A quote from "Making a bigger deal of the smaller words: Function words and other key items in research writing by Chinese learners"

Second, we cannot comment on whether the usages discussed in this paper are still communicatively effective despite being marked, as that is an empirical question that can only be answered in complex ways through further investigations. [aka my study!] What we contend is that written academic genres such as dissertations require a high level of accuracy in expression and stylistic appropriacy, and most academic writers in China aim for international intelligibility and maximal acceptability in their writing. Even if the specific usages of small words do not in themselves cause critical problems in comprehension, learners would do well to avoid them if they want to come across as language professionals, particularly since “small issues” can have undesirable cumulative and additive effects. The EXJA journal articles that we use as a yardstick are, as we mentioned at the beginning of our paper, representative simply of “good English” rather than native-speaker English, as we did not make nativeness a selection criterion in our corpus compilation. Finally, what we can say is that the choice between “local flavor” or “expert-like writing” is not a clear-cut, either/or option. 

As mentioned in the discussion of besides, the corpus-based approach allows us to formulate the following specific pedagogical strategies: (a) unlearn the clearly unacceptable and more spoken-like features; (b) maintain the use of specific constructions that are used correctly; (c) practice the use of novel or underused constructions in order to expand the active vocabulary. Perhaps what we can do is to offer our apprentice writers various alternatives—add to their rhetorical repertoire rather than subtract. Overextended uses of perfectly good academic phrases (e.g., according to) could be handled with sensitivity, to avoid discouraging learners from “trying their hand” at scholarly writing: Expert corpora such as EXJA are an affordance, a resource for teachers to show learners alternative ways of expressing what they want to say, providing authentic samples of structures that apprentices can learn from. Through structured exposure to genre-relevant samples of language use, apprentices can hone their intuitions of how certain phrases are used by expert writers, and learn the alternatives by example. This is different from a word list/phrase list approach, where learners are given a catalogue of putative academic formulae; such an approach tends to lead to misuse, abuse, or overuse, as learners presented with such lists frequently make the mistaken assumption that “more is better.” The more nuanced approach suggested in this paper is for instructors to give credit for good (“correct”) usages, and then offer alternatives so that apprentices can learn a greater variety of ways of saying the same thing, as well as learn when not to use a particular word or phrase, through increased exposure to expert texts and practices.