Assessment changes on tap

Colorado hustles to roll out new testing plan

State testing is going to take up a lot less space on school calendars in 2016 than it did last spring.

The main 2016 “testing window,” recently announced by the Department of Education, is scheduled for April 11-29.

That calendar is very different from last spring’s, when state assessments of various kinds stretched across three periods totaling more than 12 weeks, starting in March and ending in May.

The testing window isn’t the only thing that’s changing because of a law passed by the 2015 legislature and changes announced shortly thereafter by the multi-state PARCC testing group.

Other key changes include elimination of state tests for high school seniors, different tests for ninth and 10th graders and a cutback is social studies testing.

But while CDE is moving quickly to set up the new system, lots of questions remain to be answered, and key elements need to be approved by the federal government. That means students, parents and teachers don’t yet have a full picture of what state testing will look like in 2016.

Here are the details of the changes in the CMAS system, how they could affect different kinds of tests and students and what questions remain to be answered.

Time on test

The testing window is the period of time during which a school has to start and complete testing; every day of the window isn’t necessarily taken up by tests.

Next spring schools are supposed to fit language arts, math, science, social studies and alternate tests within the April 11-29 window. (This year the first part of language arts and math tests were given in March, followed by social studies and science in April and the second part of language arts and math at the end of the school year.)

Test time chart
Source: Colorado Department of Education

That may seem like a lot of tests, but PARCC also is reducing the total time consumed for language arts and math tests and number of testing sessions, known as “units” in assessment lingo. (See this story for details on PARCC’s plans for 2016.)

School districts like the single, smaller window. Last spring’s separate windows required a lot of duplicate set-up time, testing directors say.

“We really appreciate the change,” said Norm Alerta, director of assessment and evaluation for the Cherry Creek Schools. “That will help with a lot of issues.”

And CDE is providing an escape hatch for districts that can’t fit all those tests into three weeks. Districts can use a window of up to six weeks for language arts and math tests if needed because of limitations on the number of laptops and other devices available for testing use.

“We expect the vast majority to use” the three-week window, said Joyce Zurkowski, CDE executive director of assessment.

High school testing

Expansion of state tests into the 11th and 12th grades during the 2013-14 and 2014-15 school years sparked much of the public outcry faced by legislators during the last session.

Laura Johnson, 17, works on a computer between classes at Florence High School / File photo
Laura Johnson, 17, works on a computer between classes at Florence High School / File photo

The lawmakers’ ultimate solution, House Bill 15-1323, significantly reduced high school testing while making few changes at other grade levels.

Seniors – Students are off the hook for state tests in their last year of high school, so there will be no science and social studies to boycott next fall.

Juniors – The new law eliminates the full language arts and math tests in 11th grade, but the school year won’t be test-free for these students. Officials at CDE haven’t decided when to give high school science and social studies tests. But they acknowledge that spring of junior year is the logical time to test, rather than in freshman or sophomore years. And all juniors will still have to take the ACT test or an equivalent.

Sophomores – It’s too early to say what 2016 testing will look like for 10th graders. The new law proposes eliminating language arts and math tests for 11th graders, replacing those with a shorter college-and-career readiness test like the Accuplacer. That test also has to be aligned to state academic standards. However, this is a change that must be approved by the federal government.

And even if Washington signs off, it isn’t known yet which readiness test will be used, nor which 11th grade college entrance exam will be given. The new law requires those contracts be put out to competitive bid, which hasn’t happened yet. CDE officials acknowledge that the likely competitors are the ACT organization and the College Board.

Freshmen – Students will continue to take PARCC language arts and math tests in the first year of high school, just as their predecessors have done for years. CDE will have to make changes in the math tests, given that they won’t be given later in high school, as was the case for the last two years.

Social studies tests

To be decided
  • Whether language arts and math tests in 9th grade only will meet federal requirements for high school testing
  • Which college and career readiness tests sophomores and juniors will take
  • When high school science and social studies tests will be given
  • Which schools will give social studies tests next year
  • Extent of flexibility in testing some ELL students
  • Definition of “grade level” for early literacy tests
  • Double testing of students in any pilot assessment program

During the legislative testing debates these tests looked they were headed for the chopping block. But a compromise detailed in another testing measure (Senate Bill 15-056) saved the exams by drastically reducing the number of students who have to take them every year.

Now the tests will be given in one third of state schools every year, to one grade each in elementary, middle and high school. Stay tuned for CDE to decide which schools will be on the 2016 list.

English language learners

The new law proposes to allow districts to give state tests in other languages to English language learners for up to five years, instead the three now allowed. ELL students enrolled in a school for fewer than 12 months wouldn’t have to take English language arts tests.

The legislation also proposes to not use the English language arts scores of ELL students who have been in the U.S. for fewer than two years as part of school and district accountability calculations.

Some of these changes have to be approved by the U.S. Department of Education.

Early learners

One non-controversial element of the testing law was the streamlining of how students are evaluated for school readiness and how the reading skills of K-3 students are determined under the READ Act. There were concerns about duplication of assessments and about the need to retest students who read a grade level.

If schools evaluate students within the first 60 days of school, the results of READ Act tests can be used for the literacy component of school readiness. The law also streamlines some of the readiness and READ Act record keeping.

Students who demonstrate reading proficiency on the first test don’t have to take additional tests during the year. However, CDE has to set cut scores to define grade-level reading, and those won’t be determined until late summer.

Mouse or pencil

Students taking tests

Last spring the state allowed districts to request paper tests for 3rd grade language arts and math and for math in other grades. Those got only limited use around the state.

The new law allows districts to request paper tests for any grade. The state has no role in deciding, so parents anxious to have paper tests will have to make their case to their districts by next fall.

But CDE is working with test providers on the logistics of having all tests available on paper. “It’s a pretty significant shift for science and social studies,” Zurkowski said,

At least one set of paper tests also must be available for READ Act assessments.

Opting out

The right of parents to pull children out of state tests and the impact of low test participation rates on school and district ratings were hot issues during the legislative session.

Another sign at rally
Sign at anti-testing rally

The Senate and House ultimately couldn’t agree on an opt-out bill, but HB 15-1323 contains compromise language that gives districts some guidance on how to handle the issue.

Here’s how CDE explains it: “Each district must adopt a written policy and procedure allowing a student’s parent to excuse a student from participating in one or more state assessments. If a parent excuses his/her student from participating in an assessment, the district must not impose negative consequences on students or parents, including prohibiting school attendance, imposing an unexcused absence, or prohibiting participation in extracurricular activities. At the same time, the district cannot impose an unreasonable burden or requirement on a student to discourage the student from taking an assessment or encourage the student’s parent to excuse his/her student from the assessment.”

Evaluation & accountability

The testing law was driven partly by concerns that low scores on the new tests and high opt-out rates might unfairly affect teacher evaluations and school and district ratings. (Federal law requires states to penalize districts where participation rates fall below 95 percent on two or more tests.)

Evaluation illustration
PHOTO: Tajuana Cheshier/ Chalkbeat TN

As on many other issues, the new law offered a compromise. How the U.S. Department of Education reacts to the changes remains to be seen.

Teachers – Results from last spring’s state tests cannot be used to calculate the student growth measures used in teachers’ 2014-15 school year evaluations. Instead, districts can only use growth data derived from local tests.

The state’s evaluation system requires that 50 percent of a teacher’s evaluation be based on “multiple measures” of student academic growth, including more than data derived from state test results. A 2014 law created a one-year timeout for that requirement, meaning individual districts could use 50 percent, 0 percent or any percentage in-between when doing evaluations for the school year that just ended. Districts took that flexibility to heart (see story).

Starting with the upcoming 2015-16 school year, 50 percent of teacher evaluations must be based on student academic growth measures. The new law provides an out in years when state test results aren’t promptly available. If growth data isn’t given to districts at least two weeks before the school year ends, districts don’t have to use that data for evaluations but can instead use it in evaluations for the following school year.

Schools and districts – The upcoming school year will be time-out for accreditation ratings. No new ratings will be announced this autumn, meaning schools and districts will retain the ratings they were assigned at the end of 2014.

Districts and schools that have been rated in the two lowest categories for five consecutive years are subject to state intervention. That clock has been stopped for one year.

There will be ratings issued again in the fall of 2016. They will go into effect, and the accountability clock will restart, on July 1, 2017.

See this Chalkbeat story for details about the district ratings issued last November, and use this database to find individual district ratings.

Pilot programs

Another compromise element of the testing law was creation of pilot programs through which districts and groups of districts could try out new ways of testing students and holding schools accountable.

The idea, if it gets off the ground, is that two programs will be chosen from the first group of pilots, and that one of those might eventually become the new state testing and accountability system.

This plan will require multiple levels of federal approval, the first of which may affect districts’ interest in trying the experiment. That question is whether students who participate in a pilot have to continue taking regular state tests as well.

Getting those questions answered

More info

CDE is working to get most of the open questions answered by late summer or early fall.

The department already has opened informal talks with the federal department, trying to feel out what Washington is thinking. “We first have been having conversations with them” and have sent a letter detailing the changes in Colorado law, said Associate Commissioner Jill Hawley. As to how DOE will react, “We can’t answer what that will look like,’ she said.

Hawley and Zurkowski said it could be anytime between early July and early August before Colorado gets “formal guidance” from DOE.

The federal Elementary and Secondary Education Act requires annual testing of every state’s students and an accountability system for districts and schools, among other requirements.

Colorado has some flexibility under a formal agreement called a waiver. If DOE doesn’t approve of changes in the state’s system, that could lead to revocation of the waiver, re-imposing the more onerous requirements of ESEA on the state.

States that don’t follow federal requirements are subject to possible loss of federal education funding, although that’s a multi-step process that could take time to play out.

Are Children Learning

More Memphis area students are graduating high school. But what does that mean?

PHOTO: Jacinthia Jones
The 2018 spring graduation for the Memphis Virtual School was held May 22 in the Hamilton High School auditorium.

The number of students graduating from high schools in Shelby County and across the state has been rising for the last 10 years, but recent allegations of widespread improper grade changes in Memphis last year called into question if graduation rates were marred.

The results of a deeper probe of seven schools with high numbers of grade changes on transcripts is expected this month. But Shelby County Schools officials said a number of strategies have contributed to the district’s growing number of graduates and they believe better monitoring of grade changes would protect the integrity of those numbers, including sudden jumps.

“It’s our goal to aggressively increase academic performance and graduation rates at a more rapid pace, and we’ve implemented a number of strategies to do so,” the district said in a statement. “Therefore, it would be imprudent to see jumps in graduation rates alone as an indicator of improper grading practices.”

Grade changes had an impact on how many students graduated at Trezevant High School, the first school implicated in the controversy. Fifty-three students over four years obtained a diploma without passing the necessary classes, an investigation found.

Leaving high school with a diploma greatly increases a student’s chances of finding a job with a living wage and avoiding jail. But Tennessee policymakers have been pushing for more education beyond high school since college graduates and those with job certifications through technical colleges and similar schools have an even better chance of higher incomes later in life.

School districts often tie student performance to their graduation rates, citing better academics as one factor in rising graduation rates. In addition, federal law requires states to report their districts’ rates every year to monitor if some groups of students are lagging behind their peers.

Marisa Cannata, who consults with districts through Vanderbilt University on how to improve high schools, said getting a high school diploma “doesn’t mean that they’re college-ready.” The only thing the number of students who graduated truly measures is “accumulating credits in a timely manner.”

“I think of them as only one indicator of how well a school is serving a student,” she told Chalkbeat. “True improvement is going to be reflective in multiple indicators.”

Nonetheless, the district’s rising graduation trends reflect a similar upward trajectory for state and national graduation rates. The rate is calculated by dividing the number of students who graduate after four years by the total number in a high school cohort.

Tennessee is ahead of the pack in figuring out how to get more students to stay in and complete high school, said Jennifer DePaoli, the lead author on a recent national report analyzing federal graduation rate data.

“Tennessee is a state that we would say has really proven itself when it comes to raising student graduation rates,” she told Chalkbeat, adding it “still has some room to grow.”

In 2013, Tennessee was applauded in a national graduation report for outpacing the national average in nearly every category, including students from low-income families and students with disabilities. But in DePaoli’s report released last week, Tennessee’s growth in graduating its students has slowed, and has the 8th highest percentage of black students who didn’t graduate on time. The state’s graduation rate for students from poor families still ranks among the highest in the nation, however.

Before 2013, most students in the former suburban district, commonly referred to as legacy Shelby County Schools, consistently exceeded the state and national average with as many as 96 percent of students graduating on time. The number of students graduating from Memphis City Schools, which dissolved in 2013 after city school board members voted to consolidate with the county district, lagged behind the national and state average, hovering between 62 and 72 percent.

Legacy Shelby County Schools and Memphis City Schools graduation rate compared to U.S. (2008-2012)

Source: Tennessee Department of Education; Graphic by Sam Park

Since then, more students have graduated from high school. After the merger in 2013, the county split again into seven school systems.

One of Shelby County Schools’ goals is to have 90 percent of students graduating on time by 2025. The district, which is the largest in Tennessee, now sits at 79.6 percent for the class of 2017. Official numbers for the class of 2018 are expected to be released this fall.

Shelby County Schools, municipal districts, and the Achievement School District compared to U.S. (2013-2017)

Source: Tennessee Department of Education; Graphic by Sam Park

In the middle of all that, Tennessee raised the bar for students to graduate. The state had been stung in 2007 by a national report saying the existing state standards were weak and misled parents about how their students ranked against their peers nationwide. So, Tennessee started phasing in new graduation requirements in 2009 that increased the number of credits needed to graduate and introduced the current end-of-course exams.

Also, the state changed how schools and teachers are evaluated. In 2009, Memphis City Schools got a $90 million grant from the Bill & Melinda Gates Foundation to overhaul how the district recruits, trains, and evaluates its teacher workforce.

In 2010, the Tennessee Department of Education got a $500 million federal grant to recreate how it measures school success and partially tie teacher evaluation scores to student test results.

The state-run Achievement School District was born from that grant and started taking over low-performing schools in 2012. (The district didn’t have graduating seniors at high schools until 2014.)

In recent years, Shelby County Schools began to use data to help target students who might be at risk of dropping out. That kind of early warning system is part of a growing national effort to use mounds of student data to remove barriers to graduating, such as getting help with schoolwork, or pointing families to community resources to reduce absences early in a student’s high school career.

The district has also added reading specialists for ninth grade students who are behind and night and online classes for high school students so they wouldn’t have to wait until summer to retake failed courses. And before a student fails a class, district leaders have increased the number of offerings during the semester for a student to recover their grade.

In Memphis-area schools, 11 of the 48 in the region have fewer students graduating now than they did in 2008. Four of them dropped more than 5 percentage points:

  • Wooddale High School
  • Raleigh Egypt High School
  • Bolton High School
  • Ridgeway High School

Though there are 13 schools that have seen significant growth in the number of students who have graduated since 2008, they haven’t kept up with the district’s average ACT score, a common indicator of a student’s readiness for college.

But graduation rates and the ACT don’t actually measure the same things, said DePaoli.

“A lot of people would like to argue if graduation rates go up, we should be seeing gains in ACT scores and things like that,” she said. “We would like to see those things track together, but I don’t think there’s enough alignment there.”

Still, she said, “if kids aren’t getting higher scores on the ACT but the graduation rate is increasing, there is something to be really fearful of.”

Five Memphis area schools have now exceeded the district average for students graduating. Here are the 13 with the most growth:

  • B. T. Washington High School*
  • Oakhaven High School*
  • Martin Luther King College Preparatory High School (formerly Frayser High School)**
  • Hamilton High School
  • Sheffield High School
  • Westwood High School
  • Kingsbury High School
  • Manassas High School
  • East High School*
  • Craigmont High School*
  • Fairley High School**
  • Mitchell High School
  • Whitehaven High School*

*Schools that now exceeds Shelby County Schools’ graduation rate
** Taken over by the Achievement School District in 2014

Below you can look at your high school’s graduation rates over the years.

What went down

‘There was no cyber attack,’ investigator says of Tennessee’s online testing shutdown

PHOTO: Manuel Breva Colmeiro/Getty Images

Questar’s unauthorized change of an online testing tool — not a possible cyber attack, as earlier reported by the company — was responsible for shutting down Tennessee’s computerized exams on their second day this spring, the state’s chief investigator reported Wednesday.

An independent probe determined that “there was no cyber attack,” nor was any student data compromised, when thousands of students could not log onto the online exam known as TNReady on April 17.

Instead, investigators said, Questar was mostly responsible for this year’s testing miscues. The main culprit was a combination of “bugs in the software” and the slowness of a computerized tool designed to let students turn text into speech if they need audible instructions.

Comptroller Justin P. Wilson reviewed early findings of his office’s internal review and the external investigation by a company hired by the Education Department during a legislative hearing in Nashville.

Education Commissioner Candice McQueen also told lawmakers that Tennessee is docking Questar about $2.5 million this year out of its $30 million contract because of the online problems that plagued many students and schools during the three-week testing window.

Payments being withheld are punitive, as well as to cover the state’s costs to address the problems, she said, adding that other discounts could follow.

Last week, McQueen announced that the state plans to launch a new search this fall for one or more testing companies to take over TNReady beginning in the 2019-20 school year. She said a track record of successful online testing is a must.

The text-to-speech tool worked fine last fall when a smaller number of high school students tested online. But the state said Questar made a “significant and unauthorized change” to that feature before the launch of spring testing that affects the vast majority of Tennessee students.  

“We now know this decision led to the severity of other issues we experienced during online testing,” the Education Department said in a statement.

House Speaker Beth Harwell and Rep. Jeremy Faison asked the comptroller to review the state’s contract with Questar, particularly related to reports of a possible cyber attack. Wilson’s office also looked into other technical snafus that disrupted student testing for days, prompting the legislature to pass emergency laws that make this year’s scores inconsequential.

“We believe that the student testing issues occurred primarily because of how Questar set the student assessment system up to work,” said Brent Rumbley, the comptroller’s information systems audit manager.

Education Commissioner Candice McQueen testifies during Wednesday’s hearing, where specialists in the state comptroller’s office also testified.

On the second day of exams, Rumbley said, those issues manifested themselves in a suspiciously high volume of internet traffic to the testing platform.

“That’s what led the Department of Education and Questar to believe that there may have been a cyber attack,” he told lawmakers. “This traffic eventually shut the system down.”

Even though Questar upgraded the processing capability of its equipment in response, students and educators continued to report problems logging in, staying online, and submitting tests until Questar turned off the text-to-speech tool beginning May 1.

The comptroller’s office also found that Questar was ill-prepared to handle the fallout from the technical glitches. For instance, the company struggled to manually recover the high number of tests that students couldn’t submit online. And school personnel calling the customer service line experienced wait times as long as 60 minutes, prompting many to just hang up.

New details emerged Wednesday about other testing problems, too.

On April 25, a Questar employee “inadvertently overrode” custom rosters statewide that allowed schools to match students with available testing devices. “As a result, teachers and test coordinators had to scramble to get students the tests they should take,” Rumbley said.

The next day, more problems erupted when an internet cable was severed by a dump truck in a traffic accident in Hawkins County.

“According to the vendor that manages the fiber optic line, 21 districts were without internet from approximately two to four hours,” said Rumbley, adding that neither Questar nor the department could have prevented the outage that day.

Lawmakers will get an expanded look at the Education Department and its testing program in November when Wilson’s office presents the results of a year-long performance audit, along with findings from a massive survey of Tennessee educators about TNReady.

The two-hour hearing gave lawmakers a platform to take jabs at McQueen and her department for their handling of testing.

Rep. Bo Mitchell admonished the Education Department for tweeting on the second day of testing that Questar “may have experienced a deliberate attack” that morning.

“This gets into the public trust and throwing out information to the public from the Department of Education that the failure was a hack … Whose decision was that to put that out into the public domain without any proof?” asked Mitchell, a Democrat from Nashville.

McQueen clarified that the department never used the word “hack,” but reported that the testing system was experiencing a “pattern of data that was consistent with a cyber attack.” The description was based on what was known as the time, she said.

Sen. Janice Bowling, a Republican from Tullahoma, said Questar’s $2.5 million penalty “seems like a smack on the wrist” given the disruption caused by the company’s mistakes.

McQueen responded that the state is withholding almost $11 million invoiced by Questar for online testing as it continues negotiations. She added that the state’s biggest testing expenses stem from printing and transit costs for paper materials used by about half of its students this year. The state is transitioning to computerized testing and has decided to slow the switch for a second time in the wake of this year’s challenges.

Justin P. Wilson

Questar officials told Chalkbeat last week that the company plans to pursue the state’s new contract next year, but Rep. Craig Fitzhugh told McQueen that he doesn’t want the Minnesota-based company involved after it completes its current contract.

“I don’t think we can let Questar get in the ballgame again,” said the Ripley Democrat.

The proposal will be competitively bid, said Wilson, adding that Questar’s past performance will be taken into account.

For more on how Tennessee got here, read why state lawmakers share blame, too, for TNReady testing headaches.