Best practice for academic writing: write and cite or write first?

Question

I'm an undergraduate student of psychology. My teachers assign writing assignments from time to time, and usually require some minimum of cited sources in APA format (as expected).

When I start writing, I often find myself stuck. For many of my past writing assignments for school (including English Composition essays, which did not always require rigorously cited sources), I found that I wrote my best essays after brainstorming an idea, and then basically writing the whole thing in one sitting. (With excellent results, if I may say so myself.)

With these newer assignments, however, I've been having some trouble. I haven't figured out whether it's better to write everything first, and then go back and insert citations later*, or if I should "write and cite" at the same time (slowing down my writing considerably). I often end up doing both and neither at the same time, which makes everything so much harder, consumes so much more time.

• "Write and cite" takes longer to write, but I'll have all or nearly all of my citations in place when I'm done writing.
• "Write first" is easier to write, but then I'll have to carefully review the paper after it's written to make sure everything I've claimed is supported.

I figured it'd be a good thing to figure out now, as there will be many such assignments in my future (including, God-willing, a doctoral thesis).

---

* I usually write in Microsoft Word, and make good use of their citation manager. (Which, incidentally, is why I did not use the apa tag -- this question is not specific to APA citation format. Citation formatting is not something I even think about anymore: I just put the relevant info into Word and let the program sort it out :-)

Answer 1

My answer is that you should read widely and then write and cite, because you want the foundation you're building upon to be rock solid. The risk of writing first and inserting later (which is a common approach and easier) is that if you write your paper first, (telling yourself you want the flow of it), you'll find yourself wanting to cherry pick citations (whether they're good or bad) to fit your existing paper.

As a psychology student, I'm sure you see the problem with cherry picking sources?

Write and cite. At the end of the paper, you will still need to go back and add in additional sources.

You're not writing creatively in these assignments. You're not writing fiction, it's not stream of consciousness. It's closer to technical writing. You're not creating a persuasive argument in the way you were taught in school. You're not using rhetoric, not to the same extent anyway. You're synthesizing rigorously tested ideas into a new framework. It's different.

I urge you to view your paper as 50-90% 'existing knowledge and work' and 10-50% 'your framing of that.' Your job is to take what has been established in the field and bring it together in a compelling way to help people see your thesis.

Answer 2

"Write and cite" is good practice that you should start getting accustomed to early on. The longer the piece you write, the more sources you would have to juggle. Now, imagine there are twenty articles you would be citing, four of them say something relevant to the point you're making in one single paragraph. Working by "write first, cite later", you'd need to meticulously go over those four articles (if you remember which four those were - otherwise it's all the twenty), to find which article said which particular thing you're referring to. Repeat for each and every paragraph. That's a lot of work. Now imagine you're writing your thesis, citing several hundred sources. "Write first" becomes impossible. Don't forget that every piece of information and every idea you gleaned from somewhere needs to be cited. And as @DPT states, a large part of your writing would be looking back at the existing literature.

Forget writing a complete piece in one sitting. You won't write even your thesis proposal, let alone your thesis, in one sitting. There's way too much research into existing literature that you'd need to do. And since you don't write everything all at once, things bleed into each other, you don't always remember which article a particular idea came from. So if you don't cite at once, you have to go looking for it all over again.

I find it useful to summarise each source as I read it, copy out the ideas I will need to refer to later, and attach a citation to that. This process also helps me arrange everything in my head. But more relevant to your question, when I come to actually writing whatever it is I'm writing, I can copy-paste the already-formatted citation from the pre-made page, instead of breaking my chain of thought to find the relevant information for the citation manager.

Answer 3

If you don't keep track of your citations as you write, you will forget where you found the information. It will only get worse as you get further along and are using more and more citations (often from sources with similar names and covering a lot of the same material), with stricter citation requirements. Going back to figure out what goes with what is tedious and error prone.

Therefore my suggestion is to do quick and dirty "citations". Your style guide requires citations in a certain format, but your early drafts can use whatever format you find fastest. In the past I've just dumped the information in parentheses or used comments. For anything online what I put down is the URL. For paper resources it's the title and page number. You can copy and paste if you're using the same resource multiple times in a paper.

When you're ready, it should be easy to edit your paper to replace what you have with real in-text citations and an entry in the works cited.

Answer 4

Sometimes when someone asks "which should I do first when writing?" the answer is "whichever one you want." For example, I'm writing a historical novel requiring a lot of research. While I'm a research-first type of writer, it would be wholly appropriate for a different writer to plunge ahead with the story then come back much later and change the details (what clothes they're wearing, what species of tree are they sitting under, etc). In most cases, any secondary changes you'd need to make wouldn't be critical.

In an academic paper though, you can't write about, say, typical behaviors in a population then go back and find studies, case work, or fieldwork that supports your statements. Not in the same way that you can write a fictional scene where characters sit down to lunch and later you can check that the foods you chose were available then, or substitute ones that were, then put in the type of utensils they used.

Your "story" needs to come from the research.

And even if it does, if your conclusions result from the research you did, it's easier to keep track of citations before the end.

My recommendation is to make a list on your computer of all your citations along with various quotes and/or notes. Include page numbers. If you don't have access to digital forms of the works you're using, then you can put page numbers and brief notes in each section with the full citation.

Now, sit down to write in the free-flowing way you prefer. Each time you reference something, either copy and paste the quote/notes and the citation, or put some sort of placeholder there so you can easily find it when it's time. You won't use every quote/note (or maybe not even every source), but the most important ones will be there easy to grab. And you'll have enough pointers to find more if you decide to flesh out that section.

The part of your self that likes to sleep and eat and bathe will thank your writer self later.

Answer 5

I feel like you've identified a version of the architect-gardener personalities for writers, where "write and cite" is more the former, and "write first" is more the latter. I'm going to use these ideas as a way to explore your problem, but in reverse-order.

"Write first" is exploratory and creative. This is appropriate when you don't know what you're doing, or can't articulate something well (yet). Success with this writing style will result in a more enthralling read. Some writers are just like this. Its weakness, as you mentioned, is the fact that it has to be read and re-read and edited repeatedly to trim off some of its exploratory branches and give the necessary citations.

Do not do this; this will ruin you. You will badly misjudge your time and effort, being taunted by with a feeling of closeness to completion when all you actually have is a passable "story" that has to be worked on for a shocking amount longer to turn it into something that passes for your goal.

"Write and cite" is the other side of this coin, and you already intuit its efficacy because of the arduous inclusion of citations. "Slow and steady wins the race". You will (of course) have to re-read and improve.

But why not both?

A lot of people will say "if you brainstorm an idea and then write it in one sitting and that works, then just do that", but that won't work when citations are necessary. I propose you take notes as you brainstorm. Jot bullet points or key words, or put them on cards so you can shuffle them around.

What this does is to take your gardener exploration strength and immediately create building blocks ideas. For each of those major points, you can reach for citations.

So it looks like this:

1. Gardner: Brainstorm, but write keywords or very rough notes.
2. Gather up notes into core points.
3. Clean up notes, prune away any which don't appear viable to your core point(s).
4. Order your notes.
5. Fill in any obvious blanks.
6. Ensure each core point has its own citation(s).
7. Gardner: Weave your points into one narrative.

With this, you open and close with your strengths.

This also allows the the possibility for you to brainstorm by beginning with a conclusion and assembling the set of citations which supports it. Yes, I know this is awful science, but makes for a very focused piece for a student (and it's easier to grade).

Answer 6

I'm a psychologist. I'll use an analogy to illustrate my approach to academic writing.

In a novel, you have characters who act out a story. The story has a plot that moves from some situation at the beginnung to another situation at the end. In between, the characters and their actions influence the progression of the story and the turns it takes. To write your story well, you have to be intimately familiar with your characters. You have to know how they would act in a specific situation, how they differ from each other, the relationships they have with each other, and so on.

In an academic research paper, you have other publications and you have your hypotheses. The other publications define the situation at the beginning, before you begin the experiment. But something about the published research is wrong. And you want to find the truth. Or the published research is not yet sufficient, and you want to complement it. So you develop your experiment (your "plot") from the relationships among the different published research papers. Your hypothesis is either the "antagonist" to those publications or a "companion" to them. Between these "characters", the "story" of your experiment takes shape and develops towards its results. In the end, your results either "win" against their "opponents" or are united with them in a unified new theory.

And just as when you write a novel, you have to be intimately familiar with the "characters" (that is, the published research) to narracte the "story" (that is, the argumentative structure and how you came to design your experiment in the way that you did) of your research.

And when you are intimately familiar with the published research – that is, when you are an expert in your field – writing and citing become one.