Design
Learn more about the application's core design and motivation.
Contents
- Target group
- Results from the Market Research
Summary of functional requirements
Problem Description
Target Group
The target group that best fit the goals of this application consists of non-native English learners. In particular, high school students who already have an upper-intermediate level of language proficiency and are about 15 years old. We took the English learning competency level as our measurement of proficiency, as defined by the Common European Framework of Reference for Languages (CEFR). We then designed our application for learners with a proficiency around the B1-B2 level. While B1 refers to an intermediate or ‘threshold’ level of language, B2 refers to an upper-intermediate level (CEFR, 2020).
After reaching the B1 Level, students are able to express und understand English sufficiently for basic interactions in a more informal environment. They are also able to produce simple, short texts about topics in which they feel familiar, and to communicate with native speakers in these topics. Therefore, one can say that students of the level B1 are able to engage and solve problems in everyday life situations (CEFR, 2020). However, B1 is not sufficient to complete, for instance, a course of study or to work in an English speaking environment. English learners at level B2, on the other hand, have more advanced skills in comparison to those on a B1 level. They are able to understand the main topics of complex texts and to interact in a more flexible way, also concerning topics which are not familiar to them (CEFR, 2020).
Therefore, we expect that the target users are able to understand questions and the explanations concerning tasks and to communicate on a basic to advanced level. It is also important to note that our target group focusses on high school students at the age of 15. Thus, we are taking into consideration that they are likely to accustom to the use of smartphones and instant message applications, such as Telegram. As they are also receiving instructions in English at school, this application focusses on the practice and training of already established language skills.
Results from the Market Research
In the mentioned market research we examined pre-existing applications for language learning, specifically for the learning of English. Since the goal of this project is the development of a language learning application supported by methods of Artificial Intelligence (AI), more specifically, Adaptive Language Learning (ALL), we were even more interested in the use of AI and ALL tools in those applications.
According to our market research, the most important gap is to be found in applications that focus on group learning scenarios. Specifically, on learning environments, in which one is fully dependent on the other group members to fulfill a task. This is interesting as learning in groups, also described as cooperative learning, can have positive influences on learning languages (Zhang, 2010). We also found that, so far, there is no application using an environment that relies entirely on conversational learning and not only e.g. a click-a-button method. While there are applications that used methods of machine learning and conversational agents, we did not find any that also relied on group learning. Furthermore, we also did not find group learning environments that support conversations and communication among learners or between learners and a bot.
This concluded that for our application it is important to design tasks that, on the one hand, use the benefits from group learning and, on the other hand, enforce a natural communication to practice those skills for the learners. To foster the group learning scenario and give the learners an opportunity to benefit thereof, it is important that we create tasks where they have to engage with each other and rather than merely working on their own. The possibility for cheating behavior should also be minimized. To create the opportunity for the learners to practice natural communication, the task coordination is performed by the bot through a conversation with the learners. This communication should look as natural as possible to the students. Moreover, It is also important to support constant communication between the bot and the learner and with other learners.
Gamification, as also seen in many other applications, is an important part of successfully learning a second language, outside a ‘classic’ learning environment (Flores, 2015). Gamification describes the inclusion of aspects of games into a non-game environment (Flores, 2015). To enhance the motivation of our users, our tasks should be integrated into a more complex framework and should work in a game-like scenario.
Altogether, our target group corresponds to students who already posses an intermediate proficiency in English. The application should focus on the practicing of language skills, while supporting the students in a natural communication with each other and the bot.
Learning Scenario
To make the users learn English effectively, we designed the tasks using two main scenarios; a group scenario and a narrative scenario. A group scenario provides users the environment of cooperative learning (or collaborative learning). Cooperative learning refers to a particular set of classroom techniques that encourage learner interdependence as a route to cognitive and social development (Johnson 1991). In this environment, since users with a higher level of and lower level randomly mixed, users have opportunities to act as resources for others and suppose a more active role in learning (Mary 1989). Those who have a higher level could learn by teaching others. For example, in task 2 where one person has to explain about a given word to the others in the group, users should try to describe the word even though they already get used to it. In the process, users need to recall the definition, consider a situation where those words would be used, and ensure their understanding of English. On the other hand, those how have a lower level could improve their English by practicing in various random conversational situations.
In this cooperative learning situation, users will be given a common goal and work together to get the goal. In the meantime, users will be able to support each other and get social interaction which is not a common function other language learning applications could provide. According to research at Sakarya University (Bahrani & Sim 2012), social interaction is a crucial source of language learning which can occur outside of the classroom setting. Furthermore, by cooperating together, users could feel to be engaged with each other. As Astin said, the best learning environment for higher education is one in which it is possible to increase engagement (Astin 1984).
The second method we used to make users learn a language effectively is a narrative scenario, especially an escape scenario where users are supposed to be abducted by Alien and escape by solving a few tasks together. Considering the age of our target group who can easily lose their concentration, we expected this gamification of education will help users keep interested and lessen stresses for language learning. Applying gamification in education as learning tools is a promising approach since they strengthen important skills such as problem-solving, collaboration, and communication as well as knowledge (Dicheva, Darina, et al 2015).
Moreover, given a reward, a hint of a key, of completing each task successfully, users continuously being engaged and motivated. It is largely agreed that reward is an important role in terms of creating motivation (Baranek 1996). Each time users obtain a hint of the key, they could picture where they step at among the whole story and be prompted to solve further tasks to collect remaining hints.
Design Process
Escapeling currently offers four language practice tasks. These are:
Click on any of the above tasks to discover the specifics about its design process.
Summary of Functional Requirements
In summary, the chat bot should provide functionality for a group interaction in which 15 to 16 year old students (language level B1) can train their communication skills in the English language outside of a school setting. This requires that users can form and/ or join groups of up to four members when using the application. Generally, the chat bot should in its interaction with the users use language and task descriptions that match the B1 language level of the target group. Additionally, gamification is included in the design of the application. The chat bot needs to track users’ progress in solving the tasks and provide them with clear feedback and rewards when tasks are successfully completed to further motivate and engage them. Once the group has been formed the application needs to allow users to chose one of the two developed tasks.
Detailed functional requirements follow from the in-depth description of the two developed tasks provided above. For both tasks it is necessary that each user is chosen in turn as the main active user, who either attempts to correct the sentence (task one) or provides a description for the current word (task two). This means that the application needs to track which users have already performed the task in any given round and select a new user for each turn. The application should model users’ skill level for relevant language components: vocabulary (trained through task two) and grammar (trained through task one). Each of these broad language components is represented through a set of specific sub-skills. Performing the tasks should train some of these sub-skills and the application needs to monitor individual and group level skills to adequately chose a task difficulty. Additionally, each task has specific functional requirements necessary to provide an adequate user experience.
For task one the chat bot should guide users through the individual parts of the task (identifying error and correcting it) to ensure that users understand what is expected of them. Once a user attempts to solve a part of the task the application should provide immediate feedback and if necessary corrections so that the user can learn the correct solution to the task.
The second task requires that the bot chooses a new word from the list of available words which the active user should explain in the chat. If a user is tasked with explaining a word they do not know they need the option to chose a different word. This means that in providing a word to explain to a user the application additionally should provide the option to reject this word. The application should detect the use of the to-be-explained word in the chat by the user who is supposed to explain it in order to prevent this basic form of cheating in the task. The bot should in such a case mark the task as incorrectly answered. In case one of the other users correctly guesses the word the application needs to detect the solution in the chat and mark the round in the task as solved.
References
Astin, A. W. (1984). Student involvement: A developmental theory for higher education. Journal of college student personnel, 25(4), 297-308.
Bahrani, T., & Sim, T. S. (2012). Informal Language Learning Setting: Technology or Social Interaction?. Turkish Online Journal of Educational Technology-TOJET, 11(2), 142-149.
Baranek, L. K. (1996). The effect of rewards and motivation on student achievement.
CEFR (Common European Framework for Reference of Language) 2020, Global scale - Table 1 (CEFR 3.3): Common Reference levels. https://www.coe.int/en/web/common-european-framework-reference-languages/table-1-cefr-3.3-common-reference-levels-global-scale
Dicheva, D., Dichev, C., Agre, G., & Angelova, G. (2015). Gamification in education: A systematic mapping study. Journal of Educational Technology & Society, 18(3).
Flores, J. F. F. (2015). Using gamification to enhance second language learning. Digital Education Review, (27), 32-54.
Johnson, D. W. (1991). Cooperative Learning: Increasing College Faculty Instructional Productivity. ASHE-ERIC Higher Education Report No. 4, 1991. ASHE-ERIC Higher Education Reports, George Washington University, One Dupont Circle, Suite 630, Washington, DC 20036-1183.
Mary McGroarty (1989) The Benefits of Cooperative Learning Arrangements in Second Language Instruction, NABE Journal, 13:2, 127-143, DOI: 10.1080/08855072.1989.10668555
Zhang, Y. (2010). Cooperative language learning and foreign language learning and teaching. Journal of Language Teaching and Research, 1(1), 81-83.