📄 Developing the Content of Webtoon Comic Application as Supporting Media in Learning English Grammar at the 2nd Semester English Students of IAIN Palopo

Summary

The authors start from an observation made at IAIN Palopo in June 2019: students spend a lot of time on their smartphones but rarely use them for study. They respond by building an English grammar comic for second-semester English students and publishing it on Webtoon, a popular mobile comic platform that the authors say has no educational language titles. The work is a Research and Development project following the ADDIE model (Analyze, Design, Develop, Implement, Evaluate). A Google Form needs-analysis questionnaire shaped the design. The comic sits in the Slice of Life genre and covers syllabus topics approved by the course lecturers. Five experts rated it on display, content/materials, language and the application itself, and 30 randomly sampled students gave their perceptions in a try-out. Expert feedback led to fixes in writing conventions, consistency of language, and guidance on using the app. The authors conclude that the comic is suitable as a supplementary grammar resource. The project matters mainly as an example of reusing an entertainment platform students already open every day.

Key points

• Design: R&D with the ADDIE model, carried out June 2019 – August 2020 in the English Department of IAIN Palopo. The try-out sample was 30 students chosen at random.
• Needs analysis: the questionnaire had 16 items. Students preferred a learning method built on media and learning facilities (58.1%) and visual or image media (80.6%). The most frequently reported difficulty was low confidence in learning (51.6%).
• Content:
  • The grammar covered was pronouns of the demonstrative and superlative kind, modal verbs, gerunds, and question tags, all presented through a picture-and-picture method.
  • The comic was planned as 6 episodes, counting an introductory one, and can be downloaded for offline reading.
  • The comment section was meant to work as a class discussion space.
• Evaluation: five experts used a 5-point scale. Mean scores were mapped to bands from Very Poor to Very Good, a scheme adapted from Suharto (2006).
• Expert corrections:
  • A self-introduction used "it's" where "I am" was needed to stay consistent with the following lines.
  • A word had a capital letter it should not have had.
  • Several smaller grammar slips were fixed.

Interesting ideas

• I1. The expert revision stage is documented as before/after panel pairs, each tied to one error type (pronoun inconsistency, wrong capitalization), which makes the corrections concrete and countable.
• I2. One correction made a self-introduction line agree with the conversational move it belonged to, replacing "it's" with a first-person "I am" across consecutive turns.
• I3. The paper ran a needs survey before building anything (preferred media, main difficulties) and let those answers decide the format.
• I4. Mean ratings are turned into named quality bands, but the printed bands leave gaps between them (e.g., 1.7 to 1.8) and one table has an impossible interval. This shows why band boundaries should be contiguous and validated.
• I5. Grammar points are placed inside an everyday story with constant characters, so each rule appears in a coherent situation instead of isolated sentences.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

• I1 (learn): Collect such minimal before/after pairs as positive and negative controls when testing self-induced repair rules, so each rule is checked on exactly the error it claims to fix.
• I2 (frames): Treat the speech act (introducing oneself) as a constraint on expected subject person. A clash between the act and the pronoun can then be flagged as a coreference or pragmatics inconsistency with a clear explanation.
• I3 (answer): Before extending the school-course readers, tally which question types and failure categories dominate the test sets. Development effort can then follow measured needs rather than intuition.
• I4 (learn): When evaluation scores are reported as categories, define bands with no gaps and check them automatically, so no result falls outside every label.
• I5 (back): Generate short dialogues from a stable world of the text (fixed entities and events) to illustrate a target construction such as question tags, so the examples stay consistent and can be traced back to the graph.