- Home
- Daniel T Willingham
Raising Kids Who Read Page 16
Raising Kids Who Read Read online
Page 16
Here’s how we know that. Suppose you looked at fifty experiments that had taught kids reading comprehension strategies. In some experiments, kids got just a few lessons in how to use them. In other experiments, kids got more practice. You would expect that more practice would lead to bigger gains in reading. But it doesn’t. Just a few sessions—five or ten—give the same benefit as fifty. That finding, which several researchers have observed, is hard to square with the idea that reading comprehension strategies directly improve comprehension. We have this idea that comprehension is a skill, like hitting a baseball, and comprehension strategies are like things that the coach tells you: “Keep your eye on the ball,” “Your hips provide the power of the swing,” and so on. We practice these strategies, and our skill improves.
But remember how comprehension works. It depends on the particular content of sentences, so it’s not open to that kind of strategic instruction. Here’s an analogy. Suppose that reading is like putting together a piece of furniture you bought at Ikea. Like a text, furniture has parts that must fit together in a particular way, and if you do the job right, all the parts coalesce into something larger—a functional object.
Suppose you lay out all the pieces of your unassembled desk, and find strategy-like instructions (figure 9.3). These instructions don’t tell you how to actually build the piece of furniture. You need to know whether piece A is supposed to attach to piece B or C. Rather, these instructions concern what to think about when you’re executing instructions (the ones that tell you that part A attaches to part B).
Figure 9.3. Furniture assembly strategy instructions.
Source: © Joe Gough—Fotolia.
Reading comprehension strategies are similar. They tell you what to do: monitor your comprehension, relate prior knowledge to what you’re reading, relate sentences in the text to one another. They don’t tell you how to get those things done. They can’t, because how to do them depends on the particulars. Comprehension requires relating sentence A to sentence B, but I can’t give you generic instructions about how to relate them. Their relationship depends on the contents of sentence A and sentence B.
For someone who thinks that assembling furniture is merely a matter of attaching pieces to one another for a while, this big-picture overview is good advice. Likewise, if a child doesn’t appreciate that the purpose of reading is communication and that she’s meant to understand what she reads, comprehension strategy instruction is a great idea. For example, a student who has difficulty decoding may view decoding as synonymous with reading. Decoding is really taxing, so if I’m decoding, then I’m doing my job. Reading comprehension strategy instruction tells the child, “Decoding is not enough. You are supposed to understand what you’re reading. Just as you do when you listen to a story, when you read, you must relate the beginning, middle, and end.”
A New Demand: Working with Texts
The National Assessment of Educational Progress, more commonly known as the “Nation’s Report Card,” defines “basic” fourth-grade reading skill as the ability to “locate relevant information, make simple inferences, and use their understanding of the text to identify details that support a given interpretation or conclusion.” In other words, comprehension is no longer defined as understanding the text. Comprehension means being able to use the text to aid reasoning. And of course texts become longer and more complex with each grade. If that weren’t enough, upper elementary and middle school is when teachers often start to expect that students will do more reading independently at home, so that class time can be devoted to other things.
Later, in high school, students may learn that researchers in different disciplines treat texts differently. It’s not just that researchers know different things and are interested in different facets of a text. Each discipline has norms about what’s interesting and important. For example, sourcing is fundamental to a historian’s work: Who wrote this text, with what goal in mind, and for what audience? Scientists care little, if at all, about sourcing or the author’s perspective. But budding scientists must learn how scientific journal articles are structured: what goes into the methods section, what sort of speculation is permissible in the discussion section, and so forth. As students learn more about a discipline, they learn what merits special attention according to the conventions of the discipline and what is secondary (figure 9.4).
Figure 9.4. Historic letter. This is the first page of a letter from Albert Einstein to President Roosevelt about the possibility of developing atomic bombs. This letter would be read in different ways by a historian, a scientist, and a theologian.
Source: WikiMedia Commons, http://commons.wikimedia.org/wiki/File:Einstein_Szilard_p1.jpg.
“Comprehension” comes to mean different things as students get older. Initially it means no more than “understand the story.” Later, we want students to put texts to other purposes: find facts for research, memorize information for a test, or analyze the author’s technique to persuade or evoke emotion. Students who were strong readers in early grades may find themselves challenged as they had not been before by these tasks.
Most of this work happens at school, not at home, but it’s useful for parents to have it on their radar. You want to be aware that your child is taking on these new tasks and to ensure that he is getting the proper instruction and support at school. The teacher may assume that students have learned something—how to use reference materials, for example—in an earlier grade, when in fact your child never received that instruction. (That’s obviously a greater risk if you switched schools.) It’s best to be in touch with your child’s teacher so that you know what she’s doing in class and can ask how you can support this work at home.
Digital Literacy
Some education commentators have suggested that we need to think of “reading comprehension” differently, because it (along with writing and other aspects of literacy) has been so profoundly changed by the broad availability of digital technologies. The ability to create, navigate, and evaluate information on various digital platforms is generically called “digital literacy.” What should we make of this? Is the idea of “reading comprehension” outdated?
Different aspects of digital literacy need to be evaluated separately. Consider first the idea of a general tech savviness. I think there’s little doubt that exposure to and practice with digital technologies teach kids certain conventions common across these technologies: menuing systems, hierarchical file structures, and the like. This knowledge is important exactly because these conventions are respected across applications and across devices. But they are pretty easy to learn. Software is engineered to be simple to use, and kids learn this stuff rapidly. Some adults like to joke about how helpless they are in the face of these newfangled gadgets compared to kids who seem such naturals. But kids in fact vary widely in their tech knowledge, and age-related differences, when they exist, are not due to limited learning abilities on the part of oldsters; they are due to youngsters’ greater motivation and opportunities to learn from their peers.
The second aspect of digital literacy is the ability to evaluate information. The web is often praised for its effect in democratizing publishing. Twenty years ago, the owners of publishing companies were gatekeepers of information. My comprehensive knowledge of, say, rare animal species in the Pacific Northwest would remain unknown if I could not persuade the gatekeepers to publish it. Now I can publish whatever I like over the web or as an e-book and let consumers decide whether it’s valuable. That’s great, but the gatekeepers did serve a function: most had an interest in ensuring some quality control. Certainly they fulfilled this role imperfectly—falsehoods were and are published by mainstream sources—but as standing institutions, publishers are more accountable than individuals publishing websites on a lark, and their records for accuracy are easier to track. On the web, readers must take greater responsibility for evaluating the reliability of what they encounter.
In the mid-2000s the need for greater student education on
this matter gained publicity through a website describing the endangered Northwest Tree Octopus, a fictitious species said to live in trees (figure 9.5). The website deftly mimics the prose used in science textbooks (“Because of the moistness of the rainforests and specialized skin adaptations, they are able to keep from becoming desiccated for prolonged periods of time.”). Aside from the absurdity of a cephalopod living out of water, there are hints scattered throughout that it’s a hoax—for example, the octopus’s main predator is said to be the sasquatch, and the website is endorsed by the organization “Greenpeas.”
Figure 9.5. Tree octopus. A screenshot from the website describing the fictitious tree octopus.
Source: http://zapatopi.net/treeoctopus/.
Yet when researchers at the University of Connecticut asked twenty-five seventh graders to evaluate the site—students named by their schools as their most proficient online readers—every single one fell for the hoax. When they were told that it was false, most struggled to find evidence that could have told them that, and some even insisted that website was legitimate. Other research has shown that students rarely critically evaluate information they find on the web. They probably would be no more critical of a paper pamphlet on the tree octopus; the point is that there is more misinformation on the web than is generated by traditional publishers, and so kids need to be more discerning when reading there.
In the last few years, there have been greater efforts to teach students how to be critical readers of information on the web. Students have been taught techniques like evaluating the author’s credentials, tracing the domain to evaluate whether the website is commercial or originates in the education community, checking how recently the web page was updated, and looking for other websites that have linked back to the target site. Teaching students these evaluation skills is still in its infancy, but thus far it has been tough going. Some studies indicate that it helps students understand the importance of evaluating websites, but their evaluations of websites don’t actually improve.
What to Do at Home
As before, most of what I’ve encouraged you to do by way of making your home knowledge rich and your child knowledge hungry still applies. This age does bring two new concerns. First, most kids begin to spend significant amounts of time with digital devices, so we’ll evaluate what that might mean for acquiring background knowledge. Second, we need to consider strategies you might employ if you think that a lack of background knowledge is impairing your child’s reading comprehension.
Knowledge in the Digital Age
As kids move into middle school, they don’t just spend more time with digital devices; they also change what they do with them. They still watch a lot of video content, but they add video gaming, texting, and surfing the web. What are the consequences of these activities to reading and to background knowledge?
Reading Volume
One change wrought by the digital revolution is that kids are actually reading much more than they used to, even though reading is commonly thought to be in decline. A 2009 study at the University of California, San Diego, examined the number of words to which the average American is exposed per day (figure 9.6).
Figure 9.6. Average daily word consumption, by media. Note that the measure is “words,” so they might be spoken, sung, or written.
Source: From “How much information?” by R. Bohn, J. Short, Global Information Industry Center, University of California, San Diego. Data from Appendix B, © UCSD (2009).
The volume of words received by computer is enormous. And although “by computer” includes words read and words heard, when the data were collected in 2008, most Americans did not have Internet access speed adequate for video or audio streaming. Most of the words they received would have been in print. These data were collected from adults, and are now half a decade old. Still, I think it’s reasonable to suggest that kids do an enormous amount of incidental reading on digital platforms, especially in text messages, which this study did not include. So, does all this reading make them better readers?
We don’t have firm data on this question, but reading theory would predict little benefit to comprehension from this type of material. Reading improves comprehension through the acquisition of broader background knowledge, but most of what the average kid reads on screens is not content rich. It’s information within games, text messages, social network updates, and the like. But that sort of reading should (according to theory) have a positive effect on fluency. That prediction has not been tested as far as I know.
Knowledge Everywhere
Wait a minute. How can I be so sure that kids are not benefiting from this increase in reading? Doesn’t it depend on what they are reading? You can read anything online, from a Shakespeare concordance to pornographic satire of The Hunger Games series.
Although we can say with wide-eyed innocence, “Hey, they could be reading Shakespeare,” we suspect otherwise. One wag put it this way: “I possess a device, in my pocket, that is capable of accessing the entirety of information known to man. I use it to look at pictures of cats and get in arguments with strangers.” Survey data confirm that teens use computers for a relatively limited number of activities. The most common are
Social networking
Playing games
Watching videos
Instant messaging
This survey is dated (it’s from 1999), but at that time, these four activities accounted for 75 percent of teens’ computer time. Today instant messaging has been replaced by texting, to which the average teen devotes about ninety minutes daily. We can’t generalize to every child, but I don’t think it’s the case that teen use of digital devices is so varied that we cannot make any claims about its likely impact.
Well, perhaps they are not seeking out rich sources of information but find themselves exposed to such information anyway. After all, the hallmark of the digital revolution is that it has made information cheap, in the best sense of the word. You can’t help but get damp if you’re in a flood, and the Internet is a fire hose of information.
Kids could learn in this incidental way, and there’s evidence that they do, at least from certain sources. Toddlers and preschoolers who watch educational television really do learn about numbers and letters, as well as social lessons (e.g., about sharing). But overall, kids learn less from video than you’d think. Infants and toddlers seem to have a harder time learning from video than from a live person, a phenomenon established enough that it is called the “video deficit.”
Mostly, the idea that new technologies leave people more knowledgeable goes unsupported. It might be right, but at the moment, it’s unsupported. For older kids, the relationship between television viewing and academic achievement is negative, not positive; but note, that effect is carried by heavy viewers. Kids who watch only a little TV show no academic cost. (And for all kids, TV content, not just volume, matters.) More generally, kids who report being heavy users of all media (television, music, gaming, and the others) also report getting lower grades—but the relationship of grades and leisure reading is positive. Still, simple correlations like this don’t tell us much about such complex behaviors.
Do We Need to Know Less?
Kids may not choose to read Shakespeare, but they could find information about his life or plays with ease. Those of us who grew up before the digital age truly find it miraculous that we can instantly find the name of the president during the War of 1812, whether jambalaya typically includes shellfish, or how to translate European shoe sizes to American. Whatever you want to know, however obscure, you can find it, and find it almost immediately.
Given the importance I placed in chapter 1 on knowledge as a driver of reading comprehension, we might ask whether easy access to information means that digital technologies render knowledge in one’s head less important for reading (figure 9.7).
Figure 9.7. Marissa Mayer. In 2010 when she was vice president for search products and user experience at Google, Mayer wrote, “The Internet has rel
egated memorization of rote facts to mental exercise or enjoyment.”a Mayer is now president and CEO of Yahoo!
Source: Photo © Yahoo. http://commons.wikimedia.org/wiki/File:Marissa_Mayer_May_2014_%28cropped%29.jpg.
aMayer (2010).
There are three reasons that Googling information (or Yahooing it, or Binging it) is not a substitute for knowledge in your head when you are reading. First, if you recognize that you’re missing something—you realize the writer has omitted some information that you need to make an inference—it’s not always obvious which information is called for. Recall the example from chapter 1: “Trisha spilled her coffee. Dan jumped from his chair to get a rag.” If you don’t have the necessary knowledge in your head to understand why Dan jumped from his chair, you might search for “coffee” on the web. You’re going to find an enormous amount of information: where coffee is grown, social customs around the world about how it’s drunk, different ways to prepare it, and so on. That’s a lot to sort through before you deduce which property the author assumed you knew.
A second problem with tracking down information is that you don’t always know that you’re missing anything. If that happens, it will likely be in the development of the situation model. I illustrated that point with the Carol Harris/Helen Keller story.
Finally, halting reading to find a definition or bit of information is disruptive. The more often you do it, the more likely you are to lose the logical thread of what you’re reading and the more likely you are to quit reading it. So a substantial amount of knowledge needs to be known, not just findable.