Effects of classroom practices on reading comprehension, engagement, and motivations for adolescents

Guthrie, J., & Klauda, S.L. (2014). Effects of classroom practices on reading comprehension, engagement, and motivations for adolescents. Reading Research Quarterly, 49(4), 387-416.


This study is the most recent installment of the work of John Guthrie and his colleagues on the relationship between reading engagement and reading comprehension. I was among many listeners who packed into a conference hall a few years ago to hear Guthrie discuss his findings that suggested that the amount of time spent by children in active, engaged reading might have a stronger relationship with reading comprehension than even socioeconomic status of children, which is known to be one of the most powerful factors influencing reading achievement. In Guthrie’s studies with middle elementary children, he basically found that those children who reported more engaged reading time but who were from low socioeconomic status groups actually had higher reading comprehension scores than children from higher socioeconomic status groups who did not report spending as much time in engaged reading. These were powerful findings, and those of us in the room felt that we were hearing something very important that we could actually take to our own work and maybe use it to make a difference.

Much of Guthrie’s work has revolved around gathering support for a system he calls Concept-Oriented Reading Instruction (CORI), which is instruction designed to enhance and facilitate students’ engaged reading. If the relationship between engagement and reading comprehension that Guthrie claims exists is a valid relationship, then CORI should also lead to enhanced reading comprehension. Previously, the CORI model was tested mainly with middle elementary grade students and focused on science content integrated with literacy skills; this study extends that work to middle school students and focuses on history content integrated with literacy skills.

The CORI model has four basic components (as summarized on page 388 of the article): 1) competence support, 2) providing choice, 3) emphasizing importance of reading, 4) arranging collaboration. The activities in CORI are devised specifically to provide those four things to students, with the idea that these things promote students’ engagement with content area text materials and in turn promote reading comprehension. The hypothesis is that instruction focusing on these four components will promote engagement and comprehension better than what the researchers call “traditional” instruction.

A strength of the article is the large amount of detail provided on CORI, including detailed descriptions in the article and several appendices at the end. The reader can get a fairly clear and complete picture of what this instruction was supposed to look like. To me, it looks like a very systematic and purposeful approach that required extensive advance planning and a high level of teacher engagement as well as student engagement. I like CORI in principle, and I also like it mostly in the way it is implemented here, but I cannot help but wonder whether it is the planning, the intentionality, and the engagement required to implement the model that makes it work rather than the actual elements that comprise it. Teacher engagement may be as important as student engagement as a factor that leads to high levels of reading comprehension. It is not that I don’t think CORI’s four factors are good elements to have in reading instruction, but it may be harder to draw a direct line between those four things and student outcomes than this article suggests.

The methodology described in the study complicated the problem for me. The study compared outcomes for CORI versus “traditional” literacy instruction, but it was done using a “switching replications” design (p. 392) rather than a more typical experimental or quasi-experimental design with two carefully controlled groups. As I understand it, in the switching replications design, all students get both forms of instruction, with half getting the experimental instruction (in this case, CORI) and half getting the “business as usual” instruction in the first phase, and then basically switching the two groups in a second phase. Both halves of the student group thus received both kinds of instruction, but in two different orders. Data were gathered before the first phase, after the first phase and before the groups were switched, and then at the end after the switch. Students got four weeks of each kind of instruction.

The authors in general claim the superiority of CORI for both groups, no matter which order they received the two kinds of instruction, and the data as presented here do seem to support their conclusions. I was left wondering some things, however. One concern was whether with this design the effects of CORI might infiltrate and influence the effects of the traditional instruction. This could be especially possible for the group that had CORI first. If CORI is effective, might it not have effects that would last into the next phase? Furthermore, the teachers were the same for both phases, and those teachers had training (albeit fairly brief training) in the CORI model. Might that not influence teaching in all models, especially if the teacher buys into CORI (we have no information here on that). As a teacher, I would find it extremely difficult to keep from letting good principles that I knew could help my students from contaminating “business as usual” instruction. These kinds of concerns are really not addressed in the article.

Finally, I had concerns about the “control group” instruction itself. In many ways it looked like a comparison between apples and oranges. The CORI instruction was integrated history and literacy, and the students read informational text. The description of “traditional instruction” sounded like more literature-oriented instruction with narrative texts. The tasks involved in these two kinds of instruction are very different! They are discussed in the Common Core State Standards as different skills, and the demands required for both teaching and learning these skills appear to be unique. I would have found this study more convincing if two forms of informational text instruction had been compared. Of course, that might also have compounded the problem with contamination that I described above.

I’ll continue to follow this important body of research, in spite of my concerns with this study. It seems clear that engagement is an important factor in literacy learning, even though we still need to keep working on the best ways to promote that engagement when we work with learners.