The other day I was reviewing a paper that looked quite interesting, but unfortunately was written in such poor English that I could not really understand what was going on. I felt sorry for the author(s). I then recalled a recent discussion with a colleague of mine about how important so-called transferable skills are for students: We know that most of them won’t end up in academia, so stuff like critical thinking, structuring an argument, or reading a regression table a are pretty important. Among these, coherent and comprehensible English must rank very high. For those who stay in academia, I’d argue that it’s the most important skill, because it’s central to communicating with other researchers and having your work understood. Only this way can others build on what we do. Ironically, however, teaching English is typically not a focus at universities, if it is done at all. Like so many things, we just kind of assume students (have to figure out how to) do it.
Image: CC-by-nc Moiggi Interactive
The other day I was finishing off supplementary material for an accepted article, and had a major panic for half an hour. It all started with my adding a simple frequency table of the outcome variable: a binary variable. When I checked the PDF it turned out that I have miscoded the outcome variable (at least this is what it looked like) — instead of 60% 1s, I had 40% 1s. What to do? No, I didn’t think the substantive results would have been completely different, so I could have done major work on the page proof, replacing every number in the paper. For a moment I considered ‘unseeing’ what I discovered and bet on the likely case that nobody ever would replicate my findings despite my making all the code and data available. I could even have removed that line where I promise the replication code during the page proof. Ethically defensible this would not have been. Retraction passed my mind. Fortunately, it turned out that there was a benign reason. After going back to and quadruple checking the questionnaire, the raw data, and all the recoding and code, it turned out that I simply wrongly labelled that table of the outcome variable. Relief and feeling silly for panicking.
Apparently there are still researchers out there (no, I won’t name you) who have not heard of Zotero and Zotfile. Zotfile takes Zotero to another level by managing PDF files, including the ability to extract highlights and comments from PDF files. Try them.
No, I don’t mean you should read your paper at a conference, that’s just too boring to listen to (so even if you have something interesting to say, we might not be paying attention). You should read your manuscript aloud before submitting it to a journal (or an abstract before you submit it to a conference). Reading aloud is quite useful to check the manuscript because doing so slows you down: you read it more carefully — and you might spot things you want to change.
Qualitative studies are often described as small N studies because the number of respondents is small. I argue that this is the wrong perspective: What we really have in qualitative data, say interviews, is lots of data (points) clustered within individuals. Rather than focusing on the number of respondents, we should probably focus on the number of relevant statements (i.e. statements about our quantity of interest), and describe this number (along with the number of respondents). When computer aided qualitative data analysis (CAQDA) is used, I guess the number of tags relevant to our quantity of interest is that number. Seen this way, many qualitative studies are no longer small N studies, but we’re still faced with unstructured, messy data that may be difficult to analyse, and of course we don’t have independent observations — so generalization remains a challenge.