The instructions for a research essay are usually short: pick a topic, use sources, cite them, argue something. What the instructions leave out is the sequence, and the sequence is where most of the trouble starts. Students who write the thesis before the reading end up bending sources to fit it. Students who read without a question end up with forty pages of notes and no argument. This page lays out the stages in the order that works, and follows one topic from a vague idea to a finished paragraph so you can see what each stage produces.
The example topic is community composting programs and household food waste. It is invented for the purpose; the sources referred to are placeholders and any numbers are illustrative. The point is to watch the method, not to learn about composting.
Stage 1: From topic to question
A topic is a noun phrase. A question is something a reader could disagree with the answer to. Getting from one to the other takes three narrowing moves.
| Move | Example |
|---|---|
| Topic | Food waste |
| Aspect | Household food waste and municipal composting |
| Question | Do curbside composting programs reduce the amount of food households throw away, or do they only redirect the same waste into a different bin? |
The final question is answerable within an essay because it has a specific outcome (amount of food wasted), a specific intervention (curbside composting), and a built-in tension (reduction versus redirection). It also predicts the shape of the research: you will need sources that measured waste before and after programs began, and sources that explain why households change or do not change what they buy and cook.
If the question could be settled by looking something up, it is too small. If the question needs a decade of data you do not have, it is too large. For a fuller method of narrowing, How to Choose a Research Paper Topic You Can Actually Finish works through more examples.
If your paper needs a summary on top, read how to write an abstract.
Stage 2: Finding and vetting sources
Search with the question’s own words first, then with the words specialists use. For the example, a first search on “composting food waste households” returns mostly how-to guides; a second search on “curbside organics collection” and “food waste reduction behavior” reaches the studies. Note both sets of terms; you will need them again.
For each candidate source, run the five-question check before reading it fully.
- Who wrote it, and what is their claim to know? A municipal waste department, a university research group, and an advocacy organization all have knowledge, but they also have different reasons to publish.
- When was it published, and does the date matter for this question? For a program-evaluation question, a study from before widespread curbside collection is background, not evidence.
- What kind of document is it? Peer-reviewed article, government report, news feature, blog post, industry white paper. The type sets how much weight it can carry.
- What did they actually measure or observe? A study that weighed bins is not the same as a study that asked people how much they waste.
- Does it cite its own sources? A document that makes claims without references is an opinion, whatever letterhead it sits on.
A workable source set for a 2,000-word essay on the example might look like this:
| Source (illustrative) | Type | Role in the essay |
|---|---|---|
| Author A (2020) | Peer-reviewed study weighing household bins before and after a program | Main evidence |
| Author B (2021) | Peer-reviewed survey of household food-purchasing habits | Main evidence on behavior |
| City waste department report (2022) | Government report on program tonnage | Background, with caution about what tonnage can and cannot show |
| Author C (2019) | Peer-reviewed study finding no change in purchasing after a program | Counter-evidence |
| Author D (2023) | Review article on waste-reduction interventions | Definitions and framing |
Five sources, two of them carrying the argument, one opposing, two supporting the frame. That balance is what “well researched” looks like at essay length. The department report is included with a note on its limits: it measures what went into the compost bin, which says nothing by itself about whether total waste fell. Noticing that limit early is what keeps Stage 4 honest.
Stage 3: Notes you can rearrange
Take notes in units that can be moved. A page of continuous notes on a source is hard to reuse; a set of separate cards, each holding one idea, can be sorted into an outline. Digital or paper does not matter. The format does.
Each note card carries four things:
Source + page: Author A (2020), p. 14
Idea in my words: Households in the pilot area threw out about a tenth less
food by weight after one year; the drop was larger in
households that attended a kitchen workshop.
Quote (only if the wording matters): —
My reaction: Supports reduction, but the workshop confound is important —
was it the bin or the workshop?
The “idea in my words” line is doing two jobs. It forces you to understand the point well enough to restate it, and it means the sentence that ends up in the essay is already yours, which keeps paraphrase from drifting into copying. The “reaction” line is where the argument starts to form. Write it every time, even when the reaction is only “not sure this matters yet.”
By the end of Stage 3, the example topic has perhaps twenty cards. Sorting them into piles produces the first outline without anyone deciding to outline. Cards about measured reductions go in one pile, cards about purchasing behavior in another, cards about what tonnage figures hide in a third.
Stage 4: Writing the thesis from the evidence
The thesis is written now, after the reading, and it has to answer the Stage 1 question in one sentence that a reasonable person could dispute.
The piles from Stage 3 suggest an answer with a qualification: programs seem to reduce waste, but the reduction depends on something beyond the bin. That becomes:
Curbside composting programs reduce household food waste only when they are paired with instruction that changes how people shop and cook; the bin alone mostly redirects waste rather than preventing it.
Test the sentence three ways. Could someone disagree? Yes: Author C would, and a program manager who believes in bins would. Does it say what the body must prove? Yes: that reductions occurred, that they tracked instruction, and that bin-only programs show redirection. Does it use the sources you actually have? Yes, all five have a place.
A weaker thesis would be “Composting programs have both benefits and drawbacks.” Nobody disputes that, so nothing needs proving, so the essay has nowhere to go. For more on shaping the sentence, How to Write a Thesis Statement That Holds the Essay Together is the companion piece.
Stage 5: Outlining by claim
An outline by topic lists subjects (“Background on composting,” “Studies,” “Behavior”). An outline by claim lists the sentences each part of the essay will prove. The second kind can be checked against the thesis before a paragraph is drafted.
Intro
Context: growth of curbside programs; the assumption that diversion equals reduction
Question: reduce or redirect?
Thesis (as above)
1. Program tonnage figures show diversion, not reduction.
– City report (2022): what tonnage measures
– Author D (2023): why diversion and reduction are different outcomes
2. Where waste was weighed, reductions appeared mainly alongside instruction.
– Author A (2020): weighed bins, workshop effect
– Author B (2021): purchasing changed after education, not after bins
3. Bin-only programs show little change in what households buy.
– Author C (2019): no purchasing change after rollout
– Response: this supports the thesis rather than undermining it —
bins without instruction redirect
4. Why instruction changes the outcome
– Author B (2021) and Author D (2023): planning and portioning habits
– Implication for how programs should be designed
Conclusion
Restate the thesis in light of the evidence
What follows for cities running bin-only programs
Two features of this outline do most of the work. Each numbered part is a full sentence with a claim; if any of them could be deleted without weakening the thesis, it should be. And the opposing source is not quarantined in a “counterarguments” section at the end; it is placed where it does the most good, with the response written into the outline so the draft does not stumble there.
Stage 6: Drafting paragraphs
A research-essay body paragraph has a stable shape, and once it is familiar, drafting is mostly filling it in.
- Claim in your own words, one sentence.
- Evidence from a source, cited, paraphrased or briefly quoted.
- Explanation of what the evidence shows and why it supports the claim.
- Link forward or back to the thesis, sometimes merged with the explanation.
Here is part 2 of the outline drafted in that shape:
The clearest reductions in household food waste appear where a composting program was accompanied by instruction. In a study that weighed bins in a pilot district over a year, households discarded roughly a tenth less food by weight, but the decline was concentrated among those who had attended a short kitchen workshop (Author A, 2020). A separate survey found that participants changed their shopping habits after an education session and reported no comparable change after the bin alone arrived (Author B, 2021). Taken together, the two studies suggest that the bin creates an opportunity and the instruction creates the behavior; without the second, the first mostly changes where food waste goes rather than how much there is.
The claim opens the paragraph; the sources arrive second; the last sentence does the interpreting. Note that the paragraph paraphrases rather than quotes. Quotation is reserved for wording that cannot be improved or that you intend to analyze word by word. How to Use Quotes in an Essay Without Losing Your Own Voice covers when quoting earns its place.
Draft the numbered parts first, in whatever order feels easiest, then the introduction, then the conclusion. The introduction is easier to write when you know what it is introducing, and the conclusion should say something the introduction could not have said, because the evidence had not yet been shown.
Cite as you write. A draft full of “[source?]” placeholders is a draft with future errors, and the note cards make citing on the spot easy because the page number is already there.
Stage 7: Two revision passes
Revise in two separate passes, on different days if possible. Mixing them produces polished sentences in paragraphs that should have been cut.
Pass one: argument. Read only the first sentence of each paragraph. Together they should form a compressed version of the essay. If a first sentence is a quotation, a fact, or an author’s name, rewrite it as a claim. If two paragraphs make the same claim, merge them. If a paragraph makes no claim the thesis needs, cut it, however good the sentences are. Then check that the thesis in the introduction matches what the body proved; theses often need adjusting after the draft, and adjusting the thesis is easier than rewriting the body.
Pass two: paragraphs and sentences. For each paragraph: is there evidence, is it cited, is there an explanation sentence after the evidence? For each sentence: read it aloud; anything you stumble over needs shortening. Remove hedges that pile up (“it could perhaps be argued that”) and restatements that pad. Then check citation format against the style guide for every source type you used.
Proofreading comes after both passes. Polishing a sentence you will later delete is wasted work.
A checklist for the finished essay
- The question is stated in the introduction and answered in the thesis.
- The thesis could be disputed, and the body shows why it should not be.
- Every body paragraph opens with a claim in your voice.
- Every piece of evidence is followed by a sentence explaining what it shows.
- At least one source that complicates the thesis is treated fairly and answered.
- Sources used in the text match the reference list exactly.
- The conclusion says what follows from the argument, not just what the argument was.
- Read aloud once, start to finish.
What changes at larger scale
The stages above do not change for a longer paper; they lengthen. A 6,000-word term paper uses more sources, converts the numbered outline parts into headed sections, and spreads Stage 2 over several weeks rather than a few days. For that scale, Term Paper vs. Essay: Differences and a Semester Plan has a week-by-week version. For a dissertation, the same logic runs across a year, and How to Plan and Write a Dissertation Without Losing Your Way shows how the stages become chapters. The habits are the same at every size: read before you decide, write the claim first in every paragraph, and let the opposing source into the room.
Frequently asked questions
What is the difference between a research essay and a research paper?
In most courses the terms overlap. Where a distinction is drawn, a research essay is shorter, usually 1,500 to 3,000 words, keeps an essay’s flowing structure without numbered sections, and argues a position using a handful of sources. A research paper is longer, uses headings, and may report original data. The method on this page fits both; the essay just uses fewer sources and no headings.
How many sources does a research essay need?
Enough to show you have read around the question, not just found agreement. For a 2,000-word essay, five to eight sources is typical: two or three that carry the main evidence, one or two that complicate or oppose it, and a couple that supply background or definitions. If the assignment gives a minimum, treat it as the floor for sources actually used in the text, not for the reference list.
Can I use a source that disagrees with my thesis?
You should. A research essay that cites only sources on its own side reads as one-sided, and instructors notice. Give the opposing source a fair paragraph, state its strongest point, then explain why your position holds or how it needs to be qualified. Handled honestly, the disagreeing source is often the most persuasive part of the essay because it shows the reader you have tested your claim.
How do I keep the essay from turning into a summary of my sources?
Start every paragraph with a sentence in your own voice that makes a claim, then use the source to support it. If a paragraph opens with an author’s name or a quotation, the source is leading and you are following. A quick test on the finished draft: read only the first sentence of each paragraph. Together they should form a short version of your argument.