NHPS
Does your balloon fly high? Or does it bobble near the ground?
The answer may determine whether your school stays open in its current form.
Those two bubble shapes will signify the difference between a top-performing and a “failing” school when the district issues grades next week.
The school district plans to grade six to eight schools as pilots for its nascent school reform drive. Schools would be given individualized management plans — possibly including extra resources or different work rules — based on their performance. Officials plan to name the six to eight schools at a 4 p.m. press conference Monday in City Hall.
Melissa Bailey Photo
Schools chief Reggie Mayo (pictured) explained the new grading system at Monday’s school board meeting. He unveiled what the three-tiered grading system will look like, and how it will work. Click here to view his presentation.
While other districts grade schools with a simple A to F scale, New Haven has chosen a more complex method. It looks like a bunch of balloons on a chart. (A sample chart, with fictitious schools, is pictured at the top of the story.)
Each balloon corresponds to one school. If your school has a large balloon that sails to the top right corner of the chart, it would qualify as a top-performer. If your school’s balloon is smaller, and sinks toward the bottom left corner, it could be a “failing” school.
Three factors go into the balloon’s placement: student performance on standardized tests, student growth on those tests, and student “engagement.” The basic goal is to grade schools into three different tiers. (Click here to read more about that.)
The horizontal axis is for student growth, with the highest growth toward the right. The vertical axis is for student performance, with the highest performance toward the top. The size of the balloon corresponds to how engaged students are. A big balloon means kids are well-adjusted and don’t miss many days of school. Schools with poor attendance, and where students face socio-emotional problems often associated with poverty, would get smaller balloons.
After mapping the schools onto this chart, Mayo plans to place them into tiers. Schools in the top-right corner of the map would fall into Tier I. Schools in the bottom-right corner would fall into Tier III. The middle-ranging schools would fall in Tier II. There’s no formula for how the tiers would be separated; Mayo said he’ll lean toward grading a greater proportion schools in Tier III, acknowledging that more need improvement.
Tier I schools, the top-performers, would be given more autonomy. Middle-ground schools would be placed in Tier II. Tier III “improvement” schools would get extra resources. A select few failing schools, dubbed Tier III “turnaround” schools, would be closed and reopened under new work rules, possibly as charters. Mayo plans to pick six to eight schools, at least one in each of those four categories, for a test-drive to take effect this September. Starting in November, all schools will be graded annually.
For the initial batch of grades, Mayo is also reserving some wiggle-room to use “qualitative factors” to place the schools into tiers. That means when placing a school into a tier, he’ll take into account three subjective factors: the quality of the school leadership; how ready the school is to take on reforms; and what resources are available to support the reforms. He said those “subjective” factors are needed for the first pilot year, when schools will be asked to make major changes on a rushed timeline. He also said he hopes he’ll be able to make some subjective decisions in future tiering, too.
On Monday, only the six to eight pilot schools will be placed into a tier. The rest of the district’s 47 schools won’t get a grade, but they will get a sense of where they stand by looking at where they land on the chart.
How exactly will the grades be calculated?
Katya Levitan-Reiner, a data analyst and recent Yale School of Management graduate brought on board in December to aid the district’s reforms, gave the break-down of the factors going into a school’s grade.
Student performance. For elementary schools, this grade will be based on the past three years of Connecticut Mastery Tests. Here’s how it works: Take the percent of students who scored at least proficient on the CMT in a given subject in 2008-09. Average each school’s reading and math percent together. Do the same for years 2007-08 and 2006-07. Then combine the three years’ scores together, using a weighted average that places more emphasis on the most recent year. The same calculation will be done using the higher standard of “at goal” instead of “proficient.” The result will be a percentage from 1 to 100, signifying how many students who are “at goal” or “proficient” on the CMT tests.
For high schools, do the same calculation, using the Connecticut Academic Performance Test instead.
Student progress/growth: This year, only elementary schools will get a “growth” grade. The calculation starts on a student level. Here’s how it works: Take student Joe. Measure how well he improved on the 2008-09 CMT test, compared with other New Haven students who started at the same academic level the year before. Give him a growth percentile — e.g., he grew more than 66 percent of his peers in that cohort. Repeat for students Maria, Tyquan, Gerry — each student in the school. Then take the median of all of those students’ growth percentiles. That’s the Median Student Growth Percentile (MSGP) for 2008-09. Repeat the calculation for year 2007-08. Average the two years’ MSGPs, with more weight to the more recent year. Voila — you have an overall score, from 0 to 100, showing how much students improved at one school, compared to the rest of the district.
Mayor John DeStefano was quick Monday night to ask why the students’ growth is being measured compared only to students within the district, instead of statewide. Mayo said the district could look statewide in the future, by importing a lot of data from the state government, but decided not to do it this year.
High schools don’t have a growth metric yet.
College success. High schools will also be judged on another factor: how many students complete their first year of college within two years after graduating high school? The New Haven Public School system got these numbers from the National Student Clearinghouse. They’re not the best numbers: There’s only a 85 percent accuracy rate. But they plan to look at the past three years of graduates’ post-high school success as one factor in grading high schools.
Student engagement. For elementary/middle schools: take the school’s attendance rate over the past three years, weighted to place emphasis on the most recent year. Also look at with how well kids fared on the Teacher Child Rating Survey, where teachers grade their own students on questions like: does Sally have many friends? Does she cope well with failure? Average the attendance and the survey scores.
For high schools: combine rates of attendance and retention over the past three years, with more emphasis on the most recent year.
Big balloons correspond to high numbers on this scale.
At the end of all the calculations, the elementary schools are ready to be mapped.
Since high schools don’t have a growth metric, they won’t be mapped with a balloon on a chart this year, said Levitan-Reiner. However, they will be able to look at the numbers for how their school performed on three metrics.
At the end of Monday’s statistics lesson, one parent came forward with a lingering concern.
Dominic Maldonado (pictured), a grandfather of two boys at Wilbur Cross High School and Christopher Columbus Family Academy, said he worries about the “psychological impact” of grading the schools. What if his student finds out he’s a high-performer in a Tier III school? Maldonado said he’d be inclined to move his kid out of a school that was graded in the lowest category.
Board members acknowledged the negative impression a Tier III label may carry. However, they said being judged a “turnaround” school should not be seen as a curse. It will be an opportunity to get more resources, said Superintendent Mayo. The board agreed to give parents a better sense of how these grades will affect their students’ educations when they unveil the first grades next week.
How about putting the additional "rescue" resources into the troubled neighborhoods that the students come from, not just into the troubled schools?
School test score achievement, and later life achievement, is linked far more with social conditions, like poverty, crime, noise, stress, litter, unemployment, lack of food, lack of social capital, no health insurance, etc., than it is to anything about the schools themselves.
It's nice to have a bunch of balloon charts on test scores, but let's make sure they are meaningful.
Trying to improve achievement without addressing the underlying causes of most of the achievement (or lack thereof) is a recipe for failure.