Literature Review Examples and Samples

Most weak literature reviews fail in the same way: they summarize one source per paragraph. The difference between summary and synthesis is easiest to see side by side. Both versions below discuss the same three invented studies.

Summary (weak)

Hale (2018) surveyed 400 nurses and found that night shifts were linked to poorer sleep. Moreno (2020) interviewed 25 nurses about fatigue. Park and Adeyemi (2021) studied rotating schedules in two hospitals and found that sleep quality declined over six months.

Synthesis (strong)

Studies consistently associate night and rotating shifts with poorer sleep among nurses, whether measured through large surveys (Hale, 2018) or tracked over time (Park & Adeyemi, 2021). Interview evidence adds a dimension the surveys miss: nurses describe fatigue as cumulative across a run of shifts rather than tied to any single night (Moreno, 2020). Together, these findings suggest that schedule sequence, not just shift type, may matter, a possibility no study has yet tested directly.

The strong version opens with a claim, uses sources as evidence, compares methods, and ends by pointing to a gap. Look for those four moves in every example below.

Note: the sources, authors, years and findings in this sample are fictional, invented for illustration. Do not cite them.

A thematic structure groups studies by recurring topic. It is the most common way to organize a review. This excerpt is one theme from an invented review on community gardens and neighborhood social ties.

Gardens as Meeting Places

A recurring theme in the literature is that community gardens create opportunities for contact between neighbors who would not otherwise meet. Survey studies report that gardeners know more of their neighbors by name than non-gardeners in the same areas (Brandt & Osei, 2016; Lindgren, 2019), and observational work describes shared tasks such as watering rotas and harvest days as natural occasions for conversation (Tavares, 2020). However, the strength of this effect appears to depend on how gardens are governed. Gardens run by open committees were associated with more cross-group ties than those allocated by waiting list, where plots tended to stay with long-standing residents (Kowalczyk et al., 2021). Most of this evidence is cross-sectional, so it remains unclear whether gardens build social ties or attract residents who already have them. The theme therefore points to governance and self-selection as the two issues any new study must address.

Note: the sources, authors, years and findings in this sample are fictional, invented for illustration. Do not cite them.

Why this example works

• Claim-first topic sentence: the paragraph opens with the pattern, not with an author.
• Grouped citations: two studies support one claim, showing convergence.
• A qualifier: "However" introduces a condition (governance) that explains variation.
• Critical evaluation: it flags the cross-sectional design and the self-selection problem.
• A forward-looking close: the final sentence says what the theme means for new research.

A chronological structure traces how research on a topic developed over time. It works when the field moved through distinct phases, and fails when it becomes a list of dates. This invented excerpt reviews research on online course completion.

Research on completion in online courses has moved through three broad phases. Early studies were largely descriptive, documenting completion rates and comparing them with face-to-face courses (Ingram, 2009; Sato & Bell, 2011). Their main contribution was to establish that low completion was a persistent problem, but they offered little explanation of why. A second phase, beginning in the mid-2010s, used platform log data to predict which learners would drop out, identifying early inactivity and missed first assignments as strong warning signs (Duarte, 2015; Ng et al., 2017). These models improved prediction but treated dropout as a learner characteristic rather than something courses could change. The most recent phase has shifted toward intervention, testing whether reminders, peer cohorts or redesigned first weeks improve completion (Abara & Quist, 2021; Feld, 2022). Results so far are modest and vary by course type. Each phase has therefore answered the previous phase's question while raising a new one: we now know who is at risk, but not which interventions work for whom.

Note: the sources, authors, years and findings in this sample are fictional, invented for illustration. Do not cite them.

Why this example works

• Phases, not years: the structure is built around shifts in research questions, with dates as supporting detail.
• Each phase is evaluated: every stage gets both a contribution and a limitation.
• Causal logic between phases: the limitation of one phase explains the focus of the next.
• A synthesized ending: the final sentence turns the history into a current gap.

A methodological structure groups studies by how they were done. Use it when findings seem to depend on method, which is itself an important finding. This invented excerpt reviews research on mindfulness apps and university student stress.

Conclusions about mindfulness apps differ noticeably depending on research design. Randomized trials, which compare app users with waitlist or active control groups, generally report small reductions in perceived stress over four to eight weeks (Hollis et al., 2019; Rahimi & Torres, 2022). Large observational surveys report larger associations between app use and lower stress (Wexford, 2020), but because students choose whether to use the apps, these studies cannot separate the effect of the app from the characteristics of students who adopt it. Qualitative interview studies help explain the gap between these designs: students describe using apps intensively at first and then abandoning them, often during exam periods when stress peaks (Mbeki, 2021). Taken together, the evidence suggests that controlled designs may give the more cautious estimate of benefit, and that engagement over time, rarely measured in trials, is a likely moderator.

Note: the sources, authors, years and findings in this sample are fictional, invented for illustration. Do not cite them.

Why this example works

• Method as the organizing idea: trials, surveys and interviews each get a clear role.
• Explains disagreement: it shows why survey effects look larger (self-selection) instead of just reporting the difference.
• Uses qualitative evidence to reconcile quantitative results, a sophisticated move examiners reward.
• Hedged conclusion: "suggests" and "likely" match the strength of the evidence.

APA style governs headings, in-text citations and the reference list. This short sample of a literature review section shows APA 7 conventions: Level 1 and Level 2 headings, parenthetical and narrative citations, "et al." for three or more authors, an ampersand inside parentheses, and multiple citations ordered alphabetically and separated by semicolons. The topic is an invented study of reading for pleasure among adolescents.

Literature Review (Level 1: centered, bold)

Reading for pleasure is commonly defined as reading that young people choose to do for its own sake rather than for assessment (Calder, 2017). Research has examined its relationship with vocabulary growth, well-being and school attachment, but the strength of the evidence varies across these outcomes.

Reading for Pleasure and Vocabulary (Level 2: flush left, bold)

The most consistent finding concerns vocabulary. Several longitudinal studies report that adolescents who read for pleasure several times a week show greater vocabulary gains over two to three years than infrequent readers, even after accounting for family background (Calder, 2017; Duvall & Ito, 2020; Petrakis et al., 2022). Duvall and Ito (2020) found the association was stronger for fiction than for other text types, although Petrakis et al. (2022) did not replicate this difference.

Reading for Pleasure and Well-Being (Level 2)

Evidence on well-being is more limited and mostly cross-sectional. Survey studies associate frequent leisure reading with lower reported loneliness (Okonjo, 2019), but no study has yet established the direction of this relationship.

References (fictional entries, formatted in APA 7 for demonstration)

• Calder, M. (2017). Choice and motivation in adolescent reading. Journal of Literacy Development, 12(3), 45–61.
• Duvall, R., & Ito, K. (2020). Text type and vocabulary growth in secondary school readers. Reading Research Review, 8(2), 112–130.
• Okonjo, A. (2019). Leisure reading and loneliness in early adolescence. Youth Wellbeing Quarterly, 4(1), 20–34.
• Petrakis, L., Hamid, S., & Ward, J. (2022). Reading habits and language outcomes: A three-year cohort study. Educational Pathways, 15(4), 301–319.

Note: the sources, authors, years and findings in this sample are fictional, invented for illustration. Do not cite them.

Why this example works

• The opening paragraph defines the concept and previews where evidence is strong and weak.
• Level 2 headings are themes, not author names.
• Narrative citations ("Duvall and Ito (2020) found") are used when a specific study's result is the point; parenthetical citations when several studies support one claim.
• Real APA references would include a DOI where one exists; these fictional entries omit it.

For the heading levels and outline that sit behind a review like this, see the literature review outline guide.

In a journal article or course research paper, the literature review is often only a few paragraphs, sometimes folded into the introduction. It has to move quickly from context to gap. This invented example comes from a short paper on bike-lane installation and local retail.

Concerns that removing on-street parking for bike lanes will harm nearby shops are common in public consultations, yet the evidence base is thin. Studies of single corridors have reported stable or rising retail spending after installation (Asante & Roe, 2019; Quintero, 2016), and intercept surveys suggest that merchants tend to overestimate the share of customers who arrive by car (Lund, 2018). However, these studies typically compare one street before and after installation without a comparison street, making it hard to separate the effect of the lane from citywide trends. The present study addresses this limitation by comparing sales-tax receipts on twelve streets that received bike lanes with twelve matched streets that did not.

Note: the sources, authors, years and findings in this sample are fictional, invented for illustration. Do not cite them.

Why this example works

• Four sentences move through the full arc: context, evidence, limitation, response.
• The limitation (no comparison group) is methodological and specific, so the study's design follows logically from it.
• It shows why short reviews still need critical evaluation, not just a list of findings.

A literature review in a research proposal has a narrower job than a thesis chapter: it must convince reviewers that your study is needed and well designed. It is typically shorter, ends with a clear justification, and often uses future tense for the proposed work. This invented excerpt is from a proposal on telehealth follow-up after knee surgery.

Telehealth follow-up after orthopedic surgery has expanded rapidly, and early evaluations suggest it can match in-person visits on patient satisfaction while reducing travel time (Garrow & Lin, 2020; Sefton, 2021). Fewer studies have examined clinical outcomes. The two trials that measured range of motion after knee replacement found no meaningful difference between telehealth and in-person follow-up (Abiodun et al., 2021; Merritt, 2022), but both recruited patients from urban centers with reliable internet access and excluded those over 75. As a result, the evidence says little about older or rural patients, the groups for whom travel savings would matter most and for whom technical barriers may be greatest. The proposed study will address this gap by comparing range-of-motion recovery and missed-appointment rates between telehealth and in-person follow-up among patients aged 65 and over at three rural hospitals. Including older and rural patients will test whether existing findings generalize to the populations most likely to be offered telehealth.

Note: the sources, authors, years and findings in this sample are fictional, invented for illustration. Do not cite them.

Why this example works

• Focused, not exhaustive: it covers only the literature needed to justify the study.
• The gap comes from sample limitations that are named precisely (urban, under 75).
• The gap and the proposed sample match exactly, which reviewers check first.
• It ends with significance: why answering the question matters.

For the other sections of a proposal, see writing a research proposal.

All five samples follow the same paragraph-level template. Use it to draft each section of your own review, then revise for flow.

For a full section-by-section outline with introduction and conclusion, including example conclusions, see the literature review outline guide.

Standalone review articles need an abstract summarizing scope, method, findings and implications. This invented example accompanies the mindfulness-app review above.

This review examines research on smartphone mindfulness apps and perceived stress among university students. Peer-reviewed studies were identified through three databases and grouped by design: randomized trials, observational surveys and qualitative interviews. Randomized trials generally reported small short-term reductions in stress, while observational studies reported larger associations that are likely inflated by self-selection. Qualitative studies indicated that engagement declines over time, particularly during exam periods. The review concludes that mindfulness apps may offer modest benefits and that sustained engagement is an under-studied moderator. Future trials should measure engagement directly and extend follow-up through periods of academic pressure.

Note how each sentence maps to one element: topic, method, three findings, conclusion, future research. Every finding is fictional. For more on structure, see how to write an abstract.

• Borrow structure, not content. Copy the moves (claim, evidence, complication, evaluation, implication), then fill them with your own reading.
• Never cite a sample. The sources here are fictional. Sources in real published reviews must be read in the original before you cite them.
• Match the example to your task. A thesis chapter needs the depth of the thematic and methodological samples; a short paper needs the compactness of the research-paper example.
• Check your discipline's conventions. Humanities reviews often engage with arguments and interpretations rather than empirical findings, and may use MLA or Chicago.
• Save or print this page if you want an offline literature review sample; all examples are on a single page.

If you want feedback on how your own draft compares with these models, we coach, edit and explain; you remain the author of your review.