Introduction to Systematic Literature Reviews

Claudius Gräbner-Radkowitsch; Anna-Katharina Kothe

Introduction to Systematic Literature Reviews

literature-review

research-design

reproducibility

reading

Tell narrative reviews, systematic reviews and meta-analyses apart, then work through the systematic review process — research question (PICO), inclusion/exclusion criteria and coding, and a documented, reproducible search strategy.

Authors

Claudius Gräbner-Radkowitsch

Anna-Katharina Kothe

Published

June 17, 2026

Modified

June 17, 2026

1 Introduction

Research synthesis and literature review are a key step in almost every study. As Newton put it, “If I have seen further it is by standing on the shoulders of Giants.” Before you can add to a field, you need a reliable picture of what is already known — and, just as importantly, where the gaps are.

Research synthesis is “the practice of systematically distilling and integrating data from many studies in order to draw more reliable conclusions about a given research issue” (Cooper et al., 2019). Literature reviews are the central tool for this: they distil key insights from academic research for practitioners and identify research gaps. This page introduces the main kinds of review and then walks through the core stages of the most rigorous variant — the systematic review.

2 Three Key Types of Review

“Literature review” is an umbrella term. Three types matter most:

Narrative review — a non-systematic overview of a topic.
Systematic review — a systematic, transparent and replicable synthesis.
Meta-analysis — a statistical synthesis of quantitative studies.

They differ across several dimensions:

	Narrative review	Systematic review	Meta-analysis
Goal	Overview/synthesis of a topic	Systematically identify, evaluate and synthesise all relevant evidence on a specific question	Statistically combine quantitative results from multiple studies
Approach	Flexible, typically non-systematic selection	Rigorous, transparent and replicable; explicit inclusion/exclusion criteria	Statistical synthesis of studies with similar variables/outcomes
Scope	Selective; significant works	Comprehensive; aims to include all relevant studies	Selective; only studies with compatible quantitative data
Quality assessment	Limited/informal	Formal appraisal using explicit tools	Formal, rigorous appraisal of methodological and statistical compatibility
Transparency / replicability	Low	High	High
Time investment	Low–moderate	High	Very high
Typical question	What is known about leadership theories in organisations?	How effective are diversity-training programmes at reducing workplace discrimination?	How strong is the relationship between transformational leadership and employee performance?
Output	Thematic/chronological narrative	Structured findings with transparent documentation	Effect sizes and forest plots

The rest of this page focuses on the systematic review, the workhorse for evidence-based questions with enough empirical studies to synthesise. (The authoritative procedural standard is the PRISMA guidelines; treat what follows as an accessible on-ramp, not a replacement.)

3 The Systematic Review Process

3.1 1. Define the research question

Having a good question is the key to any successful research endeavour. A widely used scaffold is the PICO framework:

Element	Question	Example
P (Population)	Who is being studied?	middle managers, SMEs, multinationals
I (Intervention / Exposure)	What main variable or intervention is examined?	leadership training, flexible work arrangements
C (Comparison)	What is it compared to?	traditional management, fixed schedules
O (Outcome)	What consequences or results are measured?	productivity, employee satisfaction

Example question: “In multinational corporations (P), how does transformational leadership (I) compared to transactional leadership (C) affect employee innovation behaviour (O)?”

Avoid these common pitfalls: wordy expressions, ambiguous terms, merging several questions with “and”, suggestive (leading) questions, and questions that cannot actually be answered.

Note

PICO is strongest for clinical-style, intervention questions. Variants such as SPIDER, SPICE and ECLIPSE are often better suited to qualitative and mixed-methods research in the social sciences. Choose the scaffold that fits your question.

3.2 2. Define inclusion and exclusion criteria

Decide which studies are of potential interest for your question by asking:

which variables/topics are covered?
which kind of studies (quality, method, …)?
which locations and time periods?
which publication types (books, journals, newspapers)?
which samples (students, seniors, gender, …)?

The goal is to develop a coding scheme you (and others) can use to extract relevant information and decide whether a study is of interest. A simple scheme records, per study, fields like Study ID, publication year, main method, key topic, abstract-screening decision (keep/drop), full-text-screening decision, and role in the main narrative. Coding schemes can be developed deductively, inductively or iteratively; test the scheme regularly, write a short manual with examples, and make sure it aligns with your research interest.

3.3 3. Delineate a search strategy

Where to search?

Database selection: Web of Science, Scopus, Google Scholar, …
Additional sources: snowball sampling, citation networks, expert suggestions.

How to search?

which keywords to use (refer back to your inclusion/exclusion criteria)?
where to search for those keywords (e.g. the Topic field, which searches title, abstract and keywords)?
which papers to use for snowball / citation-based search?
whom to ask for additional papers?

In a database like Web of Science, keywords are combined with the Boolean operators AND, OR, NOT, and several sub-searches can be merged — each sub-search requiring its own rationale. Mind the caveats: every database has restrictions (Web of Science, for example, skews toward English-language articles in ranked journals, introducing publication biases), and depending on your question you may need to search in several languages.

Reproducibility is the whole point

Always document your decisions and justifications, and make your search reproducible — for you and for others. In practice this means recording your exact search strings, keeping a tool such as a spreadsheet plus a reference manager (e.g. Zotero) for documentation, being explicit about limitations, and reporting the number of records retrieved, de-duplicated, screened and finally included at each stage.

🎓 MA / PhD

At MA/PhD level a systematic review is typically intended as a primary contribution, not a background chapter. That raises the bar: pre-register the protocol (e.g. on PROSPERO or OSF), report screening with a PRISMA flow diagram, and conduct a formal risk-of-bias / quality appraisal of included studies. AI-assisted tools (Elicit, SciNet, Perplexity, Research Rabbit, …) can support — but not replace — the search; citation-based tools (which traverse reference and citation networks) and syntax/keyword-based tools have different blind spots, so combining them, and documenting their use, is good practice. For quantitative syntheses, a systematic review is also the front end of a meta-analysis.

4 Summary

Research synthesis distils and integrates findings across studies; literature reviews are its central tool, surfacing key insights and research gaps.
Three key types differ in rigour and aim: narrative (flexible overview), systematic (transparent, replicable synthesis), meta-analysis (statistical synthesis).
A systematic review proceeds by defining a focused question (PICO and its variants), setting inclusion/exclusion criteria and a coding scheme, and designing a documented, reproducible search strategy.
Throughout, document every decision so the review can be reproduced — by you and by others.

5 What to Read Next

Qualitative vs Quantitative Research — choosing question types that suit different review scaffolds.
Mixed Methods and Methodological Pluralism — synthesising across diverse study types.

6 Further Reading

Cooper, H., Hedges, L. V., & Valentine, J. C. (Eds.). (2019). The Handbook of Research Synthesis and Meta-Analysis (3rd ed.). Russell Sage Foundation.
Page, M. J., et al. (2021). “The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.” BMJ, 372:n71. prisma-statement.org.
Petticrew, M., & Roberts, H. (2006). Systematic Reviews in the Social Sciences: A Practical Guide. Blackwell.

Citation

BibTeX citation:

@inreference{graebner2026-intro-to-systematic-literature-review,
  author = {Gräbner-Radkowitsch, Claudius and Kothe, Anna-Katharina},
  editor = {Gräbner-Radkowitsch, Claudius},
  title = {Introduction to {Systematic} {Literature} {Reviews}},
  booktitle = {Method and Inquiry},
  date = {2026-06-17},
  url = {https://euf-methodology-hub.netlify.app/resources/intro-to-systematic-literature-review},
  langid = {en}
}

For attribution, please cite this work as:

Gräbner-Radkowitsch, Claudius, and Anna-Katharina Kothe. 2026. “Introduction to Systematic Literature Reviews.” In Method and Inquiry, edited by Claudius Gräbner-Radkowitsch. June 17. https://euf-methodology-hub.netlify.app/resources/intro-to-systematic-literature-review.

--- title: "Introduction to Systematic Literature Reviews" description: "Tell narrative reviews, systematic reviews and meta-analyses apart, then work through the systematic review process — research question (PICO), inclusion/exclusion criteria and coding, and a documented, reproducible search strategy." author: - "Claudius Gräbner-Radkowitsch" - "Anna-Katharina Kothe" date: 2026-06-17 date-modified: 2026-06-17 image: ../assets/images/literature-review.png learning-objectives: - "Distinguish a narrative review, a systematic review and a meta-analysis by goal, scope and rigour" - "Formulate a focused review question using the PICO framework and avoid common question pitfalls" - "Define inclusion/exclusion criteria and build a coding scheme to screen studies" - "Design and document a reproducible search strategy across databases and supplementary sources" categories: - literature-review - research-design - reproducibility - reading - BA - MA level: - BA - MA content-type: reading citation: id: graebner2026-intro-to-systematic-literature-review type: entry-encyclopedia container-title: "Method and Inquiry" title: "Introduction to Systematic Literature Reviews" author: - name: "Claudius Gräbner-Radkowitsch" - name: "Anna-Katharina Kothe" editor: - name: "Claudius Gräbner-Radkowitsch" url: https://euf-methodology-hub.netlify.app/resources/intro-to-systematic-literature-review --- # Introduction Research synthesis and literature review are a key step in almost every study. As Newton put it, *"If I have seen further it is by standing on the shoulders of Giants."* Before you can add to a field, you need a reliable picture of what is already known — and, just as importantly, where the **gaps** are. **Research synthesis** is *"the practice of systematically distilling and integrating data from many studies in order to draw more reliable conclusions about a given research issue"* (Cooper et al., 2019). Literature reviews are the central tool for this: they distil key insights from academic research for practitioners and identify research gaps. This page introduces the main *kinds* of review and then walks through the core stages of the most rigorous variant — the **systematic** review. # Three Key Types of Review "Literature review" is an umbrella term. Three types matter most: - **Narrative review** — a non-systematic overview of a topic. - **Systematic review** — a systematic, transparent and replicable synthesis. - **Meta-analysis** — a statistical synthesis of quantitative studies. They differ across several dimensions: | | Narrative review | Systematic review | Meta-analysis | |---|---|---|---| | **Goal** | Overview/synthesis of a topic | Systematically identify, evaluate and synthesise all relevant evidence on a specific question | Statistically combine quantitative results from multiple studies | | **Approach** | Flexible, typically non-systematic selection | Rigorous, transparent and replicable; explicit inclusion/exclusion criteria | Statistical synthesis of studies with similar variables/outcomes | | **Scope** | Selective; significant works | Comprehensive; aims to include all relevant studies | Selective; only studies with compatible quantitative data | | **Quality assessment** | Limited/informal | Formal appraisal using explicit tools | Formal, rigorous appraisal of methodological and statistical compatibility | | **Transparency / replicability** | Low | High | High | | **Time investment** | Low–moderate | High | Very high | | **Typical question** | *What is known about leadership theories in organisations?* | *How effective are diversity-training programmes at reducing workplace discrimination?* | *How strong is the relationship between transformational leadership and employee performance?* | | **Output** | Thematic/chronological narrative | Structured findings with transparent documentation | Effect sizes and forest plots | The rest of this page focuses on the **systematic review**, the workhorse for evidence-based questions with enough empirical studies to synthesise. (The authoritative procedural standard is the **PRISMA** guidelines; treat what follows as an accessible on-ramp, not a replacement.) # The Systematic Review Process ## 1. Define the research question Having a good question is the key to any successful research endeavour. A widely used scaffold is the **PICO** framework: | Element | Question | Example | |---|---|---| | **P** (Population) | Who is being studied? | middle managers, SMEs, multinationals | | **I** (Intervention / Exposure) | What main variable or intervention is examined? | leadership training, flexible work arrangements | | **C** (Comparison) | What is it compared to? | traditional management, fixed schedules | | **O** (Outcome) | What consequences or results are measured? | productivity, employee satisfaction | *Example question:* "In multinational corporations **(P)**, how does transformational leadership **(I)** compared to transactional leadership **(C)** affect employee innovation behaviour **(O)**?" **Avoid** these common pitfalls: wordy expressions, ambiguous terms, merging several questions with "and", suggestive (leading) questions, and questions that cannot actually be answered. ::: {.callout-note} PICO is strongest for clinical-style, intervention questions. Variants such as **SPIDER**, **SPICE** and **ECLIPSE** are often better suited to qualitative and mixed-methods research in the social sciences. Choose the scaffold that fits your question. ::: ## 2. Define inclusion and exclusion criteria Decide which studies are of potential interest for your question by asking: - which **variables/topics** are covered? - which **kind of studies** (quality, method, …)? - which **locations** and **time periods**? - which **publication types** (books, journals, newspapers)? - which **samples** (students, seniors, gender, …)? The goal is to **develop a coding scheme** you (and others) can use to extract relevant information and decide whether a study is of interest. A simple scheme records, per study, fields like *Study ID, publication year, main method, key topic, abstract-screening decision (keep/drop), full-text-screening decision,* and *role in the main narrative*. Coding schemes can be developed **deductively, inductively or iteratively**; test the scheme regularly, write a short manual with examples, and make sure it aligns with your research interest. ## 3. Delineate a search strategy **Where to search?** - **Database selection:** Web of Science, Scopus, Google Scholar, … - **Additional sources:** snowball sampling, citation networks, expert suggestions. **How to search?** - which **keywords** to use (refer back to your inclusion/exclusion criteria)? - **where** to search for those keywords (e.g. the *Topic* field, which searches title, abstract and keywords)? - which papers to use for **snowball / citation-based** search? - whom to ask for additional papers? In a database like Web of Science, keywords are combined with the Boolean operators **AND, OR, NOT**, and several sub-searches can be merged — **each sub-search requiring its own rationale**. Mind the **caveats**: every database has restrictions (Web of Science, for example, skews toward English-language articles in ranked journals, introducing publication biases), and depending on your question you may need to search in **several languages**. ::: {.callout-important} ## Reproducibility is the whole point **Always document your decisions and justifications, and make your search reproducible — for you and for others.** In practice this means recording your exact search strings, keeping a tool such as a spreadsheet plus a reference manager (e.g. Zotero) for documentation, being explicit about limitations, and reporting the number of records retrieved, de-duplicated, screened and finally included at each stage. ::: ::: {.callout-note .level-advanced} ## 🎓 MA / PhD At MA/PhD level a systematic review is typically intended as a **primary contribution**, not a background chapter. That raises the bar: **pre-register** the protocol (e.g. on PROSPERO or OSF), report screening with a **PRISMA flow diagram**, and conduct a formal **risk-of-bias / quality appraisal** of included studies. AI-assisted tools (Elicit, SciNet, Perplexity, Research Rabbit, …) can support — but not replace — the search; **citation-based** tools (which traverse reference and citation networks) and **syntax/keyword-based** tools have different blind spots, so combining them, and documenting their use, is good practice. For quantitative syntheses, a systematic review is also the front end of a **meta-analysis**. ::: # Summary - **Research synthesis** distils and integrates findings across studies; literature reviews are its central tool, surfacing key insights and **research gaps**. - Three key types differ in rigour and aim: **narrative** (flexible overview), **systematic** (transparent, replicable synthesis), **meta-analysis** (statistical synthesis). - A systematic review proceeds by **defining a focused question** (PICO and its variants), **setting inclusion/exclusion criteria** and a coding scheme, and **designing a documented, reproducible search strategy**. - Throughout, **document every decision** so the review can be reproduced — by you and by others. # What to Read Next - [Qualitative vs Quantitative Research](qualitative-vs-quantitative-research.qmd) — choosing question types that suit different review scaffolds. - [Mixed Methods and Methodological Pluralism](mixed-methods-and-methodological-pluralism.qmd) — synthesising across diverse study types. # Further Reading - Cooper, H., Hedges, L. V., & Valentine, J. C. (Eds.). (2019). *The Handbook of Research Synthesis and Meta-Analysis* (3rd ed.). Russell Sage Foundation. - Page, M. J., et al. (2021). "The PRISMA 2020 statement: an updated guideline for reporting systematic reviews." *BMJ*, 372:n71. [prisma-statement.org](https://www.prisma-statement.org/). - Petticrew, M., & Roberts, H. (2006). *Systematic Reviews in the Social Sciences: A Practical Guide*. Blackwell.