Cognitive science has demonstrated repeatedly that knowledge is essential for making sense of texts. Suggesting otherwise is misleading—and potentially harmful to students who need it the most.
"'There’s never been a study to show that increasing knowledge of the world boosts reading scores.' Such a broad and sweeping statement is demonstrably incorrect, deeply misleading, and carries significant downside for reading instruction if taken uncritically."
What exactly does this "reading instruction" look like? I'm thinking about the overwhelmed, time-strapped teacher (me) and HOW I should teach reading. If I did my knowledge-building through vocabulary development and writing related to my social studies and science curricula, is that enough knowledge, and can I call it a day? Then, can I find "text sets" to support this knowledge for students to read during the ELA block for further exploration? The "what" is the knowledge. But when it comes to teaching, the "how" also really matters. How do we help students navigate complex text to analyze what's going on and synthesize a response to that text? That's the real teaching part.
In the classroom it was essential for me to build (and assess) student background knowledge before I asked them to apply it in a skills based manner (project/product/performance). Also important but often overlooked is the need to contextualize students’ knowledge and skills within a unit of study and in relation to the wider world around them.
I do believe that knowledge-rich curriculum favors the advantaged. But I can't imagine what any alternative would be other than trying to provide a content-rich background for students who didn't get a head start on it.
"Hey, I have an idea! Let's teach the disadvantaged kids what the advantaged kids come to school knowing."
Agreed! And leave behind the soft bigotry of NO expectations! But--those of us who teach these "disadvantaged kids" know there are only so many minutes in the day, and the opportunity costs loom large. So we have to get this right every day in every way possible.
This is very right. Makes me think of a small example from Pathways. We realized our students would never be proficient at "Describing characters in a story (e.g. their traits, motivations, or feelings," a Common Core standard from grade 3, because they didn't have a vocabulary of character traits to choose from. Is the character gullible? Arrogant? Bold? They needed these words, specifically, not just the habit of describing characters. So for a whole semester we taught them. Big results. And how enjoyable, to have a class discussion analyzing a character from The Outsiders or Roll of Thunder, Hear My Cry when everyone had the knowledge to participate.
Still, one of many simultaneous interventions we did, and so very hard to isolate its impact. A good direction for future federal funding of research, perhaps.
Robert, I would further add that Barshay misrepresents Hirsch. She implies that Hirsch began with the theory that shared knowledge is good for democracy, which was then applied to cognitive science ("Now it’s a cognitive science argument that a core curriculum is also good for our brains and facilitates learning.) But that's backwards—Even in his first article and book about cultural literacy, Hirsch first surveys the decades of research (and this was in the 80s!) that prove that mature literacy is dependent on background knowledge and is not an isolated skill. It is from this evidence that he then concludes that the best kind of knowledge to teach is shared cultural knowledge, and he takes culture to primarily be national.
Therein lies the REAL question, not whether background knowledge is needed for literacy, but what constitutes background knowledge? I think another reason why these studies appear inconclusive, as Barshay suggests, is because so much knowledge is taught at home. Education pundits and researchers have blinders on when it comes to the outsized influence of the family on what kids learn.
This is a fantastic analysis of the root problem with research on reading instruction. However, one line stood out to me that I feel misinterprets Rudine Sims Bishop’s work. Specifically, the statement, “Moreover, an emphasis on ‘mirrors over windows’—prioritizing students' personal experiences in reading and writing—may offer a false positive. When students read and write about what they already know, they appear to be stronger readers and writers than they truly are, as the ‘Baseball Study’ would predict. However, when faced with unfamiliar domains, their comprehension falters.”
Dr. Bishop’s "mirrors and windows" metaphor is often misunderstood. She is not arguing that students should only read about experiences that mirror their own, nor that they should be limited to what they already know. Instead, Bishop advocates for both mirrors (texts that reflect students' own identities and experiences) and windows (texts that offer new perspectives and diverse experiences). Her central point is that literature should allow students to both affirm their own lives and identities and to explore the lives of others—this dual approach fosters empathy and a broader understanding of the world. As Bishop writes in her work, “When children see themselves in books, they feel affirmed. But when they see others, they begin to understand lives outside their own” (Bishop, 1990). This is key to building not just knowledge, but also a deeper sense of empathy.
The blog’s suggestion that emphasizing students’ personal experiences could result in "false positives" ignores the richness that comes from learning through diverse perspectives. Bishop's framework, in fact, aligns beautifully with a knowledge-building curriculum. Consider, for example, how a Civil War soldier’s account of a significant battle might challenge students to step outside their own experience, expanding their understanding of history and empathy for others’ struggles, even if those experiences are unfamiliar. This helps develop both critical thinking and empathy—crucial aspects of a robust knowledge-building curriculum.
In sum, Bishop’s work invites us to provide students with both familiar and unfamiliar stories, recognizing that learning about the lives of others enhances our collective understanding of the world, while also enriching students’ own sense of self. This is not just about affirming what students already know but expanding their knowledge and empathy through engagement with a diverse range of texts.
This quote—"It's not what good readers do, it's what they know."—is particularly revealing. You are creating an unhelpful dichotomy in how you word this as an either/or statement. Once again, we see a deliberate attempt to downplay the science supporting strategy instruction. Instead of acknowledging their role, you pivot to a coherent curriculum rather than admit that we need strategies too, alongside to help build the knowledge. Peng’s 2024 meta-analysis identified the most effective reading strategies, including main idea identification, yet this critical finding is conspicuously absent from every knowledge-centric discussion on the topic. This is an and/both and attempts to position it otherwise frustrate and confuse teachers. Good readers need both - they need to know stuff and they need to know what do as they read. And/both.
There is a clear, coordinated effort to minimize the role of strategies alongside knowledge-building. No one is arguing against the importance of knowledge, but the dismissal of well-evidenced, effective strategies—when properly taught—is both misleading and harmful. The call for "knowledge first," "knowledge only," or "knowledge instead of strategies" ignores decades of research demonstrating that strategy instruction is a stronger causal factor in improving reading outcomes.
Why not highlight the strategies that have consistently shown strong effects in rigorous, replicated studies? Researchers like Vaughn, Elleman, and Wijekumar have produced extensive evidence on the power of strategy instruction. Instead of sidelining these findings, we should be championing a comprehensive approach that incorporates both knowledge and the explicit teaching of effective strategies—especially for students who need them the most.
I don't think Pondiscio or the others he mentions, such as Daniel Willingham, are creating a dichotomy. You are right that it might not be an either/or statement. But to say it is "and/both" isn't entirely accurate either, because of all the studies that show that teaching reading strategies are helpful, but only can go so far in making up for a lack of background knowledge. Furthermore, strategies can be taught and then move one. Teaching strategies over and over again is not helpful.
As a history teacher, I help kids learn strategies all the time: how to examine and use evidence to make an argument, figure out cause & effect, compare and contrast multiple perspectives and examine change over time. But that's not what they go home and tell their parents about. They tell their parents about the stories -- about what happened when Theodore Roosevelt invited Booker T. Washington to dinner. And that Booker T. Washington wasn't lynched on his trip north because his would-be attackers didn't think to look for him on the fancy Pullman car. They know about Pullman because they learned about the Pullman strike earlier in the year. And neither Pullman, Washington nor Roosevelt are just names to them, because I share stories with them that humanize these people and put them into historic context.
So sure, teach strategies. But then move on to what is far more interesting to kids: science, social studies and literature. Kids are excited to learn about stuff--the world, people, animals, nature. They are not interested in learning about reading strategies. So lead with the good stuff and teach the strategies to help them learn the stuff, not the other way around.
I've also heard it stated "Teach the strategies in the SERVICE of knowledge-building," but this message (however you state it) sometimes gets lost in the frenzy of emphasizing knowledge-building because so much "skill-building" related to strategy instruction has been so awful for so long. When I teach students "paragraph shrinking" to help them "get the gist" there's a lot of "we do" related to complex text until my students become expert at the "I do." I love your Booker T. Washington story! But it reminds me of another strategy: teaching kids to separate what is important from what is interesting (or just cool). Oftentimes, the stories are memorable but not entirely germane to the substance of the text. When I taught my third graders a biomimicry unit, the stories related to the scientists were fascinating, but not something they necessarily needed to include in their summaries.
As IES goes the way of the dodo, take advantage of the practice guide for 4th - 6th (by Sharon Vaughn et al.) which says there is strong evidence for:
Routinely use a set of comprehension-building practices to help students make sense of the text
Part 3A. Build students’ world and word knowledge so they can make sense of the text
Part 3B. Consistently provide students with opportunities to ask and answer questions to better understand the text they read
Part 3C. Teach students a routine for determining the gist of a short section of text
Part 3D. Teach students to monitor their comprehension as they read
The knowledge-centric will occasionally give a nod that strategies are ok but do not give equal airtime to how to teach them well (for a calculated reason- to sell certain knowledge branded curriculum)- to the importance of correcting what Harriett describes as poorly taught strategies (she’s right). It misses the mark to say those of us worried about the attack on science and how science supports strategies are misunderstanding. We track is carefully. Here are the studies https://docs.google.com/spreadsheets/d/1Bw_JYU07S8050THEdx3eG4G061xxgvLsPpmEZvM87Zw/edit
"'There’s never been a study to show that increasing knowledge of the world boosts reading scores.' Such a broad and sweeping statement is demonstrably incorrect, deeply misleading, and carries significant downside for reading instruction if taken uncritically."
What exactly does this "reading instruction" look like? I'm thinking about the overwhelmed, time-strapped teacher (me) and HOW I should teach reading. If I did my knowledge-building through vocabulary development and writing related to my social studies and science curricula, is that enough knowledge, and can I call it a day? Then, can I find "text sets" to support this knowledge for students to read during the ELA block for further exploration? The "what" is the knowledge. But when it comes to teaching, the "how" also really matters. How do we help students navigate complex text to analyze what's going on and synthesize a response to that text? That's the real teaching part.
In the classroom it was essential for me to build (and assess) student background knowledge before I asked them to apply it in a skills based manner (project/product/performance). Also important but often overlooked is the need to contextualize students’ knowledge and skills within a unit of study and in relation to the wider world around them.
I do believe that knowledge-rich curriculum favors the advantaged. But I can't imagine what any alternative would be other than trying to provide a content-rich background for students who didn't get a head start on it.
Perhaps. But isn't one of THE problems in education is that virtually EVERYTHING favors the advantaged?
Hey, I have an idea! Let's teach the disadvantaged kids what the advantaged kids come to school knowing.
Oh, and let's break the smart kids' glasses. Level that playing field amirite?!
"Hey, I have an idea! Let's teach the disadvantaged kids what the advantaged kids come to school knowing."
Agreed! And leave behind the soft bigotry of NO expectations! But--those of us who teach these "disadvantaged kids" know there are only so many minutes in the day, and the opportunity costs loom large. So we have to get this right every day in every way possible.
This is very right. Makes me think of a small example from Pathways. We realized our students would never be proficient at "Describing characters in a story (e.g. their traits, motivations, or feelings," a Common Core standard from grade 3, because they didn't have a vocabulary of character traits to choose from. Is the character gullible? Arrogant? Bold? They needed these words, specifically, not just the habit of describing characters. So for a whole semester we taught them. Big results. And how enjoyable, to have a class discussion analyzing a character from The Outsiders or Roll of Thunder, Hear My Cry when everyone had the knowledge to participate.
Still, one of many simultaneous interventions we did, and so very hard to isolate its impact. A good direction for future federal funding of research, perhaps.
Robert, I would further add that Barshay misrepresents Hirsch. She implies that Hirsch began with the theory that shared knowledge is good for democracy, which was then applied to cognitive science ("Now it’s a cognitive science argument that a core curriculum is also good for our brains and facilitates learning.) But that's backwards—Even in his first article and book about cultural literacy, Hirsch first surveys the decades of research (and this was in the 80s!) that prove that mature literacy is dependent on background knowledge and is not an isolated skill. It is from this evidence that he then concludes that the best kind of knowledge to teach is shared cultural knowledge, and he takes culture to primarily be national.
Therein lies the REAL question, not whether background knowledge is needed for literacy, but what constitutes background knowledge? I think another reason why these studies appear inconclusive, as Barshay suggests, is because so much knowledge is taught at home. Education pundits and researchers have blinders on when it comes to the outsized influence of the family on what kids learn.
This is a fantastic analysis of the root problem with research on reading instruction. However, one line stood out to me that I feel misinterprets Rudine Sims Bishop’s work. Specifically, the statement, “Moreover, an emphasis on ‘mirrors over windows’—prioritizing students' personal experiences in reading and writing—may offer a false positive. When students read and write about what they already know, they appear to be stronger readers and writers than they truly are, as the ‘Baseball Study’ would predict. However, when faced with unfamiliar domains, their comprehension falters.”
Dr. Bishop’s "mirrors and windows" metaphor is often misunderstood. She is not arguing that students should only read about experiences that mirror their own, nor that they should be limited to what they already know. Instead, Bishop advocates for both mirrors (texts that reflect students' own identities and experiences) and windows (texts that offer new perspectives and diverse experiences). Her central point is that literature should allow students to both affirm their own lives and identities and to explore the lives of others—this dual approach fosters empathy and a broader understanding of the world. As Bishop writes in her work, “When children see themselves in books, they feel affirmed. But when they see others, they begin to understand lives outside their own” (Bishop, 1990). This is key to building not just knowledge, but also a deeper sense of empathy.
The blog’s suggestion that emphasizing students’ personal experiences could result in "false positives" ignores the richness that comes from learning through diverse perspectives. Bishop's framework, in fact, aligns beautifully with a knowledge-building curriculum. Consider, for example, how a Civil War soldier’s account of a significant battle might challenge students to step outside their own experience, expanding their understanding of history and empathy for others’ struggles, even if those experiences are unfamiliar. This helps develop both critical thinking and empathy—crucial aspects of a robust knowledge-building curriculum.
In sum, Bishop’s work invites us to provide students with both familiar and unfamiliar stories, recognizing that learning about the lives of others enhances our collective understanding of the world, while also enriching students’ own sense of self. This is not just about affirming what students already know but expanding their knowledge and empathy through engagement with a diverse range of texts.
This quote—"It's not what good readers do, it's what they know."—is particularly revealing. You are creating an unhelpful dichotomy in how you word this as an either/or statement. Once again, we see a deliberate attempt to downplay the science supporting strategy instruction. Instead of acknowledging their role, you pivot to a coherent curriculum rather than admit that we need strategies too, alongside to help build the knowledge. Peng’s 2024 meta-analysis identified the most effective reading strategies, including main idea identification, yet this critical finding is conspicuously absent from every knowledge-centric discussion on the topic. This is an and/both and attempts to position it otherwise frustrate and confuse teachers. Good readers need both - they need to know stuff and they need to know what do as they read. And/both.
There is a clear, coordinated effort to minimize the role of strategies alongside knowledge-building. No one is arguing against the importance of knowledge, but the dismissal of well-evidenced, effective strategies—when properly taught—is both misleading and harmful. The call for "knowledge first," "knowledge only," or "knowledge instead of strategies" ignores decades of research demonstrating that strategy instruction is a stronger causal factor in improving reading outcomes.
Why not highlight the strategies that have consistently shown strong effects in rigorous, replicated studies? Researchers like Vaughn, Elleman, and Wijekumar have produced extensive evidence on the power of strategy instruction. Instead of sidelining these findings, we should be championing a comprehensive approach that incorporates both knowledge and the explicit teaching of effective strategies—especially for students who need them the most.
I don't think Pondiscio or the others he mentions, such as Daniel Willingham, are creating a dichotomy. You are right that it might not be an either/or statement. But to say it is "and/both" isn't entirely accurate either, because of all the studies that show that teaching reading strategies are helpful, but only can go so far in making up for a lack of background knowledge. Furthermore, strategies can be taught and then move one. Teaching strategies over and over again is not helpful.
I'm thinking in particular of a brilliant example given by David Liben at the very beginning of Episode 3 of the Knowledge Matters podcast, "Know Better Do Better." You can listen or read the example on the transcript here: https://knowledgematterspodcast.buzzsprout.com/2200553/episodes/15949751-the-tail-is-wagging-the-dog-know-better-do-better?t=0
As a history teacher, I help kids learn strategies all the time: how to examine and use evidence to make an argument, figure out cause & effect, compare and contrast multiple perspectives and examine change over time. But that's not what they go home and tell their parents about. They tell their parents about the stories -- about what happened when Theodore Roosevelt invited Booker T. Washington to dinner. And that Booker T. Washington wasn't lynched on his trip north because his would-be attackers didn't think to look for him on the fancy Pullman car. They know about Pullman because they learned about the Pullman strike earlier in the year. And neither Pullman, Washington nor Roosevelt are just names to them, because I share stories with them that humanize these people and put them into historic context.
So sure, teach strategies. But then move on to what is far more interesting to kids: science, social studies and literature. Kids are excited to learn about stuff--the world, people, animals, nature. They are not interested in learning about reading strategies. So lead with the good stuff and teach the strategies to help them learn the stuff, not the other way around.
I want to come to your class.
I've also heard it stated "Teach the strategies in the SERVICE of knowledge-building," but this message (however you state it) sometimes gets lost in the frenzy of emphasizing knowledge-building because so much "skill-building" related to strategy instruction has been so awful for so long. When I teach students "paragraph shrinking" to help them "get the gist" there's a lot of "we do" related to complex text until my students become expert at the "I do." I love your Booker T. Washington story! But it reminds me of another strategy: teaching kids to separate what is important from what is interesting (or just cool). Oftentimes, the stories are memorable but not entirely germane to the substance of the text. When I taught my third graders a biomimicry unit, the stories related to the scientists were fascinating, but not something they necessarily needed to include in their summaries.
As IES goes the way of the dodo, take advantage of the practice guide for 4th - 6th (by Sharon Vaughn et al.) which says there is strong evidence for:
Routinely use a set of comprehension-building practices to help students make sense of the text
Part 3A. Build students’ world and word knowledge so they can make sense of the text
Part 3B. Consistently provide students with opportunities to ask and answer questions to better understand the text they read
Part 3C. Teach students a routine for determining the gist of a short section of text
Part 3D. Teach students to monitor their comprehension as they read
The knowledge-centric will occasionally give a nod that strategies are ok but do not give equal airtime to how to teach them well (for a calculated reason- to sell certain knowledge branded curriculum)- to the importance of correcting what Harriett describes as poorly taught strategies (she’s right). It misses the mark to say those of us worried about the attack on science and how science supports strategies are misunderstanding. We track is carefully. Here are the studies https://docs.google.com/spreadsheets/d/1Bw_JYU07S8050THEdx3eG4G061xxgvLsPpmEZvM87Zw/edit