How to write a research paper introduction (and a thesis chapter 1)

What should the introduction of a research paper include?

Only four things go in. First, the broad area and why anyone should care about it. Then what earlier studies have found, and you cite each one. After that, the part those studies have left open. Last, what your own paper does, written as an aim, a question or a hypothesis.

In our guide on how to write a research paper we call this shape a funnel, and that guide also gives the usual word counts for each IMRaD section. Here we go one step deeper, to which sentence does which job.

Keep the full methods out. Keep the results out, and the history of the field also. Some journals want a separate literature review or theory section, and in management and education this is quite common. In that case your introduction becomes shorter. Even then, the gap has to be there.

What is the CARS model for research introductions?

John Swales is an applied linguist at the University of Michigan. He studied how the introductions of published articles are put together. In Genre Analysis: English in Academic and Research Settings (Cambridge University Press, 1990) he explained them as three “moves”. Inside every move he listed some smaller “steps”. He called the whole pattern CARS, short for Create a Research Space.

In Move 1 any or all steps can come, so take only what your paper needs.

The real argument happens in Move 2. Swales gives four options here and “indicating a gap” is just one, so know the other three also. It is also the weakest one when the gap is simply “nobody has studied this in Tamil Nadu”. Counter-claiming is many times stronger, where you show that earlier work is wrong or incomplete in some specific way. Question-raising is also good when the evidence so far leaves some question open. For how to write that sentence, see our guide to writing a research gap statement.

Move 3 changes the most from one journal to another. In some journals the introduction ends with a line giving the main finding, and in others you will never see that. Some papers give an outline of the article section by section, though short papers mostly skip it. Read three recent introductions in your target journal and follow them.

Swales went back to the model in Research Genres (2004). With Christine Feak he also wrote a textbook, Academic Writing for Graduate Students. The USC Libraries writing guide follows its 3rd edition (2012), and there the first step is called “claiming importance”. The logic is the same and only the labels are a bit different.

CARS is good for checking a draft after you write it. If you build the introduction step by step from the labels, the writing comes out stiff, and reviewers can make out how it was done.

Research paper introduction example, annotated

We made up this topic, so whatever it claims is not a real finding and you should not quote it. Wherever you see square brackets, a real introduction would need a citation.

Paragraph 1

Cyclone early-warning systems on India’s east coast are widely credited with reducing deaths in severe storms [cite disaster management reports]. Alerts that once travelled by siren, radio and village volunteers now also reach households as SMS and voice messages in the local language [cite]. Research on these alerts has concentrated on delivery: how many phones received the message, and how quickly [cite, cite].

What it does: Move 1. It claims importance and backs it with evidence, makes a general point and then reviews what earlier research measured. There is no line on “the importance of disaster management for the nation”.

Paragraph 2

Receiving a warning is not the same as acting on it. Studies of evacuation decisions show that people weigh a warning against other things [cite, cite]. They weigh it against their own experience of past storms, the advice of neighbours and what they stand to lose by leaving. For fishing households the loss is concrete and immediate: a day at sea, and sometimes the boat and nets left on the shore.

What it does: Still Move 1, but narrower, bringing in a second body of research. So the gap comes out of two things the reader has already accepted.

Paragraph 3

Yet studies of SMS alerts in fishing communities have rarely followed the message past delivery. There, the decision is usually whether to go to sea at all, not whether to evacuate. The few studies that ask about response rely on recall surveys conducted months after the event [cite]. Such surveys cannot show which part of a message people trusted and which they ignored.

What it does: Move 2. The turn comes at the word “Yet” in the first line. There is a gap (very few studies look at response) and also a counter-claim about the method used. Reviewers will read this paragraph two times.

Paragraph 4

This study examines how fishing households in two coastal blocks of Odisha interpreted and acted on SMS alerts during one cyclone season. We used daily diaries that the households kept, plus follow-up interviews, to find which features of a message led families to stay ashore. The study area and methods are in Section 2. Section 3 gives findings, and in Section 4 we discuss what they mean for how alerts are worded.

What it does: Move 3. The aim takes one sentence and the method gets one clause, not full page. If articles in your journal don’t have a structure sentence, remove it.

What are the common mistakes in a research paper introduction?

These are roughly in the order I see them in drafts.

The history-of-the-world opening

“Since ancient times, water has been essential to human civilisation.” “India is an agricultural country.” Nobody will argue with such lines, so they add very little. Suppose your paper is on drip irrigation subsidies in Maharashtra. Then begin with irrigation subsidies itself.

The gap is buried or missing

The draft spends three pages going through studies, one study in each paragraph. Only at the end does the gap come, in half sentence. Your supervisor may ask “so what is new here?” and the introduction should already have answered that.

No aim sentence

Some introductions finish with “Therefore the present study” or just a list of variables. Then the reader has to guess what the paper wants to find out. Write your aim as one complete sentence, and use the same wording you use for it in the abstract.

Results in the introduction

Coefficients and percentages, or a paragraph on what you found, should go in the results and the discussion. Only one exception is there. Some journals expect the main finding in one sentence in Move 3.

Too long for the paper

If the introduction takes up a quarter of the paper, mostly there is a thesis chapter sitting inside it, because usually a paper introduction is only three to five paragraphs. Pick five recent articles from your target journal and count their paragraphs yourself.

Go through these checks before the draft goes to your guide:

• The first sentence talks about your own topic. It does not start with India, the world or the history of the subject.
• Somebody reading can put a pen under the gap sentence. It comes in the second half of the introduction, not on page one of a three-page opening.
• In the last paragraph there is one sentence that more or less starts “This study…” and says what the study is doing.
• No numbers from your results are there, unless your field expects one line on the main finding.

How is the introduction to a literature review or a research proposal different?

A literature review chapter does not have to open with CARS. The opening gives the scope and tells the order of the sections. The argument and the gap come at the end. Our literature review examples and format guide has a sample opening and closing you can look at.

A standalone review article does need an introduction close to CARS, as it must explain why a review is needed at this point. For systematic reviews, the PRISMA 2020 checklist wants two things in the introduction. The first is the rationale for the review in the context of existing knowledge (item 3). The second is a clear statement of the objectives or questions the review addresses (item 4).

In a research proposal also the three moves stay, only two things are different. In Move 1 the evidence must show that a problem exists, and showing that the topic is interesting is not enough. Move 3 is written in future tense. How the introduction fits into a two-to-four-page research proposal is there in our proposal guide. When you write a synopsis for your DC or RAC, this same opening turns into the background and the statement of the problem.

How is chapter 1 of a thesis different from a paper introduction?

The size is different and so is the reader. Someone reading a journal decides within two minutes whether to continue. An examiner goes through chapter 1 slowly and later checks each chapter against what chapter 1 promised.

That is why chapter 1 has parts a paper never needs, like significance, scope, definitions, a chapter plan and more. Our guide to PhD thesis structure gives the usual order of these parts. The most difficult one, the statement of the problem, has a separate guide.

You will still find the CARS moves in chapter 1, only spread over more pages. The background is Move 1. The problem statement and the gap make up Move 2. Then the objectives, scope and chapter plan are Move 3.

Many scholars are pasting the old synopsis as chapter 1 and that is a mistake. Your work has changed a lot since the first year. Keep the core problem as it is, but update the evidence and the tense, and do the final revision of chapter 1 at the very end. I’ve seen too many theses where chapter 1 gives five objectives and the results chapters answer only four.