John Warren, a University of Minnesota sociology professor, spent years researching the correlation between high school exit exams and dropout rates.
He, along with other researchers, speculated that retention — requiring students to repeat a grade — may increase dropout numbers. But no data existed on state retention rates.
In 2009, Warren and sociology graduate student James Saliba decided to measure retention rates in public school students in grades one through eight.
The study, published in the Educational Researcher in November, is the first of its kind to measure retention rate by state rather than nationwide.
Warren said retention rates are guided by state education policies and there’s no federal policy on it. According to Warren, the national-level estimates are insufficient.
“Local and state policies affect the rates, so there’s no reason to think that they’re similar across states, and it turns out they’re not,” he said. “It’s really something that varies from state to state.”
The research was the first to use a mathematical data collection strategy. According to the University News Service, it’s also the first to use the U.S. Department of Education’s Common Core of Data — a program that collects annual data on public schools in the U.S.
The study found that, in general, the rate of retention is highest among first grade students, but overall varies from state to state.
“I think first grade is kind of a formative year, it’s where you begin to do the building blocks of reading and math and it’s always hard to tell when a kid is ready to move on,” Warren said. “I think there is a sentiment, right or wrong, that it might do some good to hold a kid back in first grade to help them build those foundational skills.”
Despite the new technique, the study concedes that the data collection model is most accurate when used to measure retention in first through eighth grades rather than high school grades, which often rely on credit load to determine year in school.
This can make grade placement difficult. A student may be in his or her sophomore year of high school, for example, but a junior in terms of credits earned.
"There is sometimes a difference between what a school administratively classifies a student as compared to what a student would think of themselves as,” Warren said.
Warren said he hopes future research will look into this conflict and find ways to measure retention rates in high school. He said he also would like to see researchers examine retention rates across social groups and analyze data over a longer period of time.
“I hope I look back in several years,” he said, “and a lot people have published improvements on what we’ve done.”