In an era when multimedia and web technologies are becoming increasingly more ubiquitous and affordable, the concept if not the definition of video (n.d.) has been extended far beyond the moving images that are seen in a recording or broadcast. In this study, video refers to the multi-modal medium that delivers moving images with synchronized audio and sometimes on-screen texts. Examples of on-screen texts are video captioning, scrolling news ticker in television programs, and pop-up ads banner in streaming digital videos (e.g., YouTube).
The capability of delivering information in multiple modes allows video to communicate more effectively, flexibly, and engagingly than single-modal media (e.g., books, pictures, voice recordings, etc.). In contemporary theories on communications and new literacy (e.g., Kress, 2003, 2010), different modes have different affordances in conveying certain types of information. With the capability to deliver information in multiple modes, video can be designed to distribute information, which is dispatched in small trunks, through the optimal modes in the optimal sequence. Moreover, the information is not simply split into different modes in the video. The information trunks in each mode can constantly interact with those in other modes, allowing video to convey more sophisticated, nuanced meanings.
Due to the multi-modality characteristic of the video media, it has been frequently used to create authentic instructional scenarios in teacher education. Literature indicates several benefits for using video in teacher education. First, an instructional context can be represented in flexible ways and in highly dynamic, subtle forms. This is because video usually carries a greater number and variety of verbal (e.g., spoken language and on-screen texts) and non-verbal cues (e.g., sounds, music, human behaviors, lighting effects, camera panning, etc.) than single-modal media with homogeneous information. When watching videos of real classrooms, teachers are exposed to a more authentic instructional context by receiving heterogeneous data (Heimann & Pittenger, 2000; Wang & Hartley, 2003).
Second, the scenarios represented in a video, either real or fictional, can serve as the ground for critical reflective learning. Video not only provides preservice teachers more time and opportunities to analyze and interpret authentic classroom activities that they usually don’t have access to, but also becomes an anchor point that “stimulates and facilitates both individual and collaborative processes of learning and knowledge building,” (Chambel, Zahn, & Finke, 2004, p.349). Concrete examples in the video are catalysts for effective teacher reflection, and provide a common ground for colleagues to exchange ideas and feedbacks (Tan & Towndrow, 2009; Tripp & Rich, 2012). Novice teachers benefit tremendously from such critical reflective learning processes by receiving the opportunities to apply pedagogical concepts to their understanding of real teaching practices (Borko et al., 2008; Schrader et al., 2003).
Third, video can scaffold novice teachers in capturing and contextualizing events occurring in the classroom that may not be readily accessible due to the overwhelming data in a real classroom. Despite video delivers multimodal and sophisticated information, it can also be designed in ways that suppress overwhelming or distracting data so as to allow novice teachers to focus on the pedagogical aspects. Teacher learning is especially salient when video watching is coupled with learning activities that lead to profound reflections such as peer discussions (Hannafin & Kim, 2003; Sherin & Van Es, 2005; Van Es & Sherin, 2006).
Since video was introduced to teacher education in the United States in the 1960s, it has been extensively used to provide meaningful, immersive, and complex contexts for teacher learning (e.g., Copeland & Decker, 1996; Schrader et al., 2003; Sherin & Han, 2004; Yadav & Koehler, 2007). However, for over four decades, until the advent of digital video media in the early 2000s, video had been used typically as a subordinate learning resource (see Sherin, 2004; Wang & Hartley, 2003). This can be attributed to various physical, cognitive, and pedagogical limitations of older video media formats (e.g., VHS tape and DVD) that made it difficult to conduct contextualized, systematic, and reflective learning.
With older video media, teachers participating in a TPPD program were mandated to be simultaneously present in the same physical space. And various resources were required in order to play the videos: the video player device (e.g., VHS player, DVD player), the audio-visual equipment (e.g., monitor, projector, speaker), and video media (e.g., VHS tape, DVD). The required resources were often unavailable for individual teachers to use out of the class – particularly the older video media requiring special equipment to be reproduced. In fact, many copyrighted videos were not allowed to be reproduced. Finally, instructors and students did not often meet outside the regularly scheduled class time anyway (Wu & Lee, 1999).
When the instructor and students watched TPPD videos together in the class, learning was often inefficient. As previously mentioned, students might feel reluctant to interrupt the video playback in order to discuss an instructional scenario in the video. Therefore, discussions about the video content tended to happen after the video finished playing (Cadiz et al., 2000; Lee & Wu, 2006). However, analogue videos were highly self-contained and it was extremely time-consuming to access its content in non-linear fashions. In the normally limited class time, therefore, the discussions about the video were often brief and generic. And reflection and concrete feedback about the video was often hindered because many specific contexts in the video were no longer referenceable without replaying the video.
In addition, video was often criticized for making learning passive by overwhelming viewers’ cognitive load with its formal features (i.e. production techniques and conventions), though some researchers argued video viewing could also be positive as long as the viewers’ attention and comprehension was actively engaged either by the video’s form and content or through scaffolds such as preplay and repetition (e.g. Bickham, Wright, & Huston, 2001; Calvert, Huston, & Wright, 1987). If novice teachers did not understand certain segments in a training video due to the lack of explanations in the video itself, the only way to resolve the puzzle was to hear expert teachers explaining them. Unfortunately, mentors or expert teachers were not always available to answer all the questions that novice teachers might have. A large number of teacher training videos, particularly field recordings of real classrooms, could not ensure students would understand the video content adequately.
The invention of web-streaming and social media technologies sheds light on how to use video more effectively in teacher education – by means of computer-mediated communication (CMC). Web-streaming videos are distributed over the Internet so that teachers can watch them asynchronously in different locations. More importantly, teachers can: 1) access any portion of the video conveniently and watch repeatedly, 2) annotate the video content and search with semantic data, and 3) use the video more interactively with other types of learning resource. Social media helps build and maintain a sustainable community of learning and practice which facilitates collaborative knowledge building, on-demand social scaffolding, and emotional support (Beck & Kosnik, 2006).
It is necessary to clarify the distinctions between CMC and computer-supported collaborative learning (CSCL) in order to explain why the former’s perspectives are adopted in this study. Both terminologies refer to the concept of using computer and Internet to scaffold interpersonal information exchange. However, the CSCL point of view emphases the interactions between human learners and the technology-based learning environment, and has been criticized for placing human learners in a passive position who are “positively or negatively affected by the social network characteristics of the medium,” (Sherblom, Withers, & Leonard, 2013, p.33). In comparison, CMC assumes learners have more dynamic roles during the communication because they are not only affected by the technology framework but also acting as strategic agents by constructing messages and developing relationships within the constraints of the infrastructure (Sherblom, 2010). In the context of this study, the CMC perspective is more favorable because it highlights the dynamic and constructive nature of collaborative teacher learning.
In order to design a systematic and flexible framework for understanding VDD in TPPD, it is necessary to review the most influential use models of video in teacher education. There were some excellent literature reviews on this subject conducted in the past but their perspectives were instructional design in general such as what pedagogical models were facilitated by video, what affordances and challenges those models had, etc. (e.g. Wang & Hartley, 2003; Sherin, 2004 ). The literature review in this study, in comparison, emphasizes the relation of the video components with other learning resources, as well as how video contributes to the eventual learning outcomes. The following use models will be reviewed: modeling, micro-teaching, interaction analysis, video cases and field recordings, hypermedia programs, video annotation, and video-driven discussions.
What is video modeling?
Video is frequently used to demonstrate expert teaching to novice teachers, which is often referred to as video modeling or video coaching. In this use model novice teachers model expert teachers’ field teaching in the video. According to Sherin (2004), two approaches are typically used:
1) Expert teacher is videotaped in field teaching and the recording is shown to novice teachers to model. Sometimes novice teacher is also videotaped in teaching the same content as a comparison to expert teaching (e.g. Masats & Dooly, 2011).
2) Expert teacher is videotaped in field teaching. Novice teachers watch video excerpts with expert teacher’s commentaries so as to understand why certain practices are effective or ineffective (e.g. Van Es & Sherin, 2008).
In both approaches, the eventual goal is not simply having novice teachers observe or mimic expert teaching because that does not help them understand why and when to apply the expert strategies. Instead, novice teachers should be encouraged to “reflect and think critically on their own teaching strategies,” (Masats & Dooly, 2011, p.1154) in order to generate meaningful questions for subsequent learning. Group discussion is proven to be a highly effective method to achieve that purpose in F2F settings (e.g., Barrett & Rasmussen, 1996; Van Es & Sherin, 2008; Watters & Diezmann, 2007).
A transformed approach of expert modeling is to expose novice teachers to multiple expert perspectives. In this case, novice teachers model the expert thinking rather than expert teaching and the video may or may not be a classroom instruction. For example, a group of novice teachers can gather to watch video clips of several experts addressing a related issue, then work in a group to discuss the expert perspectives and formulate their own ideas (e.g., Schwartz, Lin, & Holmes, 2003). The goal is to ensure novice teachers make productive reflections based on the multiple expert perspectives and generate genuine solutions from the teachers’ own angles. Social support and collaboration play a critical role in the idea formation process.
What is hypermedia programs
Hypermedia program, sometimes called multimedia program in a broader context, began to be used in teacher education in the early 1990s but was not widely adopted until the late 1990s. Hypermedia usually refers to nonlinear media that combines graphics, audio, video, plain texts, and hyperlinks, in comparison to linear media such as traditional books, VHS tapes, etc. which must be accessed in a given order. For example, in a traditional novel book, one has to read it from front to back, from beginning to end, and therefore books are linear. In comparison, in a website which is nonlinear, one can read the webpages in any preferred order through the hyperlinks, and therefore websites are nonlinear. Multimedia is a more generic term for hypermedia without emphasizing the nonlinearity characteristic. For convenience sake, this study will use hypermedia programs for both concepts.
Unlike any of the use models previously reviewed, hypermedia program is more of a technical rather than pedagogical model. Prior to the appearance of web-based video annotation and video-driven discussion tools (discussed next) in the late 2000s, hypermedia programs did not benefit teacher education more than what traditional video-based learning models offered except that they were capable of combining multiple pedagogical models into the same learning module. This capability provides the following benefits:
First, hypermedia programs can compile related knowledge objects of various media types into the same learning module. Traditional media objects such as books and videotapes are highly self-contained and their linear information flows are designed independently. Therefore, it is extremely difficult to design a learning module where one can read a book and watch a videotape simultaneously, switching back and forth between the media yet still being able to receive the knowledge objects in a meaningful order. In comparison, in hypermedia programs all the digital media (e.g. texts, graphics, audio, video, etc.) are nonlinear and the knowledge objects in each medium can be accessed randomly. Therefore, the knowledge objects in each medium can be selected and re-assembled into a new learning module which in turn can be accessed linearly or nonlinearly as designed. With hypermedia, it is much easier to design constructive curricula that can help novice teachers develop more flexible, adaptive pedagogical thinking skills (Cannings & Talley, 2002; Hill & Hannafin, 2001; Schrader et al., 2003; Wang & Hartley, 2003).
Some researchers also proposed the notion of hypervideo, a sub-genre of hypermedia programs, where “video is not regarded as a mere illustration, but can also be structured through links defined by spatial and temporal dimensions,” in hypermedia spaces (Chambel et al., 2004). Hypervideo differs from regular digital video in that the audio-visual content in hypervideo has mutual references with external learning resources, which allows the learner to have higher controls over the knowledge objects and conduct more collaborative, reflective learning.
Second, hypermedia programs use digital data that can be duplicated and distributed more conveniently than older media such as books and analogue videotapes. As a result, the information became significantly more accessible, abundant, current, and dynamic (Hill & Hannafin, 2001; Van Es & Sherin, 2006).
Finally, modern technology infrastructures such as the World Wide Web (WWW) diminish the boundaries between individual learners imposed by the resources. Previously it was difficult to bring heterogeneous learners in terms of time, location, subject, and media into the same learning scenario. But this could easily happen with the help of Internet (Hill & Hannafin, 2001).
By the time when hypermedia programs entered teacher education, the WWW had been invented for almost a decade. However, for another decade TPPD programs still used hypermedia primarily in traditional learning settings rather than over the Internet. The reason is the technology infrastructure at the time still couldn’t provide satisfactory distribution channels for video. As a result, the profits brought by those new technologies did not “go beyond what traditional teacher education approaches attempted to achieve,” (Wang & Hartley, 2003, p.129).
The use models reviewed so far have been extensively covered by the extant literature. The two emerging use models to be reviewed next, video annotation and video-driven discussion, have evolved from hypermedia programs and is becoming increasingly influential in the recent years.
After reviewing the major use models of video in teacher education, the larger picture becomes clear. The usage of video in TPPD can be roughly divided into three eras. The first era preceded the late 1990s when TPPD curricula were conducted in F2F settings only. During this era the use of video was inconsistent, relatively simple, and isolated from other learning resources. The second era, from the late 1990s to the late 2000s, was transitional and marked by the advent of hypermedia programs that succeeded in making TPPD curricula more flexible but failed to unleash the true power of digital media due to the lack of effective distribution and collaboration channels. The third and current era started from the late 2000s when digital media and social technologies began fundamentally transforming TPPD learning environments. Video-based TPPD curricula are becoming more collaborative, comprehensive, and holistic in this era.
In the current era, the context of TPPD curricula is gradually moving from F2F to CMC due to the latter’s support for comprehensive curricula and flexible pedagogical support. In addition, promoting social support and collaborative learning has become the overarching theme in both research and practice. Video-driven discussion, as well as its sibling video annotation, is the most common instrument to foster collaborative learning because it can help novice teachers connect pedagogical theories to their preexisting knowledge.
What is micro-teaching?
Micro-teaching scales down teaching in different perspectives such as instruction time, class size, and instructional strategies. Initially proposed in the 1960s when portable video equipment was introduced, the goal of micro-teaching is to give teachers confidence, support, and feedback by letting them test-teach a short slice of a lesson to a group of real or pretended students (Allen, 1966). Typically, micro-teaching follows the procedures below (Sherin, 2004):
1) Participating teachers are introduced to a specific teaching skill, for example, lecturing or leading a discussion.
2) One participant teaches a mini lesson to a small group of students who are usually peer teachers. The teaching is videotaped.
3) The participant uses the video to analyze his or her success with the target skills. The analysis is usually conducted with peer teachers or a supervisor, where feedback is received regarding the teaching performance.
4) The participant restructures the lesson based on the feedback and re-teaches it to a new group of students.
5) The participant repeats Step 2 to 4 as needed until he or she masters the skill.
In micro-teaching, video is clearly the key component because it captures teachers’ field practice and later becomes the basis for feedback and reflection. And it is critical for novice teachers to effectively receive feedbacks and conduct reflections on their micro-teaching in order to improve their teaching strategies. As previously mentioned, in F2F settings there are usually inadequate time and opportunities for novice teachers to receive concrete, profound feedbacks from colleagues, faculties, or supervisors. As a result, F2F teacher learning tends to be superficial (Abendroth et al., 2012; Cadiz et al., 2000; Lee & Wu, 2006). In comparison, CMC avoids those problems by providing the flexibility of giving and receiving feedbacks through asynchronous communication. In CMC, teachers are able to conduct more meaningful professional discourse and conduct more profound reflections (Abendroth et al., 2012; Lee & Wu, 2006).
What is video annotation?
As illustrated earlier, it is very difficult to examine learning procedures and outcomes that involve analogue videos due to the media’s linear nature. In order to address this problem, the traditional approach was to create video transcripts that contain both verbal and non-verbal information in the video (e.g. human gestures, actions, sounds, environmental descriptions, and so on) and analyze the transcripts with reference to the corresponding video time (e.g., Jordan & Henderson, 1995). Yet teachers still found it challenging to “systematically observe, analyze, or reflect deliberately,” (Rich & Hannafin, 2009, p.53) on videos using video transcripts because video transcripts are still linear (though have much better accessibility than video) and carry only a portion of the information in the original video.
Thanks to the progress of web and digital video technologies in the recent years, video annotation finally brought the video and text media together by allowing teachers to insert annotations to specific points of the video. Teachers are now able to conduct reflective, contextualized, and extended learning with other teachers via the Internet. As illustrated earlier, hypermedia programs support and combine multiple pedagogical models. But it appears video annotation tools have more resemblance to the interactional analysis and video cases models.
Rich and Hannafin (2009) reviewed and compared several noteworthy video annotation products for promoting teacher reflective learning. According to the authors, video annotation tools provide the following learning benefits: 1) referencing between videos and diversified resources, 2) scaffolding reflection and analytical thinking, 3) encouraging collaborative reflection, and 4) reducing learning curves compared to more complicated technologies such as LMS. However, at the time when Rich and colleague reviewed those tools, video annotation was still in the early stage and there were large rooms to improve particularly in terms of the tool’s capability in connecting different media resources, the limitation in collaboration, and the lack of analytical frameworks to scaffold teacher reflection. Similar conclusions can also be found Martin & Siry (2012).
What is interaction analysis?
Interaction analysis initially appeared in the 1960s (Flanders, 1970; Wragg, 1970) without using videos, but became more feasible with the advent of video cameras in the late 1970s. This practice, like micro-teaching, believes complex classroom behaviors can be understood by breaking them down into more essential components. But instead of videotaping a fake audience, interaction analysis often involves a real class. The transcripts of the videotape are often used to identify particular student and teacher behaviors as well as analyze how those behaviors are related.
Different from micro-teaching which was designed to improve teaching practice through feedback, interaction analysis was at first conducted on the individual basis with a strong behaviorist process-product approach to help novice teachers discover what teaching practices were more effective but not necessarily why. But with the increasing influence of the social constructivism theory in teacher education since the mid 1990s, many researchers started to explore how to improve pre-service teachers’ conceptual understanding of teaching through collaborative knowledge building. They proposed the directing construct that had been long missing in teacher education of intense inquiry, discussion, and reflection in an inclusive community of learning/practice to help novice teachers connect concrete teaching practices with pedagogical theories (e.g., Beck & Kosnik, 2006; Reiman, 1999). The importance of collaborative knowledge building in interaction analysis is explained in Jordan and Henderson (1995):
Group work is also essential for incorporating novices because Interaction Analysis is difficult to describe and is best learned by doing. […] newcomers are gradually socialized into an ongoing community of practice in which they increasingly participate in the work of analysis, theorizing, and constructing appropriate representations of the activities studied. (p.43-44)
What is video-driven discussion?
While video annotation tools continue evolving in recent years, video-driven discussion emerged as a related but more generic use model. The two use models resemble each other in terms of the workflow that multiple participants insert data to the same video and collaboratively construct knowledge based on each other’s inputs. And both of them can encompass one or more pedagogical use models (e.g., modeling, video cases, interaction analysis, field recordings). However, video annotation focuses on promoting teacher reflection and analytical thinking, without emphasizing collaborative knowledge building (Nelson & Slavit, 2008). In comparison, video-driven discussion is specifically designed for collaborative inquiry-based learning.
The theory behind video-driven discussion is video watching does not automatically lead to cognitive or behavioral improvements in teaching practice. A supportive professional community is required to assist teachers explore how to improve teaching and transform practice via critical feedbacks from others (Borko et al., 2008). In F2F settings, video clubs can provide such a professional community (e.g., Sherin & Han, 2004; Sherin & Van Es, 2008; Van Es, 2012) but with various constraints such as time, space, communication, and so on. By moving the community support to the Internet, those constraints can be mitigated or lifted.
CMC is an indispensable component in video-driven discussion because it is the vehicle for pedagogical support from peers and mentors. When the CMC component is removed, video-driven discussion becomes identical to non-collaborative video annotation because the user herself would be the only recipient of the text input. Yet, even with the CMC component enabled, collaboration does not always happen in video-driven discussion. For example, in a video discussion tool called Vialogues (EdLab, 2009), many teachers posted messages only to themselves without talking to each other though the reply feature was built in the software (Figure 1). The similar situation happens in some video annotation tools which support collaborative features (e.g., videoANT, University of Minnesota, 2015, see Figure 2). Therefore, the presence of the CMC component
Figure 1. Teachers made comments only to themselves in Vialogues
Figure 2. videoANT, a tool designed for video annotation, can also be used for discussion
As pointed out earlier, CMC assumes positive roles of the participants during learning. In an exploratory study involving Vialogues (Hsiao, Malhotra, Chae, & Natriello, 2013), the authors discovered that when online class discussions are both pre-structured and coordinated, participants tend to be more focused and engaged longer on the learning tasks. This finding seems to suggest by encouraging certain interpersonal communication patterns in video-driven discussion, it is possible to achieve specific instructional goals. Similar findings were reported in forum-based (without video) asynchronous discussions (e.g., Mazzolini & Maddison, 2007). And further empirical evidence can be found in the field recordings literature reviewed earlier that supervisors should provide systematic support to supervisees to improve teaching on the behavioral level (Hoekstra & Korthagen, 2011).
Today, video-driven discussion tools are still scarce especially those specifically designed for education. In addition to Vialogues, a few video applications originally designed for other purposes also have discussion. Examples include videoANT (University of Minnesota, 2015), Grokit Answers (Grokit, 2015), and VideoNotes (VideoNot.es, 2013). Related literatures are also limited particularly on systematic studies that examine how various learning resources (e.g., textbook, lectures, videos, and so on) and pedagogical support (e.g., instructor-provided learning guides, prompts for reflection during discussion, feedback, and so on) collectively affect the learning outcomes.
Video Cases and Field Recordings
What is video cases?
Video teaching cases originated from text-based teaching cases initially developed in the late 1980s. Text-based cases contained narratives (sometimes with expert commentaries) about issues in teaching and learning within a given context. And teachers participated in a series of coordinated case discussions after studying those cases. In the mid 1990s, video cases started to appear (e.g., Copeland & Decker, 1996) though it never became the industry standard possibly because texts still worked better for story telling (see Sherin, 2004).
In TPPD curriculum, video cases are frequently used as the basis for teacher reflection and to help teachers develop professional knowledge base (Sherin, 2004). Group discussion is often the pedagogical tool used to promote reflective learning contextualized or anchored upon the video cases (e.g., Schrader et al., 2003; Sherin & Han, 2004). Video cases deliver highly complex, interactive scenarios that novice teachers can examine with ample time and pedagogical assistance. Generally speaking, the goal is to allow novice teachers to experience and reflect on classroom behaviors that they often don’t have access to or are unable to capture. The support of a strong community of practice is desired to engage teachers in productive discussions of video cases (Cannings & Talley, 2002; Schrader et al., 2003).
What is field recordings?
Field recording has been continuously used in TPPD programs for over four decades, though its usage fundamentally transformed across time (Sherin, 2004). It was initially used as a teacher assessment tool to save administrators’ time in F2F supervision. However, this traditional use case is later found contradictory to research findings that systematic supervisor support is usually required in order to help novice teachers deal with personal factors and improve teaching practices (e.g. Hoekstra & Korthagen, 2011). Since the mid 1990s, field recordings has also been used to provide pedagogical support for teachers such as to promote reflective and collaborative learning in teacher club contexts (e.g., Richardson, 1998; Tochon, 1999).
The reason for reviewing video cases and field recordings in the same use model is because their usages are increasingly more similar in recent years. Both models involve real classroom videos and the activities to conduct reflection, feedback, and discussion based on the video. The difference is video cases primarily involve pre-selected teaching scenarios that represent typical instructional issues, while field recordings arises directly from classroom teaching. Strong collaboration and sustaining community support are emerging themes in both models which are required for significant teacher change. The key is to prompt teachers to go beyond the video and connect pedagogical theories with their professional experiences. To achieve that goal, video-based discussions are often coordinated by veteran teachers, professors, or supervisors who can help inexperienced teachers understand the rationales and expert thinking behind teaching practices (e.g., Sherin & Han, 2004; Koc et al., 2009).