Articles & Research
Online Training of English Language Teachers as a Way to Improve Adult Students’ Results in an Online School (Evidence from the Skyeng Experiment)
The article examines the organization of English teachers’ professional development with the aim of improving the educational outcomes of their students. The author proposes a way of determining the need for professional development, a way of organizing the professional development process, and the means of measuring the effectiveness of that process. Analysis of the empirical material made it possible to establish the foundational didactic principle of the proposed approach: the teacher’s timely professional development on the job. At the beginning of the article the author reflects on the changes brought by the growth of the EdTech market and of online tools in education, which open up both new requirements and new opportunities for solving the problems of every participant in the educational process. Among the pressing problems facing the participants in education, the author singles out the learner’s individual educational request, the demands that companies place on the competencies of employees and candidates in the market, the influence on the learner’s result exerted through the teacher’s continuing professional development, and the automation of teaching aids and of the tools used to analyze them. One of the chief demands of the present day is, of course, the development of particular competencies connected with a person’s social-emotional and digital activity, the capacity to transform quickly in response to circumstances, and the ability to learn and to relearn. To the author, an ecosystem solution to the problems listed appears the most effective path—one intended not to fragment but to unite efforts, coordinating the goals of each participant and satisfying their requests. It is proposed to influence the learner’s result—formulated by the learner personally in accordance with market or individual educational needs—by providing teachers with timely training and support for delivering a good first lesson and the lessons that follow. As evidence of the effectiveness of the approach under consideration, the body of the article examines an experiment run by the team of the training center responsible for teaching more than 10,000 English teachers at the Skyeng online school. The present article offers an exhaustive description of the project, of the proposed tools, the training formats, and the metrics for calculating the effectiveness of the experiment; charts and other visual explanations are used for clarity. The article closes with a glossary that will be useful to everyone experimenting in education, but especially to teachers, methodologists, and managers of educational programs.
Keywords: distance learning, teaching methodology, professional competence, English as a foreign language.
1. INTRODUCTION
The rapid development of information and communication technologies and of educational technologies compels us to rethink the popular saying “A student enrolls in a school, but stays with the teacher.” Surveys show that what matters to adult learners is not only the reputation of the school and of the teacher, but also the educational organization’s ability to offer instruction at a convenient time, and the teacher’s ability, in the course of instruction, to take account of the learner’s needs, interests, and possibilities. The needs of adult learners occupy a special place in an educational organization, because they shift with the changing situation in the labor market and with a number of other factors, an important one of which has been the epidemiological situation in 2020 and the associated rise in the demand for online learning. Employers, meanwhile—from small businesses to large corporations—are generating demand for a new level of competencies, for which it is not enough to receive an education once and use its results for the whole of one’s professional career: one has to upgrade one’s qualifications regularly and acquire additional skills throughout one’s professional life.
In this situation it is especially important for learners to be able to study under conditions convenient for them (often online), with a teacher capable of teaching to a high standard—taking the learner’s needs into account and applying effective methods and means of instruction, such as the communicative approach. Online learning, meanwhile, which is increasingly becoming for the learner a way of reaching a goal, places special demands on the teacher.
This variability in learners’ needs and requests, together with the technological development of education, means that teaching adult learners successfully no longer allows a teacher to be simply an educator: the requirements for one’s professional competence have to be revisited regularly, and one has to be an educator in online education—in other words, an educator in a constantly changing world. Given technological progress and the changing organization of work, both experienced and beginning teachers need not retraining but the correction of their lesson and the introduction into it of methodological changes that raise the lesson’s quality and determine the learner’s final result.
Beyond mastering the technical side of online teaching, a teacher needs to be able to: • diagnose the learner’s goals; • establish contact with the learner when working online; • know and observe the structure of the online lesson; • give personalized assignments; • monitor the balance between Russian and English; • introduce new grammar and vocabulary with techniques that are effective in the online lesson; • deliver the first lesson, and the series of first four lessons, according to the school’s methodology, so that these lessons can subsequently be assessed and the work corrected.
Thus, after a first lesson, a new teacher has to incorporate into their habitual lesson, in a very short time, a whole set of new techniques and features, on the basis of an analysis of the new experience. An experienced teacher, on the other hand—one to whom new learners with changing requests come—must come to the lesson prepared for all kinds of challenges and unexpected situations. Qualifications of this kind are built up deliberately. Let us examine how the school prepares teachers for the specifics of working online.
2. ORGANIZATION OF ONLINE FOREIGN LANGUAGE TEACHING.
AN ANALYSIS OF THE EXPERIENCE OF THE SKYENG ONLINE SCHOOL
All new Skyeng teachers are required to take an introductory training course, in which they learn about the specifics of teaching one-to-one lessons online and the fundamentals of the communicative methodology; but as a teacher’s professional experience grows, so does their need for informational and methodological support and for regular training. A one-to-one lesson between an adult learner and a teacher is a highly complex process, from preparation to the assessment of results. This process includes the selection (and possible reassignment, depending on the learner’s wishes) of the teacher, the approval of the schedule of individual lessons, registration in the school’s systems, the questions that arise along this path, and various administrative approvals. It is important that this path be simple for the learner, and that the learner begin moving toward their results from the very start—which is why the first lessons occupy a special place in teachers’ regular and introductory training.
A distinctive feature of Skyeng’s online lessons is that every lesson is measured against a set of concrete indicators—lesson metrics.
The metrics include (Appendix 2): • lesson quality, and first-lesson quality separately (the sum of methodological and disciplinary criteria analyzed both manually and by a machine algorithm); • teacher discipline (the percentage of planned lessons that actually took place); • student retention (the percentage of learners who continue their studies); • the intensity of lessons with a learner (how often lessons take place over a given period), and others.
These metrics are used to determine the quality of the learning process. Taken together, they make it possible not only to assess the effectiveness of lessons but also to identify problems of a methodological nature and to determine which aspects of methodology a teacher needs help with. For example, the “share of good first lessons” metric, calculated across the whole school and for each teacher, shows that a change needs to be introduced into the teacher’s lesson on the job, in a timely manner and with a statistically significant effect. A threshold value has been established from the combined metrics. If a teacher’s lesson score falls below the threshold, this indicates that the teacher needs professional development, while the specific lesson-quality criteria show which methodological aspects require particular attention.
Thus the need for professional development is determined automatically, on the basis of measurable lesson-quality indicators. Such a system is objective: it takes account of the teacher’s professional skill, saves resources (since only those teachers who genuinely need it are sent for training), and subsequently makes it possible to evaluate the effectiveness of a professional development course, because a successful course leads to higher lesson scores on the same metrics.
3. EXPERIMENTAL TRAINING OF ENGLISH LANGUAGE TEACHERS
IN THE SPECIFICS OF WORKING ONLINE
Let us consider the experimental training that was designed to compensate for the gaps in teachers’ methodological knowledge identified by means of the metrics. These gaps formed the basis of the content of the experimental training. In planning the process, we proceeded from the following premises:
1. The organization of the training must not interfere with the teachers’ normal
work process;
2. The experimental training had to be large enough in scale that with
its help we could influence more than 1,000 teachers;
3. The results of the experimental training had to be reflected in the lessons
conducted by the teachers who had completed it. That reflection of the training in the lessons had to be measurable and statistically significant;
4. By the learner’s results we mean the following: together with the
teacher the learner reached a micro-goal or a larger goal, learned new words in the app, and completed homework assignments successfully.
3.1. An Experiment in Regular Training
The task of the experiment was to influence the quality of the first lesson, which is analyzed by means of manual review and machine data analysis against criteria (chose the recommended lesson, stated the lesson goals, introduced new vocabulary correctly, etc.) that correlate with the metrics of the experiment (Appendix 2)—metrics that depend on the learner’s satisfaction with the learning process and, subsequently, on the learner’s decisions.
Experiment metrics: • GL1 (the share of good first lessons across the school); • ret.2 (the rate at which learners return for the second lesson); • ret.5 (the rate at which learners return for the fifth lesson).
The profile of the teachers who took part in the training: average age 24; 4–5 months of tenure at the school; mostly without a specialized linguistic education; teaching only learners at the beginner level (A1 on the European scale of language competencies). These teachers are characterized by low motivation. They have neither sufficient time resources nor the motivation to study anything outside their regular work routine. A survey we conducted showed that on average a teacher is prepared to set aside 2–4 hours per week for additional professional training and other activities. In other words, this is the time a teacher is willing to spend on training inside the company.
Conditions of the experimental training: on the basis of the metrics, 1,000 school teachers were designated as trainees—those whose lessons with beginner learners were due to take place shortly. The experimental training lasted 2 weeks, and 4 weeks were allowed to bring the cohort to the point at which the results could be summed up. It is important to note that no administrative leverage was used in the experimental training—that is, there was no way to compel the teacher, as an employee, to carry out the task. Additional conditions: besides improving the figures on the metrics listed, the training had a second objective—to increase learner retention (retention) at the 5th lesson—which meant that time-bound target indicators had to be set. The following target results were defined: GL1 at +5% of the company average, and retention of learners at the 5th lesson at +4% of the average; once these metrics were achieved, the results of the experimental training would be extended to all of the school’s teachers. Besides the core metrics, which would tell us the outcome of the experiment—and in order to influence them more precisely—we singled out a class of leading metrics, whose task was to inform us promptly about the state of the process.
Leading metrics: • the learner’s ratings—of the lesson, the content, and the teacher; • the percentage of teachers who applied the new practice in the lesson; • the percentage of teachers who followed the lesson-selection recommendations; • the percentage of teachers who clicked through on the links.
Once a learner appears in a teacher’s schedule, the experiment team has 24 hours in which to influence the outcome of that lesson in some way, and a further 48 hours to give the teacher feedback on the first lesson—feedback that includes data on the outcome of the first lesson and recommendations for conducting the second.
Teachers could also, if they wished, take the full first-lesson course that the Teacher Training Center team had developed on the internal platform. It is worth noting that in the introductory training the teachers had already completed a program of first-lesson practice sessions and had received basic preparation. But teachers with little tenure at the school (4–5 months) have difficulty applying the knowledge they have acquired about conducting the first lesson (the teacher forgets how to run the lesson to our standards, because they do so relatively rarely), and the teachers’ effectiveness on first lessons consequently falls, as we see in the final metrics. A tool is therefore needed that will support teachers before every first lesson. This is a non-trivial task, solvable only through the joint work of several departments, so it is worth noting the specializations of the departments that worked on it. The departments that worked on the problem were: • Content—development of the visual and textual content of the first lesson; • the Teacher Training Center (TTC)—organization of the teacher support process within the test, and resource support (administrative tasks, analytics).
Two instruments of influence were developed to solve the task: the first-lesson content and the ideal first lesson map. The first-lesson content is a tool for teacher and learner alike, within which the content team realized the visual and substantive side of the course—one that helps teachers of any level of experience to conduct successful lessons, reducing (and most often eliminating altogether) gross methodological errors. Silence time in the lesson, for instance, is an important criterion that must be kept within the established benchmarks. The first-lesson content includes assignments of this kind, designed to reduce silence time on the teacher’s part. And for the learner it is a lesson with a clear goal and a clear outcome.
The lesson content can be described as follows: • 10+ slides whose structure and content are aimed specifically at the first lesson—familiarization with the future plan, diagnosis of the learner’s goals and language level; • the lesson has a specific goal, and as a result the learner comes away with a clear outcome and a plan for the next lesson; • the lesson was designed by instructional designers and placed on the school’s internal platform; • the learner rates the lesson on a number of internal metrics: a rating for the teacher, a rating for the content, and a rating for the lesson as a whole.
The ideal lesson map is an instrument, from the Teacher Training Center (TTC), for influencing changes in the teacher’s lesson: within it the teacher receives methodological material in which the resources and tools most essential to the lesson are collected and structured—data about the learner, possible scripts for the first message, materials for lesson preparation, recommendations on lesson content, the methodological route of the ideal lesson in an action/application format, and questions for self-reflection. The map is one large example—in audio, video, and text—of how the first lesson should be conducted. Inside the map teachers can find short audio fragments of real lesson situations: practical, constantly updated examples drawn from the teaching practice of the Skyeng school’s experts. All the materials are thus gathered in a single document, and working through them takes no more than 30 minutes.
The mechanics of the process:
1. The teacher receives an email 24 hours before a first lesson with
a learner; the email contains a confirmation link, the ideal lesson map, and an example lesson that the content team recommended teaching (all the links carried click counters, so that the number of teachers who followed them could be determined);
2. The system registers whether the teacher followed our
recommendations and chose, or did not choose, the lesson we recommended;
3. The team conducts a manual assessment of the teacher’s lesson, which
is then compared with the machine assessment and used to shape the feedback;
4. Depending on the teacher’s choice (the recommended lesson or
a different one), a chatbot in the corporate Slack messenger sent templated feedback and recommendations for conducting the second lesson in the form of audio examples;
5. On the strength of all these actions the team recorded the data,
which it processed to obtain a detailed result of the experiment.
The tools were delivered to the teachers within the established conditions and time constraints. From the link-click data we learned whether the teacher had chosen the recommended lesson or not—on this depended the feedback the teacher would receive from the Teacher Training Center. The feedback was identical for all teachers, but the tone of its content varied according to the teacher’s results. It contained praise or a reprimand, together with a new set of live audio examples and recommendations on how to keep developing the relationship with the learner and what to adjust in the second lesson. Later this tool grew into a process of automatic notifications, in which the teacher receives methodological support at every stage of interaction with the learner up to and including the 5th lesson—which in the end affected the result of the experiment.
3.2. Results of the Experimental Training
Over the 14 days of the experiment we tested our model for influencing changes in the lesson on 1,000 teachers, and it led to the following results:
1. More than 70% of the teachers followed our recommendations and, on
average, rated the process at 4.6 out of 5;
2. The learners rated the lesson content at 4.7 out of 5.
Two weeks into the test we saw a positive dynamic in retention to the 2nd lesson, and once the results were summed up, retention to the 5th lesson was 6%+ higher than the average value, while first-lesson quality, or GL1, was 10%+ higher; the impact on intensity was assessed as well, and a positive effect was noted. A backtest of the sample’s results was also performed, and a positive difference was likewise observed between the teachers’ current and previous results in the sample. We influenced changes in the teacher’s lesson, and this showed up in the learners’ behavior and in their results up to the 5th lesson. Learners stay with their teacher more often and are satisfied with the learning process. On the strength of the experiment’s results, a decision was made to scale the automated process to 10,000 teachers.
After the experiment ended, 1,612 teachers took part in a regular survey on the process of support and guidance before the first and second lessons.
Recall that 1,000 teachers took part in the test; but because the process was scaled, the survey covered all teachers—both those who had taken part in the experiment and those receiving support after its completion and the scale-up, respectively—which, in our view, makes its results more representative.
Overall snapshot: • 86% of the teachers reported that after receiving our recommendations their lesson had gone at a 5 or a 4; • 87% of the teachers reported that they had used the support tools they received in the corporate Slack messenger (13% said they had received them in their personal account); • 80% of the teachers reported that it is convenient for them to receive notifications in this way (20% said it was inconvenient); • 60% of the teachers reported that they had used the lesson content offered (40% used only the ideal lesson map); • 60% of the teachers reported that they had used the proposed recommendations in their lesson (14% said they had not, and 26% found it hard to answer).
4. CONCLUSIONS
The model for influencing the changes in the teacher’s lesson that are needed for the learner’s successful result—a model under which, when a trigger fires (a lesson in 24 hours), the teacher receives automated recommendations from a chatbot in the corporate Slack messenger before every first and second lesson—proved its effectiveness against the chosen set of experiment metrics, namely: the share of good first lessons (GL1) and retention to the second lesson. The knowledge and tools necessary for conducting a successful first lesson and the lessons after it are sent to the teacher automatically, and at the moment when the practices are to be applied—which has a positive and statistically significant effect on the lesson-quality metrics and on the learner’s results. It also has a positive effect on the teacher’s satisfaction, because the knowledge and tools do not take up the teacher’s time, as additional training would: they are of timely practical value.
Thus, on the basis of an analysis of the empirical material of the study, it can be concluded that the factors affecting the success of this and of alternative models of influence include: • flexibility in the design of the tools themselves (the problems of learners and teachers were both taken into account); • the quality of the preliminary research (the collection of data on the behavior of learners and teachers in various situations); • the depth of the criteria for segmenting the participants in the experiment (teachers and learners); • the precision with which the core and leading metrics of the experiment were chosen; • the quality of the administrative side of the process; • the degree of automation.
REFERENCES (IN RUSSIAN)
1. Леонтьева И.А., Ребрина Ф.Г. Применение дистанционных электронных учебных
курсов в образовательном процессе высшей школы // Вестник Челябинского государственного педагогического университета. 2018. № 3. С. 114-124.
2. Нетология-групп, Исследование российского рынка онлайн -образования и
образовательных технологий. 2020.
3. Кузнецова Л.В. Бизнес-образование как звено непрерывного образования //
Креативная экономика. 2017. C. 13-32.
4. Евенко Л.И., Лагенен Н.Е. Новый шаг в дистанционном обучении бизнесу //
Бизнесобразование. 2016. № 2(38). С. 13-15.
5. Бондаренко Н.В. Непрерывное образование как фактор накопления
человеческого капитала // Научные труды: Институт народнохозяйственного прогнозирования РАН. 2015. № 13. С. 5-10.
6. Медведев Д.А., Мангушева Л.С. Система дистанционного обучения WebTutor //
Анализ и современные информационные технологии в обеспечении экономической безопасности бизнеса и государства. Москва: РЭУ им. Г.В.
Плеханова, 2016. С. 504-506.
7. Рихтер Д., Кляйнкнехт М., Грошнер А. Что мотивирует учителей участвовать в
профессиональном развитии? Эмпирическое исследование мотивационной ориентации и использования формальных возможностей обучения // Педагогика и педагогическое образование. 2019. № 86. С. 3-7.
8. Топкая Э.З., Челикб Х. Восприятие учителями английского языка, не
являющимися носителями языка, о профессиональном развитии: последствия для этапов карьеры // Процедуры — Социальные и поведенческие науки. 2016. № 23.
С. 6-9.
9. Васильев К., Рощин С., Мальцева И. и др. Развитие навыков для инновационного
роста в России. Москва: АлекС. 2015.
10. Юркофский М.М., Блюм-Смит С., Бреннан Р. Расширение результатов: изучение
различных концепций обучения учителей в онлайн -опыте профессионального развития // Педагогика и педагогическое образование. 2019. № 82. С. 2-8.
REFERENCES
1. Bondarenko, N.V. (2015). Nepreryvnoe obrazovanie kak faktor nakoplenija
chelovecheskogo kapitala [Continuing education as a factor in the accumulation of human capital]. Nepreryvnoe obrazovanie kak faktor nakoplenija chelovecheskogo kapitala [Nauchnye trud y: Institut narodnohozjajstvennogo prognozirovanija RAN [Proceedings: institute of economic forecasting RAS], 13, 5-10 (in Russian).
2. Evenko, L.I., and Lagenen, N.E. (2016). Novyj shag v distancionnom obuchenii biznesu
[A new step in distance learning for bus iness]. Biznesobrazovanie [Business education], 2 (38), 13-15 (in Russian).
3. Kuznetsova, L.V. (2017). Biznes -obrazovanie kak zveno nepreryvnogo obrazovanija
[Business education as a link of continuous education]. Kreativnaja jekonomika [Creative economy], 13-32 (in Russian).
4. Leontyeva, I.A., and Rebrina, F.G. (2018). Primenenie distancionnyh jelektronnyh
uchebnyh kursov v obrazovatel’nom processe vysshej shkoly [Application of distance e-learning courses in the educational process of higher education]. Vestnik Cheljabinskogo gosudarstvennogo pedagogicheskogo universiteta [Bulletin of the Chelyabinsk State Pedagogical University], 3, 114-124 (in Russian).
5. Medvedev, D.A., and Mangusheva, L.S. (2016). Sistema distancionnogo obuchenija
WebTutor [Distance learning system WebTutor]. Analiz i sovremennye informacionnye tehnologii v obespechenii jekonomicheskoj bezopasnosti biznesa i gosudarstva [Analysis and modern information technologies in the economy of business and state security], Moscow: PRUE. G.V. Plekhanov, 504-506 (in Russian).
6. Netologija-grupp, Issledovanie rossijskogo rynka onlajn -obrazovanija i
obrazovatel’nyh tehnologij [Netology-groups, Research of the Russian market of online education and educational technologies]. (2020). (in Russian).
7. Richter, D., Klein knecht, M., and Groschner, A. (2019). Chto motiviruet uchitelej
uchastvovat’ v professional’nom razvitii? Jempiricheskoe issledovanie motivacionnoj orientacii i ispol’zovanija formal’nyh vozmozhnostej obuchenija [ What motivates teachers to participate in pro fessional development? An empirical investigation of motivational orientations and the uptake of formal learning opportunities]. Pedagogika i pedagogicheskoe obrazovanie [Teaching and teacher education], 86, 3-7 (in Russian).
8. Topkayaa, E.Z., and Ç elikb, H. (2016). Vosprijatie uchiteljami anglijskogo jazyka, ne
javljajushhimisja nositeljami jazyka, o professional'nom razvitii: posledstvija dlja jetapov kar'ery // Procedury — Social’nye i povedencheskie nauki [Non-native English language teachers’ perceptions of professional development: implications for career stages]. Procedury — social’nye i povedencheskie nauki [Procedia — social and behavioral sciences], 23, 6-9 (in Russian).
9. Vasiliev, K., Roshchin, S., Maltseva, I. et al. (2015). Razvitie navykov dlja
innovacionnogo rosta v Rossii [Skills development for innovative growth in Russia ].
Moscow: AlekS (in Russian).
10. Yurkofsky, M.M., Blum -Smith, S., and Brennan, R. (2019). Rasshirenie rezul’tatov:
izuchenie razlichnyh koncepcij obuchenija uchitelej v onlajn -opyte professional’nogo razvitija [Expanding outcomes: exploring varied conceptions of teacher learning in an online professional development experience ]. Pedagogika i pedagogicheskoe obrazovanie [Teaching and teacher education], 82, 2-8 (in Russian).
APPENDIX 1
Glossary
Continuing professional development involves not only training in new knowledge and practices but also familiarity with professional terminology and an understanding of what it means; the glossary presented below was created for precisely this purpose. Teachers who have only just set out on their professional path will find a great deal of new information here and will be able to immerse themselves in detail in the specifics of online learning. Experienced teachers, experts, and methodologists, meanwhile, will find new opportunities for personal and team development relevant to their professional tasks.
[1] Information and communication approach (method) in foreign language teaching: the use of computer technologies in the process of studying a foreign language. The study of a foreign language with the use of computer technologies. [2] Personalization, to personalize: when the teacher helps the learner connect new words, topics, texts, or grammar with their own life. [3] A new level of competencies (21st century): in the changing circumstances of the 21st century, value attaches to a person’s cognitive, social-emotional, and digital activity, their transformative power with respect to circumstances, their adaptability to change, and their ability to learn and to relearn. The sustainability of society and the formation of a progressive culture depend on the development of new literacies—financial, civic, legal, communicative, medical, and so on.
[4] Communicative approach: an approach to foreign language teaching in which language is, above all, a means of communication rather than an object of study. To this day there are disagreements about the methodology employed. Some experts hold the opinion that the only way to teach a person to communicate is to place the student in a situation that demands communication, while others believe that the study of a language can also have communication as its ultimate goal. Group work and pair work, as well as the encouragement of free use of the foreign language, are distinctive features of the communicative approach. [5] Goal setting: the process of selecting one or more goals and setting deadlines for achieving them. [6] Personalized assignments: an approach within the communicative methodology of foreign language teaching which holds that, for a language to be mastered, it must have real significance for the learner and be used in real-life situations. In this approach the teacher sets a task that reflects real communication situations. The typical sequence of tasks in this lesson model includes three stages: “pre-task”—exercises that prepare the learners to perform the main stage of the communicative task; “task”—the core component of the lesson; and “post-task”—a series of exercises aimed at giving the learners communicative practice in the various types of foreign-language speech activity. [7] Introducing a new technique into the lesson: when, after a training event and at the instruction of the Teacher Training Center, the teacher has introduced a change into their lesson—a specific action that can later be detected and credited by an assessor from the Teacher Training Center. [8] One-to-one online lesson: an organization of the learning process in which one teacher teaches one learner. [9] Regular teacher training: a process within which, as needed or when a trigger fires, the teacher receives training from the Teacher Training Center, whose task is to use various training formats to bring about a positive change in this problem area of the lesson. [10] On-the-job teacher training: a process in which the teacher, in order to upgrade their qualifications, develop a particular skill, or acquire a particular piece of knowledge needed to deliver lessons that are successful within the communicative methodology, does not need to be relieved of work duties—the training arrives on time and targets the specific problem. [11] Experimental training: a process in which teachers are divided into two groups—an experimental group and a test group (a classic A/B test); the first group takes the training and the second does not; their work results are then compared on the metrics chosen for the test, the effect is calculated, a judgment is made about the statistical significance of the result, and, in case of success, a decision is taken to scale. [12] Statistical significance: an assessed measure of confidence that the result obtained is not due to chance. [13] Data analysis: a field of mathematics and computer science concerned with constructing and studying the most general mathematical methods and computational algorithms for extracting knowledge from experimental (in the broad sense) data; the process of inspecting, filtering, transforming, and modeling data with the aim of extracting useful information and making decisions. Data analysis has many aspects and approaches and encompasses different methods across various fields of science and practice.
[14] Teacher profile: a set of characteristics that make it possible to understand the various segments of teachers better, distinguishing them within the general mass of all teachers (1,000+), so that more engaging formats and content of training can be selected for them, and their problems and requests understood better.
[15] Motivation: the thoughts and feelings that make us want to do something and help us keep doing it. Teachers should motivate learners during lessons.
[16] Cohort: a group of people in whom the same particular event occurred during the same period of time.
[17] Application in the lesson: an action aimed at introducing a new action or change into one’s own lesson—one that was in turn presented at a training event (in our case, a workshop: a practice-oriented format aimed at drilling specific actions, for example how to introduce grammar, work with vocabulary, etc.). [18] Feedback: the process of giving and receiving comments on the results of learners’ work, which are used to improve results and lead to the achievement of the learning goal.
[19] The forgetting mechanism (Ebbinghaus’s experiment): the experiments established that after the first error-free repetition of a series of such syllables, forgetting proceeds very quickly at first. Within the first hour alone, up to 60% of all received information is forgotten; 10 hours after memorization, 35% of what was studied remains in memory. Thereafter the forgetting process proceeds slowly, and after 6 days about 20% of the total number of syllables originally learned remains in memory, and the same amount remains after a month. When designing training materials, and when using training to influence teachers effectively, one should keep in mind the peculiarities of the forgetting process.
[20] Silence time in the lesson: calculated automatically for both the teacher and the learner in order to compute an overall value for the lesson; used as a lesson-quality metric; correlates with other metrics; and influences the learner’s decisions at the end of a series of lessons. [21] Lesson structure and components: the main parts of a lesson plan—for example, aims, procedure, warm-up, materials, interaction patterns, anticipated problems, assumptions, timetable fit, personal aims. [22] Instructional design: an academic discipline whose foremost task is the development of distance-teaching methodology through the adoption of modern information technologies. Instructional design assembles a coherent system out of goals, learning material, and the available means of transmitting knowledge. [23] Reflection: when the teacher thinks over the lesson after delivering it, or the learner thinks over the learning process, in order to decide what worked, what did not, and how to improve teaching and learning in the future. [24] Lesson assessment: a structured review of the lesson as a whole and of its individual aspects.
It is used both to analyze a teacher’s lesson—manually or by machine—and as a means of training teachers in pedagogical craftsmanship. The lesson-analysis procedure includes the evaluation of the following parameters: the lesson aim and its achievement, a characterization of the study group, the teacher’s contact with the learner, the speech of the teacher and of the learners during the lesson, the forms of assessment and their effectiveness, and so on.
[25] Chatbot: a conversational program most often used in messengers and on websites. It can perform a whole range of tasks, from entertainment to advising users on important matters. Chatbots that communicate with visitors by means of keywords—a specific command is written for each phrase.
The drawback of such chatbots is that if a question is phrased differently, the chatbot may fail to understand it and fail to answer. For example, if a user asks a question without using a single keyword, the robot may reply with messages such as “sorry, I did not understand.”
[26] Backtesting: the testing and analysis of historical data on teachers, learners, and lessons with the aim of discovering new hypotheses and solutions.
APPENDIX 2
Metrics
Educational, economic, and any other kinds of activity are subject to evaluation, both in process and in outcome. In order to evaluate the effectiveness of a training course, metrics for assessing the school’s instructional and economic activity were formulated, tested and validated in experiments, and taken as the basis.
Retention: shows what proportion of learners moved into a new month, or to the second, third, or any other lesson (this depends on the experiment and on which metric is selected—for example, retention to the second lesson is ret.2). It is the principal indicator of the performance of any company on the market, and it shows very plainly how well—or how poorly—a product’s value and mission have been thought through.
Discipline: shows how many of the lessons planned by a teacher and a learner over a given period of time actually took place.
This metric has a very strong influence on learner retention and on the teacher’s lifetime at the school.
Intensity: shows how often a teacher and a learner have lessons over a given period of time. This metric has a very strong influence on the learner’s result and on the learning program.
Lesson quality: shows how well, and which of, the lesson-quality criteria are met by the teacher—for example: the warm-up, the introduction of new grammar and vocabulary, the delivery of instructions, the balance between Russian and English, the balance of speaking between learner and teacher, etc.
Student assessments: show the learner’s rating for the lesson, for the level and quality of the teacher’s lesson delivery, and for the lesson as a whole. Three separate ratings, and they considerably influence the decisions the learner makes.
TRAINING EFL TEACHERS ONLINE: HOW TO IMPROVE
RESULTS AMONG ADULT LEARNERS (BASED ON THE
EXPERIMENT IN SKYENG SCHOOL)
Nikita S. Berezin
Skyeng School (Tbilisi, Georgia)
The paper considers the organizational aspect of EFL teachers’ professional development with a view to improving the educational results of their students. The author suggests a way of determination of the necessity for professional development, a way to organize the process of professional development and the means of measuring the effectiveness of this process. The fundamental didactic principle of the proposed approach is the timely on-the-job professional teacher’s development. At the beginning of the paper, the author discusses the changes that came along with the development of the EdTech market, online tools in education, opening up both new requirements and new opportunities to solve the problems of all participants of the educational process. Among the burning problems of the participants of the educational process are the following ones: the individual educational request of the student, the requirements of companies for the competencies of employees and candidates in the market, the influence on the student’s result by the teacher’s training, the automation of teaching aids and tools for their analysis. One of the main demands of our time is, surely, the development of separate competencies, such as: social, emotional and digital activities of a person, the ability to be flexible in relation to circumstances, the ability to learn and relearn. The ecosystem solution of the listed problems is considered by the author as the most effective way, which is designed not to separate, but to unite the efforts that coordinate the goals of each of the participants and satisfy their needs. It is proposed to influence the student’s result, formulated by themselves in accordance with the market or individual educational needs, by providing teachers with timely training and support for conducting a good first and subsequent lessons.
As the evidence of the effectiveness of the considered approach, the main part of the article views the experiment of the training center team responsible for the training of 10,000+ English teachers at the Skyeng online school. This part provides an exhaustive description of the project, the proposed tools, training formats and metrics to calculate the effectiveness of the experiment; for the purpose of clarity, charts and other visual decryptions are added. The paper ends with a glossary, which will be useful to all experts experimenting in education, but especially for teachers, methodologists and managers of educational programs.
Key words: distance learning, teaching methodology, online learning, professional competence, professional education, EFL.
How to cite this article: Berezin, N.S. (2020). Training EFL teachers online: how to improve results among adult learners (based on the experiment in Skyeng school). Discourse of Professional Communication, 2(4), 63-80 (in Russian). https://doi.org/10.24833/2687-0126-2020-2-4-63-80. This work is licensed under a Creative Commons Attribution 4.0 International License