I’d really like to be able to find out how well “our” usage guides, those in the HUGE database that is, have been selling over the years. This is confidential information, I was recently told by the firm handling the estate of Kingsley Amis. So I’ve been experimenting with other ways of finding the information I need, and one idea I hit upon (for what it is worth: please let me have your thoughts on this!) was to look at the ratings on Amazon UK’s website. And there I found that Caroline Taggart, with Her Ladyship’s Guide to the Queen’s English (2010), did slightly better than Steven Pinker’s The Sense of Style. See for yourself: the first Customer Review is for Taggart, the second for Pinker.
Pinker also scored lower ratings than Taggart did. On Amazon’s US site, shown below, Pinker’s ratings are slightly higher (4.4 stars out of 5) than those on the UK site (4.2), and they are based on more customer reviews (204), which makes the figure more reliable.
And here, too, we see that there were buyers who don’t seem to have liked the book much: about ten of them (5%) only rated it one star. So Taggart still does better, even though only 37 British customers took the trouble to rate the book (only 8 did so on Amazon US, still giving her a 4.5 between them). Well done, Caroline.
The differences between Amazon UK and Amazon US are worth exploring further. The British writer Simon Heffer’s Strictly English, very critically reviewed by Geoff Pullum and David Crystal, scores much lower on Amazon US (2.7 stars only) than on Amazon UK (4.2 stars). Perhaps American readers don’t like the book much. Then again, the book was rated by many more people on the UK site: 66 compared to a paltry 7 American buyers. Perhaps the higher UK rating and the fact that more buyers expressed their opinions suggests that the British general public liked the book better than the linguists who criticised it. Food for thought! But the real question, I suppose, is: how reliable are these figures?
Bridging the Unbridgeable 
I’m interested in the question at the end of your post: “How reliable are these figures?”
If you’re asking whether the figures accurately report how individual readers voted on the U.S. and U.K. websites of Amazon.com, I’m prepared to believe that they do. I don’t think the people at Amazon simply make up the numbers.
But if you’re asking how much weight we should give to the opinions of several dozen or several hundred self-selected readers, even assuming that they all vote honestly and give their true opinions of the books, that’s a different question altogether. Do we believe that the collective opinions of all these readers says something worthwhile about the books?
As H.L. Mencken wisely observed, “Democracy is a pathetic belief in the collective wisdom of individual ignorance.”
Thanks for your comment, Lee. I actually agree with what you say. But I’m going to test what people have to say about these books by comparing their comments with a collection of responses on a usage survey I carried out a few years ago. The people who responded were also self-selected in a way, and there is therefore a similar bias that I want to take a look at. I’ll report my findings again, and will be looking forward to your commants again. Thanks again!
Ingrid
It’s interesting to see how the number of comments varies between US/UK and how a book’s popularity varies. For a clearer indication of actual sales – though I’m sure you’ve thought of this – the Amazon ranking helps. Pinker is near the 1,000 top sellers mark, while Taggart is around 33,000! (but Amazon rankings, I believe, take mentions in the press, and other things into account apart from pure numbers).
For absolutely reliable sales figures, you need a mole at a publisher, or someone who has access to Nielsen book data: http://www.nielsenbookdata.co.uk/controller.php?page=1
Regards, J
Thank you Jeremy! That is a really good suggestion. As for Nielsen Book Data: I’ve applied for a trial version, without any reply, and am waiting for a reply from Customer Service. I doubt it if I can get access in a formal way, so if anyone can help?