A literature review matrix (also called a synthesis matrix or evidence table) is a grid with one row per source and one column per theme. You fill it in while reading; you write from it afterwards. Read a thematic column top to bottom and the marks — supports, contradicts, extends — plus your notes are already the evidence for one body paragraph of the review.
What a literature review matrix is, and where it sits
Every research paper that reviews literature has a hidden middle stage. You read twelve sources. Then, at some point, the notes have to become an argument. Most students try to jump that gap from memory, and the result is the summary parade: one paragraph per author, in the order the PDFs were opened. A literature review matrix is the missing bridge. It forces two decisions while you read — what each source actually found, and which recurring ideas the sources share — so that the writing stage starts with themes instead of a stack of papers.
The geometry is worth stating precisely, because the whole tool depends on it. Rows are sources; columns are themes. One row holds one source: its citation, its year, its method, its key finding. One column holds one idea that several sources speak to — say, goal setting or environment and distraction. Where a row and a column meet, you record what that one source says about that one theme. Synthesis lives in the cells; summary lives in the rows; the gap shows up as empty space. That is the difference between synthesis vs summary, compressed into a table.
If you have already read the full workflow in how to write a literature review, this page zooms in on the matrix itself: how to design the columns, how many sources a theme needs, what to do when sources flat-out disagree, and how to convert a filled cell into a sentence. If your assignment is the per-source artifact that comes before all this — summary and evaluation, no themes required — that is an annotated bibliography, and the rules live in how to write an annotated bibliography.
A literature review table template you can copy
Column design is the only hard part, so steal this starting layout. A workable literature review table template has two kinds of columns and rows: fixed identity columns that every row completes, and thematic columns that depend on your topic. The identity columns rarely change between projects:
| Column | What goes in it | Why it earns its place |
|---|---|---|
| Source / citation | Author surname and a two-word hint at the study | Feeds the in-text citations you will write later |
| Year | Publication year, always, even for one-author rows | Trends, replications and stale findings stay visible |
| Method | Design and sample in a phrase — RCT, 214 students | Later, contradictions are usually explained by method or sample |
| Key finding | The result in your own words, one sentence max | This is the raw material of your prose; quotes are optional |
| Theme columns (2–4) | One recurring idea per column | Each column becomes one body paragraph |
| Gaps | What the source leaves open, unmeasured, untested | The last sentence of your review lives here |
Three design rules keep the template honest. First, cap thematic columns at four: if everything is a theme, nothing is. Second, a theme name has to predict what a cell under it can say — goal setting passes that test, motivation swallows the whole field. Third, keep the year column even when the citation already holds the year: it is how you notice that three of your four sources predate 2020, which is either a strength or your gap. Keep author-year formatting consistent from row one; the APA citation generator handles APA 7 formatting if that is your style.
Build yours: the free synthesis matrix builder
The builder below is the template with the busywork removed. Add a row per source (four fields), set two to four theme columns, then walk each row and mark each cell: supports (the finding backs the theme), contradicts (the finding cuts against it) or extends (it adds a condition or boundary). Everything stays on your device — the page saves your matrix in this browser as you type, so closing the tab loses nothing.
Synthesis matrix builder
One row per source · thematic columns · export as text
Theme columns (2–4)
Want the gaps built in? Name your last column Gaps and record what each source leaves open.
Source rows
No sources yet. Add a row — or load the example below to see what a finished matrix feels like.
Example rows are invented for illustration — swap in your own sources before you write a word of the review.
Live table view
A synthesis matrix example, filled in
Below is the same matrix the Load example matrix button pre-fills, so you can study it even on a phone. The topic is study-habit interventions, and — to be clear — all four sources are invented for illustration. Notice what makes it a synthesis matrix example rather than a reading log: the three thematic columns cut across all four rows, marks disagree on purpose, and every note is a fact the source would recognize as its own.
| Source | Year | Method | Key finding | Goal setting | Environment | Timing | Gaps |
|---|---|---|---|---|---|---|---|
| Alvarez — review planner | 2019 | RCT, 214 undergraduates | Planner group kept up weekly review; controls crammed | supportsplanner turns goals into scheduled sessions | — | supportsspaced sessions beat night-before cramming | No measure of stress |
| Okafor & Ime — phone-free halls | 2021 | Quasi-experiment, 3 schools | Phone-free study halls lifted focus for distracted students | — | supportsthe room did the work willpower could not | contradictsfixed blocks beat student-chosen timing | One country, two terms |
| Lindqvist — goal diaries | 2022 | Diary study, 58 first-years | Specific daily goals predicted longer sessions; effect faded without feedback | supportsspecificity is what matters, not ambition | — | extendstiming only works when feedback follows | Self-reported data |
| Moreno & Chen — timer app | 2024 | Mixed methods, 97 students | Pomodoro timers helped half the sample, hurt anxious students | — | extendsa timer works as an artificial deadline | contradictsself-timed beats fixed for anxious students | Why timers backfire is unclear |
Read the Timing column down and watch it happen: Alvarez supports spacing, Okafor contradicts it, Lindqvist adds a condition, Moreno contradicts from the other side. That column is not just marked — it is already an argument with a shape. Two paragraphs of the finished review are sitting in it, and the contradiction in the middle is where a marker looks for your best page.
How to organize sources for a literature review
Organizing sources is where most reviews go wrong, because the obvious strategies are all traps: alphabetical order produces a list, chronological order produces a timeline, and everything-since-2019 produces a heap. The matrix organizes by idea instead. Here is the sequence that works when you already have a folder of PDFs:
Skim for findings, not for facts
Abstract, results, limitation section — ten minutes per source. You are hunting for the key finding and the method, nothing else yet.
Log each source as one row
Citation, year, method, key finding. If a row takes longer than five minutes to complete, you are reading the whole paper again — stop.
Let the columns emerge from repetition
After six rows, some ideas keep reappearing in the key findings. Those repeated ideas — never the author names — become your thematic columns.
Mark every cell, including the blanks
Supports, contradicts or extends. An untouched column after ten rows is not a failure; it is a theme your sources do not care about, or your gap.
How many sources per theme is the question every student asks next. The honest answer: enough to show a pattern, which in practice means at least two — and a paragraph gets easier to defend at three. Beyond that, quality beats padding; four sources that genuinely engage a theme outweigh nine that mention it in a limitation. If rows came from a messy export, tidy the author-year details with the bibliography generator first, then paste them in. Where the review sits inside the larger draft is covered in how to write a research paper.
The summary trap, and the synthesis fix
Here is the same four-source set written two ways. The first version reads like every draft written straight from a folder of PDFs; the second was drafted straight down the columns of the matrix above. Flip between them and watch where the organizing idea moves.
Several studies have examined study habits. Alvarez (2019) conducted a randomized controlled trial with 214 undergraduates and found that a spaced-review planner helped students keep up with weekly reading. Okafor and Ime (2021) studied three secondary schools and reported that phone-free study halls improved focus for distracted students. Lindqvist (2022) ran a diary study in which specific daily goals predicted longer study sessions. Moreno and Chen (2024) tested a timer app and found mixed results. Overall, these studies show that study habits can be improved in various ways.
- The year is the organizing principle — the paragraph is a timeline of reading, not an argument.
- Sources never meet — each sentence contains exactly one citation, so no cell ever talks to another.
- The closer is empty — "in various ways" could follow any four sources on any topic; no gap, no stake.
Across the four studies, two levers for better study habits recur: what students plan and where they study. On planning, Lindqvist (2022) found that specific daily goals predicted longer sessions, and Alvarez (2019) showed that a planner which schedules those goals keeps the habit alive past the second week. Environment works differently: Okafor and Ime (2021) found that phone-free halls carried students who lacked self-control, while Moreno and Chen (2024) found timers helped focused students but backfired for anxious ones. The contradiction turns on who is being timed — fixed structures seem to rescue the distracted and squeeze the anxious. None of the four studies measured well-being alongside grades, which is the gap this review fills.
- Themes lead, sources follow — the paragraph opens on the column, not on a single author.
- Cells combine into sentences — two citations inside one sentence is the sound of synthesis happening.
- The contradiction is named, then used — the Okafor–Moreno split becomes the interesting part, and the gap ends the paragraph with a job to do.
Matrix, literature comparison chart or evidence table?
Students arrive with the same tool under three names and one tool they actually need to distinguish. A literature comparison chart is usually the matrix under a friendlier name — rows and columns comparing sources on shared points. The quick map:
| Artifact | Shape | Use it when |
|---|---|---|
| Synthesis matrix | Rows = sources, columns = themes | Any thematic review; the default for coursework |
| Literature comparison chart | Rows = sources, columns = features you choose | Comparing products, policies or frameworks feature by feature |
| Evidence table | Matrix plus level-of-evidence and outcome columns | Nursing and health reviews where design quality must be graded |
| Annotated bibliography | No grid — one entry per source, in sequence | When the brief wants proof of reading, not synthesis |
If your brief says "comparison table" or "evidence table", build the matrix first anyway — the themes column set transfers, and the extra columns (sample size, outcome, level of evidence) bolt on without redesign. For nursing in particular, the worked example in how to write a nursing research paper shows the evidence-table variant in its natural habitat.
Thematic synthesis: from matrix cells to review paragraphs
The move from filled matrix to written review is smaller than it feels, because each column already contains one paragraph’s worth of evidence, ordered or not. The conversion has three moves. Open the paragraph on the theme: "Timing interventions split the sample" — never "Alvarez (2019) conducted…". Turn cells into sentences: a supports and an extends pair naturally combine, because one finding and its boundary condition belong together. Land on the contradiction or the blank: a column of three supports and one contradicts ends its paragraph on the contradicts cell; a column of blanks ends on the gap.
Contradictory findings deserve more than a footnote, because the instinct is to bury them. When two cells in a column disagree, resist declaring a winner. Write the disagreement as a function of the row attributes you recorded: sample, method, setting, year. Okafor’s fixed blocks worked on teenagers in supervised halls; Moreno’s self-timed students were undergraduates studying alone. Same column, different populations — which is precisely the sentence a literature review exists to say out loud. A contradiction explained by method is a finding; a contradiction ignored is a marked-down draft.
Once the paragraphs exist in skeleton form, the rest of the paper is plumbing. Each theme paragraph maps onto one section of the research paper outline, and the gap you exported in the Gaps column is the raw material for the last sentences before your research paper introduction states its thesis. Drafting order from here: gap sentence, then thesis, then the sections your outline already promised.
Four matrix mistakes, flipped
Each card below is a failure mode worth catching early. Tap one to see the fix.
Columns that mirror the sources
Tap to flip
The fix
If each column is an author, you built a reading list, not a matrix. Columns are ideas the sources share; rows are the sources. One author per column means zero synthesis is possible.
No year column
Tap to flip
The fix
Without years you cannot see whether a finding is fresh, replicated or stale — and your gap may already have been filled by a 2025 paper you filed under "maybe". One narrow column, expensive to skip.
Themes too broad
Tap to flip
The fix
"Motivation" fits every cell in every row, so it sorts nothing. Cut the name until it predicts what a cell under it can say: "goal setting", not "motivation"; "supervision", not "support".
Exporting without a gaps column
Tap to flip
The fix
The gap is the sentence the whole review is graded on. Keep a Gaps column — or a Gaps row — so the blank space in your matrix becomes a claim instead of a shrug.
Is the matrix ready to write from?
Six checks before you open a blank document. If any fails, it is a five-minute fix now and a restructure next week.
- Every row carries a citation — author and year — that can be pasted straight into the reference list.
- Each thematic column has at least two sources marked in it.
- Contradictory findings are marked as contradictions, not smoothed into agreement.
- Cell notes are the source’s facts in my words — no opinions, no copied sentences.
- A Gaps column records what the sources leave open, unmeasured or untested.
- Every theme column maps to one planned paragraph in the review outline.
Matrix full, draft still blank?
A finished matrix is half the review; the writing is the other half, and it has a deadline. Hand your exported matrix to a quoted, deadline-bound writer and get a draft back you can edit in your own voice.
Five-question matrix check
Wrong answers cost nothing and teach the most. One of these five questions has appeared on a real marking rubric near you.
1. What separates a synthesis matrix from an annotated bibliography?
2. In a synthesis matrix, what does one column represent?
3. Two sources in the same theme column disagree. Your move?
4. Why does the template insist on a year column?
5. How many sources does a theme need before you draft its paragraph?
Frequently asked questions
What is a literature review matrix in one sentence?
A grid with one row per source and one column per theme, filled in while you read, so that each column can later be read top to bottom as the evidence for one paragraph of the review.
Annotated bibliography vs matrix — which should I build first?
If the assignment asks for an annotated bibliography, that is the deliverable — build it, and treat any matrix as private scaffolding. If the assignment is the review itself, start the matrix immediately: the annotated bibliography proves you read; the matrix proves you thought. Many students keep both, since the bibliography’s evaluative notes enrich the matrix notes.
How many sources belong in the matrix?
As many as the review genuinely cites — typically eight to twenty for a coursework review, but the matrix scales either way. The columns and rows grow independently: adding a source never forces a new column, and discovering a theme never forces new reading. Start with the six sources you are sure you will cite and let the structure pull in the rest.
Can I build this table in Excel or Google Sheets instead?
Yes — the export button produces formatted plain text that pastes cleanly, and the same columns and rows work in any spreadsheet. The builder’s advantage is the supports / contradicts / extends marking and the contradiction counter, which a plain sheet will not do for you. Whatever the surface, the discipline is identical: rows are sources, columns are themes.
What do I do with two studies that contradict each other?
Keep both cells, mark both honestly, and let the row attributes explain the split: different samples, methods, settings or years. Then give the contradiction a job — it is the strongest candidate for your gap paragraph, because a disagreement between credible studies is exactly where new research earns its place. Never average findings or silently drop the inconvenient study.
Was this guide useful?