As I was going through my Goodreads account, looking through the ratings and reviews of books I’d read, I realised that almost all of them were highly rated!
I couldn’t help but think what a poor reviewer I was, contributing to a ratings inflation. Did I have an inherent bias of giving high scores to every book I read?
I sought to find out.
Though I’m sure my reviews are biased up because of various reasons, like the scores other people gave (if the average was 4.1 and I was between a 3 and a 4, for the most part I’d lean toward the 4) and cognitive dissonance (if finished the book, it couldn’t have been that bad), I found the biggest reason the ratings I’d given of books I’d read were high was because of survivorship bias.
This was a point really driven home to me when I looked at the books that I’d started reading but never finished (thankfully many ebooks and library books; but unfortunately also a few books I’d spent good money on).
Many of these books I had no interest whatsoever to pick up again. They were awful, awful books.
But I also know that many of these awful books were possibly only awful to me.
I could picture myself, in a different world, in a different time, living a different life, actually reading these awful books and going, hey, this book’s pretty good!
Also, knowing how some books are rescued by significantly better late chapters than early ones, I could not in good conscience doom a book to a 1-star review without having first completed it.
Here’s the catch though: if you had the option to read a book you’re finding really good, versus one that you’re finding really bad, and you only had time for one, which would you choose?
For me, as I’m sure it’d be for many of you, it’s the good one.
Which means the “awful” book never gets an opportunity to be completed and rated poorly!
In the end, I found that for me all of the “good” books get rated, while most of the “bad” books don’t. Which was an interesting insight, and gave me some food for thought when I look at reviews: a book with a few, bad reviews, is probably really bad. But a book with many reviews even if not greatly rated, may be worth the chance. (Note though that some books with many of reviews are part of school curriculum, My reading tastes tend to deviate significantly from high schoolers, so I make it a point to also see who’s giving these ratings. If you’d ever built a recommendation system, this is why weighting more heavily ratings of “people like you” work so well.)
(Granted, there are some books that are so truly awfully awful, ones I couldn’t imagine liking in a million different lives. For these, I put aside the “rate only when complete” rule and make sure that 1- or 2-star review is up there as a warning for others, though I try to make it a point to at least skim the later chapters for possible redeeming qualities.)
Donn, fragments 
Leave a Reply