PSY DISCUSSION 5

Strengthening the Student Toolbox

Study Strategies to Boost Learning

BY JOHN DUNLOSKY

t's the night before her biology exam, and the high school student has just begun to study. She talces outher highlighter and reads hertextbook, marking it up as she goes along. She rereads sentences that seem most important and stays up

most ofthe night, justhoping to get a good enough grasp ofthe material to do well on the exam. These are studystrategies that she may have learned from her friends orherteachers orthatshe simply took to onherown. She is not unusual in this regard; many students rely on strategies such as highlighting, rereading, and cramming the night before an exam.

Quite often, students believe these relatively ineffective strate-

John Dunlosky isa professorofpsychologyandthedirectorofexperimental training atKent State University. His research focuses on self-regulated learning and how itcan be used to improve student achievement across the lifespan.

gies are actually the most effective, 1 and at least on the surface they do seem sound, perhaps because, even after pulling an all­ nighter, students manage to squeal<byonexams. Unfortunately, ina recent review ofthe research, my colleagues and I found that these strategies are notthateffective,2 especially if students want to retain their learning and understanding ofcontent well after the exam is over-obviously, an important educational goal.

So, why aren't students learning about the best strategies? I canonly speculate, butseveral reasons seemlikely. Curricula are developed to highlightthe contentthat teachers should teach, so the focus is on providingcontent and not on training students how to effectively acquire it. Put differently, the emphasis is on whatstudents need to learn, whereas little emphasis-ifany-is placed on training students how they should go about learning the content and what skills will promote efficient studying to support robust learning. Nevertheless, teaching students how to learn is as important as teaching them content, because acquir-

12 AMERICAN EDUCATOR I FALL 2013

ing both the right learning strategies and background knowledge is important-ifnot essential-for promoting lifelong learning.

Another reason many students may not be learning about effective strategies concerns teacher preparation. Learning strat­ egies are discussed in almost every textbook on educational psychology, so many teachers likely have been introduced to at least some of them. Even so, my colleagues and I found that, in large part, the current textbooks do not adequately cover the strategies; some omitdiscussion of the most effective ones, and mostdo not provide guidelines on howto use them in the class­ room or on howto teach students to usethem. In some cases, the strategies discussed have limited applicability or benefit.3 So I sympathize with teachers whowantto devote some class time to teaching students how to learn, because teacher preparation typically does not emphasize the importance of teaching stu­ dents to use effective learning strategies. Moreover, given the demands ofday-to-day teaching, teachers do not have time to figure outwhich strategies are best.

The good news is that decades of research has focused on evaluating the effectiveness of many promising strategies for helping students learn. Admittedly, the evidence for many of these strategies is immense and not easily deciphered, especially given thetechnical nature ofthe literature. Thus, to help promote the teaching and use of effective learning strategies, my col­ leagues* and I reviewed the efficacyof1Olearning strategies:

1. Practice testing: self-testing or taking practice tests on to-be-learned material.

2. Distributed practice: implementing a schedule ofpractice that spreads out study activities over time.

3. Interleaved practice: implementing a schedule ofpractice that mixes different kinds ofproblems, or a schedule of snidy that mixes different kinds ofmaterial, within a single study session.

4. Elaborative interrogation: generating an explanation for why an explicitly stated fact orconcept is true.

5. Self-explanation: explaining how newinformation is related to known information, orexplaining steps taken during problem solving.

6. Rereading: restudying text material again after an initial reading.

7. Highlightingand underlining: marking potentially important portions ofto-be-learned materials while reading.

8. Summarization: writing summaries ( ofvarious lengths) of to-be-learned texts.

9. Keyword mnemonic: using keywords and mental imagery to associate verbal materials.

10. Imagery for text: attempting to form mental images oftext materials while reading or listening.

Before describing the strategies in detail, I will putinto context a few aspects ofour review. First, our intent was to survey strate-

*My collaborators on this project were cognitive and educational researchers Katherine A. Rawson, Elizabeth J. Marsh, Mitchell J. Nathan, and Daniel T Willingham. Willingham regularly cont ributes to American Educator in his "Ask the Cognitive Scientist" column.

gies thatteachers could coach snidents to usewithout sacrificing too much class time andthat any student could use. We excluded a variety of strategies and computer-driven tutors that show promise but require technologies that may be unavailable to many students. Although some ofthe strategies we reviewed can be implemented with computer software, they all can be used successfullybya motivated student who (at most) has access to a pen or pencil, some index cards, and perhaps a calendar.

Second, we chose to review some strategies ( e.g., practice test­ ing) because an initial survey suggested that they were relatively effective,4 whereas we chose other strategies ( e.g., rereading, highlighting) because students reported using them often yet we wondered about their effectiveness.

Finally, the strategies differ somewhat with respect to the kinds of learning they promote. For instance, some strategies (e.g., keyword mnemonic, imagery for text) are focused on improving students' memory for core concepts or facts. Others (e.g., self-explanation) may best serve to promote students' comprehension ofwhat they are reading. And still others ( e.g., practice testing) appear to be useful for enhancing both memory and comprehension.

Inthefollowing sections, I discuss each ofthe learning strate­ gies, beginning with those thatshow the most promise for improv­ ing student achievement.

The Most Effective Learning Strategies We rated two strategies-practice testing and distributed prac­ tice- asthe most effective ofthose we reviewed because they can help students regardless ofage, they can enhance learning and comprehensionofa large range ofmaterials, and, most important, they can boost student achievement.

AMERICAN EDUCATOR I FALL 2013 13

Practice Testing

Test, exam, and quizare four-letter words that provoke anxiety in many students, ifnotsome teachers aswell. Such anxiety may not be misplaced, given thehigh stakes ofstatewide exams. However, by viewing tests as the end-all assessments administered only after learning is complete, teachers and students are missing out onthe benefits ofone of the most effective strategies for improving student learning.

In 1909, a doctoral studentat the University ofIllinois dem­ onstratedthat practice tests improve student performance, 5 and morethan 100 years ofresearch has revealed thattaking practice tests (versus merely rereading the material to be learned) can substantially boost student learning. For instance, college stu­ dents who reported using practice tests to study for upcoming exams earned higher grades, 6 andwhen middle school teachers administered daily practice tests for class content, their students performed better on future tests that tapped the content they had practiced duringthe daily tests. 7

All of the strategies we reviewed can be used successfully by a motivated student who (at most) has access to a pen or pencil, some index cards, and perhaps a calendar.

The use ofpractice tests can improve student learning in both direct and indirect ways.8 Consider two students who have just read a chapter in a textbook: Both students review the most important information in the chapter, butone student reads the information again, whereas the other student hides the answers and attempts to recall the information from memory. Compared with the first student, the second student, by testing himself, is boosting his long-term memory. Urns, unlike simply reading a text, whenstudents correctly retrieve an answer from memory, the correct retrieval can have a direct effect on memory.

Practice tests canalso have an indirect effect onstudent learn­ ing. When a student fails to retrieve a correct answer during a practice test, that failure signals that the answer needs to be restudied; in thisway, practice tests can help students make better decisions about what needs further practice andwhat does not. In fact, most students who use practice tests report that they do so to figure out what they know and do not know.9

Based on the prevailing evidence, how might students use practice tests to best harness the power ofretrieval practice? First, student learning can benefit from almost any kind ofpractice test, whether it involves completing a short essay where students need to retrieve content from memory or answering questions in a multiple-choiceformat Research suggests, however, thatstudents will benefit most from tests that require recall from memory, and not from tests that merely ask them to recognize the correct answer.t0 Toey may needto work a bit harder to recall key materi­ als ( especially lengthy ones) from memory, but the payoffwill be great inthe long nm. Another benefit ofencouraging students to

recall key information from memory is that it does not require creating a bank oftest questions to serve as practice tests.

Second, students should be encouraged to take notes in a manner thatwill foster practice tests. For instance, as they read a chapter in their textbook, theyshould beencouraged to make flashcards, with the key term on oneside and the correct answer on the other. When taking notes in class, teachers should encourage students to leave room on each page ( or on the back pages of notes) for practice tests. In both cases, as the material becomes more complex ( and lengthy), teachers should encour­ age students to write down their answers when they are testing themselves. For instance, when they are studying concepts on flashcards, they should first write down the answer ( or defini­ tion) of the concept they are studying, and then they should compare their written answer with the correct one. For notes, they can hide key ideas or concepts with their hand and then attempt to write them out in the remaining space; by using this strategy, they can compare their answer with the correct oneand easily keep track oftheir progress.

Third, and perhaps most important, students should continue testing themselves, with feedback, until they correctly recall each concept at least once from memory. For flashcards, if they correctly recall an answer, they can pull the card from the stack; ifthey do not recall it correctly, they should place it at the back ofthe stack. For notes, they should try to recall all ofthe impor­ tant ideas and concepts from memory, and then go back through their notes once again and attempt to correctly recall anything they did not get right during their first pass. If students persist until they recall each idea orconcept correctly, they will enhance their chances of remembering the concepts during the actual exam. They should also be encouraged to "get it right" on more than one occasion, such as by returning to the deck ofcards on another day and relearning the materials. Using practice tests may not come naturally to students, so teachers can play an important role in informing them about the power of practice tests and how they apply to the content being taught in class.

Not only canstudents benefit from using practice tests when studying alone, but teachers cangive practice tests in the class­ room. The idea is for teachers to choose the most important ideasand then take a couple minutes atthe beginningorendof each class to test students. After all students answer a question, teachers can provide the correct answer and give feedback. 'Ihe more closely the practice questions tap the same information that will be tested on the in -class examination, the better stu­ dents willdo. Thus, this in-class "testing time" should be devoted to the most critical information that will appear on the actual exam. Even using the same questions during practice and during the test is a reasonable strategy. Itnot only ensures that the stu­ dentswill be learningwhatteachers have decided is most impor­ tant, but also affirms to students that they should take the in-class practice quizzes seriously.

Distributed Practice

A second highly effective strategy, distributed practice is a straightforward and easy-to-use technique. Consider the follow­ ing examples:

A first-grader studies for a spelling test. Using a worksheet to guide her practice, she might take one oftwo approaches. She

14 AM ERICAN EDUCATOR I FALL 2013

could practice spelling the words by writing each one several times directly below the word printed on the sheet. After practic­ ingone word repeatedly, she would move onto the next oneand practice writing that word several times below it. This kind of practice is called massed practice, because the studentpractices each word multiple times together, before moving to the next one.

An alternative strategy for the student would be to practice writing each word only once, and after transcribing the final word, going back and writing each one again, and so forth, until the practice is complete. This kind ofpractice is called distributed practice, because practice with any oneword is distributed across time ( and the time bet\'lleen practicing any one word is filled with another activity-in this case, writing other words).

In this example, the student either masses or distributes her practices during a single session. Now, imagine an eighth-grader trying to learn some basic concepts pertaining to geology for an upcoming in-class exam. He might readover his notes diligently, in a single session thenight before the exam, until he thinks he is ready for the test- a study tactic called cramming, which practi­ cally all students use. Or, as an alternative, he might study his notes and texts during a shorter session several evenings before the exam and then study them again the evening before. In this case, the studentdistributes his studying across two sessions.

Students will retain knowledge and skills for a longer period oftime when they distribute theirpractice thanwhen they mass it,11 even ifthey use the same amount oftirne massing and dis­ tributingtheirpractice.* Unfortunately, however, manystudents believe that massed practice is better than distributed practice. 12

One reason for this misconception is that students become familiar and facile with the target material quickly during a massed practice session, but learning appears to proceed more slowly with distributed practice. For instance, the first-grader quickly writes the correct word after practicing it several times in succession, but when the same practice is distributed, she may still struggle after several attempts. Likewise, the eighth­ grader may quickly become familiar with his notes after reading them twice during a single session, but when distributing his practice across two study sessions, he may realize how much he has forgotten and use extra time getting back up to speed.

Inboth cases, learning itself feels tougher when itis distributed insteadofmassed, butthe competencyandlearning thatstudents may feel ( and teachers may see) during massed practice is often ephemeral. By contrast, distributed practice may take more effort, but it is essential for obtaining knowledge in a manner that will be maintained (or easily relearned) over longer, educationally relevant periods oftime.

Most students, whether they realize it or not, use distributed practice to master many different activities, but not when they are studying. For instance, when preparing for a dance recital, most would-be dancers will practicetheroutine nightly until they have it down; they will not just do all the practice the night before the recital, because everyone knows that this kind of practice will

"To learn more about massed versus distributed practice, see Daniel T. Willingham's article, "Allocating Student Study Time," in t he Summer 2002 issue of American Educator, available at www.af t.org/newspubs/periodicals/ae/summer2002/ willingham.cfm.

likely not be successful. Similarly, when playing video games, students see their abilities and skills improve dramatically over time in large part because they keep coming back to play the game in a distributed fashion. In these and manyothercases, students realizethat more practice orplay during a current session will not h elp much, and they may even see their performance weaken near the end ofa session, so, of course, they take a break and return tothe activity later. However, for whatever reason, students don't typically use distributed practice as they work toward mas­ tering course content.

The use of practice tests can improve student learning in both direct and indirect ways.

Not using distributed practice for study is unfortunate, because the empirical evidence for the benefits of distributed (over massed) practice is overwhelming, and the strategy itself is rela­ tively easy to understand and use. Even so, I suspect that many students willneedto learn how to use it, especially for distributing practice across multiple sessions. Tue difficulty is simply tl1at most students begin to prepareand study only when they are reminded that the next exam is tomorrow. By that point, cramming is their only option. To distribute practice over time, students should set aside blocks oftime throughout each week to study the content for each class. Each study block will be briefer than an all-night cramsession, and it should involve studying ( and using practice tests) for material that was recently introduced in class and for material they studied in previous sessions.

To use distributed practice successfully, teachers should focus on helping students map out how many study sessions they will

AMERICAN EDUCATOR I FALL 2013 15

need before an exam, when those sessions should take place ( such as which evenings of the week), and what they should practice during each session. For any given class, two short study blocks per week may be enough to begin studying new material and to restudy previously covered material.

'� \::: .. ';;; •�' - ..... � ' .. . .r--­

·- , .. .,,.,.. .,..� ....... -� ,.., t �• J -�7l-. I

--� .....

► -

-

Students wil I retain knowledge for a longer period of time when they distribute their practice than when they mass It.

Ideally, students will use practice tests to study the previously covered material. H they do, they will quickly retrieve the previ­ ously learned material after just a han.dful of sessions, which will leave more time for studying new material. Of course, students may need help setting up their study schedules ( especially when they are younger), and they may need some encouragement to use the strategy. But by using distributed practice (especially if it is combined with practice testing), many students will begin to master material they never thought they could learn.

Teachers can also use distributed practice in the classroom. The idea is to return to the most important material and con­ cepts repeatedly across class days. For instance, if weekly quiz­ zes are already being administered, a teacher could easily include content that repeats across quizzes so students will relearn some concepts in a distributed manner. Repeating key points across lectures not only highlights the importance of the content but also gives students distributed practice. Administer-

16 AMERICAN EDUCATOR I FALL 2013

ing a cumulative exam that forces students to review the most important information is another way to encourage them to study content in a distributed fashion. Admittedly, using cumu­ lative exams may seem punitive, but if the teacher highlights which content is most likely to be retested (because it is the most important content for students to retain), then preparing for a cumulative exam does not need to be daunting. In fact, if stu­ dents continue to use a distributed practice schedule throughout a class, they may find preparing for a final cwnulative exam to be less difficult than it would be otherwise because they will already be well versed in the material.

Strategies with Much Promise

We rated three additional strategies as promising but stopped short of calling them the most effective because we wanted to see additional research about how broadly they improve student learning.

Interleaved Practice

Interleaved practice involves not only distributing practice across a study session but also mixing up the order of materials across different topics. As I discussed above, distributed practice trumps massed practice, but the former typically refers to dis­ tributing the practice of the same problem across time. Thus, for spelliJ1g, a student would benefit from writing each word on a worksheet once, and then cycling through the words until each has been spelled correctly several times. Interleaved practice is similar to distributed practice in that it involves spacing one's practice across tiJne, but it specifically refers to practicing differ­

ent types of problems across time. Consider how a standard math textbook ( or most any science

textbook) encourages massed practice: In a text for pre-algebra, students may learn about adding and subtracting real numbers, and then spend a block of practice adding real numbers, followed by a block of practice subtracting. The next chapter would introduce multiplying and dividing real numbers, and then practice would focus first 011 multiplying real uumbers, and then on dividing them, and so forth. Thus, students are massing their practice of similar problems. n1ey practice several iustances of one type of math prob­ lem ( e.g., addition) before practicing the next type ( e.g., subtrac­ tion). In this example, interleaving would involve solving one problem from each type (adding, subtracting, multiplying, and dividing) before solving a new problem from each type.

One aspect of massed practice that students may find appeal­ ing is that their performance will quickly improve as they work with a particular problem. Unfortunately, such fluent perfor­ mance can be misleading; students believe that they have learned a problem well when in fact their learning is fleeting.

Interleaved practice has not been explored nearly as much as practice tests or distributed practice, but initial research out­ comes have shown that interleaved practice can dramatically improve student achievement, especially in the domain of prob­ lem solving.

A study in which college students learned to compute the volume of four differeut geometric solids illustrates this advan­ tage. 13 In two practice sessions (separated by a week), a student either had massed practice or interleaved practice. For massed practice, students had a brief tutorial on solving for the volume

ofone kind ofsolid ( e.g., a wedge), and then immediately prac­ ticed solving for the volume of four different versions of the particular solid (e.g., finding the volume of four different wedges). They then received a tutorial on finding the volume of another kind ofsolid (e.g., a spherical cone), and immediately practiced solving four versions of that solid ( e.g., finding the volume of four different spherical cones). Tuey repeated this massed practice for two more kinds of solids.

For interleaved practice, students first were given a tutorial on how to solve for the volume ofeach ofthe four solids, and then they practiced solving for each ofthe four versions ofsolids in turn. They never practiced the same kind of solid twice in a row; theypracticed solving for the volume ofa wedge, followed by a spherical cone, followed by a spheroid, and so forth, until they had practiced four problems ofeach type. Regardless of whether practice was massed or interleaved, all students prac­ ticed solving four problems ofeach type.

How did the students fare? Toe results presented in Figme 1 (on the right) show that during the practice sessions, perfor­ mance finding the correct volumes was considerably higher for massed practice than for interleaved practice, which is why some students (and teachers) may prefer massed practice. Toe reason not to stick with massed practice is revealed when we examine performance on the exam, which occurred one week after the final practice session. As shown in the bars on the far right of Figure 1, students who massed practice performed horribly. By contrast, those who interleaved did three times better on the exam, and their performance did not decline compared with the original practice session! Ifstudents who interleaved had prac­ ticed just a couple more times, no doubt they would have per­ formed even better, but the message is clear: massed practice leads to quick learning and quick forgetting, whereas interleaved practice slows learning but leads to much greater retention.

Research shows that teachers can also use this promising strategy with their students. Across 25 sessions, 14 college stu­ dents with poor math skills were taught algebra rules, such as how to multiply variables with exponents, how to divide vari­ ables with exponents, and howto raise variables with exponents to a power. In different sessions, either a single rule was intro­ duced ora rule thathad alreadybeen introduced was reviewed. Most important, during review sessions, students either (a) practiced the rule from the previous session ( which was analo­ gous to massed practice), or (b) practiced the rule from the previous session intermixed with the practice ofrules from even earlier sessions (which was analogous to interleaved practice).

During the first practice sessions, the two groups achieved at about the same level. By contrast, on the final test, performance was substantially better for students who had interleaved prac­ tice than for those who had massed practice. This interleaving advantage was evident both for application ofthe rules to new algebra problems (i.e., different versions of those that the stu­ dents had practiced) and on problems that required the novel combination ofrules. Given that the review sessions were basi­ cally practice tests, one recommendation is sound: when creat­ ing practice tests for students ( whether to be completed in class or at home), it is bestto mix up problems ofdifferent kinds. Even though students initially may struggle a bit more, they will ben­ efit in the long run.

Why does interleaving work so well? In contrast to massed practice, interleaving problems requires distributing practice, which by itself benefits student achievement. Moreover, massed practice robs students ofthe opportunity to practice identifying problems, whereas interleaved practice forces students to prac­ tice doing so. When students use massed practice, after they correctly solve a problem or two of a certain type, they can almost robotically apply the samestepstothe next problem. lhat is, they do not have to figure out what kind of problem they are solving; they just have to apply the same rules to the next prob-

For interleaved practice, when a new problem is; presented, students need to first figure out which kind of problem it is and what steps they need to take to solve it.

100

90 ■ Massed Interleaved

80

70

~ 50 111... ::,

tl 40 <l:

30

20

10

0 Practice Performance Test Performance

Accuracy at solving problems during practice session and on the delayed criterion test.

SOURCE, JOI-IN DUNLOSKY. KATHERINE A. RAWSON, ELIZABETH J. MARSH, MITCHfil J. NATHAN. AND DANl£1.'[ WUINGHAM, "IMP1\0V1'1G snJDENTS' LEARNING WITH EFFECTIVELEARNING TECH>IQUESc PROMISING DIRECTIONS FROM COGNITIVE AND EDUCATIONAL PSYCHOLOGY,• PSYOKJI.OGICAL SOENCE JN THE PUBLJ<: INTfREST 14. NO. 1 (20 \ 3~ 40. DATA FROM DOUGROHRER AND KEW TAYI.OR, "THE SHUFR.INGOF MATHEMATICS Fl!OBlfMS IMFl!OVES LEARNING.• JNSTRUCOONAl SOENCE 35, NO. 6 (2007): 481-498. REPRINTED BY PERMISSION OFSAGE PU8UCATIONS.

AMERICAN EDUCATOR I FALL 2013 17

lem. For interleaving, when a new problem is presented, stu­ dents need to first figme outwhich kind ofproblemit is and what steps they need to take to solve it. This is often a difficult aspect ofsolving problems.

Interleaving has been shown to improve performance (as compared with massed practice) in multiple domains, including fourth-graders learning to solve math problems, engineering

students learning to diagnose system failures, college students learning artists' styles, and even medical students learning to interpret electrocardiograms to diagnose various diseases. Nev­ ertheless, the benefits do not extend to all disciplines; for instance, inonestudy,15 college students learned French vocab­ ulary from different categories (body parts, dinnerware, foods, etc.), and students did just as well when their practice was massed within a category as when itwas interleaved across cat­ egories. In another study, interleaving did not help high school students learn various rules for comma usage. 16

Certainly, much more research is needed to better under­ stand when interleaving will be most effective. Nevertheless, interleaved practice has shown more than enough promise for boosting student achievement to encourage its use, especially given that it does not hurt learning. To that end, I suggest that teachers revise worksheets that involve practice problems, by rearranging the order of problems to encourage interleaved practice. Also, for any in-class reviews, teachers should do their best to interleave questions and problems from newly taught materials with those from prior classes. Doing so not only will allow students to practice solving individual problems, but it also will help them practice the difficult tasks of identifying problems and choosingthe correct steps needed to solve them.

Elaborative Interrogation and Self-Explanation

Elaborative interrogation and self-explanation are two addi­ tional learning strategies that show a lot ofpromise. Imagine a student reading an introductory passage on photosynthesis: "It is a process in which a plant converts carbondioxide and water into sugar, which is its food. 1he process gives off oxygen." Ifthe student were using elaborative interrogation while reading, she would tryto explain why this fact is true. In this case, she might think that it must be true because everything that lives needs some kind offood, and sugar is something that sheeatsas food. She may not come up with exactly the right explanation, but trying to elaborate on why a fact may be true, even when the explanations are notentirelyon the mark, canstill benefit under­ standingand retention.

Students who solve new problems that involve transferring what was learned during practice perform better when they use self-explanation techniques.

Ifthe student were using self-explanation, then shewould try to explain howthisnewinformation is related to informationthat shealready knows. In this case, perhaps she might consider how the conversion is like how herown body changes food into energy and other (not-so-pleasant-as-oxygen) fumes. Students can also self-explain when they solve problems ofany sort anddecide how to proceed; they merely explain to themselves why they made a particular decision.

While practicing problems, the success rate ofsolving them is no different for students who self-explain their decisions com­ pared with those who do not However, in solving newproblems that involve transferring what one has learned during practice, those who initially used self-explanation perform better than those who did not use this technique. In fact, in one experiment where students learned to solve logical-reasoning problems, final test performance was three times better ( about 90 percent versus lessthan 30 percent) for students who self-explained during prac­ tice than for those who did not.17

One reason these two strategies can promote learning and comprehension and boost problem-solving performance is that they encourage students to actively process the content they are focusing on and integrate it with their prior knowledge. Even young students should have little trouble using elaborative inter­ rogation, because it simply involves encouraging them to askthe question "why?" when they are studying. 1hedifference between this type of"why" andthe "why" asked in early childhood ( when this is a common question to parents) is that students must take the time to develop answers. 1hisstrategy may beespecially useful as students are reading lengthy texts in which a set of concepts

18 AMERICAN EDUCATOR I FALL 2013

builds across a chapter, although admittedly the bulk of the research on elaborative interrogation has been conducted with isolated facts. At a minimum, theresearch has shown thatencour­ agingstudents to ask «why" questions about facts or simple con­ cepts that arise in class and in lengthy discussions benefits their learningand understanding.

In mostofthe research onself-explanation, students are given little instruction on howto use the strategy; instead, they are just told to use a particular question prompt that is most relevant to whattheyare studying. For instance, iftheyaresolving a problem, they might be instructed to ask themselves, "Why didI just decide to do X?" ( where Xis anymove relevant to solvingthe problemat hand). And if they were reading a text, they might be instructed to ask, "What doesthis sentence mean tome? What new informa­ tion does the sentence provide, and how does it relate to what I already know?" To take full advantage ofthis strategy, students need to try to self-explain and not merely paraphrase (or sum­ marize) whatthey are doing orreading, because the latter strate­ gies ( as I discuss below) do not consistently boost performance.

Rereading has inconsistent effects on student learning, and benefits may not be long-lasting.

Some potential limitations ofusing these strategies are rather intuitive. For instance, students with no relevant knowledge about a newcontent area mayfind itdifficult- ifnotimpossible- to use elaborative interrogation, because these students may not be able to generate any explanation about why a particular (new) fact is true.* Thus, although research shows that sn1dents as young as those in the upperelementary grades carisuccessfully useelabo­ rative interrogation, the technique may not be so useful for younger students with low levels of background knowledge. As students learn more about a particular topic, elaborative inter­ rogation should be easier to use and will support more learning.

As for self-explanation, it should not be too difficult, or require much time, to teach most students how to take advantage ofthis strategy. Nevertheless, younger students or thosewho need more support maybenefit from some coaching. For instance, as noted above, paraphrases and self-explanations are notthe same and lead to different learning outcomes, so teachers should help younger students distinguish between an explanationofan idea and its paraphrase. Even so, a gentle reminderto use elaborative interrogation or self-explanation may be all most students need to keep themusing these strategies as theylearnnewcourse con­ tent and prepare for examinations.

Because they show promise, I recommend that teachers tell their students about these strategies and explain the conditions

*For more on why reading comprehension depends largely on knowledge, see "Building Knowledge" and "How Knowledge Helps" in the Spring 2006 issue of American Educator, available at www.aft.org/newspubs/periodicals/ae/sprlng2006/ index.elm.

under which each may be most useful. For instance, they might instruct students to use elaborative interrogation when studying general facts about a topic, or to use self-explanation when read­ ing or solving practice problems inmath and science.

Teachers should keep inmind thatthese two strategies did not receive the highest rating in our team's assessment oflearning strategies. 18 Our lower marks for these strategies, however, stemmedfrom the fact thatwe wanted to see evenmore evidence that established their promise in several key areas relevant to

<:

z.1no 3J/.l.J~~~

education. Only a couple ofexperiments have demonstrated that elaborative interrogation can improve students' comprehension, and only a few investigations have established their efficacy within a classroom. So, in writing our review, we were conserva­ tive scientists who wanted every piece in place before declaring that a strategy is one that students should absolutely use. Never­ theless, other cognitive scientists who have studied the same evidence enthusiastically promote the use of these strategies, 19

and as a teacher myself, the overall promise ofthese strategies is impressive enough that I encourage my students to use them.

Less Useful Strategies (That Students Use a Lot) Besides the promising strategies discussed above, we also reviewed several others that have not fared so well when con­ sidered with an eye toward effectiveness. lhese include reread­ ing, highlighting, summarizing, and using imagery duringstudy.

Rereading and Highlighting

These two strategies are particularly popular with students. A survey conducted at an elite university revealed that 84 percent ofthe students studied by rereading their notes or textbooks.20

Despite its popularity, rereading has inconsistent effects on sn1- dent learning: whereas sn1dents typically benefit from rereading

AMERICAN EDUCATOR I FALL 2013 19

when they must later recall texts from memory, rereading does Summarization not always enhance students' understanding of what they read, Summarization involves paraphrasing the most important ideas and any benefits of rereading ( over just a single reading) may not within a text. It has shown some success at helping undergradu­ be long-lasting. So, rereading may be relatively easy for students ate students learn, although younger students who have difficul­ to do, but they should be encouraged to use other strategies ( such ties writing high-quality summaries may need extensive help to as practice testing, distributed practice, or self-explanation) when benefit from this strategy. they revisit their text and notes. In one study, 22 teachers received 90 minutes of training on

how to teach their students to summarize. The teachers were

trained to provide direct instruction, which included explicitly describing the summarization strategy to students, modeling Students need to know that the strategy for students, having students practice summarizing

highlighting is only the and providing feedback, and encouraging students to monitor and check their work. Smdents completed five sessions (about

beginning of the journey. 50 minutes each) of coaching, which began with them learning to summarize short paragraphs and slowly progressed to them using the strategy to take effective notes and ultimately to sum­

U1e use of highlighters seems universal-I even have a favor­ marize a text chapter. Students who received coaching recalled ite one that I use when reading articles. As compared with sim­ more important points from a chapter as compared with stu­ ply reading a text, however, highlighting has been shown to have dents who were not coached. And other studies have also shown failed to help students of all sorts, including Air Force trainees, that training students to summarize can benefit student

children, and undergraduate students. Even worse, one study performance. reported that students who highlighted while reading per­ Nevertheless, the need for extensive training will make the formed worse on tests of comprehension wherein they needed use of this strategy less feasible in many contexts, and although to make inferences that required connecting different ideas summarizing can be an important skill in its own right, relying across the text. 21 In this case, by focusing on individual concepts on it as a strategy to improve learning and comprehension may while highlighting, students may have spent less time thinking not be as effective as using other less-demanding strategies.

about connections across concepts. Still, I would not take away Keyword Mnemonic and Imagery for Text highlighters from students; they are a security blanket for read­

ing and studying. However, students need to know that high­ Finally, the last two techniques involve mental imagery (i.e., lighting is only the beginning of the journey, and that after they developing internal images that elaborate on what one is study­ read and highlight, they should then restudy the material using ing). Students who are studying foreign-language vocabulary, more-effective strategies. for example, may use images to link words within a pair ( e.g., for

Effectiveness of Techniques Reviewed IITa&retffl

Technique Extent and Conditions of Effectiveness

Practice testing Very effective under a wide array of situations

Distributed pract.ice Very effective under a wide array of situations

Interleaved practice Promising for math and concept learning, but needs more research

Elaborative interrogation Promising, but needs more research

Self-explanation Promising, but needs more research

Rereading Distributed rereading can be helpful, but time could be better spent using another strategy

Highlighting and underlining Not particularly helpful, but can be used as a first step toward further study

Summarization Helpful only with training on how to summarize

Keyword mnemonic Somewhat helpful for learning languages, but benefits are short-lived

Imagery for text Benefits limited to imagery-friendly text, and needs more research

the pair "la dent-tooth," students may mentally picture a dentist (for "la dent") extracting an extra-large tooth). This strategy is called key­ word mnemonic, because it involves developing a keyword to represent the foreign term (in this case, "den­ tist" for "la dent") that is then linked

to the translation using mental imagery.

Imagery can also be used with more complex text materials as well. For instance, students can develop mental images of the content as they read, such as trying to imagine the sequence of processes in photosyn­ thesis or the moving parts of an engine. lhis strategy is called imag­

ery for text. Mental imagery does increase

retention of the material being stud­ ied, especially when students are tested soon after studying. However, research has shown that the benefits

of imagery can be short-lived,23 and the strategy itself is not widely appli-

20 AMERICAN EDUCATOR I FALL 2013

cable. Concerning the latter, younger students mayhave difficul­ ties generating images for complex materials, and for thatmatter, much content in school is not imagery friendly, such as when the ideas are abstract or the content is complexenough that it cannot be easily imagined. Certainly, for students who enjoy using imageryand for materials that afford its use, it likely will nothurt (and may even improve) learning. But as comparedwith some of the better strategies, the benefits of imagery are rela­ tively limited.

Even the best strategies will only be effective if students are motivated to use them correctly.

U sing learning strategies can increasestudent under­ standing and achievement. For some ideas on how the best strategies can be used, see the box"Tips for Using Effective Learning Strategies" (on the right).

Ofcourse, all strategies are not created equal. As shown in Table 1 (on page 20), while some strategies are broadly applicable and effective, such as practice testing and distributed practice, others do not provide much-ifany- bang for the buck. Importantly, even the best strategieswill only be effective ifstudents are moti­ vated to use them correctly, and even then, the strategies will not solve many of the problems that hamper student progress and success. With these caveats in mind, the age-old adage about teaching people to fish (versus just giving them a fish) applies here: teaching students content mayhelp them succeed inany given class, butteaching them howto guide their learning ofcontent using effective strategies will allow them to success­ fully learn throughout their lifetime. D

Endnotes 1. Robert A Bjork, JohnDunlosky, and Nate Kernell, "Self-Regulated Learning: Beliefs, Techniques, and Illusions,• Annual Review ofPsychology 64(2013): 417--444.

2. John Dunlc,sky, Katherine A Rawson, Elizabeth J. Marsh, Mitchell J. Nathan, and Daniel T. Willingham, "Improving Students' Learning with Effective Learning Techniques: Promising Directions from Cognitive and Educational Psychology,• Psychological Science in the Public Interest 14, no. 1 (2013): 4-58.

3. Dunlosky etal., "Improving Students' Learning."

4. Henry L Roediger Ill and Andrew C. Butler, "The Critical Role of Retrieval Practice in Long-Term Retention," Trends in Cognitive Sciences 15, no. 1 (2011): 20-27; and Nicholas J. Cepeda, Harold ?ashier, Edward Vul, John T. Wixted, and Doug Rohrer, "Distributed Practice in Verbal Recall Tasks: A Review and Quantitative Synthesis,• Psychological Bulletin 132, no. 3 (2006): 354--380.

5. Edwina E. Abbott, " On the Analysis of the Factor of Recall in the Learning Process," PsychologicalMonographs 11 (1909): 15~177.

6. Regan A. R. Gurung, "HOl'II Do Students Really Study(and Does It Matter)?,• Teaching of Psychology 32 (2005): 23~241 .

7. Mark A McDaniel, Pooja K. Agarwal, Barbie J. Huelse,, Kathleen B. McDermott, and Henry L Roediger Ill, "Test-Enhanced Learning in a Middle School Science Classroom: The Effects of Quiz Frequency and Placement," JournalofEducationalPsyd,ology 103, no. 2 (2011 ): 39~14.

8. Henry L Roediger Ill and Jeffrey D. Karpicke, "The POl'ller ofTesting Memory: Basic Research and Implications for Educational Practice,• Pelspectives on Psychological Science 1, no. 3 (2006): 181-210.

9. Nate Kernell and Robert A. Bjork, "The Promise and Perils of Self-Regulated Study,• Psychonomic Bulletin and Review 14, no. 2 (2007): 21~224.

10. See, for example, John A. Glover, "The 'Testing' Phenomenon: Not Gone but Nearly Forgotten," Journal ofEducational Psychology 81, no. 3 (1989): 392-399.

11. Cepeda et al., "Distributed Practice in Verbal Recall Tasks.•

Tips for Using Effective Learning Strategies

Based on our review of the literature, here are a handful of suggestionsfor teachers to help students take advantage of more-effective strategies:

• Give a low-stakes quiz at the beginning of each class and focus on the most important material. Consider calling it a "review" to make it less intimidating.

• Give a cumulative examination, which should encourage students to restudy the most impor­ tant material in a distributed fashion.

• Encourage students to develop a "study plan­ ner," so they can distribute their study through­ outa class and rely less on cramming.

• Encourage students to use practice retrieval when studying instead of passively rereading their books and notes.

• Encourage students to elaborate on what they are reading, such as by asking "why" questions.

• Mix it up in math class: when assigning practice problems, be sure to mix problems from earlier units with new ones, so that students can prac­ tice identifying problems and their solutions.

• Tell students that highlighting is fine but only the beginning of the learning journey.

12.Jennifer A. McCabe, " Metacognitive Awareness ofLearning Strategies in Undergradu­ ates, • MernoryandCognroon 39, no. 3 (2011): 462--476.

13. Doug Rohrerand Kelli Tayloc, "The Shuffling of Mathematics Problems Improves Learning,· InstructionalScience 35 (2007): 481--498.

14. Kristin H. Mayfield and Philip N. Chase, "The Effects ofCumulative Practice on Mathemat­ ics Problem Solving." JournalofAppliedBehaviorAnalysis 35, no. 2 (Summer 2002): 105-123.

15.Vivian I. Schneider, Alice F. Healy, and Lyle E. Bourne Jr., "What Is Learned Under Difficult Conditions Is Hard toForget: Contextual Interference Effects in Foreign Vocabulary Acquisition, Retention, and Transfer,• JournalofMemory and Language 46, no. 2 (2002): 41~0.

16. Zane Olina, Robert Reiser, Xiaoxia Huang, Jung Lim, and Sanghoon Park, " Problem Format and Presentation Sequence: Effects on Learning and Mental Effortamong US High School Students," Applied Cognitive Psyd,ology 20, no. 3 (2006): 29~309.

17. Dianne C. Berry, " Metacognitive Experience and Transfer of Logical Reasoning, • Quarterly JoumalofExperimenta/Psyd,ology35, no. 1 (1983): 3~9.

18. Dunlosky etal., "Improving Students' Learning.•

19. Henry L Roediger Illand Mary A Pye, "Inexpensive Techniques to Improve Education: Applying Cognitive ~logy to Enhance Educational Practice,• Journal ofAppliedResearch in Memo,yand Cognition 1, no. 4 (2012): 242-248.

20. Jeffrey D. Karpicke, Andrew C. Butler, and Henry L. Roediger Ill, • Metacognitive Strategies in Student Leaming: Do Students Practise Retrieval When They Study on TheirOwn?,• Memory 17, no. 4 (2009): 471-479.

2 1. Sarah E. Peterson, "The Cognitive Functions of Underlining as a Study Technique," Reading Research and Instruction 3 1 (1992): 4~56.

22. Steven D. Rinehart, Steven A. Stahl, and Lawrence G. Erickson, "Some Effects of Summarization Training on Reading and Studying,• Reading Research Quarterly21, no. 4 (1986): 422--438.

23. Alvin Y. Wang, Margaret H. Thomas, and Judith A. Ouellette, "Keyword Mnemonic and Retention ofSecond-Language Vocabulary Words," Journal ofEducational Psychology 84, no. 4 (1992): 520--528.

AMERICAN EDUCATOR I FALL 2013 21