How Georgia calculates its “growth” measure
School and district growth scores are based on what are called “student growth percentiles.” Essentially, student growth percentiles are calculated for each student in each tested subject by comparing his or her standardized test scores at the end of one year to the performance of other students who had similar test scores to him or her in the past.
Schools and districts are given a “median growth percentile” that offers a sense of how much a typical student at the school grew in a year compared to his or her academic peers.
Check your school
Find out the "growth" result for students in your school on our searchable database, on myajc.com
Students at some of metro Atlanta’s top schools may have few problems passing state tests, but they’re not learning as much in some subjects as their peers at other schools, new data released Thursday shows.
The Gwinnett School of Mathematics, Science and Technology, for instance, typically tops Georgia’s performance charts, but under the new measure unveiled by the Georgia Department of Education the high school is a low performer in one key area of math. Cobb County’s Walton High School is another highly-ranked school that isn’t tops on the new list, which reveals performance using what the state is calling student “growth.”
The data is designed to show how much students at each school learn in a year, regardless of whether they passed state tests. The measure is based on test scores, but it’s part of a shift from grading schools almost entirely on pass/fail rates, which favor schools with low poverty.
For the first time, anyone can go online and check out a school's growth performance by grade level, subject and demographic factors like race and poverty. Schools can be compared against each other, though officials caution this data shouldn't be the only measure parents consider in choosing a school.
Most of the schools in DeKalb County have lagged under the pass/fail measure, but growth is giving administrators there reason to cheer.
“If you look at economically disadvantaged students, particularly at the high school level, we are pleased with what we are seeing,” said Trenton Arnold, a regional superintendent who used to oversee testing in the district, where three of four students live in poverty.
But, in general in Georgia, schools with more students from low-income families tended to have lower growth results than schools with a smaller proportion of low-income students.
Poor kids in DeKalb typically lagged their better-off peers in growth, too, but not by as much as in test pass rates. For example, the pass rate on the high school End of Course Test in coordinate algebra for economically disadvantaged students was 21 percent, about 25 percentage points behind wealthier students. But the median growth measure for poor students on the exam was 46, just 9 points behind wealthier students.
Meanwhile, some high-performing schools didn’t look so great under the new measure.
At Walton High School in Cobb County, for instance, 99 percent of students passed the American literature test, but the median growth percentile was just 39.
Diane Harris, who has a rising junior at the school, wondered what to make of the results.
“Is it really an indictment of the American literature program? What does it mean?” she said. “Walton High School, given this data, do they need to come up with a strategy? What is their corrective action plan?”
The director of accountability and research for Cobb, Ehsan Kattoula, said Cobb high schools did well overall with growth. Asked about results like the literature percentile at Walton, he said, “You might have a high-achieving student and still have small growth. … If a student gets a 99 on the previous test, they’re not going to have much growth.”
The Georgia Department of Education contends that even high-achieving students can grow under this new measure because it allows them to demonstrate growth in comparison to other high-achieving students.
Jonathan Patterson, the associate superintendent over testing in Gwinnett, said the growth measure is too new to know whether a math result at the district’s School of Mathematics, Science and Technology is cause for concern. Every student there passed the Mathematics II test, but the median growth percentile in the course, which covers geometry, algebra and statistics, was 40, eight points below the district median and nine below the state median. He noted, though, that in other subjects, such as biology, the school performed well in pass rates and growth.
Georgia is one of about 40 states using students’ academic growth as a factor in rating schools, though how growth is calculated and how much weight it’s given varies by state. And Georgia is one of about 20 states using students’ academic growth as a major factor in evaluating teachers.
Growth data is already part of the College and Career Ready Performance Index, Georgia’s system for rating schools and districts. Starting this school year, students’ academic growth will also count for half of teachers’ job evaluations, a cause for concern among some educators.
The old focus on pass rates alone, a mandate under the federal No Child Left Behind Act, focused attention on “the saddest” outcomes, said Melissa Fincher, the associate superintendent over testing at the Georgia Department of Education. That revealed a stark truth: Test scores were closely linked with poverty.
Schools that traditionally topped the rankings for pass rates may get some unwanted attention for middling growth, Fincher said. “That’s going to be very difficult conversations for them to have.”
This new measure is intended to give teachers and principals another way to look at performance. It also gives parents an idea how much their kids are learning.
“For some parents, they’re going to find the education their children are receiving is in line with their expectations. For others, they may find their child at the end of the day isn’t getting the education they want,” said Bradford Swann, state director for StudentsFirst Georgia, which has lobbied for the use of the growth measure. “It’s a tool in the battle of evaluating your child’s education.”
The data released Thursday is based on test scores from the 2012-13 school year, but officials say in the coming years they hope to provide more timely results.
Jason Miller, a principal in Lee County and past president of the Georgia Association of Elementary School Principals, said the growth data can be helpful by drawing a clearer picture of student performance.
“But it doesn’t tell us why a student is succeeding or not,” he said. Miller was critical of plans to base half of some teachers’ evaluations on growth.
Alan Long, a principal in Jefferson County and president of the Georgia Association of Secondary School Principals, said the new data could show that some students who may not pass tests are still growing academically.
“It’s important to be able to showcase your school and to say you are doing some good things,” he said.
About the Author
