How to Conduct a Systematic Literature Review Using NVivo [Step-by-Step]

Last updated on May 14th, 2026 at 05:22 pm

Estimated reading time: 10 minutes

Are you drowning in a hundred scholarly articles and not sure how to extract the data you need? You’re not alone. Systematic literature reviews are one of the most time-consuming parts of any PhD — but NVivo can cut that time significantly.

In this guide, I’ll walk you through exactly how to conduct a systematic literature review using NVivo, from fetching articles in academic databases all the way to exporting a clean framework matrix. I’m using the topic effects of social media on teenagers as a worked example throughout.

If you’re new to NVivo, start with my guide on how to do thematic analysis in NVivo 15 before continuing here.

A systematic literature review is a structured method for identifying, screening, and synthesising all relevant research on a specific question. Unlike a narrative review, it follows a transparent, reproducible process — typically guided by the PRISMA framework, the international standard for reporting systematic reviews.

Most researchers follow eight main steps:

1. Formulate a research question
2. Develop a review protocol
3. Conduct systematic searches
4. Screen and select studies
5. Assess study quality
6. Extract data
7. Synthesise and analyse findings
8. Interpret findings and write the review

This article picks up at Step 6 — data extraction — and shows you how NVivo makes that process faster and more organised than working manually in spreadsheets.

Before you open NVivo, make sure you have:

• A defined research question and inclusion/exclusion criteria
• A list of relevant articles from your database searches (PubMed, Scopus, Web of Science, etc.)
• PDF copies of the full articles
• Zotero installed (free — zotero.org)
• NVivo 14 or NVivo 15 (both work for everything in this guide)

Step 1 – Search and Fetch Articles from Online Databases

Start by searching your topic in an academic database. For the topic effects of social media on teenagers, I searched PubMed — one of the most widely used databases for health and social science research.

Run your search using your defined keywords and apply your inclusion/exclusion criteria to filter the results. Download or save the full-text PDFs for the articles that pass your screening.

A note on accessing full-text articles: If you can’t access a PDF through your institution, try ResearchGate, Unpaywall, or Google Scholar’s ‘All versions’ links. You can also email the corresponding author directly — most researchers are happy to share their work.

If you’re unsure which databases to search or how to build a search string, Scribbr’s systematic review guide walks through the process clearly.

The use of social media modifies teenagers’ sleep related behaviour

Step 2 – Import Your Articles into Zotero

Once you have your articles, don’t import PDFs directly into NVivo. Instead, use Zotero as an intermediary — it organises your references and captures all the metadata (title, author, year, DOI, journal) that NVivo will use later.

Here’s how:

1. Open Zotero and create a new collection named after your topic (e.g. Effects of Social Media on Teenagers)
2. Drag and drop your PDFs into that collection
3. Zotero automatically fetches metadata for each article — title, authors, year, DOI, and more
4. Wait for Zotero to finish processing before moving on

This step is worth doing properly. The metadata Zotero pulls in will populate NVivo’s classification sheet and framework matrix, saving you hours of manual entry later.

Step 3 – Export an RIS File and Import It into NVivo

NVivo reads references in RIS format. Here’s how to get your Zotero collection into NVivo:

1. In Zotero, right-click your collection and select Export Collection
2. Choose RIS format and click OK — save the file somewhere easy to find
3. Open NVivo and create a new project (e.g. Systematic Review — Social Media)
4. In NVivo, go to Import → Bibliography
5. Browse to your RIS file and click Import

Let’s create a new project then call this systematic review.

NVivo will import all your articles along with their full metadata. You’ll see them in the Files section, and their reference details will appear in the classification sheet automatically.

NVivo save tip: Always use manual save in NVivo. Auto-save can corrupt your project. Set a reminder to save every 15 minutes.

For more detail on how NVivo handles qualitative data imports, see the official NVivo documentation from Lumivero.

Step 4 – Create a Query Protocol in NVivo

Before you start extracting data, you need to know what you’re looking for. A query protocol is a simple list of the key concepts, themes, or questions you want to find evidence for across your articles.

For the social media and teenagers topic, a query protocol might include:

• Methods / study type (quantitative, qualitative, mixed methods)
• Depression
• Anxiety
• Sleep behaviour
• Recommendations
• Proposals for future research

Think of each item in your protocol as a search keyword for NVivo’s text-search query tool. Add all keywords and synonyms you want to explore — the more thorough your protocol, the more comprehensive your data extraction will be.

.

Step 7 – Export a Framework Matrix

Once you’ve coded data from across your articles, you can export a framework matrix — a table that shows what each article said about each theme you coded.

1. In NVivo, go to Create → Framework Matrix
2. In the Rows, select all your imported articles
3. In the Columns, select your codes (e.g. Depression, Study Type, Recommendations)
4. Click Auto-Summarise to populate the cells
5. Right-click the matrix and select Export to save it as an Excel file

 So I get this kind of a table click on the auto summarize. 

Step 8 – Clean Up Metadata and Visualise Your Findings

After exporting the framework matrix, go back to NVivo to clean your classification sheet and create visual summaries of your literature base.

To visualise metadata:

• Go to the file classification and right-click on a metadata field (e.g. Year)
• Select Visualise to generate a chart
• NVivo can produce bar charts and pie charts of your article metadata

To clean the metadata:

• Remove fields you don’t need (e.g. Secondary Author, Edition) by right-clicking and deleting them
• Keep fields like Title, Year, DOI, Journal, Authors

To export the classification sheet:

• Press Ctrl+Shift+E to export the sheet as an Excel file
• This gives you a complete metadata record for all articles in your review — useful as an appendix in your dissertation

Can I use NVivo 14 instead of NVivo 15 for a systematic review?

Yes. Everything covered in this guide works in both NVivo 14 and NVivo 15.

Do I have to use Zotero? Can I use Mendeley instead?

Zotero is recommended because it exports RIS files cleanly and handles metadata reliably. Mendeley can also export RIS files, so it will work — but Zotero is free and integrates more smoothly with NVivo in my experience.

How many articles can NVivo handle for a systematic review?

NVivo can handle large projects with 100+ documents. Performance depends on your computer’s RAM, but most PhD-level systematic reviews (20-80 articles) work smoothly.

What is a framework matrix and why does it matter?

A framework matrix is a table in NVivo that cross-references your articles (rows) against your themes or codes (columns). It’s the key output of your data extraction process — it shows you what each article contributed to each theme and helps you synthesise findings across your literature.

Do I need to read all my articles before using text-search queries?

No — that’s the point. NVivo’s text-search query tool lets you search across all your articles at once, so you can locate relevant passages without reading every article in full. You still need to read the highlighted passages in context before coding them.

Key Takeaways

• Import articles into Zotero first, then export an RIS file into NVivo — don’t import PDFs manually
• Build a query protocol before you start — it keeps your extraction focused and consistent
• Use text-search queries in NVivo to locate relevant passages across all articles at once
• Code your extractions to named containers so you can synthesise findings by theme
• Export a framework matrix to get a comparison table across all your articles
• Visualise and export your classification sheet for a clean metadata record for your dissertation

Need Help With Your Systematic Review?

If you’re stuck on data extraction or working to a tight deadline, I offer a done-for-you qualitative data analysis service for PhD students. You send me your files and your research question — I handle the analysis and deliver structured findings you can write up immediately.

You might also find these posts useful as you move through your review:

• qualitative coding in NVivo — if you’re still building your coding skills
• inductive thematic analysis in NVivo — for when your review includes qualitative primary studies
• how to do thematic analysis in NVivo 15 — the full step-by-step NVivo workflow from data import to final themes

 

.