Wednesday, September 5, 2012

Standardised results fail to pass the good education test





Anna Bernasek 

As September dawns, the “back to school” season begins in America. Millions of families prepare their offspring for new schools and new classes in the coming academic year.

So how is the American school system doing? The answer isn’t particularly reassuring. Formerly the most admired education system in the world, US education seems costly, inefficient and increasingly ineffectual compared with other systems.

Education is a big business. Americans spend more than $US600 billion a year on public elementary and secondary education. Meanwhile, survey after survey shows American students on average are slipping in educational attainment compared with the rest of the world, at least as far as standardised tests can measure.

The centrepiece of American education reform to date has been to link teacher evaluations to student test scores. In 2009, $US4.35 billion worth of federal grants went to states with teacher evaluation systems based on student test performance.

So popular is the idea of measuring teachers based on student test scores that the Obama administration is hoping to spread the use of standardised tests to the university level.

The idea is to use the standardised tests as a way to compare results to costs at colleges. Over the past 30 years the cost of college education has soared in the US. College remains a major financial challenge for many families, leaving many wondering if it’s really worth the expense.

Standardised testing has broad support. In addition to President Barack Obama, even fierce opponents such as conservative Republican governor of New Jersey Chris Christie believe it’s a great idea.
Christie just signed a teacher tenure reform bill tied to student test scores. Now when teachers are evaluated for tenure after four years, student test scores will be taken into account. Although the actual weight of test scores is still to be determined, Christie wants test scores to count for half of a teacher’s tenure evaluation.

In neighbouring New York, after a two-year fight, the state passed a new teacher evaluation system in February. School districts can link up to 40 per cent of a teacher’s pay to student test scores.

The appeal is that test scores are simple. They can be easily ranked and applied to teachers and schools as well as students. They seem objective. But while measurement is crucial in business and economics, you have to measure the right things, the right way.

Consider how reliable standardised tests actually are as an indicator of future performance. On that score standardised tests have plenty of flaws. Often they don’t accurately represent what students know or understand. They certainly can’t measure a student’s interest or engagement in a subject. And standardised tests tend to promote narrow, literal interpretations over nuance and creativity.

As standardised tests gain currency with students, teachers and schools, the school curriculum becomes dominated by test preparation. Unfortunately teaching to the test rather than teaching other intellectual skills students need to grow and develop can actually limit or hinder academic achievement.

There are some unintended consequences. For one, cheating seems to have become more widespread. With so much now riding on the results, Americans have experienced cheating scandals involving not only students at our best schools but even teachers and administrators as well.

And then there’s the business of test marking and scoring. In the US that’s a $US700 million market dominated by a handful of big companies: Harcourt Educational Measurement, CTB/McGraw-Hill, Riverside Publishing (a Houghton Mifflin company), and NCS Pearson. Recently Rupert Murdoch’s News Corp announced plans to enter the market. These companies not only supply tests but also test-preparation materials, curriculum content and teacher evaluation. It’s a self-reinforcing market where sales of one service lead to sales of the others.

There’s no doubt teachers need to be evaluated in the job. High-performing teachers need challenge and reinforcement, and low-performing teachers need to improve or move on.

But standardised test scores are a poor substitute for thorough supervision. They are far too narrow a measure to form a true picture of teacher performance.

There are other ways to do it. In fields where education is crucial, for example law or medicine, standardised test results are used to weed out the most unprepared or ill-equipped students. But individual evaluations by knowledgeable professionals are the gold standard for determining who are the top law and medical students, as well as the top schools and teachers. Beyond a threshold, test scores aren’t particularly important in these professions.

So while Americans struggle to reform their education sector, they haven’t spent enough time answering a basic question: what does a good education consist of? Without that answer, all the testing in the world won’t do much good.

No comments:

Post a Comment