Student Perceptions of Behavior Control in a Bimodal Presentation Classroom

1. Communicative Competence and Presentation Skills in an EFL Classroom

Communicative competence has evolved into a central objective in EFL education, enabling learners to develop communication and critical thinking skills essential for academic and professional success. The theoretical foundations of communicative competence rest on a few influential frameworks that have shaped language pedagogy over the past several decades. Ahmed (2023) reports that the foundational work of Hymes (1972) defines communicative competence as the knowledge of a language and the ability to use it effectively in communication. Following Hymes’ theory, Canale and Swain (1980) established a comprehensive model that guided all future language teaching methods. Their model posited that the ability to communicate effectively required competence in grammar, “including knowledge of lexical items, grammatical rules, syntax, morphology, semantics, and phonology” (Ahmed, 2023, p. 15), a sociolinguistic competency, and a strategic competency. In 1983, Canale added discourse to the list of competencies. Bachman and Palmer (1996) further refined these concepts through their language ability model, where they divided language competence into two basic categories: organizational knowledge and pragmatic knowledge. These theoretical frameworks underscore the multidimensional nature of communicative competence, encompassing not only linguistic knowledge but also pragmatic, strategic, and organizational competencies that enable effective communication in diverse contexts.

Research demonstrates that live oral presentations serve as a powerful pedagogical tool for fostering communicative competence in EFL classrooms (Al-Issa & Al-Qubtan, 2010; Ati & Parmawati, 2022; Brooks & Wilson, 2014; Ljubičić, 2020). Studies indicate that live oral presentations help students develop competencies through interactive activities, improve speaking abilities through authentic language use, and build actional competence through communicative language teaching (CLT) (Afifah et al., 2024; Gürbüz & Cabaroğlu, 2021; Hallemans & Copeland, 2025).

The success of live oral presentations requires learners to integrate multiple skills, including language skills, organization skills, presentation delivery skills, and the ability to use technology to produce meaningful content (Hallemans, 2021). Through the preparation and delivery of presentations, students practice organizing ideas coherently, selecting appropriate language for the target audience, and managing the challenges of real-time communication by employing technology to enhance and deliver their presentation, all essential components of effective communicative competence.

2. Benefits and Challenges of Live and Recorded Presentations

Despite their pedagogical value, live oral presentations can significantly contribute to speaker anxiety, which may inhibit student performance and learning outcomes. Studies show that students believe fear and nervousness resulting from presentation anxiety negatively affect performance, potentially undermining the learning objectives that presentations are designed to achieve (Gürbüz & Cabaroğlu, 2021; Hallemans, 2021; Hallemans & Copeland, 2025; Razawi et al., 2019). The anxiety associated with live oral presentations represents a significant barrier to maximizing the developmental benefits of this pedagogical approach. Students who experience high levels of presentation anxiety may avoid meaningful language practice, limit their linguistic risk-taking, and fail to fully engage with the communicative opportunities live oral presentations provide.

Emerging research suggests that technology-mediated presentations, such as asynchronous recorded oral presentations, offer promising solutions to the anxiety challenges inherent in traditional live oral presentations (Hallemans, 2021; Hallemans & Copeland, 2025; Mishu et al., 2022). Video-based assignments allow students to practice with multiple takes, building confidence in their output and reducing presentation anxiety through increased preparation time and the lack of pressure from a live audience. The asynchronous nature of recorded presentations provides learners with greater control over their performance conditions, enabling them to manage their anxiety more effectively while still engaging in meaningful oral communication practice.

A study by Mishu et al. (2022) specifically demonstrated that the delivery of online presentations, both asynchronous and synchronous, positively reduced anxiety and stress for introverted learners, wary of speaking with an audience in front of them, suggesting that format flexibility can accommodate diverse learner needs and preferences. However, Böhme (2009) and Smirnova and Nuzha (2013) have argued that students still needed the chance to deliver an oral presentation before a live, face-to-face audience, as the real-time communication practice was critical for the development of oral communication skills. These findings suggest that integrating recorded presentations into the EFL curricula along with live presentations may help educators balance the developmental benefits of oral presentations with the need to create supportive, low-anxiety learning environments.

The use of technology in oral presentations provides a critical opportunity for students to practice skills that align with the demands of contemporary professional environments. These industry-relevant skills, including digital literacy and problem-solving capabilities are vital in the 21st century (Larson & Miller, 2011). Live presentations require students to become adept in presentation software, while recorded presentations add the element of video production and online dissemination. Both methods require students to engage in the problem-solving processes, especially when navigating the complexities inherent in modern technology.

3. Clearly Defined Assessment Criteria

Low-anxiety learning environments are critical for the success of an oral presentation classroom. Research shows that part of reducing anxiety and providing an effective classroom environment for learning involves producing and providing a clearly defined rubric for presentation assessment (Copeland, 2021; Hallemans & Copeland, 2025; Lan et al., 2024; Razawi et al., 2019; Viswanathan & Ng, 2025). Critical components for assessment of oral presentations include the appropriate use of language, clear organization, non-verbal presentation delivery skills, and the integration of technology in the presentation. In other words, students need to prepare a script using appropriate grammar and vocabulary, organize the information so that it is easy to understand, use appropriate posture, gestures, and eye contact while managing technology intended to complement their presentation.

Clearly defined rubrics increase transparency and provide students with greater ownership and awareness of their learning stage. Assessment using clearly defined rubrics helps students exert greater control over the learning process, which causes a reduction in anxiety as well as negative perceptions of learning, allowing students to control and take responsibility for their learning and its outcome (Panadero & Jonsson, 2013).

While the literature provides substantial evidence for the benefits of both live and recorded presentations independently, a significant gap exists in the research examining pedagogical approaches that systematically combine both formats within a unified assessment framework. Searches for studies on classrooms that assign both live and recorded oral presentations yielded limited results, suggesting this represents an underexplored area of EFL pedagogical research. Hallemans and Copeland (2025) studied student preferences between the two different modes of delivery, but not the reported learning outcomes and perceived behavior control regarding future use of these skills. This gap is particularly significant given the complementary benefits of each format: recorded presentations reduce anxiety and provide opportunities for multiple practice attempts and self-correction, while live presentations develop real-time communication skills and authentic interaction abilities. This bimodal approach could also influence intended future behavior by reducing the fear of making presentations, developing self-belief through constructive and open feedback and assessment, as well as creating the communication and technological skills required for an effective presentation. The absence of research on this type of learning environment points to the need for an empirical investigation of a classroom that utilizes blended methodologies, leveraging the strengths of both formats, and promoting a positive perception of the utility of these skills in their future careers.

4. Theoretical Framework: The Theory of Planned Behavior

The Theory of Planned Behavior states that intention to perform a behavior depends on three factors: positive attitude towards performing the behavior, the subjective belief that others approve, and perceived behavioral control or confidence in their ability to carry out the behavior with ease (see Figure 1). These factors influence the actual behavior, especially for deliberate actions (Ajzen, 1991). The design for a bimodal oral presentation class that incorporates both live and recorded assignments with a clearly defined and explained assessment criteria can help predict future planned behavior if appropriately executed. Creating a positive attitude towards presentations through scaffolded activities of recorded and live assignments helps boost perceived behavioral control through skill-building exercises, such as creating videos and PowerPoint presentations. Video presentations help students gain confidence in their abilities, enabling them to self-correct and re-record until the desired outcome is achieved. At the same time, live audience presentations allow students to receive real-time feedback on their delivery. Together, this feeds their positive attitude toward performing the behavior while a clearly defined rubric and audience feedback help develop their subjective norms by seeing their progress as encouragement to continue. Their confidence in their abilities as a result of the classroom design should impact their perceived behavioral control.

Fig. 1.

The Skills Added to the Theory of Planned Behavior Model

1. Participants

A total of 223 South Korean students from a single medium-sized university took part in the study. The students were all enrolled in a required first-year academic speaking EFL class. The classes were leveled with a TOEIC-style pretest. Most students were enrolled in the intermediate level, which was defined by the university as scores between 401-700 on the pretest, with 174 students reporting in this category. There were also 37 beginner-level students, who had scores of 400 and under. Finally, 12 students had been enrolled in the advanced level classes with pretest scores over 701 points. The students also reported their majors, which covered a large spectrum (see Table 1). There were 87 engineering majors representing 39.0% of the participants. The next largest group was the social sciences with 40 participants. This was followed by 35 business, 22 software, 19 humanities, 11 music and art, seven law, and two education majors.

Table 1.

Breakdown of Students by Major

The survey was distributed by two professors at the same university in South Korea. Both teachers have experience teaching university-level courses with 16 and 17 years at their present positions. The course employs a university developed syllabus, similar assessment rubrics, and a classroom textbook that has been used since 2016. Both teachers started to use video-based presentations during the COVID-19 teaching period in 2020 and decided to integrate video presentations into the offline curriculum, in concert with live presentations when they returned to the classroom.

The class was a three-credit-hour course that met twice a week over 15 weeks during the semester. The textbook had seven units. Each unit took four classes to complete. The first class dealt with speaking activities designed to activate background knowledge of the unit topic. During the next class, the teachers worked on presentation organization, including effective introductions, detailed bodies, and comprehensive conclusions. The third class covered two specific presentation skills drawn from the textbook, as well as reviewed preparation methods necessary for the live or recorded presentation assignments. The lesson also included instructions regarding the use of technology, including presentation software, video recording, and sharing tips. During the fourth class of the unit, students alternated between delivering a live presentation or watching the submitted recorded presentations. The first, third, fifth, and seventh presentations were live presentations while the second, fourth, and sixth were recorded presentations.

2. Data Collection

A survey was made available to the students at the end of the semester through an announcement in the university’s online learning platform during the fall semesters of 2022, 2023, and 2024. Since the survey was offered through an announcement instead of as a part of the weekly lesson plan, the teacher was unable to see who followed the link to Google Forms, which administered the data collection. The only personal data collected was the student's major and course level. In the announcement, it was clearly written in both English and Korean that the survey was voluntary and anonymous in nature (see Appendix). The data collection began after the grades had been completed and posted through the school’s online system.

The survey consisted of 20, 4-point Likert-type questions. A response of four corresponded to strongly agree while three was agree, two was disagree, and one was strongly disagree. The first 19 prompts in the survey consisted of students rating the skills taught in the class based on “What skills do you feel you developed during the class?” grouped by language, presentation, organization, and technology skills. The final question of the survey asked, “I found this course content useful for my future.”

3. Data Analysis and Assumptions

The data was exported from Google Forms and imported into IBM SPSS Statistics (Version 28). The first research question can be determined through an analysis of descriptive statistics. This requires a Cronbach’s α for internal consistency. The second research question explores how the four skill factors, language, presentation, organization, and technology, integrate into the Theory of Planned Behavior (see Figure 1). The items from the survey explore whether students perceive self-development in each of the 19 skills. The 19 items were initially grouped into four skill domains based on the course objectives and the presentation assessment rubric used throughout the semester. A factor analysis was conducted to group the four skills into distinct factors, thereby enabling the application of a multiple regression analysis. There are six assumptions must be met prior to completing these statistical tests: Cronbach’s α for each factor, sampling adequacy, factor extraction, factor structure, commonalities, and factor loading (see Table 2). These assumptions are tied to the ability to break the items into factors and whether there is a large enough sample size to complete the tasks. There are five assumptions that need to be met before data analysis.

Table 2.

Factors, Factor Cronbach’s α, Means, Standard Deviations, Factor Loads, and Extraction Communalities

The first assumption for both research questions is the Cronbach’s α test. The purpose of this test is to determine the internal reliability of the data. Bland and Altman (1997) stated that for scale items, a Cronbach’s α between .700 and .950 was considered optimal. The data set had an overall Cronbach’s α of .905, indicating that the overall study contained internal reliability. The internal reliability for each skill was determined and ranged between a Cronbach’s α of .799 and .915. All the Cronbach’s α scores were within the acceptable range, indicating that the statistical tests could continue.

The second assumption is whether the data has an appropriate sample size to conduct a statistical test. The Kaiser- Meyer-Olkin (KMO) measure of sampling adequacy returned a score of .873. According to Shrestha (2021), a score between .800 and 1.000 indicates an adequate sample size to complete a factor reduction. The Bartlett’s Test of Sphericity checks if there is enough relation between the variables to attempt to detect a structure. A significant finding, “indicates that a factor analysis may be worthwhile” (Shrestha, 2021, p. 6). The data set returned an approximate χ² (171) = 2655.950, p = .000. Both statistical tests indicate that a factor analysis is possible.

Third, all the skill items were included in a scale reduction to create factors. A Kaiser (eigenvalue) criterion was run, and four factors had values over 1.000 eigenvalues. A score of 1.000 eigenvalue is considered the cut-off for significance within new multiple prompt factors (Shrestha, 2021). The four factors accounted for 68.91% of the variance in the model. Peterson (2000), in a meta-analysis, found that the average score for published factor reductions was a 56.6% variance explained. This study’s variance explained is approximately in the 80th decile according to Peterson (2000).

Next, the items were then placed into the four factors, using principal component analysis rotated by varimax with Kaiser normalization (see Table 2). The items aligned almost perfectly with the way they were labeled on the survey. There was one exception, 9. Presentation skills: Use of note cards was grouped with the organization skills factor based on the rotated component matrix.

The fifth verification on whether a factor analysis could be used in further statistical tests was the extracted communalities of the items. Field (2013) noted that average extracted commonalities with at least .60 for large samples and .70 for small samples are suitable for factor analysis. The range of extracted communalities for the study was .48 to .85 with an average of .70, meeting this criterion for using the statistical method.

The final necessary condition for factor reduction is that the model includes factor loads of sufficient size. The factor loads make up the rotated component matrix. Peterson (2000), in analyzing 401 factor reduction studies, found that a threshold of .32 was required for absolute factor load. The lowest factor load for all the variables was Presentation skills: Use of note cards, with a factor loading of .485, well above Peterson’s minimum threshold.

1. Students’ Perceptions of Skill Development in a Bimodal Classroom

The first research question asks, “What skills do students perceive they developed from this bimodal classroom?” To answer the question, the descriptive statistics were used. The participants were surveyed on how they thought they developed after completing a bimodal presentation class. The skills were grouped into areas covered during the semester and fit into four general categories: language skills, presentation skills, organization skills, and technology skills. Out of the 19 prompts, all but four returned a response between three and four or agree and strongly agree that these skills developed as a result of the class (see Table 2). The other four prompts were closer to agree than disagree denoted by a mean score above 2.5.

The first skill category, language skills, had five items. The responses ranged from an average of 2.93 to 3.43 out of a possible high score of four on the survey (see Table 2). Speaking confidence had the highest mean at 3.43, SD = .850, denoting the highest level of agreement of skill improvement. It was followed by use of vocabulary (M = 3.22, SD = .754), intonation and stress (M = 3.21, SD = .706), and pronunciation (M = 3.18, SD = .702). Grammar was the only response below an average rating of three, with a mean = 2.93 (SD = .841). Speaking confidence, as the highest rated item, indicated that students believe that the bimodal classroom helped them overcome some of their anxiety related to public speaking, as was found in Hallemans and Copeland (2025). It is possible that the lower grammar score is a result of the lower level of emphasis on grammar in the textbook, which was used to teach the class, when compared to the testing-based English classes Korean students take in high school, as described in Jung and Crookes (2025). Overall, these responses indicate that students in a bimodal presentation classroom perceive growth in their language skills.

The second category on the survey, with four items, is presentation skills. The responses for these questions ranged between agree, three, and strongly agree, four, on the survey, with means ranging from 3.10 to 3.30 (see Table 2). The three highest presentation skills were eye contact (M = 3.30, SD = .721), gestures (M = 3.28, SD = .719), and body language (M = 3.22, SD = .760). The final survey item, which was originally categorized as a presentation skill, was the use of note cards (M = 3.10, SD = .794). While creating factors, it was grouped with the third factor: organization skills, due to the rotated component matrix results. Students perceived an improvement in the use of all the presentation skills listed on the survey. This finding was corroborated by Soureshjani and Ghanbari (2012), who also found that teachers and students believed body language, gesture, and eye contact were important factors leading to effective oral presentations.

The third category on the survey was organizational skills. There were six items in this area: overall organization of a presentation, use of openers, elements of a strong introduction, preparing an outline/brainstorm, elements of a strong conclusion, and use of closers (see Table 2). Students reported development in all six areas, with means ranging from 3.12 to 3.30. The highest perceived improvement came in the overall organization of a presentation (M = 3.30, SD = .669). Then, participants perceived improvement in elements of a strong conclusion (M = 3.26, SD = .676) and elements of a strong introduction (M = 3.25, SD = .684). Use of closers (M = 3.20, SD = .733) and use of openers (M = 3.17, SD = .744) followed in levels of student perceived improvement. Finally, preparing an outline/brainstorm (M = 3.12, SD = .740) was also above a three, in agreement with perceived improvement. Therefore, students had perceived improvements in all the categories of organizational skills. This is important as Al‐Issa and Al‐Qubtan (2010) found that the development of organizational skills through presentations is critical for future careers.

The final area of the survey, technology skills, had the lowest mean (see Table 2). Use of visual aids/PPT was the only prompt that received a mean above three (M = 3.06, SD = .844). The other three items in the category received perceived improvement means between 2.65 and 2.80, making them the lowest of the survey. However, all three areas were above the inflection point of 2.50 between agree, rating of three, and disagree, a rating of two. All three items also had the highest standard deviation scores, indicating that there was more of a range in responses than on other items in the survey. Making a video had a mean of 2.80 (SD = .943), followed by use of a website to share a video (M = 2.70, SD = .918), and video editing (M = 2.65, SD = .937). While overall, there is a perception from students that these skills were improved by the class, it is not as strong, and is more varied from student to student. Al‐Issa and Al‐Qubtan (2010) opined, “Oral presentations are ideal tools for introducing students to advanced and sophisticated technology and training and encouraging them to use it” (p. 231). The survey results indicate technology might be an area to add more emphasis on in future iterations of the course.

Since the participants tended towards agreement with all items on the survey, students believed that they improved their presentation skills in the bimodal presentation classroom. With an average score across the prompts of 3.27, the presentation skills represented the highest value of student perceived development. Organization skills had an average mean of 3.20 which was very close to language skills at 3.19. Technology skills had the lowest average of means at 2.80 indicating a slightly lower perception of development in this area of the rubric. Overall, the first research question of whether students perceived development in their skills after taking the bimodal presentation course can be answered in the affirmative.

2. Predicting Future Use Through the Theory of Planned Behavior

The second research question asked, “How do student perceptions of their skills development in language, presentation, organization, and technology influence their perceived behavioral control regarding future use of these competencies?” To test the second research question, a multiple regression was performed to see how the factors inform perceived behavioral control (see Figure 1).

The final question on the survey was, “I found this course content useful for my future” (M = 3.60, SD = .576). This result indicates that students tended to strongly agree with the statement. Based on the response, students perceived the ability to use what they learned toward future endeavors. This ‘can do’ attitude could also be read as a perceived behavioral control to use the learning in the future. Perceived behavioral control is an important component of a student’s behavioral intention to use what was learned, as described in the Theory of Planned Behavior (Ajzen, 1991).

Since the assumptions for factor reduction were met, a standard multiple regression was performed to determine how the different skill groups taught in the class predicted perceived behavioral control for the future (see Figure 1). The correlation coefficient was determined between the variables. A positive statistic indicates that when one of the variables increases in magnitude, the other also increases in value. Prajapati et al. (2010) reported Cohen’s guideline for correlation coefficients was as follows: above .50 indicated a large effect size, between .30 and .50 as a medium effect size, and under .30 as a small effect size. The correlations between perceived behavioral control and each of the four skills fall within Cohen’s guideline for medium effect size which is between .300 and .500 (see Table 3). Hemphill’s (2003) analysis of 380 meta-analytic studies found that a correlation coefficient of over .30 was the baseline that defined the magnitude of the top third of studies. A correlation coefficient of .50 is in the 89th percentile of all the studies analyzed. The highest correlation was between perceived behavioral control and presentation skills with a correlation coefficient of r = .435, p ≤ .005 (see Table 3). Perceived behavioral control and organization skills had a correlation coefficient of r = .358, p ≤ .005. Perceived behavioral control and language skills had a correlation coefficient of r = .336, p ≤ .005. The lowest correlation coefficient related to perceived behavioral control was technology skills with r = .303, p ≤ .005.

Table 3.

Results of a Standard Multiple Regression to Predict Perceived Behavioral Control From the Four Factors

The inter-term correlations between the four factors were also determined (see Table 3). The correlation coefficient between presentation skills and organization skills was the greatest, displaying a large effect size of r = .505, p ≤ .005. Language skills’ correlation coefficient with organization skills, r = .495, p ≤ .005, and presentation skills, r = .494, p ≤ .005, were a medium effect size. The inter-item correlation coefficient between organization skills and technology skills was r = .383, p ≤ .005, indicating a medium effect size. The inter-item correlation coefficient between technology skills and language skills, r = .209, p ≤ .005, as well as presentation skills, r = .205, p ≤ .005, both displayed a small effect size, but are still within the middle third of published studies of magnitude of correlation coefficients. All of the correlation numbers indicate that there is a proper size in power for a regression statistical test (Amjad et al., 2020).

The overall regression using the four skills as predictors of a student’s perceived behavioral control was statistically significant at an adjusted R² = .38, F(4, 218) = 44.74, p ≤ .005. Therefore, the four skills combined represent 38% of the variance in the model for perceived behavioral control of the bimodal presentation classroom (see Figure 2). All four of the skill predictors were significant at the 95% level or better. First, the strongest effect on perceived behavioral control was presentation skills which had a standardized coefficient β = .30, t(222) = 4.22, p ≤ .005. The second highest effect was technology skills with a standardized coefficient β = .19, t(222) = 2.97, p ≤ .005 in terms of perceived behavioral control. Next, language skills’ effect on perceived behavioral control was a standardized coefficient β = .16, t(222) = 2.33, p = .021. Finally, organization skills had a standardized coefficient β = .15, t(222) = 2.04, p = .042 in relation to perceived behavioral control.

Fig. 2.

The Skills Added to the Theory of Planned Behavior Model

Based on the results, all of the skills positively affect the perceived behavioral control, predicting future intention and behavior to use the skills taught in the course. The percent variance of the four skills was the only part of the Theory of Planned Behavior model from Ajzen (1991). However, the model predicts that the attitude towards the behavior and the subjective norm, also represent a portion of the variance associated with the perceived behavioral control and were not covered by this study.