Teaching the foundational skills of reading is often a lively and physical task: students clapping out the syllables in words and practicing letter sounds in chorus and teachers demonstrating the way that the mouth forms different shapes for different sounds. This year, though, it will likely look very different.
According to Education Week’s database of more than 900 districts, which is not nationally representative, 48 percent are doing all of their instruction remotely. Young students at these schools as well as those doing a mix of in-person and virtual instruction will be learning to read through screens—in virtual classrooms with their teachers, working on computer programs and apps, or through some combination of the two.
There’s a robust evidence base for how to teach children to read in person: Decades of research has shown that explicitly teaching students how letters correspond to spoken sounds—and teaching phonics—is the most effective way to help them learn to decode words. But there’s little evidence on how this best practice should be translated to the remote environment.
It is clear, though, that many teachers will be using different materials than they do in the classroom—finding resources that can support live teaching over Zoom, or relying more on digital reading programs.
Many companies offering core reading curricula have updated and expanded their digital offerings during remote learning. Schools and teachers should take the same steps to evaluate these resources that they would print materials, experts say.
Prior Education Week reporting has shown that some of the most popular curricula and interventions used in classrooms don’t teach letter-sound connections in a systematic way, raising the possibility that some students who are still learning the alphabetic code may be left with gaps in their understanding.
There are also adaptive, digital programs that students can work through independently. Some of these programs do align with evidence-based methods, said David Liben, a literacy expert and advisor to Student Achievement Partners, a nonprofit consulting group. They follow a scope and sequence and are systematic.
Still, he said, they haven’t been designed to be used as core instruction. If students aren’t also getting strong foundational skills instruction from a teacher, “then you’re not going to get good results from the supplementary program,” he said.
Most also haven’t been evaluated in a home-based setting, without a teacher present. Other tools teachers might use, like apps and digital books, vary widely in quality, researchers say.
When evaluating how to use some of these tools, and in what combination with live online teaching, “there are not hard and fast rules,” said Devin Kearns, an associate professor of special education at the University of Connecticut.
“This is where I would say you really need a teacher instead of an app, even if a teacher is using an app. … Teachers have a unique knowledge of kids—the specific kids, the environment—and a lot of skill in responding to immediate student needs that the programs still don’t have.”
Digital Programs and Apps
When teachers of young children do have the opportunity for some live interaction with students—over videoconference, for example—researchers suggest sticking to the kind of explicit, systematic instruction that has been proven effective for teaching how to read words in an in-person setting.
But many teachers won’t have the same amount of face-to-face time that they’ve had in previous years, and schools say they’re relying more on digital tools. In a nationally representative EdWeek Research Center survey, 63 percent of educators involved in K-2 reading said that they or the teachers they work with are using tech-based reading programs somewhat or much more frequently than they were before the school shutdowns.
The survey also asked which core and supplemental programs respondents had used to teach students how to read during remote learning. Two of the most popular resources were digital programs that target lessons to students based on the specific skills they need practice with: Lexia and iReady.
Both programs offer practice in phonemic awareness and phonics, as experts recommend, and collect data on student performance that teachers can use to tailor instruction outside of the platform. iReady also offers an assessment that is normed to performance on some state standardized tests. Lexia has conducted its own, peer-reviewed research on Lexia Core5, the company’s reading product for students in grades pre-K-5, which has shown positive effects on early reading skills. Still, there are few independent studies of the program. Research on iReady’s instructional program has found that students who use it perform better on the iReady assessment, but hasn’t evaluated whether it raises student achievement on other measures.
In general, most research on technology-based programs for teaching early reading has looked at how effective these programs are in combination with classroom teaching.
One 2013 review from education researchers Alan C.K. Cheung of the Chinese University of Hong Kong and Robert Slavin of Johns Hopkins University looked at 20 studies spanning students in grades 1-6. The strongest effect sizes came from studies in which teachers worked with students in smallgroup settings, using technology that was closely aligned to their curriculum. Cheung and Slavin found a smaller positive effect for stand-alone supplemental programs, like Lexia. On the whole, though, the average effect size across all studies was much stronger for younger students (grades 1-3) than older students.
Other papers have also made the case that teacher implementation, unsurprisingly, plays a big role in reading program effectiveness. Two meta analyses, from 2012 and 2014, both found that programs that included teacher training and support were more effective than those that did not. (These papers included studies with a range of K-12 students, though, not just young learners.)
It’s hard to know how effective digital reading programs will be if students are working through them at home. These programs are designed to be a part of, or a complement to, in-person instruction.
The effectiveness could be compromised, and there’s also the potential for students to feel isolated and withdraw from learning, said Timothy Shanahan, a professor emeritus at the University of Illinois at Chicago and an author of the National Reading Panel report. “Trying to stretch these [programs] to be more than they are, more than intended, might be OK, but I’m worried about it,” he said.
Liz Brooke, the chief learning officer for Lexia, said that the company doesn’t recommend kids spend more time on the program at home than they would at school—the suggested limit for early readers is still 60 minutes a week, she said. Still, Brooke noted that usage went up this spring from previous years.
For stand-alone mobile apps, the research base is thinner. Studies have shown that it’s possible these tools can help children improve in foundational skills, like alphabet knowledge and word reading. But there are many choices, and quality varies greatly, Kearns said.
Evaluating Digital Reading Materials
So how can teachers evaluate a program or app, or decide which parts of it to use?
Kearns suggests that they start by looking for the core instructional components that they would expect in any in-person curriculum.
“I would look at the program … and say, does this include explicit, systematic phonics instruction? Do students learn individual letter sounds? Are they organized in a logical scope and sequence? Do students receive a lot of practice? Does it move from words to sentences to text?” Kearns said.
But even a well-designed digital program that follows a structured sequence can’t replicate the range of activities and feedback that teachers can provide face-to-face. Researchers identified three areas where apps and computer programs can fall short of in-person instruction, and offered suggestions for what to prioritize during synchronous teaching time.
1. Types of Questions and Possibilities for Feedback
Computers are only able to assess certain types of knowledge. For example, there are different dimensions to “knowing” a letter, said Holly Lane, the director of the University of Florida Literacy Institute. A teacher could show a student several letters and ask, which one is the letter “a”? Or, the teacher could also show a student the written letter “a” and ask students to say the letter’s name.
These two questions are assessing different kinds of knowledge. But Lane said that digital programs tend to lean more heavily on the former—presenting two, or a series of options, for students to choose from. Given that, it’s also possible that students might be able to advance with lucky guesses and not get enough practice with skills that they’re still learning.
A teacher also has more options available for correcting a student mistake and figuring out why the child is making the error, said Natalia Kucirkova, a professor of early-childhood education who studies digital books at the University of Stavanger in Norway and a professor reading and children’s development at the Open University in the UK. Say a student uses the wrong /a/ sound in the word “cat.” A computer could note the answer as wrong, Kucirkova said, but it wouldn’t necessarily be able to explain why “cat” has a short “a” as well as a teacher could.
When students are still learning a new skill, it’s important that they have time to practice in front of a live teacher, not just with a program, Kearns said. “Any activity that is better when teachers provide feedback, or when teachers listen to students and adjust instruction in the moment based on student response, that’s something that teachers are really essential for.”
2. Hearing Kids Read and Pronouncing Words
One of the reasons that digital programs rely on multiple-choice questions, Lane said, is that they can’t listen to kids pronounce words in the same way a teacher can.
“We are getting toward a point where you can have the computer listen to kids read, and the computer can determine whether the kid says the right word or not,” Kearns said, but most programs aren’t there yet.
Computers also expose students to a smaller range of word pronunciations. In a classroom, kids hear all of the slight variations in how their peers and their teachers say the same word; in a digital program, they often only get one example, said Kucirkova.
Lane raised another potential concern: Some digital programs include playback pronunciations for letters that are slightly off. For example, she said, the recorded voice sometimes pronounces the sound for the letter p as “puh,” exaggerating the sound to make it easier to hear. But adding the “uh” sound after /p/ distorts the letter’s actual sound, she said, and can make it harder for students to understand how to blend “p” into a word.
3. Differentiation Weaknesses in Online Reading Programs
While some digital reading programs call themselves “adaptive,” Kearns said, most don’t respond in the moment, moving a child forward or backward based on the answers to individual questions. Instead, they move students on at the end of whole units or sections of the program.
Even if teachers are using digital programs, it’s important that they’re still involved in initial diagnostic and continuous assessment practices, said Lane. That way, they can make sure that students aren’t starting with skills that they’ve already mastered, or haven’t skipped over ones that they need more practice with.
It’s also important to note that some programs don’t give teachers this choice. A 2015 analysis from the Joan Ganz Cooney Center looked at 183 literacy apps in popular app stores, and found that only 17 percent allowed users to select the difficulty level of the program.
4. Digital Books: Decodable vs. Leveled
In addition to tools for instruction and assessment, many teachers are looking for ways to give students virtual access to a classroom library.
In a traditional school setting, experts suggest, students who are just learning how to read should practice in decodable books. These short texts are written with a high proportion of words that are phonetically regular—meaning they follow common sound-spelling rules—and mostly include words with phonics patterns that children have already learned.
More commonly given to young children, though, are leveled texts. These are books categorized by their perceived difficulty. At the lowest levels, for kids who are just learning to read, these books often feature repetitive text patterns and literal illustrations. While they may include phonics patterns that children have already learned, they aren’t specifically designed to do so.
Some research has shown that which kind of text students are exposed to more often—decodable or leveled—can affect how they try to tackle words. Decodable text trains students to sound out words when they read, while predictable leveled text can encourage them to rely on other cues.
Still, decodable books are just one part of the diverse text diet that young students should get, researchers say. Kids should also be listening to stories read aloud and talking about them, which builds their vocabulary, knowledge, and comprehension skills. And they should have access to authentic texts that they can try to tackle as they build their decoding skills.
Outside of these general best practices for text selection, there are specific criteria to look for when judging the quality of digital books.
Digital books for children often come with more features than the standard adult e-reader. Many give the option to hear the story read to you, or to click on specific words and look up their definitions in kid-friendly dictionaries, said Kucirkova. These kinds of scaffolds can be helpful, she said, allowing children who are still developing their decoding skills, or are learning how to read English, to engage with complex stories.
But other technological enhancements are more like “bells and whistles,” Kucirkova said. Activities that take children’s attention away from the story—a game, or a drawing exercise, for example—can lower their ability to comprehend what they read. “It has to do with the cognitive load of the child. It becomes too much to process,” she said.
The International Collective of Research and Design in Children’s Books, of which Kucirkova is a member, offers a best practice design framework that has research-based guidelines for creating and identifying high-quality books.
In the EdWeek Research Center survey, two of the materials educators were most likely to say they were using to teach reading online were Epic! and Raz-Kids. Both of those essentially function like online libraries.
Epic! has decodable books and leveled readers both available, as well as other trade books. The site also offers audiobooks.
Raz-Kids is a leveled reading program that uses digital books. Teachers can assign books to students by reading level or on certain topics. The program also includes digital assessments: comprehension quizzes, rubrics that gauge a student’s ability to retell the story they read, and running records scored using the three-cueing system. (Raz-Kids has recently received criticism for books alleged to perpetuate racial stereotypes. Lisa O’Masta, the president of Learning A-Z, which publishes Raz-Kids, said in an interview that these materials have since been altered or removed, and that the company has expanded its review process.)
Teacher knowledge and discernment is important in selecting books, no matter the source, said Kucirkova. Young children can also get more out of the reading experience, she said, if adults in the home are able to read with them.
“It is the combination of the human and the digital scaffolding that makes the biggest difference for the child’s learning,” Kucirkova said.