A large number of Beach Court Elementary School students who scored proficient in fifth grade over a three-year period saw their scores drop out of the proficient category in sixth grade, an analysis conducted for Education News Colorado by I-News shows.
In 2009, for example, 76 percent of Beach Court fifth-graders scored proficient on state math tests. Just 29 percent of those same students scored proficient in math the following year when they entered sixth grade in a variety of middle schools.
By contrast, in Denver Public Schools overall, sixth-graders in 2009 scored 1 percentage point higher in math then they did fifth grade the year before – 47 to 46 percent.
In conducting the analysis, I-News studied student test score records obtained from the Colorado Department of Education. DPS has declined to provide any data or other information until after the state wraps up its investigation.
Earlier this week, Education News Colorado reported that Beach Court Principal Frank Roti has been placed on administrative leave while the state investigates testing anomalies at the school. Hallett Fundamental Academy is also being investigated. The I-News analysis of test scores did not find test score drops at Hallett similar to those at Beach Court.
Sources confirmed the district’s analysis of Colorado Student Assessment Program results included an examination of erasure marks on student answer sheets. Results showed the two schools far exceeded district averages in the number of wrong answers erased and replaced with correct responses.
The I-News analysis of Beach Court scores found that:
- Between half and three quarters of fifth-grade students in 2007 through 2009 saw their math scores drop at least one level when they left the school and tested in sixth grade.
- Between a third and almost half of fifth graders dropped a level in reading over three years of testing and a level in writing over two years of testing after leaving the school.
As a result, the percentage of fifth-grade students at Beach Court achieving proficiency in math, reading and writing dropped by a half or more between 2009 and 2010 when they were sixth graders in a different school
I-News looked at how fifth graders who took the tests at Beach Court in 2007 through 2009 fared on the CSAP tests the next year when they entered a new school in sixth grade. The analysis compared how the same students scored – unsatisfactory, partially proficient, proficient or advanced. Between one and three fifth graders each year did not have scores the next year as sixth graders.
The biggest declines took place between 2009 testing at Beach Court and 2010 tests in sixth grade. Thirty-seven of the 49 fifth graders, or 75 percent, fell at least one level in math. The biggest drop was from proficient to partially proficient. Twenty of the 21 fifth graders fell below proficiency when they took the math test in sixth grade. In addition, 13 of the 17 fifth graders who scored advanced on math at Beach Court fell to proficient or lower after they left the school.
For reading in 2009, 22 of 49 students dropped a level the next year and 25 of 49 dropped a level in writing. As with math, the biggest declines were from scoring proficient at Beach Court to scoring partially proficient in sixth grade.
The declines bucked the districtwide trends.
For all DPS schools, 46 percent of fifth graders scored proficient or better in 2009 in math, rising to 47 percent when they became sixth graders in 2010.
For all of DPS, the percent scoring proficient or advanced in reading rose from 48 percent in fifth grade in 2009 to 54 percent in sixth grade in 2010. In writing, the scores were the same in fifth grade in 2009 as they were in sixth grade for 2010 – 41 percent proficient or advanced.
About half of the fifth graders who took math tests at Beach Court in 2008 dropped a level the next year. For reading and writing, it was about 30 percent of fifth graders who lost ground the next year.
Just under half of the fifth graders dropped a level after taking reading and math tests in 2007 at Beach Court. A break down on proficiency levels was not available for writing. Only scale scores were used in the data base analyzed by I-News.
I-News also analyzed scores for Hallet Fundamental School, but the database used by I-News did not include scores to compare the 2010 and 2011 years that are the focus of the investigation. The analysis found that most scores stayed the same or rose between fifth and sixth grades.
Reflections of a former Beach Court teacher
Bernadette Lopez taught third grade at Beach Court from 2005 to 2008, when she left to join the teacher-led Math and Science Leadership Academy. Lopez has since left teaching to enroll in law school.
During her three years at the school, Lopez said CSAP tests were closely monitored. Teachers picked up their boxes of tests in the morning before testing and the boxes were immediately picked up after testing was over.
She also pointed out that no teacher had access to the tests for all of their students. That’s because of testing accommodations allowed under state testing rules. So students who qualified for extra time, for example, or other special circumstances would be under the supervision of a different proctor.
“I never had access to 100 percent of my students’ tests,” she said.
Lopez said she heard about the cheating investigation and “felt really bad for the teachers who have worked so hard … it tarnishes what we did.”
“When I was there, things were on the up and up,” she said. “I never saw anything that would be suspect.”
But she also noted, “I don’t know what happens after someone picks up the boxes” of tests – “you turn in your box … you never see it again.”
“It’s almost too bad the district made such a big deal out of the school – it felt like there was so much pressure to keep scores up,” she said.
Lopez said the second year she taught at Beach Court, 89 percent of her third-graders scored proficient or advanced in reading; the next year, the figure was 100 percent. “I didn’t even think it was possible for all of my students to do so well.”
Still, if someone were trying to cheat, why falsely create 100 percent proficiency, which could create suspicion, she asked.
Lopez and other teachers pointed out that Frank Roti, the Beach Court principal, did not always have the best relationships with teachers.
“If teachers were aware of cheating going on, it probably would have been reported,” she said. On the other hand, if teachers felt nothing would really happen to a principal, they may not have felt it would make a difference.
“You’re going to have different kids every year, your scores are going to fluctuate,” she said. “If there is something to it, the pressure got to somebody.”
By the numbers
[iframe https://docs.google.com/spreadsheet/pub?key=0Apwd4It_epL5dFhwYkdzZGRsR1RSM0lETWcxdHlMRkE&single=true&gid=0&output=html&widget=true 675px 350px]