Oklahoma Pre-K Program Found Effective
Oklahoma's five-year effort to make prekindergarten available to all 4-year-olds in the state is paying off, especially for Hispanic and African-American children, a new evaluation shows.
Conducted by researchers at Georgetown University, in Washington, the study focuses specifically on the prekindergarten program provided by the Tulsa public school system—the state's largest district, with 40,800 students and more than 1,600 pre-K children.
The findings show that, on average, children's test scores increased 16 percent after participating in the one-year program, with the greatest gains made in cognitive and language skills. Smaller increases in motor skills were found, but no significant improvements in social and emotional development.
When the results are broken down by race and ethnicity, they show that Hispanic children are benefiting the most from the program, with their overall scores increasing 54 percent. African-American children showed a 17 percent increase in scores. Participation in the program was not shown to have any remarkable effects on white children.
A similar pattern emerges when the findings are viewed by economic status. Test-score gains were insignificant for children who didn't qualify for federally subsidized lunches, the common poverty benchmark for students. For children eligible for reduced-price lunches, language scores increased by almost 35 percent. Those qualifying for free lunches posted a 31 percent increase in cognitive skills, a 15 percent increase in motor skills, and an 18 percent increase in language skills.
The results could be used to support the theory that targeting services to minority and disadvantaged children is a more effective strategy than opening programs to all children. But the authors of the study, William T. Gormley, a professor of government and public policy at Georgetown, and Deborah Phillips, a psychology professor there, suggest that diversity is what is making the difference.
As has been the case with Georgia's lottery-financed pre-K program, universal access brings "the political advantages of widespread public support," the authors write.
"It is also possible," they write, "that some of the classroom benefits that accrue to disadvantaged children are attributable in part to the presence of more advantaged children in the same classroom."
'What Really Matters'
While the Oklahoma universal pre-K initiative has not received as much national attention as Georgia's or New York state's, the program has grown quickly since it began in 1998. By the 2001-02 school year, 494 of the state's 543 school districts were receiving state money to participate.
Unlike in Georgia and New York, where services are provided in both schools and community-based child-care centers, Oklahoma's pre-K classrooms are entirely housed in public schools. And instead of focusing on specific curriculum goals, the state decided to emphasize teacher training as the pathway to high-quality pupil readiness for school.
Oklahoma's law requires every pre-K teacher to have a bachelor's degree and a certificate in early-childhood education. Pre-K teachers also receive the same pay and benefits as public school teachers, further erasing the boundaries that typically exist between preschool and K-12 teachers.
The Oklahoma model, Mr. Gormley said in an interview, "asserts that what really matters in early childhood is the quality of the teacher and the education of the teacher."
A recent study of state-financed prekindergarten programs, conducted by Yale University researchers Walter S. Gilliam and Carol H. Ripple, showed that 12 out of the 33 states surveyed required teachers to have both a four-year degree and a teaching certificate.
A separate analysis of research on pre-K programs released earlier this month by Marcy Whitebook of the University of California, Berkeley, concluded that requiring teachers to have bachelor's degrees is the best step policymakers can take toward building programs that give children the skills they need for kindergarten. ("Study: Pre-K Teachers Need 4 Years of College," Oct. 1, 2003.)
New Research Approach
In Oklahoma, both part- and full-day pre-K programs are offered, with a majority of prekindergartners—57 percent—attending half-day classes, and the rest attending for a full day.
Mr. Gormley and Ms. Phillip's study finds that the full-day program was most beneficial for Hispanic children, with Hispanic 4-year-olds in those classes showing an overall increase in scores of 73 percent. African-American children attending the full-day program showed modest gains.
But among white children, only those attending the half-day program showed an increase in achievement—specifically, a 19 percent jump in language-development scores.
Mr. Gormley said he was most confident about the findings on Hispanic pupils, because the numbers of children attending the half- and full-day programs were roughly equal.
He added that the apparent lack of benefits for non-Hispanic white children in the full-day program might be due to the possibility that a greater number of disadvantaged and minority children in the full-day program were getting more attention from the teacher.
The study on Tulsa's pre-K effort is different from most studies on the effects of early-childhood programs. Typically, researchers compare the progress of children in a certain program with that of children who were eligible but did not enroll in it.
While widely accepted, Mr. Gormley said, that method always leaves a "nagging doubt" that those parents who enrolled their children in the program were in some way different from those who did not—a factor researchers call "selection bias."
In Tulsa, however, a test called the Early Childhood Skills Inventory was given in 2001 to 4-year-olds about to begin the prekindergarten program and to 5-year-olds who had been in pre-K and were about to enter kindergarten. The design solved the selection-bias problem, the researchers say, because all of the children's parents had chosen the pre-K program.
Ramona Paul, the assistant superintendent of professional services for the Oklahoma Department of Education, said it's possible that effects on social and emotional skills were not found because the test did not capture those changes in children. But she added that she was especially pleased to see that the program was benefiting Hispanic children.
Vol. 23, Issue 9, Pages 19,22