Print

Hartford School's Test Scores

Cheating Or Success?

Pattee, Christine

September 03, 2013

I was a volunteer reading tutor at Betances Elementary School in 2011, and was trained in the research-based reading instruction methods used in its Early Reading Lab, so it's no surprise to me that their third-grade reading scores are topping the charts.

Nevertheless, the state Department of Education is right to investigate the dramatic climb in scores at Betances. There are two possibilities for the increase in Betances' scores. Either the teachers told the kids the answers on the day of testing, as has been documented in other schools, or else Betances' literacy staff — principal, program director, classroom teachers, reading tutors, paraprofessionals — were teaching reading the right way, which means the test results are an accurate reflection of the students' successful learning.

There's no magic to those methods, which start with systematic attention to sound-letter connections (phonemic awareness), and include a solid grounding in phonics. Implementation of these methods, however, is expensive — new textbooks and instructional materials, plus requiring a major turnaround on the part of administrators and policy-makers from previous decades of whole language/balanced literacy instruction.

When research-based reading instruction was developed and tested in the 1970s and 1980s by Louisa Moats and others, multiple classrooms of students in schools at varying economic levels were taught reading using either the existing whole language approach or the experimental phonemic awareness/phonics approach. Scores in the experimental groups "were significantly higher statistically" in the same way that scores at Betances jumped.

There was a similar increase in reading scores at Breakthrough Charter [now Magnet] School during the years 2001-07 when I tutored reading there. Breakthrough administration and staff are dedicated to literacy improvement. When I started in 2001, Breakthrough reading scores were in the middle of Hartford schools' averages; by 2007, they were near the top.

As any statistician (I'm a retired public health statistician) will tell you, an average goes up whether you raise the scores at the bottom or the top. In Breakthrough's case, large increases in the scores of a small group of students at the bottom of the chart, coupled with the school's excellent literacy program for those who could read, was a winning combination.

At Betances, I've observed literacy instruction from preschool on: whole-class instruction in phonemic awareness, phonics, vocabulary, fluency and comprehension; small group instruction for those who didn't get it the first time in class, and individual tutoring for the very few who are severely reading disabled. This means that virtually every student will score at least a little bit better on standardized tests, and many will score a lot better, leading to a substantial increases in scores.

It is imperative that the state Department of Education investigate this situation quickly and accurately. It's appropriate to flag unusually large increases in year-to-year scores for further review, but that review needs to be based on hard evidence. For instance, erasures were found on Betances test papers, but kids take those tests with a No. 2 pencil with an eraser on top, so erasures will be found somewhere on the papers.

A legal firm was hired to determine whether there is tangible evidence of wrongdoing, such as incriminating emails or teachers instructing a child to change an answer, but there is a more direct way to investigate suspect score changes. Because of the individualized testing used at Betances to sort students into instructional groups, it should also be possible to match score changes in these ongoing tests with scores that seem outside the norm of individual students on the state tests. Parallel improvement in the scores is solid evidence that the student learned what he or she was taught, meaning the state scores are honest and true.

I haven't had contact with Betances School since 2011, so I have no personal knowledge of the current situation, and I never participated in its testing program. Reading programs that work are few and far between, however, and it would be a tragedy not only for Betances, but for the whole Hartford reading program, if test scores that are actually a reflection of what students have learned — to read, write and spell — are tarred with the brush of a cheating accusation.