Portal de Periódicos da CAPES

The digitalization of science education: Déjà vu all over again?

2020; Wiley; Volume: 57; Issue: 9 Linguagem: Inglês

10.1002/tea.21668

1098-2736

Knut Neumann, Noemi Waight,

Mobile Learning in Education

This special issue set out to provide a platform for reporting on empirical research that examines the use and impact of 21st century cutting-edge digital technologies and ecologies on science teaching, learning, and assessment. For decades technologies, and more recently digital technologies, have been said to revolutionize education, STEM education and, more specifically, science education. This movement started in the 1980s, when personal computers began to become available in classrooms. Personal computers, together with the respective software opened up a wealth of new possibilities in teaching and learning. Simulations provided dynamic visualizations of complex content in order to better support students in mastering understanding of these contents (Marks, 1982), and interactive learning programs based on videos allowed students to work through the curriculum at their own pace enabling a more individualized learning experience (Leonard, 1985). The former quickly developed into a vast amount of software tools designed to support students' science learning—from tools designed to model authentic phenomena (Doerr, 1996) to simulations of laboratory environments to engage students in authentic inquiry (Niesink et al., 1997). The latter, partially fueled by the advent of the internet and related technologies such as the hypertext protocol, developed into a variety of computer-based learning environments—from online (e)learning environments provided for remote, self-determined learning (Ajadi, Salawu, & Adeoye, 2007) to intelligent tutoring systems designed to automatically monitor and support students' learning (Graesser, Conley & Olney, 2012). More recent developments are driven by substantial increases in computing power on the one hand and research findings suggesting that (digital) technologies, such as simulations or modeling tools, alone are not necessarily helping students' learning, but instead need to be embedded in meaningful curriculum (e.g., Zhang, 2012; for an overview see Krajcik & Mun, 2014 ) on the other. Many new developments in the field engage students in authentic learning experiences through simulations of the real world (Barab et al., 2009), and integrate multiple individual technologies, such as simulations, modeling, or data analysis tools into a carefully sequenced curriculum activities (Gerard, Spitulnik & Linn, 2010). Most recent developments even integrate an automated tracking of students' learning and respective supports either through the learning environment itself or through the teacher (Gobert et al., 2013; Gobert & Sao Pedro, 2017). And the future is envisioned even brighter: Augmented Reality devices are expected to create authentic learning experiences, and Artificial Intelligence to allow for more open, exploratory (e)learning environments that automatically guide students in their learning based on their specific needs. All these technologies are said to be soon delivered through low cost personal smart devices, such as phones or tablets affordable to everyone. These developments, however, raise questions beyond the ones asking whether these new technologies will support better science learning or how these technologies need to be designed and/or used to support better science learning. Much like understanding the reciprocal relationship between science and technology that has long been a major goal of STEM education, research on digital technologies in education, STEM education and, more specifically, science education, has been driven by the wish to understand how (digital) technologies can support better teaching, learning, and assessment and how a better understanding of science teaching, learning, and assessment can help improved learning technologies. However, as discussed in the introduction to this special issue, equally important to understanding the relationship between science, engineering, and technology is understanding the impact that science, engineering, and technology have on the world we live in. In fact, as we are living in a world driven by ever accelerating scientific and technological progress, understanding the impact that this progress has on the natural world as well as our society, may very well be the more important aim. After all, few people will understand the science and technology behind self-driving cars, but many should be able to understand the technical challenges (e.g., modeling other road user behavior) and related ethical issues (e.g., responsibility in case of a crash) related to this technology. In terms of science education, this means that in addition to researching aspects of the use of digital technologies in science teaching, learning, and assessment, researchers should also focus on the broader impacts that the use of such technologies may have on science teaching, learning, and assessment as an ecology of its own right. Imagine, for example, teachers using automated essay scoring software, which will automatically grade essays written by students. Teachers may exhibit bias toward certain students or groups of students, but what about the algorithms underlying such software? They are widely considered neutral, but are they really? Or are they biased as well, and if so, how? Which students and language are represented in these algorithms? What would the teacher need to know about these algorithms in order to recognize and avoid bias? What implications does this have for teacher education? When digital technologies were still tools mastered by teachers it seemed reasonable to focus on the effectiveness of these tools and the conditions under which they were effective. But now that these technologies have become an integral part of our lives and, subsequently, science classrooms, it seems imperative to develop a deeper understanding of how these technologies impact science teaching, learning, and assessment, in particular in terms of the potential associated risks with respect, but not limited, to the goal of helping all students develop scientific literacy. This is why we conceptualized this special issue to explicitly call for submissions examining not only the use of 21st century, cutting-edge digital technologies but also the impact on science teaching, learning, and assessment. When technologies and more specifically digital technologies were first used in science teaching, learning, and assessment the focus was on showing how they better support student learning. The initial findings were not optimistic and for the most part the impact on learning has revealed numerous gaps (e.g., findings that the use of digital technologies does not lead to improvement in teaching and learning per se or that digital technologies mostly benefit students who already have a deeper understanding of the content). Hence, the holistic impact of these technologies is still unknown. Instead, research appeared to suggest that it depends on how these technologies were being used. Accordingly, one theme we were looking for in this special issue's studies was research aiming to produce knowledge about the conditions under which specific technologies are effectively used in science classrooms and the implications for how these technologies are best used. A second theme was the impact cutting-edge technologies might have on science education. We used the term ecologies in conjunction with digital technologies to acknowledge that cutting-edge technologies currently finding their way into science classrooms are often not just based on a single technology but rather represent a digital ecology—a learning space integrating different technologies and associated curricular and pedagogical practices that frame the context of learners and the learning environment. This applies to many e-learning environments as well as to science classrooms themselves. Science teaching, learning, and assessment as it happens in science classrooms, often draws on multiple, different digital technologies. Teachers may use modeling tools in conjunction with data collection tools, smart phone apps supporting students in constructing explanations. They may also use grading software for grading, learning management systems to organize lesson content and allow access to students, and online word processors to organize the collaborative work. A third theme we were interested in was the extent to which research still focuses on individual technologies (or technologies that may be perceived as individual) versus digital ecologies bringing together different technologies or even science education as a digital ecology itself. As a result of our call, we received 45 submissions, covering a broad range of topics from students' learning in mixed-reality environments to teachers' integration of innovative technologies. Below, we briefly summarize and discuss the findings of the papers that were selected into this special issue, followed by a summative discussion in light of the three themes outlined above. We conclude by summarizing the state of research on the use and impact of 21st century digital technologies and ecologies on science teaching, learning, and assessment, and formulating suggestions for issues that researchers should attend to in the future in order for digital technologies to support a 21st century science education that can meet the vision outlined in the introduction to this special issue. The papers in this special issue cover a wide range of different topics—from engaging elementary school students in scientific inquiry to promoting science education reforms. The digital technologies involved range from simulations, to game-based technologies, to social media. In terms of research focus, the majority of papers zero in on improving science teaching and learning or understanding how to improve science teaching and learning. The paper "A quasi-experimental study comparing learning gains associated with serious educational gameplay and hands-on science in elementary classrooms" by Hodges et al., for example, presents findings from two rigorous studies investigating the efficacy of a serious educational game, engaging students in inquiry learning in comparison to a regular inquiry learning unit specifically designed to meet the same learning goals as the game. The paper goes a step further, exploring additional student characteristics, such as students' reading level to examine the extent to which efficacy may depend on individual student characteristics; that is, whether games may be particularly effective for students who are proficient readers, whereas the comparison unit may be more effective for those who are less proficient in reading (as a game will generally involve more reading). Exploring such an aptitude-treatment interaction (for details on aptitude-treatment interactions, see Koran & Koran, 1984) is an important, yet rarely attended to, aspect of science teaching in learning. Many studies work from the hypothesis that a certain feature will be generally effective, although it is much more realistic to assume that different types of interventions will resonate with different types of student preferences or characteristics (e.g., some students may prefer game-based learning, where others will not). Similarly, it is important to explore how specific features of an intervention contribute to students' learning. The paper by Saleh et al., titled "synergistic scaffolds for collaborative inquiry in a game-based learning environment" is one example of such research. Focusing on supporting students' collaborative inquiry learning, the paper investigates the complex interplay of hard (i.e., predefined) scaffolds and soft (just-in-time) scaffolds provided by a facilitator. Based on a synergistic scaffolding framework (i.e., a framework delineating the role of multiple co-occurring scaffolds), the paper presents findings on how hard and soft scaffolds work together to successfully support students' learning. As such, the paper provides important insights into how to effectively design an important feature of digital learning environments such as games: the scaffolding of students' learning. The paper by Mikeska, Howell, and Croft, "Simulations as practice-based spaces to support elementary teachers in learning how to facilitate argumentation-focused science discussions" focuses on simulations and, importantly, aligns nicely with the previous two papers, in that it is centered around the learning of the critical inquiry practice of argumentation: the construction and use of arguments. More specifically, the paper is about engaging preservice elementary teachers (PSETs) in supporting elementary students in argumentation. In order to do so, the authors use what they call a "simulation." This simulation, however, is different from simulations commonly used in science teaching and learning such as simulations of phenomena (e.g., Moore, Chamberlain, Parson, & Perkins, 2014) or scientific laboratory investigations (e.g., Quellmalz, Timms, Silberglitt, & Buckley, 2012). Instead this tool simulates a classroom situation in which the PSETs engage with a group of avatars representing elementary school students. The paper presents findings from a study of how the PSETs support students in argumentation (i.e., the construction and critique of arguments) and the teaching moves the PSETs are using. The study also presents findings from interviews with the PSETs about their perception of the usefulness of the simulation for their learning process. This paper highlights the potential of digital technologies to create authentic learning experiences for PSETs that are otherwise hard to create (like micro-teaching situations). Collectively, these papers indicate that this is where the field has moved in terms of using digital technologies for science teaching and learning (i.e., the first theme discussed above): The creation of unique learning settings that present students with an authentic learning experience to support the learning of specific knowledge, skills or abilities that are otherwise hard(er) to develop. To that end, this movement reflects the increased focus on engaging students in learning of such knowledge, skills or abilities (e.g., engaging teachers in developing specific teaching strategies instead of just teaching them about the strategies). Two of the papers in this special issue focus on the assessment of student science learning. Both papers highlight recent advancements in the field of assessment and how they can be utilized for a more reliable and valid assessment of the knowledge, skills, and abilities learners are envisioned to develop in the 21st century. The paper titled "Identifying patterns of students' performance on simulated inquiry tasks using PISA 2015 log-file data" by Teig et al. addresses the use of simulations as a means for a more valid assessment of students' inquiry skills. In particular, the paper focuses on a specific feature that assessments using computer-based simulations offer: log-file data. Log-file data allow for analyzing how students approach the task solution (i.e., the solution process) and not just the final task solution (i.e., the product). Drawing on data from the PISA 2015 test, which included several simulation-based inquiry tasks, the authors examine log-file data to identify patterns of students' interactions with computer-based assessment and to determine whether unique characteristics of these interactions emerge as distinct profiles of inquiry performance. This paper highlights the potential of digital technologies for assessing student science achievement, in particular, the assessment of complex patterns of knowledge, ability, and skill, which require a more complex assessment setting (Teig & Scherer, 2016). Interestingly, the use of digital technologies for assessing students' science learning offers not only the possibility for creating authentic settings to assess complex patterns of knowledge, skills, and abilities (e.g., Neumann, Schecker, & Theyßen, 2019) and analysis of respective student behavior (instead of the products of student behavior): Since a wealth of data recorded are easily (i.e., digitally) available for analysis, this opens up the possibility for automated analysis. Automated scoring has been a subject in science education research for a while. More recently, however, research has begun exploring automated scoring techniques that draw on the developments in artificial intelligence, the most prominent being machine-learning approaches. Machine learning is a widely used technology where the machine "learns" from existing data to establish and refine algorithmic models, just as humans learn from experiences. Based on these models, the computer can then predict student knowledge, skills, and abilities based on new data (Zhai et al., this volume). The paper "From substitution to redefinition: A framework of machine learning-based science assessment" by Zhai et al. presents findings from a systematic review of the use of machine learning in science assessment to answer the question of the extent to which machine-learning has transformed science assessment. The paper delineates a framework to describe the uses of machine learning in science assessment and, based on that framework, characterizes common trends. Finally, "Idle chatter or compelling conversation? The potential of the social media-based #NGSSchat Network as a support for science education reform efforts" by Rosenberg et al. diverges from the others in that this paper does not focus on whether or how specific technologies lead to improved teaching and subsequently learning or enable more reliable and valid assessment of students' science learning (e.g., the first of our themes). Instead, the authors explore how social media, as one of the pillars of the digital ecology we live in, can help address a systemic problem in science education reform; that is, initiating changes beyond individual classrooms (hence the paper addresses aspects of both our second and third theme). More specifically, the paper examines the potential of #NGSSchat, a twitter chat, to support learning and coordination at the systems level to enable the implementation of the vision outlined in the Framework for K-12 Science Education and the Next Generation Science Standards (NGSS Lead States, 2013). The paper analyzes posts from regularly occurring, moderated twitter chats to understand the depths and types of conversations, who participates in these chats, and how to initiate and sustain professional networks involving partners with a rich variety of expertise for implementing science education reform (e.g., teachers, administrators, and researchers). Each paper in this special issue makes its own, unique contribution to the field. The paper "A quasi-experimental study comparing learning gains associated with serious educational gameplay and hands-on science in elementary classrooms" by Hodges et al. identifies a higher efficacy of game-based learning, but also suggests that gains may depend on students' mind set. The paper by Saleh et al., "Synergistic scaffolds for collaborative inquiry in a game-based learning environment," highlights the importance of soft scaffolds for ensuring the efficacy of hard scaffolds when the latter do not have the intended effect. This finding raises the question of how soft (i.e., just-in-time) scaffolds can be automatically created by computers as students engage in digital learning environments. The paper by Mikeska, Howell, and Croft, "Simulations as practice-based spaces to support elementary teachers in learning how to facilitate argumentation-focused science discussions," showcases how simulated teaching situations can help understand how preservice teachers (and likely, in-service teachers too) engage in supporting students' learning of argumentation, what aspects of argumentation they focus on, and what teaching strategies they are using. Interestingly, interviewed about their perspective on the simulation, the preservice teachers asked for more immediate feedback. In terms of assessing student science learning, in their paper "Identifying patterns of students' performance on simulated inquiry tasks using PISA 2015 log-file data," Teig et al. identify three patterns for how students interact with the simulation task, namely: strategic, emergent, and disengaged. These patterns revealed different characteristics of students' exploration behavior, inquiry strategy, time-on-task, and item accuracy—depending on the task. Only the disengaged patterns showed similar characteristics on both tasks. The results also suggest that whereas students exhibiting a particular pattern more likely show the same or a similar pattern on other tasks, there is also substantial variation among patterns students exhibit across tasks—raising questions that warrant further research. In terms of using machine-learning techniques to analyze the (rich) data from science assessments (using digital technologies) the "From substitution to redefinition: A framework of machine learning-based science assessment" by Zhai et al. reveals that machine learning has not yet transformed science assessment. The authors identify, however, some interesting trends, such as the use of machine learning: (a) to analyze continuous, longitudinal data such as observations of student behavior (for an example, see Spikol et al., 2018); (b) to serve a more personalized learning by analyzing data from network or mobile devices; or (c) to distill the features of quality teaching from existing big data sets of science teaching and learning. The paper "Idle chatter or compelling conversation? The potential of the social media-based #NGSSchat Network as a support for science education reform efforts" by Rosenberg et al. shows how social media can be successfully utilized to engage different stakeholders such as teachers, administrators, and researchers in substantive discourse about the implementation of science education reform with each other, effectively creating wide-ranging professional networks. The results also show, however, a tendency toward homophily (especially within substantive discourse) and inconsistencies regarding sustained participation—both of which indicated a need for further research. As a whole, the papers illustrate the richness of research on digital technologies in the context of science teaching and learning. As a whole, the manuscripts also indicate that the field continues to be focused on the efficacy of digital technologies in science learning (i.e., our first theme). What has changed is the collection of digital technologies. In the past century, digital technologies included all kinds of animations, visualizations, or dynamic representations—subsumed under simulations. That is, what has been simulated has been science content, with the aim to help students learn this content. The digital technologies covered in the papers of this special issue differ in that they simulate authentic settings to support or assess students' (or, more generally, learners') learning of complex patterns of knowledge, skills, and abilities (as defined by Mislevy, 2016). This reflects the shift away from focusing on single knowledge, skills or abilities in science teaching and learning and assessment of science teaching and learning. In terms of science teaching and learning, the field also appears to increasingly take into account the role of the individual learner and their environment in learning—accounting for the likelihood that there likely is no single learning environment that supports effective learning for all, but that one type of learning environment (e.g., an educational game) will resonate with one group of students (e.g., girls), but not with another (e.g., boys). Overall, the papers in this special issue highlight the potential of simulations of authentic settings to support student learning, suggesting that more research is needed in order to understand what type of simulation may be suitable for which group of students, and how simulations need to be designed to best support students' learning. This is not about the specific implementation or design of a game, but more generally about simulation feature types (or, more generally, technologies), such as how feedback to learners is provided, and how subsequent learning is supported (i.e., scaffolded). In terms of the assessment of science teaching and learning, the papers reflect the benefit of simulation for creating authentic settings to assess complex patterns of students' behavior. The papers in this special issue also reflect the potential of machine-learning techniques to use digital technologies in science teaching and learning efficiently. This is despite the systematic review by Zhai et al. indicating that the current use of machine learning techniques is still mostly framed in the field's effort to improve the psychometric features (i.e., reliability and validity) of traditional assessments, or simply to increase efficiency through automated scoring instead of having humans score student responses. The potential of machine learning, however, lies in that it can allow for assessing student learning as it happens—in response to the call for continuous, less invasive assessments of students' learning (e.g., Penuel, Roschelle, & Shechtman, 2007). Machine learning techniques can help not only with automatically scoring students responses but also to incorporate a wide range of heterogeneous data created as students interact with digital technologies. Machine-learning techniques could, for example, help assessing student mindset in situ in a setting such as that described by Hodges et al., recognizing students' misinterpretation of hard scaffolds, and providing soft scaffolds as a correction in the game-based learning environment described by Saleh et al., or by monitoring students' behavior and providing them with the feedback that they requested as in the study by Mikeska et al. machine-learning techniques can even help bridge between different digital technologies used in the classroom. As highlighted by Zhai et al., machine learning can draw on data generated by different digital technologies to monitor learning across formal, informal, and even nonformal learning opportunities. The increasing use of smartphones, tablets, or similar devices to access different technologies creates a digital ecology, in which students live. This allows for monitoring their science learning when they use a (smartphone) app designed to help them construct explanations in scope of science instruction at school, as well as when they read up on why the sky is blue on Wikipedia in the afternoon. This vision raises a range of questions regarding how such use of digital technologies will transform science education (i.e., our third theme) and what the long-term implications such use will have for our society (i.e., our second theme). Since this vision may quickly become reality, we think these questions should be explored rather sooner than later. The paper by Rosenberg et al. provides an example of what such research may look like and how such research will likely need to also explore new methodologies to address its questions. Focusing on a core digital technology (i.e., twitter) of the digital ecology that we live in (where we can connect to anyone, anywhere, anytime), the paper explores the transformative nature that digital technologies and ecologies can have on science education (or science education reform in this case). Without #NGSSchat, it would likely not have been possible to regularly connect such diverse participants, who are diverse not only in their role (e.g., teacher vs. researcher) but also in terms of location and hence to create such rich professional experiences for, in fact, all participants. However, in order to explore the potentially transformative nature of twitter, or more specifically #NGSSchat, the authors introduced other methodological approaches that are not native to science education research. What this approach offers are new questions that also require new methodologies (from other domains) that address gaps, difficult issues, and new modes of interrogation that challenge the practices of science teaching and learning with digital technologies. In summary, each of the papers in this special issue provides a glimpse of what becomes possible through the use of digital technologies and ecologies in science education. In particular, if taken together, the papers highlight what would be possible if different technologies could be integrated to form a digital ecology that is not used for science teaching, learning, and assessment, but in which science teaching, learning, and assessment takes place, and which provides students, teachers, administrators, and so on, with the means to organize more personalized and hence more inclusive science education. This transformation raises new questions that require further exploration. Some of these questions may be difficult or even impossible to answer using established research design and methods. However, the importance of understanding the digital transformation science education is undergoing and the impact it has on our society warrants the hardships that come with these challenges. Overall, recent technological developments provide digital ecologies for and create respective ecologies of science education that can help improve science teaching, learning, and assessment—not in terms of an increased efficiency, but in terms of offering new and, more importantly, inclusive possibilities, to allow students make experiences that would otherwise be impossible (or at least difficult to achieve), to make science education more equitable and to take science education to a new level. Future research should explore these opportunities to help understand how these technologies impact science teaching, learning, and assessment and how they can meet the high expectations.