http://gcq.sagepub.com/Gifted Child Quarterly

http://gcq.sagepub.com/content/46/4/265The online version of this article can be found at:

 DOI: 10.1177/001698620204600403

2002 46: 265Gifted Child QuarterlyRobert J. Sternberg and Elena L. Grigorenko

The Theory of Successful Intelligence as a Basis for Gifted Education  

Published by:

http://www.sagepublications.com

On behalf of: 

  National Association for Gifted Children

can be found at:Gifted Child QuarterlyAdditional services and information for    

  http://gcq.sagepub.com/cgi/alertsEmail Alerts:

 

http://gcq.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://gcq.sagepub.com/content/46/4/265.refs.htmlCitations:  

What is This? 

- Oct 1, 2002Version of Record >>

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

I~

The Theory of Successful Intelligenceas a Basis for Gifted Education

Robertj. Sternberg Elena L. GrigorenkoYale University Yale University and Moscow State Univers

A number of models are available for use in giftededucation. The theory of successful intelligence, oneTEtew

such model, provides a basis for identification, inter-vention, and evaluation in gifted programs. This arti-

cle describes the model, allowing practitioners and

other interested parties to apply the model in a giftededucation environment.

The article contains four main parts. The first

part presents the theory of successful intelligence and

data in support of it. The second part shows how toedimplement the model in schools and presents data in

support of the success of its school implementation.

The third part relates the theory of successful intel-

gence to other models ofgifted education. The fourth

part briefly draws some conclusions.

The theory of successful intelligence provides a model

for identification, intervention, and evaluation in giftedprograms. The goal of this article is to describe the model _ _$so that practitioners arnd others who iglt be interested in

the theory will have a model for its application in the =_gifted education environment.select to

The Theory of Successful

Definition ofSuccessful Intelligence

Successful intelligence is the ability to succeed in life

accodin toone's own definition of success, within one's

sociocultural context, by capitalizing on one's strengths

and correcting or compensating for one's weaknesses; iorder to adapt to, shape, and select environments; through

a combination of analytical, creative, and practical abilities

(Steinberg, 1997a, 1999b, 1999c).:2

sity

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0' *

Kinds of Giftedness Accordingto the Theory ofSuccessful Intelligence

According to this theory, therefore, an individual issuccessfully intelligent by virtue of developing the skillsneeded to achieve success as she or he defines it. Peoplewho are gifted are those who are particularly well able toachieve such success. They do so by combining analytical,creative, and practical abilities. People may be gifted withrespect to any one of these abilities or with respect to theway they balance the abilities in order to succeed.

People who are analytically gifted are particularly wellable to analyze, judge, critique, compare and contrast,evaluate, and explain. They typically do well in school andon standardized tests. Tests of IQ measure largely analyti-cal abilities, as well as memory abilities. These people havethe kind of intelligence that is most likely to lead them tobe labeled as gifted in school. The fact that they are wellable to learn and analyze ideas does not necessarily meanthat they are well able to come up with their own ideas orto apply what they have learned in everyday life.

A person who is creatively gifted is particularly well ableto create, invent, discover, explore, imagine, and suppose.Conventional tests of intelligence do not really measure cre-ative intelligence, nor are they intended to. Tests such as theTorrance Test (Torrance, 1974) measure creativity in some-what restricted situations, but primarily the component of itthat is related to fluency (rapid production of ideas).However, creativity is the ability to generate ideas that arenovel, high in quality, and task appropriate. Therefore, tomeasure creativity, we typically use tasks that are somewhatdifferent, such as writing short stories, drawing pictures, for-mulating advertisements, and solving novel scientific prob-lems (Sternberg & Lubart, 1995; Sternberg & O'Hara, 2000).

A person who is practically gifted is particularly wellable to use, utilize, apply, implement, and put into practice.Such a person shows intelligence in highly contextualizedsituations. The person may or may not be notable for his orher formal knowledge, but often is distinguished by his orher tacit knowledge, that is, knowledge ofwhat one needsto know to succeed in an environment that usually is notdirectly taught and that often is not even verbalized. Forexample, a practically gifted individual might be aware ofhow his or her actions affect others and of the nonverbalsignals others emit that show how they feel about things.

A person who is gifted in a balanced way may not beextremely high in analytical, creative, or practical intelli-gence. Rather, he or she may be particularly well able tobalance the levels ofthe three abilities, knowing more pre-cisely than most people when and how to use them.

Processes Underlying the Theory

According to the theory of successful intelligence(Sternberg, 1985, 1997a, 1999b), a common set of processesunderlies all aspects of intelligence. These processes arehypothesized to be universal. For example, although the solu-tions to problems that are considered intelligent in one culturemay be different from the solutions considered to be intelli-gent in another culture, the need to define problems and for-mulate strategies to solve these problems exists in any culture.

Metacomponents, or executive processes, plan what to do,monitor things as they are being done, and evaluate thingsafter they are done. Examples of metacomponents are rec-ognizing the existence of a problem, defining the nature ofthe problem, deciding on a strategy for solving the problem,monitoring the solution of the problem, and evaluating thesolution after the problem is solved. In writing a paper, stu-dents need to recognize the existence of some problem(e.g., what does it mean to be gifted?), define the problem(e.g., define giftedness in terms of some old model, newmodel, or combination of old and new models), decide ona strategy to present the model in the paper, and so forth.

Performance components execute the instructions of themetacomponents. For example, inference is used to decidehow two stimuli are related, and application is used to applywhat one has inferred (Sternberg, 1977). Other examples ofperformance components are comparison of stimuli, justifi-cation ofa given response as adequate although not ideal, andactually making the response. For example, a student writinga paper might need to infer the implications of a theory ofgiftedness for designing assessments of achievement.

Knowledge-acquisition components are used to learn how tosolve problems or simply to acquire declarative knowledge inthe first place (Sternberg, 1985). Selective encoding is usedto decide what information is relevant in the context ofone'slearning. Selective comparison is used to bring old informa-tion to bear on new problems. And selective combination isused to put together the selectively encoded and comparedinformation into a single and sometimes insightful solutionto a problem. For example, in reading before writing a paper,a student would have to selectively encode what informationis relevant for the paper and what information is not.

Although the same processes are used for all threeaspects of intelligence universally, these processes areapplied to different kinds of tasks and situations dependingon whether a given problem requires analytical thinking,creative thinking, practical thinking, or a combination ofthese kinds of thinking. Data supporting the theory can-not be presented fully here, but are summarized elsewhere(Sternberg, 1977, 1985a; Sternberg et al., 2000).

S I * 0O S ..

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0. S

Intelligence as Developing Expertise

Successful intelligence is viewed as a form ofdevelop-ing expertise (Sternberg, 1998a, 1999b). In other words, itis not a fixed entity, but a flexible and dynamic one. Allintelligence tests measure only an aspect-typically a lim-ited aspect-of developing expertise. Developing expert-ise is defined here as the ongoing process ofthe acquisitionand consolidation of a set of skills needed for a high levelof mastery in one or more domains of life performance.Good performance on intelligence tests requires a certainkind ofexpertise, and, to the extent this expertise overlapswith the expertise required by schooling or by the work-place, there will be a correlation between the tests and per-formance in school or in the workplace. But, suchcorrelations represent no intrinsic relation between intel-ligence and other kinds ofperformance. Rather, they rep-resent overlaps in the kinds ofexpertise needed to performwell under different kinds of circumstances.

There is nothing privileged about the intelligencetests or any other tests of abilities. One could as easily use,say, academic achievement to predict intelligence-relatedscores. According to this view, although ability tests mayhave temporal priority relative to various criteria in theiradministration (i.e., ability tests are administered first and,criterion indices of performance, such as grades orachievement test scores, are collected later), they have nopsychological priority. All of the various kinds of assess-ments are of the same kind psychologically. What distin-guishes ability tests from other kinds ofassessments is howthe ability tests are used (usually predictively), rather thanwhat they measure. There is no qualitative distinctionamong the various kinds of assessments. All tests measurevarious kinds of developing expertise.

Conventional tests of intelligence and related abilitiesmeasure achievement that individuals should have accom-plished several years back (see also Anastasi & Urbina, 1997).Tests such as vocabulary, reading comprehension, verbalanalogies, arithmetic problem solving, and the like are all, inpart, tests of achievement. Even abstract-reasoning testsmeasure achievement in dealing with geometric symbols,skills taught in Western schools (Laboratory ofComparativeHuman Cognition, 1983; Serpell, 2000). One might as welluse academic performance to predict ability-test scores. Theproblem regarding the traditional model is not in its state-ment ofa correlation between ability tests and other forms ofachievement, but in its proposal ofa causal relation wherebythe tests reflect a construct that is somehow the cause of,rather than merely temporally antecedent to, later success.The developing-expertise view in no way rules out the con-

tribution of genetic factors as a source of individual differ-ences in who will be able to develop a given amount ofexpertise. Many human attributes, including intelligence,reflect the covariation and interaction of genetic and envi-ronmental factors. But, the contribution ofgenes to an indi-vidual's intelligence cannot be directly measured or evendirectly estimated. Rather, what is measured is a portion ofwhat is expressed, namely, manifestations of developingexpertise, the kind of expertise that potentially leads toreflective practitioners in a variety offields (Schon, 1983).

The upshot of this view is that successful intelligence,and giftedness in it, is not wholly inborn (see alsoCallahan, 2000; Grigorenko, 2000; Sternberg &Grigorenko, 1997). Genetic factors interact with environ-mental ones to produce variable levels of developingexpertise. Good schools help children maximize theirdevelopment of such expertise.

Overall Structureofthe Theory ofSuccessful Intelligence

The overall structure of the theory of successful intelli-gence is shown in Figure 1. The figure shows how compo-nents are applied to experience and then to real-worlddecisions about how to respond to environmental challenges.

Components of intelligence (metacomponents, per-formance components, knowledge-acquisition compo-nents) are interactive: Metacomponents activateperformance components and knowledge-acquisitioncomponents, which then provide feedback to the meta-components. When the components are applied to rela-tively familiar kinds ofproblems for which the structure isabstracted (i.e., the problem solver knows more or lesswhat to expect), the components reflect analytical abilities.

Creative abilities are invoked when the componentsare applied to relatively novel kinds ofproblems and prob-lem situations. What start out as relatively novel kinds ofproblems and problem situations eventually may becomeautomatized, so that they require little processing effort.Reading, driving, and holding a phone are examples ofautomatized actions.

Practical abilities are invoked when one applies thecomponents to experience in order to adapt to, shape, andselect environments. In other words, one confronts prob-lems of how to change oneself to suit the environment,change the environment to suit oneself, or simply find anew environment in which to apply one's skills.

People may be gifted in one ofthese areas, two ofthem,or in all three (or, of course, in none of them). But, theirareas of giftedness may develop with time because the abil-

S .. 0 66 3 I S

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

The "Triarchy" of Intelligence

Adapt to Environment NTEXTS

Shape Environments - - Select Environments

| EXPERIENCE|

Relatively Novel Automatized TasksTasks and Situations and Situations

COMPONENTS

,,/ | ~~PerformancelComponents

Metacomponents \_ Knowledge-Acquisition

\ ~~~Components

Figure 1. Structure of the theory ofsuccessfulintelligence: The "triarchy" ofintelligence.

cities are forms of developing expertise. With the right kindof schooling and home socialization, a person can develophis or her abilities further than in an environment that islimiting of opportunities for intellectual development.

Is the Theory Empirically Supported?

Gifted education is replete with theories that, how-ever attractive they may be, have little or no predictiveempirical validity data to support them as a whole. Thesetheories basically express the armchair speculations oftheirauthors. The theories are akin to arguments in philosophyor literature that can be supported by argumentation or byselective citations to past literature, but have not been

shown to predict anything, a requirement for a theory tobe scientific. There is nothing wrong with literary orphilosophical claims; they simply are different in kindfrom scientific claims. When we use a new drug on chil-dren, we would want the usefulness and safety of the drugto have been demonstrated through scientific research, notspeculation, however imaginative and persuasive that spec-ulation might be. Why should we have different standardsfor educational programs? We propose our theory herebecause it has been shown to be supported scientifically.Some ofthe evidence is summarized below. More detailedaccounts can be found elsewhere (Sternberg, 1985, 1997a,1999a; Sternberg et al., 2000).

In one study (Sternberg, Grigorenko, Ferrari, &Clinkenbeard, 1999), we used the so-called SternbergTriarchic Abilities Test (STAT; Sternberg, 1993) to inves-tigate the internal validity of the theory (in other words,whether the division into analytical, creative, and practicalabilities is justifiable). The test comprised 12 subtests andwas taken by 326 high school students, primarily fromdiverse parts of the United States, identified as gifted bytheir schools. Analytical, creative, and practical abilitieswere each measured by four subtests, three multiple-choice tests and one essay test. The multiple-choice tests,in turn, involved, respectively, verbal, quantitative, andfigural content. Consider the content of each test:1. Analytical-Verbal: Figuring out meanings of neolo-

gisms (artificial words) from natural contexts.Students see a novel word embedded in a paragraphand have to infer its meaning from the context.

2. Analytical-Quantitative: Number series. Studentshave to predict what number should come next in aseries ofnumbers.

3. Analytical-Figural: Matrices. Students see a figuralmatrix with the lower right entry missing. They haveto determine which of the options fits into the miss-ing space.

4. Practical-Verbal: Everyday reasoning. Students arepresented with a set ofeveryday problems in the life ofan adolescent and have to select the option that bestsolves each problem.

5. Practical-Quantitative: Everyday math. Students arepresented with scenarios requiring the use of math ineveryday life (e.g., buying tickets for a ballgame) andhave to solve math problems based on the scenarios.

6. Practical-Figural: Route planning. Students are pre-sented with a map of an area (e.g., an amusementpark) and have to answer questions about navigatingeffectively through the area depicted by the map.

7. Creative-Verbal: Novel analogies. Students are pre-

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

Iii i0 1 1

sented with verbal analogies preceded by counterfac-tual premises (e.g., money falls offtrees). They have tosolve the analogies as though the counterfactualpremises were true.

8. Creative-Quantitative: Novel number operations.Students are presented with rules for novel numberoperations, for example, "flix," which involvesnumerical manipulations that differ as a function ofwhether the first of two operands is greater than,equal to, or less than the second. Participants have touse the novel number operations to solve presentedmath problems.

9. Creative-Figural: In each item, participants are firstpresented with a figural series that involves one ormore transformations. They then have to apply therule of the series to a new figure with a differentappearance and complete the new series.Correlations between pairs of these tests depend on

what scores one uses. In our study of 199 high school stu-dents (Sternberg, Grigorenko, Ferrari, & Clinkenbeard,1999), raw simple correlations were .47 between analyticaland creative tests, .41 between analytical and practical tests,and .37 between creative and practical tests. In general, mul-tiple-choice tests tended to correlate more highly with eachother (median r= .52) than did essay tests (median r = .21).However, when we used structural-equation modeling tocontrol for method and error variance, the correlational pic-ture changed, with correlations for the latent abilities of-.07between analytical and creative, .00 between analytical andpractical, and .06 between creative and practical abilities.

We found that a confirmatory factor analysis on the datawas supportive ofthe theory ofsuccessful intelligence, yield-ing separate and uncorrelated analytical, creative, and practi-cal factors. The lack of correlation was due to the inclusionofboth essay and multiple-choice subtests. Although multi-ple-choice tests tended to correlate substantially with multi-ple-choice tests, the correlation ofmultiple-choice tests withessay tests was much weaker; similarly, essays correlated witheach other (although not as strongly as multiple-choice testscorrelated with each other), suggesting the presence of sub-stantial method variance. We found the multiple-choiceanalytical subtest to load most highly on the analytical factor,but the essay creative and practical subtests loaded mosthighly on their respective factors. Thus, ideally, measure-ment of creative and practical abilities probably should beaccomplished with other kinds of testing instruments thatcomplement multiple-choice instruments.

We have now developed a revised version of this test,which, in a preliminary study of 53 college students, showsoutstanding internal and external validation properties. This

test supplements the creative and practical measuresdescribed above with performance-based measures. Forexample, creative abilities are additionally measured by hav-ing people write and tell short stories, do captions for car-toons, and use computer software to design a variety ofproducts. Practical skills are measured additionally by aneveryday situational-judgment inventory and a college-stu-dent tacit-knowledge inventory. These tests require indi-viduals to make decisions about everyday problems faced inlife and in school. We found that the creative tests are mod-erately correlated with each other and the practical tests arehighly correlated with each other. The two kinds of tests aredistinct from one another, however. An exploratory factoranalysis revealed separate analytical, creative, and practicalfactors. Interestingly, the performance-based assessmentstend to cluster separately from multiple-choice assessmentsmeasuring the same skills (similar to our earlier findings thatessay measures tended to be distinct from multiple-choicemeasures). These results further suggest the need for meas-uring not only a variety of abilities, but also for measuringthese abilities through various modalities of testing.

In a second and separate study, conducted with 3,278students ranging in grade level from upper elementary tohigh school in the United States, Finland, and Spain, weused the multiple-choice section of the STAT to comparefive alternative models ofintelligence, again via confirma-tory factor analysis. A model featuring only a general fac-tor ofintelligence fit the data relatively poorly. The theoryof successful intelligence, allowing for intercorrelationsamong the analytic, creative, and practical factors, pro-vided the best fit to the data (Sternberg, Castejon, Prieto,Hautamiki, & Grigorenko, 2001).

In sum, studies with large numbers of participants inthree different countries have supported the theory of suc-cessful intelligence. Other studies in other countries, such asRussia and Kenya, have also supported the distinctionbetween academic-analytical and practical abilities (e.g.,Grigorenko & Sternberg, 2001; Sternberg et al., in press). Wethus believe we have solid (as well as peer-reviewed) empiri-cal data that support the theory of successful intelligence.

Applying the Theoryof Successful Intelligencein Gifted Programs

Identification ofChildrenfor Successful Intelligence

Because it is very easy to identify children as giftedsimply by using an IQ test or its equivalent, program

0 .. 0 00 I I 0 '.

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

I.

supervisors may be tempted to use IQ tests exclusively inidentification. To many people, they give an appearance ofobjectivity, fairness, reliability, and validity that othermeasures may not easily match. For a variety of reasons,this appearance is deceptive (Gardner, 1983; Sternberg,1997a). We believe, however, that IQ tests and theirequivalents can provide one among several useful bases foridentification. While they provide useful information, inmany cases, about children's analytical abilities, they saylittle or nothing about creative and practical abilities(Sternberg, 1985, 1997a, 1999b).

To identify gifted students in terms of the theory ofsuccessful intelligence, we prefer to use a variety ofmeasures:* standardized tests of memory and analytical abilities

(tests of IQ, SAT, ACT, SSAT, etc.);1* standardized tests of achievement that measure more

developed memory and analytical competencies (suchas the Iowa Tests of Basic Skills or the StanfordAchievement Tests);

* teachers' grades and comments, which assess memoryand analytical abilities, as well as motivational andother variables relevant to achievement;

* STAT (Sternberg Triarchic Abilities Test), asdescribed above, a nonstandardized test that can beused in conjunction with other measures but neveralone, which measures analytical, creative, and practi-cal abilities;

* evaluations of existing products, projects, and portfo-lios for analytical, creative, and practical skills;

* SI Student Questionnaire, which is a questionnaire forstudents regarding their preferences and skills (seeTable 1);

* SI Teacher Questionnaire, which is comparable to theSI Student Questionnaire, except that evaluations aredone by teachers instead of students (see Table 1); and

* tasks created by teachers, such as having students writestories or reports, draw pictures, create advertise-ments, solve novel problems, solve practical problems,and so forth.We believe that converging measures such as those

above provide a better basis for identification than can anyone measure. There is no magic formula for combininginformation from these eight converging operations.Rather, each district must decide for itself how to weighthe various criteria for assessing giftedness. In our view,the assessments are of roughly equal importance, and wesuspect little would be gained by a weighting scheme. But,we do not at the present time have data to bear out thisclaim.

Instruction and Assessment ofthe AchievementofGifted Childrenfor Successful Intelligence

We combine instruction and assessment of achieve-ment into one section in this article because we believe thatinstruction and assessment ought to be of one piece-instruction must match assessment and vice versa. Moregenerally, identification, instruction, and assessment allmust match. If identification does not match instructionand assessment, then one may end up with children who areidentified for one set of abilities failing because the instruc-tion and assessment do not match their strengths. Ifinstruc-tion does not match assessment, children very quickly cometo view the instruction as a game and start paying attentiononly to what they will be assessed on, not what is done inthe classroom. Ultimately, children should think to learn sothat they learn to think, and they can learn equally fromclassroom instruction and from classroom assessments.

Teaching and assessing achievement for successfulintelligence involves three basic sets of ideas. These sets ofideas include a set of principles, a set of techniques, and aset of skills to be developed.

Principles of teaching and assessingfor successful intelligence.There are seven main principles of teaching for successfulintelligence (Sternberg, 1998b). These principles must beadhered to in all instruction and assessment:* The goal of instruction is the creation of expertise

through a well and flexibly organized, easily retriev-able, knowledge base.

* Instruction should involve teaching for analytical, cre-ative, and practical thinking, as well as for memorylearning.

* Assessment should also involve analytical, creative,and practical, as well as memory components.

* Instruction and assessment should enable students toidentify and capitalize on their strengths.

* Instruction and assessment should enable students toidentify, correct, and, as necessary, compensate forweaknesses.

* Instruction should help students (a) adapt to the envi-ronment (change themselves to suit the environmentbetter), (b) shape the environment (change the envi-ronment to suit them better), and (c) select new envi-ronments.

* Good instruction and assessment integrate, ratherthan separate, all of the elements of intelligence.All students receive all kinds ofinstruction (analytical,

creative, and practical). Such instruction helps studentscapitalize on strengths and correct or compensate forweaknesses.

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0. S

Pedagogical techniques in teaching for successful intelligence.We have found in our various research and developmentprojects that, when teachers initiate, monitor, and evaluatetheir teaching behavior in terms of an easy-to-learn set ofcues, it is easier for them to adopt successfully the principlesof teaching for successful intelligence (Sternberg, 1998b;Sternberg & Grigorenko, 2000). We have found that teach-ers feel more comfortable and unthreatened when weemphasize to them that these are all cues that they knowhow to use and probably have used. What we provide is aframework to balance their teaching (including instructionand assessment) for successful intelligence. Examples aregiven here for some of the cues in each of the domains ofreading/language arts, mathematics, and science:1. For emphasis on analytical thinking:

a. Analyze (1) the plot of a story, (2) a mathematicalword problem, (3) a scientific theory.

b. Evaluate (1) the meaning of a poem; (2) whetheran arithmetic problem, such as one requiringmultiplication of decimals, has been solved cor-rectly; (3) whether a simple experiment, such asdropping a heavy object and a light object off atower on a windy day to see whether they land atthe same time, tests what it is supposed to test.

c. Compare and Contrast (1) two characters in a story,such as Tom Sawyer and Huckleberry Finn; (2)mathematical operations (e.g., addition and sub-traction); (3) two approaches to analyzing scien-tific data (e.g., qualitative and quantitative).

d. Explain (1) the meaning of a word such as exempt;(2) why a mathematical problem, such as a wordproblem, can be solved in a certain way; (3) a sci-entific principle such as the law ofconservation ofenergy.

2. For emphasis on creative thinking:a. Create (1) a poem; (2) a mathematical word prob-

lem ofa certain kind (e.g., mixture); (3) a theory toexplain a natural phenomenon, such as rainbows.

b. Design (1) a visual representation of the sequenceofevents in a story, such as "The Giver"; (2) a testmeasuring the material you have studied in math-ematics, such as multiplication of fractions; (3) anexperiment to test a hypothesis about a physicalphenomenon, such as effects ofgravity on objectsof different weights.

c. Imagine (1) how a given story might have endedhad it been written 100 years later; (2) what com-putation would be like if we used a base-12,rather than base-10, system of numeration; (3)how bees communicate with each other.

Ta b I e 1

Student/Teacher SI Questionnaire

Read the list below and tick off the activities you [your student]like[s] doing. Do not tick off the activities you [your student]does] not like to do.

Analytical* Analyzing characters when reading or listening to a story* Comparing and contrasting points ofview* Criticizing the quality ofwork* Thinking clearly and analytically* Appealing to logic* Evaluating various points ofview* Judging people's behavior* Explaining difficult problems to others* Solving logical problems* Making inferences and deriving conclusions* Sorting and classifying* Thinking about things

Creative* Designing new things* Coming up with ideas* Using imagination* Playing "make-believe" and "pretend" games* Thinking of alternative solutions* Noticing things people usually tend to ignore* Thinking in picture and images* Inventing (new recipes, words, games)* Supposing that things were different* Thinking about what would have happened if...* Composing (new songs, melodies)* Enjoying acting and role-playing

Practical* Taking things apart and fixing them* Learning through "hands-on" activities* Making and maintaining friends* Understanding and respecting others* Putting into practice things that have been learned* Resolving conflicts* Advising friends on their problems* Convincing someone to do something* Learning by interacting with others* Applying knowledge* Working and being with others* Adapting to new situations

Now, go back through the list and, after each statement, indicatehow well you believe you [your student] do[es] the particularactivity specified. You should mark E for Excellent, VG for VeryGood, G for Good, F for Fair, and P for Poor.

0 0 0O I I S '.

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0. 0

d. Suppose (1)Johnny Tremaine fought for the British(How might the story have ended?); (2) mathe-matics had not been invented (How would ourworld be changed?); (3) the ozone layer becomesdepleted (What changes can we then expect?)

3. For emphasis on practical thinking:a. Use (1) a lesson from Treasure Island in your own

life; (2) principles ofcompound interest to deter-mine how much money you would make if youplaced a certain amount of money in a savingsaccount at a given rate of interest; (3) what youknow about the speed of sound to discuss prob-lems with supersonic transport.

b. Apply (1) your knowledge of phonics to soundout a new word, such as exacerbate; (2) yourknowledge of arithmetic to compute how muchmore you will have to pay for front-section seatsthan for back-section seats to a football game; (3)what you learned about testing water to the waterin your own home.

c. Implement (1) a practical plan for improving yourunderstanding ofdifficult texts (e.g., seeking mainideas, summarizing, etc.); (2) the Pythagoreantheorem in figuring out how much time you willsave by taking a route to a destination that is ahypotenuse ofa triangle; (3) what you know aboutnutrition by designing a menu for a week that isnutritious and that you would be able to follow.

d. Put into Practice (1) Tom Sawyer's persuasion tech-nique for convincing people to do things that theymight otherwise not want to do (e.g., whitewash-ing a fence); (2) what you have learned about per-centages to figure out what the discount is on anitem of sale merchandise (such as a portable radio);(3) what you know about nitrates and nitrites todecide whether you want to buy foods with highlevels of these substances as preservatives.

Teaching for successful intelligence also involves stan-dard teaching for memory, with prompts, such as (a) recall(e.g., who, what, where, when, why, how), (b) summarize, (c)recognize, (d) choosefrom among the options, (e) list, and (f) identify.

In scoring materials, we devise rubrics that are appro-priate to the subject matter and grade level being assessed.There is no one generalized rubric that will apply to allkinds of products because different products require dif-ferent knowledge and skills. Thus, although we work withteachers to formulate rubrics, neither we nor anyone elsecan provide a "one-size-fits all" rubric.

Processes to be developed through teachingfor successful intel-ligence. Instruction and assessment should involve utiliza-

tion, at various times, of all seven of the metacomponentsof the problem-solving cycle: (a) problem identification,(b) problem definition, (c) formulation of problem-solv-ing strategies, (d) formulation ofmental and external rep-resentations and organizations of problems and theirassociated information, (e) allocation of resources, (f)monitoring of problem solving, and (g) evaluation ofproblem solving.

Instruction should also involve utilization, at varioustimes, of at least six performance components, including(a) encoding of information, (b) inference of relationsbetween chunks of information, (c) mapping of higherorder relations among relations, (d) application of infor-mation and relations between chunks of information, (e)comparison of alternatives, and (f) response.

Finally, instruction should involve utilization, at vari-ous times, of at least three knowledge-acquisition compo-nents, including (a) selective encoding (distinguishingrelevant from irrelevant information), (b) selective com-parison (relating old information to new information), and(c) selective combination (putting together disparatepieces ofinformation to reach a conclusion).

Optimal instruction is in the zones of (a) relative nov-elty and (b) automatization for the individual. What is rel-atively novel or ready for automatization differs from oneindividual to another.

Processes always act on mental representations.Instruction and assessment should take into account indi-vidual differences in preferred mental representations,including verbal, quantitative, and figural, as well asmodalities for input (visual, auditory, kinesthetic) and out-put (written, oral, performance-based).

Does the theory result in improved instructional outcomes?Just as many theories in the gifted education field lacksolid empirical support, so too many theories used as thebasis for instructional intervention have little or no solidand peer-reviewed empirical data showing that they workin the classroom. Teachers' satisfaction with programsbased on the theories or even students' satisfaction withsuch programs are important, but do not in themselvesconstitute solid empirical support. Teachers or studentsmay enjoy a program that does not, in fact, produce betteroutcomes than conventional programs or even satisfactoryinstructional outcomes.

In one set of studies, we explored the question ofwhether conventional education in school systematicallydiscriminates against children with creative and practicalstrengths (Sternberg & Clinkenbeard, 1995; Sternberg,Ferrari, Clinkenbeard, & Grigorenko, 1996; Sternberg,Grigorenko, Ferrari, & Clinkenbeard, 1999). Motivating

I 30 06 00~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0. 0

this work was the belief that the systems in schoolsstrongly tend to favor children with strengths in memoryand analytical abilities.

We used the Sternberg Triarchic Abilities Test, asdescribed above. The test was administered to 326 chil-dren around the United States and in some other countrieswho were identified by their high schools as gifted by anystandard. Children were selected for a summer program in(college-level) psychology if they fell into one of five abil-ity groupings: high analytical, high creative, high practical,high balanced (high in all three abilities), or low balanced(low in all three abilities). Students who came to Yale werethen divided into four instructional groups. Students in allfour instructional groups used the same introductory-psy-chology textbook (a preliminary version of Sternberg[1995]) and listened to the same psychology lectures. Whatdiffered among them was the type ofafternoon discussionsection to which they were assigned. They were assignedto an instructional condition that emphasized memory,analytical, creative, or practical instruction. For example,in the memory condition, they might be asked to describethe main tenets of a major theory of depression. In theanalytical condition, they might be asked to compare andcontrast two theories of depression. In the creative condi-tion, they might be asked to formulate their own theory ofdepression. In the practical condition, they might be askedhow they could use what they had learned about depres-sion to help a friend who was depressed.

Students in all four instructional conditions were eval-uated in terms of their performance on homework, amidterm exam, a final exam, and an independent project.Each type of work was evaluated for memory, analytical,creative, and practical quality. Thus, all students were eval-uated in exactly the same way.

Our results suggested the utility of the theory of suc-cessful intelligence. First, we observed when the studentsarrived at Yale that the students in the high creative andhigh practical groups were much more diverse in terms ofracial, ethnic, socioeconomic, and educational back-grounds than were the students in the high-analyticalgroup, suggesting that correlations of measured intelli-gence with status variables such as these may be reduced byusing a broader conception ofintelligence. Thus, the kindsofstudents identified as strong differed in terms ofpopula-tions from which they were drawn in comparison with stu-dents identified as strong solely by analytical measures.More importantly, just by expanding the range of abilitieswe measured, we discovered intellectual strengths thatmight not have been apparent through a conventional test(Sternberg, Ferrari, Clinkenbeard, & Grigorenko, 1996).

We found that all three ability tests-analytical, creative,and practical-significantly predicted course performance.When multiple-regression analysis was used, at least two ofthese ability measures contributed significantly to the predic-tion of each of the measures of achievement. Perhaps as areflection of the difficulty of de-emphasizing the analyticalway of teaching, one of the significant predictors was alwaysthe analytical score. However, in a replication of our studywith low-income African American students from NewYork, Deborah Coates of the City University ofNew Yorkfound a different pattern of results. Her data indicated thatthe practical tests were better predictors of course perform-ance than were the analytical measures, suggesting that whatability test predicts what criteria depends on both populationand mode of teaching. Most importantly, there was an apti-tude-treatment interaction whereby students who wereplaced in instructional conditions that better matched theirpattern of abilities outperformed students who were mis-matched (Sternberg, Ferrari, Clinkenbeard, & Grigorenko,1996). In other words, when students are taught in a way thatfits how they think, they do better in school. Children withcreative and practical abilities, who are almost never taught orassessed in a way that matches their pattern of abilities, maybe at a disadvantage in course after course, year after year.

In a follow-up study (Sternberg, Torff, & Grigorenko,1998a, 1998b), we looked at learning of social studies andscience by third graders and eighth graders. The 225 thirdgraders were students in a very low-income neighborhoodin Raleigh, North Carolina. The 142 eighth graders werestudents who were largely middle to upper-middle classstudying in Baltimore, Maryland, and Fresno, California.In this study, students were assigned to one of threeinstructional conditions. In the first condition, they weretaught the course that basically they would have learnedhad we not intervened. The emphasis in the course was onmemory. In a second condition, they were taught in a waythat emphasized critical (analytical) thinking. In the thirdcondition, they were taught in a way that emphasized ana-lytical, creative, and practical thinking. All students' per-formances were assessed for memory learning (throughmultiple-choice assessments) and analytical, creative, andpractical learning (through performance assessments).

As expected, we found that students in the successful-intelligence (analytical, creative, practical) condition out-performed the other students in terms of the performanceassessments. One could argue that this result merelyreflected the way they were taught. Nevertheless, theresult suggested that teaching for these kinds of thinkingsucceeded. More important, however, was the result thatchildren in the successful-intelligence condition outper-

i 0i111S0l11 1 11l rfl at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0. 0

i; 11 1 i S

formed the other children even on the multiple-choicememory tests. In other words, to the extent that one's goalis just to maximize children's memory for information,teaching for successful intelligence is still superior. Itenables children to capitalize on their strengths and to cor-rect or to compensate for their weaknesses, and it allowschildren to encode material in a variety ofinteresting ways.

We have now extended these results to reading cur-ricula at the middle school and the high school level. In astudy of 809 middle school students and 432 high schoolstudents, we taught reading either for successful intelli-gence or through the regular curriculum. At the middleschool level, reading was taught explicitly. At the highschool level, reading was infused into instruction in math-ematics, physical sciences, social sciences, English, history,foreign languages, and the arts. In all settings, studentswho were taught for successful intelligence substantiallyoutperformed students who were taught in standard ways(Sternberg, Grigorenko, &Jarvin, 2001).

Obstacles to the Useofthe Theory ofSuccessful Intelligence

When any new proposal is made for gifted education(or any other form of education), there are always poten-tial objections to its use. VWhat are some ofthe main objec-tions and our responses to them?

True gftedness is a function ofgeneral intelligence, not of suc-cessful intelligence (or other kinds of intelligence proposed in recenttheories). We have been asked many times what we believegiftedness "really" is. This question presupposes that gifted-ness really is something. It is not. It is a societal invention.What constitutes giftedness in a hunting and gathering cul-ture will differ from what constitutes giftedness in modernsociety. In a preliterate society, superior skills for reading willhave no bearing on giftedness because no one reads. Eachsociety invents giftedness as a label to reward those who doparticularly well in whatever the society values. They may beespecially strong in solving abstract, school-like problems, beespecially creative, be especially good in following orders, be(seemingly) exceptionally strong in magical skills, or what-ever. Giftedness is not a single thing. It is what we make itout to be. Creative and practical skills are at least as relevantfor success in many societies today as are memory and ana-lytical skills. The rapid change in the world renders creativeflexibility a key to survival (as many businesses find outevery day), and the need to adapt to, shape, and select envi-ronments renders practical skills always important. The labelofgfftedness, therefore, should take into account creative andpractical skills because everyone needs them. Other kinds of

skills are not needed by everyone and, therefore, are perhapsbetter viewed in the realm of talents.

Statewide tests only measure memory and, to some extent, ana-lytical thinking. Therefore, teachingfor creative andpractical skills canbe sef-defeating. Our data as described above show that teach-ing for successful intelligence improves achievement relativeto conventional instruction and instruction for critical think-ing pretty much without regard to how that achievement ismeasured. Students instructed for successful intelligence evendo better on memory tests than do students instructed inalternative ways. The instruction works successfully acrossgrade levels and subject-matter areas. There are several rea-sons why the instruction is particularly effective. It enablesstudents to learn material in multiple ways (analytically, cre-atively, and practically), so that the material is more easilyretrieved later on. It enables students to capitalize on strengthsand to correct or compensate for weaknesses in their learning,so that they learn more. And the instruction simply is moremotivating to students and teachers alike, increasing theeffectiveness of the learning/teaching process.

There is no standardized testfor successful intelligence, whereasthere are standardized testsforgeneral intelligence (g). Testing hasdriven the agenda for gifted (and other forms of) educationfor so long that gifted educators have sometimes forgottenthat testing should be in the service of an educationalagenda, rather than the educational agenda being in theservice oftesting. We have pilot tests, mentioned above, thatwe use in our research, and we are working (for the secondtime) with a testing company to produce standardizedmeasures. However, tests are not a panacea, whether theyare our tests or someone else's. Standardized tests can beused in conjunction with products, projects, and portfoliosthat show children's analytical, creative, and practical skills.

Teaching and assessing for successful intelligence requires awhole new set of skills, which teachers of the gifted cannot beexpected to have acquired. On the contrary, we believe that theskills are ones that all good teachers have, but often areafraid to use. There is no teaching technique that we rec-ommend (Sternberg & Grigorenko, 2000) that will betotally alien to any good teacher. Rather, teachers are sooften rewarded for teaching only for memory and occa-sionally for analytical skills that they cease using their fullrepertoire ofteaching skills. We work with a charter school,the Sanger Academy in Sanger, California, based on thetheory of successful intelligence. The teachers have, for themost part, ordinary backgrounds in teaching. What makesthem successful is some knowledge about the theory ofsuccessful intelligence and how to apply it, but even more,their dedication and drive in realizing the model forinstruction and assessment in their classrooms.

I I P 06 0 '. 0

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0li~ 0ii Ii

The successful-intelligence model disenfranchises the tradition-ally gifted, favoring students gifted in other ways. This criticismis simply incorrect. Analytical giftedness is roughly equiv-alent to traditional giftedness and is measured by the samekinds of tests that have been traditionally used for identifi-cation of the gifted. The successful-intelligence modelaugments, rather than replaces, the traditional model.

Relation of the Theoryof Successful Intelligenceto Other Models of GiftedEducation

The theory of successful intelligence is, of course,only one of many models used for gifted education. Weattempt here to review briefly some major alternativemodels and to relate them to the theory ofsuccessful intel-ligence. At the same time, it is not possible to review allsuch models. For more comprehensive reviews, please seeSternberg and Davidson (1984) and Heller, Monks,Sternberg, and Subotnik (2000).

The g-Based Model

The theory of successful intelligence is not whollyincompatible with the theory of general abilities(Gallagher & Courtright, 1984; Jensen, 1998;Spearman, 1927), nor is it wholly compatible. The the-ory of successful intelligence is based on the notion thatso-called general ability is a part of intelligence, but notall of it. In particular, what is typically called generalability is largely analytical ability. But, giftedness canoccur in other kinds of skills, especially in creative andpractical ones. Thus, the g-based model deals with part,but not the whole story, of intellectual giftedness.

The Theory ofMultiple Intelligences

Gardner (1983, 1993, 1999) has proposed a theory ofmul-tiple intelligences, which posits distinct linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic,naturalist, interpersonal, intrapersonal, and possibly exis-tential intelligences. Although we have some disagree-ments with his theory (as he does with ours), we view ourwork as basically complementary, in that Gardner's theoryspecifies domains in which intellectual gifts may operate,whereas the theory of successful intelligence specifieskinds of processes. Thus, the theories could be integrated.For example, one might speak of analytical, creative, and

practical processing ofinformation in domains such as thelinguistic domain (e.g., analytical-analyze a poem; cre-ative-write a poem; practical-discuss the implicationsof a poem for everyday life) or the musical domain (e.g.,analytical-analyze a musical score; creative-write amusical score; practical-play a musical score in a way thatwill make "emotional" contact with your audience).

The Schoolwide Enrichment Model

Renzulli and Reis (Renzulli, 1984; Renzulli &Reis, 2000) have proposed a model for schoolwideenrichment based on the notion that a gifted individualdisplays above-average ability, high task commitment,and creativity. Renzulli and Reis have distinguished inthis model between schoolhouse giftedness, or the kindof giftedness that leads to success in school, and creative-productive giftedness, or the kind of giftedness thatmakes more of a difference in terms of contributions ofadults to their domains of inquiry.

In the Schoolwide Enrichment Model, a distinction ismade among three kinds of enrichment in the classroom.Type I enrichment is designed to expose students to mate-rial not ordinarily covered in the curriculum. Type IIenrichment involves students in the pursuit of independ-ent inquiries in self-selected areas. These inquiries, how-ever, represent more traditional kinds of learningexperiences, such as reading and classroom-based ones.Type III enrichment involves independent projects andthe creation ofnew products.

Our model is, again, largely complementary to that ofRenzulli and Reis. Their model specifies "ability," withoutspecifying exactly what goes into this construct. One mightplug the theory of successful intelligence into this "slot."Their model also includes a creativity component, intowhich one might plug our investment theory of creativity(Sternberg & Lubart, 1995). The three kinds of teachingare compatible with the theory of successful intelligence,and, indeed, in some ofour instructional projects, we haveexplicitly used the three types in order to teach better ana-lytical, creative, and practical processing skills.

The Differentiated Model of Giftedness and Talent(DMGT)

The DMGT, proposed by Gagne (1999, 2000), argues thatone can and should distinguish between natural abilities andsystematically developed skills. High levels of natural abili-ties lead to giftedness, whereas high levels of systematicallydeveloped skills lead to talents. The former include abilities

A_==~~~~~~~~~~~O

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0. 0il l11 ^

such as intellectual, creative, socioaffective, and sensorimo-tor abilities. The latter include fields such as academics, arts,business, leisure, social action, sports, and technology.

We probably place more of an emphasis on environ-ment for both gifts and talents than does Gagne, but, ingeneral, we have few quibbles with his model. His intel-lectual, creative, and socioaffective abilities seem to corre-spond quite closely to our analytical, creative, andpractical abilities (Sternberg, 1985), and his list of fieldsclearly is not meant to be complete. Whether one wishesto use the terms gifts and talents in this way seems to uslargely a matter of semantics and is of no great import tous; we do not object to this particular distinction. We dobelieve, however, that so-called "natural" abilities are amatter of developing expertise, and, hence, these abilitiescan be developed into various talents (in the ways he usesthose terms); therefore, we do not see the distinction asbeing quite so clear as he makes it. What constitutes afield will differ from one society to another(Csikszentmihalyi, 1996), and the ways in which oneexpresses abilities will also differ from one society toanother, so that the same expressions that are seen asintelligent in one culture may be seen as unintelligent inanother (Sternberg et al., 2001).

From our point of view, talents are more importantthan abilities, in the way the terms are used in this model.The reason is that talents refer largely to realized abilities,and it is what a person does with his or her abilities, ulti-mately, that will determine whether or not he or she has acontribution to make. High intellectual abilities, in and ofthemselves, seem inadequate as a basis for giftednessbecause so many people with such abilities squander themand, therefore, do not, in our view, deserve the label ofgifted. The issue here is not merely semantic becauseschools may allocate their special-education resources onthe basis ofhow this label is assigned. CI

C o n c I u s i o n

The theory of successful intelligence provides aproven model for gifted education. The model has impli-cations for identification, instruction, and assessment ofachievement. All three should be viewed in terms of ana-lytical, creative, and practical abilities. The results ofdiverse studies suggest that the theory of successful intelli-gence is valid as a whole and provides successful interven-tions in classrooms.

Adopting any new model for the identification, instruc-tion, and assessment of gifted children is time-consuming

and potentially difficult. Yet, we believe that the evidencesupporting the model of successful intelligence is sufficientto suggest that adoption will be well worth the effort.

References

Anastasi, A., & Urbina, S. (1997). Psychological testing (7th ed.).Upper Saddle River, NJ: Prentice Hall.

Callahan, C. M. (2000). Intelligence and giftedness. In R. J.Sternberg (Ed.), Handbook of intelligence (pp. 159-175). NewYork: Cambridge University Press.

Csikszentmihalyi, M. (1996). Creativity. New York:HarperCollins.

Gagne, F. (1999). My convictions about the nature of humanabilities, gifts, and talents. Journal for the Education of theGifted, 22, 191-234.

Gagne, F. (2000). Understanding the complex choreography oftalent development. In K. A. Heller, F. J. Monks, R. J.Sternberg, & R. F. Subotnik (Eds.), International handbook ofg~ftedness and talent (2nd ed., pp. 67-79). Amsterdam:Elsevier

Gallagher, J. J., & Courtright, R. D. (1984). The educationaldefinition of giftedness and its policy implications. In R. J.Sternberg & J. E. Davidson (Eds.), Conceptions of gfitedness(pp. 93-111). New York: Cambridge University Press.

Gardner, H. (1983). Frames of mind: The theory of multiple intelli-gences. New York: BasicBooks.

Gardner, H. (1993). Multiple intelligences: The theory in practice.New York: BasicBooks.

Gardner, H. (1999). Intellligence reframed. New York: BasicBooks.Grigorenko, E. L. (2000). Heritability and intelligence. In R. J.

Sternberg (Ed.), Handbook of intelligence (pp. 53-91). NewYork: Cambridge University Press.

Grigorenko, E. L., & Sternberg, R. J. (2001). Analytical, cre-ative, and practical intelligence as predictors of self-reportedadaptive functioning: A case study in Russia. Intelligence,29(1), 57-73.

Heller, K. A., Monks, F. J., Sternberg, R. J., & Subotnik, R. F.(Eds.). (2000). International handbook of gftedness and talent(2nd ed.). Amsterdam: Elsevier.

Jensen, A. (1998). The g factor. Westport, CT: Praeger-Greenwood.

Laboratory of Comparative Human Cognition. (1983). Cultureand cognitive development. In P. Mussen (Series Ed.) & W.Kessen (Vol. Ed.), Handbook ofchildpsychology (pp. 295-356).New York: Wiley.

Renzulli, J. S. (1984). The three-ring conception of gifted-ness: A developmental model for creative productivity. InR. J. Sternberg &J. E. Davidson (Eds.), Conceptions ofgift-edness (pp. 53-92). New York: Cambridge UniversityPress.

Renzulli, J. S., & Reis, S. M. (2000). The schoolwide enrich-ment model. In K. A. Heller, F. J. Monks, R. J. Sternberg,

3 5 0 .~~~~~~ . 50 0L49 .. 0~

at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from

0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

& R. F. Subotnik (Eds.), International handbook of gftednessand talent (2nd ed., pp. 367-382). Amsterdam: Elsevier.

Schon, D. A. (1983). The reflective practitioner. New York:BasicBooks.

Serpell, R. (2000). Intelligence and culture. In R. J. Sternberg(Ed.), Handbook of intelligence (pp. 549-580). New York:Cambridge University Press.

Spearman, C. (1927). The abilities ofman. New York: Macmillan.Sternberg, R. J. (1977). Component processes in analogical rea-

soning. Psychological Review, 84, 353-378.Sternberg, R. J. (1985). Beyond IQ: A triarchic theory ofhuman intel-

ligence. New York: Cambridge University Press.Sternberg, R. J. (1993). Sternberg Triarchic Abilities Test.

Unpublished test.Sternberg, R. J. (1995). In search of the human mind. Orlando:

Harcourt Brace College Publishers.Sternberg, R. J. (1997a). Successful intelligence. New York: Plume.Sternberg, R. J. (1997b). What does it mean to be smart?

Educational Leadership, 54(6), 20-24.Sternberg, R. J. (1998a). Abilities are forms of developing

expertise. Educational Researcher, 27, 11-20.Sternberg, R. J. (1998b). Principles of teaching for successful

intelligence. Educational Psychologist, 33, 65-72.Sternberg, R. J. (1999a). Intelligence as developing expertise.

Contemporary Educational Psychology, 24, 259-375.Sternberg, R. J. (1999b). Successful intelligence: Finding a bal-

ance. Trends in Cognitive Sciences, 3, 436-442.Sternberg, R. J. (1999c). The theory of successful intelligence.

Review of General Psychology, 3, 292-316.Sternberg, R. J., Castejon, J. L., Prieto, M. D., Hautamiki, J., &

Grigorenko, E. L. (2001). Confirmatory factor analysis ofthe Sternberg Triarchic Abilities Test in three internationalsamples: An empirical test of the triarchic theory of intelli-gence. EuropeanJournal ofPsychological Assessment 17, 1-16.

Sternberg, R. J., & Clinkenbeard, P. R. (1995). A triarchicmodel ofidentifying, teaching, and assessing gifted children.Roeper Review, 17, 255-260.

Sternberg, R. J., & Davidson, J. E. (Eds.). (1984). Conceptions ofgftedness. New York: Cambridge University Press.

Sternberg, R. J., Ferrari, M., Clinkenbeard, P. R., &Grigorenko, E. L. (1996). Identification, instruction, andassessment ofgifted children: A construct validation ofa tri-archic model. Gifted Child Quarterly, 40, 129-137.

Sternberg, R. J., Forsythe, G. B., Hedlund, J., Horvath, J.,Snook, S., Williams, W. M., Wagner, R. K., &Grigorenko, E. L. (2000). Practical intelligence in everydaylife. New York: Cambridge University Press.

Sternberg, R. J., & Grigorenko, E. L. (Eds.). (1997). Intelligence,heredity, and environment. New York: Cambridge UniversityPress.

Sternberg, R. J., & Grigorenko, E. L. (2000). Teachingfor success-ful intelligence. Arlington Heights, IL: Skylight Training andPublishing.

Sternberg, R. J., Grigorenko, E. L., Ferrari, M., &Clinkenbeard, P. (1999). A triarchic analysis of an aptitude-

treatment interaction. European Journal of PsychologicalAssessment, 15, 1-11.

Sternberg, R. J., Grigorenko, E. L., & Jarvin, L. (2001).Improving reading instruction: The triarchic model.Educational Leadership 58(6), 48-52.

Sternberg, R. J., & Lubart, T. I. (1995). Defying the crowd: Cultivatingcreativity in a culture ofconformity. New York: Free Press.

Sternberg, R. J., Nokes, K., Geissler, P. W., Prince, R.,Okatcha, F., Bundy, D. A., & Grigorenko, E. L. (2001).The relationship between academic and practical intelli-gence: A case study in Kenya. Intelligence, 29, 401-418.

Sternberg, R. J., & O'Hara, L. A. (2000). Intelligence and cre-ativity. In R. J. Sternberg (Ed.), Handbook of intelligence (pp.609-628). New York: Cambridge University Press.

Sternberg, R. J., Torff, B., & Grigorenko, E. L. (1998a).Teaching for successful intelligence raises school achieve-ment. Phi Delta Kappan, 79, 667-669.

Sternberg, R. J., Torff, B., & Grigorenko, E. L. (1998b).Teaching triarchically improves school achievement.Journalof Educational Psychology, 90, 1-11.

Torrance, E. P. (1974). Torrance Test of Creative Thinking.Lexington, MA: Personnel Press.

End Note

1. Note that, because we view all tests of abilities as testsof developing expertise, tests of abilities inevitably meas-ure both achievement and abilities. It is impossible to sep-arate the two kinds of measures cleanly.

Author Note

Preparation of this article was supported by GrantREC-9979843 from the National Science Foundationand by a government grant under the Javits Act Program(Grant No. R206R000001) as administered by the Officeof Educational Research and Improvement, U.S.Department of Education. Grantees undertaking suchprojects are encouraged to express freely their professionaljudgment. This article, therefore, does not necessarily rep-resent the positions or the policies of the U.S. govern-ment, and no official endorsement should be inferred.Information about the Sanger Academy, which uses thesuccessful-intelligence model, can be obtained fromKenneth Garcia, Principal, Sanger Academy, 1905Seventh St., Sanger, CA 93657. Requests for informationshould be sent to Robert J. Sternberg, Director, YaleCenter for the Psychology ofAbilities, Competencies, andExpertise, Yale University, P.O. Box 208358, NewHaven, CT 06520-8358.

0 .. 0

I ill .~i I0l at Levinsky College of Education on April 15, 2013gcq.sagepub.comDownloaded from