This article first appeared in the St. Louis Beacon, April 9, 2012 - When she was working for the state of Missouri, trying to keep student testing results honest, Sherri Sampson came upon some pretty egregious examples of cheating:
- Teachers took test books home to change students’ answers.
- Teachers took pictures of the test on their cell phones and sent them to their colleagues.
- Teachers made copies of the test and sent them home as homework.
- Teachers watched students take the test and gave very obvious clues about which answers were right and which were wrong.
Now that she is executive director of assessment at the Riverview Gardens School District – which, she quickly emphasizes, was not involved in any of those cheating schemes – Sampson wants to make sure that the results of the MAP tests the district’s students take this month portray a true picture of the progress its students are making.
In unaccredited Riverview Gardens, MAP test results are particularly crucial. The district can’t get its accreditation back until students show marked, sustained improvement in what has been consistently lackluster academic achievement.
Even teachers’ jobs could be affected, if Missouri lawmakers approve proposals that tie tenure and employment to students' test scores.
But in the end, Sampson said in a recent interview, the basic issue of cheating that has been in the news a lot recently is really a question of integrity – not just the integrity of individual teachers but of the district and the whole process of education.
“Integrity is the most important thing,” she said. “To compromise your integrity is not worth it. You not only compromise your personal integrity, you compromise the integrity of the school.”
With test scores counting so heavily, there is a lot of pressure riding on the outcome. But, Sampson said, the teaching staff at Riverview Gardens, which was transformed when the state took over the district as of July 1, 2010, will be able to handle it.
“No one that we brought into the district thought this would be a bed of roses,” she said.
She dismisses the notion that cheating is widespread, though she admits that a few cases have attracted a lot of attention. In that view, she is joined by Missouri’s education commissioner, Chris Nicastro, and Michael Muenks, the state’s coordinator of assessment. He notes that because money is scarce, the Department of Elementary and Secondary Education depends to a large degree on schools' reporting of test irregularities.
“There are instances where folks I am sure are doing things they shouldn’t be doing,” he said. “But I think our districts for the most part are doing the very best they can, and we’re doing the very best we can.”
And he has another big reason to think that test results are an honest reflection of how Missouri students are doing.
“If we have widespread cheating going on,” Muenks says, “we would have much different looking scores than what we do.”
A brighter spotlight
Cheating on tests has always been on educators’ radar, but it has taken a more central spot in the past few years with revelations of large-scale scandals in several cities, most notably Atlanta.
There, the Journal-Constitution has conducted several investigations into cheating, including one last month that took a look at the situation nationwide. That story recounted allegations of suspicious test scores at the Patrick Henry school just north of downtown St. Louis. The newspaper said it had uncovered suspicious scores there dating back to 2007.
Recent stories in the Post-Dispatch have raised the question of how diligently Missouri education officials check into reports of cheating that are made by individual school districts.
And the Southeast Missourian newspaper reported Friday that some science test results from last spring in Jackson, Mo., were declared invalid after the school district reported that a fifth-grade teacher had sent home a study guide that mirrored questions on the MAP test. That teacher and another resigned in the wake of the incident, which was investigated by DESE based on the district’s report.
With more than 1.3 million MAP tests set to be given to more than 600,000 students across Missouri this month, such reports prompted Nicastro to issue a statement recently trying to reassure the public of the accuracy of the test results.
"School officials take this responsibility very seriously," she said, "and educators conduct the testing with great care. ... We are confident schools will continue to be vigilant in their administration of the MAP test, and we urge teachers to encourage students to their best on the tests.”
Over the past three years, DESE said that 50 testing irregularities were reported to the department in 2009, 69 in 2010 and 42 last year. When such reports come in, the district's superintendent involved is notified and asked to investigate. The department then reviewes the findings to see whether scores involved should be invalidated.
Last year, 181 student test scores were voided because of questions about their integrity.
In some cases, the department said, teachers have been disciplined and lost their jobs as a result; in others, schools provide additional training and oversight to make sure educators know the proper testing procedures.
Department officials noted that irregularities have been reported across the board and are not limited to any particular part of the state, to large or small districts or to schools in rural, urban or suburban areas.
Nicastro pointed out that for more than 20 years, DESE has had quality assurance procedures to make sure test results are accurate. But Muenks, the state coordinator of assessment, said that budget problems in recent years forced the state to shed 27 regional facilitators who helped ensure the tests were being conducted honestly.
To help restore funding to detect possible cheating, state Rep. Jay Barnes, R-Jefferson City, introduced legislation to give DESE the personnel and the money – estimated at $255,000 – to conduct the data forensics needed to detect cheating.
It would require statewide tests to be analyzed to detect possible patterns of cheating, then share any such findings with the superintendent and board members of affected districts at a public meeting.
Tight controls on tests
In Riverview Gardens, Sherri Sampson wants to make sure that no such reports from her district ever come to the attention of officials at DESE. She notes that she helped develop many of the safeguards used by the state, and she emphasizes that point to test coordinators in her district.
“This is what I did for the state,” she says. “My name is on that, so I’m going to make sure we do 100 percent of what it says. I put a lot more checks and balances in place than what is required by DESE.”
Those procedures begin when the MAP test materials are delivered to districts in late March, Sampson says. Principals and test coordinators have to sign for them and tell her in writing where they will be kept and who can get to them. Most of the time, she says, the booklets are locked in a room with limited access.
Each test booklet has a code number, and every test coordinator in the district has a list of codes that is matched with a list of students. When test day comes, each teacher has to sign off on the books received and match them to the students.
On test day, no substitute teachers are allowed to administer the exams, Sampson said. If classes are large enough, proctors are added to give the teacher an extra set of eyes to guarantee students’ work is truly their own.
Immediately after the test, the coding and matching begin again, and test administrators have to sign to verify that the results are accurate. Sampson is called, and she counts the returned tests again, accounting for all the materials. Then the tests are sent to CTB/McGraw Hill for scoring. Results come back in the summer.
What kind of results would prompt concerns? Sampson said that with accreditation on the line, Riverview Gardens would like to see students progress about 10 percent a year and would be very pleased with 15 percent. Anything above that, she said, would lead to serious investigation.
“Once you get into the 20-30-40 percent range,” Sampson said, “you definitely get red flags.”
When she was with the state, she recalled one district whose test scores raised a lot of flags. Once its tests were more closely monitored the next year, she said, the scores went back down.
Data, data everywhere
As Riverview Gardens works to regain accreditation, MAP scores are hardly the only numbers it uses to gauge how well students are doing. Each school has a data room where students have code numbers and brightly colored dots that show how well they have been progressing on the regular tests that the school gives them.
That way, Sampson notes, no one should be shocked by the MAP scores that come back in the summer.
“The data should never surprise you,” she says, “unless you’re talking about a kindergarten student who has never been tested before.
“I’m not hanging on tenterhooks waiting for the assessment to come back. For the most part, we know what the scores should be.”
At Gibson Elementary School, home of the Geckos, principal Germaine Stewart led a tour of the data room, where blue indicates advanced, green proficient, yellow basic and red below basic. On its most recent state evaluation, Riverview Gardens had far too little blue and green and far too much yellow and red; it did not meet any of the MAP requirements needed to get out from under state control.
“We really pride ourselves on being data-driven,” she said. “Every kid on this board is different, so it’s our job to make sure every kid gets what he or she needs. It’s an ongoing job. It never stops.”
Adds Sampson:
“It begins in kindergarten, so every student is responsible for the success of the school.”
Stewart said that some teachers had initially been wary of such an emphasis on the numbers, but they have been won over. Charts on the wall of a hallway show how strong the teachers think they are on the data-driven approach and whether they need to improve.
In Evon Simmons’ fourth-grade class, the students were working that day on how to read and interpret data – specifically, how Sally could use a sleep journal to determine how many hours she spends sleeping each week.
Using a worksheet that shows what time Sally went to bed each night and what time she woke up each morning, the students worked in small groups to come up with the answer. They drew clocks, they counted on their fingers, they used a variety of methods to come up with the answer to the questions of the day: How do you use data in your daily life? Why are data important?
Those were the lessons designed to help Simmons and other teachers achieve the curriculum challenge of the day, working with GLEs – educationese for grade level expectations, what all children should know depending on what grade they are in.
The expectations are posted by the door, to keep the teachers on track. To help keep the students on track, Simmons is working the room, helping individual students figure out Sally’s sleep schedule.
“I like to see your work,” she called out. “I like to see your thinking. Don’t erase anything on your paper.
“Justify your answer. If someone has a concern, and you don’t have the same answer, you have to justify your answer. Always justify.”
As test time approaches, any posters or other aids that students may use through the year to help master their lessons are either covered up or taken down, to avoid inadvertent hints. A final walkthrough is conducted to make sure nothing is missed.
It’s too soon to know whether the emphasis on data will help the district solve the problems that led to its losing accreditation. As Sampson notes, test scores are just one component of a far more complex situation in districts like Riverview Gardens.
But, Stewart adds, those who are working to help the district succeed realized that they had to do something new.
“We feel that if we did what we had always done in the past,” she said, “we would get what we always had gotten.”