Why the Smarter Balanced tests are so different, and maybe better
Mike Szymanski | August 21, 2015
In a memo regarding the Smarter Balanced Assessments, LA Unified officials explained to Superintendent Ramon Cortines and members of the school board how the new test is so different from its predecessor and why scores may appear lower than in the past.
Cynthia Lim, executive director of the district’s Office of Data and Accountability, said she knows parents will make comparisons between the previous California Standards Tests and the new tests. That would be unfair, she said, describing it as worse than an apples-and-oranges comparison.
“Students are being measured in different ways than in the past tests, and there were five different performance bands; now there are four,” Lim told the LA School Report. The five “Far Below Basic,” “Below Basic,” “Basic,” “Proficient” and “Advanced” — are now “Has Not Met the Standard,” “Nearly Met the Standard,” “Met the Standard” and “Exceeded the Standard.”
Lim said she has seen preliminary scores for LAUSD — about 96 percent of the scores in grades 3 through 8 and 84 percent of the test scores in 11th grade — and they are not great. She declined to provide details but admitted that the scores would be “lower than what we’ve seen in the past in terms of what we would say is proficiency.”
The way the students are taking the tests could hurt the initial scores, too. All the tests are conducted on tablets, with no paper and pencil, no bubbles to fill in, no multiple guess. And, no student gets the same questions.
“In the past, every student had the same set of questions in a one-answer format,” Lim said. “So you got a question and possible range of answers and the student had to eliminate answers. On this new assessment, there are graphics, and the types of answers aren’t given to the students; they have to take it a step beyond, and show how they came up with the computation.”
The skills include writing on the tablet, typing in answers and sometimes listening on headphones before answering questions.
It’s also the first time the district tried a “computer adaptative” test. If the student gets questions correct, the questions get tougher. If they miss them, the questions get easier. “The questions are tailored to your ability, and we get more precise information about achievements and what the students need to work on,” Lim said.
The overall impact of the new metrics, Lim said, makes it less likely that instructors will teach-to-the-test — a common complaint of teachers — because of how the questions are asked, forcing students to concentrate on demonstrating a better understanding of what they are learning rather than rote recitation. That’s the essential change ushered in by the use of the Common Core State Standards as an instruction guideline.
The new tests not only help lower-performing students but help better identify highly-gifted students, as well. If more than two-thirds of the questions are answered correctly, the questions automatically go into a higher grade level. “That will give a better idea of the potential of the student, and the teacher can adjust accordingly,” Lim said.
Another reason the scores can’t be compared with the past regimen is because all students 2 through 11th grades were tested, while now only children in grades 3 through 8 and 11th graders are tested.
Schools went through a test-run two years ago and resolved problems with tablets not working properly. But all schools took the test in the last 12 weeks of this past school year, and those testing scores will now provide a baseline for the future.
The reports of the test scores sent to parents will have detailed explanations of where a child needs improvement, with more specifics than ever provided in the past. The inter-office memo showed examples, with a fictional student, Emily, as Below Standard for “Demonstrating ability to support mathematical conclusions” and At-or-Near Standard for “Demonstrating understanding of literary and non-fiction texts.”
“Teachers can use this information to adjust their teaching because the scores are broken into these sub-categories,” Lim said.
Teachers and principals can also ask students to take interim tests.
Meanwhile, the sample letters urge parents not to compare these test scores to past ones. An example reads: “These results are one measure of Emily’s academic performance and provide limited information. Like any important measure of your child’s performance, they should be viewed with other available information—such as classroom tests, assignments, and grades—and they may be used to help guide a conversation with Emily’s teacher.”