How to interpret large volumes of patient feedback:methods from computer-assisted linguistics

In this article, we demonstrate how methods from corpus linguistics can be used to interpret large collections of patient feedback. Using a series of established corpus linguistic techniques, this study examines the key areas of concern in 29 million words of online patient feedback about the National Health Service (NHS) in England. Focusing on the theme of staff rudeness, our analysis shows how the power of computer-assisted textual analysis can be fruitfully synthesised with fine grained, humanled analyses to develop a more refined picture of the complex ways that patients evaluate healthcare services in their comments. We argue that, by affording the opportunity to study large collections of authentic language data, corpus linguistics methods can facilitate the types of objective and empirical approaches to large datasets that are now commonplace in the domain of evidence-based health communication research