# Literature Review > Plan, structure, write, and improve academic literature reviews of all types — including narrative reviews, systematic reviews, and literature review chapters in dissertations. Use this skill whenever the user needs to write a literature review, synthesise multiple sources, structure a review chapter, understand PRISMA guidelines, or integrate sources into a coherent argumentative narrative. Trig… - **Type:** Skill - **Install:** `agentstack add skill-muhammad1438-academic-writer-skills-literature-review` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [muhammad1438](https://agentstack.voostack.com/s/muhammad1438) - **Installs:** 0 - **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills) - **Latest version:** 0.1.0 - **License:** MIT - **Upstream author:** [muhammad1438](https://github.com/muhammad1438) - **Source:** https://github.com/muhammad1438/academic-writer-skills/tree/master/skills/literature-review ## Install ```sh agentstack add skill-muhammad1438-academic-writer-skills-literature-review ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # Literature Review A literature review is not a summary parade — it is a **critical, synthesising argument** that demonstrates your command of the existing discourse and carves out the intellectual space your own research will occupy. --- ## Types of Literature Reviews | Type | Purpose | Structure | |---|---|---| | **Narrative Review** | Broad synthesis and critical discussion of a field | Thematic or chronological; author-led | | **Systematic Review** | Exhaustive, reproducible survey of all evidence on a question | PRISMA protocol; strict inclusion/exclusion | | **Scoping Review** | Map the breadth of a topic; identify gaps | PRISMA-ScR; broader scope than systematic | | **Literature Review Chapter** (Dissertation) | Contextualise your own study; establish the gap | Funnel structure: broad → specific | | **Integrative Review** | Synthesise diverse methodologies | Theory-building focus | | **Critical Review** | Evaluate quality and contribution of a body of work | Argumentative; includes critique of methods | --- ## Narrative Literature Review — Structure ### The Funnel Model (Standard Dissertation Chapter) ``` ┌─────────────────────────────────────┐ │ Broad theoretical context │ (What does the field broadly know?) ├─────────────────────────────────────┤ │ Key themes and debates │ (What are the major arguments?) ├─────────────────────────────────────┤ │ Specific studies relevant to you │ (What evidence exists closest to your topic?) ├─────────────────────────────────────┤ │ Research gap │ (What is still unknown or contested?) └─────────────────────────────────────┘ │ Your study's positioning │ (How does your study address this gap?) └─────────────────────────────────────┘ ``` ### Thematic Organisation (Preferred over Chronological) Organise sources by **theme/concept**, not by author or publication date: - Group sources that agree with each other - Group sources that disagree or complicate each other - Each theme = one section with its own mini-argument Avoid the "annotated bibliography" mistake: writing a paragraph per source in isolation. Synthesis means showing how sources **talk to each other**. --- ## Systematic Review — PRISMA Framework Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) is the international standard for systematic reviews. ### PRISMA Phases 1. **Identification:** Search all relevant databases (PubMed, Web of Science, Scopus, JSTOR, etc.) using pre-defined search strings 2. **Screening (Title & Abstract):** Remove duplicates; screen by title and abstract for relevance 3. **Eligibility (Full Text):** Assess remaining papers against formal inclusion/exclusion criteria 4. **Included:** Final set of papers that meet all criteria ### Documentation Requirements - **Search strategy:** Must be fully reproducible — list every database, every search term, Boolean operators used, date range - **PRISMA Flow Diagram:** Visual representation of papers at each stage (required in most journals) - **Inclusion criteria:** Population, Intervention/exposure, Comparison, Outcome (PICO framework) - **Exclusion criteria:** State why papers were excluded (wrong population, methodology, date, language) - **Data extraction table:** Standardised extraction of key variables from each included study --- ## The Synthesis vs. Summary Distinction | Summary | Synthesis | |---|---| | Reports what one source says | Connects what multiple sources say to each other | | Source-by-source structure | Theme-by-theme structure | | "Smith (2019) found that X" | "While Smith (2019) found X, Jones (2021) complicates this by demonstrating Y, a tension also observed by Lee (2022)" | | Descriptive | Analytical and evaluative | **The test:** Can you remove any one source from your paragraph without breaking the logic? If yes — you are summarising, not synthesising. --- ## Identifying and Writing the Research Gap The research gap is the **most important sentence** in your literature review — it justifies your entire study. ### Types of Research Gaps - **Empirical gap:** Topic has not been studied in a specific population, context, or time period - **Methodological gap:** Prior studies use weak or limited methods; a more rigorous approach is needed - **Theoretical gap:** Existing theories do not explain a particular phenomenon - **Geographical gap:** Research exists in Western contexts but not in your regional context - **Temporal gap:** The field has not been updated since significant changes occurred ### Gap-Signalling Phrases - "To date, there has been little agreement on..." - "Previous studies have failed to specify..." - "What is not yet clear is the impact of X on..." - "No study has examined X in the context of..." - "While X has been extensively studied, Y remains largely unexplored..." - "Existing research is limited by its reliance on..." - "There remains a paucity of longitudinal evidence on..." --- ## Critical Analysis in the Literature Review Do NOT simply report findings — evaluate them: - **Sample limitations:** Small samples, unrepresentative populations, self-selected groups - **Methodological weaknesses:** Lack of control group, self-report bias, cross-sectional design - **Theoretical limitations:** Assumes Western frameworks that may not apply universally - **Temporal issues:** Research is outdated; conditions have changed significantly - **Conflicting evidence:** Sources contradict each other — note the conflict and analyse why ### Critical Verbs for Evaluation - challenges, problematises, overlooks, neglects, fails to account for, is limited by, underestimates, overgeneralises, does not distinguish between --- ## Synthesis Paragraph Structure **Topic sentence** → State the theme or argument this paragraph addresses **Evidence 1** → Introduce first source and its finding (paraphrase + cite) **Evidence 2** → Introduce contrasting or supporting source (paraphrase + cite) **Analysis** → What does this tension or agreement reveal? Why does it matter? **Link** → Connect to the broader argument or bridge to the next theme ### Example Synthesis Paragraph > Organisational commitment has consistently been identified as a key predictor of employee retention (Meyer & Allen, 1991; Mowday et al., 1979). However, more recent empirical work suggests that the relationship is moderated by job autonomy and perceived organisational support (Eisenberger et al., 2020; Rhoades & Eisenberger, 2002). These findings complicate early models, which treated commitment as a relatively stable, individual-level disposition (Allen & Meyer, 1996), by demonstrating that structural and relational workplace factors play a significant role. The implication is that retention strategies focused purely on attitudinal change may be insufficient without corresponding changes to organisational conditions. --- ## Common Literature Review Mistakes | Mistake | Fix | |---|---| | Annotated bibliography structure (one source per paragraph) | Restructure thematically; make sources speak to each other | | Quoting too heavily instead of synthesising | Paraphrase and attribute; reserve direct quotes for key definitions only | | Including outdated sources without acknowledging it | Note the date and explain why older foundational work is still relevant | | No critical evaluation — just description | Add evaluative language; note limitations of each study | | Missing the research gap | Explicitly state what is unknown; do not leave it implicit | | No clear organisation (sources listed randomly) | Use headings and thematic structure | | Including everything you read | Only include sources relevant to your specific research question | --- ## Review Type Decision Guide ### Choosing Your Review Methodology ``` What is your research question? ├─ Broad overview of a field? → Narrative Review ├─ Specific, answerable PICO question? → Systematic Review ├─ Mapping extent of evidence? → Scoping Review ├─ Quantitative synthesis of effect sizes? → Meta-Analysis └─ Mixed methods synthesis? → Mixed Methods Review ``` ### Narrative Review **When to use:** Broad topics, introductory sections, theoretical discussions. **Advantages:** Flexible, covers wide ground, expert interpretation valued. **Limitations:** Prone to selection bias; less replicable. **Reporting:** No formal protocol required; cite key sources; acknowledge limitations. ### Systematic Review **When to use:** Specific clinical or empirical question; evidence synthesis for policy. **Required elements:** 1. Pre-specified protocol (register on PROSPERO before searching) 2. Comprehensive search strategy (multiple databases, grey literature) 3. Documented inclusion/exclusion criteria 4. Quality assessment of included studies 5. PRISMA reporting (flow diagram + 27-item checklist) **Search strategy template:** ``` ("keyword 1" OR "synonym 1") AND ("keyword 2" OR "synonym 2") AND ("keyword 3" OR "synonym 3") ``` Run in: PubMed, Embase, Cochrane Library, PsycINFO (add field-specific databases). ### Scoping Review **When to use:** Emerging field; map concepts and evidence types; identify research gaps. **Different from systematic review:** No quality assessment; broader inclusion. **Report with:** PRISMA-ScR (Scoping Review extension). ### Meta-Analysis **When to use:** Sufficient homogeneous studies exist; quantitative synthesis appropriate. **Required:** Effect sizes and variance measures from each study. **Report:** Pooled effect size with 95% CI, heterogeneity (I², τ²), funnel plot. ## Synthesis vs. Summary The central skill in literature reviewing is synthesis — making meaning across sources, not just describing them. | Summary | Synthesis | |---------|-----------| | "Smith (2020) found X" | "Studies consistently show X (Smith, 2020; Jones, 2019)" | | Each paragraph = one paper | Each paragraph = one theme/finding | | Papers are described in sequence | Papers compared and contrasted | | Author's perspective is absent | Your analytical voice is present | | "Next, Lee (2021) argued..." | "This debate reflects..." | ### Synthesis Sentence Starters - "Taken together, these studies suggest..." - "A consistent finding across the literature is..." - "Contrasting explanations have been offered: [X], while [Y]..." - "The evidence for X is mixed: [Smith] found Y, while [Jones] found Z..." - "A gap in the existing research is..." ## Research Gap Identification After synthesising, explicitly identify gaps: 1. **Population gaps:** Studies focus on Western, WEIRD (Western, Educated, Industrialised, Rich, Democratic) samples 2. **Methodological gaps:** Mostly cross-sectional studies; no longitudinal data 3. **Contextual gaps:** No studies in low-income countries 4. **Theoretical gaps:** No studies testing [specific theory] 5. **Temporal gaps:** No studies since [major event] 6. **Replication gaps:** Key findings not independently replicated ## Integration with Other Skills - Register your review on PROSPERO first using `preregistration-guide` - Use `statistical-reporting` for meta-analysis reporting - Use `figure-table-checker` for PRISMA flow diagrams - Use `reference-verifier` to validate all included study citations ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [muhammad1438](https://github.com/muhammad1438) - **Source:** [muhammad1438/academic-writer-skills](https://github.com/muhammad1438/academic-writer-skills) - **License:** MIT Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-muhammad1438-academic-writer-skills-literature-review - Seller: https://agentstack.voostack.com/s/muhammad1438 - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.