A multi-dimensional analysis of corporate blogs.

Using a corpus-based multi-dimensional analysis, this study investigates the linguistic features and variation of corporate blogs from four industries: Service industry, the Wholesale and Retail Trade industry, the Manufacturing industry, and Information industry. The primary goal of this study is to examine the linguistic variation in corporate blogs from different industries. The data used for the current study is a 570,745-word corpus consisting of 995 textual posts from 40 top-ranked corporate blogs. In the multi-dimensional analysis, 67 linguistic features in the corpus are tagged, counted and normalized by using a Multidimensional Analysis Tagger (MAT). We imported the output files containing relevant statistical information from the MAT and then to SPSS 25.0 for a further t-test analysis. Overall, the corporate blogs are closest to the text type of general narrative exposition. The study finds that, as a hybrid genre, corporate blogs are basically informationally dense, non-narrative, interactive, and contextually independent. There are some statistically significant differences between four sub-corpora in corporate blogs. In particular, the Wholesale & Retail Trade sub-corpus is the most distinctive one that is significantly different from others on many dimensions, characterized by giving instructions, or telling stories, rather than persuading their audiences to do something. [ABSTRACT FROM AUTHOR]