End of an era

Colorado backing away from PARCC English and math tests, forging its own path

Sheridan School District sixth grader Monica Dinh takes part in a practice session last year (Photo By Craig F. Walker / The Denver Post)

Colorado will begin shifting away from standardized tests developed as part of a controversial multi-state effort and toward tests developed mostly by Colorado educators.

The move, one consequence of a contract announced Wednesday by the state education department, will end Colorado’s membership in PARCC, one of two multi-state testing collectives that were supposed to allow for easy comparison across states but have fallen short of that promise.

However, Colorado will likely keep using some PARCC questions in the math and English tests given to students in grades three through eight, said Joyce Zurkowski, the Colorado Department of Education’s executive director of assessment. Doing so would ensure the state could track student academic growth data and continue rating schools without pause.

“We’re not tossing everything out and starting from scratch,” Zurkowski said. But “we are expecting that Colorado educators will be much more involved throughout the development process — and we’re going to need more Colorado educators to be involved in the process.”

The break from PARCC grew out of a December directive from the State Board of Education that the education department find a vendor that could reduce testing time, provide results within 30 days and give the state authority over test questions and administration policies.

That effectively ruled out sticking with PARCC because Colorado is but one founding member of the group’s governing board and can’t dictate all terms of the test.

State education officials said the multi-year transition on the math and English tests will begin next spring and should cause little disruption to classrooms.

State officials announced Wednesday that the multinational testing and technology company Pearson had won a competitive bidding process to administer the math and English tests. Pearson administers the current PARCC tests along with the state’s science and social studies tests, which are given annually to students in one grade in elementary, middle and high school.

“Because Pearson has been already providing the testing services for CMAS (the state’s tests) for a number of years, the transition to the new contract should be seamless for educators and students,” Katy Anthes, Colorado’s education commissioner, said in a statement. “Educators and students are familiar with Pearson’s systems, so this will allow them to continue to concentrate on teaching and learning the Colorado Academic Standards, which is the content assessed by the tests.”

Results from next year’s tests will be comparable to past CMAS results because the standards, test questions and scoring are not dramatically changing, officials said.

Colorado has already changed math and English testing twice in the past decade, making comparing past results extremely difficult — if not impossible. Officials say it won’t be the case now because this is essentially a contract change. However, more significant test changes may need to be considered after the state’s academic standards revision process is completed in 2018.

Board member Steve Durham, a Colorado Springs Republican who led the effort to abandon PARCC, was critical of the decision to continue using Pearson as the state’s testing administrator.

He pressed the department to ensure Pearson could deliver on the board’s directive, especially limiting testing time to eight hours and delivering results quickly.

“It simply can’t run on for days or weeks as it did,” he said, referring to the amount of time some schools needed to complete the tests.

A 13-member panel comprised of Colorado school administrators and education department staff reviewed bids from Pearson and one other company, Questar.

Pearson, the world’s largest textbook and testing company, has been criticized for its extensive lobbying practices to win business from states and school districts.

Zurkowski said the panel choose Pearson because of its user friendliness, superior technology to provide interactive question online and data privacy protections.

“Knowing our current legislation, and our board’s priorities in that area, we couldn’t go backward,” Zurkowski said, referring to Pearson’s ability to protect student data.

Reaction of the state’s announcement was mixed.

Reilly Pharo Carter, executive director of the nonprofit education reform group Climb Higher Colorado, said the state’s decision was another step toward re-engaging families in the state’s testing system.

“We are seeing continued buy-in to our system of statewide assessments as a result of ongoing, thoughtful adjustments made by our state board, state Department of Education and legislature,” she said in a statement. “Most importantly, these changes simultaneously protect rigorous assessments aligned to the education standards and ensure continuity for our students, educators and systems of accountability.”

Angela Engel, a leader in the testing opt-out movement and founder of the nonprofit Uniting4Kids, said the change did little to alleviate her concerns about the role of standardized testing in schools.

“This new direction by the State Board of Education is simply another version of a bad idea,” she said in a statement. “Colorado has already changed test vendors, test names, and test styles – all with the same failed outcomes. Innovation and equity require an entirely different approach than administering a test.”

The state’s testing contract with Pearson can be renewed annually until 2024.

Under federal law, states must give English and math tests in grades three through eight and once in high school. They must also test students in science once in elementary, middle and high school.

Results are used to measure school quality. Schools and districts that repeatedly fall short are subject to state-ordered improvement plans — a process that has taken center stage this year at the state Board of Education. Results in some cases are also used to measure teacher performance.

The state had to make decisions about the future of its testing system; the state’s multiple contracts with Pearson and the PARCC organization began expiring this spring.

Colorado was one of the founding members of PARCC, which formed to develop tests measuring student’s mastery of Common Core academic standards in math and English.

PARCC became the focus of protest two years ago as part of a wave of anti-testing backlash. Large number of high school students, especially, opted out of PARCC tests, starting a long conversation that resulted in legislation that reduced the volume of state tests.

Colorado will become the latest in a long line of founding members of PARCC to leave the consortium. Five states and Washington, D.C., remain as governing members.

Because Colorado is no longer a member of the consortium, it will have no say in the development of the new questions that the group develops. State officials said they plan to only purchase test items from PARCC that Colorado teachers helped develop.

On Close watch

State’s lowest performing schools and districts taking hard look at this year’s test data

PHOTO: Denver Post file
Sixth-grade science teacher Monica Wisniewski works with Pija Williams Terralee, left, and Myth Cubbison at Kearney Middle School in Commerce City. Kearney is in Adams County School District 14.

Testing data for Colorado’s longest-struggling schools and districts show mixed trends.

Results released Thursday are from exams students took last spring, before the State Board of Education approved corrective action plans for the five districts and a dozen schools that had run out of time on Colorado’s accountability clock.

The Pueblo City 60 district saw a decline across many tests and grade levels while Westminster Public Schools showed improvements in 10 of 14 English and math tests.

Those districts, like other districts and schools facing state intervention this past school year, were already making changes before their state improvement plans were finalized. Much of that work is incorporated into the plans.

Thursday’s test data will be used toward a new state rating, one which these districts and schools must improve soon. The state plans gave most schools and districts until 2019 to earn a higher quality rating for face potential consequencs. But some, including the Adams County School District 14 and Adams City High School in Commerce City, must have a higher rating by 2018.

Find your school’s scores
Search for your school’s growth scores in Chalkbeat’s database here, or search for your school’s test results and participation rates in Chalkbeat’s database here.

The district of almost 7,500 students saw some improvements, but still is posting very low scores. For instance, 9.3 percent of fifth-grade students met the state’s learning goals in math, up from 7.9 percent last year. In that area, the district did better than the state, as fewer fifth graders did well on math tests statewide than last year.

At Adams City High School, growth scores, which represent how much students learned in a year compared to similar-performing students, decreased for both math and English. The school had an interim principal for much of the school year, which led to a student walkout in the spring.

Overall, Adams 14’s proficiency numbers are still lower than state averages.

Of these low-performing districts, the Pueblo City 60 district, which faced state action only for some of its schools but not as a district, was the only one that had decreases in growth scores for both English and math tests.

In English, the growth score was 43, down from 47 last year. That means students this year scored on average better than just 43 percent of Colorado students who had similar test scores last year.

The Pueblo district saw an increase in how many students met or exceeded expectations in eighth-grade English. One possible reason officials pointed to: innovation schools granted flexibility from some rules and state laws.

Dalton Sprouse, a district spokesman, said district officials are relatively pleased with the improvements they see in the data, especially when broken down by school.

“Given that there’s just two years of growth data, some of the fluctuation could be expected,” Sprouse said. “We see this as we’re maintaining the progress we made last year.”

Sprouse noted that two of the three schools that faced the state board earlier this year for low performance saw big increases in the number of students meeting math expectations.

“Some progress is starting to take place,” Sprouse said. “The assessment office is already working with principals to really dig into that data.”

Westminster Public Schools, another district that faced state action this year and is now on a three-year improvement plan, saw more improvements than the other districts on the state watchlist.

“We are pleased to see our focus on high expectations and personalized learning for all students is paying off,” Superintendent Pamela Swanson said in a statement.

The Westminster district, however, was also one of the only districts in the metro area where English language learners had worse growth scores than native English speakers in both math and English. Last year, there was no gap in growth on English tests.

Last year, about 40 percent of students in Westminster schools were English language learners.

In Adams 14 schools, where about 46 percent of students are English language learners, those students posted higher growth scores than native English speaking students.

Westminster did increase their overall rate of growth according to median growth scores, and reached above 50 for English language arts.

Aurora Public Schools, the only district at risk of state action next year, posted increases and also got one growth score above 50, which is critical to catch students up when they are behind grade level.

Here’s how districts that ran out of time on Colorado’s accountability clock — or districts that had schools that did — compared:

Shrinking gaps

Denver Public Schools posts record gains on latest state tests

PHOTO: Melanie Asmar
McAuliffe Manual students gather for a photo with Denver Public Schools officials at a press conference Thursday.

Denver students made more academic progress on state English and math tests last year than ever before, and the overall percentage of third- through ninth-graders who scored at grade level moved to within a few points of the statewide average, test results released Thursday show.

It’s a significant feat for the state’s largest school district, which ten years ago lagged far behind.

Notably, the diverse district’s academic growth was driven by low-income students, students of color, students with disabilities and English language learners. Students in those groups made progress at a faster rate than students not into those groups, shrinking the growth gaps between traditionally underserved students and their more privileged peers.

Superintendent Tom Boasberg called the results “wonderful.” He said that while the district’s gaps “are still large and concerning, it’s nice to see them moving in the right direction.”

Overall, more Denver Public Schools students met or exceeded state expectations on most tests in most grades. Among the biggest increases was the percent of third-graders at grade level in literacy. In 2015-16, 32 percent of DPS third-graders met that bar. In 2016-17, it jumped to 38 percent, a 6 percent increase. The statewide average was 40 percent.

Boasberg credited the district’s focus on early literacy, and its monetary investment in new curriculum and more training for early childhood teachers and paraprofessionals. A tax increase approved by voters in November includes $6.8 million to continue those efforts.

Find your school’s scores
Search for your school’s growth scores in Chalkbeat’s database here, or search for your school’s test results and participation rates in Chalkbeat’s database here.

“We’ve never had growth like that in third-grade reading,” Boasberg said.

Denver students also continued to outpace their peers across Colorado in academic growth. The state uses “median growth percentile” scores to gauge how much students learn each year.

A growth score higher than 50 means students are learning at a faster rate than peers who started the year at the same academic level as them. A growth score lower than 50 means students are learning at a slower rate than their academic peers.

Denver’s overall growth score in literacy last year was 57, up from 56 the year before. In math, the overall growth score was 53, up from 51.

“It all starts with our teachers and our school leaders,” Boasberg said of the improvements.

The district has expanded to nearly all schools an initiative that allows successful teachers to teach part-time and coach their colleagues part-time, and Boasberg said the latest scores are proof that helping teachers improve helps students, too.

Mixed results for reform efforts

Denver is nationally known for its education reform efforts, which include granting charter school-like autonomy to district-run schools, and replacing persistently low-performing schools with schools officials deem more likely to succeed.

The school board this past school year voted to close three long struggling elementary schools, including Amesse Elementary in far northeast Denver. Board members chose as a replacement a program proposed by leaders of nearby McGlone Academy. The district has held up McGlone as a rare example of a successful turnaround school.

But this year, McGlone’s scores faltered. On most tests, fewer students met expectations last year than the year before. Growth scores fell, too, to 41 in literacy and 37 in math.

Amesse posted higher growth scores: 58 in literacy and 49 in math.

Boasberg said he remains confident in McGlone’s leaders. McGlone principal Sara Gips Goodall said she’s excited by the growth at Amesse. She pointed to other measures of success at McGlone, including low student suspensions and high teacher retention.

“McGlone, over multiple years, has had very strong growth,” Boasberg said. “This year, their growth wasn’t as strong. Part of that was all of the time and effort that the school put into planning for and working with the community around the Amesse turnaround.”

He added that, “I think you have extraordinary teachers and leadership at McGlone who have an exceptional track record, and I’m confident they’ll have strong growth this year.”

Boasberg and other officials held a celebratory press conference Thursday at the Manual High School campus, which is also home to McAuliffe Manual Middle School, a replication of the successful McAuliffe International School. Both are innovation schools, which means they’re run by the district but enjoy flexibilities with scheduling, teacher hiring and firing, and more.

McAuliffe International has for years posted high test scores and had above-average growth. The school is not as diverse as the district as a whole — just 18 percent of students qualify for subsidized lunches, a measure of poverty, and 37 percent are students of color.

McAuliffe Manual opened last year with sixth grade in an effort to bring high-performing middle schools to northeast Denver, a neighborhood that historically lacked them. Nearly six in 10 students qualified for subsidized lunches, and seven in 10 were students of color.

While McAuliffe Manual trailed McAuliffe International in the percentage of students at grade-level, its growth scores were nearly as high: 72 in both literacy and math, compared to 75 in literacy and 74 in math at McAuliffe International.

There was more good news for three previously low-performing elementary schools — Goldrick, Harrington and Schmitt — in the midst of school turnaround. New principals spent the 2015-16 school year soliciting opinions and crafting plans to improve academic performance at the schools while other leaders handled day-to-day operations — a strategy known as “year zero.”

In 2016-17, the first year the new principals and their improvement plans were in place, growth scores at all three schools shot up by as much as 24 points.

Another turnaround school also showed remarkable progress. The University Prep Steele Street charter school, which replaced struggling Pioneer Charter School last year, boasted growth scores of 84 in literacy and 91 in math. The math growth was the highest in the state.

The test scores at four schools that are part of another DPS experiment, an “innovation zone” that gives the schools even more autonomy than regular innovation schools, were a mixed bag.

Two of the schools, Creativity Challenge Community and Denver Green School, posted increasingly strong scores on most tests and showed high academic growth.

But two other schools, Ashley Elementary and Cole Arts and Science Academy, saw low growth and slipping scores. The median growth percentile in math at Ashley was 32, well below the district average. At Cole, where just 5 percent of fifth-graders scored at grade-level, it was 17.

Boasberg said the scores at those two schools are concerning. But he said he appreciates what the innovation zone, called the Luminary Learning Network, is doing. District officials have talked about inviting other innovation schools to form similar zones.

“They have some very strong leadership at the zone,” Boasberg said, “and we recognize that for any one school, you are going to have some ups and downs.” He cautioned against reading too much into the scores of Ashley and Cole.

Jessica Roberts, executive director of the Luminary Learning Network, said it’s become clear that Ashley and Cole, which serve a more at-risk population, need a different type of support than the other two schools. Zone leaders are working to help them figure out how to use their increased autonomy — and freed-up funding — to boost student achievement, she said.

“We have confidence in these school leaders,” Roberts said, “and we will provide additional support in coaching hours and oversight over how their resources are used.”

Narrowing gaps

About two-thirds of Denver’s 92,000 students qualify for subsidized lunches, a proxy for poverty, and about 77 percent are non-white. More than a third are English language learners.

The district has in the past struggled to close wide gaps between how much students in those groups learn each year and how much students not in those groups learn.

White students, non-low-income students and non-English language learners have historically posted higher proficiency scores and higher growth scores, which continues to be the case. But their growth scores last year remained relatively flat.

Meanwhile, the growth scores for students of color, low-income students and English language learners increased by several points for every group in each subject.

In literacy, Latino students had a growth score of 54 and black students had a score of 53. White students had a score of 64, meaning the gaps were 10 points and 9 points, respectively. Those are smaller than in 2015-16, when the gap for both black and Latino students was 13 points.

The gaps in math last year were bigger than the gaps in literacy. Black and Latino students had a growth score of 50 in math, while white students had a score of 63, a 13-point gap. However, that gap also shrunk from the year before, when it was 16 points.

The smallest gap last year was between English language learners and native speakers in literacy. State statistics, which include “exited” English language learners who no longer need services in the count of English language learners, show no gap at all.

But DPS statistics, which break exited English language learners into their own category, show a 3-point gap between English language learners and non-English language learners.

The district has in recent years provided more training for educators who teach English language learners, worked harder to ensure all eligible students get those classes and made efforts to encourage bilingualism and biliteracy, Boasberg said.