Individual Learner Support in Digital ELT Courses: Insights from Teacher Education

Individual support is crucial to help learners improve, but is complicated in times of digital teaching and physical distancing. The present paper demonstrates how individual support can nevertheless be accomplished in digital ELT (English language teaching). It reports on two teacher education courses in Germany that were conducted in a purely digital form as a response to the unprecedented coronavirus situation in 2020. The purpose of these courses was two-fold: On the one hand, the aim was to develop preservice teachers’ (PTs’) skills of providing individual support to their prospective EFL (English as a foreign language) learners by means of screencast feedback (SCFB). On the other hand, the teacher educator (TE) sought to offer individual support to the PTs in order to promote their feedback skills and advance their English language proficiency even further, especially with respect to academic writing. The multifaceted challenges arising from this complex objective were met through a peer SCFB approach that resorted to a purposeful combination of live webmeetings and individual consultations as well as software tutorials, instructor-generated videos and step-by-step manuals. Also, regular polls and screensharing during the webmeetings helped to keep track of the PTs’ progress and understanding. In addition, online surveys gave the PTs room for regular reflection and provided the TE with opportunities for formative assessment. Finally, cognitive, affective and strategic support was offered through group work in digital breakout rooms as well as through individual consultations with the TE and the tutor. Given the novelty of the digital course design, the paper will close with a reflection on its affordances and challenges and suggest potential modifications for future teaching and research.

will report on two teacher education courses in Germany that were conducted in a purely digital form as a response to the unprecedented coronavirus situation in early 2020. A variety of tools and methods were utilized in an integrated and meaningful manner to fulfill two major goals at once. On the one hand, the aim was to develop preservice teachers' (PTs') skills of offering individual support to their prospective EFL (English as a foreign language) learners by means of multimodal screencast feedback (SCFB). On the other hand, the teacher educator (TE) sought to provide individual support to the PTs in order to promote their feedback skills and advance their English language proficiency even further. Since the PTs strive to become English language teachers in the near future, they need to build up a very high target-language proficiency during their university education, as reflected in the competent use of academic English, for instance. The content focus of the seminars was therefore set on feedback skills applied to academic writing in English in order to fulfill this two-fold challenge. This daunting objective had to be realized within the confines of a purely digital sphere. Hence, the PTs were not only asked to produce a multimedia product (a feedback video), but also to acquire the required skills in a multimedia environment that was devoid of face-to-face contact.
The present paper describes the methods and tools that I utilized in my digital teacher education courses at the University of Osnabrück, Germany, and discusses the affordances and challenges that were encountered. To set a theoretical framework, the notions of individual learner support and feedback are defined first (section 2.1) and possible ways of minimizing the adverse effects of physical distancing in digital environments are outlined next (section 2.2). The ambitious course aims of simultaneously developing PTs' multimodal assessment literacy and academic writing skills are anchored theoretically (section 2.3) before their concrete implementation in the two digital EFL teacher education courses is presented (section 3). The paper concludes with reflections on the affordances and challenges associated with the complex course design and with recommendations for fellow ELT practitioners.

Rationale
To lay the theoretical foundation and contextualize the primary goals of the two EFL teacher education seminars, the current section clarifies the key terms and highlights their relevance for foreign language learners and teachers. Special emphasis is given to research evidence regarding learner support in digital environments. This helps to delineate the research and practice gap that is addressed by the present teacher education project.

Providing individual support through feedback
Learning a foreign language requires various competencies from students, which are typically developed under the guidance of others, notably teachers or peers. To be effective, this guidance or scaffolding should take place within the individual learner's Zone of Proximal Development (ZPD). According to Vygotsky (1978), it is "the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers" (p. 86). In other words, it addresses the difference between a learner's potential level of performance (that they may attain through sufficient assistance) and the learner's present performance (that they are already able to accomplish on their own). While assistance is crucial, the ultimate aim should be to help learners move from other-regulation to self-regulation (Brown, 2020, p. 100) by providing as much mediational scaffolding as needed, but not more than needed (cf. Aljaafreh & Lantolf, 1994, p. 468).
In that respect, one of the most influential mediational activities offered by teachers or peers is feedback (Hattie, 2009, p. 12). It is learner support that is provided in response to their performance and aims to assist them in closing the gap between their current level of achievement and the desired learning goal (cf. Ramaprasad, 1983, p. 4;Sadler, 1989, p. 120). To accomplish this, feedback should not only address the strengths and weaknesses of learners' present performance, but also give concrete recommendations as to how they could improve in the future. Hattie (2012) therefore stated, " [t]he aim is to provide feedback that is 'just in time', 'just for me', 'just for where I am in my learning process ', and 'just what I need to help me move forward'" (p. 122). However, this kind of personalized assistance might be complicated in purely digital ELT, as the following section elucidates.

Reducing transactional distance in digital teaching
Quite often, teachers make extensive use of written communication in digital courses (Grigoryan, 2017, p. 84), even though this communication mode is laborious for teachers (e.g. Anson, 2015, p. 375) and prone to misinterpretations by students (e.g. Duncan, 2007;Zamel, 1985). Moreover, oral skills are neglected, although they are of utmost importance for foreign language learning and crucial for creating rapport between teachers and students (Anson, 2018, p. 34;Anson, Dannels, Laboy, & Carneiro, 2016, p. 397;Henderson & Phillips, 2014, p. 6;Stannard & Mann, 2018, pp. 99-100). For instance, Grigoryan (2017) detected statistically significant differences on the rapport dimension between online learners who received audiovisual teacher feedback as opposed to the text-only feedback groups, who scored lower. She therefore argued for the use of audiovisual feedback in online composition classes, building on Moore's (1993Moore's ( , 2013 theory of transactional distance as well as the Community of Inquiry (CoI) model advanced by Garrison, Anderson, and Archer (2000). The two central and interrelated tenets are as follows: Moore's theory of transactional distance […] explores how differences in space and time between the learner and educator create a psychological and communicative distance resulting in barriers to communication and understanding between students and instructor. (Grigoryan, 2017, pp. 89-90) According to the CoI model, deep and meaningful online learning occurs through the interaction of social presence, cognitive presence, and teaching presence. (Grigoryan, 2017, p. 91) Hence, the aim is to reduce the transactional distance by increasing social, cognitive and teaching presence. One way to do so is the use of screencast feedback (SCFB). It denotes the production of a video file by means of a screen-capture tool, electronic annotations and audio comments in order to provide multimodal feedback to a recipient (see Figure 1).

Figure 1. Screencast feedback (SCFB)
When using this asynchronous audiovisual feedback method, the assessors record their on-screen activities of navigating through an electronic assignment while editing, highlighting or annotating sections of it and simultaneously providing oral commentary with explanations and suggestions to help the recipients improve their performance in the future (Brick & Holmes, 2010, p. 339;Henderson & Phillips, 2014, p. 5;Vincelette & Bostic, 2013, p. 258). This gives the impression of a live marking session of a student's assignment (Jones, Georghiades, & Gunson, 2012, pp. 593-594). Oral and written feedback are utilized in a synergistic manner to compensate for the shortcomings of each mode: The ephemeral nature of oral feedback is surpassed (Ryan, Henderson, & Phillips, 2019, pp. 1508-1510 and the typical terseness and ambiguity of written commentary is supplanted by the richness of the oral mode (Brick & Holmes, 2010, p. 340;Duncan, 2007, pp. 273-274, 277;Edwards, Dujardin, & Williams, 2012, p. 97;Weaver, 2006, p. 381). Students do not need to decipher cryptic written comments or infer the meaning of unfamiliar error codes, while instructors do not need to engage in the time-consuming activity of writing detailed feedback prose (Anson, 2015, pp. 375-377). At the same time, the oral mode allows for a more conversational approach to feedback, which may convey a personal connection that is comparable to individual face-to-face meetings (Grigoryan, 2017, pp. 100-101). Also, numerous postediting options exist through the use of specialized software, for instance by inserting text boxes as structural elements (see "Language" and "repetitions" in Figure 1) or multimedia resources for further information about a topic or a grammar rule the learner is struggling with (see Schluer, 2020).
Consequently, social presence can be achieved through the perception of a greater 'personal touch' of SCFB; cognitive presence through the exchange of information and clarification of misunderstandings; and teaching presence through the additional explanations that are enabled by the audiovisual mode (Grigoryan, 2017, p. 105). Indeed, a number of studies have highlighted the benefits of SCFB with respect to reducing the transactional distance that would otherwise exist in online courses (Anson, 2015, pp. 376, 387;2018, p. 34;Crook et al., 2012, pp. 394-395;Edwards et al., 2012, p. 105;Grigoryan, 2017, pp. 100-101, 104;Mathieson, 2012, p. 151;Orlando, 2016, p. 158;Stannard & Mann, 2018, pp. 102, 110;see Murphy & Rodríguez-Manzanares, 2012, for a comprehensive overview of factors that are beneficial to rapport in distance education). However, even though SCFB is considered favorable for online learning, there is an almost absolute lack of studies that explore the implementation of peerto-peer SCFB. Instead, the peer review modes typically remained traditional in nature, with peers either providing oral comments or written comments to their partners (Silva, 2017, p. 329;Walker, 2017, p. 359). In fact, only two studies were found in which a peer SCFB approach was employed, i.e. Silva (2017) and Walker (2017). Neither of them, however, was situated in a teacher education context. Instead, both were mainly targeted at freshmen students in general composition and research writing courses. While the peers benefited from this approach to some extent, the scholars also identified challenges that still needed to be overcome, both when producing SCFB and when using it for the revisions (Silva, 2017, pp. 337, 339-341). Notably, not only were technical problems encountered by several learners (p. 337), but ineffective feedback practices from past instructors seemed to have been replicated by the peers (p. 335), such as very shallow and generic feedback comments that lacked further guidance and explanation.
For this reason, students should not only be trained in the use of technological tools and devices, but also in the provision (and reception) of supportive feedback. Moreover, they should be granted sufficient time for practicing its application. Otherwise, they might find a multimodal peer feedback approach too overwhelming (Silva, 2017, p. 338). Walker (2017), for instance, invested effort in training her students with respect to the software functions and the delivery of feedback, but only gave them limited time to actually produce the SCFB (p. 370). For the freshmen composition students, this turned out to be so demanding that more than two thirds (68%) of the respondents eventually stated a preference for written comments over screencasts (Walker, 2017, p. 372). As peer review itself is demanding, the addition of a digital, notably multimodal, dimension might have been too ambitious within that short time frame (Walker, 2017, p. 373). Especially for PTs, sufficient preparation and practical application are crucial, since feedback delivery is a core part of their future job. In light of the current demands of digitalization, the development of PTs' multimodal language assessment literacy should therefore be a central element of contemporary teacher education. It might be fostered through a peer SCFB approach (Walker, 2017, p. 360) if conducted in a considerate and stepwise manner. This is particularly important whenever learner products need to be assessed on various layers, as is exemplified by the complex nature of academic texts (see next section).

SCFB approach
Prospective teachers face a two-fold challenge with regard to English as a foreign language (EFL): On the one hand, they need to practice teaching that language to students; on the other hand, they are learners of English themselves and need to further improve their target-language proficiency. Some of the PTs' linguistic insecurities become most apparent when they produce academic assignments as part of their studies, such as oral presentations or written papers. Typically, TEs (notably higher education staff) then grade these assignments by utilizing oral or written modes for (rather brief) feedback. Hence, as expounded above, the feedback that PTs (as all other students) receive is often limited in scope and mode. Thus, if they are not given the chance to familiarize themselves with alternative feedback methods, the PTs will probably replicate the restricted range, amount, depth and modes of feedback that they had experienced as students at school and at university. It is therefore crucial to make PTs aware of the multitude of feedback methods that exist and to give them the opportunity to apply them. In that regard, a peer approach enables them to experience both perspectives at the same time: the producers' perspective of a language teacher and the recipients' perspective of a language learner (Schluer, 2020; in press).

Table 1 Benefits for Producers and Recipients of Peer SCFB Dimension
Benefits for SCFB producers Benefits for SCFB recipients Written skills Improvement of academic writing Improvement of academic writing skills skills through critical reflection and through application of the peer's feedback assessment of a peer's paper comments Oral skills Improvement of speaking skills and Improvement of listening skills in the explanation/ communication skills in target language the target language Scope and Feedback at macro-and micro-Global overview and focus on details: depth of structural levels: "zooming in" elaborate and explanatory feedback with feedback possible for praiseworthy or feed-forward suggestions problematic passages Facilitation of Self-monitoring (metacognition) Autonomous application of feedback due self-regulated facilitated through familiarization to its enhanced clarity, comprehensibility learning with assessment criteria and transparency Digital skills Development of digital teaching skills Development of digital learning skills Flexibility: Further editing possible, e.g. regarding Step-by-step corrections possible by availability the sequencing and scope of contents pausing and re-watching the video at and adjustability any time Preferred Suitability for different teaching styles/ Suitability for different learning styles/ processing style preferences (aural, visual, kinesthetic) preferences (aural, visual, kinesthetic) In fact, the benefits of such a peer SCFB approach are numerous (e.g. Ali, 2016;Séror, 2012;West & Turner, 2016; see also Sun &Doman, 2018, andZhu &Carless, 2018, for peer feedback in general), as shown in Table 1 (cf. Schluer, 2020, p. 5).
To provide multimodal feedback, the peers exchanged academic text drafts with each other. Academic texts were chosen for two reasons: First, the aim was to foster the PTs' academic writing skills because they are crucial for their studies. Second, academic texts are typically quite long and complex, which gives the PTs the opportunity to engage with a multitude of assessment criteria and to learn how to provide focused feedback in a constructive manner. In fact, several scholars contended that SCFB would be particularly useful for the assessment of complex works (Stannard & Mann, 2018, p. 103), i.e. when written work is "simultaneously evaluated on theoretical, empirical, compositional, stylistic, and research design components, among many others" (Anson, 2015, p. 376). Besides, SCFB is highly suitable for the assessment of visual work, such as graphic visualizations, animations (O'Malley, 2011, p. 28), games (Law, 2013), architectural designs (Comiskey, 2012), films (McCarthy, 2015, p. 162) and websites (Borup, West, & Thomas, 2015, p. 179;Perkoski, 2017, pp. 45, 47, 51-52), including blogs, e-portfolios, surveys and other digital course contents (Stannard & Mann, 2018, pp. 101, 103). The creation of digital learning materials and virtual learning environments is specifically relevant for (preservice) teachers, especially in the time of the coronavirus. Hence, learning how to produce digital contents, such as through a screencast video, will prepare the PTs for the digital demands of their future job.
Likewise, gaining confidence in learner assessment is key to the teaching profession. As Hattie (2009) found, feedback is "the most powerful single influence enhancing achievement" (p. 12). It is learner support that should assist the students in closing the gap between their current performance and the desired learning goal (cf. Ramaprasad, 1983, p. 4; see section 2.1). Resonating with the idea of selfregulated learning (SRL), feedback is thus allotted an important function in guiding students to regulate their learning process successfully (Sadler, 1989). SLR is a multidimensional construct that comprises various cognitive, metacognitive, affective and (inter-)actional aspects (Nicol & Macfarlane-Dick, 2006). To develop this multidimensional competence, Nicol and Macfarlane-Dick (2006) suggested "seven principles of good feedback practice that support self-regulation" (p. 199; emphasis omitted). It should (1) help clarify what good performance is (goals, criteria, expected standards); (2) facilitate the development of self-assessment (reflection) in learning; (3) deliver high-quality information to students about their learning; (4) encourage teacher and peer dialogue around learning; (5) encourage positive motivational beliefs and self-esteem; (6) provide opportunities to close the gap between current and desired performance; (7) provide information to teachers that can be used to help shape the teaching (p. 203). Due to its multimodal nature, SCFB has particular merits in achieving this, as the foregoing discussion has shown. Even though it does not allow for immediate interactions with the feedback providers (Grigoryan, 2017, p. 101;O'Malley, 2011, p. 30), SCFB creates the impression of an interactive exchange (Anson et al., 2016, pp. 397, 399;Bakla, 2020, p. 119). On the one hand, the oral mode easily allows for the use of conversational and mitigated language, as would be typical of face-to-face conversations. This includes the utilization of personal and possessive pronouns as well as of tentative language when giving recommendations to learners. On the other hand, the specific explanations and suggestions recorded in the SCFB may stimulate learners' profound engagement with the feedback contents and might trigger follow-up discussions (Vincelette & Bostic, 2013, p. 265). This is in line with the conceptualization of feedback as a dialogue (Nicol, 2010) and can most suitably be realized by means of a peer approach (cf. Zhu & Carless, 2018, p. 883). A purposeful exploitation of the multimodal affordances of SCFB may therefore result in personalized support that is particularly valuable for online and distance learning (e.g. Anson, 2015;Grigoryan, 2017;Mathieson, 2012).
However, as any "medium […] is only as effective as the message contained within it" (Cranny, 2016, p. 29117), TEs should strive to develop their PTs' assessment literacy in a profound manner (Harding & Kremmel, 2016, p. 415;Popham, 2009, p. 5). This multidimensional concept refers to the knowledge (what), principles (why) and skills (how to) needed to perform assessment tasks (Davies, 2008, p. 335, as cited in Harding & Kremmel, 2016, p. 418;Inbar-Lourie, 2017, p. 257). For EFL teachers, it additionally comprises knowledge about the language(s) and the multilingual realities in the classroom, including the role and use of English as a lingua franca and translanguaging (Inbar-Lourie, 2017, p. 260). While research on multilingual assessment has gained some ground in past years, the investigation of assessment as a multimodal activity is still widely unexplored (Silva, 2017, pp. 327, 342). With the rapid rise of digitalization in ELT due to the coronavirus-induced shutdown of schools and higher education institutions, multimodal assessment methods definitely deserve closer scrutiny from both practical and empirical perspectives. This is the research and practice gap that the current project is going to fill. Its implementation in digital ELT courses will be expounded in the next section.

Implementation
This section outlines the procedure that was adopted to foster the above-mentioned competencies among the PTs. To start with, the first subsection describes the general teaching conditions that students and faculty faced as a result of the coronavirus-induced lockdown.

Preliminaries
The preceding sections have shown that valuable learner support can be provided via multimodal feedback. It is therefore crucial to develop PTs' multimodal feedback literacy in teacher education courses to prepare them for the demands of their future job. For this reason, the TE (I, the course instructor) had originally planned to conduct two face-to-face seminars with PTs in pre-equipped computer rooms in summer term 2020 (running from April to September). Due to the sudden coronavirus-induced lockdown of the university's facilities, however, I had to transform these courses into digital seminars on rather short notice. Hence, not only the course product, i.e. multimodal feedback in ELT, was digital in nature, but also the entire process of developing the manifold competencies among the PTs that would ultimately lead to that end product. As shown in Figure 2, they comprised digital and didactic competencies, social and strategic competencies, language competencies for the production and reception of SCFB as well as academic competencies for composing and assessing academic texts.

Figure 2.
Overview of competencies to be developed in the digital teacher education courses Clearly, the development of all these competencies requires a careful didactic design, especially when implementing it in a digital way for the first time. However, due to the sudden transformation from faceto-face to online teaching, this constituted a major challenge for the TEs. To start with, the university's IT department first had to create the conditions that would make digital teaching a successful endeavor. Since the platform StudIP already existed as a Learning Management System (LMS) beforehand, the main task for the IT department was to increase server capacities, especially to allow for live webmeetings between instructors (TEs) and students (PTs). The weeks before course start were therefore characterized by gradual changes in the seminar design, followed by multiple modifications that continued until mid-semester. These fluctuations and flexible adaptations were caused by the piecemeal delivery of information as well as the continuously changing advice from higher education boards and ministries. Even the exact start and end dates of the semester were unclear for a very long time. Naturally, this was a result of the unpredictable nature of the coronavirus spread and the concomitant uncertainty as to how such a situation might best be handled. Without doubt, the students also did not know what to expect and what digital teaching would look like in their classes. In the meantime, higher-education staff began to develop individual solutions for implementing their seminars in a virtual space, while the IT support center gradually introduced different digital tools to them through their newly created webinars.
Generally, the IT support recommended using asynchronous tools to a large extent, especially for courses that took place during peak times (9 a.m. to 4 p.m.). The aim was to avoid server breakdowns and to offer flexibility for the students in case their technological equipment and internet connections would not allow for regular participation in live webmeetings. However, the two present courses were scheduled for the time between 4 p.m. and 8 p.m., and so it was hoped that live sessions would be feasible. To be on the safe side, the first two lecture weeks started in a purely asynchronous manner, before live meetings turned into a regular component of the overall course design. The exact procedure will be detailed in the next section.

Digital course design
The present section describes the course design that I developed for the two teacher education seminars, including the challenges that had to be mastered at course start (section 3.2.1) as well as the different tools, methods and resources that were utilized (section 3.2.2). In that regard, individual consultations (section 3.2.3) and reflective practice (section 3.2.4) turned out to be central catalysts for reaching the complex course goals.

Course start
Compliant with ministerial regulations, the two teacher education courses "Videos for ELF teaching I & II" began in mid-April and ended in mid-July. The first two weeks were conducted in an entirely asynchronous manner. In that respect, the first challenge was to clarify participation status. Normally, attendance in the first face-to-face meeting was necessary to become a course member, with a maximum set to 18 students per course. Since on-site meetings were not feasible in the digital semester, other ways had to be found to determine course membership. I thus created two major tasks for the first two lecture weeks which had to be completed by the PTs to indicate their willingness to participate in the course.
To introduce the PTs to that procedure, I produced a "welcome video" that was made available on the electronic course platform StudIP. In that video, my face was visible so that the prospective participants could see me. I did this to establish rapport with the PTs and engender confidence as well as encouragement with regard to the general course set-up in times of change and uncertainty. Moreover, the video was post-edited in the program Camtasia through the use of several basic features (intro and outro animations, text boxes, transition slides and effects etc.) to make it more appealing and to communicate essential information via two channels (written key words and oral explanations). This way, the PTs also gained some first insight into post-production possibilities that they might want to use in their own videos later on. The video was posted onto the LMS two weeks before course start and was made available on the first seminar day. In addition, the PTs received the following welcoming e-mail from me on the first day: Dear students, First of all, welcome to our seminar :-) As we experienced in the past weeks, digital competence is of utmost importance for teachers and learners. This seminar perfectly fits the current demands, as it enables you to create your own videos for EFL teaching. Your main task is to create a feedback video by means of screencasting technology (= Assignment A3). This way, you will be able to provide highquality feedback to learners, even if they cannot be in the same place as you are.
To lay the necessary foundations, you first of all need to complete Assignments A1 and A2. At the same time, their successful completion is a prerequisite for being a participant in this course. Please browse through the syllabus to read about all details of this seminar. If you no longer wish to become a member of this seminar, please remove your registration from StudIP so that another student can have the chance to participate. First lecture week (April 20): As a first step (= Assignment A1), you need to fill in the first online survey (available at https://www.soscisurvey.de/[...]) AND afterwards submit an academic paper draft by uploading it onto StudIP (as specified in manual 1). Also complete the other steps that are mentioned in the section for April 20 in our syllabus. Students who don't complete Assignment A1 (online survey) before the given deadline will give room to a person from the waiting list. Second lecture week (April 27): Please fulfill the tasks that are mentioned in the section for April 27 in our syllabus. Make sure to select a topic via the scheduler of the DFN (German Research Network; see "Terminplaner" link on our syllabus). Up to 3 persons can work on the same topic. You will need to complete a form and present your results on May 4th (webmeeting). General remarks: Please use the "forum" tool of StudIP to ask questions that might be relevant to everybody (instead of writing individual e-mails). Furthermore, you can get in touch with your tutor […] who will offer a tutorial every Tuesday from 12:00-14:00 (please register for her StudIP course "Tutorial: Videos for EFL Teaching I & II").
All required materials will become visible on StudIP shortly before the start of our sessions.
[…] Thank you and best regards, Dr. Jennifer Schluer The e-mail gives a first indication of the variety of tools and resources that were utilized in the courses and that will be surveyed in more detail below (section 3.2.2). Moreover, the bold print of the word "syllabus" was meant to emphasize its key role as the main instrument for communicating organizational issues. The syllabus was updated on a regular basis on StudIP whenever changes in course organization became necessary. It contained elaborate information about all tasks that were due and all steps to be taken. Thus, it was more comprehensive and detailed than previous offline syllabi which were typically reduced to bullet points for the topics tackled in each session. Consequently, the PTs were able to use the syllabus as their main guide for orientation about all tasks and requirements and to access the pertinent resources by means of direct hyperlinks.

Overview of tools, methods and resources
Overall, I resorted to a multitude of digital tools, methods and resources, as shown in Figure 3. The figure illustrates the purposeful combination of live webmeetings and individual consultations as well as of software tutorials, file-sharing, instructor-generated videos and step-by-step manuals. Regular polls and screensharing during the webmeetings helped to keep track of the PTs' progress and understanding. In addition, online surveys gave the PTs room for regular reflection and provided me, the TE, with opportunities for formative assessment. Finally, cognitive, affective and strategic support was offered through group work in breakout rooms as well as individual consultations with the tutor and the TE. The distinct digital key tools, resources and methods are marked in different colors and with unique icons in the figure: manuals (pink), literature (grey), surveys (dark purple), online tutorials and instructor-generated explanations (yellow), general discussions and demonstrations (light blue) and individual consultations (dark blue). A purple background stands for asynchronous delivery while a green background represents synchronous methods. The numbers in the figure indicate the order of events and have been added for easier reference.
The webmeetings were arranged on a weekly basis and conducted by means of the BigBlueButton videoconferencing application that was integrated into the StudIP platform. BigBlueButton allowed for screensharing and the live delivery of presentations. Whenever there was a need for further clarification, the participating PTs were encouraged to type questions into the "public chat". I then gave an immediate response if the question was self-evident, or asked the PTs to turn on their microphone in order to elaborate on their question orally. Furthermore, BigBlueButton enabled the creation of breakout rooms for group work as well as the implementation of quick live polls. Group work results were either inserted into the "shared notes" pad on BigBlueButton or uploaded as documents onto StudIP for permanent access. The "shared notes" pad also turned out to be suitable for impromptu brainstorming activities, since it was possible to re-organize the collected ideas and to highlight specific information with some basic text editing tools (bold print, italics, underlining etc.).
The PTs and I utilized these BigBlueButton functions from the first webmeeting onwards, i.e. from early May to mid-July. In the preceding two weeks, the PTs had to complete two assignments, A1 and A2, as stated above in the e-mail. While A1 was the initial online survey that encouraged the PTs to reflect on their prior knowledge and experience with regard to feedback as well as digital teaching and learning methods, A2 required them to familiarize themselves with the theoretical literature about feedback. In the face-to-face courses of previous semesters, the participants were asked to conduct database searches on their own in order to identify suitable publications. Given the extraordinary circumstances of the digital semester, however, I pre-selected relevant publications and asked the PTs to enroll into one of the following six groups: (1) Effective feedback (Hattie & Timperley, 2007) (2) Feedback conditions (Gibbs & Simpson, 2005)  ) I provided each group with a set of guiding questions, which the PTs were requested to answer by filling in a template that I created for A2. Moreover, they were asked to present their findings as expert groups in the next session, i.e. in the first live meeting on BigBlueButton. To do so, the presenters had to turn on their microphones and to share their screen in order to display their answers to the guiding questions. This way, they were also able to practice speaking into a microphone, as would be necessary for the audio part of their screencast feedback later on. In addition, trialing the use of the screensharing function was deemed conducive for speeding up the screensharing process during the individual consultations that were conducted from session 5 onwards (boxes 8, 11 and 14 in Figure 3), as will be further explained below.
In total, there were six assignments (A1 to A6), but the core assignment was A3, i.e. the production of SCFB for a peer's academic paper draft. Figure 4 summarizes the underlying rationale as well as the approach that was taken for A3.

Figure 4. Peer SCFB
Due to the complex and multifaceted nature of SCFB production, the PTs were introduced to the procedure in a stepwise manner and received comprehensive manuals for each phase of the process. As a first step, the PTs were asked to submit an anonymized 10-page draft of an academic paper written in English. They were informed about the anonymization of personal information in the first webmeeting and through manual 1. While initial versions of the manuals had already been created for a previous course (a face-to-face seminar reported by Schluer, 2020), they had to be revised for the digital seminars to include more details. Thus, effort was taken to ensure the accessibility of information even if students encountered interruptions during the live meetings or could not attend them. Following the basic idea of the inverted classroom model, the PTs were requested to familiarize themselves with the theoretical, methodological and technological knowledge at home (at least to some basic degree) and to apply the knowledge and discuss open questions in the live meetings (cf. Brame, 2013;Handke, 2012, p. 94;Lage, Platt, & Treglia, 2000, p. 32). More precisely, the procedure was as follows.
Prior to the webmeetings, the PTs had to consult specific materials (e.g. manuals, videos, texts) that had either been taken from academic journals or relevant websites or had been created by me, the TE. For instance, I produced two videos for the PTs in which the underlying didactic idea of the seminar was introduced and very brief sample sequences of SCFB were shown. Yet, as stressed throughout, this mainly served the purpose of giving the PTs a basic idea of what SCFB was about while encouraging them to be creative in their own production.
During the webmeetings, I conducted live demonstrations of specialized software functions. As for the text corrections, different editing, highlighting and commenting tools were introduced, such as the "track changes" function and the purposeful display or concealment of mark-ups in the correction mode. With regard to audio-and screen-recording as well as video editing, I explained all basic functions in the webmeetings and compiled a comprehensive manual (manual 6) for the PTs. A major reason why Camtasia was chosen as the video editing program for these courses was the wealth of training resources that is available for self-directed learning on the producer's website. Hence, the PTs were able to consult online tutorials for virtually every program function whenever they needed it. Apart from that, they were free to contact the tutor and me whenever they had questions.

Individual consultations
In addition to the general webmeetings, individual consultations were scheduled regularly. For this purpose, large parts of several sessions during the SCFB production period were reserved for one-toone discussions with the PTs. Each of these sessions started and ended with a collective webmeeting (about 15 to 20 minutes each) and was interspersed with individual consultations in separate meeting rooms (about 7 to 8 minutes per PT). To give everybody a chance for a weekly consultation and to allow for smooth transitions, every PT was assigned a particular time slot during which they talked to the tutor or to me on a one-to-one basis. During the pre-arranged time slots, every PT entered the individual consultation room and shared their screen via the BigBlueButton screensharing tool. This way, the tutor and I were able to observe the PTs' progress and provide scaffolded learner support. The procedure resembled that of face-to-face seminars, in which the tutor and I walked around in the computer room and talked to every PT individually about their specific plans, progress and questions regarding feedback principles, academic writing, electronic commenting in a text editor and video production in the video editor.
After these individual consultations, everybody rejoined the general webmeeting room, where I reported about recurring problems and solutions and gave specific instructions for the steps that needed to be taken before the next session. Similarly, each collective webmeeting started with the introduction of new information. For instance, I demonstrated important software functions that the PTs were asked to practice and apply while they were waiting for their individual consultation to begin. This so-called "webmeeting burger" or "webmeeting sandwich" is shown in Figure 5 and illustrates the synchronous elements 8-9, 11-12 and 14-15 of Figure 3. Similar to the face-to-face seminars, the digital individual consultations turned out to be critical in order to provide individual support to the PTs, especially since every draft and resultant video was unique and raised highly specific questions.
In the first two weeks of the individual consultations (boxes 8 and 11 in Figure 3), the PTs had the chance to discuss their assessment process with the tutor and me, the TE. This comprised a general and tentative evaluation of the peer's draft, the identification of relevant assessment criteria that might be applied to the paper and their actual application in a pedagogically sound and technologically appropriate manner. Since the individual papers dealt with unique topics, the PTs had to search for relevant information about that topic (notably within the fields of literary or linguistic analysis) and to derive appropriate criteria for assessment, e.g. concerning the analysis of novels, poems, classroom interactions, teaching materials etc., but also about general aspects on micro-and macro-structural levels of content, form and language. Some sample assessment criteria were listed in manual 3 and comprised the following ones, for instance: • Content: breadth/ scope; depth, e.g. depth of analysis; line of reasoning/ argumentation/ flow/ coherence/ idea development, including paragraph structure; understanding of topic and literature/ correctness of contents; use of relevant evidence from the published literature and appropriate use of examples; • Form: formatting; integration of resources and correct citation; • Language: word choice; grammar; fluency; clarity; conciseness; style/ voice; cohesion, e.g. use of transitional devices. Since the potential scope of feedback comments was wide-ranging, the PTs were instructed to identify those aspects that would be of greatest relevance and benefit to their peer. For example, when certain grammatical errors recurred, the PTs were asked to point them out. The overall length of the resultant feedback video, however, should not exceed 5 to 6 minutes, because this duration appeared to be the norm in the SCFB literature (e.g. Grigoryan, 2017, p. 92;Moore & Filling, 2012, p. 10;Vincelette & Bostic, 2013, p. 270) as it may help avoid overwhelming the recipient.
Finding a focus, structuring the feedback and balancing positive and negative comments turned out to be one major challenge for the PTs. Another challenge resulted from the broad range of software functions that was available for visual mark-up in the peer's draft and in the screencast. To assist the PTs in the purposeful selection of visual feedback strategies, several video editing tools were introduced in the third and fourth rounds of individual consultations (see box 14 in Figure 3). While some PTs were eager to explore the various functions and effects of the text and video editors, others were less confident due to their rudimentary computer skills. In some individual consultations, the tutor and I therefore had to reiterate the basic steps regarding file storage and software use, whereas in others in-depth questions about specialized software functions were addressed.
As soon as they had finished their work on the video files, the PTs shared their materials with the tutor and me. In previous face-to-face seminars, file sharing was relatively easy because the PTs converted their files into .mp4 videos by accessing fully licensed Camtasia versions in the computer room. In the digital courses, however, the PTs had to resort to the trial version of Camtasia that was restricted to 30 days of use and resulted in videos with large watermarks that made the reading of screen details impossible. The PTs therefore packed all their raw materials into a .zip folder (by using the Camtasia export function) and uploaded the folder onto their personal storage space of the protected university cloud MyShare. After that, they created a private link for sharing the files and submitted it via the electronic learning platform StudIP. This additional step of using cloud space became necessary due to file size restrictions of the LMS StudIP. Upon reception of the link, the tutor and I converted the videos, uploaded them onto Screencast.com and distributed the video links to the correct recipients via a private hyperlink (unlisted videos that are not searchable online; cf. Cunningham, 2017, p. 47;Vincelette & Bostic, 2013, p. 262). This way, the peers obtained high-resolution videos with anonymous peer feedback that they could use to further improve their academic writing.

Reflective practice
Immediately after they had finalized their videos, the PTs were encouraged to reflect on their production experience in survey A4. Similarly, upon reception of their peer's feedback video, they shared their perceptions and learning gain as part of the reception survey A5.
Finally, another reflective dimension was added. In the written end-of-module task, the PTs had to closely analyze the feedback video that they had created and to reflect on potential improvements and applications in their future job. This is in line with the idea of teachers as reflective practitioners (see e.g. Brandenburg, Glasswell, Jones, & Ryan, 2017;Schön, 1983;Smith, Geng, & Black, 2017). Furthermore, it helps them prepare for their final Master thesis, which often requires them to conduct a study on their own, e.g. by analyzing school lessons or specific ELT-related phenomena. In the two digital ELT courses, the PTs were asked to scrutinize the feedback strategies they had employed in the SCFB by engaging with the theoretical literature and by utilizing a software program for qualitative data analysis. For this purpose, I recommended MAXQDA 2020 because its basic functions of coding and retrieval are relatively straightforward and because it offers numerous online resources for self-directed learning (video tutorials, manuals, user support, etc.). With Camtasia and MAXQDA, two well-established programs were thus chosen that were user-friendly and rich in scope at the same time. Nevertheless, the additional live demonstration and discussion of their key functions as well as of their applicability to concrete student projects were deemed essential, together with step-by-step manuals and individual support. Further affordances and challenges as well as recommendations for future ELT practice will be discussed in the subsequent section.

Discussion and Reflection
The unforeseen coronavirus pandemic had far-reaching consequences for people's personal and professional lives. In higher education, nearly all courses had to turn digital within a very short time and so creative ways had to be found to make ELT possible. Clearly, this sudden shift generated unconventional solutions that require reflection in order to improve future teaching. The following subsections are therefore devoted to a critical reflection on the above-mentioned digital implementation. This reflection is written from my perspective as the TE of the two courses and enriched by an initial perusal of the collected PT data.

Affordances
Despite the sudden and unforeseen shift to digital ELT, it was possible to offer individual support and reduce the transactional distance that is usually associated with online education (Moore, 1993(Moore, , 2013, as cited by Grigoryan, 2017, pp. 89-90). In particular, the integrated use of individual consultations, live polls and live exchanges in the plenary as well as in small groups (breakout rooms) turned out to be important to create rapport and provide affective and strategic support. In addition, these tools and methods allowed me to monitor the students' progress as well as to gain insights into the challenges the PTs encountered and the accomplishments that they had already reached. Certainly, the availability of a virtual learning and meeting platform was a basic precondition for achieving these aims, but it was not the major driving force of the course design. Rather, didactic considerations were decisive, i.e. the learning goal and the learning needs of the PTs (cf. Harris & Hofer, 2011, pp. 214, 222;Kultusministerkonferenz [KMK], 2016, p. 51;Wannemacher, Jungermann, Scholz, Tercanli, & von Villiez, 2016, p. 5; see also Mayer, 2005, pp. 7-9, on the distinction between a technology-centered and a learner-centered approach to multimedia learning). In the two digital teacher education courses, the learning goal itself was partly digital in nature, i.e. to develop digital assessment skills among the prospective English language teachers. While scholars in past studies cut short the preparation time for the participants (Silva, 2017;Walker, 2017), I took ample time to systematically introduce the PTs to the multilayered course objectives. Furthermore, the tutor and I offered in-process support throughout the seminars to accomplish personalized scaffolding despite physical distancing.
In the end, the TE's and tutor's individual guidance enabled the PTs to provide individualized feedback and scaffolding to their peers by means of SCFB (see Figure 4). This personalized support is particularly valuable for online and distance learning, as previous studies have shown (e.g. Anson, 2015;Grigoryan, 2017;Mathieson, 2012). In the current project, the PTs likewise appreciated the personalized support they received from their peers, their tutor and me (surveys A4, A5 and A6).
Moreover, the initial survey (A1) indicated that the PTs appeared to be highly motivated and interested in the course contents due to the perceived relevance of digital teaching in the time of the coronavirus and the lockdown of educational institutions. Learning how to provide electronic learner feedback, however, required their full commitment to a complex course design and a willingness to develop several new skills in an integrated manner. Clearly, this process was not without challenges, as the following sections illustrate.

Challenges encountered by the preservice teachers
The questions and discussions during the live meetings and individual consultations as well as the PTs' occasional forum posts and e-mails provided me with some initial insight into the challenges that the PTs faced during the two digital ELT seminars.
Most of the challenges appeared to be technological in nature. First of all, the PTs had to become familiar with the tools that were available in the live meetings, but they quickly felt confident in using them due to the TE's explanations and demonstrations. Interruptions of the internet connection hardly ever occurred and did not result in a substantial loss of information because of the comprehensive manuals that had been prepared in advance. Instead, the challenges seemed to be rather organizational in nature: Given the abrupt shift to digital teaching, every lecturer found their own ways of delivering their courses and communicating with the students. As a result, many PTs felt overwhelmed by managing the flow of information, resources and tools that differed from lecturer to lecturer and from course to course. This was not a challenge arising from the set-up of the seminars reported in this paper, but from the general variability in digital course design at university. Consequently, some PTs wondered whether they had missed anything, especially in the first weeks of the semester. Other PTs did not expect a digital semester to be as laborious as traditional face-to-face seminars and enrolled into numerous courses simultaneously. However, the workload was still as high as before and their participation in the live meetings was mandatory, making parallel attendance hardly possible. Coupled with the unfamiliarity of digital learning and teaching, some PTs therefore changed their membership status in the different seminars, leading to some fluctuation in course memberships in the first weeks. To avoid late drop-outs, I utilized the syllabus as a detailed and transparent medium for communicating the course contents, goals and tasks. All requirements were made evident in the first session so that the PTs knew what was expected of them. This way, they were also able to schedule their time well in advance in order to work on the assignments and to proceed in a self-paced and structured manner.
As soon as basic organizational questions were clarified and work on the PTs' SCFB began, more specific technological hurdles became obvious. The PTs possessed different levels of expertise in video production, the use of text processing programs and even basic file management at a computer. While many questions could be clarified by reading the manuals and attending the individual consultations, some PTs still underestimated the effort it took to produce a multimodal feedback video. They invested a lot of time into the preparatory steps of commenting electronically on the peer's written draft or crafting out a detailed plan for their feedback design, but postponed the actual implementation to a very late point because they did not anticipate how much work it involved.
Clearly, SCFB is still a relatively new method that needs more understanding and concrete recommendations for practical application. Many teachers and TEs might refrain from it because it appears to be very laborious at first. For instance, in McCarthy's (2015) study, staff complained about the greater "workload to produce feedback files" (p. 164) and to disseminate them to the students, especially since the entire procedure was novel to them. However, studies also showed that after an initial investment of time, the creation of SCFB becomes faster with experience (e.g. Bakla, 2017, pp. 328-329;Séror, 2012, p. 108;Vincelette & Bostic, 2013, pp. 268, 270;Warnock, 2008, pp. 205, 210). Therefore, Hewson and Poulsen (2014) underlined that "[t]he time it takes to become proficient in utilizing the technology can be viewed as accrued time savings when later using it to provide feedback". Moreover, teachers also need to acknowledge that "the finished product is richer and denser in information than traditional forms" (Brick & Holmes, 2010, p. 341) when setting up a time cost-benefit analysis of different feedback methods (cf. Turner & West, 2013, pp. 293-294). In the end, the feedback method should always be chosen in accordance with the learning goal that is envisioned and the characteristics of the individual learners as well as of the learning environment. (Prospective) teachers therefore must be critically aware of the necessity to put didactic considerations into the foreground instead of blindly following a technological imperative (Harris & Hofer, 2011, pp. 214, 222;KMK, 2016, p. 51). As Borup et al. (2015) explained, Richer and more nuanced communication is not always the best course of action. Ultimately a teacher's expertise should be the deciding factor, as teachers must understand what their students' needs are and provide feedback in a manner that best meets them. Therein lies both the challenge and the imperative for the modern online instructor -to understand what media options are available and when to use them. (p. 181) Similarly, Stannard and Mann (2018) remind us that "using SC feedback is merely the medium, the quality and usefulness of the feedback is still tantamount and depends on the individual teacher" (p. 108). The most important point for the PTs therefore is to become proficient in feedback provision and to make considerate choices with respect to the medium that best fits a particular purpose. SCFB clearly has several advantages, but at the same time continuous critical reflection on the implementation of specific feedback methods is crucial. The PTs were therefore encouraged to reflect on the entire procedure at various points and to submit a final written reflection in which they critically analyzed their SCFB. The evidence-based analysis deepens the PTs' understanding of their SCFB strategies and helps them to suggest possible modifications that they could employ in their future career. Moreover, the written reflections give the TE additional insights into the PTs' challenges, which will be utilized to improve the digital course design even further.

Challenges encountered by the teacher educator
For me as the TE, the first major challenge was to devise a digital seminar design that would be conducive to attaining the ambitious course goals. While I was already experienced in the use of several tools and approaches for digital learning and teaching, a purely digital seminar without any face-to-face meetings was also novel for me. I therefore had to think about possible software solutions and to modify many tasks that had originally been planned for on-site meetings. Because of the general uncertainty that characterized the first few weeks of the summer semester, I had to revise the syllabus several times. Moreover, I put huge effort into the compilation of manuals and the creation of videos for asynchronous use by the PTs. Apart from the course preparation, the actual implementation demanded various skills: It involved a smooth and purposeful integration of different digital tools and applications during the webmeetings and the simultaneous management of the chat questions, shared notes, breakout rooms, polls, presentations and screensharing. In that regard, I considered it important to agree on rules for communication at the start of the seminar, i.e. that the chat was reserved for open questions and the notes section for brainstorming activities or summaries of group work results.
The webmeeting platform that afforded these tools was BigBlueButton. Since the cameras were turned off to guarantee a smooth signal transmission for everyone, I was not able to see the PTs' bodily reactions, e.g. their facial expressions that might have indicated non-understanding. However, the live polls helped me to monitor the PTs' active engagement and understanding while following the lesson contents. Another challenge which resulted from the increased transactional distance (Moore, 1993, cited in Grigoryan, 2017 was that social support among the PTs was limited because spontaneous dialogues were impeded by the physical distancing. Yet, they were able to message each other via private chats during the webmeetings or by means of the forum and chat tools of the LMS. Moreover, at a few points in the individual consultations, I observed that some PTs offered mutual support to each other that extended beyond the confines of the learning platform. For example, some PTs reported that other students had encountered a similar text editing problem and that they had tried to solve it on their own already. Apart from these self-reports, the individual consultations also helped me to identify recurring problems and to address these issues in the collective webmeetings that were scheduled directly after the individual consultations (see Figure 5).
Another challenge had to be overcome towards the end of the seminars when the PTs transmitted the Camtasia project data for conversion into .mp4 files and upload onto Screencast.com. Even though the entire process was explained in the manuals and in the live meetings, several PTs were unsure about the procedures for storing and sharing their files. This suggests that some PTs may require additional training in basic IT skills. Since they did not address these aspects at an earlier stage, several short-hand requests for help reached me in the submission week, which were nevertheless resolved in the end.

Conclusions and Recommendations
In the two digital ELT courses, individual support was not only the course goal for the PTs but also a central element of the seminar design. To that end, the one-to-one consultations with the PTs turned out to be beneficial and can likewise be implemented in other teaching contexts whenever webmeetings are feasible. Apart from the technological infrastructure, the successful implementation requires the skill and will to learn about video production and multimodal assessment methods -not only on the PTs' part, but crucially also from the TE's perspective. Only if the TE is able to model the skills that are expected of the PTs will they be able to master them in the end.
While the current seminars made use of various resources and tools, an integrated LMS solution might be strived for in the future. Especially when it comes to video production, the provision of sufficient cloud space and tools for screen-recording as well as screen-editing would be desirable. The university made Opencast accessible for lecturers at the beginning of the semester in order to directly record simple screencasts for teaching in the LMS. However, this program was not readily available for students and did not allow for complex editing, which is why a different software, Camtasia, was used. Camtasia is a well-established and user-friendly video editor that is compatible with Windows and Mac. Also, in more general terms, I recommend the use of well-established software solutions that offer numerous resources to enable self-regulated learning and customized support, such as through video tutorials and online manuals.
Regarding the social support among the PTs in the digital seminars, it might be worthwhile to organize breakout rooms for small-group exchanges as a regular component of the course design. Although I made the PTs aware of the possibility to use the normal webmeeting room for small-group exchanges even beyond the ordinary meeting slots, it is unclear whether they actually seized that opportunity. However, the individual consultations revealed that at least some PTs tried to assist each other beyond the confines of the digital courses. Nevertheless, further surveys would be necessary to elicit information regarding peer-to-peer support beyond the arranged meetings.
Thus far, the four surveys (A1, A4, A5, A6) that were completed by the PTs offer a rich database for further analysis. They shed light on the PTs' experiences and perceptions as well as their learning gains in terms of individual support in digital environments. Additionally, the peer SCFB and the written reflections provide insight into the success of these digital seminars. Based on the evidence that has been collected so far, I may conclude that the two digital ELT courses turned out to be valuable for the PTs, but they also came with several challenges. Mainly, these challenges revolved around the development of a careful didactic design that allows for mutual exchange and individual support, backed up by a variety of materials for self-paced learning as well as opportunities for constant reflection on the learning progress. Overall, a gradual familiarization with the technology and the subject matter appeared to be essential, but needs to be coupled with continuous practice of the skills in the PTs' future profession. In the seminars, we therefore also discussed concrete future applications, such as potential modifications of the SCFB approach for large classes, young learners, shorter assignments and blended learning settings. It became clear that every course design needs to be driven by didactic considerations instead of technological tools alone (Harris & Hofer, 2011, pp. 214, 222;KMK, 2016, p. 51;Wannemacher et al., 2016, p. 5). Thus, while the availability of an IT infrastructure is an elementary pre-condition for the realization of digital teaching, the integrated development of multifaceted teacher competencies is decisive to make the digital learning experience a successful one. In that regard, the current report aimed to inspire fellow ELT practitioners at schools and universities by outlining ways of providing individual support despite physical distancing in digital environments.