Mark ở vũ trụ khác · Another Universe Papers That Changed How We Learn #7 · Study methods · Research review

Grading 10 ways to study

Dunlosky, Rawson, Marsh, Nathan and Willingham · Psychological Science in the Public Interest · January 2013

The paper"Improving students' learning with effective learning techniques"
Typea 55-page review grading 10 common study techniques
Main result2 techniques rated "high utility": practice testing and spacing
Mark's labelSolid; read the low ratings carefully

Source: Dunlosky et al. (2013), Psychological Science in the Public Interest 14(1), 4–58; press releases from the Association for Psychological Science (APS) and Kent State University, January 2013. Full text checked through excerpts of the PDF in search results.

One-minute recap

Essentials

10 study methods on trialFive scientists reviewed existing research on 10 methods students commonly use.
2 rated highPractice testing and spaced practice. They help many ages on many kinds of tests.
3 rated moderateAsking "why", self-explanation, interleaving. Useful, but the evidence is still thin.
5 rated lowSummarizing, highlighting, keyword mnemonics, imagery while reading, rereading. Including the two most popular.
Still standing 12 years onMeta-analyses in 2021 and 2025 still put spacing and testing on top. The "low" group is not useless.

Source: Dunlosky et al. (2013); Donoghue and Hattie (2021), Frontiers in Education 6:581216; Mawson and Kang (2025), Behavioral Sciences 15(6):771.

Part 1

The paper's question

Are the two most popular ways to study, rereading and highlighting, the most effective? The paper says no. And the two most effective methods are rarely used, because while you study they feel harder.

Source: Dunlosky et al. (2013); survey by Karpicke, Butler and Roediger (2009), Memory 17(4). The "feels harder" point follows the review by Carpenter, Pan and Butler (2022), Nature Reviews Psychology.

Step 1Pick 10 methodsEasy to use, already studied, including the methods students use most.
→
Step 2Ask 4 questionsDoes it help across conditions, learners, materials and kinds of tests?
→
Step 3Rate "utility"High, moderate or low. A judgment by the authors, not a measured number.
Part 2

The discovery story

No lab, no test tubes. This is a "read everything, then grade it" paper, written for teachers, students and parents.

A 2009 survey of 177 students: 84 percent said rereading was one of their study methods, 55 percent ranked rereading as their number one method. how do students study? survey of 177 students, 2009 list rereading as a method 84% rank rereading number 1 55% The 2013 paper graded rereading and highlighting precisely because students report using them so much.
Starting point: a popular method is not necessarily a good one. A 2019 study points out that "rereading" in the survey often means rereading the parts not yet understood, and that toward the end students switch more to testing themselves. See the Fact check.Source: Karpicke, Butler and Roediger (2009), Memory 17(4), 471–479; the 84% figure appears in Dunlosky (2013); the 177 students and 55% follow documents quoting the survey (secondary source). Kuhbandner and Emmerdinger (2019), Memory 27(7), 952–961.
Before 2013 · Hundreds of experiments, scatteredThe spacing effect and the testing effect had been measured for decades but rarely reached learners. A 2006 review of spacing alone pooled 317 experiments from 184 articles.
2009 · A student surveyKarpicke, Butler and Roediger asked students how they study. Rereading came first; self-testing far less.
January 2013 · The paper appearsFive authors at four universities: John Dunlosky and Katherine Rawson (Kent State), Elizabeth Marsh (Duke), Mitchell Nathan (Wisconsin-Madison), Daniel Willingham (Virginia). Published in the APS special-topic journal, written for the public and policy makers.
Fall 2013 · A version for teachersDunlosky wrote "Strengthening the student toolbox" in American Educator, retelling the results for classroom teachers.
2021 and 2025 · Rechecked by meta-analysisDonoghue and Hattie pooled 242 studies on the same 10 methods. Mawson and Kang re-measured spacing using only real classrooms.

Source: Cepeda et al. (2006), Psychological Bulletin 132(3) (317 experiments, 184 articles); Karpicke et al. (2009); Kent State press release, January 2013 (authors, affiliations); Dunlosky (2013), American Educator 37(3), 12–21; Donoghue and Hattie (2021); Mawson and Kang (2025).

Four questions used to grade each method: does it help across learning conditions, kinds of learners, materials and kinds of tests. conditionsalone or in a group,how long, how often learnersage, ability,prior knowledge materialsvocabulary, concepts,long science texts testsrecall, comprehension,new problems each method was asked: does the benefit travel? the more boxes it helps in, the higher its utility
How the grading worked. A method that only helps college students remember word lists in a lab will not rate high, even with a large effect.Source: Dunlosky et al. (2013), introduction to the evaluation (four categories of variables: learning conditions, student characteristics, materials, criterion tasks).
Part 3

The paper in 3 sentences

Of 10 popular study methods, only practice testing (questions, flashcards) and spaced practice (spreading review over several days) are rated "high utility".

Asking "why", explaining to yourself and interleaving problem types are rated "moderate": useful, but not yet tested enough in real classrooms.

Summarizing, highlighting, keyword mnemonics, imagery while reading and rereading are rated "low", including the two methods students use most.

The ranking of 10 methods. High utility: practice testing, spaced practice. Moderate: asking why, self-explanation, interleaving. Low: summarizing, highlighting, keyword mnemonic, imagery while reading, rereading. HIGH2 methods MODERATE3 methods LOW5 methods practice testing spaced practice asking "why" self-explanation interleaving summarizing highlighting keyword mnemonic imagery for text rereading ← the two most popular methods are here too
The paper's report card. "Low" means the benefit is small, narrow or clearly worse than other methods, not that it is zero.Source: Dunlosky et al. (2013), utility summary table; APS press release "Which Study Strategies Make the Grade?" (January 2013).
Why did highlighting and rereading rate low?

Highlighting: people remember what they highlighted better, but they often highlight the wrong parts, and highlighting asks the brain to do little beyond reading.

Rereading: it helps a little, especially after a gap. But compared with testing yourself for the same time, it falls far behind. That relative disadvantage is its biggest strike.

Source: Dunlosky et al. (2013), sections on highlighting and rereading, checked through secondary summaries.

Part 4

How this paper has been received

Each row has a source and a lookup date.

Where and whenPsychological Science in the Public Interest, volume 14 issue 1, January 2013, pp. 4–58. The APS special-topic journal, where each issue is a long review on a question of public interest.Source: article page (DOI 10.1177/1529100612453266); APS press release, January 2013.
CitationsAbout 6,282.Source: John Dunlosky's Google Scholar profile, read through search results on 15 October 2026 (the counter shows no update date; the true figure may be higher). Other databases give other numbers.
Replicated?This is a review, so "replication" means later meta-analyses. Donoghue and Hattie (2021) pooled 242 studies with 169,179 participants: spacing (0.85) and practice testing (0.74) still came out on top. Classroom-only meta-analyses also find both work (see Part 6).Source: Donoghue and Hattie (2021), Frontiers in Education 6:581216, Table 1; Yang et al. (2021), Psychological Bulletin; Mawson and Kang (2025), Behavioral Sciences 15(6):771.
Main criticismDonoghue and Hattie: the "low" group still shows effects of 0.44 or more and should not be dismissed. Kuhbandner and Emmerdinger (2019): the "students only reread" survey has been over-read. Ponce, Mayer and Méndez (2022): highlighting does help college students remember. The authors themselves: much evidence comes from single-session studies, and some methods lack classroom tests.Source: the papers named; Kent State press release, January 2013.
Awards, textbooksNo reliable source found for awards or textbook status, so none is claimed.
VerdictSOLID The two main conclusions stand after 12 years. Read "low" as "worse", not "useless".The label is Mark's assessment.
Part 5

Fact check: what holds and what is myth

14 things often said about the paper, checked against the paper itself and later research. Click each row to see why.

SOLIDThe paper grades 10 common study methods

True: practice testing, spacing, interleaving, asking "why", self-explanation, summarizing, highlighting, keyword mnemonic, imagery for text, rereading (Dunlosky et al. 2013).

SOLIDOnly 2 methods are rated "high utility"

Practice testing and spaced practice, because they help learners of many ages and abilities on many kinds of tests (the paper; APS press release, January 2013).

SOLIDHighlighting is rated "low utility"

In the same group as summarizing, keyword mnemonics, imagery for text and rereading (the paper).

SOLIDLater meta-analysis still puts these two on top

Donoghue and Hattie (2021): spacing 0.85, practice testing 0.74, the highest of the 10. Note: the 0.85 mixes in lab studies (Mawson and Kang 2025).

SOLIDPractice testing works in real classrooms, not only in the lab

Yang et al. (2021), Psychological Bulletin: 222 classroom studies, 48,478 students, effect g = 0.499.

CAUTION"Highlighting is completely useless"

Meta-analysis by Ponce, Mayer and Méndez (2022), 36 articles, 85 effects: learner-generated highlighting improves memory a little (0.36) but not clearly comprehension (0.20); it helps college students (0.39), not clearly school students (0.24). Highlighting done by the instructor helps both memory and comprehension (0.44).

CAUTION"Rereading is useless"

It rated low because it is clearly worse than other methods for the same time, not because it does nothing. Donoghue and Hattie (2021) measured 0.47. Rereading after a gap helps more than rereading straight away.

CAUTION"Almost every student studies only by rereading"

2009 survey: 84% list rereading, 55% rank it number one (Karpicke et al.; the 55% via sources quoting the survey). But Kuhbandner and Emmerdinger (2019) found students mostly reread parts they did not understand, and later switch to testing themselves.

CAUTION"Summarizing is a good way to study"

The paper rates it low: summarizing helps people who already summarize well; most need fairly long training. Donoghue and Hattie (2021): 0.44, lowest of the 10 together with underlining.

CAUTION"Interleaving works for every subject"

Brunmair and Richter (2019), 59 studies: average g = 0.42, strongest for telling paintings apart (0.67), depending on how similar the materials are. The 2013 paper also rated it only "moderate".

CAUTION"Good methods help everyone equally"

The authors say plainly these methods are not a panacea, and work best for motivated students able to use them (Kent State press release). The 2021 meta-analysis found very large differences between studies.

FALSE"It was one big experiment"

The paper ran no new experiment. It is a review: it read existing studies and graded them.

FALSE"The utility rating is a measured number"

It is the five authors' judgment, based on four groups of questions. A numeric measure (a meta-analysis of all 10 methods) only arrived in 2021.

FALSE"The paper also graded learning styles: visual, auditory, kinesthetic"

Learning styles are not among the 10 methods graded. "Everyone has a learning style" has its own piece: Everyone has a learning style.

Source: as noted in each row; full list in Part 10.

Part 6

What the latest research says

The hardest question: outside the lab, in real classrooms, does it still work?

Average effects in recent meta-analyses. Spacing in real classrooms, 2025: 0.54. Practice testing in classrooms, 2021: 0.50. Interleaving, 2019: 0.42. Learner highlighting, memory, 2022: 0.36. Spacing in mathematics only, 2025: 0.28. 00.250.50 spacing, real classrooms (2025)0.54 practice testing, classrooms (2021)0.50 interleaving (2019)0.42 own highlighting, memory (2022)0.36 spacing, mathematics only (2025)0.28 average effect (standard deviations), each row from a different meta-analysis
The top two still stand, but smaller in real classrooms. The numbers come from different meta-analyses with different comparison groups, so only their rough size is comparable.Source: Mawson and Kang (2025), Behavioral Sciences 15(6):771 (22 reports, 31 effects, d = 0.54); Yang et al. (2021), Psychological Bulletin (222 studies, g = 0.499); Brunmair and Richter (2019), Psychological Bulletin (59 studies, g = 0.42); Ponce, Mayer and Méndez (2022), Educational Psychology Review 34 (0.36); Murray, Horner and Göbel (2025), Educational Psychology Review (27 studies, g = 0.28).

2021 · All 10 re-measured

Donoghue and Hattie: 242 studies, average effect 0.56. Spacing and testing highest; underlining and summarizing lowest (0.44) but still positive.

2022 · Why so few use them

Carpenter, Pan and Butler (Nature Reviews Psychology): spacing and testing are underused because of wrong beliefs about learning, lack of awareness, or because they run against intuition.

2025 · Real classrooms only

Mawson and Kang: spacing still helps, d = 0.54, smaller than the 0.85 that mixed in lab work. In mathematics alone (Murray et al.): 0.28.

Source: Donoghue and Hattie (2021); Carpenter, Pan and Butler (2022), Nature Reviews Psychology 1, 496–511; Mawson and Kang (2025); Murray, Horner and Göbel (2025).

Part 7

Assessing the paper

Scores are Mark's assessment based on the sources in Part 10, not data from a study.

Main conclusions hold
5/5
Easy to apply now
5/5
Evidence from real classrooms
3/5
Transparent grading
4/5
Strengths
  • Turns a large body of research into one readable table
  • Speaks plainly about the two most popular methods
  • Both main conclusions confirmed by meta-analysis
Weaknesses
  • Ratings are judgments, not measurements
  • Much evidence from lab, single-session studies
  • "Low" is easily misread as "useless"

A rare "read everything, then grade it" paper that still stands 12 years later. Read it to know where to put your study time, not to ban the highlighter.

Part 8

Lessons: move time from "highlight and reread" to "quiz and space out"

No need to study more hours. Just use the same hours differently.

1 · close the bookinstead of rereadingcards, questions, blank page 2 · space it outinstead of crammingreview before you forget 3 · mix problem typesinstead of 20 of one kindpractise spotting "which kind" 4 · ask "why"instead of colouringexplain it in your own words Studying that feels "harder" is often studying that lasts.
The first two habits are the most solid part of the paper. The other two are "moderate": use them in addition, not instead.Source: ratings from Dunlosky et al. (2013); last line follows Carpenter, Pan and Butler (2022). The concrete steps are Mark's suggestions.
Read once, then close the bookWrite down on a blank page everything you remember, then open the book and compare. What you got wrong is what to study. Replace the second reading with a self-quiz.
Plan reviews from the exam dateExam in one week: review after about 2 days. Exam in one year: review after about 3–5 weeks. A study of over 1,350 people found the best gap shrinks, as a share, the further away the test.Source: Cepeda et al. (2008), Psychological Science 19(11): optimal gap about 20–40% of the time to a 1-week test, 5–10% for a 1-year test.
Mix types once you know eachDo 3 problem types mixed together instead of 10 of type A then 10 of type B. It helps most when the types are easy to confuse.
If you still want to highlightHighlight little, after reading the whole passage, then turn each highlight into a question to test yourself later.
Check your own studying: 5 questions

1. Last week, how many minutes did I spend rereading, and how many quizzing myself without looking?

2. Did I study this subject in one night, or spread it over at least 3 different days?

3. If I close the book right now, how many main points of the last lesson can I write down?

4. My most recent highlight: have I turned it into a question yet?

5. After 1 week, redo 5 old questions: how many are right? Write it down to compare next time.

Part 9

Further reading

Dunlosky (2013), "Strengthening the student toolbox"American Educator 37(3). The author's own short version for teachers. On ERIC (EJ1021069).
Donoghue and Hattie (2021), "A meta-analysis of ten learning techniques"Frontiers in Education. Free to read; see Table 1 for each method's number.
Carpenter, Pan and Butler (2022), "The science of effective learning with spacing and retrieval practice"Nature Reviews Psychology. A recent review of the two top methods.
Papers That Changed How We Learn #6: Testing beats rereadingRoediger and Karpicke's 2006 experiment, the root of the first "high utility" method.
Common Myths #3: Everyone has a "learning style"Why "visual" and "auditory" learning is not on this report card.
Papers That Changed How We Learn #8: The 2 sigma problemBloom (1984): one-to-one tutoring far beats a big class. How true is it?
Part 10

Sources

The main sources are the paper itself and peer-reviewed research. The working environment blocks direct downloads from journal sites, so content was checked through excerpts of those same pages in search results (looked up 15 October 2026), not the full text.

  1. Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., and 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. doi.org/10.1177/1529100612453266
  2. APS (January 2013), press release "Which Study Strategies Make the Grade?". psychologicalscience.org · Kent State University (January 2013), "All Study Strategies Not Created Equal". kent.edu
  3. Dunlosky, J. (2013). Strengthening the student toolbox. American Educator, 37(3), 12–21. ERIC EJ1021069
  4. Karpicke, J. D., Butler, A. C., and Roediger, H. L. (2009). Metacognitive strategies in student learning. Memory, 17(4), 471–479. doi
  5. Kuhbandner, C., and Emmerdinger, K. J. (2019). Do students really prefer repeated rereading over testing when studying textbooks? A reexamination. Memory, 27(7), 952–961. uni-regensburg
  6. Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., and Rohrer, D. (2006). Distributed practice in verbal recall tasks. Psychological Bulletin, 132(3), 354–380.
  7. Cepeda, N. J., Vul, E., Rohrer, D., Wixted, J. T., and Pashler, H. (2008). Spacing effects in learning: A temporal ridgeline of optimal retention. Psychological Science, 19(11), 1095–1102.
  8. Brunmair, M., and Richter, T. (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin.
  9. Donoghue, G. M., and Hattie, J. A. C. (2021). A meta-analysis of ten learning techniques. Frontiers in Education, 6, 581216. frontiersin.org
  10. Yang, C., Luo, L., Vadillo, M. A., Yu, R., and Shanks, D. R. (2021). Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review. Psychological Bulletin.
  11. Ponce, H. R., Mayer, R. E., and Méndez, E. E. (2022). Effects of learner-generated highlighting and instructor-provided highlighting on learning from text: A meta-analysis. Educational Psychology Review, 34, 989–1024. springer
  12. Carpenter, S. K., Pan, S. C., and Butler, A. C. (2022). The science of effective learning with spacing and retrieval practice. Nature Reviews Psychology, 1, 496–511.
  13. Murray, E., Horner, A. J., and Göbel, S. M. (2025). A meta-analytic review of the effectiveness of spacing and retrieval practice for mathematics learning. Educational Psychology Review. White Rose
  14. Mawson, R. D., and Kang, S. H. K. (2025). The distributed practice effect on classroom learning: A meta-analytic review of applied research. Behavioral Sciences, 15(6), 771. PMC12189222
  15. Citation count: John Dunlosky's Google Scholar profile (via search results, looked up 15 October 2026). The 177 students and 55% of the 2009 survey follow a Queen's University handout and Aljabri (2024), Frontiers in Education (secondary sources).