Enhancing EFL Students’ Speaking Skills through Digital Storytelling: A Quasi-Experimental Study

Introduction

English is a compulsory subject across all levels of formal education in Indonesia, where it is taught as a foreign language (EFL). Among the four language skills, speaking is widely regarded as the most challenging because it functions as the main medium for communication and self-expression (1). Despite years of instruction, many students still show limited ability to communicate orally in meaningful ways. Previous studies found that Indonesian senior high school students often perform at moderate to low levels in speaking, particularly in terms of hesitation, inaccurate pronunciation, and limited vocabulary use (2, 3). These findings indicate a gap between classroom exposure to English and students’ actual communicative competence, especially at the secondary school level.

The urgency of improving speaking instruction is also reflected in classroom practice. Conventional teacher-centered approaches often provide limited opportunities for students to practice speaking actively, which leads to passive participation and low confidence (4, 5). Preliminary observations at SMAN 4 Palu revealed similar conditions, where students were able to produce basic spoken English but still struggled to express ideas fluently and accurately. This situation suggests that speaking instruction requires more interactive strategies that encourage meaningful communication and continuous oral practice.

Various approaches, such as task-based learning, role play, interviews, and storytelling, have been used to address speaking difficulties. However, earlier studies reported that these methods are sometimes constrained by low student engagement, limited use of technology, or insufficient multimodal support for language learning (6). In response to these limitations, Digital Storytelling (DST) has emerged as an alternative approach that combines narrative activities with digital media such as images, audio, and video (7, 8). Studies have shown that DST can increase learners’ motivation, creativity, and participation because students actively construct and present their own stories in meaningful contexts (9–11). However, most previous studies focused on general speaking development or learner engagement, while limited evidence specifically examines how DST affects pronunciation and fluency in Indonesian EFL classrooms (12, 13).

To address this gap, the present study investigates the use of Digital Storytelling to enhance students’ speaking skills, with particular focus on pronunciation and fluency. Using a quasi-experimental design with pre-test and post-test measures, this study seeks to provide empirical evidence of whether DST produces significantly better speaking outcomes than conventional instruction (14, 15). The research question guiding this study is: Does the use of Digital Storytelling significantly improve students’ speaking performance? Based on this question, the null hypothesis (H₀) states that there is no significant difference between students taught using DST and those receiving conventional instruction, while the alternative hypothesis (H₁) states that there is a significant improvement in the DST group.

Methodology

Study Design and Rationale

This study employed a quantitative quasi-experimental design using a non-equivalent control group pre-test–post-test model, which is widely recognized as appropriate for educational settings where random assignment is not feasible (7). The design enabled a controlled comparison between an experimental group receiving Digital Storytelling (DST) instruction and a control group receiving conventional speaking instruction. The structure of the design followed the standard notation O₁ X O₂ / O₃ – O₄, where O₁ and O₃ represent pre-test observations, X denotes the DST intervention, and O₂ and O₄ represent post-test observations. This approach allowed for the examination of causal relationships between the instructional intervention and students’ speaking performance while maintaining ecological validity in a real classroom context. Such a design is frequently recommended in classroom-based research because it allows treatment evaluation without disrupting existing class organization. Furthermore, the quasi-experimental design enabled the researcher to compare students’ speaking achievement before and after the implementation of the instructional strategy, thereby providing empirical evidence regarding its effectiveness. By utilizing naturally existing classroom groups, the study also reflected authentic teaching and learning conditions, increasing the practical relevance and applicability of the findings to similar educational settings. In addition, this design minimized administrative and ethical challenges that often arise when random assignment is not feasible in formal school environments.

Population, Sample Size, and Sampling Criteria

The population comprised all eleventh-grade students at SMAN 4 Palu during the 2023/2024 academic year, totaling 405 students across 13 classes, as summarized in Table 1 (Population Distribution). A purposive sampling technique was applied to select participants based on instructional equivalence, teacher recommendation, and comparable baseline speaking proficiency. Two intact classes were selected: XI D as the experimental group (n = 31) and XI C as the control group (n = 24), yielding a total sample of 55 students. Purposive sampling was deemed appropriate to ensure contextual alignment with the research objectives while minimizing instructional disruption in intact classroom settings (16). Random assignment was not feasible due to institutional policies requiring the preservation of existing classroom groupings.

Variables and Operational Definitions

The study involved one independent variable and one dependent variable. The independent variable was the implementation of Digital Storytelling (DST) as an instructional medium in speaking instruction. The dependent variable was students’ speaking skill, operationalized through two measurable components: pronunciation accuracy and fluency. Operational definitions were established to ensure measurement precision and replicability. Speaking effectiveness was defined as statistically significant improvement in post-test scores relative to pre-test scores, consistent with established educational effectiveness frameworks.

Instructional Materials and Instruments

Data were collected using speaking performance tests administered as pre-tests and post-tests. The assessment instruments consisted of structured speaking tasks requiring students to orally produce short narratives and responses. Students’ speaking performances were evaluated using analytic scoring rubrics, which are presented in Table 2 (Pronunciation Rubric) and Table 3 (Fluency Rubric). Each rubric employed a five-point scale (1–5), with clearly defined performance descriptors to ensure scoring consistency. Overall speaking achievement was further categorized using score ranges and qualitative descriptors presented in Table 4 (Score Range, Category, and Qualification). The use of analytic rubrics was justified to enhance reliability and diagnostic sensitivity in assessing speaking performance, as supported by recent studies on speaking assessment in EFL contexts (16). To strengthen scoring reliability, two English teachers independently rated students’ performances, and discrepancies were discussed until agreement was reached. This inter-rater procedure was applied consistently in both pre-test and post-test scoring.

Research Procedures

The study was conducted over three main phases: pre-testing, treatment, and post-testing. During the pre-test phase, both groups completed identical speaking tasks to establish baseline equivalence. The treatment phase lasted for six instructional meetings over six weeks, with each meeting conducted for approximately 90 minutes. During this phase, the experimental group received DST-based instruction. Students were guided through systematic DST procedures, including topic selection, script drafting, audio narration recording, integration of visual media, and oral story presentation. These procedures followed structured stages of Digital Storytelling implementation, including planning, implementing, and reporting (8). During these stages, students engaged in script writing, audio narration recording, and integration of multimedia elements such as images and background music (17). Each DST session emphasized repeated oral rehearsal, pronunciation modeling, and fluency development through narrative delivery. In contrast, the control group received conventional speaking instruction emphasizing textbook-based exercises and teacher-led oral practice without digital media integration. Following the intervention, both groups completed a post-test using parallel speaking tasks to measure learning gains attributable to the instructional treatment.

The integration of multimedia elements reflects multimodal learning principles, which emphasize the use of visual, auditory, and textual modes to enhance student engagement and language development (18). Previous studies have also shown that Digital Storytelling can improve speaking fluency, reduce anxiety, and promote learner autonomy in EFL learning environments (16, 19).

Data Analysis Techniques

Quantitative data analysis was conducted using SPSS version 26 and Microsoft Excel. Individual speaking scores were first converted into standardized percentages using Equation 1, where A represents the score obtained and N denotes the maximum possible score. Descriptive statistics were computed to summarize central tendencies and dispersion. Prior to inferential analysis, normality of data distribution was tested using the Shapiro–Wilk test, while homogeneity of variance was examined using Levene’s test, with a significance threshold of p > 0.05. To test the research hypothesis, a paired-samples t-test was employed to compare pre-test and post-test scores within groups. In addition, an independent-samples t-test was conducted to compare the post-test mean scores of the experimental and control groups in order to determine whether the DST treatment produced a significantly greater effect than conventional instruction. Statistical significance was set at α = 0.05.

Ethical Considerations

Ethical approval for the study was obtained through institutional authorization from SMAN 4 Palu and the Faculty of Tarbiyah and Teacher Training, State Islamic University Datokarama Palu. Participation was conducted with the consent of school authorities, teachers, and students. All data were anonymized, and participation posed no academic or psychological risk to students. The study adhered to standard ethical principles for educational research, including confidentiality, voluntary participation, and responsible data handling (9).

Results

Descriptive Statistics of Control and Experimental Groups

The descriptive statistics of students’ speaking scores in both groups are presented in Table 5. The table summarizes pre-test means, post-test means, mean gains, and standard deviations for pronunciation and fluency. Overall, both groups showed improvement after the instructional treatment.

Although the control group obtained higher post-test means, this group also started with higher pre-test means. Therefore, the comparison should consider the initial differences between groups. The experimental group showed consistent improvement after receiving Digital Storytelling (DST) instruction, indicating positive progress in students’ speaking development.

Assumption Testing: Normality and Homogeneity

Prior to hypothesis testing, normality and homogeneity tests were conducted to determine whether the data met the assumptions for parametric analysis. The results of the Shapiro-Wilk normality test are presented in Table 6, while the homogeneity test results are shown in Table 7.

As shown in Table 6, most significance values were above 0.05, indicating that the data were normally distributed. Although one gained-score variable showed p < 0.05, the overall distribution was considered acceptable for further parametric analysis. This result suggests that the score distributions in both groups were sufficiently normal to proceed with t-test analysis. Since only one variable slightly deviated from normality, the violation was not considered substantial enough to invalidate the use of parametric procedures.

Table 7 shows that all significance values exceeded 0.05, indicating homogeneous variances between groups. Therefore, the assumption of homogeneity was fulfilled. In other words, the spread of scores in the control and experimental groups was statistically comparable. This strengthens the validity of subsequent group comparisons because differences in results are less likely to be caused by unequal score variability.

Hypothesis Testing Results

Prior to hypothesis testing, assumption checks were conducted to ensure the suitability of parametric analysis. The core inferential findings of the study are presented in Table 8, which reports the results of the paired-samples t-test comparing the combined pre-test and post-test scores of all measured speaking components. The analysis revealed a statistically significant difference between pre-test and post-test scores, with a t-value of 23.242 and a significance level of p < 0.001 (df = 54). This paired-samples t-test was conducted on the combined dataset of students’ speaking scores (pronunciation and fluency aggregated across both groups), comparing overall performance before and after the instructional treatment.

Table 8 indicates a statistically significant difference between pre-test and post-test scores (t = 23.242, df = 54, p < 0.001). This confirms that the analysis consistently refers to the same paired dataset, and that students’ speaking performance improved significantly after the treatment. Therefore, the null hypothesis (H₀) was rejected and the alternative hypothesis (H₁) was accepted. The positive mean difference (14.982) shows that post-test scores were substantially higher than pre-test scores. In practical terms, this finding indicates that the instructional treatment was associated with improvement in students’ pronunciation and fluency.

Discussion

The present study examined the effectiveness of Digital Storytelling (DST) in improving the speaking skills of eleventh-grade EFL students at SMAN 4 Palu, with particular focus on pronunciation and fluency. The findings showed that students in both groups improved, while the experimental group demonstrated significant progress after receiving DST-based instruction. This suggests that DST can function as an effective complementary strategy for speaking instruction in EFL classrooms.

Effect of Digital Storytelling on Speaking Skill

One important finding was the significant improvement in the experimental group after treatment. This result suggests that DST created more opportunities for students to practice speaking through meaningful and contextualized tasks. Students were required to prepare stories, rehearse oral delivery, and present ideas clearly. These repeated speaking opportunities may explain why students showed better speaking performance after the intervention.

This finding is consistent with previous studies stating that speaking develops more effectively when learners use language for authentic communication rather than isolated drills (20). DST supports this principle because students use English to communicate messages, not only to complete exercises.

Pronunciation Development

Students in the experimental group also improved in pronunciation. A possible explanation is that DST activities involved recording and replaying narration, allowing students to notice pronunciation errors and revise their performance. This self-monitoring process may increase awareness of stress, intonation, and sound accuracy. Repeated rehearsal before final presentation may also strengthen pronunciation control (21).

Fluency Development

Fluency gains were also evident in the experimental group. Students became more able to speak continuously with fewer pauses and less hesitation. DST tasks encouraged learners to produce longer stretches of speech because they had to narrate complete stories. Extended speaking practice is important for fluency because students learn to organize ideas while speaking in real time (22).

Comparison with Conventional Instruction

The control group also showed improvement, indicating that conventional instruction can still support speaking development. However, teacher-centered practice may provide fewer opportunities for extended oral production and creative expression. DST appears to add value by combining speaking practice with multimedia support, student autonomy, and meaningful communication tasks.

Pedagogical Implications

The findings suggest that English teachers may consider integrating DST into speaking classes as an alternative or supplementary strategy. By combining images, audio, and storytelling, DST can increase participation and motivation while supporting pronunciation and fluency practice. For schools with limited resources, simple mobile-phone recording applications may already be sufficient to implement DST activities effectively.

Conclusion

This study found a statistically significant difference between pre-test and post-test speaking scores, indicating that the instructional treatment contributed to students’ speaking improvement. Digital Storytelling (DST) may support the development of EFL students’ speaking skills, particularly in pronunciation and fluency. Although both the experimental and control groups showed progress, students who received DST instruction demonstrated more consistent gains across the measured aspects.

From a pedagogical perspective, DST can be considered a useful complementary strategy for EFL speaking classes because it provides students with opportunities for repeated practice, meaningful communication, and multimedia-supported learning. However, these findings should be interpreted cautiously due to the limited sample size and quasi-experimental design without random assignment. Therefore, further studies with larger samples and more rigorous experimental procedures are recommended to confirm the effectiveness of DST in different educational contexts.