I recently noticed that I had obtained my first bad review from one AlexN in October last year, who granted a single star. The individual also commented, “Unnecessarily wordy which made it a slog to get through.” While I wonder about the mentality of a person prepared to give a worst possible score over a stylistic disagreement, I must thank AlexN for finally providing the impetus to write about how a published writer may interpret the star ratings granted by their readers.
The five-star rating system is well known. It provides the first five positive natural numbers of which any can be selected to evaluate the quality of a product or service. On some parts of its website, Goodreads provides vernacular interpretations of each rating; 1 means ‘did not like it’, 2 means ‘it was ok’, 3 means ‘liked it’, 4 means ‘really liked it’, and 5 means ‘it was amazing’. We can clearly see that the rating giver is assumed to be making a statement about their subjective experience of the thing or experience they paid for (or otherwise obtained).
The evaluative tool is designed to collect data from a large number of purchasers, from which the mean can be calculated and shown to future shoppers. If the central tendency of the set of star ratings is not enough, you can usually click a link to see how many reviewers selected each of the five ratings as well as any qualitative feedback, such as AlexN’s above.
In general, however, one searches for a desired thing, several items fitting the search criteria are shown, and the mean star rating for each is shown. Despite what we know to be the source of this datum, it is all too easy - in my opinion - to read the number as an objective measure of the quality of the thing for sale. One can feel positively squeamish about purchasing an item despite costing several currency units less if we find that its mean rating is, say, 0.6 stars less than that of the more expensive item. Why? Because, in something like Goodreads’ parlance, it’s ‘not as good’. Yet consider the case of my debut novel Glassworld: Out of the Darkness, which had obtained an average score of 4.8 with five reviews, and this dropped 0.6 stars to 4.2 as the result of a single low score.
Returning to the data, it is quite surprising to discover that a process can begin with an enquiry along the lines of ‘What did you think of this?’ and somehow transform the response into a number which appears to represent ‘This is how good this is’. I suppose the idea is that all opinions are basically equal, and that an objective composite or meta-opinion can be created using basic arithmetic. (Though, I should add that A****n modifies the value of a star rating based on how recent it was or how ‘trustworthy’ the online store considers the rating to be, so the above arithmetic was not duplicated exactly online.)
However, I don’t think this is defensible in the case of a book.
Perhaps, in the case of distinct manufactured goods whose actual conditions can vary based on the quality of their individual components and the expertise used to assemble them, store them and carry them to their destinations, the reviewer makes something more like an objective assessment of the product. In that case, analysing the ratings and their distribution may give us useful information about the probability that our purchase would arrive intact.
But a book, aside from any issues with the production quality or manner of carriage, is not like this. In the vast majority of cases, there is no discernible difference between my 2004 copy of Alice Walker’s The Color Purple and yours. Therefore, any difference in our evaluation of that work cannot be objective, or else we would both simply give it five stars to show that it had been printed, bound and delivered using effective methods, or less if otherwise.
Indeed, it may be because multi-department online shops feature electronic and mechanical goods alongside printed works using the same review system that we are so easily confused by their meaning. In any case, I wish to present my own system for understanding the user-provided feedback in such cases where production quality is probably not what is being evaluated. This may be of use to writers or purchasers of books.
I suggest that, instead of considering a star-rating to be either a subjective or objective review of the book, that it be the reader’s subjective review of themselves, specifically where they lie within the natural audience of the work. Thus, with apologies to Goodreads, 5 means ‘I am at the centre of the intended audience for this work’, 4 means ‘I am well within the intended audience for this work’, 3 means ‘I am on the border of the intended audience for this work’, 2 means ‘I am outside the intended audience for this work’ and 1 means ‘I am well outside the intended audience for this work’. I believe this is a reasonable interpretation because the quality of the text itself cannot change from one reader to the next, but the readers themselves can vary wildly in literacy, expectations, hermeneutics, interests, reading history, and so on.
The immediate value of this interpretation is that none of the ratings speak directly about the quality of the text, so none of them are directly critical of it. What they do is collectively build up an image of how well the book is reaching its intended audience and whether it is being promoted beyond that audience or not. This immediately creates some interesting results. For example, under the objective result system, a score of 5.0 is a perfect score. Under the reader-self-evaluation system, it implies that the book has had very little reach beyond precisely the kind of person the writer or marketing department had in mind for it. Whereas a spread of scores that encompasses some 5s, plenty of 4s, lots of 3s, some 2s and a handful of 1s suggests that the book has enjoyed a healthy reception across a diverse audience.
It should be noted that the score is retrospective upon (presumably) finishing the book. A reader who initially doubted the book might not be for them but were ultimately charmed by it might give it a 4 or even a 5-star rating, and vice versa; a reader who thought they loved that kind of thing but were ultimately disappointed might give it a lower score. Unfortunately or otherwise, the system will not speak of such conversions as readers do not say what score they think they will give the book before reading it.
In conclusion, my book, which now enjoys four 5-star reviews, one 4-star review and one 1-star review, has until now been struggling to emerge from the very heart of its audience, but has at last met someone from outside the warm appreciation of that beloved group. This is all well and good for Glassworld. The novel was written to change hearts and minds, and it may well be that readers like AlexN, who “slog” through my book but nevertheless do finish it, are precisely the unconverted to whom I most wish to preach.
So, one star reviews? Bring ‘em on! The more, the better!
No comments:
Post a Comment