What would you guess it takes to be hired at Google? A computer science degree? Academic prowess? The ability to arrive quickly and confidently at the right answer to a question?
Not so long ago, Google famously asked every job candidate for a transcript, G.P.A., and test scores before considering them for a position.
But as you might expect from a fundamentally data-driven company, Google regularly examines its own hiring methods, collecting and analyzing tremendous amounts of information from employees and adjusting its policies accordingly.
In 2011, Google released the results of a study called Project Oxygen, which showed that its old hiring model proved very little about a candidate’s potential for success.
“One of the things we’ve seen from all our data crunching is that G.P.A.’s are worthless as a criteria for hiring, and test scores are worthless — no correlation at all except for brand-new college grads, where there’s a slight correlation,” said senior vice president for people operations Laszlo Bock in an interview with Adam Bryant of the New York Times.
“What’s interesting is the proportion of people without any college education at Google has increased over time as well. So we have teams where you have 14 percent of the team made up of people who’ve never gone to college.”
What’s more, the kind of “tests” hiring teams now use in interviews focus on behavior, not knowledge.
For example, managers now ask questions like, “Give me an example of a time when you solved an analytically difficult problem” instead of “How many bowling balls can you fit inside this room?”
“The interesting thing about the behavioral interview is that when you ask somebody to speak to their own experience, and you drill into that, you get two kinds of information,” says Bock. “One is you get to see how they actually interacted in a real-world situation, and the valuable ‘meta’ information you get about the candidate is a sense of what they consider to be difficult.”
But aren’t we all trained in school, from an early age, to be able to demonstrate our competence by arriving at the right answer to a tricky question? Don’t the standards of the classroom prepare us for the standards of the real world?
“Brainteasers are a complete waste of time,” says Bock. “How many golf balls can you fit into an airplane? How many gas stations in Manhattan? A complete waste of time. They don’t predict anything. They serve primarily to make the interviewer feel smart.”
Hold on. In every high-performing nation, standardized tests are embedded in the wiring of schools, particularly in high schools. In the developed world, 76 percent of students attend high schools that use them, according to the OECD. Granted, each country uses a slightly different system.
In West Germany, for instance, education is the responsibility of the states, rather than the national government. Standardized tests are not used on a population-wide basis, and the use of standardized tests is largely restricted to counseling centers and similar specialists in the schools.
Neither achievement nor intelligence tests are often used in the schools. The Netherlands created a national curriculum development center in 1975 and has created national examinations, although they are not yet widely used. Achievement tests are used by teachers only, and intelligence test use is similar to that in West Germany. In Sweden, national standardized tests based on objective techniques are used above the primary levels. And the list goes on…
Why would so many societies be using standardized tests if they didn’t predict anything important?
Daniel Koretz, a scholar in educational measurement, has been studying standardized testing models for over two decades.
“Standardized tests as we currently have them are enormously useful,” Koretz says. “Anyone who follows education knows that there are enormous inequities in American schooling. One of the ways we know this is because of standardized tests.
“The problem is not that the testing shouldn’t be used. I wouldn’t argue that testing shouldn’t be used, we’re just using it in the wrong way, putting far too much pressure on increases in scores.”
In the United States, where K-12 standardized testing has become a multi-billion-dollar industry, highly regarded and national institutions and associations such as the National Board on Educational Testing and Public Policy and the American Educational Research Association have been writing for decades about the pitfalls of both standardized testing in general and the over-reliance on standardized testing, especially in using these tests to make high-stakes decisions about things like high school graduation and teacher evaluation.
Finland uses a standardized testing system, but more effectively than most countries. Finnish high school seniors are required to take a standardized, national matriculation exam that determines their chances of attending a Finnish university. The exam stretches out over three grueling weeks and takes about 43 hours. Whereas in the United States, tests have greater consequences for teachers and administrators than for students, in Finland, the students are the ones who feel the pressure.
Tests are hard, and they affect students’ lives. The same is true in South Korea, the other top-performing nation on international tests.
But is creating a better test really the answer? Maybe it’s not the Finnish testing system that makes a difference but the selectivity of its teacher-preparation programs. And even if it is the Finnish testing system, who’s to say that what’s being tested is what students will actually need to succeed in their careers? Even if every student in the developed world aced his or her college entrance exams, would they be able to land a job at Google? McDonald’s, even?
In his book, Creating Innovators: The Making of Young People Who Will Change the World, Harvard education specialist Tony Wagner says many of today’s K-12 and college tracks are not consistently “adding the value and teaching the skills that matter most in the marketplace.”
Because today’s job landscape is changing so fast, and because high-paying, middle-skilled vocations are fewer and father between, it is absolutely imperative for young professionals to be able to solve problems creatively and think critically.
The capacity to innovate, Wagner says, is far more important than academic knowledge.
But how do you measure innovation? How do you set standards for it, define it, teach it? Can it become part of each school system’s curriculum the way that reading, math, and writing are?
It’s too bad we were all too busy filling in bubbles on test sheets to hone our creative skills, or else we might just be able to think of an answer.
Six Ways Standardized Testing Harms Students:
1. Misused and Misled
“Before Jesus Christ was born, human beings were taking tests,” writes Amanda Ripley for Education Nation. “Civil service exams date back to China’s Han Dynasty (206 BC – 220 AD.) Hiring test-prep tutors—and cheating— go back about as far, by the way.”
And when she says “cheating,” Ripley is hardly referring to the students themselves. She’s referring to school systems, policymakers, and anyone else who cheats students out of opportunities for success by maintaining the status quo.
Schools and districts across the United States have been caught cheating— changing test answers or giving their students test problems ahead of time— including Atlanta, Philadelphia, Washington, D.C., and Texas. A March, 2011 USA Today investigation showed that the dramatic rise in D.C. test scores was due to cheating, not to effective administration. There have also been instances in which tests were scored incorrectly, failing and sending students who had actually passed the tests to summer school.
In October 2010, the Chicago Tribune reported that when Illinois adjusted the ISAT scoring system in 2006, it “lowered the number of points required to pass,” resulting in more students appearing proficient than was actually the case. A recent study by the Chicago Consortium on School Research found that, after controlling for differences in assessments and changing demographics, there was essentially no increase in scores of elementary students over two decades of test-driven reform.
If the tests we use to measure student learning are themselves invalid, then the inferences we draw and the direction we derive from them are inherently misleading.
2. Knowledge is Dead
“What you know matters far less than what you can do with what you know,” says Tony Wagner. Because knowledge is available on every Internet-connected device, and instant answers can be found on Google, tests that focus on content knowledge—the tests most often and easily standardized— disengage and, ultimately, disadvantage students.
Look at Google’s hiring team. Why should we continue to emphasize content knowledge when employers aren’t asking for it?
“Every young person will continue to need basic knowledge, of course,” he says. “But they will need skills and motivation even more… [which are] increasingly important as many traditional careers disappear.”
But don’t count on students to stand out if they are constantly being trained to fit in.
3. You Are What You Score
In her book Now You See It, Cathy Davidson points to the similarity between standardized tests and the “assembly line model,” effectively placing kids inside a one-fits-all educational mold and labeling low-scoring students as failures.
When students are already wired, as humans, to compare themselves to others, it only exacerbates the situation when their basis for comparison is designed to put some of them at a disadvantage. Standardization may enable consistent measurement, but it creates a nasty byproduct in the process: a consistently distorted self-image.
Students who ace tests internalize their performance as self-worth, and students who fail tests (and see others succeeding) internalize their performance as self-worthlessness. This trend can last throughout an entire educational career—or lack thereof.
A Harvard University study found that standardized testing actually increases the drop-out rate. Students in the bottom 10 percent of achievement were 33 percent more likely to drop out of school in states with graduation tests. The National Research Council found that low-performing elementary and secondary school students who are held back do less well academically, are much worse off socially, and are far likelier to drop out than equally weak students who are promoted.
The simple act of writing your name on a test sheet is like signing your life away.
4. Ignoring the Individual
Because standardized tests are, by definition, meant to be administered, scored, and interpreted in a standardized and consistent manner, they ignore differences in student learning style and background.
This is perhaps the least remediable aspect of the tests, and for that reason the most harmful. While not all standardized tests use multiple choice questions—many are actually performance or project based—they are designed to judge all students using the same set of criteria.
And while this is completely necessary for efficient grading, it does not take into account individual variances in learning style or background, and teaches students to follow guidelines more than it teaches them to think outside the box.
5. What’s Not Tested Is Not Taught
Educator Alfie Kohn advises parents to ask an unusual question when a school’s test scores increase: “What did you have to sacrifice about my child’s education to raise those scores?”
As schools struggle to avoid the “underperforming” label, entire subject areas—such as music, art, social studies, and foreign languages—are de-emphasized. What is not tested does not count, and 85 percent of teachers believe that their school gives less attention to subjects that are not on the state test.
One teacher had this to say about how the timing of state tests drives teaching: “At our school, third- and fourth-grade teachers are told not to teach social studies and science until March.” As “real learning” takes a backseat to “test learning,” challenging curriculum is replaced by multiple choice materials, individualized student learning projects disappear, and in-depth exploration of subjects along with extracurricular activities are squeezed out of the curriculum.
6. Students As Guinea Pigs
Koretz says that while in some nations like the Netherlands, it’s routine for schools to be visited by outside inspectors who look at far more than test scores (and, in fact, the Dutch public often hears about test scores in the context of an inspector’s report), in the United States we don’t measure and tweak. We realize, in hindsight, that a policy didn’t work, leaving many classes of students behind in the process.
“What we have is a lot of interesting ideas about better ways of holding schools accountable and very little hard research,” says Koretz. “And I would say that that’s really an ethical problem, not just a political problem. It’s a political problem because we lack information that we could use to better serve children. It’s an ethical problem because children are not consenting adults. When we drop into schools these very high powered policies that clearly change teacher’s behavior in dramatic ways, we have an obligation, in my view, to monitor what happens.”
To get hired at Google, Microsoft, BBC News, Peking University, the Commonwealth Bank of Australia, St. Mary’s Hospital, the International Grocer’s Association, even the local burger joint—or to invent a new job in ten years— students need to spend more time using their skills than measuring them.