Study Methods That Work: 9 Techniques Ranked by Research

The study methods that work best, according to the research, are testing yourself and spreading your study across several days. A widely cited 2013 review by John Dunlosky and four colleagues examined ten common techniques and rated only those two as high utility. Three more earned a moderate rating. The habits most students actually use, rereading and highlighting, landed at the bottom.

This page ranks nine techniques by the strength of that evidence, shows how to use each in a 20-minute block, and says plainly where each one fails. Short on time before exams? Our 2-week plan for finals puts these methods on a calendar.

Where does the ranking come from?

Two research summaries do most of the work here. The first is Dunlosky, Rawson, Marsh, Nathan and Willingham, “Improving Students’ Learning With Effective Learning Techniques”↗, published in Psychological Science in the Public Interest in 2013. The authors rated ten techniques as high, moderate or low utility based on how well each held up across different learners, materials, tests and time delays.

The second is the federal Institute of Education Sciences (IES) practice guide Organizing Instruction and Study to Improve Student Learning↗ (Pashler et al., 2007), which grades its recommendations as strong, moderate or low evidence.

A third source adds numbers. Donoghue and Hattie’s 2021 meta-analysis in Frontiers in Education↗ pooled 242 studies (1,619 effects, 169,179 participants) on Dunlosky’s ten techniques. The overall average effect was d = 0.56, and distributed practice (d = 0.85) and practice testing (d = 0.74) came out on top. An effect size (d) compares a group using the technique with a comparison group in standard-deviation units; higher means a bigger gain.

Seven of the nine techniques below come from Dunlosky’s list. Two (worked examples and combining words with pictures) come from the IES guide, which rated both moderate. Dunlosky’s other low-rated techniques (highlighting, the keyword mnemonic and imagery for text) are covered in the table but not given their own section.

The ranking at a glance

Table: nine study techniques ranked by evidence. Ratings from Dunlosky et al. (2013) and the IES practice guide (Pashler et al., 2007); effect sizes from Donoghue and Hattie (2021).

Two things stand out. Spacing and self-testing lead on every measure. And the effect sizes for the “low” techniques are not zero: rereading still beats doing nothing. Dunlosky’s ratings weigh how reliably a technique works across students, materials and tests, which is why imagery can post a decent average effect and still be rated low.

1. Practice testing: why does quizzing yourself work so well?

Pulling information out of memory strengthens it more than putting it in again. Dunlosky’s review rated practice testing high utility because the benefits held across ages and materials, and his summary for teachers notes that tests requiring recall from memory help most. The IES guide rates using quizzes to re-expose key content as strong evidence, its top grade.

The classic experiment shows why rereading fools people. In Roediger and Karpicke’s 2006 study in Psychological Science, students who reread a passage four times beat the self-testing group on a test five minutes later. One week later the result flipped: the group that studied once and tested itself three times recalled 61%, the study-three-times-test-once group 56%, and the reread-four-times group 40%. The rereaders had read the passage 14.2 times on average; the testing group, 3.4 times.

In 20 minutes: 5 minutes writing everything you remember about one topic on a blank page; 10 minutes checking your notes and marking gaps; 5 minutes on flashcards or practice questions for the gaps only.

Where it fails: testing yourself on material you never understood just produces wrong answers. Learn it once first. And flashcards that you flip too early turn back into rereading. Say the answer before you look.

AI chatbots can generate practice questions in seconds. Our AI study workflow shows how to use them for quizzing without letting the tool do the remembering.

2. Spacing: how far apart should study sessions be?

The same hours spread over several days beat one long session for anything you need later. Dunlosky’s review rated distributed practice high utility; the IES guide rates spacing moderate evidence and recommends a second exposure at least several weeks, and preferably several months, after the first. Its classroom example: eighth graders who reviewed US history facts 16 weeks after learning them scored almost 100% higher on a test nine months later than students who reviewed after one week.

How big should the gap be? In Cepeda and colleagues’ 2008 study of more than 1,350 people, the best gap grew with the time until the test: about 20% to 40% of the delay for a test one week away, falling to about 5% to 10% for a test a year away. For a test in a week, that means reviewing after one to three days.

In 20 minutes: today, learn and self-test topic A. In two or three days, do a 10-minute recall quiz on topic A before starting topic B. A week later, quiz A and B together.

Where it fails: spacing needs a plan, and it feels slower because you forget a little between sessions. That forgetting is part of why it works. Put review sessions in a homework planner so they actually happen.

3. Interleaving: should you mix up problem types?

Yes, especially in math and science, where the hard part is choosing the method. Interleaved practice means mixing problem types (say, area, volume and perimeter questions shuffled together) instead of doing ten of one type in a row. Dunlosky’s team rated it moderate, partly because most of the evidence came from math and problem solving, and Dunlosky notes the benefits do not extend to all disciplines.

In 20 minutes: pick six problems from three different sections of the chapter and do them in shuffled order. For each one, first write which method it needs.

Where it fails: mixing feels harder and your practice scores drop at first, which tempts people to stop. It also helps less on brand-new material. Learn each type once, then mix.

4. Elaborative interrogation: asking “why?”

Asking why a fact is true links it to what you already know. Dunlosky’s review rated elaborative interrogation moderate; the IES guide rates asking deep explanatory questions (“why,” “how,” “what if”) as strong evidence.

In 20 minutes: take ten facts from your notes. For each, write one sentence answering “why would this be true?” Check the answers you are unsure about in the textbook.

Where it fails: it works best when you already know something about the topic. With zero background, the “why” answers can be guesses that make errors stick. Check them.

5. Self-explanation: talking yourself through each step

Explaining how each step follows from the last shows you exactly where your understanding breaks. Dunlosky’s review rated self-explanation moderate. It suits math, science and any multi-step process.

In 20 minutes: take a solved problem from class. Under each line, write one sentence explaining why that step happens. Then solve a fresh problem while saying the reasons out loud.

Where it fails: it takes longer than just doing problems, and explanations that only restate the step (“then I subtract”) add little. Aim for the reason, not the action.

6 and 7. Worked examples and words plus pictures

These two come from the IES guide, which rates both moderate.

Worked examples, then problems. Alternate between studying a fully solved problem and solving a similar one yourself. Beginners learn a method faster this way than by struggling alone. The catch: as you get more skilled, the benefit shrinks, so switch to plain practice once you can solve problems without looking. Our math homework help guide uses this routine.

Words plus pictures. Pair your notes with a diagram, timeline, graph or flowchart, and connect abstract ideas to a concrete example. This is the honest version of “visual learning”: it helps nearly everyone, not just people who call themselves visual learners.

8 and 9. Why do summarizing and rereading rank so low?

They make material feel familiar without testing whether you can use it. Dunlosky’s review rated summarization low, noting that students need training to write good summaries, and rated rereading low because it helps recall more than understanding and its benefits may not last. Dunlosky’s summary also cites a study in which highlighting hurt students’ ability to draw inferences. Highlighting was rated low for similar reasons.

Familiarity is the trap. After a second read, the page looks known, which feels like learning. Then the test asks you to produce the answer from nothing, which you never practiced.

Upgrade instead of dropping them: reread once, then close the book and recall. Summarize from memory, not from the page, then check.

What about visual, auditory or hands-on learners?

The evidence does not support studying “to your learning style.” In a 2008 review in Psychological Science in the Public Interest, Pashler, McDaniel, Rohrer and Bjork found no adequate evidence that matching teaching to a person’s preferred style (visual, auditory, kinesthetic) improves learning.

Preferences are real. People do enjoy some formats more, and enjoying a method can help you start. But the techniques that work, recall, spacing and explaining, work for everyone. If you like diagrams, draw them, then test yourself from the diagram with the labels covered. If you like hands-on work, build the model, then explain it without notes.

Picking the right method: if this, then that

Table: matching study methods to the task.

If starting is the real problem, technique is not the fix yet. Our guide to homework motivation covers that, and students with attention difficulties can find adapted versions of these methods in our ADHD study tips.

Honest limits of this ranking

Most of the studies behind these ratings were run with college students and specific materials, often in labs. Dunlosky’s team rated techniques partly on how widely they had been tested, so “moderate” can mean “promising but less studied” rather than “weaker.” And no technique replaces sleep, a reasonable workload and actually understanding the material first. Use a classroom timer to keep sessions short and honest, and the students’ hub for the rest of our study guides.