What Is an Annotated Bibliography?
An annotated bibliography is a list of citations, each followed by a brief descriptive and evaluative paragraph — the annotation. It differs from a standard reference list in that it requires you to engage critically with each source: what does it say, how reliable is it, and how does it relate to your research?
In STEM, annotated bibliographies are used as:
- Standalone assignments in research methods modules
- Preliminary work before a literature review or systematic review
- A tool to document and justify sources used in a larger project
- Preparation for a thesis or dissertation proposal
Three Types of Annotation
| Type | What it does | When to use |
|---|
| Descriptive / Informative | Summarises the content, methods, and findings — no evaluation | When the assignment asks only for a summary of sources |
| Evaluative / Critical | Summarises AND critiques — assesses strengths, limitations, methodology quality | Most academic annotated bibliography assignments |
| Reflective / Analytical | Summarises, evaluates, AND reflects on how the source relates to your specific research question | Thesis/dissertation preparation; research methods modules |
Check your assignment brief carefully. Some modules ask only for descriptive annotations; others require full evaluative or reflective annotations. The distinction affects both length and depth — an evaluative annotation for a STEM paper is typically 150–250 words; a descriptive one may be only 75–100 words.
What to Include in Each Annotation
A complete evaluative annotation addresses five questions:
- What does the source argue or report? — the main thesis, finding, or claim
- What methodology was used? — experimental, observational, systematic review, computational?
- What are the main strengths? — large sample, rigorous methods, peer-reviewed, high-impact journal
- What are the limitations or weaknesses? — small sample, narrow population, dated, potential bias, conflict of interest
- How does this source relate to your topic? — does it support, challenge, or contextualise your argument?
Worked Examples by Citation Style
APA 7th Edition — Evaluative Annotation
Example entry
Chen, L., Patel, R., & Kim, S. (2023). Machine learning approaches to early sepsis prediction in intensive care units: A systematic review. Critical Care Medicine, 51(4), 412–428. https://doi.org/10.1097/CCM.0000000000005789
This systematic review synthesises 42 studies evaluating machine learning (ML) models for early sepsis prediction, covering logistic regression, random forests, and LSTM neural networks applied to ICU electronic health record data. The authors conducted a comprehensive search of PubMed, Embase, and Cochrane through December 2022, applying PRISMA reporting guidelines and assessing study quality using the PROBAST tool. The review's key finding — that gradient boosting models achieved the highest AUROC (mean 0.89, 95% CI: 0.86–0.92) across external validation datasets — is directly relevant to this project's comparison of ML classifiers. A significant limitation is the heterogeneity of sepsis definitions across included studies (Sepsis-2 and Sepsis-3 criteria were both used), which complicates cross-study comparisons. Nevertheless, this is the most methodologically rigorous review currently available and provides the primary benchmark for model performance expectations in this field.
IEEE — Evaluative Annotation
Example entry
[1] M. Rodriguez, A. Singh, and T. Nguyen, "Convolutional neural networks for real-time defect detection in additive manufacturing," IEEE Trans. Ind. Electron., vol. 70, no. 3, pp. 2841–2850, Mar. 2023, doi: 10.1109/TIE.2022.3174284.
Rodriguez et al. present a CNN architecture achieving 97.3% defect classification accuracy on in-situ images from a fused deposition modelling (FDM) process, with inference times of 12 ms per frame — compatible with real-time monitoring. The dataset (18,500 labelled images across seven defect categories) is larger than previous comparable studies and the authors provide ablation results demonstrating that the custom attention module contributed a 4.1% accuracy gain over a baseline ResNet-50. A limitation is that the system was validated exclusively on a single FDM platform; generalisation to SLS or SLA processes is untested. This paper is highly relevant to Chapter 3 of this project as it establishes state-of-the-art performance benchmarks and provides the attention module architecture adopted in our implementation.
Need a professionally written annotated bibliography?
Our STEM specialists research, annotate, and format annotated bibliographies in APA, IEEE, Harvard, or any required style.
Get Help Now →
Harvard — Reflective Annotation
Example entry
Müller, K. and Wang, F. (2022) 'Antibiotic resistance gene transfer in hospital wastewater: a longitudinal cohort study', Water Research, 215, pp. 118–131.
This two-year longitudinal study tracked antibiotic resistance gene (ARG) concentrations in effluent from three German tertiary hospitals using quantitative PCR and metagenomic sequencing. The authors identified significant seasonal variation in ARG load (highest in winter months, p < 0.001) and a positive correlation between patient antibiotic prescription rates and downstream ARG concentrations (r = 0.74). Methodologically, the use of both qPCR and metagenomic approaches is a strength — they provide complementary quantitative and taxonomic data. The main limitation is the European hospital setting, which may not be representative of lower-resource contexts with different wastewater infrastructure. This source directly informs my comparative analysis of ARG fate in Sub-Saharan African hospitals, providing baseline European data against which local findings can be benchmarked. The seasonal variation finding raises a hypothesis I will test: whether the Nairobi sites show a similar pattern driven by wet/dry season antibiotic usage changes.
Formatting Rules
- The citation is formatted exactly as it would appear in a reference list — same style, same punctuation
- The annotation begins on a new line, indented (usually consistent with hanging indent style)
- Entries are arranged alphabetically by author surname (APA, Harvard) or in citation order (IEEE)
- Each annotation is a single paragraph — no bullet points within the annotation itself
- Written in third person and present tense for summarising ("the authors argue that…" "the study examines…")
Common Mistakes
- Pure summary without evaluation: describing what the paper says without assessing its quality or relevance
- Vague praise: "this is a very useful and important paper" — specify why it is useful and what exactly it contributes
- No limitations identified: every source has weaknesses — acknowledging them demonstrates critical reading
- Annotation too long: focus on the most important points — 150–250 words is usually sufficient for a full evaluative annotation
- Inconsistent citation format: mixing APA and Harvard formatting within the same document
- Annotations copied from the abstract: an abstract summary is not an evaluation — you must add your own critical assessment
Frequently Asked Questions
How is an annotated bibliography different from a literature review?
A literature review synthesises multiple sources into a coherent argument about the state of knowledge — it is integrated, thematic prose. An annotated bibliography treats each source separately with its own entry and annotation. A literature review is the output; an annotated bibliography is often the preparation for writing it. Some modules require both: the annotated bibliography as a formative exercise and the literature review as the summative piece.
How many sources should an annotated bibliography include?
This varies by assignment. Research methods modules commonly ask for 10–20 sources. As preparation for a master's thesis, 30–50 annotated sources is typical. Systematic review preparation may require annotating 50–100 included studies. Always follow the brief — "at least 15 peer-reviewed sources" is a minimum, not a target.
Can I include non-peer-reviewed sources?
Most academic annotated bibliographies should be primarily peer-reviewed journal articles and books. Government reports, technical standards, and institutional publications are acceptable if they are authoritative and relevant. Wikipedia, general websites, and opinion pieces are not appropriate primary sources for academic STEM annotated bibliographies unless explicitly permitted by the assignment brief.