Research Paper Introduction Example
Three complete research paper introductions with every sentence labelled, the four-move structure they share, and the openings that cost marks.
Why Introductions Stall
You know the introduction needs a hook, some background and a thesis. What you cannot tell from that list is how specific the hook has to be, where the background stops, or why your draft reads like a description of the topic instead of a reason for the paper.
Four Moves, In Order
Three complete introductions below — social sciences, STEM and humanities — with every sentence labelled by the move it performs, so you can hold yours against a worked version in your own field rather than against a list of abstract advice.
How it works
- Draft the thesis first, after your analysis exists
- Name the gap that makes the thesis worth stating
- Work backwards to two to four sentences of background
- Write the hook last, on a specific citable fact in your subfield
- Reread to check every sentence is narrower than the one before it
The four moves, in order
A research paper introduction narrows. It starts with something a reader already accepts, arrives at a question nobody has answered, and ends by telling them what this paper does about it. Four moves, almost always in this order.
| Move | Its job | Length |
|---|---|---|
| 1. Hook | Establish that the topic matters, in terms specific to your field | 1–2 sentences |
| 2. Background | Give only the context needed to understand the gap | 2–4 sentences |
| 3. Gap | Name what is unknown, contested or untested — this is the hinge | 1–2 sentences |
| 4. Thesis or question | State what you argue, ask or hypothesise, and often how | 1–2 sentences |
The gap is the move students skip, and skipping it is why an introduction reads like a summary of the topic rather than a reason for the paper to exist. If a reader finishes your introduction and could not say what was missing from the literature, the gap is not there.
Example 1 — social sciences
An empirical paper on social media and adolescent sleep. Each sentence is labelled with the move it performs.
Adolescents in high-income countries now sleep roughly 40 minutes less per night than their counterparts did in 1990, a decline that tracks almost exactly with the spread of smartphone ownership. [Hook]
Researchers have established a reliable correlation between evening screen exposure and delayed sleep onset, attributing it both to blue-light suppression of melatonin and to the cognitive arousal of social interaction. Most of this work, however, treats "screen time" as a single aggregate measure, counting a messaging app and a video feed as equivalent exposures. [Background]
Whether different categories of social media use carry different sleep costs remains largely untested, despite the obvious practical implications for guidance given to parents and schools. [Gap]
This study uses seven-day actigraphy and app-level usage logs from 214 participants aged 14 to 17 to compare the effects of passive consumption and active communication on sleep onset latency. We hypothesise that passive scrolling delays sleep onset more than equivalent time spent in direct messaging. [Thesis / hypothesis]
Note what the hook is not: it is not "Social media has become an important part of modern life." It is a specific, quantified claim that a reader in the field would accept and that already points at the variable under study.
Example 2 — STEM
A materials science paper. The moves are identical; the register is flatter and the gap is a technical limitation rather than a conceptual one.
Solid-state electrolytes promise lithium batteries that cannot catch fire, but every candidate material to reach commercial testing has failed on the same metric: interfacial resistance at the anode. [Hook]
Garnet-type oxides such as LLZO offer the ionic conductivity required, and sintering routes for dense pellets are now well characterised. Reported interfacial resistances, however, span three orders of magnitude across laboratories using nominally identical preparations. [Background]
The source of that variance has not been isolated, which makes it impossible to tell whether the remaining barrier is intrinsic to the material or an artefact of surface preparation. [Gap]
Here we quantify the effect of ambient exposure time on LLZO surface chemistry using XPS and impedance spectroscopy across controlled exposure intervals, and show that lithium carbonate formation accounts for the majority of reported variance. [Thesis]
The STEM convention
Empirical papers usually end the introduction with "Here we show…" or "This paper demonstrates…" and state the finding outright. Withholding your result for suspense is a humanities habit that reads as evasive in a sciences paper.
Example 3 — humanities
A literature paper. There is no hypothesis, so the fourth move is an interpretive claim — but it is still a claim someone could disagree with, which is what makes it a thesis.
The house at 124 Bluestone Road is introduced in the first line of Beloved as "spiteful," a judgement delivered before any character appears to make it. [Hook]
Critics have read the novel's hauntings largely as figures for the unassimilable trauma of slavery, following Morrison's own account of the book as a response to an absent archive. This reading treats the supernatural as metaphor, a vehicle for historical memory that the realist mode cannot carry. [Background]
Less attention has been paid to the novel's narrative withholding — the way it repeatedly declines to narrate the events it circles, and attributes that refusal to the characters rather than to the text. [Gap]
I argue that Beloved treats omission itself as a form of testimony, and that the novel's unnarrated centre is not a gap in memory but an assertion about what memory is permitted to do. [Thesis]
A humanities thesis has to be arguable. "Beloved explores memory and trauma" is a topic, not a thesis — nobody would contest it. The version above takes a position another critic could reject, which is the test.
Write the thesis first, the hook last
Introductions are read first and should be written nearly last. The reason is mechanical: until you know what you concluded, you cannot write a sentence that promises it.
- Draft the thesis. One sentence, written after your analysis exists, saying what you found or argue.
- Name the gap that makes the thesis worth stating. If you cannot, the thesis is probably a summary.
- Work backwards to the background — include only what a reader needs to see the gap. Two to four sentences, not a literature review.
- Write the hook last. It is the hardest sentence and the easiest to write once you know exactly where it has to lead.
- Reread for narrowing. Each sentence should be more specific than the one before it. If any sentence widens, cut it.
The introduction you drafted first is usually the wrong one
If you wrote your introduction before the paper, go back and check whether it still promises what the paper delivers. A mismatch between the thesis in the introduction and the argument in the body is one of the most commonly flagged problems in marked work.
Openings that lose marks
The dawn of time. "Throughout history, humans have always…" — unfalsifiable, unspecific, and it tells a marker you had nothing to open with.
Open on a specific, citable fact inside your subfield that already points at your variable.
The dictionary. "Merriam-Webster defines identity as…" — a lexical definition is not an argument and signals no engagement with the field.
If a term genuinely needs defining, define it the way your discipline does, with a citation to whoever defined it that way.
The announcement. "In this essay I will discuss three things about X." A list of topics is not a thesis, because it states no position.
Say what you conclude, not what you will cover. Signposting the structure can follow in one further sentence.
- No rhetorical question as a hook. "What if everything we know about memory is wrong?" is a stock AI opening and markers now read it as one.
- No full literature review. Background is two to four sentences; the review is its own section.
- No results in a humanities introduction and no suspense in a sciences one — the conventions are opposite.
- No citations you have not read. Introductions attract fabricated references because they are the densest citation point in the paper.
How long should it be
Roughly ten per cent of the paper, with the proportion shrinking as the paper grows.
| Paper length | Introduction | In practice |
|---|---|---|
| 1,000 words | 100–150 words | One tight paragraph |
| 2,500 words | 200–300 words | One or two paragraphs |
| 5,000 words | 400–500 words | Two to three paragraphs |
| 8,000+ words | 600–800 words | Separate paragraph per move |
| Dissertation chapter | 5–8% of the chapter | Often its own numbered section |
If your introduction is running long, the usual cause is that the background has turned into a literature review. Move it. The introduction needs only enough context for the gap to be visible.
Want the method, not just the tool?
How to Write an Essay Introduction (With Hooks That Work) — The four moves every strong introduction makes, six hooks with real examples, how to tell a thesis from a statement of fact — and why you should write it second.
Try it on your own document
Write, cite, humanize and check your AI score in one editor. The free plan includes limited AI chat and AI detection each day, plus three humanizer runs — no card needed.
Open the editor →