January 25, 2006

Study: Literacy program has lasting impact on Indiana students

Maribeth Schmitt

The team, led by Maribeth C. Schmitt in Purdue University's College of Education, surveyed 548 children in 253 Indiana schools that had Reading Recovery programs in place for at least two years. The purpose of the study was to assess the long-term impact of the program, which helps low-achieving first-graders improve their reading skills to an average level of achievement or beyond.

The team found that in both analyses the team conducted — oral text reading and standardized reading — the vast majority of students who had completed a Reading Recovery program did as well or better than their peers one, two and three years after finishing the program.

Schmitt, who is director of the Purdue Literacy Network Project and established Reading Recovery for Indiana in 1992, said the results were not surprising but confirmed what she had known for years and what had been previously proven in other states.

"We undertook the study at the request of Indiana legislators and school administrators who wanted to know if early-intervention programs like this have a positive outcome in the long term for students and if the amount of money being spent on Reading Recovery is worth the expense," she said. "What we found was that the program does help students years down the road and that schools are getting a substantial return for their investment. The findings were quite affirming."

Reading Recovery is used by more than 400 schools in Indiana and more than 10,000 schools nationwide. It is implemented in the first grade, and students are helped through tutoring sessions with trained teachers that last 30 minutes a day. The intervention usually lasts 12-20 weeks.

Schmitt worked on the research with Anne E. Gregory, an assistant professor of education at Boise State University and former Purdue doctoral student. The study was funded by the Indiana Department of Education and was published in the current edition of Literacy Teaching and Learning: An International Journal of Early Reading and Writing.

To assess the impact of the program, Schmitt and Gregory surveyed students in second, third and fourth grades who had completed Reading Recovery and compared them to children of the same age who had not been enrolled in Reading Recovery.

In oral-text reading, the study found that 88 percent of the former successful Reading Recovery children in second grade were reading at or above their grade level, compared to 83 percent of their classmates. In third grade, 96 percent of the Reading Recovery students were reading at or above their grade level, compared to 89 percent of other students. And among fourth-graders, 90 percent of those who had completed the program were performing at or above their grade level, compared with 92 percent of those who hadn't been in the program.

All students were selected randomly for comparing samples of the average and the Reading Recovery group from the total population in the schools.

Schmitt attributes the high success rate of Reading Recovery to three things.

"First is the high-quality training that the teachers receive," she said. "Second is that the program relies on research-based strategies. And finally, the individualized, one-on-one structure of the teaching allows students not just to learn how to read, but it also helps them become better problem solvers and rely less on the teacher when they struggle."

Schmitt and Gregory also found that in oral-text reading at all grade levels, the lowest-achieving children who had participated in Reading Recovery were performing above the lowest-achieving students who hadn't been in the program. Schmitt said this suggests that the children who complete the intervention no longer dominate the low end of the achievement scale.

When evaluated on standardized reading tests, the researchers found that most former Reading Recovery students outperformed their peers in all areas tested. Schmitt and Gregory viewed students' performance on the Gates-MacGinitie Reading Test total composite reading score, a vocabulary subtest, a comprehension subtest and the Indiana State Test of Educational Progress, known as ISTEP.

On the total reading test, 86 percent of the second-grade former Reading Recovery second-graders scored within the average range, as did 85 percent of the third-graders and 80 percent of the fourth-graders.

Similar numbers were seen among former Reading Recovery children in vocabulary achievement and reading comprehension. And on ISTEP, the percentage of former Reading Recovery students scoring in the normal range was similar to that of the general population.

Schmitt also recently published another study that measured awareness of strategies useful for before, during and after reading a story, such as activating background knowledge, previewing and predicting for comprehension. She compared the same group of former Reading Recovery children to the random sample population and found no differences between their level of strategy knowledge two and three years later.

"That study showed that students overwhelmingly remembered and employed those alternative strategies for understanding when they came across unfamiliar or difficult material," she said.

"Reading is a complex process that involves many different cognitive processes, much like playing the piano. One of the main reasons Reading Recovery is so successful is that the individualized attention helps participants be better problem solvers and teaches them how to help themselves, which is a vital skill that all children must learn."

Reading Recovery is one of three programs offered by the Purdue Literacy Network Project in the College of Education. The others are the Literacy Collaborative and the Professional Development Initiative.

The Purdue Literacy Network Project provides professional development to help teachers create successful school literacy programs for students in elementary grades to assure all children are given the chance to become readers and writers early in their schooling.

Writer: Kim Medaris, (765) 494-6998, kmedaris@purdue.edu

Source: Maribeth C. Schmitt, (765) 494-5683, mschmitt@purdue.edu

Purdue News Service: (765) 494-2096; purduenews@purdue.edu

The purpose of this study was to contribute to and strengthen previous work that examined the long-lasting effects of Reading Recovery in statewide efforts aimed at bolstering early literacy achievement and reducing early learning difficulties. Specifically, the study explored the literacy achievement of Reading Recovery participants whose series of lessons had been successfully discontinued during their first-grade year at points 1, 2, and 3 years beyond receiving the intervention in Indiana—providing a picture in time for where the children are now. The participants included randomly selected children who had either successfully completed Reading Recovery or who had not participated in the intervention (i.e., cohort sample) from the three grade levels in 253 schools in Indiana. The two assessment instruments used to gauge literacy performance included the running record of oral text reading (Clay, 1993) and the comprehension and vocabulary subtests of the Gates-MacGinitie Reading Tests and the score for total test. The fourth-grade former Reading Recovery children’s results on the state achievement test taken in third grade were collected from their school records to establish their achievement distribution 2 years beyond the intervention. Results indicate a considerable majority of the former successful Reading Recovery children were reading text at or above their grade level and that 1, 2, and 3 years beyond the intervention, Reading Recovery children were performing roughly as well as or better than their cohort sample peers on the task of oral text reading. Analysis of the Gates-MacGinitie Reading Test data indicated the vast majority of the previously successful Reading Recovery children performed within the calculated average bands of the cohort sample groups at each grade level, indicating the formerly struggling learners were continuing to progress with their peers in literacy. In addition, the former Reading Recovery fourth graders achieved a normal curve distribution with a mean of the 45th percentile on the Indiana State Test of Education Progress (ISTEP), a considerably different pattern from their first-grade 15–20% achievement range.

To the News Service home page