What’s new in comprehension research?

    by Freddy Hiebert | July 12, 2006

    For those of you who are checking the dates of the Frankly Freddy installments, you’ll note a hiatus between the last installment and this one. Yes, I took off last week for the 4th of July and to attend the annual meeting of the Society for the Scientific Study of Reading in Vancouver.

    My experience at SSSR this past weekend in Vancouver led to me the decision that it is going to be one of my regular conferences—not just one that I attend when the location is convenient. If you’re wondering why I came to that decision, visit the SSSR website to read some of the abstracts.

    One of the papers that I thought was very germane to the Institute was Joanne Carlisle’s report on Reading First in Michigan. In that Michigan was the first state to get its Reading First grant, Joanne and her colleagues at U of M (who serve as evaluators of the effort) were able to get started earlier than most efforts. Thus, they have two years of data that have been analyzed and have just finished the collection of the third year’s data. The results are very positive, as is evident in the following snippet from a press release at the University of Michigan website (see the full press release):

    Researchers studied performance on three reading subtests of the Iowa Test of Basic Skills from the end of year one, when many reform efforts produce early payoffs, to the end of year two, when progress tends to slow. The team has found continuing gains across the board thus far from year to year.

    The results show that a significantly larger percentage of first, second and third graders were reading at or above grade level after two years than after one year in Reading First schools. Similarly, a significantly smaller percentage of first, second and third graders were substantially underachieving in reading after two than after one year in Reading First schools.

    The Michigan effort demonstrates what can happen when supervisors and teachers do considerable and concerted work with primary-level students. It’s important to remember that the Michigan success presented in this report is derived from a norm-referenced, standardized test.

    A second paper at SSSR that I found particularly salient to the Institute was presented by Peter Dewitz, Jennifer Jones, and Susan Leahy. As the title of their paper indicates—Reading Comprehension Instruction in Five Basal Reading Programs: Durkin Revisited—they conducted an analysis of the nature of comprehension instruction in five current basal programs. Durkin’s classic studies (of classroom instruction as well as the manuals of teachers’ guides) pertained to students in the middle grades (i.e., fourth through sixth). They concluded that, while basal reading programs offer some guidelines for direct instruction, there is minimal guided practice. Much of the instruction is in the form of teachers asking students questions. Dewitz et al. concluded that “The instruction in basal programs lacks the intensive massed practice found in the original research studies for five key strategies” (self-questioning, summarizing, narrative structure, main idea and making inferences).

    The proficiency levels of primary-level students can be improved through hard work by teachers and their educational leaders (the Michigan Reading First report). The next challenge is to build on this proficiency and support high levels of thinking with text. The Dewitt et al. analysis suggests that the programs and support systems for teachers that support this proficiency and thinking about text need to be considered carefully. Whether current basal programs can provide the guidance for middle-grade students and their teachers that ensures that students build on their initial gains remains the next area of study for educators and educational researchers.