As this research aimed to investigate the relationship between the preservice teachers’ stated beliefs about teaching speaking skills and their actual teaching practices, observations were a crucial data collection method (Adler & Adler, 1994). This instrument is essential because it “provides direct evidence of behavior, is (in theory) non-interventionist, and allows a large amount of descriptive data to be collected” (Borg, 2006, p. 227).

This study followed the observation procedures stated by Cohen et al. (2018). We utilized semi-structured observations, with the focus determined by each participant’s stated beliefs. Nonetheless, this approach did not confine our observational process; we remained open-minded, aiming to explore the situation as naturally as possible. The lead researcher assumed the role of “nonparticipant observer” (Creswell, 2013, p. 213), maintaining descriptive and reflective field notes to write details that could not be audio recorded, such as teacher nonverbal communication.

Cohen et al. (2018) argue that interviews are a potent tool for delving into intricate and profound issues. This type of interview “elicits teacher’s verbalizations while [he or she] looks at a replay of herself/himself performing a task” (Fang, 1996, p. 57). Rahman and Singh (2023) used this tool to confirm the relationship between oral corrective feedback beliefs and practices.

We conducted a round of stimulated recall interviews (SRI) with each participant at the end of their classes. The interviews were conducted in Spanish, and, with the participants’ written consent, they were audio-recorded, resulting in more than five hours of recorded data. The questions were based on the participants’ performance and the activities conducted in class. The objective of employing SRIs was to gain insights into the participants’ decision-making processes, instructional strategies, and reflective practices, thereby providing a deeper understanding of their pedagogical approaches and developmental needs.

We maintained field notes while observing the participants’ classes, following the protocol proposed by Phillippi and Lauderdale (2018). Field notes enabled us to capture the behaviors, activities, events, and various aspects observed (Schwandt, 2015), thereby ensuring a rich empirical basis for analysis.

Additionally, field notes helped us gather our own insights and reflections, thereby enriching the data for lucid interpretation and comprehension. Field notes followed the checklist suggested by Spradley (1980), with an emphasis on

Table 2 summarizes the data collection procedure.

Table 2

Methods	Time for each participant	Total time
Classroom observation	Nine lessons for each teacher, 45 minutes for each lesson	34 hours
Interview	Once for each teacher, lasting one hour	5 hours
Field notes	Once for each observed class	45 reports

Collaborative work. Malena’s belief in the power of collaborative work was evident in her sessions, where she encouraged group activities and peer feedback among students, facilitating a shared learning environment. Her classes were based on this activity, and the way she arranged the seating favored group activities. She justified this type of task:

Malena’s belief is consistent with Harmer’s (2007) suggestion to provide different seating arrangements to foster the flow of spoken interactions and enhance the students’ confidence in speaking (Brown, 2001; Luoma, 2004). Students in small groups can share ideas and participate actively in the discussions generated in class. Correa et al. (2017) provide practical insights into classroom management, particularly regarding how physical arrangements impact language learning and student engagement. This belief suggests that educators should consider alternative seating arrangements to enhance interaction and participation in language classes.

Malena’s class seating arrangement (Figure 3) involves small groups in which students face their classmates, allowing for better conversation. Furthermore, Malena formed the groups taking into account the students’ peer preferences. Her role was to facilitate and monitor the discussions.

Figure 3

Games to teach. Rafaela, Francesco, and Violet employed games as a central teaching technique. This aligns with their belief that games can significantly enhance language retention and make learning enjoyable for all age groups. They used more game activities when teaching children. In one of the observed lessons, Rafaela said:

During the interview, Rafaela listed the reasons for using games to teach speaking skills to kids:

Meanwhile, Violet commented that using games helps students become more engaged in class:

Both preservice teachers highlighted the importance of engagement and enjoyment in learning through games. These findings suggest that incorporating games into the teaching of speaking skills can make learning more effective and enjoyable.

Use of technology. Francesco and Pepe utilized technological tools, including language learning apps and interactive software. During a typical class session, Francesco might guide his students through interactive lessons using apps that focus on gamification and collaborative work, which help students practice their speaking skills. Meanwhile, Pepe often used voice recording software that allowed students to practice speaking. These activities were combined with peer feedback, allowing students to practice interacting with each other and demonstrating the importance of making mistakes in the language learning process. By employing technological tools, Francesco and Pepe aimed to create a classroom atmosphere that was not only educational but also engaging and responsive to the needs of modern learners. According to Santos and Ramírez-Avila (2023), digital technologies can provide immediate feedback and personalized learning experiences, which are crucial for developing language skills, enhancing interaction, and fostering communicative competence—essential components in language acquisition.

English immersion strategy. Rafaela consistently applied an English-only policy in her classroom to immerse students in the language, reflecting her belief that immersion aids quicker and more natural language acquisition (Cook, 2001). She spoke no Spanish words; everything was done in English. During the interview, she stated: “I believe it is important to speak 100 % in English because you gradually get used to the language and start learning. Maybe not through translation, but perhaps through logic.”

Repetition to teach pronunciation. Finally, a standard methodology among Malena, Violet, Francesco, and Pepe was the use of repetition to teach pronunciation. This approach was crucial for developing accurate and confident speech abilities in students across different levels. The following two episodes demonstrate the use of repetition in the classroom.

Pepe, Malena, and Violet used repetition in various moments of the class, each with the same goal: to have students practice their pronunciation. Violet mentioned,

Although belief and practice were congruent, this stated belief contradicts the principles of CLT, which prioritize meaningful communication over accuracy and fluency (Richards & Rodgers, 2014). The reliance on repetition underscores a traditional view that repetitive practice can solidify language structures and aid fluency. However, it does not necessarily prepare learners for spontaneous language use in diverse communicative contexts.

English immersion strategy. While there was significant alignment in many areas, a divergence was observed in the application of the English immersion strategy by Malena and Violet. Despite its proven effectiveness and popularity among some educators (Cook, 2001), as demonstrated by Rafaela’s implementation, Malena and Violet deviated from this approach, going against their stated beliefs. Both preservice teachers used Spanish to give instructions about the activities, which were then conducted in English. Episode 3 is from Violet’s first speaking class.

This utterance demonstrates her effective teaching strategy and reflects academic perspectives on language teaching methodologies. Specifically, Violet’s decision to initially provide instructions in Spanish to ensure clarity and comprehension mirrors the theoretical insights offered by Bateman (2008), who explained that the first language can be strategically used for clarification or giving directions in class. This bilingual approach acknowledges the advantages of using students’ first language to establish a solid basis for participation in later English-speaking activities. By ensuring that students fully understand the instructions, Violet effectively facilitates a smoother transition to English, enhancing the communicative efficacy of her lessons. This practice underscores the dynamic interplay between teachers’ beliefs and the pedagogical theories that support using L1 as a scaffold in EFL teaching, thereby validating Bateman’s (2008) assertion in a real-world educational setting. During the interview, Violet stated:

Using the L1 reflects an adaptive teaching strategy, recognizing the students’ diverse linguistic backgrounds and the necessity of scaffolding their learning experiences to facilitate language acquisition. The switch from L2 to L1 demonstrates Violet’s commitment to providing the best for her students. As Borg (2003) argues, “teachers are active, thinking decision-makers who make instructional choices by drawing on complex, practically oriented, personalized, and context-sensitive networks of knowledge, thoughts, and beliefs” (p. 81).

Timing. Malena, Francesco, and Violet share a common belief in the importance of timing in error correction. They prefer to correct errors at the end of an activity or a speaking task rather than interrupting immediately. This approach is based on their belief that immediate corrections can hinder fluency and discourage learners. By allowing students to complete their thoughts, these educators aim to foster a more supportive and less intimidating environment that encourages communication and risk-taking in language use. The following episode is from Francesco’s and is about the parts of the body. He started the class with a song and then asked the students to repeat the parts of the body that were mentioned in the song.

In Episode 4, Francesco exemplifies his belief with Malena and Violet regarding the timing of error correction. When the student mispronounces “mouth” as [maʊt] instead of [maʊθ], Francesco waits until the student completes their sentence before providing the correct pronunciation. This method not only corrects mistakes but also preserves the students’ self-esteem and encourages ongoing participation by allowing students to express their thoughts first. These participants’ beliefs align with Gharaghanipour et al.’s (2015) assertion that correcting students at the end of their utterances can help reduce speaking anxiety. This convergence may be related to the preservice teachers’ experiences of feeling comfortable in class.

Timing. In contrast, Rafaela and Pepe believe in the effectiveness of immediate correction. They argue that addressing errors as they occur helps students immediately understand their mistakes and learn the correct usage in context. This method is believed to prevent the reinforcement of incorrect habits and promote accuracy in language use from the beginning (Dawood, 2014). However, during their classes, they corrected the errors at the end of the activities. They waited for the students to finish the thought, and later, they wrote the mispronounced words on the board and asked them to repeat.

In Episode 5, despite her belief in the benefits of immediate feedback, Rafaela’s approach to error correction aligns more closely with Francesco’s method, as seen in practice rather than in theory. Rafaela and Pepe advocate for immediacy in correcting mistakes to prevent the entrenchment of incorrect language habits and reinforce correct usage in the communicative context. However, in practice, Rafaela allows her students to complete their statements before addressing pronunciation errors, which does not align with her stated belief. This discrepancy between belief and practice could reflect a pedagogical compromise. By waiting until the end of the activity, Rafaela ensures that the students are not interrupted and can maintain their flow of speech, thus fostering a more fluid communication process.

Finally, in addressing our research question on how beliefs influence teaching practices, we observed that the preservice teachers who value interactive learning are more likely to integrate communicative language teaching methods into their classrooms. Their practices are influenced by a complex array of factors including their beliefs, the teaching context (e.g., the class seating arrangement, available technology, and lesson timing), institutional policies (e.g., the requirement to align with the CLT approach and the use of L1), and the characteristics of the learners (e.g., age, interests, learning styles, and motivation).