Drama for L2 Speaking and Language Anxiety: Evidence from Brazilian EFL Learners

Abstract

Anxiety is a dimension of L2 speaking that has been heavily investigated over the past several decades, but there is a paucity of research investigating instruction aiming at lowering anxiety. While research suggests drama lowers L2 learners’ anxiety, it is unclear to what extent anxiety is affected by drama. This article reports results from a mixed methods study examining whether drama impacts foreign language anxiety (FLA). The participants were 24 Brazilian adolescents who took part in two distinct four-month EFL programmes: a drama and a non-drama programme. An adapted version of the Foreign Language Classroom Anxiety Scale (FLCAS) was used as pre- and post measures. Analyses from FLCAS indicate a significant reduction in FLA levels among learners in both groups over time, with a slightly better improvement among learners in the drama group. Further analysis provides evidence that drama can enhance comfort levels when speaking the L2. Implications for research and language teaching are discussed.

Keywords

Drama language anxiety L2 speaking theatre classroom research

Introduction

English as a Foreign Language (EFL) programmes for adolescents have been gaining increasing popularity in many countries, and Brazil is no exception. It is not uncommon for parents to enroll their children in private language schools so that they can achieve proficiency in English. Given that advanced proficiency levels in English, particularly in speaking, are typically required in Brazilian higher education institutions and the workplace, these parents aim to ensure that their children are rightly equipped with English proficiency for their future. Relative to other skills, speaking is considered to be one of the most threatening dimensions of language learning (Horwitz et al., 1986), and anxiety may cause EFL learners who are not yet proficient to perform poorly. Because anxiety is an affective variable that can prevent learners from developing speaking skills (Dörnyei, 2005; Oya et al., 2004; Philips, 1999), exploring EFL instruction that has the potential to alleviate anxiety, particularly foreign/second language anxiety (henceforth FLA), is paramount.

One classroom instructional approach with potential to reduce anxiety is drama. Drama is not new and has been used for several decades as a strategy to engage L2 learners (Dodson, 2002; Kao and O’Neill, 1998; Maley and Duff, 1982; Miccoli, 2003; Stinson and Freebody, 2006; Stinson and Winston, 2011; Via, 1976), and to develop speaking skills in the L2. Past research suggests that drama increases learners’ willingness to participate in class activities, leading to the development of overall speaking skills (Bournot-Trites et al., 2007), as well as more specific dimensions such as fluency and comprehensibility (Galante and Thomson, 2017).

Although drama instruction has been used in L2 learning as a complementary strategy for the provision of extensive speaking practice, only a limited body of research has focussed on how drama, in particular, affects overall anxiety (Coleman, 2005; Kao, 1994; Piazzoli, 2011) and specific dimensions of anxiety such as FLA (Atas, 2015; Sağlamel and Kayaoğlu, 2013). This past research provides evidence to hypothesize that FLA can be reduced after student participation in a drama programme. However, given that a comparison/control group was not used in past research, it remains unclear whether drama can in fact reduce FLA compared to other instructional approaches. This is one main goal of the study in that it investigates whether the use of drama in the L2 classroom is more effective at alleviating FLA compared to another with a communicative learner-centred approach. By using a mixed methods approach, results from two data sources are analysed for convergence purposes and they provide further information about the added values (if any) of drama. In addition, given that participants in past research included learners from Taiwan, Korea, Australia, and Turkey, an investigation of FLA among Brazilian learners is an important contribution to the field.

Anxiety

For decades, anxiety has been frequently cited as being an affective variable with a negative impact on L2 speaking (Aida, 1994; Matsuda and Gobel, 2003; Horwitz, 2001; Oya et al., 2004; Philips, 1992; Saito and Samimy, 1996). The concept of anxiety has been defined as trait, state and situation-specific (MacIntyre, 2007; Speilberger et al., 1970): trait anxiety as a person’s predisposition to become anxious, indicating a behavioural pattern; state anxiety as temporary and relating to apprehension experienced at a particular moment; and situation-specific anxiety as recurrent and particularly attributed to measures limited to a specific context. However, these three types of anxieties might not necessarily be independent dimensions, as indicated by one recent study that found that trait and situation-specific anxieties can be related (Dewaele, 2013).

Horwitz and colleagues developed a scale to measure FLA in particular: the Foreign Language Classroom Anxiety Scale (FLCAS), a scale of 33 items that relate to communication apprehension, fear of negative evaluation, and test anxiety. They found the FLCAS has high levels of internal consistency (r = .93) and reliability (r = .83), motivating several researchers to use this scale in subsequent studies. The FLCAS has become a standard measure of FLA and has made significant contributions to the field of Applied Linguistics (see research timeline in Horwitz, 2010).

Previous investigation of FLA includes diverse areas such as the relation between FLA and other types of anxiety (MacIntyre, 2007; MacIntyre and Legatto, 2010), the effects of FLA and language achievement (Horwitz, 2001; Matsuda and Gobel, 2003), and the relation between FLA and language skills such as reading and listening (Elkhafaifi, 2005; Saito et al., 1999). There have been several suggestions for strategies to alleviate FLA in the L2 classroom, including the introduction of pre-tasks to prepare learners for a main language task, and interactions in pairs or small groups (Arnold, 1999; Koch and Terrell, 1991; Young, 1991), but research on their effectiveness is scarce. In addition to these strategies, there is evidence that drama as an instructional approach also alleviates overall anxiety (Kao, 1994; Coleman, 2005; Piazzoli, 2011) and FLA (Sağlamel and Kayaoğlu, 2013; Atas, 2015), but results have been inconclusive and there is a need for further exploration.

Drama As an Instructional Approach in L2 Learning

Although drama has been used in L2 learning for many years, its implementation has been somewhat unsystematic with a vague definition of what drama is. Such lack of clarification has resulted in the terms ‘drama’ and ‘theatre’ being used interchangeably. Although complimentary, differentiating drama from theatre can inform how each impacts the L2 learning environment (Galante and Thomson, 2017).

In the L2 classroom, drama, also known as process drama, emphasizes the experience of a task, which typically involves group activities and negotiation, as well as natural and spontaneous use of the language being learned (Kao and O’Neill, 1998; Moody, 2002). It can include discussions, feelings about a given issue, explorations of identity and even an impromptu presentation. In contrast to drama, theatre is typically viewed as a product with use of scripted language. Furthermore, theatre emphasizes a final staged performance, which is often the major goal of the L2 learning experience (Moody, 2002). Similar to many tasks in L2 textbooks, scripted play presentations and role-plays require that learners take pre-established roles and use the L2 in a way limited to a particular situation.

Both drama (process) and theatre (product) impact L2 learning in distinct ways. While drama offers opportunities for learners to use prior knowledge of the L2 in a creative manner, theatre focusses on accuracy and aspects integral to the performance such as vocal projection and comprehensible speech. Both drama and theatre can be integrated (Moody, 2002). For example, learners can discuss ideas about a theme (process) or/and develop a presentation (product). Maley and Duff (1982) go beyond to suggest that the implementation of both approaches is crucial for a more comprehensive understanding of the potential of drama in L2 learning. Following these suggestions, this study considers a balance between both drama and theatre, though treated simply as drama.

Research Into the Effects of Drama on Anxiety

Research investigating whether drama reduces anxiety started to burgeon in the 1990s, after Kao (1994) pioneered an investigation of the effects of drama on overall anxiety in L2. In Kao’s study, Taiwanese EFL learners (n = 23) took part in a drama programme, and analysis of teachers’ perceptions indicate positive effects on L2 learning, including improvement in factors such as confidence in speaking English. Teachers also reported that learners who were afraid of speaking the L2 gained more confidence over time while others who had both lower self-esteem and language proficiency seemed to benefit less from the programme. In another study with Korean adolescents studying EFL (n = 60), participants reported feeling more relaxed speaking English after a five-day drama programme (Coleman, 2005). In an investigation with process drama, Piazzoli (2011) examined whether anxiety levels of adult Australian learners of Italian would lower after taking part in six drama workshops. Results suggest that anxious learners benefited from the process drama workshops, gaining more confidence in speaking the L2, particularly among those who were initially reluctant to engage in speaking tasks.

More recently, two studies narrowed the investigation by focussing on whether L2 learners’ FLA levels are reduced after taking drama lessons. Using the FLCAS in a pre- and post-test design, Atas (2015) found a significant reduction of FLA levels over time (p = 0.002, M = 3.06 and M = 2.40 in pre- and post-tests respectively) among Turkish high school EFL learners (n = 24) after taking a six-week drama EFL programme. Likewise, Sağlamel and Kayaoğlu (2013) also applied a pre- and post-test design using a modified version of the FLCAS and found that anxiety levels of Turkish EFL adult learners (n = 22), who reported having anxiety prior to the drama intervention, reduced significantly (p < 0.005, M = 2.99 and M = 2.64 in pre- and post-tests respectively) after a six-week creative drama programme.

Although all of these studies provide evidence that drama alleviates FLA, they have limitations typically expected in classroom research: small sample sizes, little control over learners’ language proficiency levels, lack of a control/comparison group, and a short drama intervention period. Particularly, it remains uncertain whether FLA is positively affected by drama when a comparison group is introduced. This study aims to specifically address some limitations by adding a comparison group, extending the period of instruction to four months, focussing specifically on FLA, controlling learners’ demographic information and proficiency levels, and obtaining two types of data sets to assess convergence.

Methodology

Research Questions

The methodology and research questions focus on the effects of drama on FLA. Following a concurrent mixed methods design with both QUAN and QUAL data sets to provide better understandings of a particular phenomenon (Creswell and Plano Clark, 2011), the research questions are as follows:

After four months of EFL instruction, do learners in a drama programme have lower FLA levels compared to learners in a non-drama programme?

What classroom practices (if any) help with comfort levels when speaking the L2?

Participants

Both learners and teachers participated in the study, although data was collected from learners only. Participants were part of the English programmes in two separate locations of a large private English institute in São Paulo, Brazil. Both were located in the metropolitan region of the city in upscale neighborhoods.

Teachers

Four teachers delivered the EFL programme: two each in the treatment and comparison groups. All teachers were female non-native speakers of English, born in São Paulo, Brazil, ranging in age from 26 to 38. Levels of EFL teaching experience varied from three to 16 years. The two teachers who delivered the drama programme had no professional training in drama or theatre but had previously received a professional development workshop on how to use drama in L2 learning. They had also taught the drama programme before the study.

Learners

Data from 24 pre-intermediate EFL learners (11 females and 13 males; ages 12-16, M = 13.8) were collected in both schools. School 1 had two drama groups (Group 1 with five males, and Group 2 with four males and four females) and School 2 had two comparison groups (Group 1 with two females and three males, and Group 2 with five females and one male). All learners were born in Brazil, came from similar socioeconomic backgrounds, and had proficiency levels equivalent to level B1 in the Common European Framework of Reference for Languages (Council of Europe, 2001).

The Two EFL Programmes

Both EFL programmes (drama and non-drama) were delivered in the same term, over a period of four months. The programmes were delivered twice a week and distributed across 37 days, with two 2-hour classes each week. Both programmes included the same textbook (Granger, 2004) and followed a communicative task-based approach. Supplementary materials developed by a team of experts in the school included a laboratory booklet that focussed on listening and pronunciation, a booklet with extra tasks, and a teacher’s guide.

Drama EFL Programme (Treatment)

The drama programme followed a curriculum I developed six years prior to the study. This was integrated into the seventh level of the EFL course to encourage positive learner affect, and speaking and pronunciation practices. The tasks were adapted from three drama education books (Booth and Lundy, 1985; Kao and O’Neill, 1998; Spolin, 1989) and allowed a focus on particular dimensions of oral communication, including the use of prior and new knowledge of lexical items, pronunciation, fluency, improvisation, prosodic features, and confidence in speaking. Beyond these tasks, the drama programme included a balance between process drama and theatre strategies such as improvised dialogues as well as rehearsals with a presentation of a short play at the end of the programme, with approximately 50 per cent of the programme dedicated to both these practices. Typically, the drama EFL programme developed in the following sequence: the teacher explained the drama project to learners and once leaners were clear on what was expected of them, they would work on the drama tasks. This usually evolved into an informal presentation of a play or scene at the end of the programme. Approximately halfway through the programme, a sample of plays and scenes were presented to learners, who made a decision about which to choose. Learners collaborated with teachers to adapt the plays or scenes of interest. Care was taken to ensure language level, content, and the number of characters and lines were appropriate for all learners. The rehearsal required that learners study the meaning of their lines rather than memorize them. Improvisation and character development were also encouraged. One drama class chose to perform two scenes (The New James Bond and The Bank), while the other class chose an episode of a sitcom (The Big Bang Theory), with both preparing a final presentation of approximately 15 minutes.

Non-Drama EFL Programme (Comparison)

The non-drama programme included the same core materials as the drama group with an oral presentation project in place of the drama component. This included several steps (i.e. in-class discussions, research, presentation rehearsal) that culminated in oral presentations in front of the class with themes suggested by the learners. While care was taken to include a project that would allow ample opportunities for speaking in class, the oral presentation project did not explicitly include learner affect. Approximately 50 per cent of the non-drama EFL programme was dedicated to the project.

Instruments

As previously noted, one standard measure to investigate FLA is Horwitz et al.’s (1986) FLCAS, a scale that has been repeatedly used in previous years (Atas, 2015; Ay, 2010; Cheng et al., 1999; Horwitz, 2001). To ensure this measure would be appropriate to the research context, participants, and research questions, a modified version of the original FLCAS was used. The modifications included: 1) suitable wording for EFL adolescent learners (e.g. the terms foreign language and language were substituted with English); 2) a scale reduced to 26 items in order to exclude seven questions related specifically to test anxiety, which was not a concern of this study; and 3) translation into Portuguese, the learners’ first language, to ensure understanding of the items. Based on a five-point Likert scale, the modified bilingual version (English-Portuguese) of the FLCAS sought responses ranging from strongly agree to strongly disagree (see modified version in the appendix). Following methodology used in previous studies (Aida, 1994; Ay, 2010; Kim, 2009; Matsuda and Gobel, 2003; Yan and Horwitz, 2008), a numerical value to each response to the 26 items was applied: from 1 strongly disagree to 5 strongly agree. The numerical values were reversed when items of the scale were worded negatively. The scale was administered to all 24 learners in both treatment (drama) and comparison (non-drama) groups at the outset of the programme, at T1 (week 1), and four months later, at T2 (week 17), the end of the programme.

Besides the modified FLCAS, all EFL learners from the drama group (n=13) participated in a semi-structured interview at the end of their programme. The purpose of the interview was to enhance understandings of classroom practices perceived by the EFL learners that alleviated FLA (if any). Two main questions were asked: ‘What were the moments during this course in which you didn’t feel comfortable speaking English and why was this the case?’, and ‘What were the moments during this course that made you feel comfortable speaking English and why was this the case?’ These questions stemmed from items in the modified FLCAS scale and required that learners recalled both FLA-inducing situations and situations that reduced FLA, but focus was given to the classroom practices that helped alleviate FLA.

Findings and Discussion

Following a mixed methods convergent design (Creswell and Plano Clark, 2011), both quantitative (bilingual FLCAS scale) and qualitative (semi-structured interviews) data sets were collected and analysed independently. Results were then compared.

The quantitative analysis included pooling answers to the 26 items of the modified FLCAS scale for each learner (n = 24). After gathering the mean ratings, IBM SPSS 24 software was used to calculate Cronbach’s alpha for both T1 and T2. A repeated measures analysis of variance (ANOVA) test and an independent samples t-test were carried out. The qualitative analysis of the semi-structured interviews first included verbatim transcriptions of the recordings of all 24 EFL learners. They were then coded and organized into the main theme that referred to comfort levels in speaking the L2. Results from both the modified FLCAS and semi-structured interviews are reported below.

FLA Levels: Drama versus Non-drama

The EFL learners’ answers to the 26 items were separated into groups (treatment and comparison) and time (T1 and T2). To assess internal consistency, Cronbach’s alpha coefficients were computed for all responses, with scores of .77 at T1 and .88 at T2, indicating an acceptable level of internal validity. A repeated measures ANOVA test with Time (two levels) as the within-subject factor and Group (two levels) as between-subject factor was carried out. A significant effect was found for Time, F(1, 22) = 14.35, p = .001, partial η² = .40, indicating overall improvement over time. No significant effect was found for Time and Group interaction, F(1,22) = .19, p = .67, partial η² = .01, suggesting that improvement was in tandem across participants. In addition, no significant difference was found between groups overall, F(1, 22) = 2.02, p = .17, partial η² = .08. Table 1 shows descriptive statistics with mean values at T1 and T2.

Table 1.

Descriptive Statistics of Mean FLCAS Scores between Groups at T1 and T2.

	Group	Mean	Std. Deviation
FLA T1	Treatment (drama), n=13	2.34	.31
	Comparison (non-drama), n=11	2.11	.36
FLA T2	Treatment (drama), n=13	2.07	.38
	Comparison (non-drama), n=11	1.89	.46

Given that the scale ranged from 1 to 5 points, with higher numbers indicating higher levels of FLA, the analyses indicate both groups experienced low FLA levels at both T1 and T2, but with a significant improvement over time.

In addition to the repeated measures ANOVA, an independent samples t-test was carried out to confirm the results. For each participant, an improvement score was calculated as a difference between FLA scores at T1 and T2. An independent t-test for the significance of the improvement mean scores was carried out. The test confirms results from the repeated measures ANOVA, with no significant difference between the treatment group (M = .26, SD = .32) and the comparison group (M = .21, SD = .29); t(22) = −.41, p = .68. Although both tests failed to establish a significant difference between groups, there is an indication that participants in the treatment group had a slightly better improvement relative to the comparison group, as shown in Table 2.

Table 2.

Descriptive Statistics of FLCAS Improvement Scores between Groups.

	Group	Mean	Std. Deviation
Improvement in FLA	Treatment, n=13	.26	.32
Scores over time	Comparison, n=11	.21	.29

These results suggest that learners from both groups experienced lower FLA levels over time, but the drama group had slightly better improvement, although not statistically significant.

Classroom Practices that Enhance Comfort Levels Speaking the L2

As previously noted, semi-structured interviews with all 13 EFL learners in the drama group were carried out to identify classroom practices (if any) that enhance comfort levels speaking the L2.

The only classroom practice reported by most participants as contributing to comfort levels speaking English was drama. No other classroom practice was mentioned. One interesting result refers to playing different roles and that ‘being in character’ has helped participant 5 feel more comfortable:

The presentation is very fun and…I think it’s more easy to talk and…interact with the others…I will be the James Bond girl …Linda. I feel more comfortable… speaking English because it’s not me… it’s Linda (Participant 5).

It seems that embracing another identity, that of the character played, helped this learner feel good when speaking English. Similarly, learner 6 also stated that drama was the most helpful classroom practice:

In this course I feel more comfortable and talk uh…in the front of the people. I think the…drama project makes us more comfortable to talk in front of people… It’s a kind of theatre…and conversation that I think it’s the activity to help us (Participant 6).

This account shows that this participant does not have fear of speaking the L2 in front of an audience because of drama. Another participant stated drama helped prepare speech, which may have contributed to comfort levels:

I felt comfortable…specifically when I prepare what I’m going to say…the presentation…the drama project…we’re going to present a scene of…uh…a bank…a bank that this bank was robbed. Because I prepare what I’m going to say…I felt comfortable (Participant 7).

Considering the interview questions did not explicitly state FLA or drama per se, the fact that drama was the most frequently reported classroom practice that contributed to comfort levels speaking the L2 was an important result. Among participants who reported being shy, drama was also mentioned as affording benefits that other classroom practice might not. For example:

I think the play…the play help you to…be a little bit less shy. I think speak is a good way to…don’t be shy…the play is very good for it…because we need to act and speak English loud…and act too…we make more body language (Participant 2).

When we…we begin to rehearse…uh…it was very funny… when I know my lines it’s easy to me to talk because I know what I’m going to say but uh…when I rehearsing…it’s very funny so I don’t care if I will…uh…make mistakes (Participant 3).

It is noteworthy that among participants who reported being shy, drama was also the classroom practice reported to overcome their shyness when speaking the L2. These results are important as they suggest drama can offer advantages unique to this type of instruction, including for shy learners. Other reported themes were not necessarily related to classroom practice and included teachers, fellow classmates and the class environment.

Discussion

Analyses from the repeated measures ANOVA and t-tests indicate that learners in both treatment and comparison groups experienced significant improvements in FLA levels over time, with only a slightly higher reduction among learners in the drama group. These results do not seem to confirm results from previous studies using a similar modified FLCAS. If a comparison group had not been used in this study, the results from participants in the treatment group would have accorded well with results in both Atas’s (2015) and Sağlamel and Kayaoğlu’s (2013) studies in the sense that learners’ FLA levels were significantly reduced after taking part in a drama programme. However, it is not possible to conclude that this reduction was due to the drama instruction. Relative to the comparison group, FLA levels among participants in the drama group did not have a statistically significant reduction, indicating other variables might have contributed to this result. For example, the type of instructional approach used in previous studies is unclear and may not be comparable to the one used in this study. In addition, while past studies used a six-week intervention, this study used a four-month intervention and it could be that drama may only alleviate FLA in the short term.

As the focus of this study was not on the effects of drama on anxious learners, the sample represented included learners in two classrooms who might or might not be anxious. While FLA mean scores at T1 were 2.34 in this study, participants in previous studies (Atas, 2015) had scores of 3.06 (Atas, 2015) and 2.99 (Sağlamel and Kayaoğlu, 2013). It may be that drama offers more benefits for anxious learners compared to less anxious learners. One important result of this study is that it shows that drama enhanced levels of comfort when speaking the L2 especially among learners who reported being shy. It is important to note that shyness is a trait and was not considered the same as FLA.

While the results of this study cannot state that drama reduced FLA levels significantly compared to a comparison group, it can hint at benefits related to FLA such as comfort levels in speaking the L2. Three other studies that investigated overall anxiety when speaking the L2 (Coleman, 2005; Kao, 1994; Piazzoli, 2011) also support this study’s conclusion that drama affords benefits to oral communication. It is worth noting that these three studies did not focus on a particular type of anxiety. It could be that trait, state and situation-specific anxieties are affected differently after drama instruction or that drama benefits learners who have high FLA levels only. Qualitative analysis indicates that drama as an instructional approach helps increase comfort levels when speaking the L2, particularly among shy students. The shy learners reported drama can enhance comfort levels speaking the L2, allowing them to ‘be more comfortable…relaxed’ (Participant 4), and ‘comfortable…specifically when I prepare what I’m going to say (Participant 7). This also confirms results from previous studies that suggest that drama can help learners feel more relaxed (Coleman, 2005), confident (Kao, 1994), and less worried (Piazzoli, 2011) speaking the L2, at least in classroom interactions. It seems that drama can be beneficial for learners from different linguistic backgrounds (e.g. Portuguese, Turkish, Korean) and in different settings (e.g. EFL, ESL).

Besides enhancing comfort levels when speaking the L2, results of this study also suggest that drama brought added values. One particular advantage relates to communication skills such as accuracy, eye contact, and body language, similar to results in previous research (Coleman, 2005; Kao, 1994; Piazzoli, 2011). Another unique advantage is associated with identity in L2 learning, although this was not a focus of the study. One learner reported feeling ‘more comfortable… speaking English because it’s not me… it’s Linda’. Similar to previous research (Piazzoli, 2011), drama enhances comfort levels because it puts learners in a role; thus, any anxiety related to speaking the L2 can be assigned to the character rather than the learner. Furthermore, the drama programme in this study provided several opportunities for learners to develop the L2 in a friendly environment, possibly increasing affect. Likewise, a previous study (Coleman, 2005) also indicated drama can facilitate managing anxiety and apprehension, allowing for a more relaxed environment in which to use the L2.

Conclusion

EFL learners around the world strive to achieve high proficiency levels, particularly in speaking skills. Because anxiety can be a debilitating factor when learners are asked to perform in the L2, the provision of instructional approaches that help alleviate FLA is needed. Unlike other studies (Atas, 2015; Sağlamel and Kayaoğlu, 2013), this study cannot suggest that drama reduces FLA, particularly because a comparison group was introduced. However, similar to previous studies (Coleman, 2005; Kao, 1994; Piazzoli, 2011) this study suggests that drama can make positive contributions to L2 learning, particularly in relation to comfort levels speaking the L2.

Like any study, this study has limitations. Although previous research on the effects of drama on L2 learning has revealed positive findings, several limitations have limited the generalizability of the results particularly because of methodological issues (Stinson and Winston, 2011). A novelty of this study was the introduction of a comparison group as an attempt to provide more reliable results. However, some limitations should be noted: the small sample size may have limited the power of the statistical analyses. While variables were controlled (e.g. all participants belonged to similar linguistic, cultural and socio-economic backgrounds, and were similar in age), there was no random selection of participants who had trait anxiety or FLA. Nor was their assignment to treatment or comparison groups random.

Overall, despite its limitations, this study makes important contributions to the literature on the effects of drama on FLA. It expands the scope of previous research on drama and L2 learning mainly by introducing a comparison group. Drama as an L2 instructional approach can increase comfort levels when using the L2, even among learners who are shy, but may not alleviate FLA levels, especially among learners who do not report being anxious. That may seem obvious given learners who do not show high FLA levels at T1 have little room for improvement at T2.

Language instructors can explore the effects of drama in their own context by implementing strategies from both process and theatre and get anecdotal information from students as to the potential benefits of drama. Given research with learners from various linguistic backgrounds and different contexts reported benefits of drama, language classrooms across the globe will help learners with L2 speaking skills. Of course, drama cannot be seen as a method and adaptations will be necessary to best suit language learners.

Future research on drama and FLA will benefit from studies with a clear definition for drama, long-term drama instruction, control/comparison groups, and a large sample of participants with different levels of FLA. Studies using the FLCAS scale may also benefit from adding qualitative data gathering so that understanding of the effects of drama can be further explored. Furthermore, future studies can expand our understanding of the potential of drama in L2 learning. While more research is needed to validate the findings reported in this article, the knowledge gained here can be applied and examined in other L2 contexts.