Delivering American University's news and views since 1925. | Sunday, December 16, 2018

Committee considers college testing

Tests would measure how well universities are teaching

As statistics show disturbing trends in college completion rates and poor performance by graduates on basic college-level skills tests, the U.S. Department of Education is looking for a way to measure how well universities are teaching their students.

A federal committee commissioned by Secretary of Education Margaret Spellings is studying the use of standardized testing in colleges and universities to evaluate and compare how much students are learning at schools across the country.

Late last year, a federal study that measured basic reading skills classified only 31 percent of college graduates as "proficient," The Eagle previously reported.

"That we have such problems is clear," said Kati Haycock, director of the Education Trust and a member of the department's commission to study higher education. "The way to better results is less clear."

According to a memo from the commission's chairman, Charles Miller, there is currently "gathering momentum" for the use of standardized testing to measure what students learn and what types of skills they acquire in college.

Miller's memo points to several recently developed tests, including the Collegiate Learning Assessment, which some schools have administered to gauge their own performance, and suggests the model could be adopted by more schools.

The Collegiate Learning Assessment, which was designed by the Council for Aid to Education, was implemented in 2003 to compare learning at the 15 different schools that make up the University of Texas system. Miller, who was the chair of the system's board of regents at the time, first introduced the idea of a "system-wide accountability framework" for the Texas schools in 1999, according to Geri Malandra, UT assistant vice chancellor for accountability and institutional planning. Malandra spoke before Congress in December 2005 about the need for such a test in colleges and universities.

Although the assessment proved successful for evaluations within the UT system, Haig Mardirosian, AU's acting Dean of Academic Affairs said in an e-mail that using a similar test to compare student learning among colleges nationwide would be ineffective, although the tests can serve some purpose. Unlike the SAT students take before entering college, which is designed to measure aptitude, standardized testing is not the best method to measure learning, he wrote.

"The place where standardized tests have some value, for example the GRE or the GMAT, is as an indicator of very specific skills or knowledge sets that will be applied in specific ways in graduate school," he said. "That is a measure separate and distinct from how a student has learned to think or analyze or create ideas or synthesize thought," things which can't be easily tested.

The distinct and individual nature of different schools can also create a hurdle for testing.

"Authentic learning ... is a highly personalized interaction of a student, a teacher, and a set of institutional values and expectations," Mardirosian said. "Faculty members here represent a set of particular outlooks and beliefs, a set of goals that are then coupled with what AU stands for. This core set of values, by definition, cannot be the same at another place."

This type of variation is difficult to measure in a comparable way across colleges through testing alone, according to Mardirosian.

"There are lots of reasons and ways to compare various learning institutions, but standardized test results are probably not the reliable means of doing that."

David Sadker, an education professor at AU said standardized testing couldn't appropriately measure a group of schools that each has their own unique strengths. Creating a test that measures traditional knowledge would not account for other important skills like creativity or insight, Sadker said. It would be up to the company making the test to identify specifically what students should be learning by regulating what will be tested.

This idea of what Sadker calls a shrinking curriculum- tailoring classroom curriculum to match certain test objectives- is also troubling to educators.

"If standardized tests were to be the arbiter of what students learn, then we'd have to significantly change our teaching mission to 'teach to the test,' if we cared about our students performance on such tests," Mardirosian said.

High test scores would become a new method for ranking schools, and could be a crucial part to standing out as a top university, much like the U.S. News and World Report college rankings, Sadker said.

"The way to be known as a great university would be to have great test scores," he said, although high test scores only mean students are prepared for what is being tested, not necessarily true knowledge or learning.

There is no serious thought being given to implementing a testing system at AU, according to Mardirosian, and there are currently no plans to test the system nationwide.

"To my knowledge, nobody--including the Commission--has yet proposed a single national test for all of higher education," Haycock wrote in an e-mail. "And frankly, I very much doubt that will happen."

The best way to university improvement may be through tests, but those assessments may be best administered through the university, she said.

"My own sense is that change occurs fastest when institutions step up and take responsibility for working on improving student outcomes," she said. "They're not the only players, of course; government policy matters. Family supports can matter. What students do, of course, matters a lot. But what institutions do matters, too."

The direct impact of the tests on students will depend on how the tests are administered and how seriously students take them.

The developers of the Collegiate Learning Assessment offer suggestions for how to entice students to take the exam, including free movie passes, collegiate T-shirts and pens or money.

But Peter Ewell, vice president of the National Center for Higher Education Management Systems, said the only way to ensure students give it their best shot is to incorporate the test into the curriculum.

"That is the general way in which you can get students to do well in any kind of an examination, to make it part of their coursework, to make it count," he said.

Accountability is another major theme for the commission. The rising costs of higher education have students and their parents asking if the quality matches the price, according to a memo from the commission's chairman, Charles Miller.

"We need to assure that the American public understand through access to sufficient information, particularly in the area of student learning, what they are getting for their investment in a college education," he

said.

The commission will make recommendations to Education Secretary Margaret Spellings in August.

Eagle Contributing Writer Eddie Shimkus contributed to this article.


Never miss a story.

Get our weekly newsletter in your inbox.