Putting 'Value Added' Data to Good Use
When a statistician like William L. Sanders says you run the best school in the state, people take notice. Which partly explains why so many Tennessee districts have asked Principal Joel Giffin to explain his accomplishments.
Mr. Giffin presides over the only middle school in the 3,800-student Maryville district, about 20 miles southwest of Knoxville. His school's test scores rank well above the state average, and in 1996 the U.S. Department of Education gave the school its Blue Ribbon Award, including special recognition for technology programs. The system even boasts a former Tennessee governor, U.S. secretary of education, and presidential hopeful among its alumni: Lamar Alexander.
But what impresses Mr. Sanders is that the 1,070-student Maryville Middle School hasn't become complacent. Graphs of the gains made by its students show steady progress being made at each grade, regardless of students' previous performance. From its highest achievers to its most disadvantaged students, the school is leaving a positive mark on its children.
In fact, the University of Tennessee researcher has played a catalytic role in these gains. By showing the effect every class and teacher has had on each student, the data provided by Mr. Sanders' Tennessee Value-Added Assessment System have helped Maryville Middle School diagnose its strengths and weaknesses.
"What we used to do 10 years ago is we'd sit around the last day and say, 'What kind of year did we have?' " the 61-year-old Mr. Giffin says. "But that was kind of a shot in the dark. So when Dr. Sanders came with his system, it was like somebody answered from heaven."
An Added Focus
Four years ago, for example, Mr. Giffin looked at his school's TVAAS data in mathematics and saw a challenge. Although the top four-fifths of 6th graders were making gains well above the national norm, those at the bottom were making little progress. So the school planned an extra math period just for the lowest performers.
With data in hand, Mr. Giffin persuaded district officials to hire an additional teacher for the class, in which students were tutored and allowed to complete their homework. Now the 6th graders from the lowest fifth of performers are making gains two to three times the national-norm gains in math, Mr. Giffin says.
It was the TVAAS data, he adds, that showed the approach was working.
"Achievement levels alone tell you where the students are," he explains. "But the value-added data tells you what an individual school or an individual teacher has done for those students."
The data also help Maryville stay true to one of Mr. Giffin's core educational beliefs: Children come to school with different levels of ability.
"You can't take every kid and make him Tiger Woods," Mr. Giffin says, using golf as an analogy. "But you can take one with a 30 handicap and try to reduce it to a 29 or a 28, and if you're doing that, you're doing a great job. If we try to make you into Tiger Woods, you'll quit the first day."
As a result, the principal has honed a system of tracking students into courses based on their performance. The system pushes students in each track to achieve at higher levels, so they can move to a more advanced track later on. The hope is to have as few students as possible in the lowest track by the 8th grade.
"Originally, tracking had some big problems," Mr. Giffin concedes. "There was not a good system to get you in the right track, and once you got in, you couldn't get out. Here we'll change you any day of the week, or of the year."
The payoff is clear: Over the past three years, the number of his 8th graders taking high school algebra has jumped from about 100 to more than 150, and they're maintaining a competitive edge against 9th graders across the state. At the same time, most of those who don't make it to algebra are still showing gains above the national-norm gain for 8th grade math.
"Once you recognize what data does for you, if you've got the philosophy and willingness to change the school," Mr. Giffin says, "you can drastically affect the learning of individual students. It's that simple.''
Vol. 18, Issue 34, Page 27