Category: Writing

How are abstract, introduction, and conclusions different? (they really are)

Darrion Nguyễn (@lab_shenanigans) shares the above meme and writes

Abstract: I will do this thing
Introduction: I’m finna do this thing
Conclusion: So about that… I kinda did the thing. However, it can only be done only under some specific circumstances, which are beyond the scope of this paper and would require additional studies 🙃

Hey, Darrion, guess I'm gonna have to go ahead and disagree with you there! One big caveat: Clearly, as so many things, this is ultimately a matter of taste or group tradition.

But (!), the way I see these things, the ambiguity (amtriguity?) vanishes, and each of the three parts Abstract, Introduction, and Conclusion can be used to fulfill a distinct and useful purpose.

Let's start at the beginnning, with the Abstract. Remember that the abstract is available to everyone visiting a journal's website, regardless of whether they have a subscription to read the full paper or not. Some authors prefer to treat the abstract like a movie trailer, just teasing the article, not giving away too much and certainly not the resolution. I think that's a wasted opportunity: I want the curious reader to get a complete summary of the paper without having to first figure out how to get your library to subscribe, or to actually buy the article. Most importantly, I want the curious reader to know what the insights are, the take-aways.

The purpose of the Abstract is to provide a complete summary of the article, including research question, methods, approach, results, and conclusions.

Next is the Introduction, and there's a whole post about that: https://science.banuti.de/how-to-write-an-introduction/ Without simply repeating what I wrote there:

The purpose of the Introduction is to present an interesting research question and demonstrate you are able to solve it.

Finally, we have the Conclusions.

The purpose of the Conclusions are to identify your lessons learned, things that surprised you. Find 3-5 points that you want the reader to take away from the paper, order them from most important to least important.

No need to summarize the paper (if you really, really want, maybe include a very short sentence stating the research goal identified at the end of the introduction, but don't repeat what you did).

Also no need to spell out all the things that you didn't do because they are beyond the scope. That goes without saying: things you didn't do are not in your paper.

So, sure, go ahead and treat Abstract, Introduction, and Conclusion all the same at your own peril; you are unnecessarily sacrificing a big opportunity to make your paper much clearer and poignant.

Happy writing!

db

What’s the difference between thesis and paper?

We talked about how to structure an academic article, but now you want to finally (finally!) write up your thesis. Is it the same thing, just longer?

Well, no. But also, yes. Kinda. Also, it depends.

Generally, the structure of a thesis is indeed similar to the structure of an academic article. Abstract, Inroduction, Methods, Results, Discussion, and Conclusion can be used to outline your thesis structure.

But beware: Your school has their own requirements (and, ideally, template) that your thesis has to follow before it will be accepted. Look it up!

Let's recall what we said about the purpose of a paper:

The main purpose of a paper is to convey a new idea, and show that it has merit.

At this point, we have to distinguish between a Masters thesis and a doctoral dissertation. Indeed, the purpose of an M.Sc. thesis is quite different:

The main purpose of an M.Sc. thesis is to demonstrate that you are a problem solver with domain knowledge.

Do you see the difference to the article? In your thesis, you can (and, I think, should) also write about things that did not work out. Your M.Sc. thesis is the story of how you tackled a problem, and overcame challenges. Your idea A did not work out? What did you do then? Your experiment was stuck in stage B, how did you find an alternative approach? Etc. It is an account of your path.

Most importantly, it is not strictly necessary that everything worked out. Would it be desireable? Sure. Would it be more satisfactory? Of course. But for me, it is not necessary for an M.Sc. thesis to come out with a positive result, as long as you could demonstrate without doubt that you are an independent problem solver with domain knowledge.

That's different for a Ph.D. dissertation, which is more closely related to an academic paper:

The purpose of a Ph.D. dissertation is to demonstrate domain knowledge and to use it to advance the art.

You have to do something new, something no one has ever done. And something must work. It does not have to be revolutionary, it does not have to be better than previous methods/ideas/models/techniques, but you have to show that you did something that no-one has ever done.

Happy writing!

db

Things you should never ever write in a paper (or thesis)

I've seen all of these before. Maybe it's because some schools encourage filling pages, or because of simple lack of practice, or because it is honestly meant.

"Obviously", "It is common knowledge", "Trivially", etc. -> condescending, or you're too lazy to look up references. Easy to fix: just delete!

"We obtain the value by using the formula from the book" -> sounds like you don't know what you're doing. Call the equation or principle by its name instead! Also, in case you really don't know what you're doing -- read up on it!

"The function xyz of the software package we did not program returns us a value, called 'enthalpy', which we can use to" -> again, sounds like you don't know what you're doing. Talk about your tools once, in the Methods section, and never again after. But do make sure you do know what you're doing.

"However, [17] failed to do xyz" -> never talk bad about a fellow researcher, it makes you look like an ass. Talk about what they did instead, then identify a knowledge gap that you address. Whatever wasn't addressed before is your playing field!

"I call this the 'John Smith' effect, after myself"'. Just don't.

Happy writing!

db

How to write an introduction

After having discussed the overall Structure of Papers, it is now time to look into the different parts of the paper structure, and see if there are more patterns we need to be aware of (spoiler: there are).

We said about the purpose of a paper:

The main purpose of a paper is to convey a new idea, and show that it has merit.

Then, we can say about the role of the introduction:

The main purpose of the introduction is to identify a problem worth solving, and to show that you are indeed able to do so.

The shape of the introduction is much more formalized than any other part of a paper, which helps and guides us writing one. In particular, we find

  1. State the "BIG Problem" your paper addresses (even if only ever very indirectly). Start out with some BIG Problem (TM), that everyone can relate to. Climate change? Great. COVID? You got it. Gun violence, polarization of the society, etc. 1-2 sentences.

  2. Your topic, hyperspecific. Guide the reader from the BIG Problem statement to your particular hyperspecific research question. You want to solve COVID? Did you know that this one particular protein is superimportant? You want to solve global warming? Did you know that the molecular interaction of carbon dioxide at the critical point is super relevant? 1-2 sentences.

  3. Literature review. This is a tricky one: you have to give the reader an overview, but also demonstrate that you talk the talk and walk the walk and know the lingo of your domain. The meat of the introduction. Describe what is known, what others (and maybe yourself!) have done, what the current state of knowledge with regards to your hyperspecific topic is. 1-3 pages.

  4. What isn't there? The Knowledge Gap! Clearly identify one aspect that no one has addressed yet, something that isn't known yet, some open question. 1 paragraph.

  5. How are you going to fill that gap? Your Research Goal. State it very clearly, "The goal of this paper is to xxx". 1 paragraph.

Some like to give an outline of the paper in that last paragraph, me personally, I think it's a waste of space. Why? It's always the same. Always: The Structure of Papers.

Happy writing!

db

The structure of academic articles

They're all the same! And it's a feature, not a bug!

Ask yourself: what is the purpose of an article, or, as I'll call it from here on, a paper? Sure, it could be vanity, or someone's boss asked them to write something. But surely

The main purpose of a paper is to convey a new idea, and show that it has merit.

That's it! And everything else should serve that purpose. It turns out that over the last few hundreds of years, Science has converged to a standard format to convey ideas and show their merit. Enter, the standard structure of a paper!

  1. Title. This is marketing. Make something snappy that is interesting, but also explains what this is about. 2 to 8 words.

  2. Abstract. (Executive summary) There are different views on this, but for me it's an executive summary, not only a teaser. Put the why, how, what, and results in here. 200-400 words.

  3. Introduction. (Why are you doing this?) Explain why you wrote the paper. What is the problem you are solving? What is the big picture? Which other ideas were there, and what specific question did they not answer? How are you going to solve this? 1-3 pages.

  4. Methods. (How did you do it?) Which tools and theories do you use in this paper? How do you know they work? A few paragraphs to a few pages, depends on what you're doing!.

  5. Results. (What did you find?) What does your analysis show? Neutral, objective discussion of the findings. This is the core of the paper, see how much space you need (and the journal gives you) to document everything.

  6. [Discussion] (What does it mean?) Not always there, but this is the spot to interpret your results rather than just report them. Maybe 1-2 pages, if necessary.

  7. Conclusions (What did you learn?) Did something surprise you? If there are 3 points you would like the reader to realize, what are they? 3-5 short paragraphs or bullet points, ordered in decreasing importance.

Of course there are intricacies to each of these points in turn, but this is a good starter to get you oriented!

db