Agrowing number of business leaders,politicians, and educators are unitedaround the idea that students need “21stcentury skills” to be successful today. It’sexciting to believe that we live in times

that are so revolutionary that they demand new anddifferent abilities. But in fact, the skills students needin the 21st century are not new.

Critical thinking and problem solving, for example,have been components of human progress throughouthistory, from the development of early tools, to agri-cultural advancements, to the invention of vaccines, toland and sea exploration. Such skills as informationliteracy and global awareness are not new, at least notamong the elites in different societies. The need formastery of different kinds of knowledge, ranging fromfacts to complex analysis? Not new either. In TheRepublic, Plato wrote about four distinct levels of intel-lect. Perhaps at the time, these were considered “3rdcentury BCE skills”?

What’s actually new is the extent to which changesin our economy and the world mean that collectiveand individual success depends on having such skills.Many U.S. students are taught these skills—those whoare fortunate enough to attend highly effective schoolsor at least encounter great teachers—but it’s a matterof chance rather than the deliberate design of ourschool system. Today we cannot afford a system inwhich receiving a high-quality education is akin to agame of bingo. If we are to have a more equitable andeffective public education system, skills that have beenthe province of the few must become universal.

This distinction between “skills that are novel” and“skills that must be taught more intentionally andeffectively” ought to lead policymakers to differenteducation reforms than those they are now consid-ering. If these skills were indeed new, then perhaps wewould need a radical overhaul of how we think aboutcontent and curriculum. But if the issue is, instead,that schools must be more deliberate about teachingcritical thinking, collaboration, and problem solving toall students, then the remedies are more obvious,although still intensely challenging.

What Will It Take?The history of U.S. education reform should greatlyconcern everyone who wants schools to do a betterjob of teaching students to think. Many reform efforts,from reducing class size to improving reading instruc-tion, have devolved into fads or been implementedwith weak fidelity to their core intent. The 21stcentury skills movement faces the same risk.

To complicate the challenge, some of the rhetoricwe have heard surrounding this movement suggeststhat with so much new knowledge being created,content no longer matters; that ways of knowing infor-mation are now much more important than informa-tion itself. Such notions contradict what we knowabout teaching and learning and raise concerns thatthe 21st century skills movement will end up being aweak intervention for the very students—low-incomestudents and students of color—who most needpowerful schools as a matter of social equity.

The debate is not about content versus skills. There

16 E D U C AT I O N A L L E A D E R S H I P / S E P T E M B E R 2 0 0 9

To work, the 21st century skills movement will require keen attention to curriculum, teacher quality, and assessment.

Andrew J. Rotherham and Daniel Willingham

21st Century

Rotherham pp16-21_2:EL Template 7/31/09 9:10 AM Page 16

A S C D / W W W. A S C D . O R G 17

The Challenges AheadSkills:

Rotherham pp16-21_2:EL Template 7/31/09 9:10 AM Page 17

is no responsible constituency arguingagainst ensuring that students learn howto think in school. Rather, the issue ishow to meet the challenges of deliveringcontent and skills in a rich way thatgenuinely improves outcomes forstudents.

What will it take to ensure that theidea of “21st century skills”—or moreprecisely, the effort to ensure that allstudents, rather than just a privilegedfew, have access to a rich education thatintentionally helps them learn theseskills—is successful in improvingschools? That effort requires threeprimary components. First, educatorsand policymakers must ensure that theinstructional program is complete andthat content is not shortchanged for anephemeral pursuit of skills. Second,states, school districts, and schools needto revamp how they think about humancapital in education—in particular howteachers are trained. Finally, we neednew assessments that can accuratelymeasure richer learning and morecomplex tasks.

For the 21st century skills effort to beeffective, these three elements must beimplemented in concert. Otherwise, thereform will be superficial and counter-productive.

Better CurriculumPeople on all sides of this debate oftenspeak of skills and knowledge as sepa-rate. They describe skills as akin to afunction on a calculator: If your calcu-lator can compute square roots, it cando so for any number; similarly, if astudent has developed the ability to“think scientifically,” he or she can do sowith any content. In this formulation,domain knowledge is mainly importantas grist for the mill—you need some-thing to think about.

Skills and knowledge are not sepa-rate, however, but intertwined. In somecases, knowledge helps us recognize the

underlying structure of a problem. Forexample, even young children under-stand the logical implications of a rulelike “If you finish your vegetables, youwill get a cookie after dinner.” They candraw the logical conclusion that a childwho is denied a cookie after dinnermust not have finished her vegetables.

Without this familiar context, however,the same child will probably find it diffi-cult to understand the logical formmodus tollens, of which the cookie rule isan example. (If P, then Q. Q is false.Therefore, P is false.) Thus, it’s inaccurateto conceive of logical thinking as a sepa-rate skill that can be applied across avariety of situations. Sometimes we failto recognize that we have a particularthinking skill (such as applying modus

tollens) unless it comes in the form ofknown content.

At other times, we know that we havea particular thinking skill, but domainknowledge is necessary if we are to useit. For example, a student might havelearned that “thinking scientifically”requires understanding the importanceof anomalous results in an experiment.If you’re surprised by the results of anexperiment, that suggests that yourhypothesis was wrong and the data aretelling you something interesting. But tobe surprised, you must make a predic-tion in the first place—and you can onlygenerate a prediction if you understandthe domain in which you are working.Thus, without content knowledge weoften cannot use thinking skills properlyand effectively.

Why would misunderstanding therelationship of skills and knowledgelead to trouble? If you believe that skillsand knowledge are separate, you arelikely to draw two incorrect conclu-sions. First, because content is readilyavailable in many locations but thinkingskills reside in the learner’s brain, itwould seem clear that if we must choosebetween them, skills are essential,whereas content is merely desirable.Second, if skills are independent ofcontent, we could reasonably concludethat we can develop these skills throughthe use of any content. For example, ifstudents can learn how to think criti-cally about science in the context of anyscientific material, a teacher shouldselect content that will engage students(for instance, the chemistry of candy),even if that content is not central to thefield. But all content is not equallyimportant to mathematics, or to science,or to literature. To think critically,students need the knowledge that iscentral to the domain.

The importance of content in thedevelopment of thinking creates severalchallenges for the 21st century skills

18 E D U C AT I O N A L L E A D E R S H I P / S E P T E M B E R 2 0 0 9

Curriculum, teacher expertise,and assessment have all been weak linksin past educationreform efforts.

Be careful what you give children, for sooner or l

Rotherham pp16-21_2:EL Template 7/31/09 9:10 AM Page 18

movement. The first is the temptation toemphasize advanced, conceptualthinking too early in training—anapproach that has proven ineffective innumerous past reforms, such as the“New Math” of the 1960s (Loveless,2002). Learning tends to follow apredictable path. When students firstencounter new ideas, their knowledge isshallow and their understanding isbound to specific examples. They needexposure to varied examples before theirunderstanding of a concept becomesmore abstract and they can successfullyapply that understanding to novel situations.

Another curricular challenge is thatwe don’t yet know how to teach self-direction, collaboration, creativity, andinnovation the way we know how toteach long division. The plan of 21stcentury skills proponents seems to be togive students more experiences that willpresumably develop these skills—forexample, having them work in groups.But experience is not the same thing aspractice. Experience means only thatyou use a skill; practice means that youtry to improve by noticing what you aredoing wrong and formulating strategiesto do better. Practice also requires feed-back, usually from someone moreskilled than you are.

Because of these challenges, devisinga 21st century skills curriculum requiresmore than paying lip service to contentknowledge. Outlining the skills in detailand merely urging that content betaught, too, is a recipe for failure. Wemust plan to teach skills in the contextof particular content knowledge and totreat both as equally important.

In addition, education leaders mustbe realistic about which skills are teach-able. If we deem that such skills ascollaboration and self-direction areessential, we should launch a concertedeffort to study how they can be taughteffectively rather than blithely assume

that mandating their teaching will resultin students learning them.

Better TeachingGreater emphasis on skills also hasimportant implications for teachertraining. Our resolve to teach theseskills to all students will not be enough.We must have a plan by which teacherscan succeed where previous generationshave failed.

Advocates of 21st century skills favorstudent-centered methods—forexample, problem-based learning and

project-based learning—that allowstudents to collaborate, work onauthentic problems, and engage withthe community. These approaches arewidely acclaimed and can be found inany pedagogical methods textbook;teachers know about them and believethey’re effective. And yet, teachers don’tuse them. Recent data show that mostinstructional time is composed of seat-work and whole-class instruction led bythe teacher (National Institute of ChildHealth and Human Development EarlyChild Care Research Network, 2005).Even when class sizes are reduced,teachers do not change their teachingstrategies or use these student-centeredmethods (Shapson, Wright, Eason, &Fitzgerald, 1980). Again, these are notnew issues. John Goodlad (1984)reported the same finding in his land-mark study published more than 20years ago.

Why don’t teachers use the methodsthat they believe are most effective?

Even advocates of student-centeredmethods acknowledge that thesemethods pose classroom managementproblems for teachers. When studentscollaborate, one expects a certainamount of hubbub in the room, whichcould devolve into chaos in less-than-expert hands. These methods alsodemand that teachers be knowledgeableabout a broad range of topics and areprepared to make in-the-moment deci-sions as the lesson plan progresses.Anyone who has watched a highly effec-tive teacher lead a class by simultane-

ously engaging with content, classroommanagement, and the ongoing moni-toring of student progress knows howintense and demanding this work is. It’sa constant juggling act that involveskeeping many balls in the air.

Part of the 21st century skills move-ment’s plan is the call for greater collab-oration among teachers. Indeed, this isone of the plan’s greatest strengths; wewaste a valuable resource when we don’tgive teachers time to share theirexpertise. But where will schools findthe release time for such collaboration?Will they hire more teachers or increaseclass size? How will they provide thetechnology infrastructure that willenable teachers to collaborate with morethan just the teacher down the hall?Who will build and maintain and editthe Web sites, wikis, and so forth? Thesechallenges raise thorny questions aboutwhether the design of today’s schools iscompatible with the goals of the 21stcentury skills movement.

A S C D / W W W. A S C D . O R G 19

We don’t yet know how to teach self-direction, collaboration, creativity, and innovation the way we know how to teach long division.

r later you are sure to get it back. —Barbara Kingsolver

Rotherham pp16-21_2:EL Template 7/31/09 9:10 AM Page 19

For change to move beyond adminis-trators’ offices and penetrate classrooms,we must understand that professionaldevelopment is a massive undertaking.Most teachers don’t need to bepersuaded that project-based learning isa good idea—they already believe that.What teachers need is much morerobust training and support than theyreceive today, including specific lessonplans that deal with the high cognitivedemands and potential classroommanagement problems of using student-centered methods.

Unfortunately, there is a widespreadbelief that teachers already know how todo this if only we could unleash themfrom today’s stifling standards andaccountability metrics. This notionromanticizes student-centered methods,underestimates the challenge of imple-menting such methods, and ignores thelack of capacity in the field today.

Instead, staff development plannerswould do well to engage the bestteachers available in an iterative processof planning, execution, feedback, andcontinued planning. This process, alongwith additional teacher training, willrequire significant time. And of coursenone of this will be successful withoutbroader reforms in how teachers arerecruited, selected, and deselected in aneffort to address the whole picture ofeducation’s human capital challenge.

Better TestsThere is little point in investing heavilyin curriculum and human capitalwithout also investing in assessments to

evaluate what is or is not being accom-plished in the classroom. Fortunately, asElena Silva (2008) noted in a recentreport for Education Sector, the poten-tial exists today to produce assessmentsthat measure thinking skills and are alsoreliable and comparable betweenstudents and schools—elements integralto efforts to ensure accountability andequity. But efforts to assess these skillsare still in their infancy; education facesenormous challenges in developing theability to deliver these assessments atscale.

The first challenge is the cost.Although higher-level skills like criticalthinking and analysis can be assessedwith well-designed multiple-choicetests, a truly rich assessment systemwould go beyond multiple-choicetesting and include measures thatencourage greater creativity, show howstudents arrived at answers, and evenallow for collaboration. Such measures,however, cost more money than policy-makers have traditionally been willingto commit to assessment. And, at a timewhen complaining about testing is anational pastime and cynicism aboutassessment, albeit often uninformed, ison the rise, getting policymakers tocommit substantially more resources toit is a difficult political challenge.

Producing enough high-qualityassessments to meet the needs of asystem as large and diverse as U.S.public schools would stretch thecapacity of the assessment industry, andincentives do not exist today for manynew entrants to become major players

in that field. We would need a coordi-nated public, private, and philanthropicstrategy—including an intensiveresearch and development effort—tofoster genuine change.

Substantial delivery challenges alsoremain. Delivering these assessments ina few settings, as is the case today, ishardly the same as delivering them atscale across a state—especially the largerstates. Because most of these assess-ments will be technology-based, mostschools’ information technology systemswill require a substantial upgrade.

None of these assessment challengesare insurmountable, but addressingthem will require deliberate attentionfrom policymakers and 21st centuryskills proponents, as well as a deviationfrom the path that policymaking is ontoday. Such an effort is essential. Whymount a national effort to changeeducation if you have no way ofknowing whether the change has beeneffective?

A Better, But Harder, WayThe point of our argument is not to saythat teaching students how to think,work together better, or use new infor-mation more rigorously is not a worthyand attainable goal. Rather, we seek tocall attention to the magnitude of thechallenge and to sound a note ofcaution amidst the sirens calling ourpolitical leaders once again to the rockyshoals of past education reform failures.Without better curriculum, betterteaching, and better tests, the emphasison “21st century skills” will be a superfi-cial one that will sacrifice long-termgains for the appearance of short-termprogress.

Curriculum, teacher expertise, andassessment have all been weak links inpast education reform efforts—a factthat should sober today’s skills propo-nents as they survey the task of dramati-cally improving all three. Efforts to

20 E D U C AT I O N A L L E A D E R S H I P / S E P T E M B E R 2 0 0 9

Devising a 21st century skills curriculum requires more than payinglip service to content knowledge.

Rotherham pp16-21_2:EL Template 7/31/09 9:10 AM Page 20

create more formalized common stan-dards would help address some of thechallenges by focusing efforts in acommon direction. But common stan-dards will not, by themselves, beenough.

The past few decades have seen greatprogress in education reform in theUnited States—progress that has espe-cially benefited less-advantagedstudents. Today’s reformers can build onthat progress only if they pay keenattention to the challenges associatedwith genuinely improving teaching andlearning. If we ignore these challenges,the 21st century skills movement risksbecoming another fad that ultimatelychanges little—or even worse, sets backthe cause of creating dramatically more

powerful schools for U.S. students,especially those who are underservedtoday.

ReferencesGoodlad, J. I. (1984). A place called school.

New York: McGraw-Hill.Loveless, T. (2002). A tale of two math

reforms: The politics of the new math andNCTM standards. In T. Loveless (Ed.),The great curriculum debate (pp. 184–209).Washington, DC: Brookings.

National Institute of Child Health and

Human Development Early Child CareResearch Network. (2005). A day in thethird grade: A large-scale study of class-room quality and teacher and studentbehavior. Elementary School Journal, 105,305–323.

Shapson, S. M., Wright, E. N., Eason, G., &Fitzgerald, J. (1980). An experimentalstudy of the effects of class size. AmericanEducational Research Journal, 17, 141–152.

Silva, E. (2008). Measuring skills for the 21stcentury. Washington, DC: EducationSector. Available: www.educationsector.org/usr_doc/MeasuringSkills.pdf

Andrew J. Rotherham is Cofounder andPublisher of Education Sector and writesthe blog Eduwonk.com; arotherham@educationsector.org. Daniel Willinghamis Professor of Psychology at the Univer-sity of Virginia and the author of WhyDon’t Students Like School? (Jossey-Bass, 2009); willingham@virginia.edu.

EL

A S C D / W W W. A S C D . O R G 21

Skills and knowledgeare not separate, but intertwined.

Rotherham pp16-21_2:EL Template 7/31/09 9:10 AM Page 21