Test Scores Show Room for Improvement

Darien’s standardized math test scores are good, but they could be better, say public school officials.

Every September, Darien public school students take out their number two pencils, and sit through a number of standardized tests. Grades three through eight take the Connecticut Mastery Test (CMT), and tenth graders take the Connecticut Academic Performance Test (CAPT). Whether or not the tests serve their purpose as accurate measures of academic aptitude is debatable. Nonetheless, yearly administration is required per the No Child Left Behind Act, and an analysis of the scores can often help teachers revise their curricular. So on Monday, it was the school officials who pulled out their number twos for a test of their own: making sense of the scores.

There are a number of ways to look at the data, said Assistant Superintendent for Elementary Education Judith Panolfo. The focus of Monday’s discussion was a review of the CMT and CAPT mathematics “content strands.” Simply put, a content strand is a type of math problem—fractions, for example. Low scores in a particular content strand serve as a red flag for teachers, who can tweak their instruction accordingly, in the hopes of improving scores.

While this year, Darien's scores were “adequate”—above the state average—Pandolfo and other school officials said they could be better. At every grade level, a number of content strands stand out as “problem areas.”

Among the elementary students, grades three through five, the problems were “Estimating Solutions to Problems” and “Approximating Measures.”

“Across the state, these are two areas that children will generally have a harder job in being successful,” said Pandolfo.

Darien’s scores were lower than the majority of schools in its District Reference Group or DRG (Darien, New Canaan, Westport, Wilton, Redding, Ridgefield, Weston, Easton); however, Pandolfo said a town-to-town analysis isn’t as beneficial as looking at the cohorts.

“Schools’ scores will fluctuate because groups of students are different. We want to know that the same group of students, when they are tested each year, continue to improve,” said Pandolfo, who added that the scores generally show cohort improvement.

Improving scores in estimation and approximation is a matter of regular integration, said Pandofino; currently there are only a limited number of lessons dedicated to the two content strands.

Board of Ed. member Susan Perticone questioned whether the low scores are predictable, given elementary school children’s stage in neural development.

“Estimation has been a problem for as long as I have been studying this test,” said Perticone. “Does the state ever consider that this may not be an appropriate strand for elementary student who aren’t good with abstract thought processes?”

Yes and no, said Pandolfo.

“Some struggle; others don’t. If we can increase students exposure and get them to perform a bit better, we’ll try to do that.”

Among the middle school students, Assistant Superintendent for Secondary Schools Stephen Falcone singled out “Computation with Whole Numbers and Decimals,” as a content strand with room for improvement. This type of problem requires recording answers in a grid, with boxes for whole numbers, decimals and fractions, said Falcone. Students don’t have a lot of familiarity with the format, he said; subsequently, Middlesex math teachers are looking for opportunities to incorporate the grid into class work.   

Geometry, particularly volume, proved a challenge for the high school students. Falcone noted that students had not completed the unit at the time of the test.

“These subjects are not covered in detail by late February,” said Falcone. “Some curriculum realignment is being done to introduce and reinforce some concepts of volume at an earlier time.”

Test questions at both the middle school and high school level are quite sophisticated, said Falcone. More often than not, reading and comprehending the question is more challenging than the math required. Proof to his point, Falcone posed the following CMT question to the board:

The Bushnell Park carousel in Hartford opens in early May and runs through October from 11 a.m. to 5 p.m. One cycle of the carousel consists of three stages: loading people, the actual ride and unloading people. It takes about eight minutes to complete one cycle. The actual ride on the carousel takes three and a half minutes. If the carousel rotates four times per minute, how many rotations could it make from 11 a.m. to 5 p.m.?

The board chuckled, unanimously.

Secretary Clara Sartori asked whether time constraints limit students’ ability think creatively and “play” with a problem like the carousel example

Falcone said that time may be an issue, but that for every student who failed to correctly answer the problem (or even attempt it), there was an “elegant example,” of a student with clear understanding.

Members looked at a number of PowerPoint slides covered with complex data and analysis: scores, statistics and percentages.

“Our deficiency is rising in understanding the scores,” said newly elected Chairman Kimberly Westcott.

In closing, Pandofino said that teachers' grappling with the data, though perhaps testing, was key to the students' ultimate improvement.

“By going through this with faculty and talking about the childrens’ performance, they’re getting a higher level of understanding of how the children are thinking mathematically,” she said, “and hopefully our scores will go up as we improve.”

Pencils down.