 5 of those 1 star ratings are from people who were horrified that the author had a character’s sword easily slice through a silk curtain. So now you have 15 one-star ratings for reasons that you might not remotely care about, whereas the people who liked it might have liked it for reasons more in line with your interests. I am of course simplifying by referencing a single incident as the reason for a rating. Hopefully most people who rate a book with 1 star will have more than one reason for doing so, although anybody who reads reviews knows that this isn't always the case. But the gist of the idea is that people are giving the book a bad rating for reasons that wouldn't matter to you.
Even if our above example was 2.28 stars with most of the ratings being a consistent 2 stars, you have no way of knowing without context that those ratings were based on the same reasons. Even if they were, the rating would only have meaning if you care about the same things the majority cares about. It also completely ignores the other factors that can influence a rating. Some people may have been influenced by others and/or may have taken into account what their friends would think when choosing their rating. You also have the vindictive sorts of people who hate an author and go around giving everything they write 1 star, as well as the cheerleader squads who give everything an author writes 5 stars. These sorts of things are easier (usually) to detect if those people write a review.
I’ve never gone by star ratings on any site. I’ve bought things with low average reviews because the reasons behind the low reviews were things I didn’t care about, and I’ve been happy with my purchases. I’ve decided against buying highly rated products because some reviews indicated flaws that were unacceptable to me. A subjective opinion is worthless to me without a reason behind it.

Most of that missed my point. I'll repeat:
Individual ratings would be unclear, true. But the aggregate would, at least, be talking about similar things. If they give 2 Stars, they are saying that it was (in their own opinion), at least "okay".
Their opinion may be based on characters, plot, syntax, locale or any number of different things. The point is that if they are all using the same guidelines, then any book they consider "okay" will get a 2. And books they think are bad will always get a 1. That's how I know the aggregate will be of people talking about the same thing.
This is true by simple definition. They are trying to give similar, partly-or-wholly subjective, evaluations.
I also said:
Yes, everyone has their own criteria, but if a majority of people think a book is "bad", then it's probably not going to blow my socks off. There is no way to determine that, however, with the current reality.
In other words, all I'd expect is for the rating to suggest things that suck.
I made very clear that I know the criteria will be all over the place. And that I was talking about the ratings, not the criteria.

I'm with you here. I admit it raises my hackles a bit when people do the whole, "I only give good ratings because I only read books I like."
It's not like I don't try to pick books I think I'll like, or that I actively seek out bad books or anything. And I would love to be able to rave about most of the books I read, like I see some people doing.
But I also do try to be experimental sometimes, and I also read a lot of MG and YA books, and I think I'm more likely to give them a shot because they're usually quick so even if I hate it it only takes a couple of days to get through.
I do tend to be a bit more rigorous when I'm considering longer books/series which would be more of an investment.

I understand that you’re talking about the ratings. Your argument is that, when people use the same rating scale, it gives the rating meaning and that rating has value. My argument is that it doesn’t. We have a difference of opinion and I was attempting to explain my perspective.
If two people rate a book 2 stars, using the same scale, then you consider that “the same thing”, even if they rate it 2 stars for completely different reasons, yes? As you stated in the paragraph I quoted, if people’s opinions are based on “any number of different things”, and they’re using consistent guidelines, then the “aggregate will be of people talking about the same thing”. That is where we disagree, and that is the point I have been trying to make.
I only consider it “the same thing” if they rate the book 2 stars for the same reasons. And that rating is still only relevant to me if the reason(s) are relevant to me. A number without context doesn’t tell me anything, and a rating scale does not count as “context”. Two bags of potato chips are not the same, even if they were measured on the same scale and weigh the same amount, if one bag contains regular potato chips and the other contains barbecue potato chips.
If an average star rating, using a consistent scale, is sufficient for you to make a decision about a book, then obviously you’re correct that Goodreads ratings will not help you whereas another site’s ratings with a consistently-used scale would be of value to you. Not only do people use different scales on Goodreads, but I’ve read posts indicating that some people even use the stars for different purposes altogether, such as to indicate their interest in reading a book they’ve never read. You also don’t have to look far to find unpublished books that have lots of ratings – ratings that, while easily ignored now, will be confusing once the book is published since the average star rating will be based on reviews that don’t reflect the actual content of the book.

When I say "talking about the same thing" I mean, specifically, that they both agree that a score of 1 means they did not like it and a score of 2 means it wasn't amazing, but they didn't dislike it.
That's what I meant and tried to explain, again. That's all. Nothing about the reasoning or prejudices or expectations or anything else.
They agree that 1 is bad and 2 is not bad.
Other people have already listed their own criteria, that includes 2 being bad, as well. That's the variation that causes the numbers to lose all meaning.

Our conversation began when you said the ratings are meaningless because people aren’t using the same scale. My response was to argue that the ratings would be meaningless anyway, even on the same scale, because of their subjective nature.
You then disagreed that their subjective nature made them meaningless, because at least they provide a consistent opinion about the book (i.e., “I didn’t like it”). It’s meaningful for you to know if the average person did or didn’t like a book. I then disagreed that this was meaningful because there are too many different reasons behind an average rating to take the number at face value -- it doesn’t tell me whether or not I’ll like the book because I might not care about the same things.
At that point, you took my disagreement as misunderstanding and started repeating yourself. And then I repeated myself. And I think that’s where we are – you find value in knowing the aggregate opinion, I don’t. So we disagree and that’s fine. I’ve expressed my opinion, you’ve expressed yours, there’s nothing more we can do.

If I can't know that (because people might use 2 or even 3 to also mean bad), then there is no valid data.
The rest is flummery.

i so fully agree, i have been reading this thread, and as new member to the whole site all i can say is I read people's comments to chose a book rather than the rating system eg i found the handmaiden's tale just plain yuk and so many others loved it. Comments are the only real way to chose a book because every one has their own value system.
I have a large family of book readers and only one nephew and i agree on the books we read. We all still discuss the books but realise that we will agree to disagree
This discussion leaves me wondering what proportion of readers promote the interest/enjoyment of a book's plot over the quality of the writing itself. I have read books that were enjoyable/entertaining despite having what some would qualify as poor grammar or syntax, while other books by masters of the English language bored me to tears. So, how many readers rate a book solely on how interesting/enjoyable/entertaining it was? Does 'bad' mean for some 'boring', while 'amazing' would describe the degree to which the reader enjoyed the book, despite any potential flaws in the written English? If it is so, it could explain why there are such rating disparities between readers of the same book.

I rate mine variably, I suppose. The story is always the important thing. A beautifully written book with no story is a waste. A fabulous story with so many writing errors that it's unreadable is also a waste.
I can forgive the odd writing issue if the story's already sucked me in, but I really struggle with 'beautiful writing' and no story.
By odd writing issue, I mean occasional spelling or tense errors or the odd piece of clunky dialogue, not constant problems that drag me out of the story.
If I like the story I'll keep reading. If I don't, I struggle to finish. My ratings probably reflect this - I'm much more likely to give a good story with a few writing issues three or four stars and a piece of perfectly written twaddle two stars.
If it's fabulously written, and has a great story, it's a five star read for me.

Leonie already basically covered my answer, but: I take both into consideration, but my #1 consideration is how much I enjoyed it. And it's #1 by a long shot, so much so that a meh book as far as technical writing goes can get a 5 star from me. Since I read simply to enjoy/be entertained, that has to be what I rate on. I will say though, when I don't enjoy a story but it's clear that the writer is good, I'll be a little more generous in my rating and also tell people who I think might like the story about it.

Anyway, I don't think I can recognize really good writing (not the one that is just correct, but is said to be more). I disliked Lolita by example and couldn't find anything special in the sentences (though it may be that because I am not fluent enough in English, I fail to enjoy subtleties). And there's only once when I found a book wonderfully well-written in my native tongue (though the story wasn't exceptional).
The thing that can annoy me much is the believability. I'm more likely to find that a book is bad if there are many things that shouldn't happen in the book, like a woman recognizing that a model walking on a catwalk is her friend and wanting to commit a kind of suicide because he didn't want to acknowledge her during this time. And it wasn't a comedy.

I've never rated (or suggested to someone) a book highly simply because it was properly formatted, with just-so syntax. Although I can't see giving high ratings to poor grammar. Poor grammar slows you down, makes you re-read, leaves you occasionally confused... not the kind of thing that deserves a high rating.
Whether you like character, plot, description, action, subtext... all are subverted by poor grammar, sloppy syntax, and general lack of professional standards. It can not be avoided.
Which isn't to say that they can't have their charms... but I can't remember a poorly structured novel that I found impressive.
I rate Harlan Ellison highly, but not specifically because of his exceptional use of language, but because of his wonderful imagination.
Oscar Wilde is one of my very top authors. And he has a way with language that few have ever come close to.
Virtuosity, on its own, is pointless, dull, lifeless. It requires a subject, to act on. If the amazing writer is accurately and informatively describing the life of a nematode... I'm probably not going to enjoy it.

But with a competent orchestra, a conductor can make a thing of beauty.

Mine is 100% based off enjoy-ability. Bad writing, however does change my overall ranking because well a book with bad pacing, confusing writing, insert other thing here is just not as enjoyable to read. I personally have a very very high tolerance level for poor grammar, as in, I understand it and it doesn't break the flow until it's truly bad. So this means that many books I read and will not have a problem with the writing, or it might get a slight downgrade for, might make something neigh unreadable for someone more sensitive.

Writing quality, in terms of grammar, spelling, and general readability, is pretty important to me. Errors of that sort pull me out of a story and make me start rewriting passages in my head. On my Kindle, there’s an option to report errors if I’m reading Amazon-provided content. I often use that, unless there are really a lot of errors. When there are really a lot of errors, then I reason to myself that reporting them will help the author’s book look better to future readers, potentially increasing their ratings and encouraging them to write more books with horrible grammar. I wouldn't want that! :) In any case, when I do take the time to report an error, it further hinders my ability to get back into the story after finding the error.
Anything that pulls me out of the story will affect my enjoyment of the book. It's hard to appreciate a good plot if I keep getting distracted from it. If there are a few errors but I really enjoy the story in spite of them, then those errors won’t necessarily cause me to reduce my rating. I might, however, mention finding them when I write my review if I think there are enough to be troublesome. I can’t quantify exactly what amount of errors I consider “troublesome”, though. There’s just a point where I see yet another error and suddenly I’m exasperated and wondering to myself, “Does this author have any idea how to write? Do they know what an editor is?” I think writing errors tend to be more annoying if I already don’t care for the book for other reasons. Annoyance piles upon annoyance, and something that might have otherwise been a small annoyance will seem more significant.

In fact, most 'reviews' here, including almost all of mine, aren't best described by that name. They're more like reactions, commentaries....
However! That's a good thing! I can't tell you how many books I've read because a professor, or the NYT, or someone like that said I should, that then disappointed and/or annoyed me. Otoh, I've read hundreds of wonderful books based on GR readers' commentaries. This community addresses what a potential reader wants to know about a book.

My question is the Good Reads ratings a conglomerate of us or is there a smaller group that actually rates the books? Because if it is a conglomerate of us then for every book that is rated higher than you rate, there is someone else that rated it higher than the average Good Reads rating.


5 = it was amazing! recommended
4 = I really liked it
3 = I liked it
2 = it was ok
1 = I didn't like it / couldn't finish it
If I like a book enough to start reading it, it begins with a "3", and moves up & down from there. Not surprisingly, the rating I've used the most is "3".
Sometimes I feel bad about giving a book a 2 or a 1. The book could be competently written but deals with subjects that I don't want to read about or the author's sympathies are with the wrong characters in the novel. Sometimes I just cannot suspend disbelief enough to enjoy a particular fantasy novel (see: The Sum of All Men). And, I don't like stories where they kill the baby. Those stories lose a star in my system.
Alternatively, time travel novels often gain a star if they are at all competent because, well, I love time travel tales!

I also find a few reviewers who like the books I liked, and didn't like the books I didn't like, and I mark them "follow reviews". I pay more attention to those reviewers than I do the crowd.

But it's true that here and there I see reviews from people that just turned out to read give opinion of a book only few have heard about.

I also disagree that ratings in the aggregate are meaningless, regardless of whether or not the people rating them are using the same scale. Generally, closer to 5 is better than closer to 1 regardless of your criteria. I don't trust books rated 5 because they are usually self-published authors and their friends or people they've given free copies to who feel obligated to rate the book highly. The aggregate ratings are a beginning point for me though, not an end point. If something is rated below about 3.5, I'm hesitant. Above that, I'll read a sampling of opinions from the 2-4 ratings.
Personally, I like half-star ratings like LibraryThing has. Sometimes I'm torn between a 3 and a 4 and have to make a call that I'm not entirely comfortable with. A half-star would fill that need for me. Ironically, a 10 point rating scale would leave me with the same dilemma because even though a 7 out of 10 is the functional equivalent of 3.5 out of 5, I'd still like that wiggle room of a little more than 7 but not quite 8.

- ratings tend to come from people who have finished the book, and it's rare to finish a book unless you liked it to some extent
- the above is even truer of series, which would typically "shed" dissatisfied readers by the second tome, leaving the sequels with more favorable ratings.

But I also have more one star books finished than DNF. Because I usually hope that the book will improve later, or that I will appreciate it when the setting is done. There's also the case of books so bad it's quite interesting to read them or just to wonder how they can be so popular.


And then there are books where I would've given 2-stars to if I'd bailed when I wanted, and ended up downgrading it to a 1-star by the time I was done.
In my experience, limited as it may be, the few books I pushed my way through that I was glad I finished are in the minority to the ones I wish I'd bailed on when I first had a mind to.


But I do explain, in the review space, why I bailed. I like to help other potential readers.

I've done the same. I haven't scored (yet) anything that I stopped reading. As I got older, I've stopped reading more and more. Life's too short to keep reading something that I do not enjoy. (fiction-wise... with non-fiction, enjoyment isn't the only goal)
But I'm new here and still trying to remember books that I have read, to add ratings to.

But I rarely bail so early. (I'm not even sure I'd consider that bailing as much as a failure to launch.)
I'm rather loathe to bail on a book once I start, so it takes a lot for me to bail. The last book I bailed on I'd read 57% - though I probably only hung in that long 'cause a friend picked it for a buddy read.
The one before that was 48% and the one before that 42%.
As far as I'm concerned, my shots are rather more than fair... ;)

I've stopped reading books halfway through and I've read a page and decided it wouldn't be worth it.
Poor plotting may take half a book to determine, but bad writing stands out on the first page.
(I've given books more time then they deserved, because I really liked the concept)


If I haven't found the "essence of its storyline", by a quarter of the way through... it's badly written and I expect that either I never will find it or it isn't worth finding.

I used to be this way, and I still am to a large degree, but I've found that a good test is to let it sit for awhile... put it down with the intention of coming back to it... and, often, I'll find that I don't even remember or care enough about what I had read to bother continuing.
Also, sometimes you can find some good plot summaries on Wikipedia, if you just want to know for knowledge and/or completionist sake.
Books mentioned in this topic
The Runelords (other topics)To Kill a Mockingbird (other topics)
Tigana (other topics)
Thomas the Rhymer (other topics)
Authors mentioned in this topic
Harlan Ellison (other topics)Robert Jordan (other topics)
Robert Jordan (other topics)
Brandon Sanderson (other topics)
Terry Goodkind (other topics)
4 = good
3 = ok
2 = bad
1 = terrible "
This is how I rate books. 1 star means so bad I just couldn't finish it. 2 is really bad but I managed to complete it. I try to be ..."
Agreed, good list and how I would see rating books.