Today, Chalkbeat begins taking a closer look at recently released state test scores. The scores are one way to measure whether schools are doing a good job serving different groups of students, including students from low-income families. They are the focus of this story.
On the whole, Colorado students in grades 3 through 8 who come from low-income families scored lower on literacy and math tests taken this past spring than students from higher-income families did. That gap has persisted for years here and across the nation.
But the raw scores don’t tell the whole story. While they show whether or not students are at grade-level, they don’t show how much academic progress students made in the year up to taking the tests. Take, for instance, a fifth-grader who jumped from reading at a first-grade level to a fourth-grade level. The student made a lot of progress but still isn’t at grade level.
The state aims to measure such academic progress with something called a “growth percentile,” or growth score. State education officials have focused on growth scores as a better gauge of teaching and learning – rather than raw test scores, which tend to be correlated to family income level.
That’s why the state heavily weights growth scores when assigning quality ratings to districts and schools. That’s also why we’re featuring them in the searchable database at the bottom of this story.
Growth scores work like this: Each student’s raw test score is compared with the scores of students who performed similarly to them in previous years. The growth score is a percentile: A 99 means a student did better than 99 percent of students with similar test score histories.
The state also calculates growth scores for entire districts, schools, and groups of students by ranking from highest to lowest all of their growth scores and then finding the midway point among students, or the median.
In the charts below, you’ll see median growth scores for 15 metro area districts. For each district, you’ll see two scores: One for students from low-income families who receive free or reduced-price school lunch, and one for students who don’t.
To qualify for subsidized lunches in Colorado this school year, a family of four must earn less than $46,435 annually.
In most districts, students from low-income families made less academic progress than students from higher-income families. But Derek Briggs, a University of Colorado professor and testing expert, said that shouldn’t be a foregone conclusion.
“It should be just as possible for schools that are in disadvantaged communities to show growth as those in more advantaged communities,” he said.
The database below goes even deeper to show school-level median growth scores for students who receive free or reduced-price lunch, known as FRL, and students who do not. The comparison provides an indication of whether schools are serving students from different income levels equitably or not. Briggs cautioned that growth scores are just one piece of the puzzle when measuring school quality.
A note about the numbers themselves: The median growth score for the state is always about 50. The state considers growth scores between 35 and 65 “typical,” meaning the students who earned those scores made a typical amount of academic progress that year.
Scores higher than 65 suggest students are making above-average academic progress – evidence, perhaps, that a school’s curriculum is working for that group of students or their teachers are getting the training, support, and resources they need to be effective.
Scores lower than 35 suggest the opposite.
Look up your elementary or middle school in the database below. The database also allows you to pull up several schools at once and see their scores side-by-side.
If you see the “-” symbol, that means the score for that group of students at the school is not publicly available. The state obscures test results for small groups of students in what has been a controversial effort to protect student privacy.