Ok, I now had my data: 100 “Rants” from TripAdvisor. What next?
I did what (I am sure) most discourse analysts do. I read through my data. I thought about my data. I studied my data. And then I read, studied, and thought about my data some more. I repeated this process several times. In fact, I became so familiar with my data that, after a while, I could recite entire excerpts verbatim from my 60+ pages of downloaded hotel reviews.
Because my most recent project prior to this one had been about how speakers complain in a F2F professional setting, I thought it might make sense to consider my TripAdvisor “Rant” data from the perspective of complaints. Specifically, I was curious about which of the language features characteristic of spoken complaints, as reported in the research literature, were also applicable to complaints in this specific CMC (computer mediated communication) genre.
What I found was interesting. First of all, although there had been dozens of books and articles published about how people complain in all sorts of languages (from Chinese to IsiXhosa), I could find no studies about how people use languages to complain in online environments. This was around 2008. I also found that, at this point, only 2 scholars had examined online reviews from a linguistic perspective (Pollach and Mackiewicz). So it seemed like I had found a research topic that was worth exploring in greater detail.
My analysis of TripAdvisor complaints turned up the following findings:
1) About 1/3 of reviewers actually included 1 or more positive comments within their primarily negative (1-star) review. Presumably those reviewers wish to offer a balanced perspective, and do not want to come across as being categorically negative.
2) Online complaints tended to co-occur with other speech acts, such as recommendations, advice, and warnings…which is different from findings about F2F complaints. This tells us that the kinds of speech acts that are most likely to co-occur with complaints depends on the larger context in which those complaints occur.
3) Lots of reviewers made explicit reference to their expectations, which are often based on the hotel’s price, brand, or their own pre-travel research. Consumer complaints seem to be closely related to their expectations.
4) Some reviewers recognize that multiple audiences may be reading their reviews online, and take the opportunity to address two audiences within a single review text: other consumers, as well as the owners of the business being reviewed. This is unique to CMC. Before, when we complained (either in person, or in letters written to business), it was difficult – and often impossible – to address multiple audiences simultaneously.
More details about online complaints can be found in my article:
Vasquez, C. (2011). Complaints online: The case of TripAdvisor. Journal of Pragmatics, 43, 1707-1717.