“Tracing the Origins” is a multi-part series. Each post explores how a promising finding in education research became distorted and misinterpreted in its widespread implementation.
Frenemies: Background Knowledge and Comprehension Strategies
This post untangles the fascinating history and current misunderstandings around what teachers nowadays commonly refer to as “reading comprehension strategies.” Surprisingly, modern misconceptions around reading comprehension have their roots in promising research from the 1970s in Cognitive Strategy Instruction.
Rosenshine (1997) wrote “the research using cognitive strategies, from 1970-1990, has produced incredible results, results for which we as a profession can be justly proud.” However, things seemed to go downhill, quickly. Dole (2009) states “in the transition from research to practice, strategy instruction has morphed into so many things that it no longer has a shared meaning.”
As such, reading comprehension strategies are out, and an emphasis on background knowledge is in. In an excerpt from her new book, The Knowledge Gap, author Natalie Wexler captures this shift:
“What if the best way to boost reading comprehension is not to drill kids on discrete skills but to teach them, as early as possible, the very things we’ve marginalized—including history, science, and other content that could build the knowledge and vocabulary they need to understand both written texts and the world around them?”
Wexler’s assessment, that the teaching of reading comprehension has become narrowly constricted to practicing a growing set of disconnected skills similar to multiple choice exams (“find the main idea…make inferences”) is absolutely spot-on. What is missing, however, is a careful consideration of what strategies are actually effective for aiding comprehension of knowledge-rich text, given existence of sufficient prior knowledge. Or put simply, how can we best help students meaningfully interact with text to deepen comprehension?
When we look back in history at the original research, it becomes apparent that knowledge-building and research-informed cognitive strategy instruction were never enemies, and actually went hand-in-hand.
Teacher Effects and Cognitive Strategy Instruction
Lately, there has been an incredible resurgence of interest in Teacher Effects research that flourished from 1960-1980. This research studied teachers whose students made the most and least gains on achievement tests, and then compared the instructional procedures used by the effective (expert) and ineffective (novice) teachers to isolate specific, research-based principles of instruction.
The results of these Teacher Effects studies were used to form the instructional models articulated by Brophy & Good (1986) and Rosenshine & Stevens (1986). However, there were obvious limitations to this Teacher Effects research.
“The concepts from the teacher effects research were very useful when we could break a task into series of explicit steps, guide student practice on those steps, and provide support, feedback, and practice to enable students to respond at a high level of success.
But the concepts from the teacher effects research seemed less useful for teaching tasks that could not be broken into explicit steps, tasks such as reading comprehension.”
To better understand how to teach tasks in which each step could not be explicitly specified, a companion to Teacher Effects research emerged in the 1970s-80s: Cognitive Strategy Instruction. This research was based on the idea that humans develop cognitive strategies for processing incoming information and meta-cognitive strategies for monitoring and evaluating the understanding of that information.
Cognitive strategies were to be explicitly taught initially, and Dole (2009) writes “with time and practice, the use of both cognitive and metacognitive strategies can become less effortful and can be carried out efficiently and effectively at an automatic level.”
To gradually release responsibility to students while teaching a strategy, teachers would scaffold concrete procedural prompts. For example, in teaching the strategy of question-generation, the teacher might scaffold this by modeling the process while thinking aloud, and guiding practice using a concrete prompt such as a list of question words (who, what when). In this way, the Teacher Effects and Cognitive Strategy Instruction research built off each other, and were practically joined at the hip.
“When we compare the instructional procedures from the earlier teacher effects literature with those that emerged from this cognitive strategy research, we find that these two sets of instructional procedures complement each other…both sets of concepts are important for instruction.”
But here’s the crucial part that was seemingly missed as publishers began swiftly introducing a wide variety of additional “comprehension strategies” and “skill sheets.”
“Procedural prompts are most useful when the student has sufficient background knowledge and can understand the concepts in the material. Procedural prompts and the use of scaffolds cannot overcome the limitations imposed by a student’s insufficient background knowledge.”
-Rosenshine, Meister & Chapman (1996)
To summarize so far:
- There were two primary strands of educational research in the 1960s-1980s that were tightly connected and responsible for promising findings:
- Teacher Effects: identified effectiveness of frequent review, guided practice, and questioning. Led to research-based principles of instruction.
- Cognitive Strategy Instruction: identified effectiveness of teachers’ scaffolding of concrete prompts while teaching abstract, higher-order cognitive strategies such as question-generation and summarization.
- These cognitive strategies were to be explicitly taught at first, and were recognized to be contingent on the degree of existing background knowledge.
Emerging from Cognitive Strategy Instruction research was a promising reading comprehension procedure – Reciprocal Teaching – that would initially be taught whole group, but could over time be done effectively in pairs (Palincsar & Brown, 1984; Rosenshine & Meister, 1993).
The core idea behind Reciprocal Teaching, or “ReQuest” was to teach students four specific cognitive strategies to aid reading comprehension (Dole, 2009). These strategies included question-generation, summarization, predicting, and clarifying. According to Rosenshine and Meister’s (1993) review of 19 studies involving Reciprocal Teaching, two of the four strategies were consistently found to be the most effective in boosting comprehension:
“Based on this research, the strategy of question-generation, taught individually and in combination with other strategies, had the highest success rate, followed by summarization.”
To support students in generating questions, Rosenshine, Meister & Chapman (1996) recommended teachers model and teach using two types of prompts:
- Signal Words – prompting with “when…why…how”
- Generic Question Types/Stems
- “What is another example of _____”
- “How are ____ and ____ alike/different”
- “How does ____ affect ____”
And to support summarization, the Kintsch & van Dijk (1976) procedures were reviewed (deletion, superordination, selection, and invention) along with the legs-and-table procedure in which students list main details to generate a summary sentence.
These strategies, question-generation and summarization, are actually heavily emphasized in Hochman and Wexler’s (2017) The Writing Revolution as key to improving reading comprehension.
“It’s as important for students to learn how to generate questions as it is for them to learn how to answer them…summarizing can be a challenging and rigorous activity that provides powerful benefits.”
And so, how did we get from the research-based foundation of Cognitive Strategy Instruction, to where we are now – the Wild, Wild West of research comprehension strategies and skills.
What Went Wrong – Skills or Strategies?
Dole, Nokes, and Drits (2009) reviewed the history of Cognitive Strategy Instruction and concluded “much of the fidelity of cognitive strategy implementation has been lost in the translation from research to practice.” One of the biggest confusions arose over the confusion between strategies and skills.
“During the 1980s, when so much research was being conducted on cognitive strategies, teachers taught reading comprehension as a sequence of separate skills that were identified in the basal reading programs.” – Dole, Nokes, and Drits (2009)
Over time, educators were led to mistakenly believe that with repeated practice in more and more skills (finding main idea, sequencing, inferring, visualizing, identifying theme etc.), students’ comprehension would magically improve. As a result, an increasing number of skills were thrown at students with very little explicit teaching. It became a matter of quantity over quality.
“Teachers did no teaching; instead, students practiced the skills and teachers “tested” whether students could use them. In other words, when teachers directed students to “find the main idea” and to “create a summary of a story,” there was no help or assistance for students who could not find the main idea or create a summary.” – Dole, Nokes, and Drits (2009)
In addition, as the number of reading skills and strategies metastasized in workbooks and test-preparation booklets, the deliberate focus of building students’ precious background knowledge was buried.
“In the rush to teach cognitive strategies, teachers work on the strategies without regard to the content of the text…These teachers may forget that the goal of strategy instruction is improved understanding of a given text, and improving the ability to comprehend across texts, not learning the strategies.” – Dole, Nokes, and Drits (2009)
Where Do We Go from Here?
I hope to have shown over the course of this analysis that there are redeeming qualities of cognitive strategy instruction, so long as a few strategies are implemented with care and attention to the original guiding research. Students do not need 300 different strategies and skills for reading, they probably will do just fine with two.
In order to replace rampant, ineffective practices we see in schools, I suggest we radically simplify and dramatically refocus discussions around reading comprehension strategies, while continuing the much-needed emphasis on developing students background knowledge.
- The cognitive strategies of question-generation and summarization are found to best support reading comprehension. Shanahan has a wonderful two-part series on teaching summarization.
- For Cognitive Strategy Instruction to be effective, sufficient background knowledge must be present.
- Cognitive strategies should be explicitly taught and practiced extensively as a whole-group before being gradually released.
Next post in this series: Tracing the Origins: The Slippery Slope of Cooperative Learning
I’m on Twitter.
Rosenshine, B. (1997). The case for explicit, teacher-led, cognitive strategy instruction. MF Graves (Chair), What sort of comprehension strategy instruction should schools provide. LINK
Rosenshine, B., & Meister, C. E. (1993). Reciprocal teaching: A review of 19 experimental studies. Center for the Study of Reading Technical Report; no. 574. LINK
Rosenshine, B., Meister, C., & Chapman, S. (1996). Teaching students to generate questions: A review of the intervention studies. Review of educational research, 66(2), 181-221. LINK