We modeled customer rating bias caused by the framing effects of rating scales. By defining rating bias and further classifying rating inflation and deflation, we argue that framing-based biases are characterized by two parameters, namely the reference point and the scalar of a rating scale, which allow researchers not only to test whether rating bias exits and, if so, assess its magnitude, but also to identify the sources of rating bias. We further classify framing-based bias into the reference point-based, scalar-based, and both the reference point-and-scalar-based to capture a wide range of rating bias that may have occurred in the marketplace. We employed a computer simulation method to portray the distribution of customer rating bias at the aggregate level when the framing effects are factored into customer rating behavior.