The Use of Storybird to Improve Students’ Narrative Writing Skills: A Classroom Action Research in Grade VIII at SMP N 3 Sleman, Yogyakarta

Research Design

This study employed a Classroom Action Research (CAR) approach aimed at improving the narrative writing skills of eighth-grade students through the use of the digital learning medium Storybird in English language instruction. Classroom action research was selected because it enables teachers and researchers to implement direct interventions in classroom learning problems while simultaneously reflecting on the outcomes of the actions undertaken. Classroom action research can be understood as a systematic inquiry process intended to solve practical problems in educational settings through cycles of reflective action. This approach allows researchers to develop contextual solutions to learning challenges experienced by students in everyday classroom practice.

Through this mechanism, the learning process becomes not only the object of investigation but also a space for continuous pedagogical innovation. The research design followed the cyclical model proposed by Stephen Kemmis and Robin McTaggart, which consists of four principal stages: planning, action, observation, and reflection. During the planning stage, the researcher identified problems related to students’ writing abilities and developed a learning strategy based on the use of Storybird. The action stage involved implementing the planned instructional activities in narrative writing lessons. The observation stage was conducted to monitor the teaching and learning process, students’ engagement, and changes in writing ability that occurred during the implementation of the action.

Finally, the reflection stage involved analyzing the results of observations and learning evaluations to determine the effectiveness of the intervention and to formulate improvements for the subsequent cycle. This research was conducted in two action cycles, with each cycle consisting of three instructional meetings.

Research Setting and Time

The study was conducted during the first semester of the 2024/2025 academic year at SMP Negeri 3 Sleman, located in the Special Region of Yogyakarta, Indonesia. The research took place from July 24 to October 15, 2024, following the Ainah et al. English class schedule for eighth-grade students, which was held once a week on Tuesdays. Each learning session lasted 45 minutes, in accordance with the standard instructional time allocation applied at the school. The research was carried out through two learning cycles, each consisting of the stages of planning, action, observation, and reflection.

Research Participants

The participants of this study were 32 students of class VIII B at SMP N 3 Sleman. The selection of this class was based not only on preliminary observations indicating that students experienced difficulties in generating ideas, organizing paragraphs, and applying appropriate linguistic features in narrative writing, but also on several pedagogical considerations. First, the English teacher identified class VIII B as having relatively lower baseline writing performance compared to other classes at the same grade level. Second, the class was considered suitable for intervention due to its accessibility and the teacher’s willingness to collaborate in implementing and reflecting on the classroom action research cycles.

In addition to the students, the study involved an English teacher as a collaborator who actively participated in classroom observations, reflective discussions on the implemented actions, and the evaluation of the teaching and learning process. This collaborative involvement was essential to ensure the relevance and practicality of the intervention within the actual classroom context.

Research Procedures

The research procedures were implemented through several stages as follows. 1. Reconnaissance (Problem Identification) At the initial stage, the researcher conducted classroom observations and interviews with the English teacher and several students to identify problems related to the teaching and learning of writing. The findings indicated that students had difficulties generating story ideas, developing narrative structures, and showed limited motivation in writing activities. 2. Planning : Based on the preliminary findings, the researcher designed a learning strategy utilizing Storybird as a digital medium to assist students in generating story ideas through visual illustrations.

At this stage, the researcher also prepared instructional materials and research instruments, including the syllabus, lesson plans, student worksheets, writing assessment rubric, observation guidelines, and interview guidelines. 3. Action and Observation : The action stage involved implementing Storybird-based writing activities during narrative text instruction. During the teaching and learning process, the researcher and collaborator observed students’ activities, their level of engagement in the classroom, and the development of their writing skills. 4. Reflection : At this stage, the researcher and collaborator evaluated the outcomes of the implemented actions based on observation data, interview results, and students’ writing test scores. Reflection was conducted to determine the effectiveness of the instructional strategy and to formulate improvements for the following cycle.

Data Collection Techniques and Instruments

This study was conducted using a Classroom Action Research (CAR) design, in which quantitative and qualitative data were integrated to support the cyclical stages of planning, action, observation, and reflection. Quantitative data were obtained from students’ learning outcomes to measure improvements in achievement across cycles, while qualitative data were collected through classroom observations, field notes, and interviews to document students’ learning behaviors, participation, engagement, and 10.61194/ijss.v7i3.2320 challenges encountered during the implementation process. The integration of these data sources was not intended to constitute a formal mixed-methods research design; rather, it served to strengthen reflective analysis and provide comprehensive evidence for evaluating the effectiveness of instructional actions and informing revisions in subsequent cycles.

This methodological description is applied consistently throughout the abstract, methodology, and conclusion sections. Meanwhile, quantitative data, obtained from students’ writing test scores, were used to measure the progression of students’ writing performance across cycles. The combination of these two data strands allowed for a more comprehensive evaluation of the intervention, where qualitative insights explained the learning process and classroom dynamics, while quantitative results provided empirical evidence of changes in students’ writing outcomes. This integration ensured that each cycle of planning, action, observation, and reflection was systematically informed and continuously improved. 1. Observation : Observation was conducted to examine the teaching and learning process, student interactions, and the implementation of Storybird in classroom activities.

Observation data were recorded using observation checklists and field notes. 2. Interviews : Semi-structured interviews were conducted with the teacher and students to obtain in-depth information regarding their learning experiences, difficulties in writing, and their perceptions of using Storybird in the learning process. 3. Writing Test : Writing tests were administered to measure students’ writing ability before and after the implementation of the instructional intervention. The tests consisted of a pre-test, administered before the action was implemented, and post-tests, administered at the end of each cycle. Students’ written work was assessed using an analytic scoring rubric covering five aspects: content, organization, vocabulary, language use, and mechanics.

Data Analysis

Data analysis was conducted through both qualitative and quantitative approaches. 1. Qualitative Data Analysis : Qualitative data obtained from observations, field notes, and interviews were analyzed through several stages, including data reduction, coding, categorization, interpretation, and conclusion drawing. This analysis aimed to understand changes in students’ learning behavior and the dynamics of classroom interaction during the implementation of the intervention. 2. Quantitative Data Analysis : Quantitative data were derived from students’ writing test scores. The analysis involved calculating the mean scores and the percentage of improvement between the pre-test and post-tests. The data were processed using IBM SPSS Statistics to provide an objective description of the improvement in students’ writing skills.

Validity and Reliability

The validity of the research was ensured through five classroom action research (CAR) validity criteria. Democratic validity was established by actively involving students and the collaborating English teacher in providing feedback throughout the implementation of the Teams Games Tournament (TGT) model. Outcome validity was evaluated by examining improvements in students’ negotiation text writing performance across research cycles and comparing the achievement of predetermined learning objectives. Process validity was maintained through systematic classroom observations, field notes, and reflective discussions conducted after each instructional session to ensure that the intervention was implemented as planned.

Catalytic validity was achieved by encouraging both students and the teacher to reflect on their learning and teaching experiences, leading to greater awareness of effective writing practices and classroom participation. Dialogic validity was strengthened through regular consultations and discussions with the collaborating teacher and academic supervisor to review research procedures, interpretations, and findings. To ensure reliability, students’ writing tests were assessed independently by two raters: the classroom English teacher with more than five years of teaching experience and a university lecturer specializing in English language education and writing instruction.

Prior to scoring, both raters participated in a calibration session to review the scoring rubric, discuss assessment criteria, and score several sample negotiation texts to establish a common understanding of performance standards. Inter-rater reliability was calculated using percentage agreement, resulting in an agreement level of 86%, which exceeded the commonly accepted threshold of 80% and indicated satisfactory scoring consistency. In cases of scoring discrepancies, the final score was determined by averaging the scores assigned by both raters. Furthermore, method triangulation and investigator triangulation were employed by integrating data from observations, interviews, field notes, and writing tests to enhance the credibility and trustworthiness of the research findings.

Research Findings

Reconnaissance Stage

The reconnaissance stage aimed to identify the main problems occurring in the teaching and learning process of writing narrative texts. This stage was conducted through classroom observation, interviews with the English teacher and students, and an initial writing assessment. The research was carried out in class VIII B of SMP Negeri 3 Sleman between July and August 2024. The classroom observation revealed several instructional challenges affecting students’ writing performance. First, many students had difficulty maintaining concentration during the learning process. Some students appeared disengaged, talked with peers, used their mobile phones for unrelated activities, and occasionally did not pay attention to the teacher’s explanations.

Second, the instructional media used in the classroom were relatively limited and repetitive, primarily consisting of presentation slides and online quizzes. This situation reduced students’ Figure 1. Example of Students’ Narrative Writing Before the Intervention

The Action to solve the problems

of the English writing in grade VIII B at SMP N 3 Sleman No Problems The students found it difficult to maintain their focus during the learning process. Actions Implementing a Genre-Based Approach and utilizing Storybird as an instructional medium to attract students’ attention, stimulate their interest, and provide various writing activities. 2 The teacher used monotonous instructional media. Utilizing Storybird as an instructional medium to increase students’ attention and interest in learning writing. 3 The students lacked ideas and experienced difficulties in organizing them into a coherent text. Utilizing Storybird to help students generate ideas and organize them effectively in their writing. 4 The students had limited vocabulary mastery.

Using stories available on Storybird, providing greater exposure to vocabulary, and conducting vocabulary exercises. 5 The students experienced difficulties in using correct grammatical structures in their writing. Providing Storybird stories and picture books to enhance students’ understanding of text organization and reduce grammatical errors. 6 The students’ understanding of writing mechanics was still limited.

Encouraging students to identify and correct their mistakes related to punctuation, capitalization, and the use of commas. 1 Expected Results ✓ Students would become more actively engaged in writing activities. ✓ Students would demonstrate greater focus and interest in writing activities. ✓ Students would experience fewer difficulties in generating and organizing ideas. ✓ Students would improve their vocabulary mastery. ✓ Students would better understand text organization and apply appropriate grammatical structures in their writing. ✓ Students would reduce errors related to writing mechanics. Summary of

Reflection

Results in Cycle 1 of the Implementation of Storybird in Writing Learning Data Source / Evidence Observation and reflection results No Research Aspect 1 Effectiveness of Storybird in stimulating ideas 2 Students’ perception of Storybird learning activities Interview Students 3, 17 September 2024 3 Students’ interest in Storybird features Interview Students 4, 17 September 2024 4 Teacher’s perception of Storybird media Interview Students 5, 17 September 2024 5 Students’ difficulties in vocabulary and language use Interview Students 6, 17 September 2024 6 Students’ vocabulary problems during writing Field Note 13, 17 September 2024 7 Teacher evaluation of students’ writing ability Field Note 14, 17 September 2024 8 Students’ comprehension of narrative writing structure Improvement of writing aspects between Pretest and Cycle 1 Interview and observation results 9 Writing assessment results Research Findings Observation and reflection data indicated that Storybird appeared to support idea generation and text elaboration during Cycle 1, as students were observed developing additional paragraphs in their narrative writing.

Students expressed positive responses to the learning activity. One student stated “Seru banget miss” and explained “yah seru nya, belajarnya pake hape miss storybird gambar itu, terus banyak aktifitasnya tentang nulis, pusing sih tapi seru kok miss.” (Interview Students 3, 17 September 2024) Students felt that Storybird made writing easier and more enjoyable because of the pictures. A student stated “Itu seru banyak gambar nya lucu lucu kita bisa milih apa aja, jadi gampang buat nulis.” (Interview Students 4, 17 September 2024) The English teacher stated that Storybird was a good medium because students looked happy and enthusiastic when using it.

The teacher commented “Storybird is already good… the students also look happy when using the media because the pictures are interesting and there are lots of them.” (Interview Students 5, 17 September 2024) Some students still experienced difficulties in distinguishing verb forms. One student explained “Sudah jelas kok miss, namun saya masih sedikit pahamnya, yang bedain v1 sama verb 2 itu miss masih bingung.” (Interview Students 6, 17 September 2024) Students frequently asked about vocabulary when writing narrative texts. A student asked “Ms, what is the English for hedgehog, Ms?” (Field Note 13, 17 September 2024) The teacher stated that students had progress in developing ideas and class participation, but many errors remained in vocabulary, grammar, and understanding regular and irregular verbs.

The teacher suggested increasing vocabulary, grammar, and punctuation activities in the next cycle. Students began to understand the narrative writing sequence and recognized what should be written first in narrative text, although many students still had difficulty connecting sentences and applying time conjunctions, spelling, and punctuation correctly. The mean score of all writing aspects increased from the pre-test to Cycle 1. Improvements occurred in content, organization, vocabulary, language use, and mechanics, although the improvement in language use and mechanics was lower compared with other aspects. based approach combined with Storybird as instructional media was designed to address several major difficulties experienced by the students in learning writing.

The planned actions focus on increasing students’ engagement, stimulating idea generation, improving vocabulary mastery, strengthening grammatical understanding, and enhancing awareness of writing mechanics. Consequently, these strategies are expected to reduce students’ learning difficulties and support the improvement of their overall writing performance.

Implementation of the Action

The intervention was conducted through two action research cycles following the model proposed by Stephen Kemmis and Robin McTaggart. Each cycle consisted of four stages: planning, action, observation, and reflection.

Report Cycle 1

Planning

During the planning stage, the researcher designed instructional activities integrating Storybird into narrative writing lessons. The instructional design followed the GenreBased Approach, which included four stages: building knowledge of the field, modeling of the text, joint construction of the text, and independent construction of the text. Several instructional materials and research instruments were prepared, including lesson plans, worksheets, observation sheets, interview guidelines, and writing tests.

Action and Observation

The results of Cycle 1 (Actions and Observations) indicate that the implementation of Storybird-based learning activities supported students in understanding and practicing narrative text writing (see Table 3). Through the stages of building context, modelling of text, group construction, and independent construction, students gradually demonstrated improved comprehension of narrative text structure and language features. The observations also revealed that visual media from Storybird increased students’ interest and engagement during the learning process, although some students initially showed shyness and hesitation when responding to questions. However, several students still experienced difficulties, particularly in applying simple past tense and organizing narrative structures accurately.

Overall, the instructional actions in Cycle 1 showed positive progress in students’ participation and writing ability, although further instructional support was still required to strengthen their grammatical accuracy and structural understanding. This Figure 2 illustrates how students collaborated in constructing narrative texts based on the visual prompts provided by Storybird. In the third meeting, students were asked to write narrative texts individually based on the fable theme. The individual writing task served as the post-test for Cycle 1. Reflection The results presented in Table 4 show that the use of Storybird as a learning tool has a positive impact on improving students’ writing skills.

The increase in the mean scores from the pre-test to Cycle 1 across all writing aspects demonstrates that students showed measurable progress in content development, organization, vocabulary, language use, and mechanics. In addition, interview and observation data reveal that students felt more interested and motivated to write because the visual features in Storybird helped them generate ideas for their narratives. However, several students still experienced difficulties in vocabulary mastery and grammatical accuracy, particularly in distinguishing verb forms and applying simple past tense. Therefore, although the first cycle showed meaningful improvement in students’ writing performance, further instructional support focusing on vocabulary, grammar, and punctuation was still required in the next cycle.

Based on Table 5, the mean scores across all five aspects of writing content, organization, vocabulary, language use, and mechanics demonstrate a consistent improvement from the pre-test to Cycle 1. This indicates that students showed measurable progress in their writing performance following the initial implementation of Storybird. Among the five aspects, organization and vocabulary exhibited the most notable increases, suggesting that the visual scaffolding provided by Storybird effectively supported students in structuring their ideas and selecting appropriate lexical items. Improvements in the content aspect also indicate that students became more capable of developing relevant and coherent ideas in their narrative writing. Meanwhile, gains in language use and mechanics, although present, were comparatively more moderate.

This suggests that while the intervention facilitated idea generation and text organization, students still required further support and practice in mastering grammatical accuracy and writing conventions. Overall, the findings from Cycle 1 reflect a positive trend in students’ writing development; however, the variation in improvement across different aspects highlights the need for continued refinement of instructional strategies in subsequent cycles, particularly to strengthen students’ proficiency in language use and mechanics.

Report of Cycle 2

Planning The planning stage of Cycle 2 was developed through discussions between the researcher and the collaborator based on reflections from Cycle 1, with the aim of improving students’ writing skills through the use of Storybird. The researcher designed three meetings and prepared instructional components aligned with the Merdeka Curriculum, including lesson plans on narrative texts using the Genre-Based Approach stages: Building Knowledge of the Text, Modeling of the Text, Joint Construction of the Text, and Independent Construction of the Text. In addition, supporting instruments such as student worksheets, observation sheets, interview guidelines, and post-tests were prepared to facilitate learning activities and evaluate the improvement of students’ writing skills after the implementation of Storybird in Cycle 2.

Action and Observation The findings presented in Table 6 show that the use of Storybird as a learning tool has a positive impact on improving students’ narrative writing skills. Through the stages of building knowledge, modelling, joint construction, and independent construction, students gradually demonstrated better understanding of narrative text structure and language features. The learning activities also encouraged active participation, collaborative discussion, and increased enthusiasm among students during the writing process.

Furthermore, students showed noticeable Summary of Research Results in Cycle 2 (Action and Observation Using Storybird) No 1 Meeting / Stage General Implementation of Cycle 1 Date 24 September, 1 October, and 8 October 2024 2 Meeting 1 – Building Knowledge of Field and Modelling of Text 24 September 2024 3 Meeting 1 – Analyzing Narrative Text Structure and Language Features 24 September 2024 4 Meeting 1 – Grammar Practice (Verb 1 to Verb 2) 24 September 2024 5 Meeting 1 – Arranging Narrative Paragraph 24 September 2024 Students arranged eight random sentences into a coherent narrative paragraph with their desk mates. 6 Meeting 2 – Reviewing Narrative Text and Introducing Storybird Writing Meeting 2 – Group Narrative Writing Using Storybird 1 October 2024 The researcher reviewed narrative text concepts, structure, and purpose.

Students then worked in groups using Storybird to create narrative texts based on pictures. Students worked in groups of 4–5 to create narrative stories using Storybird with a minimum of five slides. Meeting 2 – Presentation and

Discussion

of Students’ Narrative Text Meeting 3 – Review and Preparation for Individual Writing 1 October 2024 7 8 9 1 October 2024 15 October 2024 Learning Activities The teaching and learning process in cycle 1 was conducted in three meetings which purposed to increase the students writing skills using storybird. The research implemented the actions according the lesson plans created with the topic Narrative Text. The collaborator led notes and captures on the teaching and learning process. The researcher displayed pictures from Storybird and asked questions such as “Do you know this story”, “what the title of story” and “Describe the story with one words”. Students also completed vocabulary exercises related to the story.

Students analyzed narrative text titled “the shepherd Boy and the wolf” in LKPD activities 4 and 5 by identifying structure and language features such as past tense, adjectives, and conjunctions of time. Students individually changed verb 1 to verb 2 using exercises provided in the LKPD along with a list of regular and irregular verbs. Each group published their Storybird narrative and presented their story to the class while the researcher discussed and corrected their writing. The researcher reviewed narrative text and prepared students for individual writing activities. Research Findings / Observations The actions focused on improving students’ writing skills in narrative text through Storybird-based activities and structured writing exercises. Students demonstrated improvement in identifying narrative text structure through discussion.

A student stated “wow that's what Upin Ipin told” and another student explained the conflict and ending of the story (Field Note 15, 24 September 2024). Students actively participated and showed better understanding than in the previous cycle. A student asked “Miss, the conjunction of time is like one day and once upon a time, right Miss?” and the researcher responded positively (Field Note 16, 24 September 2024). Students became more confident and answered quickly. When asked about verb forms, students responded correctly such as “talked Miss” and “heard Miss” (Field Note 17, 24 September 2024). Students actively discussed the structure and identified orientation, complication, and resolution. A student explained “there is once upon a time, then it starts introducing the characters so it's orientation right Miss” (Field Note 18, 24 September 2024).

Students were able to recall narrative concepts, answering that narrative text is “text about imaginary story Miss” and its purpose is “to amuse the reader Miss” (Field Note 20, 2 October 2024). Students were active, enthusiastic, and exchanged ideas with peers. One student asked “Miss, I want to ask, if for example we want to make a story about animals in the forest and then their conflict is fighting over food in the forest and then it is resolved by the king of the forest, the lion, like being reconciled, can we do it, Miss?” (Field Note 22, 2 October 2024). Students’ writing results improved in language use, structure, and writing mechanics, and many students actively participated in discussions. Students showed enthusiasm and active participation in answering questions related to narrative text structure (Field Note 24, 15 October 2024). Ainah et al.

No 10 10.61194/ijss.v7i3.2320 Meeting / Stage Meeting 3 – Individual Narrative Writing Using Storybird Date 15 October 2024 Learning Activities Students wrote narrative texts individually with a fable theme while the researcher monitored and corrected their writing. Research Findings / Observations Students were able to complete their narrative writing tasks independently, although many still asked about verb 2 forms and narrative structure during the writing pr Mean Score of Student’s Achievement in Cycle 1 and Cycle 2 Test Cycle 1 Cycle 2 Content R C 3,04 3,04 3,75 3,72 Organization R C 3,1 3,09 3,85 3,88 Vocabulary Language Use R C R C 2,92 3 3 3 3,83 3,86 3,65 3,68 Mechanics R C 2,78 2,84 3 3 Table 9.

Comparison of Writing Scores Test Pretest Cycle 1 Cycle 2 Content R C 2,61 2,70 3,04 3,04 3,75 3,72 Organization R C 2,58 2,65 3,1 3,09 3,85 3,88 The findings presented in Table 7 indicate that the use of Storybird is associated with an improvement in students’ narrative writing performance during the second cycle. Qualitative evidence from classroom observations and interviews indicated that students demonstrated greater enthusiasm, collaboration, and confidence when engaging in narrative writing activities. In addition, improvements were observed in several writing components, including vocabulary, language use, organization, and writing mechanics.

The quantitative data also showed an increase in mean scores across all assessed writing aspects from Cycle 1 to Cycle 2 Vocabulary R C 2,44 2.47 2,92 3 3,83 3,86 Language Use R C 2,38 2.38 3 3 3,65 3,68 Mechanics R C 2,25 2.26 2,78 2,84 3 3 (see Table 8). However, given the Classroom Action Research design without a control group, these findings should be interpreted as classroom-based improvements associated with the implementation of Storybird rather than as evidence of statistically significant effects or definitive causal impact.

Improvement of Students’ Writing Scores

The improvement of students’ writing performance can be observed through the comparison of the mean scores across the pre-test, Cycle 1, and Cycle 2. As shown in the Table 9, the mean scores for all five writing aspects improved from the pre-test to the post-test of Cycle 1 and further to the post-test of Cycle 2. The most substantial gains were observed in the organization aspect (O), which demonstrated a significant increase across the three assessment points. This improvement was substantiated by the comparison of mean scores between the pre-test and the post-test of Cycle 2. A comparable pattern of progress was evident in the content aspect (C), where the mean score rose from 2.61 in the pre-test to 3.04 in the post-test of Cycle 1 and subsequently increased to 3.70 in the post-test of Cycle 2.

In the vocabulary aspect (V), the mean scores advanced from 2.47 in the pre-test to 3.00 in the post-test of Cycle 1 and reached 3.83 in the post-test of Cycle 2 following the implementation of Storybird. Similar improvements were recorded in the language use aspect (L), with mean scores increasing from 2.38 in the pre-test to 3.00 in the post-test of Cycle 1 and to 3.65 in the post-test of Cycle 2, as well as in the mechanics aspect (M), which improved from 2.26 in the pre-test to 2.84 in the post-test of Cycle 1 and to 3.00 in the post-test of Cycle 2. To support of comparison of mean scores on each aspect, the researcher aspects, the researcher also analyzed the data by looking at the Paired Sample T-test. The detail quantitative data analysis of the pre-test and post-test can be seen on the table below.

Table 10 presents that paired-samples t-test was conducted to evaluate whether the observed improvement in students’ overall narrative writing scores (composite score) from pre-test to post-test at the end of Cycle 2 was statistically significant. This test was selected as the appropriate parametric procedure because the design involved repeated measures on the same 32 students, allowing the analysis to focus on within-subject change scores while controlling for individual differences and thereby increasing statistical power compared with an independent-samples test. Prior to interpretation, the assumption of normality of the difference scores was examined and considered tenable given the moderate sample size and the known robustness of the t-test to mild departures from normality.

The analysis revealed a statistically significant mean increase of 27.50 points in overall writing scores from pre-test (M = 57.61) to post-test (M = 80.03), t (31) = 23.066, p < .001, 95% CI [25.07, 29.93]. The effect size was very large (Cohen’s d_z = 4.08), indicating substantial practical improvement in addition to statistical significance. These results support the conclusion that implementation of Storybird across the two classroom action research cycles was associated with meaningful gains in students’ narrative writing performance. However, because the study employed a pre-experimental classroom action research design without a control or comparison group, alternative explanations such as practice effects, task familiarity, or teacher scaffolding cannot be ruled out; therefore, causal claims regarding the intervention’s effectiveness should be advanced with appropriate caution.

The scores were obtained from the accumulation of the students’ scores on the five aspects of writing at four assessment points: the pre-test, at the end of Cycle 1, at the end of Cycle 2, and in the post-test (see Figure 3). In the pretest, the mean score was 57. This rose to 71 at the end of Cycle 1. Following the implementation of Storybird in Cycle 1, the mean score increased further to 80 at the end of Cycle 2. These results demonstrated a clear improvement in the students’ writing skills. This version is now grammatically accurate, terminologically precise, and stylistically appropriate for a research article while directly resolving every point raised in the reviewer’s comment.

The findings of this study indicate that the integration of the Storybird platform contributed to a progressive improvement in students’ narrative writing skills across two classroom action research (CAR) cycles. This improvement is reflected not only in the increase in overall mean scores but also in the differentiated development across specific writing components. Notably, organization, vocabulary, and content exhibited more substantial gains, whereas language use and mechanics improved at a comparatively slower rate. This uneven pattern suggests that the impact of the intervention was selective rather than uniformly distributed across all dimensions of writing competence. From a theoretical standpoint, this pattern can be explained through the framework of cognitive scaffolding in multimodal Table 10.

Paired Samples T- Test of The Pre-Test and Post Test Paired Differences Mean Pair 1 PRE-TEST POST-TEST 27.50000 Std. Deviation Std. Error Mean 6.74417 1.19221 95% Confidence Interval of the Difference Lower -29.93153 Upper 25.06847 t df Sig. (2-tailed) -23.066 31 .000 Figure 3. Improvement of Students’ Writing Scores Across Research Cycles Ainah et al. digital environments. The visual features embedded in Storybird function as external cognitive supports that facilitate higher-order writing processes, particularly idea generation and narrative structuring. Students demonstrated enhanced ability to organize narrative elements such as orientation, complication, and resolution and to elaborate ideas into coherent texts. However, lower-order linguistic features, including grammatical accuracy and writing mechanics, did not develop at the same pace.

This indicates that while visual scaffolding effectively enhances conceptual and organizational aspects of writing, it does not automatically lead to mastery of linguistic forms, which require explicit instruction and sustained practice. Consequently, the observed improvement should be interpreted as domainspecific rather than comprehensive across all writing skills. This interpretation is consistent with (Lou et al., 2026), who found that multimodal scaffolding in digital storytelling improves narrative coherence but requires additional support for linguistic precision, as well as (Alemi & al.), 2022), who reported stronger gains in idea development than in grammatical accuracy. In relation to existing literature, this study supports the broader claim that digital storytelling tools enhance writing performance (Kazazoğlu, 2021; Shamsulbahri & Aziz, 2020; Zhang & Zhang, 2021).

However, its primary contribution lies in offering more nuanced empirical evidence regarding the specific dimensions of improvement. While previous studies often emphasize overall writing gains, the present findings clarify that Storybird’s principal strength resides in its ability to support narrative coherence and idea development through visual prompts. Conversely, its effect on grammatical accuracy remains limited without targeted pedagogical intervention. Similar patterns have been reported by (Ambarwati et al., 2026), (Köroğlu, 2023), (Ajabshir et al., 2024), and (Mahulette et al., 2020), all of whom highlight stronger improvements in content, organization, and creativity compared to linguistic accuracy. From a pedagogical perspective, the effectiveness of Storybird is not inherent in the tool itself but is contingent upon instructional design.

The increased student engagement observed in Cycle 2 suggests that iterative refinements such as clearer instructions, structured tasks, and guided feedback played a crucial role in enhancing learning outcomes. This finding aligns with Tetik and (Koder et al., 2025; Murad et al., 2023; Winnaandhini, 2025) and (Munajah, 2023), who emphasize that digital storytelling yields optimal results when integrated with scaffolded pedagogy and explicit teacher guidance. (Hastomo, 2015) similarly underscores that instructional media must be supported by appropriate pedagogical strategies to achieve effectiveness. Therefore, digital tools should be embedded within a structured instructional framework rather than treated as standalone solutions.

In terms of language development, the observed improvements in vocabulary and organization support the notion that visual media function as external cognitive aids (Abdullah & A., 2020). The images provided by Storybird enable learners to associate meaning with linguistic forms, thereby facilitating vocabulary acquisition and contextual usage. However, the relatively limited progress in language use and mechanics indicates that visual input alone is insufficient for developing grammatical competence. Empirical evidence from Ajabshir et al. (2024), further confirms that while digital storytelling enhances complexity and lexical diversity, grammatical accuracy improves more significantly when combined with explicit instruction, targeted feedback, and peer review.

This underscores the necessity of integrating digital storytelling with focused language exercises to achieve balanced writing development. Methodologically, the CAR design characterized by 10.61194/ijss.v7i3.2320 iterative cycles of planning, action, observation, and reflection (Stringer, 2007) enabled continuous refinement of instructional practices. The transition from Cycle 1 to Cycle 2 demonstrates that pedagogical adjustments, including increased teacher guidance, incorporation of individual writing tasks, and more systematic feedback, were instrumental in improving both student engagement and performance. Thus, the study contributes not only by demonstrating the application of Storybird but also by illustrating how iterative pedagogical refinement within a CAR framework can optimize its implementation in authentic classroom contexts.

This finding is consistent with recent CAR-based studies (Ambarwati et al., 2026; Köroğlu, 2023), which attribute progressive improvements to reflective and adaptive teaching strategies. Nevertheless, these findings should be interpreted with caution. Although there is a clear association between the use of Storybird and the improvement in students’ writing skills, the research design does not allow for strong causal inference. Other factors, such as repeated exposure to writing tasks, increased familiarity with narrative structures, and the role of teacher facilitation, may also have contributed to the observed outcomes. Therefore, the results are better understood as indicative of context-specific pedagogical enhancement rather than definitive evidence of causal effectiveness.

This limitation aligns with the broader conclusions of (Winnaandhini, 2025) who notes that while digital storytelling consistently supports narrative development, causal claims remain constrained in the absence of controlled experimental designs. Comparable cautions appear across recent studies of digital storytelling and technology-enhanced writing interventions in language education, which similarly report positive associations in classroom-based or quasi-experimental settings yet emphasize that outcomes are frequently moderated by confounding variables including instructional support, technological access, learner prior knowledge, and contextual conditions, thereby underscoring the necessity of randomized controlled trials or extended experimental designs to substantiate causal effects (Lashari et al., 2022).

Implications The findings of this study offer several pedagogical implications for English language teaching, particularly in the context of narrative writing instruction. First, the use of Storybird can support students in generating ideas and organizing narrative texts more effectively, especially at the junior high school level where learners often rely on visual stimuli. However, its implementation should be accompanied by structured instructional guidance to maximize its benefits. Second, collaborative learning activities, such as pair or group work, were found to enhance students’ engagement and facilitate idea exchange. This supports previous findings (Anggraini, 2020) that peer interaction contributes to deeper understanding and more active participation. Nevertheless, collaboration should be carefully managed to ensure equal participation among students.

Third, the provision of feedback plays a crucial role in improving students’ writing skills across multiple aspects, including content, organization, vocabulary, language use, and mechanics. Consistent with (Miranty & al, 2025), timely and constructive feedback enables students to identify and correct their errors, thereby supporting continuous improvement. Fourth, the inclusion of guided practice, such as grammar and vocabulary exercises, remains essential. While Storybird supports higher-level writing processes, lower-level linguistic accuracy still requires explicit instruction (Erden, 2022). Therefore, a balanced approach combining digital media and traditional instruction is recommended.

From a broader perspective, although the findings suggest positive outcomes, the scalability of this approach may be influenced by contextual factors such as technological access, teacher readiness, and Ainah et al. classroom conditions. Consequently, further studies are needed to examine its applicability in different educational settings.

Limitation of the Study

This study has several limitations that warrant acknowledgment. First, empirical research on the implementation of Storybird within a classroom action research framework remains relatively scarce, which restricts the extent to which the present findings can be compared with or generalized from existing literature. Second, external classroom factors, including scheduling disruptions, school activities, and time constraints, affected the consistency of the intervention and may have influenced students’ learning focus and overall outcomes. Third, the study was conducted in a single Grade VIII class comprising 32 students at one junior high school. Consequently, the findings are specific to this particular classroom context and cannot establish causal relationships, as the classroom action research design did not include a control group.

All conclusions and recommendations must therefore be interpreted strictly within these contextual boundaries. Future research is recommended to involve larger and more diverse samples across multiple classes or schools and to employ experimental or quasi-experimental designs to strengthen causal inference while building upon the contextual insights provided by the present study.