Advances in the digital technologies available to support learning are among the most dramatic developments since the publication of HPL I.1 Digital technologies can support learners in meeting a wide range of goals in different contexts, for example:
- A first-grade teacher concludes that her students are disengaged when working on drill and practice of mathematical operations using digital workbooks. They also rarely complete the associated homework assignments. The teacher introduces a program that targets the same operations in the context of a game. Not only are many of the students excited about the game, but they often choose to play it and have improved their mathematical skills.
- A manager of an aircraft repair shop receives reports of errors made by workers. The software used to train the workers involves reading and memorizing procedures for troubleshooting, replacing, and repairing the devices they are responsible for. The manager believes that the workers need a deeper understanding (mental model) of the device mechanisms and purchases an intelligent tutoring system that offers individualized instruction and virtual reality simulations; it also explains device mechanisms and common misconceptions. Use of the system results in a significant reduction in errors.
- An individual went bankrupt after his business failed during a major recession. He found a job in a rural area that required a number of new skills, including knowledge of agriculture and statistics. He has completed free online courses to fill these knowledge gaps, has earned a dozen certificates, and now has a leadership role in his new field.
These examples suggest the range of ways technologies can support learning in varied sociocultural contexts. The game in the first example was appropriate for repetitive drill and practice on numerical operations, whereas the intelligent tutoring system was needed to acquire deep mental models of aircraft devices. The free online courses supported self-regulated learning by the individual who needed to change fields. As these examples suggest, learning technology is most useful when it is designed to meet specific needs and contexts.
HPL I noted that technologies may be used to: (1) incorporate real-world problem solving into classroom curricula; (2) scaffold students’ learning; (3) provide students and teachers with more opportunities for feedback, reflection, and revision; (4) build local and global communities of individuals who are invested and interested in learning; and (5) expand opportunities for teachers’ learning. Since that report was published, new technologies have been developed and researchers have expanded understanding of how digital technology can most effectively be used to foster learning.
In this chapter, we discuss ways to align learning technologies with goals for learning, drawing on research on new technologies that have shown promise for stimulating active learning and supporting learning in school and in the workforce. We also discuss the use of technologies for supporting older learners and close with a discussion of access to learning opportunities.
Learning technologies open up significant possibilities for supporting learners. Researchers in the field use the term affordances to refer to opportunities that a technology makes possible related to learning and instruction (Collins et al., 2000). In this section we first examine the nature of the affordances of learning technologies and then explore research on how technology can support several aspects of learning.
An affordance has been defined as a feature or property of an object that makes possible a particular way of relating to the object for the person who
uses it (Gibson, 1979; Norman, 2013). For example, a door knob affords its users a way to twist and push, whereas a length of string affords users a means to pull and tie. Contemporary digital environments have features such as multimedia displays with texts, pictures, diagrams, visual highlighting, sound, spoken messages, and input channels (clicking, touching) for entering information that can afford important learning opportunities for users. Box 8-1 summarizes information delivery and input features and other technological
We pointed in Chapters 3 and 4 to types of learning that require a significant amount of practice and repetition of items (e.g., perceptual patterns, words, concepts, facts, rules, procedures), including perceptual and motor learning and some kinds of memory learning. We have noted that such learning
is most lasting when it takes place in a variety of contexts and on a schedule that is distributed over time (Koedinger et al., 2012; Pashler et al., 2007). Interactivity and feedback are two affordances that are particularly helpful for supporting these types of learning.
Traditional computer-based instruction, or what used to be called “computer-assisted instruction,” provides interactivity and feedback. For example, there is a mature industry that provides computer-based vocabulary instruction in which the computer displays a picture and two to four words. The learner selects the word that names the picture and receives immediate feedback (correct versus incorrect). The computer could present thousands of trials with this simple procedure, following particular schedules of item presentation with interactivity and feedback. These training trials have been used in classrooms and labs and to support homework outside of class. The training can be accessible throughout the day if it is available on a mobile device.
One drawback to this type of computer-based instruction is that some learners may lose motivation when using a repetitive format. One way to enhance motivation is to add the affordance of adaptivity. For example, the FaCT system is adaptable in that it offers the learner optimally spaced training trials rather than massed training (see Chapter 4 for discussion of spacing) and stops the training on a particular fact if the learner performs correctly on it three times (Pavlik et al., 2016). This approach can result in more efficient learning because learners do not waste time studying facts they already know.
Another approach is to gamify the learning by adding expanded feedback (e.g., total score points) and communication with other people in the form of leader boards and competition with partners (Clark et al., 2014; Tobias and Fletcher, 2011; Wouters et al., 2013). Yet another way to sustain motivation is to allow learners to select topics that interest them. Some topics may be very important but unappealing, so a possible downside of allowing too much choice is the risk that learners never get around to acquiring critical knowledge or skills.
People need more than the foundations of literacy, numeracy, and other basic skills to handle the complex technologies, social systems, and subject matter typical of 21st century tasks (Autor and Price, 2013; Carnevale and Smith, 2013; Griffin et al., 2012; National Research Council, 2012b). Deeper learning involves understanding complex concepts and systems and is manifested in, for example, the use and construction of models (see Chapter 3), the ability to integrate information from multiple documents and experiences (Wiley et al., 2009), and the ability to explain correct versus incorrect system behavior (VanLehn et al., 2016). Deeper learning is needed for complex
problem solving, reasoning, inferential thinking, and transfer of knowledge to new situations (Hattie and Donoghue, 2016).
The technology affordances of linked representations and open-ended learner input are particularly important for this type of learning, as are the interactivity, feedback, and adaptivity affordances of traditional computer-based training. The value of technology for representing a situation from multiple linked perspectives is evident in the example of helping learners understand a system, such as an electronic circuit. An intelligent technology can allow a learner quick access to perspectives, including a picture of the circuit as it appears in a device, a functional diagram of the components and connections, descriptions of the properties of each component, formulas that specify quantitative laws (e.g., Ohm’s and Kirchoff’s laws), explanations of device behavior, and the simulated behavior of the circuit as a whole when one component in the circuit is modified (see computer systems developed by Dzikovska and colleagues  and Swartout and colleagues ). The quick access will allow the learner to link these elements. Open-ended learner input is also important for conceptual learning with system models. For example, much can be ascertained about the state of a learner’s system model and misconceptions about electronics by asking the learner to explain her reasoning through natural-language tutorial dialogue (Graesser, 2016; VanLehn et al., 2007) or to create a circuit to achieve a specified function (VanLehn et al., 2016).
Intelligent tutoring systems can also support deep learning with models (Sottilare et al., 2014; VanLehn, 2011), as has been demonstrated for topics in algebra, geometry, programming languages, engineering, and the sciences. Noteworthy examples in mathematics are the Cognitive Tutors (Anderson et al., 1995; Ritter et al., 2007) and ALEKS (Doignon and Falmagne, 1999; Hu et al., 2012), which have been scaled up for use in thousands of schools. Intelligent tutoring systems have been widely used and have produced impressive learning gains in the areas of digital literacy (Kulik and Fletcher, 2016) and information technology (Mitrovic et al., 2007).
Intelligent tutoring system environments have also shown promise in domains that have strong verbal demands. Such tutoring tools have included open-ended learner input and the ability to communicate with other people, in addition to most of the other affordances. For example, AutoTutor (Graesser, 2016; Graesser et al., 2014; Nye et al., 2014) uses natural language conversations to assist students in learning about a variety of topics. AutoTutor is associated with learning gains in both physics (VanLehn et al., 2007) and computer literacy (Graesser et al., 2004) for college students, beyond the gains associated with reading a textbook for an equivalent amount of time. The agent is a talking head that speaks, points, gestures, and exhibits facial expressions. The learning gains from natural language interactions have been strongest for underachieving college students and for tests that tap deeper inferential reason-
ing rather than shallow knowledge. However, the research also suggests that conversational interactions with AutoTutor are not ideal for high-achieving college students, who tend to be more autonomous and self-regulated learners, or for use in simulation environments that are intended to push the student to acquire very precise models of the subject matter. AutoTutor is also not the best choice for perceptual, motor, and memory-based learning.
Intelligent tutoring systems have been developed for a wide range of subject matters and proficiencies and have benefited learners in schools, universities, and the workforce. Hundreds of studies have shown the effectiveness of intelligent tutoring systems in promoting deeper learning for some populations of learners on core literacy and numeracy skills, complex STEM topics, and 21st century skills (Kulik and Fletcher, 2016). However, two issues related to implementation have been noted. First, the systems are expensive to build, so using them on a large scale can be a challenge for schools, universities, and workforce programs with limited budgets. Developers of these systems are exploring ways to develop content more quickly and cheaply, as in the U.S. Army’s Generalized Intelligent Framework for Tutoring (Sottilare et al., 2014). Second, like any classroom intervention, intelligent tutoring systems need to be integrated adequately into teacher training and curricula in order to have an impact (Dynarsky et al., 2007).
The ability to work effectively in teams is among the 21st century learning objectives that have been identified in a number of venues because of its critical importance in the workplace (National Research Council, 2012b; OECD, 2013). Technologies offer many possibilities for fostering the skills of collaborative and cooperative learning, for example by supporting members of a group in seeking common ground, explaining their ideas, and understanding each other’s points of view; all of these processes are associated with successful collaborative problem solving as well as model learning (Chi, 2009; Dillenbourg and Traum, 2006). (See Chapter 7 for further discussion of collaborative learning.)
Collaborative learning can be distinguished from cooperative learning (Dillenbourg et al., 1996; Hesse et al., 2015). Collaborative learning requires interdependency, wherein group members work together to plan and organize joint activities to complete a task or solve a problem. The action of each person builds on the actions of others, and an action of one person may be taken up or completed by others in the group. In contrast, cooperative learning involves breaking a task into pieces: group members work separately, although they may coordinate activities that proceed in parallel. The completed pieces are assembled by the group (Hesse et al., 2015).
A variety of general-purpose tools have become available in the past 15
years to support communication and collaboration. These are tools that allow users to, for example, (a) add, change, or delete content with a Web browser on the Internet (such as on a “Wiki Website” or a “Wiki”); (b) upload shared/shareable word processing and spreadsheet files so that others may access, comment on, and edit them (e.g., Google docs); (c) make free voice and video calls; (d) conduct online meetings, with messaging services designed for group use (e.g., Google Hangouts, What’s App); (e) store and share electronic files on cloud-based facilities (e.g., Dropbox); or (f) participate in social media (e.g., Facebook, Twitter, and Instagram). Many of these tools are free or very low in cost.
Learning technologies have been designed to promote deeper conceptual learning as part of group collaboration. Two examples for which the developers have shown positive effects are described in Box 8-2. However, the availability of communication technologies for cooperation and collaboration does not necessarily translate into learning gains. For example, Reich and colleagues (2012) studied the use of Wikis in kindergarten through twelfth grade (K-12) classrooms by extracting a random sample of Wikis from a popular site that provides free hosting for education-related Wikis. They assessed the Wikis’ development and usage patterns to see whether the students were using them in collaborative knowledge building (and other skills). Nearly three-quarters of the Wikis showed no evidence of student-created content, and only 1 percent featured multimedia content created collaboratively by students. Equally discouraging was their finding that content created by students, as opposed to teachers, was more common among schools serving high-income students than among schools serving less-affluent populations.
Several computer technologies have been developed to train learners to acquire metacognitive and self-regulated learning strategies. Two examples that have shown promise in improving these types of learning are MetaTutor and iDrive.
MetaTutor (Azevedo et al., 2010) is designed to promote self-regulated learning in topics in biology, such as the circulatory system and the digestive system, in a hypermedia learning environment. It uses conversational agents to train students on 13 strategies, such as taking notes, drawing tables or diagrams, re-reading, and making inferences that theory suggests are important for self-regulated learning (Azevedo and Cromley, 2004). Initial studies have shown some positive impacts but not for all learning strategies. One reason may be that the instruction was delivered using a standard script; individualized training adapted to learners may be more effective.
The tool iDRIVE (Instruction with Deep-level Reasoning questions In Vicarious Environments) trains learners to ask deep questions in a STEM
context by using computer agents, representing a learner and a teacher, that engage in dialogues in which they model discussion of deep reasoning questions (Craig et al., 2006; Gholson et al., 2009). The student agent asks a series of deep questions about the science content (e.g., who, how, what-if, and what-if-not questions), and the teacher agent immediately answers each ques-
These technologies have two of the important affordances for learning described earlier. They give the learner choice, which seems to optimize motivation, and allow him to communicate with other people, which is especially productive when learners are just beginning to develop self-regulation strategies. However, such approaches have had mixed success, and it usually takes many hours of training with many examples for learners to show appreciable progress (Azevedo et al., 2010; Craig et al., 2006; Gholson et al., 2009).
We have pointed to the importance of stimulating active student learning rather than merely delivering information to the student through books and lectures (see Chapters 5 and 7). Digital technologies offer a variety of possibilities for stimulating and engaging learners.
Games are known to capture the attention of players for hours, as the players actively participate for competition or other forms of pleasure. Social media also shares these benefits. It is possible for designers of learning technologies to capitalize on these phenomena and leverage social engagement for academic learning.
Some games were not originally designed with the goal of enhancing academic learning, but case studies have found that they nevertheless provide opportunities for learning and identity formation that can spill over into other aspects of life. These findings have spawned efforts to use technologies such as digital games, social media, and online affinity groups to engage students for academic purposes (Gee, 2009). In other cases, games have been designed specifically to support learning of academic content and skills (O’Neil and Perez, 2008; Shute and Ventura, 2013; Tobias and Fletcher, 2011). Several such online games have been used at scale in both afterschool and classroom settings; examples include Atlantis, Civilization, Crystal Island, Minecraft, Sim City, and Whyville (Dawley and Dede, 2014).
An enthusiastic community of researchers of “serious” games has argued that games have educational benefits because they foster sustained engagement in learning, but a review of this work did not support the claim that skills learned in game playing improve cognition, lead to better performance on cognitive skill tests, or improve cognition (Mayer, 2016). Nevertheless, games may be more effective than alternative approaches for some specific
categories of learning outcomes. For example, “shooter” type games have resulted in transferable learning of perceptual attention skills (Mayer, 2016).
More relevant for schooling are the recent reviews of serious games that target specific academic content. A large number of quantitative studies (Clark et al., 2014; Tobias and Fletcher, 2011; Wouters and Van Oostendorp, 2017; Wouters et al., 2013) show a moderate advantage for games over other instructional approaches in fostering knowledge in science, mathematics, and literacy, as well as in promoting productive habits of mind such as intellectual openness, conscientiousness, and positive self-evaluation.
Some researchers have suggested that video games are inherently engaging and motivating to people (Prensky, 2006; Squire, 2011) and that research on video games can provide insights into the design of educational environments (Gee, 2003; Squire, 2011). Malone (1981) argued, for instance, that computer games are intrinsically motivating because they can provide optimal challenge and fantasy, while stimulating curiosity. Malone and Lepper (1987) expanded on the motivating factors of computer games by adding that such games give users a sense of control because their actions affect game outcomes. Gee (2003) identified a taxonomy of motivational factors that could be used to design video games.
However, very little empirical evidence supports these claims (Zusho et al., 2014). Although some studies have linked video game playing to motivation, this possible relationship has not been explored in educational settings. Further, the literature on adult populations (college students and other adults) suggests that users play video games for a variety of cognitive, affective, and social reasons. For instance, such games may satisfy psychological needs for competence, autonomy, and relatedness, which are associated with intrinsic motivation (see Chapter 6), but the applicability of such findings to K-12 populations is unknown (Zusho et al., 2014). Further, it cannot be assumed that gaming, or technology in general, would be inherently motivating to all learners. Whether technology is motivating to people is likely to depend on the learner, the task, and the learning context.
The entertainment industry has established the practice of connecting television shows or movies with social media sites, online games, and products based on favorite characters. Educators have found opportunities in this phenomenon for linking education and training programs to popular stories, personalities, and characters (Jenkins et al., 2006) to encourage what some refer to as transmedia learning, a “scalable system of messages representing a narrative or core experience that unfolds from the use of multiple media, emotionally engaging learners by involving them personally in the story” (Raybourn, 2014, p. 471). For example, the U.S. Army has used transmedia
campaigns that include online games and social messaging for training in cultural literacy.
The U.S. Department of Education’s Ready to Learn initiative promoted the development and evaluation of transmedia learning experiences for children ages 2 to 8. Under this program, the Corporation for Public Broadcasting developed PBS Kids Lab, an online portal containing collections of games featuring popular characters from Public Broadcasting System children’s shows (such as Sid the Science Kid and Curious George). The games can be played on computers, smartphones, electronic tablets, or smartboards. The PBS Kids Lab Website also helps users link game content related to mathematics and literacy curricula to activities for home, school, or afterschool settings (Herr-Stephenson et al., 2013). For example, designers of the Ready to Learn transmedia experiences have built videos, games, and digital device applications (apps) to support model learning by stimulating discussions between young children and their caretakers and helping children formulate questions and express their ideas (Mihalca and Miclea, 2007). Ready to Learn transmedia interventions that combine media-based and nonmedia activities into coherent curriculum units have shown positive effects on early reading and mathematics skills for preschoolers from homes with low income (Pasnik and Llorente, 2013; Pasnik et al., 2015; Penuel et al., 2012).
Historically, the use of technology with young children has been controversial, largely because of concerns about possible negative effects of extensive screen-viewing time on children’s development (American Academy of Pediatrics, 1999). More recently, organizations concerned with young children’s health and well-being have taken the position that technology can be designed and used in developmentally appropriate ways that enhance learning (American Academy of Pediatrics, 2015; National Association for the Education of Young Children and Fred Rogers Center, 2012). The American Academy of Pediatrics (2016) has recommended that parents and caregivers develop a family media plan that considers the health, education, and entertainment needs of each child and the entire family. It has offered the following age-based guidelines:
- For children younger than 18 months, avoid use of screen media (with the exception of video-chatting).
- For children 18 to 24 months, parents may introduce digital media. However, they should choose high-quality programming and watch with their children, in order to engage with them and help them understand what they are seeing.
- For children ages 2 to 5, viewing (of high-quality programs) should be limited to less than 1 hour per day. Parents should begin to help their children understand how the material applies to the world around them.
- For children ages 6 and older, parents should place consistent limits on the time spent using media and on the types of media; they should ensure that media use does not take the place of adequate sleep, physical activity, and other behaviors essential to health.
Educators view digital technology as a mixed blessing for academic learning. For example, writing teachers report that online activity has helped equip teen learners to understand multiple points of view, but they worry that the informal style of text messages and Internet posts has crept into students’ scholastic writing and that common practices such as “retweeting” and “copy and paste” have desensitized students to the seriousness of plagiarism (Purcell et al., 2013). Another concern is that becoming accustomed to skimming short snippets of online content may reduce students’ willingness to read and ponder longer text (Purcell et al., 2013). At present, there is little experimental research that sheds light on whether and how online communication skills and habits transfer to academic settings.
The Internet allows people without programming skills to create and post content to be shared with millions of people. Consequently, people can create content, collaborate, and critique the ideas and works of others relating to any topic one can imagine. Moreover, a learner can connect quickly with a small community scattered geographically around the world to become knowledgeable about a very specialized topic and develop real expertise.
Studies of informal learning communities, such as those engaged in multiuser online games, suggest that people go through stages in their development as online creators and producers (Dawley, 2009; Kafai, 2010). At first, learners identify relevant social networks within and surrounding the virtual world that can serve as a resource for their learning. The learners lurk in the virtual worlds and observe more experienced players and the cultural norms and rules for participation. As they become more comfortable with the learning context, the learners contribute small amounts of information or time to the network. As they become more experienced and knowledgeable, they create their own material, perhaps modifying some aspects of the digital environment or making elaborations to a game. In the final stage, they lead, which includes mentoring new learners or managing networks they belong to. In this process, novice players often receive explicit mentoring or tips from fellow players (Shaffer, 2007; Shaffer et al., 2009). With gains in expertise, a player also can gain recognition from fellow players, which may also have positive effects.
For example, a few studies suggest that online learning activity can play a role in the development of a learner’s identity, self-concept, and motivation to
learn (Ito et al., 2009; Lemke et al., 2015). A review of programs that provide media-rich experiences after school indicates that such activities contribute to a student’s social and emotional growth; persistence in the face of obstacles; and skills that support collaboration, provision of mutual support, and inquiry (Lemke et al., 2015).
Wikipedia, the online free encyclopedia, and YouTube, a video- and music-sharing Website, are two examples of online innovations that have blurred the boundary between teacher or expert and learner. In 2013, Wikipedia contained more than 4 million entries in English; it is available in 285 other languages. Authorship of new entries, review, fact checking, and content editing are provided primarily by volunteers, supported by a surprisingly small number of expert editors. YouTube offers a platform for amateurs to develop free learning apps and other resources. Many ventures, such as Khan Academy, which was created by Sal Khan to tutor his young cousin in mathematics, first developed out of altruism or simply as a way to share an interest with others but have evolved into successful companies or nonprofit organizations. Research on the impact of such innovations on learning is still needed, but it is not yet clear how data that would allow for an assessment of their impact can be collected in an environment where the producing and using communities emerge over time with little control and coordination.
Makers are people who engage in building and creating. They use their hands to assemble, build, mold, or modify a physical object. Although the popularity of “making” first arose outside of formal education, making has become increasingly prevalent in formal learning. In universities, making is ingrained in the teaching of engineering, and many institutions have invested significant resources in creating makerspaces to support making activities. Makerspaces are physical spaces (e.g., a room or an entire building) where people come together to share resources, knowledge, and equipment to engage in making. Makerspaces may, for example, have tools and machines for use in welding, fabricating, crafting, three-dimensional printing, laser cutting, molding, casting, and sculpting (Barrett et al., 2015; Jordan and Lande, 2014). Makerspaces thus introduce the technology of tools used to build a physical object; these tools create experiences that contribute to their users’ understanding of how objects are assembled and how they work.
Making is a form of active learning because it is experiential and engages students in developing their own understanding of a domain through doing. Active learning strategies are generally understood to be student-centered, inquiry-based instructional approaches (Kuh, 2008). Although research on making and educational outcomes has only just begun (Jordan and Lande, 2014), the results to date point to the benefits of active, inquiry-based experi-
Digital versions of making are beginning to flourish. Informally, computer clubhouses are places for students to meet after school to develop computer programs using easy-to-learn computer languages such as Scratch. Other popular digital-making activities include developing wearables, such as jewelry or t-shirts with flashing messages. Digital making is also finding its way into schools. For example, at Design Tech High School in San Mateo, California, students engage in projects in which they identify a problem (such as lighting a campsite at night) and then use Raspberry Pi software and simple peripherals to design and prototype a solution. Wearable technology projects use Flora microcontrollers, conductive materials, sensors, and actuators in designs that respond to student-generated problems. In both cases, students’ design work is supported by industry mentors working with teachers in the makerspace.
Another new area of active research, embodied cognition, has become closely intertwined with digital technology advances. Embodied cognition is the idea that cognition is shaped by every aspect of an organism’s experience, including the bodily system and ways the body interacts with its environment (see Yannier et al., 2016). SMALLab is an example of a technological application of embodied cognition that was designed as a mixed-reality2 environment for student-centered learning. Students move within a 15 × 15 foot space equipped with a vision-based object tracking system, a top-mounted visual projection system, speakers for surround sound, and (in some applications) glow balls that students can hold or toss. A series of studies conducted using SMALLab in high school classrooms showed positive results for learning about geological layers, chemical titration, and disease transmission, in comparison to instruction without this approach (Birchfield and Johnson-Glenberg, 2010).
The military and corporate sectors have invested resources to develop and test sophisticated embodied-cognition digital technologies not available in typical K-12 and college environments. These capabilities are displayed at the annual Interservice/Industry Training, Simulation and Education Conference.3 Immersive games and simulation environments are designed to help soldiers improve in several areas that include marksmanship; sensitivity to hazardous signals in combat situations; discharge of weapons under appropriate conditions; and performance on tasks that tap perceptual, motor, memory, and
2 In mixed-reality environments, real and virtual worlds are merged (Milgram and Kishino, 1994). For example, graphics (or other digital components) are projected on a floor or wall and are merged with real-world tangible objects such as trackable handheld wands.
basic levels of cognition. Immersive environments also have been developed to train soldiers on equipment maintenance, troubleshooting and repair, and other tasks that require reasoning and more thoughtful deliberation. The technologies have included mixed-reality environments with conversational agents and avatars for the learning of language, social interactions, and collaborations that are culturally appropriate (Johnson and Valente, 2009; Swartout et al., 2013). For example, Figure 8-1 describes the Tactical Language and Culture Training System (TLCTS), which has been used by more than 40, 000 learners, mostly in the military. TLCTS is among the few systems that have been
assessed on measures of learning, engagement, and learner impressions. The impact of most embodied-cognition digital technologies is difficult to assess because the results typically are not reported outside the business and military environments where they are used.
Another new technology that can stimulate active learning is the computerized conversational agent. Digital agents are designed to engage the learner in dialogues that promote reasoning, social interaction, conscious deliberation, and model learning (D’Mello et al., 2014; Lehman et al., 2013). The design allows students to engage in a three-way conversation known as a trialogue that includes two computer agents and the student, taking on different roles (e.g., two peers with an expert or a peer with two experts). Figure 8-2 shows two agents on the screen interacting with a human in a trialogue. The results of this particular test of trialogue with conversational agents showed deeper,
conceptual learning by the student in conditions where the two agents disagreed, especially for students who experienced confusion.
Research suggests that agent technologies can stimulate active learning by means of several features. A single agent can serve as a tutor (such as AutoTutor; see Graesser, 2016) or as a peer of the learner-player. An ensemble of agents can set up a variety of social situations, which may, for example, (a) model desired behavior and social interaction, (b) stage arguments that invoke reasoning, or (c) pull the learner-player into active contributions through actions and social communication (Graesser et al., 2014). Computerized agent technologies can implement pedagogical approaches with a degree of fidelity that may be difficult or impossible with human agents who are not experts in the pedagogical approach.
Learning technology can be used to support instruction, and this section explores the evidence for this technology’s capability to support three instructional goals: linking formal and informal learning to improve learners’ outcomes, orchestrating the complexities of instruction in the classroom, and developing students’ writing through interactivity and feedback.
Researchers have explored ways educators might recruit the vast bodies of informal knowledge learners acquire from their cultural contexts and self-directed learning to help achieve formal learning objectives in schools and workplaces. Since the publication of HPL I, the role of technology in informal learning—and the potential for linking it to formal learning—have only become more salient, as daily life is increasingly mediated by digital and Internet technology. A survey conducted in 2014-2015 found that 88 percent of U.S. teens had access to a smartphone and 86 percent reported going online from a mobile device at least once a day (Lenhart, 2015). Text messages have become a central part of social communication. In this survey, the average teenager reported sending and receiving 30 texts a day. Playing video games online or on their phones was reported by 84 percent of teenage boys and 59 percent of teenage girls.
Educators have explored approaches to capitalize on this pervasive access to these technologies (Bull et al., 2008; U.S. Department of Education, 2010). One approach has been to extend the time for academic learning through mechanisms such as putting WiFi on school buses so that students with long rides can do their homework online. Web-based homework systems give students adaptive practice outside of school hours. Some teachers are experimenting with flipped classrooms by having students watch video
presentations of academic content at home as preparation for applying that content to problem-solving activities in the classroom (Siemens et al., 2015).4 By doing in class what traditionally would have been thought of as homework, students in flipped classrooms have the opportunity to work collaboratively with other students and to get coaching from their teachers when they encounter difficulties applying new knowledge and skills to specific problems.
Online “hangouts” and other informal online groups of students support academic learning for college students in large lecture courses. Early research suggested that membership in study groups can be helpful in a challenging course (Treisman, 1992). More recently, study groups have met online, and the formation and functioning of such groups among people taking massively open online courses (MOOCs) has become a focus of research (Gasevic et al., 2014). Other programs are creating in-person study groups for learners taking courses online. For such programs in public libraries, library staff assist with technical difficulties and scaffold student behaviors intended to help with deeper learning (U.S. Department of Education, 2016).
Technology can support learning outside of school in other ways as well—for example, by providing opportunities for sustained intellectual engagement. Afterschool clubs, youth organizations, museums, and arts programs are examples of settings where technology-supported activities combine learning with entertainment (National Research Council, 2009). Adults can support this type of learning not only by acting as models of technology fluency but also by helping to connect interested children and adolescents with learning-rich out-of-school activities (Barron, 2006). A number of organizations (e.g., Computer Clubhouse, Black Girls Code, 5th Dimension, code.org, and the Digital Youth Network) have developed out-of-school activities to provide mentoring and learning opportunities in digital media and computer programming for low-income, female, and minority youth.
K-12 teachers must orchestrate many types of learning to achieve school systems’ ambitious goals for college and career readiness. For example, new science standards call on teachers to help their students develop skill in evidence-based argumentation through classroom discussions and related activities that bring out students’ initial ideas about science phenomena and confront them with opposing ideas and evidence as a way to trigger conceptual growth. Executing such instructional approaches with classes of 20-40 students is a challenge when students vary markedly in their prior experiences, interests, motivation, and knowledge. Technology can help educators
coordinate the many aspects of instruction and navigate the complexities of pedagogy, whether in K-12, college, or corporate settings. Educators may use technology in their personal lives but not be comfortable with integrating it into their teaching (Bakia et al., 2009).
The committee identified three levels at which technology can be integrated into instruction. At a basic level, the educator uses technology to present content or has students use technological tools designed to engage their interest. At the second level, students can use technology to support their individual learning in ways that they, rather than their teachers, direct. At the third level, digital tools allow learners to collaborate with individuals and organizations outside the classroom; these applications require that each participant or group have a network-enabled and connected device.
New technology places additional new demands on teachers, and this in turn places demands on both preservice training and in-service professional development programs. Teacher education programs can model effective integration. Moreover, although the characteristics of effective teacher professional development for technology integration have not been systematically established (Lawless and Pellegrino, 2007), the challenge clearly suggests the importance of devoting considerable training time to that integration, rather than attempting to cover it in a few lectures or a single course. Professional development in technology integration is more successful when it is of extended duration, gives teams of teachers from the same school or program the opportunity to collaborate in using concrete practices and to comment on each other’s practice, is coherent with the other practices and change initiatives at their schools, and demonstrates ways to leverage data from digital learning systems for formative purposes (Fishman and Dede, 2016).
Educators and researchers have long recognized that the knowledge transmission model exemplified by lecture-based teaching is less than ideal for many learners and many kinds of learning. It is difficult for educators to build on students’ prior understandings when they have no window into the nature of those understandings. Without that connection into “the learner’s world,” students experiencing a lectured lesson may tune out (Medimorecc et al., 2015). Even when teachers pepper their lectures with questions, the number of students who respond tends to be small. Instructors have limited information to help them identify whether the class is following their explanations, taking notes without thinking, or merely putting on an attentive face. Such concerns, which are particularly strong in college courses that enroll hundreds of students, has inspired the development of technologies that allow each student to respond to a question in a multiple-choice format presented on a small screen or handheld device. Student responses are sent to the instructor, who can then display aggregated responses as histograms (bar charts) for the whole class to see (Abrahamson, 2006; Kay and LeSage, 2009; Mazur, 1997).
Early evidence on the use of this technology showed improvements in student engagement and learning outcomes (Mazur, 1997). The results were attributed to the opportunity for the instructor to identify and address the sources of conceptual confusions common among students in introductory physics. More recent work has shown positive results with similar approaches (Deslauriers et al., 2011). These systems are most heavily used at the postsecondary level, but their use has begun to spread to secondary schools and even elementary school classrooms (Smith et al., 2011).
Another example of the classroom communication concept is Group Scribbles, a network technology designed to support collaborative learning. Group Scribbles works like the student-response systems described above except that students can share notes, sketches, and images, not just numerical responses or selections among multiple-choice response options. Student contributions are displayed (anonymously) on an electronic whiteboard. Group Scribbles has been used in the United States to help students understand fractions, in Spanish primary classrooms (Prieto et al., 2011), and in Singapore to teach science and Chinese language classes (Looi et al., 2009).
Software systems for writing instruction and for giving students feedback on their writing are another technological support for classroom communication. These systems can be used to distribute writing assignments and learning resources, provide immediate feedback to students, provide feedback on plagiarism, and allow students to submit their writing to the teacher or to peers for evaluation and feedback. The automated feedback may allow teachers to focus on what a student’s writing reveals about a deeper understanding of the material (Cassidy et al., 2016; Warschauer and Grimes, 2008).
Automated writing assessments have also been used to analyze students’ writing at deeper levels. For example, Summary Street, a program that analyzes the coherence of sentences and statements within a summary, has shown positive outcomes, such as increases in time spent revising and in depth of content, for elementary school students (Wade-Stein and Kintsch, 2004). Writing Pal (or W-Pal) is a strategy-based training system for middle school ages through adulthood that has game components for improving skills in writing argumentative essays, which are required in some high-stakes assessments (Allen et al., 2016; McNamara et al., 2015). With this digital system, the student generates a thesis statement, supporting statements, and then a conclusion. Pedagogical agents model good writing strategies and give interactive and immediate feedback as the student writes or revises an essay designed to address the student’s challenges. The Writing Pal system was based on studies of writing interventions that showed strategy instruction to be a successful form of writing instruction (Graham and Perin, 2007).
There is some evidence that teachers may not view such systems as a substitute for teacher-generated feedback. For example, a study of three writing software systems for use in classrooms, including WriteToLearn, which incorporates Summary Street, found that although teachers appreciated the immediate feedback these systems offer, they still found it important to provide their own feedback on other aspects of their students’ writing (Means et al., 2017).
Recent technological advances in several areas have yielded both opportunities and challenges. In this section, the committee reviews the issues associated with digital dashboards, distance learning, universal design, mobile devices, and features of technologies that may be addressed through further application of principles from the science of learning.
Digital dashboards allow a learner to monitor his own progress through the learning environment. Open learning environments (Bull and Kay, 2013) allow learners to observe their own performance scores on lessons and skills over time, which can be motivating and help develop metacognitive skills. Teachers can use the dashboards in learning management systems such as Desire to Learn or Blackboard, which provide a quick glimpse of the lessons, how each student is doing on each lesson, and which students need help (Dede and Richards, 2012). The dashboard has options that allow instructors to explore this information in greater detail. For example, they may identify which questions on an assignment were problematic for a student or the extent to which a student is mastering specific areas of skill and knowledge. The dashboard also can provide more general information about a student based on multiple lessons, such as: What percentages of lessons is she completing? How much time is she devoting to the course? How often does the student get stuck and need help? How often does she use digital help facilities? The dashboards also track and display noncognitive characteristics, such as profiles of a student’s emotions and social interactions (Siemens et al., 2015).
One example is the ASSISTments system,5 which allows teachers to create materials for mathematics as well as other topics, to see how well students perform, and to interact with researchers on possible improvements based on the science of learning (Heffernan and Heffernan, 2014). ASSISTment offers three views: The Builder view guides the curriculum designer or teacher in
creating lessons. The Teacher view shows performance of each student on particular lessons. The Student view guides the student in completing tasks and viewing feedback on performance. In 2015, ASSISTments was used by more than 600 teachers in 43 states and 12 countries, with students completing more than 10 million problems. A randomized field trial of the impact of using ASSISTments for homework problems showed an increase in seventh-graders’ scores on end-of-year math achievement tests, compared to a control group that completed homework without the immediate feedback offered by ASSISTments. Lower-achieving students benefited the most from working with ASSISTments (Roschelle et al., 2016).
Digital dashboards are most likely to perform as intended when they are not optional and when users have the time and resources needed to integrate these tools into instruction. Providing the professional development necessary for instructors to use these digital dashboards effectively is a challenge. Many teachers do not yet use digital platforms frequently and systematically in their classrooms. Very simple computer-teacher interfaces may be ignored or quickly abandoned after the novelty of the technology fades (Moeller and Reitzes, 2011). For example, instructors may need a systematic curriculum to facilitate access, use, and monitoring of the digital dashboard interface as a routine part of their courses.
Distance learning has been defined as “planned learning that normally occurs in a different place from teaching and as a result requires special techniques of course design, special instructional techniques, special methods of communication by electronic and other technology, as well as special organizational and administrative arrangements” (Moore and Kearsley, 1996, p. 2). It does not necessarily require technology, but digital technologies such as e-learning, online learning, or Web-based learning provide many advantages for distance learning (Siemens et al., 2015).
Digital technology can support synchronous communication between instructors and students, such as participating in a live Webinar, using technology-based instruction in the classroom, or corresponding in a course chatroom (instructor and learners spatially separated but interacting in real time). It can also support asynchronous learning, in which the interactions between a human instructor and students are separated in time (and typically also by space), as when the instructor posts a video lecture or lesson on a course learning management system or Website. Technology can also support communication, whether synchronous or asynchronous, such as between the learner and a computer-based teaching agent or with intelligent tutoring systems like those described earlier in this chapter.
Finally, blended learning, which combines one or more forms of distance
learning and face-to-face instruction, is facilitated by technology. For example, an instructor could use a learning management system to deliver course material, videos, tests, quizzes, and grades but would periodically interact with students face-to-face (Siemens et al., 2015).
Educators have traditionally been cynical about the effectiveness of distance learning approaches compared to traditional face-to-face synchronous learning (Thompson, 1990), and indeed the early research findings were mixed. The available evidence indicates that modern, technology-rich approaches to distance learning can be as effective as traditional approaches, more effective, or less effective (Bernard et al., 2009; Means et al., 2013). Efficacy depends on the quality of the interactions among the students, the content to be learned, and the instructor.
Technology that encourages students to actively engage with course material and with other students can positively affect cognitive outcomes. In a meta-analysis, blended online and in-person instruction produced better learning outcomes, on average, than conventional face-to-face instruction, but the blended learning conditions in the studies assessed for this analysis also incorporated other changes such as additional learning resources or more time for learning (Means et al., 2013). Based on analyses of the academic progress of students taking fully online courses, a number of researchers have raised concerns about the suitability of fully online learning for less motivated, lower achieving, or less mature learners (Miron et al., 2013; Xu and Jaggers, 2011a, 2011b). Although many students learn successfully with fully online courses, a blend of online and in-person instruction is generally recommended for lower achieving and younger learners (Means et al., 2010).
Social communication has become a ubiquitous feature of modern digital platforms in which instructors, students, and sometimes parents can communicate with each other through chat, email, and discussion boards. Such computer-mediated social support is routinely integrated into MOOCs (Siemens et al., 2015) to compensate for the lack of face-to-face contact with instructors and peers. Most learning management systems include social communication media even in traditional classrooms. However, usage is currently low, with only about 7 percent of the students using it, according to one estimate (Siemens et al., 2015). Social communication may be used more in the future as learning environments become more digitally supported, self-regulated, and socially connected.
The use of mobile technologies for learning has exploded in recent years, and this trend is expected to continue (Hirsh-Pasek et al., 2015; Looi et al., 2009). Although mobile technologies share some features with other electronic learning tools, their relatively flexible platforms are unique. Small and
easily transportable devices now give users quick and easy ways to search for information, create recordings (pictures, videos, audios), and communicate with others (Looi et al., 2009). This flexibility offers several advantages over standard e-learning. Mobile applications can be adapted for different learning contexts inside and outside of school.
Well-designed mobile applications can also be adapted to a learner’s abilities and desires, which may have positive effects on the learning process and peoples’ attitudes about their learning experiences (García-Cabot et al., 2015; Hsu et al., 2013). For example, learners who were surveyed reported positive attitudes toward mobile technologies with respect to the amount of effort it takes to use the devices, social norms related to using mobile technologies, perceived playfulness of the devices (i.e., how much fun people will perceive them to be), and the extent to which mobile learning facilitates self-management (Wang et al., 2009). The researchers who conducted the survey reported some gender and age differences in social norms associated with use of mobile devices; their results are consistent with other research on differences in general acceptance of mobile technologies (Magsamen-Conrad et al., 2015).
Despite indications of the potential benefits of mobile devices for learning, systematic research on their effectiveness is limited, and the research that exists often comes from the application developers themselves (Chiong and Shuler, 2010). Downsides also have been reported. For example, if laptops are not used for specific aims and purposes, they can impede students’ ability to focus their attention on learning (Fried, 2008; Sana et al., 2013). Adherence to guidelines for the use of mobile devices may help to promote learning in different educational contexts (for an example of guidelines, see Hirsch-Pasek et al., 2015).
Educational technologies are replete with features that can facilitate learning in controlled settings but can also serve as a distraction to many students (Gurung and Daniel, 2005). For example, e-textbook developers highlight possibilities for making information available in side-boxes or through embedded links as desirable features that allow students to click out of the reading to pursue learning about certain topics. Yet students may rarely choose to interrupt their reading to do this (Woody et al., 2010). Furthermore, the links can affect fluid reading of narrative and increase the learner’s cognitive load. Similarly, text comprehension and metacognition can decrease when readers switch from print to an e-reading format (Ackerman and Goldsmith, 2011). Printed textbooks may use boldface type to support readers’ understanding by highlighting key concepts, but some students rely on reading the highlighted material and skip the narrative.
There are ways both teachers and designers can help students benefit from technology. One is to provide adequate instructions for interacting with the technology. Instructions are sometimes poorly presented, such as on a cluttered computer screen, and users often skip them. Design that prioritizes easy engagement for the user and productivity with respect to the intended pedagogical goal is important. Achieving this objective requires substantial testing with users to ensure that the learner is guided to use the technology as intended. Designers can also rely on evidence-based principles supported by decades of research from the fields of human-computer interaction, human factors, and educational technology. Mayer (2001, 2009) identified 12 empirically supported principles to guide learning from multimedia (see Box 8-3). These principles are best viewed as guidelines for the design or selection of a learning technology, rather than as universal rules that apply to all multimedia
and populations, because implementing them may require tradeoffs among competing objectives.
Universal Design for Learning refers to a framework for drawing on relevant research to design educational experiences that are optimal for all learners, including those with specific learning challenges. Removing obstacles to interacting with technology has been a key objective of Universal Design (Burgstahler, 2015; Meyer et al., 2014). For example, many people benefit from speech-interpretation agents (such as the Siri agent on Apple iPhones) and audio book formats that originated with innovations developed for blind or deaf populations. The core vision of Universal Design is to design technolo-
gies “up front” that are accessible for diverse populations, rather than using an accommodation approach, in which features are added later (after the initial design) to allow people with particular disabilities to use the technology. An example of the accommodation approach is a mouth-operated control that allows a person who does not have use of her hands to operate a computer curser.6
Burgstahler (2015) identified seven Universal Design principles that have implications for the design of technologies for learning (see Table 8-1). These principles are often violated in typical learning contexts. For example, instructors often rely on a single medium (such as a PowerPoint presentation) instead of attending to cognitive variability and promoting cognitive flexibility through engagement of multiple modalities (Mayer, 2009).
Because people can become reliant on technology, the principles of Universal Design also may be useful in helping learners to adjust when a technology breaks down or is unavailable for other reasons (Burgstahler, 2015; Meyer et al., 2014). For example, it may be desirable to require that the user have some control over the device, rather than the device being fully automated, so the user can acquire some understanding of how the device functions and how to control it and thereby adjust to device malfunctions.
Several trends suggest digital technologies can support both formal and informal learning in adults. Older adults are increasingly comfortable using technological devices, including tablets and computers (Pew Research Center, 2014). For example, over the past decade, Internet use among people over age 65 has more than doubled, and it is likely to grow as more individuals gain access to computers with Internet connectivity (Pew Research Center, 2014). Despite stereotypes depicting older adults as being uninterested in using Internet resources, many older adults report being interested in using the Internet and are capable of learning to use it (Morrell et al., 2004).
Technologies may support cognition and learning in older adults by providing cognitive aids; expanding their access to content and resources for learning; promoting social connectedness; and providing immersive, multimodal,
6 Early examples of Universal Design for Learning were motivated by the search for ways to help individuals with various disabilities, such as those who are deaf, blind, otherwise physically disabled, or psychologically challenged. Examples of such technological breakthroughs include Braille, American Sign Language, and more recently, text-to-speech generators for the blind and speech-to-text generators for the deaf. Other examples of universal design include ramps and in-vehicle lifts for those in wheel chairs and medication organizers for elderly people. In the United States, federal standards and guidelines for the education of people with disabilities have followed from civil rights mandates, such as the Rehabilitation Act of 1973 and the American Disabilities Act of 1990, with amendments in 2008.
|Universal Design Principle||Example of Universal Design in Higher-Education Practice|
|Equitable use. The design is useful and marketable to people with diverse abilities.||Career services. Job postings in formats accessible to people with a broad range of abilities, disabilities, ages, and racial/ethnic backgrounds.|
|Flexibility in use. The design accommodates a wide range of individual preferences and abilities.||Campus museum. A design that allows a visitor to choose to read or listen to the description of the contents of display cases.|
|Simple and intuitive. Use of the design is easy to understand, regardless of the user’s experience, knowledge, language skills, or current concentration level.||Assessment. Testing in a predictable, straightforward manner.|
|Perceptible information. The design communicates necessary information effectively to the user, regardless of ambient conditions or the user’s sensory abilities.||Dormitory. An emergency alarm system with visual, aural, and kinesthetic characteristics.|
|Tolerance of error. The design minimizes hazards and the adverse consequences of accidental or unintended actions.||Instructional software. A program that provides guidance when the student makes an inappropriate selection.|
|Low physical effort. The design can be used efficiently, comfortably, and with a minimum of fatigue.||Curriculum. Software with on-screen control buttons that are large enough for students with limited fine motor skills to select easily.|
|Size and space for approach and use. Appropriate size and space is provided for approach, reach, manipulation, and use, regardless of the user’s body size, posture, or mobility (Center for Universal Design, 1997).||Science job. An adjustable table and flexible work area that is usable by students who are right- or left-handed and have a wide range of physical characteristics and abilities (Burgstahler, 2015).|
and tailored learning environments. Technology can be a cognitive support, for example, by keeping track of grocery lists, upcoming appointments, or medication regimes or by providing easy access to clear explanations of recommended medical procedures (Tait et al., 2014). Older adults are also taking advantage of commercial software and online educational opportunities through universities to enhance their exposure to new fields of study (Gaumer Erickson and Noonan, 2010). Some MOOCs offered by universities serve large numbers of middle-aged and older adults, but research on such online platforms is sparse. Research is needed on the characteristics of older
Technology can also bring people together to interact and collaborate virtually, when they might otherwise be isolated in their learning because they live alone or in remote areas or have limited mobility. Social connectedness is linked to successful cognitive aging (Ballesteros et al., 2015), so collaborative learning opportunities may lead to enriched social connections that improve cognition and mitigate cognitive decline.
Providing older adults with the rich, multimodal learning contexts and immersive learning environments that technologies can afford has the potential to optimize learning in later adulthood (Kensinger, 2016). Furthermore, older adults may benefit even more than younger adults from such opportunities as multimodal presentations (Mozolic et al., 2012). Older adults benefit when instruction matches and supports their own intrinsic motivations for learning and when they have the autonomy to guide their own learning. Technology-based experiences that can be tailored to individual learners may be especially useful for older people, whose life experiences and knowledge can be used to engage the individual and scaffold his learning (Kensinger, 2016).
Although older adults may benefit from technology-supported learning and report that technology can improve their quality of life (Delello and McWhorter, 2015), some experience challenges in adopting new technologies (Kensinger, 2016). Many adults need training that focuses both on how to use the technology and on their motivation to learn and the particular benefits the technology offers them (Kensinger, 2016). There is some evidence that participation in training itself can be a beneficial cognitive intervention, a finding consistent with research showing that mentally stimulating activities can benefit older adults’ cognitive functions (e.g., Lenehan et al., 2016). For instance, training older adults to use tablet computers has been shown to help episodic memory and increase processing speed more than social activities do (Chan et al., 2014). Another study suggested that training older adults to use online social networking led to gains in executive function (Myhre et al., 2017).
Other age-related challenges associated with technology use pertain to sensory capacities (e.g., font sizes), cognitive capacities (e.g., working memory load imposed by passwords, distractions from pop-up ads), and motor abilities (e.g., ability to control a mouse or to type on a small keyboard) (Pew Research Center, 2014). Older adults who have less income and education are less likely to adopt technologies than those who are more affluent and highly educated (Pew Research Center, 2014), potentially making it more difficult to reach some of the older adults who could most benefit.
Policy makers have worried for decades about the “digital divide” between those who do and those who do not have access to a large suite of digital resources (U.S. Department of Commerce, 2014). Significant gaps in technology access related to income and education levels remain a problem. A 2015 survey of a nationally representative sample of adults ages 18 and older reported growth in ownership of smartphones and tablet computers, whereas ownership of other kinds of computing devices (such as laptop and desktop computers) was relatively constant or had fallen (Anderson, 2015b). Smartphones, the most widely used computing devices in 2014, were owned by 68 percent of adults. Smartphone ownership did not differ by racial/ethnic identity but did vary by income level, education, and geographic location. For example, more than 80 percent of adults with at least a bachelor’s degree reported smartphone ownership, compared to just 41 percent of those who had not completed high school. The urban-rural difference in smartphone ownership rate was 20 percentage points (72% versus 52%).
Some prefer the term “digital inclusion” rather than “digital divide,” to signal that degree of access ranges along a continuum and that the issue is unequal participation in online activities, rather than complete lack of access for certain groups (Livingstone and Helsper, 2007). At this point, for example, access to the Internet is widespread, but the tools to use the current generation of digital learning resources and to create content for online distribution is much less so. This continuum is evident in data about technology use among young people, despite the generalization that young people know how to use technology and learn to do so more easily because they are “digital natives” who have grown up with digital technologies (Warschauer and Matuchniak, 2010). Young people from less-privileged backgrounds who lack technology mentors tend to use their computing devices mainly for texting friends, taking photos, playing simple games, and accessing celebrity Websites, activities that do not develop key digital skills (Anderson, 2015b). Gee (2009) has argued that the digital divide is growing, not shrinking, because those with greater literacy skills and more access to supports for learning continue to accrue larger and larger benefits in areas of learning not available to people of more limited means. Moreover, the most empowering aspect of digital participation—the ability to create or modify online content—lies out of reach for many. Concerns about these digital opportunity gaps have inspired the creation of clubs and community centers with rich technology resources and social supports to enable more of the U.S. population to use a larger range of technologies.
Basic Internet access in U.S. schools has become more consistent over time for students from different backgrounds (Warschauer and Matuchniak, 2010). Moreover, schools that serve students from different income levels differ less in their technology infrastructures than do students’ home environments. Although these are positive developments, the infrastructure requirements for
the current generation of digital learning applications have risen substantially, and available evidence suggests that schools are not yet positioned to close the digital opportunity gap. The 2016 Broadband Progress Report indicated that 41 percent of U.S. schools did not have Internet access at transmission speeds (bandwidth) capable of supporting digital learning applications (Federal Communications Commission, 2016).
The gap is particularly acute for those living in sparsely populated areas and on tribal lands. Moreover, provision of devices and broadband Internet access is not sufficient: programs of professional support for teachers and leaders in schools who serve low-income students are also necessary (U.S. Department of Education and Office of Educational Technology, 2016). Children who attend schools in more-affluent communities and who have highly educated parents are more likely to use advanced technologies, such as simulations, and to encounter stimulating challenges such as opportunities to create products and address open-ended problems through technology. In contrast, children attending schools in less-privileged communities are more likely to use technology for drill-and-practice and for taking online benchmark assessments (Warschauer and Matuchniak, 2010; Wenglinsky, 2005).
In 2013, the federal government unveiled a plan to provide 99 percent of all public schools with broadband Internet access within 5 years. The plan, called ConnectED, set goals for providing bandwidth to rural areas that would support Internet upload and download speeds needed to access digital resources for learning. For example, with the proposed upgrades, it is envisioned that whole classes would be able to use next-generation learning applications at the same time. The plan called for the preparation of teachers to take advantage of this improved technology infrastructure. It also called on private companies to support the effort by donating computing equipment and support services to the nation’s poorest schools.7 However, the costs of providing bandwidth to sparsely populated areas are large, and debate continues about feasibility and how to pay for the necessary upgrades. It remains to be seen, therefore, whether plans like ConnectED are sufficiently viable to be implemented within the next decade.
Effective implementation of digital technology for learning is vital, and failure to properly consider implementation challenges may significantly limit the benefits to be gained from using technology. There is considerable evidence that use of a single instructional technology can lead to different outcomes when used by different learners in different contexts. For example, a large federally funded, randomized controlled trial that investigated the impact of
reading and mathematics software on students in 132 schools showed positive impacts at some schools and negative impacts at others (Dynarski et al., 2007). Findings such as these show that education policy makers are wise to be cautious about the promise of “single solution” technologies and avoid making major investments in a technology without identifying concrete benchmarks for success and evidence that the technology can meet them. Many factors can affect the impact of a technology when it is used on a large scale, including the characteristics of learners, the sociocultural context, the nature of the affordances the technology provides, the curriculum and materials to be used for learning, the faithfulness with which the technology is implemented, and the involvement of instructors and learners in the implementation process.
Some researchers have advocated taking a “systems approach” in implementing learning technologies, in order to take into account the multiple factors that may affect the impact of the technology. This approach is illustrated by the Texas SimCalc study (Roschelle et al., 2010). SimCalc8 is a program designed to integrate the use of technology with curriculum goals and teacher professional development, with the goal of improving middle school students’ understanding of key mathematics concepts that provide the foundation for algebra and calculus. The program includes a curriculum unit on proportionality, linear functions, and rate that is built around a storyline (managing a soccer team) and calls for small-group work, class discussion, and use of both paper materials and a mathematics software program, The software allows students to see animations of different patterns of motion and link those with corresponding representations in the form of interactive graphs and equations.
The curriculum unit’s design emphasizes coherence across these different activities, all related to the unit’s theme. The SimCalc curriculum emphasizes repeated applications of key concepts in multiple contexts. The outcome measure in the tests of learning incorporated these deeper levels of understanding. Finally, the role of the classroom instructor in supporting student learning was supported with multiple days of professional development and materials, including a teachers’ guide with suggested activities and hints on likely student responses and misconceptions. The researchers found gains in mathematics skills across classrooms that used SimCalc (Roschelle et al., 2010).
A systems approach is also taken in intelligent tutoring systems for mathematics that have been used in thousands of schools, such as Cognitive Tutors (Koedinger et al., 1997; Ritter et al., 2007). Examples such as these suggest that several elements are important to a systems approach. In these cases, the users identified learning goals and matched the use of learning software to those goals. A method for measuring outcomes was identified in advance. The roles each of the actors in the system would play were coordinated. Teachers and other learning facilitators received substantial training.
The research discussed in this chapter demonstrates that recent advances in technologies for learning can offer significant benefits, but the results will depend on the alignment of goals for learning, contexts, the type of content to be learned, characteristics of learners, and the supports available for learners and instructors. Decision makers responsible for investments in technology need evidence about the many factors that can affect implementation of instructional technologies on a large scale.
From the available evidence on uses of digital technologies in people’s learning, we draw two conclusions:
CONCLUSION 8-1: The decision to use a technology for learning should be based on evidence indicating that the technology has a positive impact in situations that are similar with respect to:
- the types of learning and goals for learning;
- characteristics of the learners;
- the learning environment;
- features of the social and cultural context likely to affect learning; and
- the level of support in using the technology to be provided to learners and educators.
CONCLUSION 8-2: Effective use of technologies in formal education and training requires careful planning for implementation that addresses factors known to affect learning. These factors include alignment of the technology with learning goals, provision of professional development and other supports for instructors and learners, and equitable access to the technology. Ongoing assessment of student learning and evaluation of implementation are critical to ensuring that a particular use of technology is optimal and to identifying needed improvements.