Keywords

Introduction

Flipped Learning (FL) practice and research has recently grown extensively at all levels of education, prompting active discourse among researchers and practitioners. Although there still exist learners left behind by a lack of technological access, FL has certainly benefited from increasing access to digital technologies, resources, and broadband connectivity along with open course ware, open educational resources, massive open online courses, or YouTube clips, which has led to the easier and wider implementation of FL. Another reason for its increasing popularity is the movement by many educational institutions to improve the quality of education by incorporating FL to bring about learners’ active engagement coupled with positive learning outcomes (Lee & Choi, 2019). This chapter will discuss the research into FL: what is known from current FL literature, what the practical implications are from the literature, what the gaps within FL research and between FL research and practices are, and how those gaps should be addressed. This chapter will begin by discussing the definitions, history, and theories underpinning FL followed by the effectiveness of FL, its affecting factors, challenges, and possible solutions and introduce a set of extended versions of FL model and future research directions.

What Is Flipped Learning? The Definitions of Flipped Learning

Flipped classroom, flipped learning, inverted classroom, or inverted learning are all terms used in connection with this instructional model. Originally called flipped or inverted classroom, these terms are being replaced by flipped learning, which will be used throughout this chapter with the abbreviation FL.

When defining FL, researchers have slightly differing perspectives regarding what is flipped and whether to include video-based pre-class learning. FL has been referred to as both flipping traditional teaching and flipping early forms of blended learning. Bergmann and Sams (2012) defined FL as an instructional strategy “where work that was traditionally done in the class is now done at home, and what was traditionally homework is now completed in class” (p. 13). An earlier definition by Lage, Platt, and Treglia (2000) similarly defined FL as happening when “events that have traditionally taken place inside the classroom now take place outside the classroom and vice versa” (p. 32). Other definitions are to gain foundational knowledge before class and then actively apply it in the collaborative classroom (Brewer & Movahedazarhouligh, 2018), individual learning as homework outside the classroom, and interactive group-based learning inside the classroom (Bishop & Verleger, 2013). With students learning and reviewing concepts at home, class time can then be freed up for active, collaborative activities within the group space and increased time with the teacher (Bond, 2020; Lo & Hew, 2017).

Some other definitions posit FL as flipping the early version of blended learning, i.e., a combination of face-to-face synchronous classroom lecture and online asynchronous text-based activities (Garrison & Kanuka, 2004). Incorporating the concept of online and offline learning, Lee, Lim, and Kim (2017) defined FL as “a newly emerging form of blended learning, where students individually watch online lectures prior to class and then engage in classroom learning activities interacting with peers and instructors” (p. 427). Chen et al. (2018) also defined FL as a hybrid approach, combining online learning and face-to-face classroom activities where, during online learning, students engage in content learning before class and thereby during face-to-face classroom activities students engage in maximized active learning.

There is yet another lack of consensus among researchers and practitioners on learning materials with different modalities for FL pre-class. Some use a broader definition of FL that includes not only video but also reading assignments, homework problems, and PowerPoint presentations as pre-class learning materials (Giannakos, Krogstie, & Sampson, 2018; Lai & Hwang, 2016; Lee & Choi, 2019). More cutting-edge materials like interactive simulations, animated readings, and intelligent tutoring systems were also reported as pre-class learning (Davies, Dean & Ball, 2013). The FL definition of Bergmann and Sams (2012) allowed videos and other forms of media for pre-class learning materials, and Brewer and Movahedazarhouligh (2018) also delineated resources such as modules, videos, or readings.

Other researchers take a stricter approach to defining FL as only utilizing online video lectures as pre-class learning materials. Lo and Hew (2017), for example, defined FL as an instructional approach combining video-based learning outside the classroom and interactive group learning activities inside the classroom. In addition, Bishop and Verleger (2013) excluded from their definition implementations that did not employ pre-class lecture videos. A systematic review of 37 FL studies published during 4 years by Giannakos et al. (2018) revealed that over 80% of reported cases employed video lectures in practice. The reasons of this stricter FL approach can be summarized as follows: firstly, many believe students learn better with video lectures than reading assignments (Akçayır & Akçayır, 2017; Lo, 2020). Video lectures are believed to function better because it is easier to repeat, pause, fast-forward, or slow down the instruction to match the learner’s pace and make content accessible at any time (Bush, 2013); secondly, the emphasis on video may be because the original version of FL that Sams and Bergman tried in their high-school chemistry course and launched into the spotlight exclusively utilized prerecorded video lectures.

However, many educators are questioning whether video lectures can flexibly be replaced with reading assignments for FL pre-class learning (Lo, Hew, & Chen, 2017). This is related to the evolving nature of FL’s definition as well as the effectiveness of learning resources with different modalities. The value of FL lies in enabling in-depth, learner-centered learning experiences by offloading content delivery onto pre-class online learning (Lee & Choi, 2019). Thus, despite the popularity of video lectures, other solutions for providing pre-class learning could also be valuable components of FL. One example could be, in a problem-based learning situation, having students explore and study out the situation individually before class, which may allow for more constructivist learning in-class within groups as they derive solutions to the problem (West, Tawfik, Gishbaugher, & Gatewood, 2020).

Synthesizing two approaches of its definition, FL can be defined as an instructional strategy or model that combines online pre-class self-paced learning for foundational knowledge acquisition with video lectures together with other equivalent solutions and offline in-class interactive learning activities for knowledge application facilitated by instructors .

The History of Flipped Learning Practice

There are disagreements regarding when and where FL originated. Some point to Harvard’s Eric Mazur, who, in the late 1990s, implemented a peer instruction method in his undergraduate physics course, having students prepare to learn outside class by doing pre-class readings and answering questions about those readings and then in class engaging students in discussing their answers with peers coached by instructors. He found that students learned better when instructors coached students’ learning rather than directly instructing them. Although it resembles FL, it seems unwarranted to regard it as the origin of FL since it was not similarly named.

The term “classroom flipped” and “flipped class” then was initiated by Wesley Baker at Cedarville College in 2000. He had practiced FL in his graphic design course since 1995 using web course management software, where lectures were uploaded and asynchronous threaded discussions were hosted. In 2000, he mentioned FL made it possible for faculty to move from “sage on the stage” into “guide on the side” (Baker, 2000, p. 11). The term “inverted classroom” was then used by Maureen Lage, Glenn Platt, and Michael Treglia, who, in 2000 at Miami University, implemented FL in their economics course, putting a high value on the individualized approach of FL for accommodating different student learning styles.

While the concept and practices of FL had existed since 2000, sporadic reports and anecdotes were shared among limited groups of educators in the form of conference papers. The research contained limited empirical evidence on implementation and effectiveness (Cheng, Ritzhaupt, & Antonenko, 2019). Full-fledged implementation and development were then pioneered by high-school chemistry teachers John Bergmann and Aeron Sams. From 2007 and 2008, they recorded lectures for students who had missed their classes. Later, these teachers prerecorded all of their chemistry lectures and utilized them in assigning student’s homework, subsequently finding that students engaged in problem-solving and experiments during class more actively. After reflecting on questions like “when are students struggling and seeking help from their instructor or peers?” and “how can instructors help them while optimizing the technological benefits of an online learning environment and the in-person benefits of human interactions in the classroom?” (Lee et al., 2017, p. 428), the two then formulated their instructional strategy into the FL model and shared their experiences in a series of publications (Bergmann & Sams, 2012; Bergmann & Sams, 2014).

Since 2012, FL has gained widespread popularity, becoming extensively recognized and implemented at all levels of education across the world. In 2012, Sams and Bergmann started the nonprofit Flipped Learning Network™ (FLN) to provide educators with the knowledge, skills, and resources for successful FL implementation. From 2016, international and coalitional support for effective implementation of FL began to emerge (Birgili, Nevra Seggie, & Oğuz, 2021). In 2016, an FL global coalition, Flipped Learning Global Initiative (FLGI), was formed by researchers, practitioners, and technologists from 49 countries. In 2018, higher education institutions such as Harvard and Stanford and MEF University in Turkey, the first fully flipped university, took the initiative for the FL global movement and collaborated in publishing Flipped Learning 3.0 Global Standards that were curated from 187 FL best practices worldwide (Flipped Learning 3.0 Global Standards Summit, 2018).

Theories and Research Evidence Supporting Flipped Learning

With its widespread implementations and reported anecdotal success, researchers have sought to stipulate theoretical rationales or evidences for FL and provide FL practitioners with suggestions for best practices. Related theories can be the didactical 3C model of Kerres and de Witt (2003) in the area of blended learning theories; the media synchronicity theory of Dennis, Fuller, and Valacich (2008) in the area of media and communication theories; the cognitive load theory from a group of studies initiated by Sweller, Merrienboer, and Paas (1998), and the schema theory of Bartlett (1995) and subsequent studies in the area of cognitive psychology theories.

Since FL is under the big umbrella of blended learning, designing FL entails designing the optimal blend between two modes of learning, i.e., online pre-class and offline in-class learning. Blended learning (BL) integrates classroom face-to-face learning experiences with online learning experiences (Garrison & Kanuka, 2004; Garrison & Vaughan, 2008). Within this framework, BL practitioners have sought to design a better blend, one of the theoretical endeavors being the didactical 3C model of Kerres and de Witt (2003). The model specifies three components that all types of BL should have: Content, Communication, and Constructive components. In this model, educators designate the relative proportion of each component and then decide a corresponding delivery format. The Content component is factual learning content that needs to be available to learners; the Communication component is interpersonal interactions between instructor and learners or among learners for more arguable learning tasks; and the Constructive component is activities that facilitate learners’ more cognitive engagement in most complex learning tasks. In FL contexts, the theory supports FL having all of the 3C components: the Content component for factual learning, which is implemented with FL pre-class, and the Communication and Constructive components for more cognitively engaging learning interacting with instructor and peers, which are implemented with FL in-class. The model suggests the proportion of components should be designed considering learning objectives and learning contexts. Once the proportion of each of three Cs has been decided, the delivery method is then chosen from the variety of technologies and communication tools offering different affordances.

Several media and communication theories inform how the components of the 3C model can best be arranged using available media. Media synchronicity theory (MST) is one of the theories informing FL design, which argues that the fit between information processing needs and the affordances of the media used to support synchronous communication determines the quality of communication performance (Dennis & Valacich, 1999; Dennis et al., 2008). MST views communication as a series of tasks to achieve shared understanding, where a task consists of two cognitive processes – conveyance and convergence. Conveyance focuses on transmitting large amounts of new information followed by receivers internalizing that information to construct their own mental models. Convergence focuses on verifying, adjusting, or negotiating participants’ mental models, enabling shared understanding. MST proposes that tasks with conveyance focus can be better supported by synchronous media, whereas tasks with convergence focus may be supported by asynchronous media. In FL contexts, pre-class learning aims at conveyance, i.e., transmitting new information and the learner internalizing that information to construct his/her mental model, for which asynchronous media such as learning materials posted online should be selected. In-class learning aims at convergence, i.e., approaching to shared understanding by interacting with instructor and peers, for which synchronous media such as face-to-face class should be selected.

The importance of FL pre-class learning and pre-class mini-lectures is supported by cognitive load theory and schema theory, which concern the role of prior knowledge or expertise of learners. Cognitive load theory (CLT) states that learning can be hindered if learner’s cognitive resources and working memory are overburdened, and learning should be designed so that learner’s cognitive load is reduced, saving capacity for more meaningful and higher-order learning tasks. CLT classifies the load into three types: intrinsic cognitive load, which is the inherent difficulty of the learning content; extraneous cognitive load, which is the load caused by the way the content is presented; and germane cognitive load, which is efforts required for integrating new information, the creation and modification of schema (Sweller et al., 1998).

Leaving extraneous cognitive load that is associated with learning materials aside, one way to reduce cognitive load is constructing and making accessible one’s cognitive schemas, for which the learner’s level of prior knowledge or expertise is a critical factor (Bartlett, 1995). A learner’s level of prior knowledge or expertise relating to the topic can determine the comfort with which complicated learning tasks can be processed simultaneously in working memory without burdening cognitive overload. Learners with more prior knowledge can have more cognitive resources available to actively process learning tasks at hand, whereas learners with less prior knowledge need more cognitive resources and thus can choose superficial learning strategies. In the context of FL, locating individualized pre-class learning that can provide background knowledge for collaborative in-depth in-class learning activities prior to FL in-class is for activating cognitive schemas to reduce intrinsic cognitive load prior to FL in-class. Ultimately this allows more resources for germane load that require more complex schema constructions during FL in-class learning .

Growth of Flipped Learning Research

With the popularity of FL practice, FL research has grown exponentially. A systematic search, as of June 26, 2021, using the keywords “flipped learning” or “flipped class*” or “inverted learning” or “inverted class*” in titles of peer-reviewed journal articles throughout the four databases of ERIC, Scopus, Web of Science, and Medline, found in total 1,629 publications. As shown in Fig. 1, the volume of FL peer-reviewed journal articles jumped significantly from 5 in 2012 to 372 in 2020. In less than 10 years, FL research has achieved tremendous quantitative growth, and this trend is expected to continue.

Fig. 1
figure 1

The quantity of FL publications since 2012 as of June 25, 2021 (n = 1629)

FL literature is presently dominated by investigated FL cases and FL effectiveness (Lundin et al. 2020). As Lundin et al. (2020) found, in the initial years of FL research, descriptive case reports on FL implementation were dominant, but these studies were mainly limited to reporting the experimental process and results of adopting FL in individual courses or subjects at individual schools. There have also been studies with reflection or opinion-based arguments around FL and with a theoretical approach discussing relevant educational theories supporting FL (Lundin et al., 2020). For this reason, the study design lacked rigor with lower levels of evidence. As educators and researchers asked for compelling evidence on the effectiveness of FL, studies have tried to assess its effectiveness with higher levels of evidence – mostly done through meta-analyses.

This exploding body of empirical literature has also led to an increasing number of review studies such as systematic reviews, scoping reviews, and meta-analyses (Birgili et al., 2021; Hew et al., 2021). Systematic syntheses of the state of research can provide researchers and practitioners with knowledge on the integrative effectiveness of a certain approach of learning design (Zawacki-Richter, 2020). With the systematically synthesizing endeavor, we can “look at the body of evidence rather than looking at any study in isolation” (Nordenbo, 2010, p. 22) and stay reflective about future research, education practices, and policies. The growing number of FL review studies reflects a scholarly and practical need to understand the FL outcomes across studies.

Effectiveness research investigates the learning outcomes achieved by FL, by either comparing FL to traditional lecture-based learning or comparing the start and end of FL using pre-post designs. The learning outcomes analyzed have been mostly academic achievement, measured by score or grade, and occasionally motivational or attitudinal outcomes such as engagement, perceptional change, and satisfaction (Brewer & Movahedazarhouligh, 2018; McLaughlin et al., 2014). Among the 25 meta-analyses, all reported that FL had higher learning outcomes, among which 24 were statistically significant and 18 reported with medium to large effect sizes. The full list of FL meta-analysis studies can be accessed in Data 1. The full list of FL review studies can be accessed in Data 2.

According to recent systematic review studies on FL, the greatest proportion of FL studies have been conducted with undergraduate students in higher education institutions, with a far smaller proportion with K-12 contexts (Akçayır & Akçayır, 2017; Cheng et al., 2019; Lundin et al., 2020). The areas reported as having implemented FL were diverse, but higher education and STEM education were two main areas (Bond, 2020). Asia and the USA were geographically dominant in FL research (Lundin et al., 2020). In this chapter, I will now discuss the practical implications from the literature, including the gaps within FL research and between FL research and practices .

Flipped Learning Benefits

The majority of studies investigating the effectiveness and benefits of FL have reported that FL promotes improvements in student academic performance mainly measured by test scores or course grades. The systematic reviews and meta-analyses synthesized this dominant cognitive outcome of FL. Recent meta-analyses with relatively large samples of studies— for example, a meta-analysis of 203 studies published between 2012 and 2019 (Tutal & Yazar, 2021), of 95 studies published between 2013 and 2019 (Zheng, Bhagat, Zhen, & Zhang, 2020), and of 198 studies published between 2012 and 2018 (Strelan, Osborn, & Palmer, 2020) — reported positive cognitive learning outcomes, i.e., growth in academic performance or outperformance by FL over traditional approaches. Twenty-four of the 25 FL meta-analysis studies in Data 1 investigated either academic performance or cognitive achievement.

The next most commonly investigated outcome of FL is student motivation or engagement, measured, for example, by the amount of discussion participation. Related outcomes were positive attitudes or perception toward FL, which were measured usually by self-report survey. Other outcomes investigated include a self-regulated learning or study habit, competency in problem-solving and social or collaborative learning, digital competency, academic efficacy and changes in behaviors, changes in professional practice, or changes in actual outcome such as patient recovery. Four out of 25 studies reported course satisfaction or motivation as FL outcomes (Data 1).

There are many other potential positive benefits of FL that have been explored. One is that it can enable flexible, individualized learning with pre-class resources at students’ own pace at any time, and they can repeat or learn additional content as they want (Birgili et al., 2021; Brewer & Movahedazarhouligh, 2018; Cheng et al., 2019; Hew et al., 2021; Lo & Hew, 2017). The benefit of flexible individualized learning can lead to learners’ self-regulated and self-directed learning by properly scheduling their time to learn pre-class assignment and ultimately being accountable for their learning (Lee, Park, & Davis, 2022). Kostaris, Sergis, Sampson, Giannakos, and Pelliccione (2017) showed that FL brought learners’ growth in their digital competencies while accessing and managing digital learning resources.

Another compelling benefit of FL is that it facilitates learner’s active and interactive learning. FL design is ultimately for securing active learning time by offloading knowledge delivery pre-class, allowing in-class time to be utilized in engaging learners in interaction between instructor and students and between peers (Birgili et al., 2021; Bond, 2020). FL can benefit large college courses such as introductory basic science courses, where learner engagement is traditionally minimal (Brewer & Movahedazarhouligh, 2018) and where the benefit of active and interactive learning leads to enhanced engagement.

Lastly and most critically, most FL studies have agreed that FL can promote meaningful and constructivist learning by involving learners in knowledge construction assignments and activities both in and out of the classroom, thereby developing higher-order thinking competencies such as critical thinking or problem-solving skills (Brewer & Movahedazarhouligh, 2018) .

Factors Affecting Flipped Learning Effectiveness

While the majority of studies have trumpeted positive benefits of FL, some studies have reported no significant difference between FL groups and non-FL groups. A few further studies reported FL impaired student learning. One possible solution to the muddy picture of FL effectiveness is to explore factors affecting FL’s effectiveness and investigate into what context and in what ways FL can be effective, as intended. FL is not a panacea and might not be effective in certain contexts or for a particular learner group. Educators’ decision-making would benefit greatly from knowing how and where FL should be implemented and which particular groups would gain maximum benefit.

Educational level has a bearing on the effectiveness of FL. According to Cheng et al. (2019), 55 publications, published from 2000 to 2016, reported the overall effects of FL on K-12 and college undergraduate students’ learning outcomes and its moderating variables and concluded that FL outperformed the traditional lecture method. However, for graduate students, the traditional way of teaching and learning was more effective than FL, which was explained by the fact that the usual format of graduate education is already similar to FL and discussing and analyzing scholarly works are their main activities.

The subject or area of learning can influence the effectiveness of FL. Arts and humanities courses benefited more from FL than did mathematics courses, perhaps because some subject areas are difficult to learn independently and students could be overwhelmed by the FL pre-class learning (Cheng et al., 2019). Learning objectives and assessment method also can be affecting factors. In the course for skill acquisition assessed by OSCE (Objective Structured Clinical Examination) or knowledge retention/comprehension assessed by multiple-choice questions (MCQs) (Chen et al., 2018; Hughes & Lyons, 2017), FL showed mixed result or no significant advantage over a non-FL approach.

One of the most critical affecting factors is learner’s pre-class learning. The influence of pre-class learning on final FL success has been empirically shown to be both significant and strong. As the proportion of students who do not complete the assigned pre-class work is reported as quite substantial (Heitz, Prusakowski, Willis, & Franck, 2015) and pre-class is the opening move to the benefits of FL, completion of pre-class assignments has a drastic effect on the differential effectiveness of FL. Typically video materials induce higher academic performance than reading materials (Lee & Choi, 2019), but the quality of video materials can be also crucial to the success of FL, implying a need to produce or prepare high-quality video materials that are attractive and not overly long. Although video materials certainly have unique and optimal features for FL pre-class, it is the quality of learning content and instructional design that those video materials deliver that matters, as explained in the classic debate raised between Richard Clark and Robert Kozma on the role of media in learning in the early 1990s. As an instructional strategy to promote students pre-class learning, quizzes at the start of an in-class session can be employed. Hew and Lo (2018) found that the use of verification quizzes for testing learner’s understanding on pre-class learning significantly influenced the effect of FL.

Learner factors, academic capability, attitudes toward FL, and their technology competency also can be significant factors as shown in Table 1. As educator factors, their pedagogical approach, such as their recognition of FL as a learner-centered approach, their empowering, engaging and encouraging students and commitment can be a critical affecting factor in the effective implementation of FL (Brewer & Movahedazarhouligh, 2018; Shi et al., 2020). Institutional support and professional development may be another factor in the same vein (Brewer & Movahedazarhouligh, 2018).

Table 1 Factors potentially affecting FL effectiveness

Remaining Challenges and Possible Solutions

FL research conducted so far has provided practical implications for FL implementations, but there remain challenges that research has not yet fully addressed. Lo and Hew (2017) classified those challenges into three: student challenges, educator challenges, and technical challenges.

Challenges related to students are mostly related to their pre-class learning issues and also general unfamiliarity and an unreceptive attitude toward a new approach to learning. Student unpreparedness of pre-class assignments is seen as the biggest challenge of FL due to lack of either motivation or of self-regulation, and overcoming this is critical because FL pre-class learning is positively and strongly correlated with FL learning outcome and nearly twice that of in-class learning (Lee & Choi, 2019). Underprepared learners are clearly an obstacle to the intended benefits of FL. Students’ disengagement from the pre-class learning, due to their natural vulnerability to distraction while watching online lectures, and helplessness during pre-class learning when they are unable to understand the materials and want to ask questions during the pre-class can be also pre-class-related challenges. The general unfamiliarity and unreceptive attitude toward FL can be another type of challenge. Students can feel uncomfortable and nervous in the new format and further resist FL in the initial stage of implementation (Brewer & Movahedazarhouligh, 2018; Giannakos et al., 2018), which can be even worse for older learners (Birgili et al., 2021).

Student challenges naturally become educators’ challenges. Significant start-up effort is required to design both pre-class and in-class sessions, tightly linking the two. Pre-class video requires levels of instructional and technical qualities attractive enough to engage reluctant students in self-regulated learning and prevent distractions. The length and load must also be appropriate. The production of a 10-min video is known to require nearly 2–3 h work, and according to McLaughlin et al. (2014), an FL educator has to spend 1.27 times more time in developing and managing an FL course as compared to a traditional course. Preparing in-class session also requires great deliberation in order to provide the necessary active learning experiences. Educators whose courses were mostly lecture-based may be bewildered as to what to prepare for the in-class time. Most importantly, educators need to create a tight link between pre- and in-class learning so that the two complement each other, which may ultimately involve redesigning a course (Lee & Choi, 2019).

Another fundamental educator challenge is the need to shift their pedagogy and philosophy about teaching and learning to minimize the knowledge deliverer’s role and maximize the facilitator’s role. However, in reality many educators resist this change and prefer traditional lecture-based teaching (Hew et al., 2021). Since FL requires a fundamental conceptual shift in thinking about teaching and learning, this challenge must be tackled strenuously.

Lastly, there are technical challenges in terms of infrastructure, devices, and technology skills, for both students and educators. Educators must constantly be aware of any learner left behind by such issues.

Research has provided some possible solutions to these challenges, as shown in Table 2. For student pre-class learning issues, strategies such as the checking of students’ pre-class learning written notes, online quizzes or video-embedded quizzes, pre-class assignments, in-class quizzes, and study schedule can be provided to support student self-regulation (Cheng et al., 2019). Educators must help students recognize that FL pre-class learning is a crucial component of FL, not a supplementary task. Regarding unfamiliarity and unreceptive attitude toward FL, course orientation with clear explanations about the FL concept and teacher’s expectations of students’ responsibility can be provided (Lee et al., 2017). For the educators’ video production challenge, educators can deliberately select and use the vast number of preexisting video learning materials (e.g., OER, MOOCs) provided by institutions and refer to production guidance on how to create high-quality video.

Table 2 FL challenges and possible solutions

For educators who usually deliver knowledge during class and do not know what to do during FL in-class, designing in-class learning activities can be a big challenge. Studies suggested a set of in-class activities such as discussion, small group activities, feedback, problem-solving, Q&A or exercises, group projects, gaming activities, personal projects, and peer assessment. Other instructional model such as team-based learning (TBL) (Michaelsen, Knight, & Fink, 2004) or peer instruction (PI) (Mazur, 1997) can also be implemented, each of which has a structured procedure. For example, TBL starts with (1) individual readiness assurance test (IRAT), followed by (2) team readiness assurance test (TRAT), (3) appeal (the opportunity to appeal that their answer choice is better), and (4) instructor feedback, and finally ends with (5) application-focused exercises. These steps can guide FL educators when designing and implementing FL in-class learning.

Fully Online Flipped Learning, Bichronous Online Learning, and HyFlex Learning

The COVID-19 pandemic, causing offline in-class sessions to be cancelled, has naturally called for a modified FL model implemented in fully online contexts. A few studies (e.g., Stöhr, Demazière, & Adawi, 2020; Jia, Hew, Bai, & Huang, 2021) reported the experiences when implementing the fully online FL model. The fully online flipped learning converts offline in-class learning into online in-class learning using video conferencing tools. Face-to-face classroom learning is replaced by synchronous virtual class learning as in Fig. 2. Video conferencing enables real-time interaction with nonverbal communication cues, such as facial expressions and body language, which offer high levels of immediacy and student engagement (Jia et al., 2021). Stöhr et al. (2020) and Jia et al. (2021) investigated whether fully online FL is as effective as the original FL and found no statistically significant difference in average academic performance between the two. However, Stöhr et al. (2020) cautioned that fully online FL led to a polarized larger spread in performance, indicating some students perform better, while others struggle even more.

Fig. 2
figure 2

Fully online FL in comparison with original FL and traditional learning

Although no specific prescriptions for what should be done during each mode of learning exist, the blending of both asynchronous and synchronous online learning has recently been christened with the name of bichronous online learning by Martin, Polly, and Ritzhaupt (2020). Bichronous online learning can be a broader concept than the fully online flipped learning model. In the former model, students participate in their learning anytime, anywhere, during the asynchronous parts of the course but then participate in real-time activities for the synchronous sessions, whereas in the latter model, the way of blending both asynchronous and synchronous online learning is the same, but it specifies that knowledge delivery should be done during the asynchronous part, and collaborative constructive learning activities should be conducted during the synchronous part. Further variation is the hybrid flexible (HyFlex) approach suggested by Abdelmalak and Parra (2016), where students choose how to participate in the course or session, either online or offline, synchronously or asynchronously, over the course and from session to session, depending on which works best for them.

Future of Flipped Learning Practices and Research

In this chapter, I have reviewed the definition, history, supporting theories, research status quo, modified versions, and future research issues of FL.

FL review studies have given many recommendations for future research, with the following four being most often mentioned. Firstly, while some FL studies have provided rich accounts of design features and associated implementation, more have not. Future studies should provide details on specific course design, materials used, pedagogical strategies, learning environment, and context for both pre-class and in-class learning. In addition, there have been calls for the research to be more rigorous and theoretically grounded. Secondly, research should explore the contexts in which FL works best, i.e., for whom, when, under what circumstances, and in what ways. Thirdly, although most meta-analyses show that FL is better than the traditional approach in terms of academic performance, studies comparing the two in terms of higher-order learning outcomes are limited.

Fourthly, a small number of studies into long-term implementation of FL have been conducted. In addition, as some studies stressed that FL involves a shift in thinking about teaching and roles of teachers, and its successful implementation requires cultural shifts, further studies on cases of a whole FL school or multiyear FL implementation would reveal whether such a pedagogical or cultural shift was induced by FL.

FL started as a local trial strategy in a US high school and in less than 10 years exploded in popularity, having been extensively recognized and implemented at all levels of education across the world by researchers. Behind this popularity is a global desire to improve the quality of education through putting learners in the center of their learning. The contributions of FL in improving educational practices thus far cannot be ignored. However, in practice, educators have to remember that flipping, per se, cannot guarantee its effectiveness but can be the catalyst to reflecting and redesigning the entire learning experience. FL research should rigorously listen to and support practical needs in order to have the FL model properly and effectively implemented and further extended.