Anxiety and enjoyment among young teenagers learning English as a foreign language outdoors: a mixed-methods study

ABSTRACT

Purpose

The aim of this study is to explore how outdoor education can influence foreign language anxiety (FLA) and foreign language enjoyment (FLE) when young teenagers are learning a foreign language (FL). We applied a dynamic perspective to investigate if using English in an outdoor context would reduce FLA and boost FLE, leading to a stronger willingness to communicate (WTC) in the FL.

Design/methodology/approach

The design was a pre-post intervention study with an experimental group (EG) and a control group (CG). Both groups engaged in fluency-oriented activities during a six-week intervention. The EG conducted the activities outdoors whereas the CG conducted them in their classrooms. A mixed-methods approach was applied collecting data through student questionnaires and group interviews with both students and their English teachers. Observations and recordings of participants’ speech were also conducted.

Findings

Statistical analyses indicated a small effect of the intervention for both FLA and FLE but the difference in gain scores between groups was found to be non-significant. The qualitative data revealed a strong appreciation of the outdoor experience in English.

Originality/value

Most previous studies have investigated emotions in FL learning inside the classroom. The current study investigates emotions in FL learning among young teenagers in an outdoor environment.

KEYWORDS:

• Second and foreign language learning
• foreign language anxiety
• foreign language enjoyment
• willingness to communicate
• outdoor education
• affordances

Introduction

Foreign Language Anxiety (FLA), defined as ‘the worry and negative emotional reaction aroused when learning or using a second language’ (MacIntyre 1999, 27) and Foreign Language Enjoyment (FLE), defined as a complex positive emotion resulting from a combination of challenge and perceived ability (Dewaele and MacIntyre 2014), have been described metaphorically as the cold shower and the fire that affect the second or foreign language (FL) learning process (Dewaele 2015). Meta-analyses have shown that FLA is moderately negatively correlated with FL achievement and Willingness to Communicate (WTC) (e.g. Botes, Dewaele, and Greiff 2020), defined as ‘a readiness to enter into discourse at a particular time with a specific person or persons, using a L2’ (MacIntyre et al. 1998), while FLE is moderately positively correlated with both FL achievement and WTC (Botes, Dewaele, and Greiff 2021). While research suggests that teachers can indeed boost their students’ enjoyment, they seem to have much less success in lowering their anxiety (Dewaele and Dewaele 2020; Dewaele et al. 2018). One possible reason for this is that FLA is much more linked to learners’ personality, in particular neuroticism or trait anxiety (Dewaele and MacIntyre 2019) which cannot be changed by teachers, while FLE is much more the result of a positive emotional atmosphere where the teacher can play a key part. Teachers can help anxious students by creating non-threatening environments where students are encouraged to experiment with the FL without fear of ridicule (Dewaele and Dewaele 2020). A realistic aim for teachers would thus be to help learners deal with their anxiety in a constructive and empathic way. There are studies that show that specific interventions, such as reminiscing about FL achievements, can temporarily lower FLA (Jin, Dewaele, and MacIntyre 2021).

Another negative emotion that has attracted growing attention is FL boredom (FLB), a negative activity-related achievement emotion with a low arousal (Li, Dewaele, and Hu 2021). Research suggests there are links between FLA, FLE and FLB (Dewaele and Li 2021; Dewaele and MacIntyre 2014; Pawlak, Kruk, and Zawodniak 2022) and that FLB is also linked to WTC (Li, Dewaele, and Hu 2021).

What previous studies on emotions have in common is the setting in which the FL was being acquired, namely classrooms where learners sit and are physically quite passive. The question is whether a change in setting might affect learners’ emotions. Could FL learning in an outdoor environment make a difference? Outdoor learning and teaching is referred to as outdoor education, defined as ‘ … an approach to teaching in which parts of the school day are spent in the local environment’ (Jordet 1998, 24). Traditionally, outdoor education takes place in a schools’ nearby surroundings so that learners can interact with the local environment and learn from their experiences (e.g. Beames and Ross 2010). Applying this approach in our study, students would actively engage in meaningful English communication activities with their classmates in local outdoor environments.

Change of learning environment can provide new affordances. According to Gibson (1986), an affordance is the intuitive, visual perception of our surroundings, e.g. a tree might invite a child to climb it. The affordance theory was developed by Heft (1988), and later Kyttä (2002, 2004) developed a framework for studying children's outdoor environments. In the FL literature, affordances or ‘action possibilities’, have been thoroughly discussed (e.g. Brown and Lee 2015; van Lier 2000). Because an outdoor environment represents an informal, open space, learners do not have to follow the norms and expectations of the classroom (Harris 2017) and will also be offered other affordances.

The aim of this study is to explore how outdoor education can influence young teenagers’ FLA and FLE when they are learning a FL.

Literature review

Studies on foreign language anxiety and enjoyment

Research on language anxiety has evolved through the 1970s and Scovel’s (1978) review of language anxiety through to the Specialised Approach which stressed how language anxiety was experienced. One pioneering study was Horwitz, Horwitz, and Cope (1986) which introduced the Foreign Language Classroom Anxiety Scale (FLCAS) and language anxiety was seen as situation-specific (Gardner 1985). The latest approach is the Dynamic Approach which studies the complex interactions between language anxiety and both learner-internal and learner-external variables. MacIntyre (2017) concluded that language anxiety is always debilitating and has no facilitative effects. MacIntyre (2007) found that FLA had a negative effect on learners’ WTC, a finding confirmed in recent research (e.g. Dewaele and Dewaele 2018).

Interest in positive emotions emerged within the context of the Dynamic Approach. Positive and negative emotions were found to be linked in highly dynamic, complex networks of learner-internal and learner-external variables which were both influencing FL performance and simultaneously being affected by it. Dewaele and MacIntyre (2014) drew on positive psychology, more specifically on the broaden-and-build theory (Fredrickson 2001, 2003, 2004). Fredrickson argued that positive emotions broaden learners’ momentary thought-action repertoires which can increase their personal resources whereas negative emotions have the opposite effect, with learners feeling under threat and absorbing less input from the environment. Dewaele and MacIntyre (2014) introduced the notion of Foreign Language Enjoyment (FLE) and proposed a FLE scale of 21 items. They combined it with eight items extracted from the FLCAS (Horwitz, Horwitz, and Cope 1986) to explore the relationship between FLE and FLCA. The study showed that FLE levels were significantly higher than the FLCA levels and that a moderate negative relationship existed between FLE and FLCA. Both emotions were linked to age, educational level, the number of languages known, the number of FL being studied, perception of FL proficiency level in the classroom and the general level of the FL. The number and range of independent variables has been expanded in more recent studies (e.g. Dewaele and Dewaele 2018, 2020). Dewaele and Dewaele (2018) investigated learner-internal and learner-external predictors of WTC in the FL classroom. Results showed that FL teachers who manage to create a positive emotional classroom environment can boost learners’ FLE levels and hence increase their WTC. Studies have also shown that the teacher has a much stronger effect on FLE than on FLCA (Dewaele and Dewaele 2020; Dewaele and MacIntyre 2019; Dewaele et al. 2018). Whereas FLCA could be seen as the cold shower that leaves learners shivering in a corner, FLE has been compared to the fuel that allows students to glow and grow (Dewaele 2015).

Although emotions permeate learners’ attitudes and motivation, they remained largely out of view for a long time. The first study to explicitly link attitudes and motivation and learner emotions showed strong correlations between attitudes and motivation to learn a FL and the positive and negative emotions students had experienced in the previous week (MacIntyre et al. 2020). FLE was also found to support flagging motivation over time (Dewaele, Saito, and Halimi 2022).

Studies on outdoor education

Outdoor education has many positive effects on learning including increased academic achievement (e.g. Fägerstam and Blom 2013), motivation (e.g. Bølling et al. 2018; Fiskum and Jacobsen 2012b, 2015) and engagement (Beames and Ross 2010). In addition, it reduces students’ stress levels (Dettweiler et al. 2017), which might be important in reducing FLA. Outdoor education may also change the students’ attitudes towards learning (e.g. Fiskum and Jacobsen 2015; Myhre and Fiskum 2021). Moreover, several studies indicate more communication between learners outdoors compared to the classroom (Fiskum and Jacobsen 2012a, 2012b; Fägerstam and Blom 2013; Mygind 2014; Myhre and Fiskum 2021). In terms of FL learning, Myhre and Fiskum (2021) found that young teenagers in Norway experienced learning spoken English outdoors as less stressful than classroom learning due to less focus on accuracy and an avoidance of constant direct observation by the teacher. This boosted their confidence and reduced their feeling of nervousness which in turn boosted their WTC in English. They found that an outdoor environment's new and varied affordances increased teenagers’ WTC. It is therefore likely that a change of learning environment could affect students’ levels of FLA and FLE.

The article addresses the following research questions:

1. Does learning in an outdoor environment contribute to:

   1. A reduction in FLA in young teenagers learning a FL?

   2. An increase in FLE in young teenagers learning a FL?

2. How might an outdoor environment affect FLA and FLE in young teenagers learning a FL?

We hypothesised that learning English in an outdoor environment would reduce young teenagers’ FLA and boost their FLE.

Method

A mixed methods design was chosen to answer the research questions. A concurrent embedded approach was used. In other words, quantitative and qualitative data collections happened simultaneously (Creswell 2009). Both data collections were planned for the same study and planning was finished prior to the data collection. Hence the two methods of measurement did not influence each other's design. The data analyses were done separately with two researchers conducting the quantitative analysis and two researchers conducting the qualitative analysis.

Research design

The current study is part of a larger pre-post pedagogic intervention study with a control group. Participants completed questionnaires before and after an intervention. Observations and recordings of students’ speech were conducted both for the experimental group (EG) and the control group (CG) to assess the intervention's effect.¹ Towards the end of the intervention, group interviews were organised with students and their English teachers (separately). After the intervention, parents completed a questionnaire about socioeconomic status. A second interview with the English teachers took place two months later and a thematic analysis of the interview data was carried out.

The students originally were in five different classes (randomly divided) with approximately 25 in each. The EG consisted of three classes, the CG of two. Each of the classes in the EG had a different English teacher whereas the classes in the CG had the same teacher.

Intervention

The intervention lasted six continuous weeks in autumn 2018 and focused on fluency-oriented activities, engaging the students in real and meaningful communication with classmates. Both groups were as far as possible offered the same English language activities, but the EG conducted them in an outdoor environment whereas the CG did so in their classrooms. Accordingly, the main difference was the setting in which the instruction took place. The EG had one double session per week, approximately 80 minutes. Because of timetabling issues, the CG had two separate sessions per week, each of about 45 minutes. The EG had their sessions simultaneously one morning per week, but the activities took place at different locations. Most places were not far from the school's premises, such as nearby parks. As the students would be walking 10-15 minutes to their location and back, they also participated in spoken language activities during their walks (‘walk and talk’).

The activities were planned by the article's first author, but the teaching was carried out by the English teachers. Different activities with a task-based approach were planned for each week, six lesson plans in total, and they were carried out in pairs or in small groups. Activities included various descriptions being the participants’ bedroom, nouns on flashcards and the surroundings (EG), as well as conversations about general topics and acting as guides.

Participants in questionnaires

A total of 106 students completed the questionnaires (N = 106, 58 females, 48 males). They were all in Year nine (13-14 year olds) at one secondary school in a town in rural Norway. The vast majority had Norwegian as their first language, a few were refugees and had different first languages. The students with Norwegian as their first language had studied English for seven years at the time of the data collection.

Sociobiographical information is presented in Tables 1 and 2.

Table 1. Descriptive data for EG and CG in terms of gender, mean age and mean grade.

Experimental group (67)					Control group (39)
Gender		Mean age (SD)	Mean grade S/W (SD)		Gender		Mean age (SD)	Mean grade S/W (SD)
Female 37	Male 30	14.29 years (0.29)	S 3.88 (0.78)	W 3.83 (0.94)	Female 21	Male 18	14.28 years (0.36)	S 3.95 (0.90)	W 4.08 (1.02)

S = spoken English, W = written English, SD = standard deviation.

Note: Performance in spoken and written English is measured on a scale from 2 to 6 where 6 reflects the best performance. For two students we did not have access to the grades.

Table 2. Descriptive data for EG and CG in terms of socioeconomic status provided by parents.

	Experimental group	Control group
Educational background
Response rate	36/67	22/39
College (1–2 years)		1
College (3 years)	1	2
Certificate of apprenticeship	6	5
University college or university (4 years or less)	9	5
University college or university (4 years or more)	20	9

Participants in recordings of speech

A total of n = 14 students volunteered to be recorded, n = 8 from EG, n = 6 from CG.

Participants in interviews

Student interviews: n = 32 students from both EG (n = 20) and CG (n = 12) were interviewed pre and post (17 boys, 15 girls).

Teacher interviews: n = 4 teachers were interviewed in the first interview (all female with Norwegian as their first language). Three had been teaching outdoors (EG), the fourth indoors (CG). Only two of the four teachers (n = 2 from EG) could participate in the second interview.

Description of questionnaires

Sociobiographical data included gender, age and socioeconomic status, namely parents’ educational background (see Tables 1 and 2). The questionnaire included seven items from Horwitz et al.'s (1986) FLCAS, a validated scale to measure FLA. FLE was measured using Dewaele and MacIntyre’s (2014) scale consisting of 21 items. FLA² and FLE items were adapted³ to the participants’ age (where necessary) and to the English subject in Norwegian schools, and translated into Norwegian by the article's first author⁴ to avoid misunderstandings. Internal consistency was very good for both scales: Cronbach's alpha = 0.91 (pre) and 0.89 (post) for FLA, and 0.90 (pre) and 0.92 (post) for FLE. The questionnaires were answered on paper in class. A four-point Likert scale with no neutral response was used, which is recommended in the literature for the aim and age group (Omrani et al. 2019). The questionnaire was tested with a few teenagers in the same age group.

Description of observations

Unstructured observations were conducted in both groups over a period of several weeks to observe if the students understood and participated in the activities and used English while conducting them.⁵ Each observation lasted for approximately 45 (CG) and 80 minutes (EG). In the CG, two observations were conducted, one in each of the two classes. In the EG, there were three observations. In this group, two different teaching programmes were observed in one class and one teaching programme was observed in which all three classes participated simultaneously at a historic arena. Due to the last observation in this group taking place at a large venue it was challenging to observe all students. Notes were taken during the observations but no coding took place.

Description of recordings of speech

Participants were equipped with an audio recorder and recordings took place three times with a two-week interval (weeks 2-4-6). Each recording lasted approximately 45 (CG) and 80 minutes (EG) (the whole sessions were recorded). We listened to a randomised sample of recordings in both the CG and the EG from the middle of the intervention (week 4) and recorded the time the groups of students spent communicating in English during the activities. This time was calculated into an overall percentage for each group (CG and EG).⁶

Description of interviews

Ten semi-structured group interviews with students in both groups (EG and CG) took place in October 2018, each having three or four participants and lasted approximately 30 minutes. All interviews were conducted at the students' school, in a room close to their classroom. The teachers put the students into groups making sure all would feel safe and not be afraid to speak. The interviews were conducted in Norwegian and hence contributions were not dependent on their English language skills.

The interview guide consisted of questions of a more general nature including the importance of learning English, the frequency of English use in previous years and the emotions it elicited. The interview guide had been tested with a few teenagers in the same age group.

The two semi-structured group interviews with the English teachers only were conducted in mid-November 2018 and at the end of January 2019. The first focused on how successful the activities had been in terms of student participation and engagement including comparing engagement during these activities to the more traditional teaching approaches the students were used to. In the second interview, they were asked to reflect on the activities, if they had observed any changes in students’ willingness to communicate in English and their emotions, and also if potential changes could be due to gender differences.

Data analysis – questionnaires⁷

A total of 130 students were included in the study, where 24 students had missing data in pre and post-tests in FLA and FLE (18.46% missing in pre and post-test for EG and CG for FLA of seven and six, respectively, and for FLE of eight and seven, respectively⁸). A complete case analysis was performed where these 24 students were deleted, giving a final dataset of N = 106. For students missing the entire pre or post-tests (13 of 24 students), we assumed a possible missing completely at random (MCAR) as these were due to causes such as illness. For the remaining students missing one or several values in either of the tests due to not ticking off answers, we assumed missing at random (MAR). Although this created a reduced dataset, we argue that deletion of students with missing values did not create any systematic biases.

Because the students answered the questionnaires manually, they were able to tick off more than one response. We gave those instances an average score (e.g. if ticked 1 and 2 = 1.5). These instances were relatively few and the vast majority of the instances were next to one another (e.g. 1 and 2, not 1 and 4). This may indicate that the students who ticked off more than one response were likely to be unsure about what response to give and hence we found it appropriate to give an estimated score. Two positively formulated FLA items (6 and 7) were reverse-coded, but none of the FLE items needed reverse coding. We conducted separate analyses for FLA and FLE scores. Descriptive data in pre and post-tests was calculated for the two groups by central tendency reported as median (Mdn) and mean (M), variability as range (Min-Max), interquartile range (IQR; 25-75th percentiles) and standard deviation (SD). A gain score approach was found to be appropriate (Gliner, Morgan, and Harmon 2003) to analyse changes pre and post, subtracting pre-test scores from post-test scores within each group (EG and CG). Due to the non-normal distribution in gain scores for the EG, a Mann-Whitney U test was used to test whether there was a significant difference between these two groups (i.e. whether they were likely to derive from the same population). The r proposed by Cohen (1988) was used as measure of effect size. The effect size r was calculated as $r={\displaystyle \frac{z}{\sqrt{N}}}$ where z is the absolute (positive) standardised test statistic and N is the number of pairs (before-after), here N = 106 (Fritz, Morris, and Richler 2012). Analysis was conducted using SPSS version 27.

Data analysis – interviews

The interviews were transcribed and imported into NVivo. Two researchers used the constant comparative method (Glaser and Strauss 1967) to jointly identify various codes and categories in the data.⁹

During the open coding process (Strauss and Corbin 1998), all statements relating to the themes language anxiety and language enjoyment were analysed separately. In the axial and selective coding processes, the two themes were analysed as a whole. Open coding revealed 67 categories for FLA and 110 for FLE, whereas the axial coding left us with a total of 32 categories. In the selective coding process, the axial categories were given different colours to group them thematically. The analysis revealed five core categories (Glaser 2001).

Ethics

The study met the research ethics guidelines set by the Norwegian National Research Ethics Committees (2014) and was approved by the Norwegian Social Science Data Services (NSD).

The students’ parents signed a consent form and the students had the option of not participating despite their parents’ consent. All data was handled anonymously.

Results

Quantitative data

To reflect the research questions, we split the results for FLA and FLE. Descriptive data for FLA and FLE in pre and post-tests for the two groups are given in Tables 3 and 4.

Table 3. Descriptive data for the EG and CG in pre-test and post-test measuring levels of FLA.

FLA	Experimental group				Control group
	Min-Max	Mdn	IQR	M (SD)	Min-Max	Mdn	IQR	M (SD)
Pre	1.00–4.00	2.29	1.57–3.14	2.37 (0.85)	1.00–3.43	2.43	1.86–3.00	2.40 (0.69)
Post	1.00–4.00	2.36	1.86–3.14	2.50 (0.81)	1.14–3.71	2.57	2.00–3.00	2.45 (0.70)

Min = minimum, Max = maximum, Mdn = median, IQR = interquartile range, M = mean, SD = standard deviation

1 meant high FLA and 4 meant low FLA.

Table 4. Descriptive data for the EG and CG in pre-test and post-test measuring levels of FLE.

FLE	Experimental group				Control group
	Min-Max	Mdn	IQR	M (SD)	Min-Max	Mdn	IQR	M (SD)
Pre	1.33–3.10	2.07	1.86–2.33	2.09 (0.38)	1.19–4.00	1.79	1.62–2.07	1.90 (0.54)
Post	1.29–3.14	2.04	1.71–2.36	2.03 (0.41)	1.14–4.00	1.81	1.52–2.00	1.90 (0.59)

Min = minimum, Max = maximum, Mdn = median, IQR = interquartile range, M = mean, SD = standard deviation

1 meant high FLE and 4 meant low FLE.

Foreign language anxiety

Overall, Table 3 shows that collectively the EG and the CG reported to have moderate FLA levels in both pre (EG M = 2.37, CG M = 2.40) and post (EG M = 2.50, CG M = 2.45) tests.

A Mann-Whitney U test did not indicate any statistically significant difference in gain scores between the EG and the CG group (U = 1114.5, p = 0.21, r = 0.12). The effect size is extremely small (Coolican 2009, 395).

Foreign language enjoyment

Overall, Table 4 shows that collectively the EG and the CG reported to have moderate FLE levels in both pre (EG M = 2.09, CG M = 1.90) and post (EG M = 2.03, CG M = 1.90) tests. Levels increased very slightly in the EG from pre to post, whereas levels remained more or less constant for the CG. After the intervention, the EG reported slightly lower FLE than the CG, and there was a very slight increase in enjoyment for EG, whereas there was no change for the CG.

A Mann-Whitney U test indicated no statistically significant difference in gain scores between the EG and the CG group (U = 1017.5, p = 0.06, r = 0.18). The effect size of 0.18 is considered a small effect (Coolican 2009, 395).

Qualitative data

We present the results of the interviews first, then the results of the observations and recordings of speech. When analysing the student and teacher interviews the following five themes were identified across the groups and hence student and teacher interviews are merged.

Performance pressure vs. boredom

Students in both groups stated that they are anxious when asked to talk audibly in the classroom when everybody listens and feel performance pressure. They explicitly said that they find the classroom's many listeners to be anxiety-provoking. They were especially fearful of making grammatical mistakes or mispronouncing words and were afraid of classmates laughing: ‘Other students look at you … everybody turns around’ (Interview 4, EG). However, the classroom was not experienced as anxiety-provoking by all students: ‘I don't experience it as being frightening, but it makes me a bit uncomfortable speaking in front of so many people in class’ (Int. 9, EG). Students in the EG said that they did not compare their performances to others when learning outdoors, as they would easily do in the classroom.

Many said they experience the traditional English lessons as boring and not very interesting. The teacher does most of the talking and listening leads to boredom and loss of concentration: ‘ … one starts to think about something else’ (Int. 8, EG). Some also felt like giving up, and did not feel like putting in much effort.

Change of context

An indoor classroom has many norms and expectations, including students sitting quietly at their desks and paying attention. The outdoor classroom, on the other hand, represents a contrast in that norms and expectations are less rigid.

Both teachers and students in the EG claimed that the outdoors contributes to increased confidence in speaking English due to more space and greater distance between the students: ‘The big space outdoors makes them more sheltered compared to the classroom where we’re close to one another and it's easy to listen to each other. They become more courageous speakers outdoors I think’ (Teacher int. 1). Students in the EG claimed that the outdoors gives opportunities to practice spoken language to a greater extent because in the classroom ‘ … we usually just sit at our desks and pay attention, there is not much speaking’ (Int. 10) and ‘ … when we were outdoors many would speak at the same time, whereas in the classroom we have to speak one at a time’ (Int. 10).

Real life experiences and engaging ways of learning

The teachers explained that the activities had been more similar to what the students would experience in real life and more engaging than traditional teaching. Students stressed that varied teaching approaches in general are important for them in order to pay attention and explained they find activities in which they need to be physically active to be particularly engaging. Students in the EG expressed their views in the following ways: ‘I become more motivated when I’m outdoors because it's more fun than sitting at my desk’ (Int. 3), ‘I lose focus when I have to sit still’ (Int. 4), and ‘I can't sit still that long’ (Int. 4). The teaching also engaged the students because it was experienced as more enjoyable than their traditional teaching: ‘I think it has been much more fun. I have paid more attention to the teaching because one needs to remember what to say’ (Int. 6, CG).

Feeling comfortable and safe learning environment

Students said that a relaxed learning environment where they do not need to fear losing face reduces FLA and increases FLE. Students in the EG claimed they were not afraid of saying something grammatically incorrect because: ‘ … the ones who listen to or look at you, may not focus that much on you when you’re outdoors’ (Int. 3). One of the outdoor teachers stated: ‘The big space outdoors makes them more sheltered compared to the classroom … They become more courageous speakers outdoors I think’ (Teacher int. 1). One student in the CG said the following about speaking audibly in the classroom versus outdoors: ‘There are many listeners, including the teacher. It becomes totally quiet when everybody's waiting for you to start speaking, but when everybody speaks at the same time it feels more comfortable’ (Int. 8). Students in both groups emphasised that they feel safer when speaking in small groups, especially with peers that they know well.

Enjoyment and stress relief

Students in the EG explained they found learning outdoors to be more enjoyable than traditional classroom learning. Talking about the activities one said: ‘They were more fun and I felt I learned a lot more’ (Int. 10). Several experienced learning outdoors as different and that it offered other possibilities for cooperation among peers than classroom learning: ‘ … when we started learning outdoors, I thought the lessons were more fun than being in the classroom doing written tasks and such’ (Int. 3) and ‘In the classroom you’re only sitting by yourself’ (Int. 4).

Change of learning environment seemed to reduce performance pressure among students: ‘Because in the classroom you must know exactly what to say etc., but when being outside the classroom one doesn't think about being assessed and such ‘ (Int. 1). Other students being outdoors said: ‘In the classroom one feels that one needs to perform, while being outdoors one forgets about this’ (Int. 1) and ‘Yes, because the focus is on putting up your hand and such and then the teacher will pay attention to you. When we were outdoors this wasn't the case’ (Int. 4). Also, the teachers noticed that some students that hardly used English in the classroom, did so outdoors.

Observations

Overall, the students were observed to understand the tasks and to be quite talkative during the activities.

EG: When visiting the historic place and acting as guides some were more active than others. As this was a task that required some preparation, the prepared ones were more successful in using English. Observations of the walk and talk activities showed some pairs were more easily distracted than others and used English less.

CG: Students were more easily distracted by their classmates and a fair amount of Norwegian was used. However, the observations detected a very friendly student-teacher relationship and a relaxed atmosphere.

Recordings of speech

Overall, the recorded material showed that the students understood the activities. In the EG the students had dialogues in English 86.8% of the time of the activity, while in the CG this was 65.6%.

Discussion

The aim of this study was to explore how learning a FL outdoors can influence FLA and FLE levels in young teenagers. Findings from the quantitative analysis suggested changes in gain scores between the EG and the CG to be borderline positive for FLE but not for FLA. A clearer picture emerged from the qualitative findings that suggested learning English as a FL in an outdoor context is experienced as positive and enjoyable for many students in this age group. The fact that the present study found little change in FLA and FLE is supported by recent research. Studies have demonstrated that levels of FLA remain quite stable (Dewaele and Dewaele 2020; Dewaele, Saito, and Halimi 2022; Dewaele et al. 2018). One reason for this is that FLA has been found to be linked to personality traits (Dewaele and MacIntyre 2019), and these cannot be easily altered by teachers unless by targeted interventions (Jin, Dewaele, and MacIntyre 2021). FLE, on the other hand, is much more malleable and can be shaped by teachers, though it has also been shown to be quite stable over time (Dewaele, Saito, and Halimi 2022). The findings will be further discussed using headlines that capture both the quantitative and qualitative results.

Feeling safe and less anxious

The outdoor spaces were regarded as non-threatening environments where the students felt safe and secure and hence not afraid to communicate in English. They experienced less focus on their performance as only peers in their group were listening. This finds support in research which has found that teachers can help anxious students by creating non-threatening environments where students are encouraged to experiment with the FL without fear of ridicule or losing face (e.g. Myhre and Fiskum 2021). Outdoor learning was also experienced as less stressful, confirming findings in Dettweiler et al. (2017) and Myhre and Fiskum (2021) due to less focus on accuracy and an avoidance of constant direct observation by the teacher. Their confidence increased and their feeling of nervousness was reduced, which in turn boosted their WTC in English.

Eagerness to learn

Many students reported they found the learning fun and enjoyable and it boosted their WTC. This confirms findings in previous studies that change of learning environment may give learners a more positive attitude towards learning (e.g. Fiskum and Jacobsen 2015; Myhre and Fiskum 2021). Reasons for this were different activities which also felt meaningful. Increased WTC in the FL when learners experience collective enjoyment is supported by recent research (Dewaele and Li 2021; Li et al. 2022).

Positive learning arena

The outdoor environments were considered positive learning arenas probably because they offered different affordances for learning from the classroom, which is supported by recent research (Fiskum and Jacobsen 2012a; Myhre and Fiskum 2021). Due to the informal open space an outdoor setting represents, learners do not have to follow the norms and expectations of the classroom (Harris 2017) and students found it easier to interact away from direct observation by their teachers and peers, including communicating in groups.

It should be noted, that for the CG, the classroom seemed to represent a positive learning arena in which students enjoyed their English classes (FLE remained more or less constant both pre and post). This may be explained by a good teacher-student relationship and a friendly learning environment (Dewaele and MacIntyre 2019), which in this study was also supported by observations. The fact that the intervention did not cause significant change in FLA can be explained by data from the interviews; the students spoke little English in class. Hence, it is possible that FLA had not yet fully developed.

Because participants were very similar in age and gender, these background variables did not influence the results. Socioeconomic status in terms of parents’ educational background was not given weight due to two factors: in a Norwegian context the students are initially randomly placed in classes, hence socioeconomic status was comparable between the two groups. Also, the response rate for these items was low (EG 53.7%, CG 56.4%).

A six-week intervention is potentially too short to have much impact, especially as students reported they were not used to practising spoken English, and might therefore explain why the difference in pre and post is so limited.

Limitations of the present study include the fact that data on the students’ FL proficiency was not included. The study was not fully randomised and hence we cannot completely isolate the effect of the intervention. Furthermore, the EG and the CG had been taught by different teachers prior to the pre-test. This could have affected participants’ scores as the groups could have had slightly different preconditions in terms of learning. Also, groups had different teachers and sessions differed (EG 1 × 80, CG 2 × 45 minutes per week) complicating comparisons. However, the study had a naturalistic design and ethically we aimed to influence the school as little as possible. We cannot completely establish the causes of the patterns we observed, which could be linked to the move outdoors, or just a change in environment and the new affordances it allows. The present study also has a number of strengths. Intervention-based research is not very common in emotions research. A mixed methods approach collecting different data gives a more holistic picture. Another strength is the teachers’ perspectives on outdoor education compared to traditional teaching methods. Finally, the teachers' perspectives on outdoor education added rich insights.

Conclusion

The current study looked into the effect of outdoor environments on learners’ emotions over a period of six weeks. While the statistical analyses showed little change in FLA and FLE, the qualitative analyses revealed that learning in an outdoor environment increased the students’ feeling of safety and it reduced performance pressure and anxiety-provoking situations, which stimulated their eagerness to learn. The findings suggest that using a FL outdoors can offer young teenagers the opportunity to develop their communicative skills in a relatively informal, fun and stimulating environment which paves the way to positive attitudes, confidence and ultimately solid progress. Finally, we acknowledge that findings in this study can be the result of change in general, and not the outdoor setting itself. The pedagogical implication is that outdoor learning offers young FL learners a unique way to develop their budding communication skills. Further research could investigate whether a longer intervention might have a stronger impact and also if students would improve their spoken FL proficiency using an alternative teaching approach like outdoor education.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Tone Stuler Myhre

Tone Stuler Myhre is undertaking a PhD in foreign language learning and outdoor education at Nord University, Norway. She holds a position as university lecturer at the Faculty of Education and Arts at Nord University, Norway.

Jean-Marc Dewaele

Professor Jean-Marc Dewaele is former president of the European Second Language Association, and is current president of the International Association for the Psychology of Language Learning. He is Editor of Journal of Multilingual and Multicultural Development. He won the Equality and Diversity Research Award from the British Association for Counselling and Psychotherapy (2013), the Robert Gardner Award (2016) and the EUROSLA Distinguished Scholar Award (2022).

Tove Anita Fiskum

Tove Anita Fiskum is associate professor at Nord University, Norway. She has a PhD in health and science and outdoor education.

Anna Marie Holand

Anna Marie Holand is associate professor at Nord University, Norway. She has a PhD in biology, in the field of evolutionary biology and statistics, and she is working in the teacher education programme (Master’s degree). She is involved in research and teaching within statistics, ecology and evolution, outdoor education, sustainable development and school garden as a learning arena.

Notes

1 Observations and recordings of speech were included as supplementary methods to see if the students understood the tasks and if they communicated in English.

2 We dropped the ‘ C ’ for classroom as the EG was actually outside the classroom.

3 The language was simplified to suit the age group and ‘English’ was used instead of ‘FL’.

4 The first author is a qualified English teacher.

5 All observations were conducted by the first author.

6 We did not measure words per minute nor number of sentences, but only recorded the time spent on speaking English.

7 The analysis was done by two researchers.

8 Some students had missing values for both FLA and FLE.

9 The coding was a collaboration between two researchers, one with a linguistic perspective and one with an outdoor education perspective. If disagreement, discussions led to joint codes.

Appendices

Appendix 1. The FLCA scale and the FLE scale

How true do you find the statements to be?

Very true / A little bit true / Not very true / Not true at all

The FLCA scale

1. I start to panic when I have to speak without preparation in English class

2. Even if I am well prepared for English class, I feel anxious about it

3. I feel that the other students speak English better than I do

4. I get embarrassed if I need to speak in English class without being well prepared

5. Even if I am well prepared for English class, I am worried about saying something odd in class

6. I don’t worry about making mistakes in English class (reverse-coded)

7. I feel it is easy to speak in English class and I feel I do so quite well (reverse-coded)

The FLE scale

1. I can be creative in English class

2. I can laugh off embarrassing mistakes in English

3. I don't get bored during English classes

4. I enjoy English class

5. I feel as though I’m a different person during the English class

6. I’ve learnt to express myself better in English

7. I’m a worthy member of the English class

8. I’ve learnt interesting things in English class

9. In English class, I feel proud of my accomplishments

10. It's a positive environment in English class

11. It's cool to know English

12. It's fun to learn English

13. Making errors is part of the English learning process

14. The peers in English class are nice

15. The English teacher is encouraging

16. The English teacher is friendly

17. The English teacher is supportive

18. There is a good atmosphere in English class

19. In English class we form a tight group

20. We have common ‘legends’, such as running jokes in English class

21. We laugh a lot in English class

Appendix 2. Lesson plans and competence aims

Examples of lesson plans

Example 1

Activity 1: Talk about the entrepreneur project (a projects the pupils had completed the previous semester)

Setting: Natural setting (experimental group)/ the classroom (control group)

Organisation: Walk & talk in pairs or small groups (experimental group)/ sit in pairs or small groups (control group)

Time: Approx. 15 minutes

Material: A notepad and a pen

Task: The pupils should speak about:

• Which topic did you work on in your entrepreneur project (company and topic)

• What was your contribution?

• Did you/your group experience any challenges?

• Tell about the final result

Instructions for the teacher: Write the task on the board and have the pupils write down the task.

Activity 2: Describe your house

Setting: A natural setting (experimental group) /the classroom (control group)

Organisation: In pairs, but the partners are shuffled

Time: Approx. 20 minutes

Material: Notepad and a pen

• The pupils spend 5–10 minutes preparing a 3-minute spoken description of their dream house. They may make notes (keywords only) – they should not be reading from a manuscript.

• Place the pupils in pairs and ask them to exchange descriptions, each pupil talking for approximately three minutes.

• Now shuffle the partners, and ask the pupils to deliver the same description, but in 2.5 minutes.

• Shuffle again, and give them 2 minutes.

Activity 3: Role play

Setting: A natural setting (experimental group) /the classroom (control group)

Organisation: In pairs

Time: Approx. 20 min

Material: Role cards (made by the first author)

Instruction to the pupils: Look at the information on your role card and talk to your partner. Find a solution to the problem.

Role cards (in envelopes):

Pupil A: You are a guest staying at a hotel. The hotel website says it is a luxury hotel, but in your room the sheets and towels are dirty, the bathroom is too small, the street outside is very noisy and … (you decide two more problems). You want to change to a better room and you want a discount. Talk to the receptionist and solve the problem.

Pupil B: You are a hotel receptionist. There is a guest staying at the hotel who complains about everything, even when there is not a problem. You can move a guest to a different room, but you cannot change the price of a room. Talk to the guest and solve the problem.

Examples of role cards:

Pupil A: You are a guest staying at a hotel. The hotel website says it is a luxury hotel, but in your room the sheets and towels are dirty, the bathroom is too small, the street outside is very noisy and … (you decide two more problems). You want to change to a better room and you want a discount. Talk to the receptionist and solve the problem.

Pupil B: You are a hotel receptionist. There is a guest staying at the hotel who complains about everything, even when there is not a problem. You can move a guest to a different room, but you cannot change the price of a room. Talk to the guest and solve the problem.

Activity 4:

Experimental group: Walk & talk about what you see/observe on your way back to the school. Remind the pupils to speak English only!

For the control group: The classroom, organisation: In pairs, time: Approx. 10 minutes, material: laptops

Pupil A: Use your laptop and look at the information about films at the local cinema. Listen to your partner's questions and use the information to answer them.

Pupil B: Use your laptop and look at the information about films at the local cinema. Ask your partner questions to find out more about the films.

Instructions for the teacher: Write the roles of pupil A and pupil B on the board. Remind the pupils to speak English only!

Example 2:

Activity 1: Being a guide – name of the local place

Setting: A historic place (experimental group) /the classroom (control group)

Materials: A list of English keywords on a piece of paper or notebook (not a computer)

Groups: Small groups of two, three or four pupils

Time: Approx. 45 minutes

Homework: The pupils make a list of keywords about × (the place) based on the attached material and by looking at the following websites: x, x, x, x. In addition, they can ask their parents for information.

Instructions:

The pupils guide each other in small groups where one is the guide and the other pupil is asking questions. In groups of four two are guides.

The pupils should imagine that they are going to guide/tell American pupils visiting Norway about × (name of place)

The pupils need to be creative and make sure that both the guide and ‘the American pupil’ are taking an active part in the conversation.

The guide knows everything about × (name of place), and the other buildings on this site etc. The guide must appear to be knowledgeable and confident.

The visiting pupil wants to know as much as possible about × (name) and needs to ask many questions, such as

Can you tell me about … .?

What is this … .?

Why is this so … .?

When did this take place … .? Etc.

The pupils swap roles after approx. 20 minutes.

Activity 2: Typically Norwegian

Setting: A natural setting (the experimental group)/ the classroom (the control group)

Materials: A list of English keywords on a piece of paper or notebook (not a computer)

Groups: Small groups of two, three or four pupils, but not the same groups as Activity 1

Time: Approx. 45 minutes

Homework: The pupils make a list of keywords about Norway by using the following websites:

https://www.visitnorway.com/?lang=uk

https://www.visitnorway.com/

http://travelguide.nettavisen.no/42-things-every-foreigner-must-learn-about-norway/

In addition, they should bring at least two pictures showing examples of typically Norwegian culture, e.g. a typical Norwegian meal, a spare time activity, a typical Norwegian house etc.

Instructions:

The pupils should imagine that American pupils are visiting Norway.

The pupils would like to know as much as possible about Norway, Norwegian culture, how we live, etc.

‘The Norwegian pupils’ know a lot about the topic and ‘the American pupils’ are eager to ask questions (like the ones in Activity 1).

‘The Norwegian pupils’ also describe/explain the pictures they have brought to class.

The pupils swap roles after approx. 20 minutes.

Competence aims after Year 10

The pupil is expected to be able to

• use a variety of strategies for language learning, text creation and communication

• use different digital resources and other aids in language learning, text creation and interaction

• use key patterns of pronunciation in communication

• listen to and understand words and expressions in variants of English

• express oneself with fluency and coherence with a varied vocabulary and idiomatic expressions adapted to the purpose, recipient and situation

• ask questions and follow up input when talking about various topics adapted to different purposes, recipients and situations

• explore and describe some linguistic similarities and differences between English and other languages the pupil is familiar with and use this in one's own language learning

• use knowledge of word classes and syntax in working on one's own oral and written texts

• follow rules for spelling, word inflection, syntax and text structure

• read, discuss and present content from various types of texts, including self-chosen texts

• read, interpret and reflect on English-language fiction, including young people's literature

• read factual texts and assess the reliability of the sources

• use sources in a critical and accountable manner

• write formal and informal texts, including multimedia texts with structure and coherence that describe, narrate and reflect, and are adapted to the purpose, recipient and situation

• revise one's own texts based on feedback and knowledge of the language

• describe and reflect on the role played by the English language in Norway and the rest of the world

• explore and reflect on the situation of indigenous peoples in the English-speaking world and in Norway

• explore and describe ways of living, ways of thinking, communication patterns and diversity in the English-speaking world

• explore and present the content of cultural forms of expression from various media in the English-speaking world that are related to one's own interests

Appendix 3. Interview guide for students

• Why do you feel it is important to learn English?

• Why do you think it is important to be able to understand and speak English?

• In which ways do you best learn to speak English?

• How motivated are you for learning English?

• How important is it for you to be able to speak English?

• What do you think about speaking audibly in class?

• Do you find it more comfortable to speak English in smaller groups? Please explain.

• When you are listening to English music or are watching a British or American film, do you listen to what is being said?

• Have you spoken English much in class at primary school?

• Did you experience the activities you have been engaged in as motivating/inspiring/fun? Why was it so?

• Were you afraid of not speaking English correctly?

• What was positive with the activities you have been engaged in?

• Were there some activities which did not work well or activities you did not like doing?

• If you are asked to speak English audibly in class, are you worried about making language mistakes?

• Can you laugh of mistakes you make when you speak English?

• Do you feel like a different person when you speak English compared to when you speak Norwegian (your native language)?

• Even if you are prepared for English class, can you feel you are nervous?

• Do you feel that the other students in your class speak English better than you do?

• Do you feel you become nervous when asked to speak audibly in English class?

• Can you feel that you panic if you are asked to speak audibly in English class without being prepared?

• Do you worry about being asked to speak audibly in English class?

• Do you think there is a good atmosphere in English class which makes it feel safe to speak English?

• Do you feel that what you learn in English class is interesting and necessary to learn?

• Do you think it is fun to learn English? If so, why is it fun?

• In which situations would you like to speak English?

• How did you experience speaking English in these activities compared to your regular English teaching activities?

• Could the activities have been different so you could have spoken more English?

• What was it that made you interested in speaking English?

• Do you sometimes experience that it feels uncomfortable in English class? If so, in which situations do you feel it is uncomfortable? What do you think are the reasons that it feels uncomfortable?

• How should English classes be so you can pay attention, do tasks and at the same time not find it boring?

• You have now had a different type of teaching in English. Do you find this teaching more or less difficult or boring than your regular English teaching?