SciSpace Academy
Tutorial

How to Conduct a Systematic Literature Review with SciSpace Agent

Dr. Jaroslaw (Jarek) KriukowEnglishSystematic review

TL;DR

1-min read

In this tutorial, Dr. Jaroslaw (Jarek) Kriukow, a research consultant, data analyst and academic tutor who is also an editorial team member at Polaris Global Journal of Scholarly Research and Trends, shares how he runs a systematic literature review using SciSpace Agent. It is aimed at researchers and students preparing their first or next SLR who know the theory but want to see the process in practice.

Key takeaways

  • Attendees follow a working researcher's real workflow from start to finish, rather than a generic checklist, and see how he structures each stage of the review.
  • A typical SLR moves from a focused question and protocol with inclusion and exclusion criteria to database searches, screening, data extraction, quality assessment and synthesis, often reported using PRISMA.
  • The tutorial shows where SciSpace Agent fits into each stage of that workflow and how it supports searching, screening, extraction and summarising.
  • Practical decisions that generic checklists often skip are made visible, helping you anticipate the judgement calls in your own review.
  • AI support can make a systematic review more manageable without lowering methodological standards, provided humans verify screening and extraction and the process is documented transparently.
  • Lessons from teaching research methods offer guidance on planning your own review so that it stays reproducible and credible.
Apply it

To apply this, draft your review protocol first, then use SciSpace Agent stage by stage while documenting how AI was used and verifying each screening and extraction decision.

How to do it

  1. Structure your SLR from the start

    Plan each stage of your systematic literature review before you begin. A clear structure keeps a systematic literature review with AI manageable.

  2. Use SciSpace Agent at each stage

    Decide where SciSpace Agent fits into each stage of your SLR workflow and apply it there. This saves time on the heavy lifting.

  3. Document your practical decisions

    Record the practical decisions you make that generic checklists skip. This makes your review transparent and defensible.

  4. Maintain methodological rigour

    Check the agent's outputs against your protocol at every stage. Rigour matters more than speed in an SLR.

  5. Apply research-methods lessons

    Use lessons from research-methods teaching to plan your own review. This helps you avoid common planning mistakes.

Try this on SciSpace

Frequently asked questions

What are the main steps of a systematic literature review?

A typical SLR involves defining a focused research question, writing a protocol with inclusion and exclusion criteria, searching multiple databases, screening titles, abstracts and full texts, extracting data, assessing study quality and synthesising findings. Reporting usually follows a framework such as PRISMA so that others can see exactly how studies were found and selected.

Can AI be used in a systematic review?

Yes, AI is increasingly used to support searching, screening, data extraction and summarising. To keep the review credible, document how AI was used, have humans verify screening and extraction decisions, and report the process transparently. AI should speed up the work, not replace the systematic, reproducible methods that define an SLR.

How is a systematic review different from a literature review?

A traditional literature review gives a broad, often selective overview of a topic. A systematic review follows a pre-defined, reproducible method to find and appraise all relevant studies on a specific question, which reduces bias. It takes longer but provides stronger evidence for conclusions and is often expected for evidence-based fields.

Related tutorials