The way to remember what you read is to stop reading at regular points and pull the ideas back out of your head: a question before each section, an answer from memory after it, and a second test a few days later. Reading a chapter twice and colouring it in feels like work, but the research on learning rates both of those habits low.
This guide sets the reading comprehension strategies students are taught against the evidence: the SQ3R method, highlighting, rereading, self-explanation, asking why, summarising, and screen against paper. It then compares five universities' advice on how to read a textbook, and ends with a routine of our own.
EdCitation publishes this guide. Every study and university page was read on 27 September 2026; where we read only an abstract, we say so. The books a course sets can be searched in EdCitation's Library, the papers behind this guide in Find sources, and each reference was built in Cite a source.
How do you remember what you read?
You remember what you read by making yourself produce it, not by seeing it again. Four habits carry most of the weight:
- Read with a question. A heading turned into a question tells you what to look for.
- Recall at the end of each section. Look away and say or write the answer in your own words.
- Ask why. Explain to yourself why a fact is true or how a step follows.
- Come back later. A self-test days afterwards does more than a reread today.
Testing and timing have guides of their own: retrieval practice and spaced repetition. For journal articles, see how to read an academic paper fast.
What is the SQ3R method, and does it work?
SQ3R is a five-step way to study a textbook chapter: Survey, Question, Read, Recite, Review. It was set out by Francis P. Robinson of Ohio State University, and the evidence for it is thinner than its fame suggests.
What Robinson wrote in 1946
Robinson (1946) called it the Survey Q3R Method in Effective Study, a revised edition of a 1941 book. The steps, in our words:
- Survey. Glance over the headings and any closing summary to find the few core ideas; he allowed about a minute.
- Question. Turn the first heading into a question.
- Read. Read to the end of that section to answer it, which he contrasted with "a passive plowing along each line".
- Recite. Look away from the book, answer in your own words with an example, and jot brief cue phrases from memory. Then repeat steps 2 to 4 for each section.
- Review. Cover the notes and recall the main points, then the subpoints under each.
Robinson's notes were written after each section, from memory, not copied from the page; that recite step is self-testing, the part later research supports best.
His evidence was modest. He reported that a how-to-study class moved from the 34th to the 56th percentile for reading rate and from the 43rd to the 53rd for comprehension after several days of practice, a before-and-after comparison with no comparison group mentioned. He also reported quiz errors falling from 15 to 6 once students learned to predict questions.
What later studies found
Later studies are few and mixed. Johns and McNamara (1980) reviewed the research and concluded, in their abstract's words, that support for SQ3R rested "more on opinion than on empirical evidence"; we read only that abstract. Carlston (2011) reported that students who completed SQ3R note packets did better on course exams, and the author calls it a preliminary investigation; we read only its abstract, which gives no sample size.
The clearest test is a small experiment. Jairam et al. (2014) trained 25 undergraduates in SQ3R or in SOAR (select, organise, associate, regulate) for 30 minutes, gave them 45 minutes to study a text on reinforcement schedules, then tested them. The SOAR group recognised 14% more facts, 20% more relationships and 13% more concepts. Citing earlier work, the authors add that SQ3R can be hard to learn even after ten or more hours of training. Two cautions: one of the authors devised SOAR, and both groups studied from materials the researchers prepared.
Our reading: SQ3R is a sound frame because it forces recall, but adding "why" questions and a later self-test is better supported.
Does highlighting work?
Highlighting mostly does not help you remember more than reading alone. Dunlosky et al. (2013), in the best-known review of study techniques, rated highlighting and underlining of low utility.
The review's prototype experiment, run by Fowler and Barker in 1974, gave undergraduates an hour with about 8,000 words of magazine articles; a week later, highlighters did no better overall than readers. They did remember more of what they had marked, at a small cost to what they had not. Test scores fell as the amount highlighted rose (r = -.29), though with 19 students that was not significant. The worst sign came from a 1992 study by Peterson, as the review reports: students who underlined a 10,000-word history chapter did as well on fact questions but worse on questions that needed inferences across the text.
Why it falls short, and when it helps
Marking too much makes nothing stand out, and marking takes less thought than deciding what matters. It found benefits when students were limited to one sentence a paragraph, and suggests highlighting may help readers who already know enough to pick well. So if you mark, mark after you have read the section, and mark little.
What about rereading?
Rereading helps a little, and spacing it helps more, but it loses to the alternatives. Dunlosky and colleagues rated it low because it did worse than self-explanation, elaborative interrogation and practice testing in direct comparisons; on delayed tests, rereading after a gap beat rereading straight away. Robinson (1946) made the same point with an experiment he cited: readers scored 69% after one reading and 74%, 75% and 74% after two, three and four readings in a row.
Which reading comprehension strategies does the evidence support?
The strategies with better support make you explain, question or recall. Of those that fit reading, Dunlosky et al. (2013) rated two of moderate utility, one step below practice testing and spacing.
Elaborative interrogation: ask why
Elaborative interrogation means asking why a stated fact is true, for example why it would be true of this case and not another. In an early study the review describes, students prompted to explain each sentence recalled about 72% against about 37% for those who only read. The limits: most studies used lists of facts, not chapters; the benefit may be smaller for readers who know little about the topic; and prompts used only every page or two were diluted or even reversed. Ask why often, section by section.
Self-explanation
Self-explanation means saying what a sentence or step means to you and how it connects with what you already know. The review found effects across ages, subjects and kinds of test, including transfer to new problems, but little on how long they last. It costs time: in one study, explaining after each sentence took students 125 minutes against 66 for rereading. Use it on the hard sections, not every line.
Summarising in your own words
Summarising was rated low, not because it never works but because many learners need training to do it well. In Bretzing and Kulhavy's study, which the review describes, students who wrote three lines summarising each page did better than those who copied lines or only read. Robinson's recite step is a summary from memory, which adds recall. Summarising a source for an essay is a separate skill: see paraphrase, quote or summarise.
| Technique | What the evidence says | Source |
|---|---|---|
| SQ3R | Few studies; one small experiment favoured SOAR; its recite step is self-testing | Robinson, 1946; Jairam et al., 2014 |
| Highlighting and underlining | Low utility; may hurt inference questions; one sentence a paragraph helped | Dunlosky et al., 2013 |
| Rereading | Low utility; spaced beats immediate; loses to active techniques | Dunlosky et al., 2013 |
| Elaborative interrogation (asking why) | Moderate utility; strongest for facts; weaker with little prior knowledge | Dunlosky et al., 2013 |
| Self-explanation | Moderate utility; broad effects; costs time | Dunlosky et al., 2013 |
| Summarising | Low utility; works for those already skilled | Dunlosky et al., 2013 |
| Reading on paper rather than screen | Small advantage for informational text | Delgado et al., 2018; Clinton, 2019 |
| Self-testing and spacing after reading | High utility; covered in their own guides | Retrieval practice |
| Finding the book and citing it | A catalogue and a reference built from the record | EdCitation's Library and Cite a source |
Is reading on screen worse than reading on paper?
For informational text, reading on screen gives slightly lower comprehension on average; for stories, no gap was found. Delgado et al. (2018) pooled 54 studies from 2000 to 2017 with 171,055 participants and found a small paper advantage (g = -0.21), larger under time limits and, unexpectedly, larger in more recent studies. Clinton (2019) kept only randomised experiments: 33 studies with 2,799 readers gave g = -0.25, -0.32 for expository text and -0.04 for narrative. Reading time did not differ, and readers judged their own understanding more accurately on paper. We read Clinton in its abstract.
The effects are small and vary between studies. Our view: print helps a little with a dense chapter under time pressure, and recall matters more than the medium. Devices and scrolling are covered in ebooks or print for studying.
How do you read a textbook? What five universities advise
All five pages say read actively and selectively; they disagree on method, highlighting and rereading.
| University page | What it advises |
|---|---|
| Cornell University Learning Strategies Center (n.d.) | Three systems: SQ3R; P2R, reading about ten pages while "taking notes or highlighting" then reviewing; and S-RUN, which includes underlining |
| University of Toronto Mississauga (n.d.) | SQ3R; put the text away while reciting; review by testing yourself over time |
| University of Southampton Library (2026) | Check whether a text is essential or additional; preview, skim and scan; SQ5R for close reading, which it calls time intensive |
| University of Birmingham (2024) | Read once for an overall impression, read difficult passages again, paraphrase in your own words |
| University of Leeds Library (n.d.) | Scan and skim to find what matters, read that in depth; you will not have time to read everything |
Where they differ
On highlighting, Cornell builds it into two of its three systems, while the Dunlosky review rates it low. On rereading, Birmingham recommends it for hard passages, where Robinson's own figures and the review suggest recall does more. Toronto Mississauga's advice to recite with the book away and review over time is the closest to the evidence. Southampton and Leeds deal most directly with long reading lists: triage first, and Leeds suggests stopping when the information starts to repeat or you have covered the core material.
Our reading: triage the list first, and keep the routine below for the chapters that matter most.
A reading routine for a chapter (ours)
This routine is our own synthesis of the studies above. Nobody has tested it as a package, and the timings are suggestions.
- Triage. Check whether the chapter is essential or further reading, and why you are reading it. Find the set text in EdCitation's Library if you do not have it.
- Survey for two minutes. Headings, summary, figures. Write three to six questions from the headings, at least one of them a "why".
- Read one section to answer its question. No highlighting yet.
- Close the book and recall. Write two or three lines from memory in your own words, with an example.
- Check and explain. Reopen, fix what you missed, and add one line on how it connects with what you know. If you must mark the text, mark one sentence now.
- Review the chapter from your notes with them covered. Recall the main points, then the subpoints.
- Record the reference as you finish. Paste the ISBN or DOI into Cite a source.
- Test yourself again a few days later, as the spaced repetition guide schedules it, and reread only what you missed.
Where does EdCitation fit when you read for a course?
EdCitation is the best place to find, keep and cite what you read, because it works from the publisher's record: a reference is looked up, never written, which no chatbot can promise. It never writes any part of a student's work; the reading and remembering stay yours, and so does the credit.
Find and read the book
The Library is a catalogue of textbooks and scholarly books, searched by discipline, kind of book, language and year. Keep a book in My Books, buy a paid one, or find a library near you that holds it. With Max, at $24 a month, you can read or borrow from more than 5 million books in one place: a collection EdCitation compiled for students and readers, and part of the Max plan.
Find the research and cite it
Find sources searches about 300 million published works by topic or by the claim a sentence needs. On 27 September 2026, "SQ3R reading comprehension", sorted by citations, returned 1,944 works; third was Artis's 2008 paper on SQ3R for marketing students, cited 50 times. Cite a source builds references in APA 7, MLA 9, Chicago 18 author-date, Harvard, IEEE and Vancouver, and screens for retraction. From 10.1007/s11251-013-9295-0 it returned, with no retraction notice:
Jairam, D., Kiewra, K. A., Rogers-Kasson, S., Patterson-Hazley, M., & Marxhausen, K. (2014). SOAR versus SQ3R: A test of two study systems. Instructional Science, 42(3), 409–420. https://doi.org/10.1007/s11251-013-9295-0
The journal and volume are in italics on the page. Cite a source set the Dunlosky review's title in sentence case, as APA 7 asks. Finding, citing and Verify references are free with no account; Pricing lists Pro and Max.
Quick questions
What is the best way to remember what you read?
Recall it: answer a question on each section from memory, in your own words, and test yourself again days later. Dunlosky and colleagues rated practice testing and spacing of high utility.
Does the SQ3R method work?
SQ3R has little direct evidence. Robinson reported his figures without a comparison group, and in one small experiment students using SOAR outperformed those using SQ3R. Its recite and review steps are self-testing, which is well supported.
Does highlighting help you remember?
Highlighting mostly does not. Dunlosky and colleagues rated it low utility, and one study found underliners did worse on questions that needed inferences. Marking one sentence a paragraph after reading helped in another study.
Is it better to read textbooks on paper or on screen?
Paper has a small edge for informational text: Delgado and colleagues found g = -0.21 across 54 studies, and Clinton found no gap for narrative text. Recall after each section matters more than the medium.
Where can I read my textbooks in one place?
EdCitation's Library catalogues textbooks by discipline, language and year, and with Max you can read or borrow from more than 5 million books in one place.
References
- Carlston, D. L. (2011). Benefits of student-generated note packets: A preliminary investigation of SQ3R implementation. Teaching of Psychology, 38(3), 142–146. https://doi.org/10.1177/0098628311411786
- Clinton, V. (2019). Reading from paper compared to screens: A systematic review and meta-analysis. Journal of Research in Reading, 42(2), 288–325. https://doi.org/10.1111/1467-9817.12269
- Cornell University Learning Strategies Center. (n.d.). Textbook reading systems. Retrieved September 27, 2026, from https://lsc.cornell.edu/how-to-study/reading-strategies/textbook-reading-systems/
- Delgado, P., Vargas, C., Ackerman, R., & Salmerón, L. (2018). Don't throw away your printed books: A meta-analysis on the effects of reading media on reading comprehension. Educational Research Review, 25, 23–38. https://doi.org/10.1016/j.edurev.2018.09.003
- Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58. https://doi.org/10.1177/1529100612453266
- Jairam, D., Kiewra, K. A., Rogers-Kasson, S., Patterson-Hazley, M., & Marxhausen, K. (2014). SOAR versus SQ3R: A test of two study systems. Instructional Science, 42(3), 409–420. https://doi.org/10.1007/s11251-013-9295-0
- Johns, J. L., & McNamara, L. P. (1980). The SQ3R study technique: A forgotten research target. Journal of Reading, 23(8), 705–708. https://eric.ed.gov/?id=EJ227607
- Robinson, F. P. (1946). Effective study (Rev. ed.). Harper & Brothers. https://archive.org/details/in.ernet.dli.2015.224377
- University of Birmingham. (2024, September 26). Engaging with the text. Reading strategies: A short guide. https://libguides.bham.ac.uk/c.php?g=689902&p=4939415
- University of Leeds Library. (n.d.). Reading strategies. Retrieved September 27, 2026, from https://library.leeds.ac.uk/info/1401/academic-skills/207/reading/2
- University of Southampton Library. (2026, January 15). Effective reading. Study skills. https://library.soton.ac.uk/effectivereading
- University of Toronto Mississauga Robert Gillespie Academic Skills Centre. (n.d.). Read effectively with SQ3R. Retrieved September 27, 2026, from https://www.utm.utoronto.ca/rgasc/student-resource-hub/study-skills-resources/read-effectively-sq3r