Thread by @CogSciGirlPhD

@TravisSowards @OpenAcademics Hi, I've been a writing instructor at university and graduate level (that's my credential for "good writer", take it with a grain of salt). Here are my suggestions:

1. Don't read academic papers for "good writing". Academic papers are often crap in terms of writing quality. 1/n

@TravisSowards @OpenAcademics A lot of scientists are lazy in the sense that they think good science can overcome bad writing. It does not.

2. Read GOOD writing everyday. There's a certain cadence to good prose. Stuff like @TheNewYorker, which is vetted for writing quality. 2/n

@TravisSowards @OpenAcademics @thenewyorker 3. Understand that an academic paper is NOT EXPOSITORY. It's argumentative. You are trying to convince the readers that your theory is correct.

4. From #3, realise that argumentative writing requires certain components which academic papers often miss. 3/n

@TravisSowards @OpenAcademics @thenewyorker These components are Knowledge Gap, Research Question, Hypothesis, and Answer/Conclusion. To define:

Knowledge gap: The big hole in the knowledge compendium of the field.
Research question: the teeny tiny part of the knowledge gap you are trying to fill. 4/n

@TravisSowards @OpenAcademics @thenewyorker Hypothesis: What you think is happening, and the mechanism behind it. Without the "mechanism" explanation all you're doing is prediction.

Conclusion: confirmation/rejection of hypothesis, and why you confirmed/rejected. 5/n

@TravisSowards @OpenAcademics @thenewyorker This is a tricky concept to grasp for many students, so to give an example using peanut butter and jelly sammich:

Knowledge gap: what makes the best peanut butter jelly sammich. (Notice that this can be answered in MULTIPLE WAYS. Knowledge gap is a BIG gap.) 6/n

@TravisSowards @OpenAcademics @thenewyorker Research question: today, we asked "which jelly flavour goes best with SMOOTH peanut butter". (Notice that suddenly the knowledge gap addressed in the question is significantly narrower).

Hypothesis: we hypothesise that strawberry is the best flavour... 7/n

@TravisSowards @OpenAcademics @thenewyorker due to the strawberry's sweet fragrance, its acidity, and its tartness that would counterbalance the creamy sweetness of peanut butter (notice that I gave a reason WHY we think it's strawberry).

Conclusion: After surveying 1 person, we concluded that strawberry is the best. 8/n

@TravisSowards @OpenAcademics @thenewyorker The person surveyed reported that the person's favourite jelly flavour for PB&J was strawberry (here we gave a reason WHY we confirmed the hypothesis).

These components must be in the abstract and the introduction, and there cannot be a lot of discrepancy amongst the... 9/n

@TravisSowards @OpenAcademics @thenewyorker readers on what the hypothesis is, what the research question is, etc. These are the foundations for your argument in the paper.

To see these in action, read legal briefs (#legaltwitter, know any good briefs?). Lawyers use previous cases and laws instead of data... 10/n

@TravisSowards @OpenAcademics @thenewyorker but the basic concept (a broad-concept question that is being addressed, the narrower question as it pertains to the case, the answer to the question, and why the answer is true).

For results, at least the question and the hypothesis need to be mentioned as a reminder. 11/n

@TravisSowards @OpenAcademics @thenewyorker 5. Plan your writing. Write an outline. It's tedious, but there's a reason almost every Language Arts teacher made you write an outline before writing a 5 paragraph essay, and it wasn't because (s)he liked grading 50 outlines from kids. This is where you hammer out the... 12/n

@TravisSowards @OpenAcademics @thenewyorker 4 components. Write them out.

6. Write. Write a lot. I don't mean write papers everyday. Write a blog. Surely you have a lot to write about your #gradschool life. This is an exercise to find your own rhythm and vocab, which is what people refer to as "voice".

@TravisSowards @OpenAcademics @thenewyorker 7. For the results section, make sure that you tell the audience what we're supposed to be looking for. Take a look at the graph below: 14/n

@TravisSowards @OpenAcademics @thenewyorker What are we supposed to be looking at? The general trend, or that one outlier? You need to point them out. If you want us to take a look at both, you need to mention both (e.g. the general trend was on the rise, which indicates that... [explanation] The outlier was... 15/n

@TravisSowards @OpenAcademics @thenewyorker due to the one person who was abnormally adept at blowing out candles with one's nose").

8. Discussion: this is where you go into details for your interpretation of the data, discuss what might've gone wrong, what you will do next, etc. This is the part where you... 16/n

@TravisSowards @OpenAcademics @thenewyorker get relative freedom with your thoughts.

Sorry for the long thread, but I hope this helps. A scholar's job is to present original thoughts/arguments. Anyone can collect data, but only YOU can come up with your arguments. Writing is a way to showcase it.