Consider, for example, an online shop listing multiple products available for purchase. Each product has an average rating from 1 to 5 calculated from all that product's customer ratings to date.
To me, a product with an average rating of 4.8 calculated from 1,000 ratings is preferable to a product with an average rating of 5 calculated from 1 rating. I'm using extreme examples here but you get the point. I would like to be able to rank the products taking this preference into account.
I realise there is a degree of subjectivity to this so I would be interested in hearing some different approaches to the issue. For example, for products with very few ratings I am considering effectively synthesising additional ratings by factoring in the average for the entire population when calculating the average, but I'm not sure to what extent this would introduce bias, how to define the "not-enough-ratings" threshold, or whether there is a better method I could be using.