Reading Programs Yield Few Gains in Comprehension
A federal study of supplemental programs that are intended to improve students’ reading comprehension has found that only one of the three programs examined actually did so.
The report, released May 5, focuses on the second and final year of research into the reading programs. It concludes that ReadAbout, a computer-based program by Scholastic Inc., improved students’ comprehension of social studies texts when the teacher had a previous year’s experience with the program. The size of the effect after an academic year of instruction was the equivalent of moving a student from the 50th to the 59th percentile, the researchers said.
ReadAbout showed no statistically significant effect, however, on tests measuring students’ comprehension of reading more generally, or of science texts, or for students whose teachers were in their first year of using the program.
No significantly positive effects were found for the other two programs in the study: Read for Real, created by Chapman University and Zaner-Boser; and Project CRISS, developed by Creating Independence Through Student-Owned Strategies.
The findings from the first year of the study, released last May, concluded that none of the programs studied significantly improved students’ comprehension, and that one had a negative effect. ("Supplementary Reading Programs Found Ineffective," May 13, 2009.)
That program, Reading for Knowledge, adapted for the study from a program created by the Success For All Foundation, was not studied in the second year because more than half the schools assigned to use it for the study declined to continue, according to researchers from Mathematica Policy Research Inc., which conducted the study for the U.S. Department of Education’s Institute of Education Sciences.
The research project was intended to probe the question of whether supplemental reading programs, used in “real world” settings, as schools typically would use them, could help young adolescents move from “learning to read” to “reading to learn.” The first year tracked use of the four programs by 268 teachers and 6,350 5th graders in 10 urban school districts in 2006-07. Schools were randomly assigned to use one of the programs or to serve as a control group.
In 2007-08, researchers followed the 5th grade cohort into 6th grade, where they did not use any of the supplemental reading programs, in order to see if any longer-term effects took shape from using the programs the previous year. They found no such effect from any of the four programs.
In the second year the researchers also added a second cohort of 4,140 5th graders to the study, who used Read for Real, Project CRISS, and ReadAbout with the same teachers who had used those programs the year before. The researchers wanted to know if teachers’ previous year’s experience with the programs helped students’ reading comprehension. It did, but only for ReadAbout and only in social studies.
“Our findings do not support the hypothesis that these four supplemental reading-comprehension curricula improve students’ reading comprehension, except when ReadAbout teachers have had one prior year of experience using the ReadAbout curriculum,” the study says.
ReadAbout teaches skills such as summarizing, making inferences, and detecting the author’s purpose through a computer-based program that adjusts to each student’s reading level.
Michael L. Kamil, a professor of education at Stanford University who specializes in literacy, said that the amount of improvement shown by ReadAbout, even if obtained annually, wouldn’t be enough to make a substantial difference in a struggling reader’s skills by the time he or she faced college or a job.
“You would run out of time in a school career before you make a significant dent in student achievement,” he said. “And that’s assuming you could get that result each year.”
But Margery W. Mayer, the president of Scholastic Education, called the results “extremely positive.”
Kristin DeVivo, Scholastic Education’s vice president of research and validation, said that the study might not have measured students’ reading comprehension fully because the test of general reading comprehension the researchers used, the Group Reading Assessment and Diagnostic Evaluation, or GRADE, isn’t designed specifically to capture their understanding of nonfiction text. But Susanne James-Burdumy, the lead author of the study and Mathematica’s associate director of research, said she and her colleagues used a form of the GRADE that is especially geared toward expository writing.
Part of the explanation for why the study found ReadAbout more effective than the other programs was that teachers implemented it more fully, and a larger portion of them reported actually using that program than did teachers assigned to use the other programs in the study, Ms. James-Burdumy said.
One question raised by the research, Mr. Kamil said, is whether teachers should use a variety of supplemental reading interventions, rather than just one, in their classrooms to address the diverse needs of young adolescent students.
Vol. 29, Issue 31