Common Core Testing, Up Close and Personal
Usually, the best way to learn about a test is to just take it yourself.
Or at least that was the thinking at the recent Education Writers Association National Seminar session, “Testing, Testing: Trying Out New Assessments.” Journalists were greeted by a thick packet of test questions created for the two national assessment consortia that put together exams aligned to the Common Core State Standards — the Partnership for Assessment of Readiness for College and Careers (PARCC) and Smarter Balanced Assessment Consortium.
The seminar was led by Andrew Latham, the director of assessment and standards for WestEd, a nonprofit research and development organization that aims to improve education. Latham said the goal of the session was to better acquaint reporters with how these tests would be different from the more traditional standardized tests.
Latham said WestEd has had contracts in the past with both Smarter Balanced and PARCC to develop and manage the development of test questions.
“I’m not an objective observer,” Latham said. “But I’m going to try to give you, as best I can, an agnostic view.”
Asking New Questions
Part of the reason test questions have had to change, Latham said, is because standards are changing. The Common Core has been touted by some educators and policymakers as more rigorous and challenging that many states’ prior standards. Latham said colleges and universities said students were not meeting their own requirements for literacy and preparation, so something needed to change.
Now, tests have to measure more difficult standards that often ask students to use higher-ordering thinking skills rather than rely on memorization or simple multiple-choice questions.
Latham pulled up an example question on a screen in front of the room: The picture showed a number line from zero to five, with fractions of one-half, three-halves and six-halves sitting above it. “Drag each fraction to the correct location on the number line,” the directions read.
“So what would you do?” Latham asked the audience. “What do you notice about this question?”
He gave some time to sort through the question, analyze and answer it. Slowly, audience members began raising their hands.
“You’re not given specific options to choose from,” one person volunteered. “You have to create the option.”
“You need to drag and drop things,” another offered. “That’s not something a kid would do on a paper and pencil test.”
After a few more examples, Latham got to the point. What were these questions trying to measure? Why were they given in non-traditional formats?
For the number line problem, it’s considered higher-level for a few reasons. One, it’s not just asking kids to identify something. Two, it’s assuming in the question itself that students can understand what it means when two fractions have the same denominator or when one fraction is equivalent to a whole number.
“One of the things about Common Core, especially in math, is patterns,” Latham said. “I think this requires a deeper knowledge, a richer knowledge of the connectedness than if you just asked someone, ‘Where is four-fifths on the number line?’”
The second question was more complicated, but it seemed simpler right off the bat: The student was only asked to choose true statements from five options.
But each option was an equation, such as “8 x 9 = 81” or “54/9 = 24/6.” Again, Latham asked the audience to analyze the problem.
“It’s not multiple choice again,” one audience member suggested. “It by and large eliminates guessing.”
This question, Latham explained, measured both multiplication and division and expected third-graders to know that an equals sign between equations means both sides must lead to the same number.
“It’s more rigorous than you’d expect,” Latham said.
More Than Math
Sometimes, the language of the questions was confusing to the audience, especially when the directions were more detailed and relied on story-type problems rather than simple calculations. It was easy to see how kids or teachers might also be confused. But Latham said that’s why the test creation process has so many steps to root out bad questions.
If problematic questions do make it through to the actual test, they should be caught during grading. This is most common for long-answer questions, known as “performance tasks,” where students are writing or showing their mathematics work. With that type of answer, students can often earn partial credit because a grader can ferret out their line of thinking.
As new standards have been rolled out in many states across the country, it’s easy to see where some teachers have gotten frustrated by the new material and how to use it to make sure kids are prepared. Audience members asked what teachers can do to get to that point faster.
Latham said lots of classroom practice helps, especially when the question formats can be woven in to include or apply to lessons teachers are already trying to teach. That way, it’s not one more thing on their plates, he said.
“Assessment, ultimately, when it’s serving instruction is nirvana,” Latham said. “[Students] also learn by doing the performance tasks and the teachers get an idea of how to teach them.”