An article about research by Longwood economics professors Dr. David Lehr and Dr. Melanie Marks appeared in the Nov. 13 issue of the Harrisonburg Daily News-Record. The research, which will be presented at a meeting of the Southern Economic Association to be held Nov. 23-25 in Tampa, analyzed 2008-09 SOL pass rates in reading and math for third-graders in nearly every Virginia elementary school at that time.
Republished, with permission of the author.
Report Finds Ethnicity, Income Best Indicators
Posted: November 13, 2013
By EMILY SHARRER
Harrisonburg Daily News-Record
HARRISONBURG — Ethnicity and family income are the best predictors of student performance on Standards of Learning tests, new research based on Virginia test scores shows.
The more black, Hispanic and free- and reduced-price lunch students a school has, the lower the school’s pass rates were, the recently published study from Longwood University economist David Lehr found.
What’s more, "commonly cited remedies" for improving test scores — like smaller class sizes, higher teacher salaries and teachers with advanced degrees under their belts — were shown to have little effect on test scores, Lehr said.
"The results show that school districts should be very careful with trying to manage the policy variables that they control and thinking that by changing those they’re going to see substantial changes in SOL pass rates," said Lehr, whose son David Lehr and Longwood economics professor Melanie Marks co-authored a paper about the research. "When you change expenditures per pupil, when you change class sizes, when you change the percentage of teachers who are experienced … you’re going to see very little change, we predict."
The findings support an oft-quoted claim of Harrisonburg City Schools Superintendent Scott Kizner, who said the city’s high free- and reduced-price lunch population and the large number of students who speak English as a second language have a significant impact on test scores.
"This study only reinforces what I have been saying the past four years," Kizner said. "Our high diversity and free-reduced lunch numbers challenges us daily to find unique ways to meet the learning needs of our students."
City Has Highest LEP Rate
In Harrisonburg, about 34 percent of the division population is limited English proficient. Kizner said about 70 percent of the Hispanic population in the school system is LEP.
The division is expected to maintain its standing this year for having the highest percentage of LEP students in Virginia, Kizner said.
Rockingham County Schools is less diverse. As of Sept. 30, only 577 students are identified as LEP, which represents 5 percent of the division’s enrollment.
In Virginia, 40.1 percent of schoolchildren, or 496,771 kids, qualified for free- or reduced-price meals in 2012-13, according to the Virginia Department of Education. Numbers for 2013-14 are not yet available.
In Harrisonburg, 73.4 percent of students are eligible for the meal benefits for 2013-14. That represents 3,952 of the division’s 5,380 students in prekindergarten through 12th grade.
In Rockingham County Public Schools, 4,631, or nearly 40 percent, of the division’s 11,876 students receive free- or reduced-price meals, based on federal guidelines for the National School Lunch Program.
‘Indicators’ Used
Using data from 99 percent of elementary schools in the state — including all of those in Harrisonburg and Rockingham County — Lehr analyzed SOL pass rates in reading and math for third-graders from the 2008-09 school year. That was the most recent year that data was available when the study began.
Lehr also looked at 22 indicators — the percent of faculty with a master’s degree, class sizes and average faculty salaries, for instance — to examine their importance in predicting pass rates.
"I thought rather than do another attempt at cause and effect, instead how about I show policy makers how they could use these tools to help them predict what will happen to SOL scores when they make policy changes," Lehr said.
His findings show that for math and reading, the percentage of students eligible for free- and reduced-price lunch and the number of black students were the two best predictors of pass rates.
The higher those percentages in a school’s enrollment, the lower the school’s pass rate turned out to be.
The percentage of Hispanic students was also a strong predictor in those subjects, he said. Lehr said that due to a lack of available data, he did not include students who speak English as a second language as a variable. He suspects that a school’s Hispanic enrollment came out as a strong predictor of SOL scores because a high percentage of those students do not speak English as a first language.
‘Good Teaching’
Some school officials, though, questioned the results of the study, noting that several schools with high black, Hispanic and free- and reduced-price lunch enrollments have impressive test scores.
Doug Alderfer, assistant superintendent for administration for Rockingham County Schools, said the county’s free- and reduced-price lunch numbers, for example, do not always correlate with SOL scores. Schools like Pleasant Valley, Peak View and South River elementary schools have high numbers of free- and reduced-price lunch eligible students, but historically have done "very, very well" on SOLs, he noted.
"My experience has been that good teaching gives you good test scores; that’s really the bottom line," Alderfer said. "We shouldn’t handicap ourselves and say just because kids are economically disadvantaged means they’re going to get low scores. I think I would be foolish to say it doesn’t have some impact, but I think it can be overcome for sure."
The solution to remedying test scores may lie in early childhood education, Lehr and Kizner agreed, to ensure students are exposed to vocabulary and books at an early age — things they might not have if they come from a low-income background.
"Children need to be taught before age 5," Kizner said. It’s critical that we really begin to have high quality early childhood programs."
Contact Emily Sharrer at 574-6286 or esharrer@dnronline.com
Leave a Comment