An annotated bibliography is a reference list where every entry is followed by a short paragraph — usually 100 to 200 words — saying what the source claims, how good the evidence is, and what it does for your project. It is a reading log with citations attached, not a pile of abstracts.
The fastest way to understand one is to read one. Below is an example in full — three entries, then the same entry reformatted for MLA and Chicago.
The example
These entries come from a bibliography on one question: does taking lecture notes by hand produce better learning than typing them? The style is APA 7, alphabetical by first author’s surname. On paper each reference gets a hanging indent and the annotation sits indented underneath; the indented blocks below stand in for that indent.
Entry 1
Mueller, P. A., & Oppenheimer, D. M. (2014). The pen is mightier than the keyboard: Advantages of longhand over laptop note taking. Psychological Science, 25(6), 1159–1168. https://doi.org/10.1177/0956797614524581
Across three experiments with undergraduates, Mueller and Oppenheimer found that handwritten note-takers outperformed laptop note-takers on conceptual questions, and argued that the cause is encoding: longhand is too slow for transcription, so writers must reframe the lecture in their own words, while typists transcribe verbatim. This is the origin of the “pen is mightier” claim that almost all popular coverage repeats. Its limits matter here: samples of 67 to 151 students from two universities’ subject pools, and immediate testing in the first two studies — only Study 3 delayed the test by a week, and there longhand won only among students who were allowed to review their notes first. I treat it as the strongest statement of the encoding hypothesis rather than as settled evidence, and read it against Morehead et al. (2019) below.
Entry 2
Morehead, K., Dunlosky, J., & Rawson, K. A. (2019). How much mightier is the pen than the keyboard for note-taking? A replication and extension of Mueller and Oppenheimer (2014). Educational Psychology Review, 31(3), 753–780. https://doi.org/10.1007/s10648-019-09468-2
A direct replication, extended with groups who took notes on eWriters and who took no notes at all, that did not reproduce the headline result: some trends pointed to longhand, but test performance did not consistently differ between any of the groups — including, in the second experiment, the group that took no notes. A meta-analysis the authors run across the direct replications puts the longhand advantage at small and non-significant, and they conclude that calling either method superior is premature. This is the most important source in my bibliography, because it forces the review to be framed as an open question rather than as a finding to be reported.
Entry 3
Sana, F., Weston, T., & Cepeda, N. J. (2013). Laptop multitasking hinders classroom learning for both users and nearby peers. Computers & Education, 62, 24–31. https://doi.org/10.1016/j.compedu.2012.10.003
This paper moves the question from handwriting versus typing to what the laptop is being used for. In two experiments, multitasking on a laptop during a lecture lowered comprehension for the multitasker and — the striking result — for students seated within view of that screen. I include it because it names a confound running through the whole literature: a study comparing “laptop” with “handwriting” may be comparing distraction levels rather than encoding processes. It is also the only result here with a direct classroom-policy implication, and the one my discussion builds on.
What each sentence is doing
A working annotation makes three moves. Entry 1 makes all three, in order:
| Move | What it answers | Where it is in Entry 1 |
|---|---|---|
| Summarize | What is claimed, on what evidence? | “Across three experiments … transcribe verbatim.” |
| Assess | How strong is it, where does it stop? | “Its limits matter here … review their notes first.” |
| Reflect | What does it do for your project? | “I treat it as … read it against Morehead et al.” |
An annotation that only makes the first move is an abstract, and your marker already has the abstract. The credit is in the second and third. A quick test: if you could have written it without knowing your research question, it is not doing any work.
The same entry in MLA and Chicago
Only the reference changes. The annotation prose is identical.
MLA 9 — MLA asks that an annotated list be titled Annotated Bibliography or Annotated List of Works Cited:
Mueller, Pam A., and Daniel M. Oppenheimer. “The Pen Is Mightier Than the Keyboard: Advantages of Longhand over Laptop Note Taking.” Psychological Science, vol. 25, no. 6, 2014, pp. 1159–68, https://doi.org/10.1177/0956797614524581.
Chicago 18 author-date, under a heading called References:
Mueller, Pam A., and Daniel M. Oppenheimer. 2014. “The Pen Is Mightier Than the Keyboard: Advantages of Longhand over Laptop Note Taking.” Psychological Science 25 (6): 1159–68. https://doi.org/10.1177/0956797614524581.
| APA 7 | MLA 9 | Chicago 18 (author-date) | |
|---|---|---|---|
| Heading on a plain list | References | Works Cited | References |
| Author names | Surname + initials | Surname, full given name | Surname, full given name |
| Year position | After the authors, in parentheses | After the volume and issue | After the authors, no parentheses |
| Article title | Sentence case, no quotation marks | Title Case, in quotation marks | Title Case, in quotation marks |
| Volume / issue | 25(6) | vol. 25, no. 6 | 25 (6) |
| Three or more authors | All listed, up to 20 | First author + “et al.” | All listed, up to a cut-off |
| Annotation goes | New paragraph, indented under the entry | New line with a paragraph indent, or run in after the entry | New indented line, or run in after the entry |
Two cautions. Chicago has two systems: notes-bibliography, headed Bibliography, moves the year out of the author slot into parentheses after the volume and issue; author-date is shown above. And numeric styles such as IEEE and Vancouver order the list by citation order rather than alphabetically, which is why annotated bibliographies are almost never set in them. Manuals get revised and departments add house rules, so the sheet your marker is holding beats this table.
Length, order, and kind
- Per annotation: most assignments ask for 100 to 200 words; some ask for two sentences. The brief beats any guide on the web, this one included.
- Order: alphabetical by first author’s surname in APA, MLA, and Chicago, exactly as a plain reference list would run. MLA also allows arranging the list by date of publication or by subject; elsewhere, group by theme only if asked.
- Kind: descriptive annotations summarize; evaluative ones also judge and place the source. Most assignments want the evaluative kind without saying so.
Four mistakes that cost marks
- Pasting the abstract. The most detectable plagiarism in this genre, and it skips the assessing entirely.
- Annotations that never mention your project. See the test above.
- Mixing styles. One style, every entry.
- Wrong metadata. Page ranges, issue numbers, and DOIs are what a marker spot-checks, because checking is quick.
Doing this without retyping every reference
The mechanical half — finding sources, keeping their metadata correct, rendering it in the right style — is worth automating. The annotations are not. In Litlas that half maps onto the product directly:
- Save and annotate together. Search indexed sources including OpenAlex, Crossref, PubMed, and arXiv, save a paper to a board for the assignment, and write the annotation in that item’s personal note in your library.
- Format the reference. APA, Chicago (author-date), IEEE, MLA, and Vancouver, rendered with citeproc-js from the paper’s stored metadata. Set the style once and every citation you insert uses it.
- Assemble it in a document. In a Litlas document — rich text or LaTeX — “cite from your library” inserts the in-text citation plus the formatted reference entry, and you type the annotation underneath.
- Or export and finish elsewhere. Export your library, or one board, as a
single BibTeX file. Your note travels in the BibTeX
notefield, so annotations arrive with the references instead of being retyped. (A single paper’s Cite panel also exports RIS and CSL-JSON, but those carry the paper’s metadata only — not your note.)
The free plan covers a course assignment: 3 boards, 100 saved items, 5 searches a day, no card.
Three things Litlas does not do, so nothing is a surprise later:
- There is no “annotated bibliography” output. Litlas holds the two pieces — a correct reference and your note — but you assemble the document.
- Chicago notes-bibliography is not one of the five styles. Chicago author-date is. If your assignment names the notes system, format that list yourself.
- Litlas has no AI of its own. The ✨ button hands a paper off to Gemini, ChatGPT, Claude, or Gemini Notebook in your own account; nothing is summarized on Litlas servers. That suits this task — an annotation written by a model that has not read your brief is exactly what markers look for.
Annotated bibliographies get assigned because they are hard to fake. The references prove you found the sources; the annotations prove you read them and formed a view. Automate the first half, and spend what you save on the second.