Enhancing Occupational English Speaking Skills among STEM Students through an Integrated EOP–Ethnopedagogical Digital Framework

English for occupational purposes in higher education

English for Occupational Purposes (EOP) is explicitly designed to prepare learners for professional communication in specific occupational contexts, with particular emphasis on authentic workplace language use.^11,12 EOP instruction is distinguished by its systematic integration of needs analysis, domain-specific authentic materials, and task-based activities that reflect communicative demands learners will encounter in their future professions.^13,14 The emergence of digital EOP platforms offering unprecedented flexibility, interactivity, and access to authentic multimedia content represents a transformative opportunity for higher education institutions seeking to deliver pedagogically relevant occupational language instruction aligned with 21st-century workplace demands.^15,16 Mobile-assisted language learning approaches have further demonstrated efficacy in developing occupational speaking competencies, with EFL learners reporting measurable improvements in fluency and communicative confidence through targeted digital speaking activities.^17,18

Furthermore, cross-cultural concepts in EOP programs address both linguistic and cultural demands in professional settings.¹³ EOP is one of the essential competencies that prospective graduates need to master to enhance their employability and readiness to enter the workforce.¹⁹ EOP instruction is designed to strengthen students’ practical skills, as they are expected to use English appropriately in specific situations and in their professional fields in the workplace.²⁰ In short, EOP instruction plays a crucial role in preparing students for their future careers by integrating authentic, workplace-oriented, and cross-cultural learning materials that enhance professional language competence and readiness for employment.

Speaking Ability in Occupational Contexts

Speaking proficiency is one of the most in-demand competencies in modern professional workplaces. In occupational settings, the capacity to communicate orally with clarity, precision, and cultural appropriateness is a fundamental prerequisite for career success.^21–23 Drawing on established assessment frameworks,²⁴ the key components of speaking proficiency include: (1) fluency natural pace and automaticity; (2) pronunciation intelligible sound production; (3) grammar accurate syntactic structures; (4) vocabulary breadth and contextual appropriateness; and (5) comprehension real-time decoding and response. The simultaneous development of all five dimensions within a single instructional module represents a substantial pedagogical challenge that digital, task-based EOP instruction is uniquely positioned to address through authentic, scaffolded, and iterative speaking practice.^25,26 Emerging immersive technologies such as virtual reality have further demonstrated effectiveness in developing EFL speaking skills through situated learning approaches that authentically simulate occupational communication contexts.²⁷

Effective workplace speaking requires more than just putting words together clearly; it also means using body language, gestures, and facial expressions to communicate.^28,23 Students must construct meaning by producing, receiving, and processing information during interaction. In addition, students need to master several aspects of speaking skills grammar, pronunciation, fluency, vocabulary, and comprehension in order to ensure effective and efficient communication.^24,29,30 In other words, effective speaking involves not only clear verbal expression but also the appropriate use of nonverbal communication, and mastery of linguistic components to achieve meaningful and successful interaction.

Ethnopedagogy of riau malay culture in language learning

Ethnopedagogy is a culturally grounded educational framework that systematically incorporates indigenous knowledge systems and community-based values into formal instructional practice.^31–34 This orientation finds robust grounding in³⁵ sociocultural theory, which posits that learning is inseparably mediated by cultural tools and interpersonal interaction within specific sociocultural contexts. Ethnopedagogy resonates deeply with the framework of Culturally Responsive Teaching (CRT), which argues that instruction is most effective when it deliberately connects content to students’ sociocultural backgrounds.^36,37 Riau Malay culture embodies politeness, cooperation, hard work, and self-confidence as foundational professional character values.^38,39 EOP EMPOWER operationalizes four of these values as active pedagogical tools that shape speaking task design and the construction of authentic professional scenarios.⁴⁰ Demonstrate that learners who perceive a coherent relationship between the target language and their own cultural identity develop stronger motivation, deeper engagement, and more sustained language learning trajectories.^41,37,42 Riau Malay culture embodies a rich system of identity values including politeness, respect, cooperation, social harmony, and dignity that collectively constitute the ethical and philosophical foundation of the Malay professional character.^38,43

Research design

This study was conducted over an eight-week instructional period involving both experimental and control groups. Prior to the intervention, all participants completed a speaking pre-test to determine baseline equivalence. The experimental group received instruction through the EOP EMPOWER platform, which integrated occupational English materials, multimedia-based speaking activities, and ethnopedagogical values derived from Riau Malay culture. Students in the experimental group engaged in structured learning activities including video-based presentation tasks, simulated job interviews, collaborative discussions, pronunciation exercises, and culturally contextualized workplace communication practices. The implementation followed the procedural stages illustrated in Figure 2, beginning with orientation and pre-testing, followed by systematic instructional treatment, guided speaking practice, peer interaction, and post-test evaluation. In contrast, the control group received conventional classroom instruction using printed textbooks and PDF-based materials without access to the EOP EMPOWER platform or ethnopedagogical integration. Both groups studied equivalent occupational English topics and were taught over the same instructional duration to maintain procedural consistency. However, only the experimental group experienced digitally mediated interactive learning and culturally responsive speaking activities embedded within the EED Framework.

Figure 2. Research design and procedural framework.

Participants

Participants were selected via purposive sampling.⁴⁴ Selection criteria required Computer Science students currently enrolled in a compulsory English course at universities with equivalent BAN-PT Level B accreditation. All participants were adults aged 18–19 years at the time of data collection. The total sample comprised 162 students: 81 in the experimental group and 81 in the control group ( Table 1). Prior to intervention, both groups demonstrated statistically equivalent baseline speaking proficiency (pre-test: experimental M = 53.05, control M = 53.02; p = .990), confirming baseline equivalence.

Instrumentation and data analysis

Speaking proficiency was assessed using²⁴ rubric across five dimensions (pronunciation, grammar, vocabulary, fluency, comprehension), each scored on a five-point scale. Three trained TEFL experts independently rated video-recorded performances. Inter-rater reliability (ICC = .92) and internal consistency (Cronbach’s α = .832) confirmed excellent instrument reliability. Content validity was established via Aiken’s V = .89. Statistical analyses included Kolmogorov–Smirnov normality tests, Levene’s homogeneity tests, paired-samples t-tests, independent-samples t-tests, and one-way ANCOVA for partial eta squared effect size.⁴⁵

Instructional intervention: EOP EMPOWER platform

EOP EMPOWER is a web-based EOP e-module grounded in Riau Malay ethnopedagogy developed by the authors, accessible via www.eop-empower.com. The platform comprises four structured units, each containing eight sequential learning stages: Lesson Learning Outcome, Warm-Up, Presentation, Application, Freer Practice, Summary, and Discussion. Figure 3 illustrates the Presentation stage, which delivers content through multimedia input, including embedded video. Figure 4 illustrates the Application stage, where learners engage in interactive practices aligned with occupational scenarios. Figure 5 illustrates how students learn and discuss the identity of Riau Malay culture (ethnopedagogy) in the Application stage, specifically politeness, which is relevant to workplace communication.

Figure 3. An example of the presentation section of EOP EMPOWER.

Figure 4. An example of the application section of EOP EMPOWER.

Figure 5. An example of Ethnopedagogy of Riau Malay Culture content in EOP EMPOWER.

Assumption checks: Normality and homogeneity

Prior to inferential testing, the normality of the data distribution was examined using the Kolmogorov Smirnov test. Table 2 presents the results of the normality test for both the experimental and control groups across pre-test and post-test scores.

Following the normality check, the homogeneity of variance between the two groups was examined using Levene’s test. Table 3 presents the results of the homogeneity-of-variance test for both the pre-test and post-test scores.

Descriptive statistics

Table 4 presents the descriptive statistics, including the mean (M) and standard deviation (SD), for both the experimental and control groups across the pre-test and post-test measurements.

Figure 6 illustrates the pre-test and post-test mean scores for both groups. Both groups had almost identical speaking proficiency before the intervention (Experimental: M = 53.05, SD = 9.27; Control: M = 53.02, SD = 9.19). Following the eight-week intervention, the experimental group demonstrated substantially greater improvement (M = 71.21, SD = 8.14; gain = 18.17 points) than the control group (M = 61.57, SD = 8.36; gain = 8.54 points), with a ratio of 2.13:1.

Figure 6. Pre-test and Post-test mean speaking scores by group.

RQ1: Within-Group improvement in speaking ability

To address RQ1, a paired-samples t-test was conducted to examine within-group improvement in speaking performance from pre-test to post-test for each group. The results are presented in Table 5.

Table 5 reveals that both groups demonstrated statistically significant within-group improvements from pre-test to post-test (p < .001). The experimental group showed a mean improvement of 18.168 points (t(80) = −19.138, p < .001), while the control group showed a mean improvement of 8.543 points (t(80) = −15.865, p < .001). The magnitude of improvement in the experimental group was more than twice that of the control group, indicating that EOP EMPOWER produced a substantially greater within-group effect on speaking development.

RQ1: Between-Group comparison of improvement

To further address RQ1, an independent-samples t-test was conducted to determine whether a statistically significant difference existed between the experimental and control groups at both the pre-test and post-test stages. Table 6 presents the results of this between-group comparison.

Table 6 shows no significant difference between groups at baseline (p = .990 > .05), confirming equivalent baseline ability. Following the eight-week intervention, a highly significant between-group difference emerged (t(160) = 6.306, p < .001), with the experimental group outperforming the control group by 9.649 mean points, indicating that EOP EMPOWER produced greater improvement in speaking development.

RQ2: Speaking component analysis

Figures 7 and 8 present per-component scores and gain scores across both groups. The experimental group achieved greater gains across all five components compared to the control group, with vocabulary (+5.2) and comprehension (+4.4) demonstrating the most pronounced improvements, followed by fluency (+3.8), grammar (+3.4), and pronunciation (+3.1).

Figure 7. Speaking component scores: Pre-test and Post-test across both groups.

Figure 8. Mean gain score per speaking component.

The experimental group demonstrated consistently greater improvement across all five components. Figure 7 reveals that the most significant improvements are shown in vocabulary and comprehension, while fluency is also markedly enhanced. These results show that the e-module helps students improve their overall speaking skills, not just sentence structure, but also understanding and communicating ideas.

Figure 9 presents a radar profile comparing the experimental group’s pre-test and post-test performance across all five speaking dimensions. The radar visualization confirms that improvement was balanced and comprehensive across all components, with the post-test profile showing markedly expanded coverage relative to the pre-test baseline, particularly for vocabulary and comprehension. This multidimensional profile reinforces the finding that EOP EMPOWER produced holistic speaking development rather than gains confined to isolated dimensions.

Figure 9. Radar profile of the experimental group: Pre-test vs Post-test.

Effect Size Analysis

To quantify the practical significance of the EOP EMPOWER intervention, a one-way ANCOVA was performed with the pre-test score as the covariate and the post-test score as the dependent variable. Table 7 presents the ANCOVA results, including the partial eta squared (η²p) effect size.

Limitations and Future Research

While this study presents significant findings, several limitations warrant consideration. First, the sample was drawn from three universities within a single Indonesian province, constraining external generalizability. Second, the intervention addressed only two professional scenarios (job interviews and presentations) over eight weeks, not fully capturing the range of communicative demands that Computer Science graduates encounter. Third, the current implementation incorporated four of the twenty-nine Riau Malay identity values; future research should explore the differential contributions of additional values to professional speaking development. Fourth, and importantly, the current design does not isolate the independent effects of the ethnopedagogical integration and the digital learning components within the EED Framework.

Future researchers are recommended to replicate and extend this study across geographically, culturally, and institutionally diverse university settings. Longer intervention durations would facilitate the development of advanced speaking sub-skills requiring sustained practice, such as critical argumentation and extended professional discourse. Mixed-methods designs incorporating qualitative data could further illuminate the mechanisms through which ethnopedagogical integration enhances speaking development and professional identity formation.