Reducing Speaking Anxiety in EFL Classrooms: An Explanatory Mixed-Methods Study

This study outlines the impact of a process enriched with mini-speeches, presentations, and scaffolded feedback on EFL learners’ foreign language speaking anxiety. The participants included 52 university freshman students and the data were collected during an oral 14-week communication course. For quantitative data, the foreign language speaking anxiety scale (FLSAS) was administered as pre and post tests and the qualitative data included reflection reports and a focus group interview at the end of the semester. The data derived from the scale were analyzed through a paired-sample t-test and the qualitative data were analyzed using a content analysis scheme. The findings revealed a significant impact of the aforementioned activities, leading to around thirty percent reduction in participants’ EFL speaking anxiety. Highlighting the effectiveness of feedback sessions, the participants suggested that the major gains of this process were a) increased self-confidence, b) increased opportunities to speak, and c) learning from their own mistakes, all of which helped them become less anxious learners. In line with these findings, the study offers several suggestions for both practitioners and research purposes.


INTRodUCTIoN
The literature of second language acquisition (SLA) shows that studying the affective factors of second language learning has attracted the attention of researchers for some time. Of these factors, anxiety has probably attracted the most attention due to its debilitating effect on learners. Horwitz, Horwitz, and Cope (1986, p. 125) define anxiety as a "subjective feeling of tension, apprehension, nervousness, and worry associated with the arousal of the autonomic nervous system". They also suggest that it is obvious foreign language learning is in itself an anxiety provoking factor, and that this negative feeling "slows down progress and impedes success in foreign language teaching" (Nimmannit, 1998, p. 37). Guiora (1984) argues that language learning itself is an unsettling situation. As the first scholars focusing solely on foreign language anxiety, Horwitz, Horwitz, and Cope (1986, p. 128) define it as "a distinct complex construct of self-perceptions, beliefs, feelings and behaviors related to classroom language learning arising from the uniqueness of the language learning process." It is clear that learners feel stressed while learning a foreign language, and this stress level intensifies when it comes to speaking, no matter how willing and motivated they are to speak in the foreign language (Horwitz, Horwitz, & Cope, 1986). Luoma (2004) says that to be able to speak lies at the heart of using a foreign language, and every learner desires to be able to speak with others and express her/himself through speech in a target language. Due to this desire, they naturally become more anxious because of the difficulties they face in speaking in the target language (Luoma, 2004). The reasons learners feel anxious while speaking becomes salient when considering the features of spoken language, for example, the sounds of speech, spoken grammar, words, processing conditions, and reciprocity (Luoma, 2004). These factors have led many research studies on anxiety to focus on speaking anxiety (Horwitz, Horwitz & Cope, 1986;Horwitz, 2001;Phillips, 1992).
The literature shows that researchers have begun to identify the sources of EFL speaking anxiety, identifying various underlying reasons for anxiety levels. Gkonou (2011) states that learners fear being negatively evaluated by their teachers should they use inaccurate language and not meet their teachers' expectations. He's (2013) study with Chinese EFL learners found that speaking on an unfamiliar topic, having little time to prepare before speaking, and being tested orally were the main reasons learners feel anxious while speaking. In addition, Greer (2000) suggests that the fear of making mistakes and sounding comical to the others and having a low level of proficiency are also sources of speaking anxiety. In the Turkish context, Öztürk and Gürbüz (2014) identified several reasons for learners' anxiety: forgetting vocabulary items, waiting for his/her turn to speak, speaking in front of the others, and incorrect pronunciation.
Following this attention on speaking anxiety, researchers have investigated how speaking anxiety can be dealt with and reduced. As Sato (2003) states, when learners' anxiety level decreases, they learn better and more effectively. Bearing this in mind, Burden's (2004) study reveals that asking learners to speak on a familiar topic, focusing more on content rather than form and giving positive feedback rather than announcing "that's wrong" helped decrease learners' level of speaking anxiety. Other factors that can reduce learners' speaking anxiety include integrating project work into the curriculum, using teacher immediacy, providing indirect rather than direct feedback, using praise, and creating a supportive classroom atmosphere (Tsiplakides & Keramida, 2009). In addition, Karakaş (2012) found that the 251 teacher is an important figure, as the teacher has an enormous effect on students' levels of speaking anxiety, suggesting teachers need to adopt a motivating and encouraging attitude and minimize competition between and comparison of learners.
As well as student and teacher-related factors, the teaching techniques, tasks, and activities used in EFL speaking classes have been studied in relation to speaking anxiety. Ataş (2015), for example, made use of drama techniques in her EFL speaking classes to see whether they reduced her students' speaking anxiety levels. Based on 24 learners' pre and post tests and reflection reports, she concludes that using drama techniques reduced speaking anxiety levels and the students felt relaxed and less concerned about making mistakes and how others reacted to their utterances. Other researchers integrated oral presentation techniques into EFL speaking classes and investigated whether giving presentations affects learners' speaking anxiety levels. Hammad (2020) employed an experimental design to investigate the impact of oral presentations on speaking anxiety on 60 female Palestinian university students. The results indicate that the experimental group that gave oral presentations performed better in a speaking achievement test and had a lower level of anxiety compared to the control group, which did no oral presentation. Moreover, Coşkun (2017) integrated oral presentations into his classes and concludes that this activity reduced learners' anxiety.
How teachers support learners giving oral presentations in front of others matters because the activity is challenging (Woodrow, 2006;Liang & Kelsen, 2018) and provokes anxiety (Ayres & Hopf, 1993). One way to encourage learners is the use of feedback or, more specifically, scaffolded feedback. As studies have shown, learners care about the feedback given by their teacher, and they especially give importance to the way feedback is given (Burden, 2004;Tsiplakides & Keramida, 2009). Storch (2017) argues that scaffolded feedback enables learners to complete challenging tasks and, in the future, to complete such tasks independently. Looking into the effectiveness of scaffolded feedback, Rassaei (2014) found that scaffolded feedback assists the development of learners' oral performance. Zarei and Rezadoust (2020) carried out a study with 90 EFL learners, dividing them randomly into three groups: one given scaffolded feedback, one given unscaffolded feedback, and one given no feedback. The pre and posttest results of the three groups indicate that using scaffolded feedback decreases learners' speaking anxiety.
In general, recent research on foreign language speaking anxiety has focused more on the methods and techniques for reducing the level of speaking anxiety in language classrooms. However, more research is needed to understand the effectiveness of these methods and techniques and to explore new ways to help learners feel less anxious while speaking. This study aims to investigate whether and in what aspects the systematic combination of presentations, mini speeches, and scaffolded feedback, all of which have been found to have anxiety-reducing effects, influence language learners' speaking anxiety level. With this purpose, the study addresses the following research questions: 1. Is there a significant difference in the speaking anxiety level of participant EFL learners at the end of a semester enriched with mini-speeches, presentations, and scaffolded feedback sessions? 2. In what ways does this process influence EFL learners' foreign language speaking anxiety?

Research design
Commonly described as a combination of qualitative and quantitative strategies, mixed-method research enables researchers to "broaden the scope of the investigation and enrich the ability to draw conclusions about the problem under study" (Dörnyei, 2007, p. 164). In this regard, in order to fully understand whether and how a process of presentations, mini speeches, and scaffolded feedback can reduce the foreign language speaking anxiety of EFL learners, this study employed an explanatory sequential mixed-method design. In this research design, a QUAN → qual sequence is followed. In this sequence, first quantitative data is collected and analyzed, and the results are refined through follow-up qualitative data (Creswell, 2012). Pre and post tests were examined for decreased speaking anxiety, then reflection reports and a focus group interview were used to focus and broaden these findings in order to understand how the process influenced the participants' EFL speaking anxiety. The following figure summarizes the research process.

Participants
The participants of the study were 52 freshman students studying in the English language teaching department of a state university in Turkey. Before embarking on the program, students are required to obtain a particular score in a national university exam that mainly tests their language proficiency. Once registered with the university, students must achieve a score of over 80 (or demonstrate an equivalent score) in the university's language proficiency exam in order to be eligible to study in the English language teaching program. If they fail to achieve that score, they must follow a one-year intensive English preparatory program. All the participants of this study achieved the required score on the university examination to start their freshman year, suggesting their proficiency level was intermediate or higher. At the time of the study, they were all taking an oral communication skills course in two groups of 26. Thirty participants were female and 22 male and their ages ranged from 18 to 21.

Data collection tools
To collect the quantitative data, the foreign language speaking anxiety scale (FLSAS) with 18 five-graded Likert-type items, from "strongly agree" to "strongly disagree", was 4 2. METHODOLOGY

Research design
Commonly described as a combination of qualitative and quantitative strategies, mixed-method research enables researchers to "broaden the scope of the investigation and enrich the ability to draw conclusions about the problem under study" (Dörnyei, 2007, p. 164). In this regard, in order to fully understand whether and how a process of presentations, mini speeches, and scaffolded feedback can reduce the foreign language speaking anxiety of EFL learners, this study employed an explanatory sequential mixedmethod design. In this research design, a QUAN è qual sequence is followed. In this sequence, first quantitative data is collected and analyzed, and the results are refined through follow-up qualitative data (Creswell, 2012). Pre and post tests were examined for decreased speaking anxiety, then reflection reports and a focus group interview were used to focus and broaden these findings in order to understand how the process influenced the participants' EFL speaking anxiety. The following figure summarizes the research process.

Participants
The participants of the study were 52 freshman students studying in the English language teaching department of a state university in Turkey. Before embarking on the program, students are required to obtain a particular score in a national university exam that mainly tests their language proficiency. Once registered with the university, students must achieve a score of over 80 (or demonstrate an equivalent score) in the university's language proficiency exam in order to be eligible to study in the English language teaching program. If they fail to achieve that score, they must follow a one-year intensive English preparatory program. All the participants of this study achieved the required score on the university examination to start their freshman year, suggesting their proficiency level was intermediate or higher. At the time of the study, they were all taking an oral communication skills course in two groups of 26. Thirty participants were female and 22 male and their ages ranged from 18 to 21.

Data collection tools
To collect the quantitative data, the foreign language speaking anxiety scale (FLSAS) with 18 five-graded Likert-type items, from "strongly agree" to "strongly disagree", was used. The scale was adapted by Öztürk and Gürbüz (2014) Toubot, et. al. 2018;Çağatay, 2015), FLSAS has been reported to be a highly valid and reliable instrument, with an inner consistency of .91. The second phase of the study collected qualitative data from reflection reports and a focus group interview. Since this phase was conducted after collection of the quantitative data, the participants were asked to provide in-depth perspectives on how their speaking anxiety decreased at the end of the teaching process. Some of the participants (elaborated on in the following section) were asked to write reflection reports and others were invited to participate in a focus group interview. Both data collection tools were based on open-ended questions prepared by the researchers to gather the participants' voices regarding the process and how it affected their anxiety in terms of speaking. Before application, the questions were controlled by a colleague in terms of their orthography and other students from the department checked the questions for comprehensibility. After making some small revisions, the questions were prepared for the data collection process.

Data collection process
All the participants were taking an oral communication skills course and at the beginning of the semester, they were asked to sign a consent form stating that they were willing to participate and they were informed that they could leave the process whenever they wished (all the students signed and none of them left before the end of the study). The FLSAS was then administered as a pre-test at the beginning of the semester, which was 14 weeks.
The 14-week oral communication skills course was two hours per week and was conducted by one of the researchers (the teacher of the course) in two groups of 26 students. The course mainly included weekly tasks prepared by the teacher to help students practice their speaking skill. As part of the course, the students were asked to prepare two mini-speeches and two presentations. For the mini-speeches, one prepared and one non-prepared, students were asked to give two-minute speeches from their seated position in a U-shape class addressing their friends. For the presentations, of around 10 minutes, the students were asked to choose two topics of their own interest, prepare slide shows, and present to the whole class from the front. This process was organized at the beginning of the semester and conducted based on a schedule. In all the aforementioned activities, the researcher acted as observer taking notes on the students' performance for the weekly feedback sessions. Every week after the lesson, the researcher organized a feedback session for the students who had given a speech or presentation that week. This feedback process was scaffolded and, following the theoretical framework, based on three mechanisms (graduation, contingency, ongoing assessment) suggested by Ellis and Barkhuizen (2005) for scaffolded assistance. In these sessions, the participants received feedback on their accuracy, fluency, non-verbal language (eye contact, mimes etc.), and mistakes in grammar, vocabulary, and pronunciation were negotiated.
Towards the end of the semester, the FLSAS was administered as a post-test to see whether their speaking anxiety had changed following the process of speeches, presentations, and scaffolded feedback. Analysis of the data showed that their foreign language speaking anxiety significantly decreased, and 15 participants with the highest level of decrease were asked to take part in the second phase of data collection, in which ten were asked to write reflection reports and five were invited to participate in a focus group interview. In the reflection reports and the focus group interview, the participants expressed their ideas on how their anxiety level had reduced and on which aspects of the process helped them to become less anxious speakers of English.

Data analysis
The quantitative data obtained via the FLSAS as pre and post tests were analyzed using a statistical program. To see whether the speaking anxiety level of participants changed at the end of the research process, a paired-sample t-test was employed. For the qualitative data analysis, Creswell's (2012) content analysis scheme was followed. The data from the reflection reports and focus-group interview were transcribed and prepared for analysis. Familiar chunks were code-labelled and the codes were grouped under broader themes. In order to increase the trustworthiness of the qualitative data, the participants were asked to check the accuracy of the transcriptions (member-checking) and a faculty member with a PhD in English language teaching assisted in the coding process.

Findings of the first research question
One focus of the current study was to explore whether the students' foreign language speaking anxiety level had decreased by the end of the semester. To do this, their anxiety levels at the beginning and end of the semester were examined using a paired sampled t-test. The results are presented in Table 1. As seen in table 1, the results indicate a significant decrease in participants' foreign language speaking anxiety level. At the beginning of the semester, they experienced a high level of speaking anxiety (M=65,57) whereas at the end of the semester enriched with presentations, mini speeches, and scaffolded feedback their anxiety had reduced (M=42,28). The mean scores show that the participants' anxiety level reduced by around 30% (Mean Difference; 23,2) over the course of the semester.

Findings of the second research question
After the analysis of quantitative data showing that their speaking anxiety level was significantly reduced, the students were asked to express in-depth ideas on how and in what aspects the teaching process of presentations, mini speeches, and scaffolded feedback sessions helped them to become less anxious learners. The data derived from reflection reports and focus group interview were analyzed using content analysis. The results are presented in Table 2. Gaining self-confidence in speaking *Speaking comfortably in front of others *Using body language *Being more aware while speaking *Having eye contact 10 8 6 4 Making speaking practice *Lack of speaking practice in the past *Variety of practice opportunities *The best course to practice 8 5 4 Learning from mistakes * Pronunciation mistakes * Grammatical mistakes * Using appropriate vocabulary items 9 3 3 The findings indicate that the scaffolded feedback sessions that took place during the teaching process played an important role in the eyes of the participants. More or less similarly seen as "serving as a mirror reflecting what I looked like while I was speaking in front of others" (S11, reflection report), these feedback sessions were perceived as moments in which participants were able to reflect on their performances and analyze their mistakes. The following utterances by S2 demonstrate this: "That was not just someone telling you the mistakes, I had the chance to analyze them, to reflect upon them. To me, it was not how much I practiced through presentations, but it was the quality of those session that helped me a lot." (S2, focus-group) Another important point regarding the feedback sessions was the way the feedback was given to the participants. As one student states below, most of the participants thought that the constructive language used during the feedback session and the positive attitude of the teacher were quite influential in decreasing their anxiety level: "She took steps gradually, I mean, each and every week, she extended her explanations, in a more detailed way, as if she was trying to help us gain a new skill… She was doing that with very careful language and an extremely positive attitude, not to scare us or offend us. That was one of the most valuable things." (S4, focus-group) According to the findings, it was gaining self-confidence in speaking English that was the major issue helping participants experience lower speaking anxiety towards the end of the semester. They felt that they began to speak more comfortably in front of their peers, learnt how to make effective use of body language and make eye contact with the listeners, and became more aware of themselves while speaking in the target language. The following excerpts, particularly that from a reflection report, summarizes how the participants gained more confidence over time: "The main reason for me is because I feel much more comfortable now while speaking. I remember myself at the beginning of the semester, with shaking hands and trembling tone of voice (laughing), but now, after all these speeches, presentations, and feedback, it is very different. I am calm and comfortable, and I know exactly what I am doing while speaking." (S5, focus-group) "To be honest, even I, myself, am surprised how much I developed my speaking. At the beginning of the semester, while I was speaking, I used to feel as if my mind went blank, there were some murmurs around me, and I was not hearing what I was saying… I can't compare how I am feeling now: much more comfortable. I am quite conscious of what I am saying, using my body language and tone of voice effectively (based on what you said in the feedback session) without any fear for being in front of others." (S8, reflection report) The third factor that decreased participants' speaking anxiety was the practice opportunities they had throughout the process. In both the reflection reports and the focus group interview, the participants highlighted the fact that the teaching process within the oral communication course provided a great number of opportunities to improve their speaking skills that they had not experienced in their previous schooling. This is reflected in the reports, for example "the amount of time I spoke in this course was more than the time I spoke during all my high school years" (S14, reflection report), "I never spoke that much during my schooling life" (S8, reflection report), and "I am in the English language teaching department but that was the only course contributing to my speaking skills this semester" (S15, reflection report). The following extract from the focus group shows how important and fruitful this process was in the eyes of some learners.
"I was anxious because I did not use to speak in class, just like many of my friends; there were no opportunities for speaking. You already know the problems in our English courses at schools, so there is nothing more to say about them. When I first came here, I suffered a lot because of that. And what made the difference is precisely the opportunities we had in this course… They were planned and organized and, accordingly, helped us a lot." (S3, focus-group) Finally, some participants expressed how learning from their mistakes helped them feel less anxious. In other words, the participants thought that as they became more aware of the fact that making mistakes is a part of speaking a foreign language, they gradually became more aware of their mistakes and therefore had the chance to repair them. The words of one focus group member summarizes the general feeling: "not doing the same mistakes over and over made me feel less anxious" (S5, focus group). Some participants highlighted the recovery of their pronunciation mistakes whereas some others focused on the development of their grammar while speaking. The following reflection report extract reflects the feeling of many of the participants regarding how learning from their own mistakes contributed to decreasing anxiety over time: "The feedback sessions at the beginning of the semester and our in-class tasks were full of my grammatical and pronunciation mistakes. I noted all of them and learnt the correct ones… Our teachers' suggestions of some vocabulary items such as linkers or the ones to fill gaps while you are speaking were quite noteworthy for me. As a result, I learnt a lot of things in time and all those pieces made me a less anxious learner." (S13, reflection report) The qualitative data derived from reflection reports and the focus group were full of similar excerpts revealing how the effectiveness of feedback sessions, gaining self-confidence in speaking, having practice opportunities, and learning from their mistakes helped them become less anxious learners by the end of this process. As well as the data from the qualitative tools, participants' attitudes and non-verbal behaviours also highlighted the positive effect this process enriched by presentations, mini speeches, and scaffolded feedback sessions had on their speaking performance.

dISCUSSIoN
Foreign language speaking anxiety has long been a subject of investigation in SLA research and, over the last two decades in particular, the studies focusing on this area have mushroomed. Most of these studies are descriptive in nature, revealing the level or causes of speaking anxiety experienced by learners (Martinez Agudo, 2013;Toubot, et. al. 2017;Erdiana, et. al. 2020) whereas others focus on the relationship or connectedness of speaking anxiety to language skills (Gkonou, 2011;Liu, 2018) and other variables (Öztürk & Gürbüz, 2014;Chou, 2018). These and similar studies also make further research suggestions emphasizing that further study is needed to identify the ways that speaking anxiety in language classrooms can be reduced. Several studies have responded to these suggestions for 258 research (Ataş, 2015;Galante, 2018;Bashori et. al., 2020;Pontillas, 2020), and the current study also provides some evidence from an EFL context on how speaking anxiety can be reduced in language classrooms.
This study found that mini-speeches and presentations combined with scaffolded feedback sessions had a significant impact on reducing EFL learners' foreign language speaking anxiety within a 14-week semester. Giving a speech in front of others is naturally anxiety provoking (Ayres & Hopf, 1993). However, integrating tasks and activities such as presentations and mini-speeches into classroom teaching can gradually reduce anxiety, as shown by Coşkun (2017) and Hammad (2020). The current study demonstrates that the anxiety-reducing effect of activities employed may increase when combined with scaffolded feedback sessions. Along with other studies (Tsiplakides & Keramida, 2009;Storch, 2017), the participants in the current study also highlighted the effectiveness of feedback when integrated and conducted in a systematic way.
More specifically, this feedback-enriched period, in which students gave mini-speeches and presentations in their classes, was found to be effective in helping students gain self-confidence in speaking. They began to feel more comfortable and to use their body language more effectively, making them more confident and less anxious while speaking. Given that feeling relaxed and comfortable in class improves learners' self-confidence and encourages them to speak (Sato, 2003), the findings of the current study are noteworthy as they demonstrate the benefits of using such tasks and activities to help learners feel comfortable, gain confidence, and feel less anxious while speaking. In addition, the participants of the study felt that this process provided a lot of opportunities for them to practice speaking, thus addressing an important factor, lack of practice opportunities, which has been found to be a prominent cause of speaking anxiety (Tsiplakides & Keramida, 2009;Öztürk & Gürbüz, 2014), contributing to their reduced anxiety.
Finally, it was found that the participants learnt a lot from their mistakes during the scaffolded feedback sessions and that being aware of their mistakes made them less anxious while speaking. Studies on the factors contributing to foreign language speaking anxiety show that the fear of making mistakes is one of the three most common reasons (He, 2013;Öztürk & Gürbüz, 2014;Sadighi & Dastpak, 2017). The findings of this study show the importance of making students aware of and comfortable with their mistakes through systematic and regular feedback in order to address the fear of making them that they experience while speaking.

CoNClUSIoN
Foreign language speaking anxiety is one of the factors that potentially hinders language learners' achievement in oral performance. The SLA literature on this aspect is saturated in terms of the description and identification of the problem. The priority now must be to find strategies and solutions to help our learners cope with and overcome it. To address this, this study reports on a teaching process of mini-speeches, presentations, and scaffolded feedback that can reduce speaking anxiety in language classrooms over a period of time. The findings show that all these activities combined with systematic feedback sessions significantly reduced participants' speaking anxiety levels by the end of the semester and helped them gain more confidence due, in particular, to having more practice opportunities and learning from their mistakes. The process not only reduced their anxiety levels but also made them more comfortable and conscious speakers in their language classrooms.
This study was carried out in a specific context and conducted with a limited number of participants. However, it outlines a way for practitioners in similar contexts to create anxiety-free language classrooms. As the findings demonstrate, integrating practice opportunities such as mini-speeches, presentations, and other activities that focus on speaking practice, and giving systematic and regular feedback on students' performances will reduce their speaking anxiety, which will be reflected positively in their oral performances. However, researchers need to focus on the impact of other techniques and strategies that can potentially reduce the negative effects of speaking anxiety. Investigation of a variety of activities and their impact on students' language achievement within a longitudinal perspective will also provide a better understanding of anxiety for both researchers and practitioners.