Agency and Affordances in Study Abroad

Abstract

Research in study abroad (SA) has developed over the past several decades, with an increasing focus on individual learners and their experiences. Despite this trend, it remains unclear what affordances beyond the classroom are available to students and whether and how learners enact their agency to make use of the available opportunities. The present study adopts an ecological framework to examine how four Japanese learners enacted their agency to capitalize on affordances for language learning beyond the classroom during study abroad. Agency is understood as a multidimensional construct, with both internal and external dimensions. Learners utilized a smartphone application to regularly report their language use and associated reflections. Further details about their experiences were elicited by post-study abroad questionnaires and interviews. The findings reveal the various kinds of language episodes reported by the students and how these learners exercised their agency in relation to their learning experiences. A discussion of what factors likely contributed to these learners’ enacting their agency is included. The study concludes with implications for supporting students’ out-of-class learning during study abroad.

1. Introduction

In Japan there are limited opportunities for learners to use English in naturalistic contexts; study abroad (SA), however, offers one potential avenue for learners to enrich their language learning experiences, especially outside of the classroom [1]. However, Japanese learners [2], as well as learners from other cultural backgrounds [3,4], often fail to take full advantage of these opportunities for language learning while they are abroad. Few studies exist that have investigated the reasons for such a lack of engagement, and even fewer have taken the learners’ experiences and perspectives as their starting point. A major constraint, both in classroom learning but in particular in learning beyond the classroom (LBC), is learners’ (perceived or real) lack of agency [5]. In order to better understand learners’ choices and behaviors in the SA setting, learner agency can serve as a fruitful starting point for investigation.

Accordingly, the present study takes an ecological approach to investigating the affordances for language learning beyond the classroom in SA settings by focusing on the reports of four Japanese learners. The students used a smartphone application to report regularly on their language episodes soon after they occurred. Further data were collected from post-SA questionnaires and interviews, and all of these were drawn on to uncover the available affordances and examples of students’ agency in capitalizing on these language episodes, using both external (i.e., social and behavioral) and internal (i.e., cognitive and affective) resources.

2. Theoretical Background

Since the 1990s, SA research has become more concerned with individual differences [6]. A number of factors, such as gender and aptitude, have been found to be influential in learners’ language and cultural learning outcomes during SA [7]. In more recent years, there has been an “explosion” of qualitative studies focusing on language learner experiences, which often include students’ perceptions of learning [8] (p. 459). While a great number of these have focused on various aspects of identity, comparatively few have examined learner agency.

Below, we review the key constructs of affordances and agency before examining studies of SA that have adopted these as their primary focus. From the gaps in the current literature, we draw our rationale and present the research questions of our study.

2.1. Ecology and Affordances

Applied linguistics has been influenced by the ecological theories of psychologists Bronfenbrenner and Gibson. Bronfenbrenner [9] adopted an ecological perspective toward understanding human psychology, arguing for the more careful consideration of the role of the environment in learning. He also explained the notion of environment as being more than “a single, immediate setting”, and, as related to human development, that it may involve connections between immediate settings in addition to “external influences emanating from the larger surroundings” (p. 22). Gibson [10], the first researcher to use the term affordance—initially applied to the natural world—explained that “the affordances of the environment are what it offers the animal, what it provides, or furnishes, either for good or ill” (p. 127). He went on to clarify that an affordance “implies the complementarity of the animal and the environment” (p. 127).

Van Lier [11,12] applied Gibson’s ideas to language learning. Believing that linguistic affordances emerge when we interact with the environment, van Lier explained that “in terms of language learning, affordances arise out of participation and use, and learning opportunities arise as a consequence of participation and use” [12] (p. 92). In this understanding, the presence of affordances depends on the perceptions and actions of a learner. From van Lier’s point of view, we cannot understand the environment and its affordances without regard to the learner. The environment and individual situations offer multiple affordances for learning, including utterances, gestures, and even physical objects.

Other researchers in the area of LBC have adopted the term in a more concrete way. For example, while Palfreyman’s research [13,14] focused on material resources (e.g., books and computers) and social resources (i.e., interaction with people), Menezes [15] presented affordances as learners’ written and oral interactions, as well as their experiences with cultural artifacts such as books, magazines, music, and movies (p. 70).

Here, we follow van Lier [11,12] and argue that affordances represent relations between an active, perceiving learner and an opportunity, including material or social resources, that enables language use. With this understanding in mind, we turn to a discussion of learners’ actions, thoughts, and emotions that are linked to their realization of affordances for learning—in other words, their agency.

2.2. Agency

Agency has been defined in a number of ways, with some emphasizing the importance of social structures, and others stressing the role of the individual. Regarding the former, Ahearn defined agency as “the socioculturally mediated capacity to act” [16] (p. 112). Discussing agency from this sociocultural perspective, Yang [17] states that “one’s agency is constantly constrained and empowered by social groupings, materials and symbolic resources, as well as other personal and social factors” (p. 9). Similarly, Lantolf and Pavlenko [18] claim that “...agency is never a property of a particular individual; rather, it is a relationship that is constantly co-constructed and renegotiated with those around the individual and with the society at large” (p. 148). In contrast, others have foregrounded the role of the individual. While Hicks believes that a sociocultural interpretation of agency “is biased towards systems and activities rather than individuals and is less suitable for understanding small personal stories” [19] (p. 39), Sullivan and McCarthy [20] recognize the importance of the individual, arguing that “a focus on lived experience can enrich [our] understanding of agency in people’s lives by complementing the centripetal tendency of much cultural theorizing with a centrifugal corrective” (p. 291).

In the field of applied linguistics, Gao [21] has adopted a somewhat balanced understanding of agency, which involves both learners’ will to act as well as the contextual situations in which they find themselves. Similarly, citing the work of Sealey and Carter [22], Mercer [23] argues for realist perspectives which “conceive of both structure (social relations and macro features of society) and agency (humans as agents in the human world) as interacting in a relationship of reciprocal causality...” (p. 428).

Agency can be understood to be multidimensional. Mercer [24] refers to two broad dimensions of learner’s agency: a learner’s sense of agency, which refers to how agentic a learner feels in general and regarding specific contexts, and a learner’s agentic behavior, including a learner’s participation and action (p. 42). In addition to observable actions, agency includes a learner’s thoughts, beliefs, and emotions. This is in line with Larsen-Freeman’s [25] observation that “agency is not only about behavior” (p. 67).

Agency is therefore well-suited for making sense of learners’ experiences as they capitalize on a range of affordances for learning while abroad, and agency and affordances complement each other within an ecological perspective. As Larsen-Freeman [25] explains, “Agency is always related to the affordances in the context, and thus inseparable from them, and affordances, in turn, are ecological rather than merely physical features of the world” (p. 65).

2.3. Research Focusing on Agency and Affordances in Study Abroad

Not many studies have examined language learners’ agency in SA settings through an ecological perspective. Furthermore, as we shall see, most existing studies focused on long-term sojourns, and the data regarding learners’ experiences were generally collected retrospectively.

Allen [26] looked at the interactive contact of 18 Americans in France and their perceptions of the value of the affordances with which they engaged. The students wrote in learning blogs twice a week during a six-week stay about their language and cultural learning as well as how they spent time out of class. Other data were collected through language learning histories, pre- and post-SA interviews, and the Language Contact Profile, an instrument designed to examine students’ L2 contact in study abroad programs [27]. Allen found that learners had more than twice as much contact with French with their host family than with their peers, and that there was very little other contact with French speakers out of class during the time abroad. This finding is not uncommon; as other researchers [28,29] have noted, learners report limited access to social networks during their sojourns. One limitation of this study, however, is that the analysis ignored non-social types of language use, such as reading and listening outside of the classroom.

In contrast to Allen’s [26] research, three studies by Benson and colleagues [28,30,31] did include non-social types of language contact. Benson’s [28] study explored how the context of SA was an emergent property of one learner’s experiences, as well as the role that agency played in mitigating the negative effects of a Hong Kongese student’s opportunities to use the language. Weekly blog entries, along with interview data collected before and after a semester of study at an Australian university, were used to write up a narrative to represent the student’s experiences. Rather than viewing context as a static backdrop to one’s SA experience, this study demonstrated how agency can transform context into affordances for language learning.

While the above studies by Allen [26] and Benson [28] involved students only writing in blogs weekly or bi-weekly, Benson et al. [30] required hourly recordings of a learner’s activities using an online diary application. The authors examined how a Colombian woman, who was taking a prerequisite to enter a master’s course in Australia, navigated her physical surroundings in order to maximize opportunities to learn English. Benson et al. argued that it is important to consider the spatial contexts of the city, which include its spaces, buildings, and various texts, which can all create opportunities for language use beyond the classroom. Importantly, this also includes non-social learning activities, such as reading or listening to texts in one’s environment. Similar to the study above by Benson [28], the authors consider the relationship between agency and context, and they “view context as something that both shapes and is shaped by learners and their actions” [30] (p. 23). While agency was an important construct in this study, the authors did not define the term.

Kashiwa and Benson [31] considered in-class as well as out-of-class learning. In particular, they investigated how SA students—seven Chinese students in Australia taking preparatory courses to enter a master’s course—conceptualized the relationship between these two learning contexts, how these conceptions developed, and how changes in their conceptions were reflected in their actions. Furthermore, they examined the role of students’ agency in transforming in- and out-of-class opportunities into affordances for learning. However, as with the study by Benson et al. [30], Kashiwa and Benson did not provide a definition of agency. In addition to interviews, diaries were collected in the form of weekly emails. These emails were responses to a prompt sent by the first author, but only four of the seven students participated in this process.

Some of the out-of-class activities in Kashiwa and Benson’s study, e.g., joining a soccer club, represent public pedagogies. This kind of informal language learning refers to “spaces, sites, and languages of education and learning that exist outside of the walls of the institution of schools” [32] (p. 1). This kind of learning is important because it is connected with learners’ living spaces as well as their intrinsic motivation for learning [33].

Collectively, these four studies were unique in that they adopted an ecological view to examine learners’ language learning experiences in SA settings. While they employed a range of data collection instruments, only the study by Benson et al. [30] included a method for students to share their language episodes regularly and in the moment. All but Allen’s [26] study focused on longer term SA students in programs lasting at least one semester. Finally, only Allen [26] and Benson [28] included a rigorous definition of agency. It therefore remains important to investigate, more concretely, how short-term SA students exercise their agency to take advantage of the available affordances beyond the classroom. In order to understand the relationship between learners and the unique settings where they use language, we need to collect data more regularly so they can provide us with more details of these occurrences.

As we have relatively little knowledge of “the ecologies of particular study abroad environments”, [1] call for more investigations into LBC during SA that examine specific settings. They share methodological recommendations, including the use of concurrent diaries and retrospective interviews to document LBC settings. As the authors continue, “there is a need to better understand how LBC experiences affect both language use IN THE MOMENT [emphasis original] as well as their broader impact on the learner and the learning process” (p. 567). They recommend using stimulated recall to “encourage recollection and interpretation” in order to examine things such as learners’ interactions in certain situations, feelings related to contact with native speakers, and their motivation (p. 568). The current study attempts to address the gaps in the literature and in the methodological approaches taken so far. The research questions for our study were therefore:

• What language episodes do students report in a study abroad context?
• What affordances for learning English beyond the classroom do learners report in a study abroad context?
• How do learners exercise their agency in capitalizing on the available affordances in learning beyond the classroom in a study abroad context?

3. Methodology

3.1. Theoretical Framework

This study takes an ecological approach to investigating students’ LBC during SA with the specific aim of examining the agentic interrelationships of the affordances in their individual learning contexts. We define these as opportunities, often centered around one or more social or material resources, that are perceived as being useful and/or acted upon by an individual for using or learning the language. Following Larsen-Freeman [25], we conceptualized affordances as instantiations of learners’ agency through which to understand how learners actively engage with their environment. In doing so, we could gain a holistic perspective on learners’ language learning experiences during SA. We adopt Thomson’s [34] definition of agency, “the capacity to act in the world” (cited in Heuman, 2014). In addition, like Mercer [23,24], we utilize a multidimensional operationalization of the construct, which includes external and internal forms of agency. These dimensions are defined below.

3.2. Research Design

Mixed research methods were used to collect and analyze the data. These included pre- and post-study abroad surveys, a post-study abroad interview, and the collection of reports of learners’ language episodes outside of the classroom via a messaging application. The collection and analysis of the data of the language episodes, the primary source of data, were largely done qualitatively. The aim of qualitative research, also known as naturalistic inquiry, is “to obtain insights into the complexities... of learning” [35] (p. 7).

3.3. Participants

Fifteen students volunteered to join this study, and of these, only 12 completed the PostSAQ. From these twelve students, we selected the eight learners to interview who had most consistently shared their language episodes with us during their sojourns. Following the interviews, we selected four students’ data to further analyze for this study. As the analyses are very detailed, including examinations of their social media messages, questionnaire responses and interview transcriptions, and following recommendations in the literature on qualitative research [36], four was deemed an appropriate number of participants for achieving both depth and breadth in the analysis.

The four students, Ayumi, Kenji, Maya, and Yuya, were selected for a few reasons. First, we wanted to have a balance in terms of gender, so two female and two male students were selected. In addition, as we thought it was important to have a variety of study abroad locations represented, students who went to universities in three different countries were chosen. Finally, we selected students who reported different kinds of experiences with using English outside of the classroom.

The four students’ pre-study abroad TOEIC scores ranged from 395 to 595. All student and institution names below are pseudonyms. General background information about these participants is presented in

Table 1. Learners’ general background.

Name	Gender	Age	Faculty	Year at University
Ayumi	Female	18	Agriculture	1st
Kenji a	Male	25	Pharmacy	3rd
Maya	Female	19	Agriculture	1st
Yuya	Male	18	Science & Technology	1st

^a Kenji was enrolled in a graduate program.

.

3.4. Setting

The four students in this study attended Central Japan University (CJU, a pseudonym), a large, private university in central Japan. In recent years, the university has made efforts to promote internationalization and English language learning. For example, the faculty of foreign studies was started and two new self-access centers (SACs)—consisting of English learning resources and dedicated staff that promote autonomous language learning—were opened. At the time of the study, one of the authors worked as a learning advisor in the university’s SACs. The Center for Promoting Internationalization, which oversaw the two SACs, was also in charge of the university’s SA programs. This center and the SAC have worked closely together to provide support for students before and after they went abroad.

The four students participated in three different short-term SA programs.

Table 2. General details of study abroad (SA) programs.

Location	Start Date	End Date	Duration	Residence	Learners
Melbourne, Australia	15 August 2017	10 September 2017	27 days	Homestay	Ayumi
Victoria, Canada	8 August 2017	27 August 2017	20 days	Dormitory	Maya
Cambridge, U.K.	13 August 2017	3 September 2017	22 days	Dormitory	Kenji Yuya

presents an overview of their SA programs.

3.5. Procedures and Instruments

3.5.1. Pre-Study Abroad

The SA students involved in this study enrolled in a special voluntary program, called Capture English Learning Opportunities (CELO), described in more detail below. These students joined a 90 min pre-departure meeting during which the first author delivered an overview of the CELO program. After students signed consent forms, they completed the pre-study abroad questionnaire (PreSAQ). The PreSAQ contained six sections, the contents of which were slightly modified from part 1 of the Pretest Version of Freed et al.’s Language Contact Profile [27]. The PreSAQ included questions about students’ general background, their SA program, experience with learning English, and experience traveling abroad.

Next, students were instructed on how to write the CELO program messages using LINE, a popular instant messaging application in Japan. The instructions about LINE included both the contents of the messages and the technology needed to send to them. Regarding the contents, students were provided with a list of possible areas of focus for their messages (see

Table 3. Example areas of focus for messages.

Noticing an opportunity for learning or using English

A description of how you used or learned English at a particular time

A reflection on what you did, including how it went and how you felt

Experience living overseas as it relates to learning English

Changes in feelings or attitudes about learning and using English, including those related to cultural perceptions and understandings

Sudden thoughts that you have about English, learning English, or culture

Successful moments with using or learning English outside of the classroom

Challenging moments with using or learning English outside of the classroom

). Students were told that, in many cases, a message would include more than one of these areas. As students were encouraged to write about their experiences outside of the classroom, many of their messages centered around language episodes. We define language episodes as experiences learners had with English beyond the classroom during SA. Most language episodes were shared directly with the researchers via the LINE messaging application; however, a few were shared in post-study abroad interviews.

3.5.2. During Study Abroad

The students were asked to send one LINE message per day to the first author while they were abroad. They could write in English, Japanese, or a combination of the two languages. We asked them to send these as soon as they could after a language episode, which we explained to them to be immediately after the episode or by the end of the day.

LINE enabled learners to report on their experiences in the moment or shortly thereafter. Thus, there was less time between the actual experiences and the reports that were sent. In accordance with the ecological theoretical framework adopted for this study, we thought that learners would be able to better capture the details of the settings and their accompanying thoughts and feelings through the use of the LINE application.

3.5.3. Post-Study Abroad

On returning, we asked students to review the LINE messages they had sent us and complete the post-study abroad questionnaire (PostSAQ). The PostSAQ included questions about students’ language episodes that were particularly significant for them. These questions were adapted from a questionnaire designed by Finch [37], which used questions to elicit critical incidents among his university students in Korea. As our questionnaire was designed for SA students, we made minor adjustments so that the questions matched the SA LBC context. The main questions in this section asked about particularly positive or negative experiences, as well as language episodes in which students felt they were especially successful or unsuccessful. Such experiences could then be discussed in more detail in the retrospective interviews.

Post-study abroad interviews were scheduled within six weeks after students’ return to Japan. The interviews had three main sections. In the first two sections of the interview, the first author asked students questions about their LINE messages and their questionnaire responses. This procedure follows recommendations [1] for using retrospective interviews in order to obtain deeper insights into learners’ thoughts that were expressed in concurrent diaries. In addition, the interview included questions about learners’ responses to the PostSAQ. Finally, the first author asked questions about students’ motivation, any changes related to their thinking about learning or using English, and new plans or goals they had for learning English.

3.6. Data Analysis

The students’ LINE messages were analyzed as detailed below.

3.6.1. Categories of Messages

First, students’ LINE messages that did not contain reports of language episodes were omitted from the analysis. In order to answer our first research question (1. What language episodes do students report experiencing beyond the classroom in a study abroad context?), we categorized the remaining messages based on the kind of language episode described. These categories were adapted from items in the Language Contact Profile (LCP) by Freed et al. [27]. Part 4 of the LCP contains a number of items related to different kinds of language use outside of the classroom. Some items from the LCP were initially employed as categories for the present study, and other categories were created in cases where the LCP did not offer suitable ones. Finally, through “an ongoing and recursive” process [38], we adapted the categories to better fit the messages written by the students (p. 260).

Table 4. Categories of learners’ LINE messages.

Having extended conversations in English

Accomplishing real-world tasks through brief exchanges in English

Listening to monologic talks (e.g., academic lectures, talks given by tour guides, etc.) and live performances (e.g., songs)

Classmates teaching or introducing each other to new things outside the classroom

Giving or preparing for presentations, speeches, or other monologic talks

Reading things outside of the classroom such as schedules, signs, menus, and advertisements

presents the categories that were used to label the messages.

3.6.2. Affordances and Agency

Next, we explain how the data were analyzed to answer the second and third research questions (2. What affordances for learning English beyond the classroom do learners report in a study abroad context? 3. How do learners exercise their agency in capitalizing on the available affordances in learning beyond the classroom in a study abroad context?).

As mentioned above, we operationalize affordances as opportunities, often centered around one or more social or material resources, that are perceived and/or acted on for language learning or language use. In the results section, affordances are presented through individual instances of each LINE message category (see Table 4). As an example to illustrate this, Ayumi wrote three messages labeled as accomplishing real-world tasks through brief exchanges using English. While each of these episodes is centered around a particular social resource in the form of an interlocutor, in Ayumi’s messages we also learn of important contextual details related to these settings. For instance, in message number 5, Ayumi reports that a gym receptionist “was busy and talked very fast and in a hurried manner”. Despite this difficulty, Ayumi still “managed to apply” for the membership. In this situation, we can see a specific instance of a service encounter, including details of the interlocutor. We also notice that Ayumi’s agency is expressed as being successful using English to get a gym membership. Such an analysis demonstrates that affordances should not simply be understood as social or material resources. Rather, they are opportunities centered around resources and, crucially, also involve students’ agency in relation to them [25].

Similar to Mercer [24], we see agency as a multidimensional construct, having both internal and external dimensions. However, whilst analyzing the results and drawing on existing theoretical frames to guide our interpretation, it became clear that the nature of our data was such that our initial distinction between internal and external agency had to be indexed in a more fine-grained manner. Drawing on the literature on language learner engagement, with its historical links to the research on agency [39], we followed Philp and Duchesne [40] in adopting four discrete dimensions: social, behavioral, cognitive, and affective. This resulted in two broad categories: external agency, including social agency and behavioral agency, and internal agency, including cognitive agency and affective agency. We operationalize these four dimensions of agency below.

Social agency is an action that transforms a social resource into an affordance for learning and/or using the target language. A social resource is a person or persons with whom the learner can interact.

Behavioral agency is an action that transforms a material resource into an affordance for learning and/or using the target language. Material resources are different from social resources in that one cannot interact with material resources in such a way that the resource is affected.

Cognitive agency refers to learners’ thoughts about learning and using the L2 that can contribute to their development in the language. Cognitive agency, as we define it, refers to both cognition and metacognition. As Oxford [41] explains, “Metacognitive simply means ‘beyond the cognitive’ and includes strategies that provide general management (control) of cognitive strategies” (p. 16). The literature on metacognition in language learning includes planning, setting learning objectives, selecting activities, monitoring, and evaluating as examples of metacognitive processes [41,42,43].

Affective agency refers to learners’ emotions in relation to learning or using the L2 that can contribute to their development in the language. Instances of affective agency are expressed in relation to other dimensions of agency and can be related to the past, including episodes which have just occurred, as well as to the future. Scherer’s [44] Geneva Affect Coder Label, which lists 36 categories of emotions, was used to help identify instances of affective agency.

3.6.3. Reliability

To ensure reliability, all of the LINE data were coded by an external rater, a former colleague of the first author. We first analyzed 25% of the LINE messages and achieved an inter-reliability rate of 73%. After this, some categories were revised to make them clearer and more distinct. The first author and external rater then coded all items and achieved an inter-rater reliability rate of 90%. We discussed our differences and agreed on all items that we had initially coded differently. Therefore, in the end, we were able to reach 100% agreement on all items.

4. Results

Below we examine how four Japanese language learners exercised their agency to capitalize on the available affordances for learning outside of the classroom during study abroad. The learners each sent between 11 and 20 messages. While Ayumi and Kenji wrote primarily in Japanese, Maya and Yuya mostly wrote their messages in English. LINE messages as well as questions or responses from the PostSAQ or interview, which were originally produced in Japanese, have been displayed in italics in the analysis below.

Table 5. Categories and number of LINE messages sent by each student.

Category	Ayumi	Maya	Kenji	Yuya
Having extended conversations	8	5	4	12
Accomplishing real world tasks	3	7	5	5
Listening to monologic talks	0	3	2	1
Classmates teaching each other things	0	4	0	0
Giving or preparing for presentations	0	2	0	2
Reading things outside of the classroom	0	0	3	0

In some cases, messages have been grouped under more than one category.

presents the categories of language episodes that were reported by these students and the LINE message numbers sent for each category.

Next, we look closer at the four learners’ messages to examine the relationship between their agency and the affordances in context, as well as connections between internal and external dimensions of agency. In addition, we present relevant comments from the PostSAQ and interview to offer greater insight into these learners’ language learning experiences. Due to issues of space, we limit our analysis to representative categories, messages, and comments by each student.

4.1. Having Extended Conversations in English

Messages categorized as extended conversations included sustained interactions with host families, international students, teaching and campus staff, as well as various strangers in public. Below, we look at examples of these reported by Ayumi, Maya and Yuya. All of these messages were selected as they contain examples of learners exercising social agency.

4.1.1. Ayumi

While Ayumi reported having extended conversations with an international student and a teacher, she wrote more messages about sustained interactions with her host family. These episodes included reports of both positive and negative experiences, and in the interview, Ayumi said that it became easier to communicate with her host family over time. We see this trend in the reports of her experiences, as, in her first message, she wrote of struggling to extend the conversation, but towards the end of her time, she reported being able to successfully talk about her home and school life in Japan.

In the following example, Ayumi includes more examples of these mixed experiences with English. She writes of conversing about various topics while playing card games:

Today at dinner time I talked a lot with my host family. Favorite foods, countries I have gone to, and so on. We played card games too and I was able to talk a lot. I really like takoyaki [fried breaded octopus formed in the shape of a small ball], but, although it’s a Japanese food, I could not quite convey what kind of thing it was and I got frustrated. I wanted to convey the taste… As this was the first time that I was able to talk like this, from now on too, at dinner time and other times, I want so speak actively.

As in other messages, Ayumi expresses her cognitive agency in monitoring her participation in this conversation. On the positive side, she says that she could speak a lot. However, when talking about Japanese food, she experienced difficulty in conveying the taste of a particular dish, and she expresses her frustration. However, Ayumi concludes the message with a positive outlook. Despite feeling frustrated, she explains that this was her first time being able to have this kind of conversation, another example of her monitoring. She offers a goal-related statement, saying that she wants to speak actively like this from now on. Fortunately for Ayumi, in other messages, she characterized her host family as being supportive in conversations, which likely aided her in her pursuits to continue conversations with them.

4.1.2. Maya

More than half of Maya’s messages involved communication with non-Japanese international students. These interactions included extended conversations, classmates teaching each other things, and brief exchanges. In the following example, Maya writes about talking to her international classmates over a meal:

In the noon. I had lunch with my classmates. We enjoyed talking each other. I think lunch time is good opportunity to know table manners of other countries.

This message not only shows Maya’s social agency, but also her affective agency, as she comments on her enjoyment in conversing with her classmates.

This positive experience communicating with non-Japanese students was not unique. In her response to PostSAQ14 (“Was there anything that caused you to have a particularly strong positive feeling in relation to learning or using English? Please explain what happened and why you think you felt that way.”), Maya answered, “Enjoying talking to people from other countries.” Maya explained in the interview that, early on, she had asked “other countries’ friends” to attend a night class with her, as her Japanese friends were not interested in attending. This led to her establishing stronger relationships with non-Japanese international students. Maya was able to exercise her social agency in the ongoing communication with this group. As Maya indicated that she enjoyed talking to these students, there is a relationship between her affective agency and her repeated social interaction with these students.

4.1.3. Yuya

Approximately half of the episodes Yuya reported involved interactions with his TAs or other university staff. As with Ayumi, these episodes included both positive and negative experiences with English. The following message is one of seven that Yuya wrote involving extended conversations with TAs. Here, he writes about his struggle to talk about his holiday:

Hi, Andy. Today I’d like to talk about my chatting. Today’s lunch time, I talked with teaching assistant about holiday. However, I was very sad because I wasn’t able to talk about my thinking. Therefore, I think I should learn more English. see you.

Yuya expresses his cognitive and affective agency in monitoring his inability to express himself and the concurrent expression of sadness. However, the reflection on his poor performance leads to his goal-related statement of wanting to “learn English more”.

Yuya also wrote two messages about interactions with the porter’s lodge staff at his dormitory in Cambridge. In this LINE message he reports a struggle to communicate with them:

Hi, Andy. Today I’d like to talk about my English. Almost morning, I talk with some Porter’s Lodge people but yesterday I didn’t it. Today I tried again, but I wasn’t able to understand. Therefore, I think I should do that every day. see you.

Here, Yuya reflects on the activity of speaking with the porter’s staff most mornings. Regarding this unsuccessful experience, his cognitive agency is expressed as he monitors his ability to understand the staff. In addition, he makes the goal-related statement that he should speak to the porter’s lodge every day. The feeling of failure drives his decision to change future behaviors and to interact more regularly with the porter’s staff.

While the above episodes present examples of his struggles to use English, Yuya considered the TAs and the porter’s lodge staff as supportive social affordances, and he regularly sought them out. In his answer to PostSAQ11 (“What thing(s) do you think were helpful for your English learning?”), he listed TAs, along with teachers and school staff, as being beneficial for his English learning. Yuya said in the interview that he spoke with the porter’s lodge staff “almost every day”, sometimes at their office and sometimes around the campus. He explained that it was easy to communicate with the porter’s staff as they “talked very slowly so [he] can make sense”. He also enjoyed these conversations as the porter’s lodge staff liked to joke. The TAs, too, being regularly around international students, were likely easy for Yuya to talk to. It may have been due, at least in part, to the supportive dispositions of the TAs and the porter’s lodge staff that, even when he struggled, Yuya wanted to meet them repeatedly and continue using English.

4.2. Accomplishing Real-World Tasks through Brief Exchanges in English

In this section we look at episodes centered around brief exchanges. These short English transactions took place while using public transportation, shopping, ordering food and drinks, asking for assistance, as well as in other situations. We examine examples of learners’ social agency reported by Ayumi and Kenji.

4.2.1. Ayumi

In her LINE messages and in the interview, Ayumi reported having uncomfortable interactions with various staff. We see an example of this in the following message, where Ayumi writes about applying for a gym membership:

Yesterday I applied to the school gym that seems like it can be used. Because the receptionist was busy and talked very fast and in a hurried manner, I was hesitant but managed to apply. I was acutely aware that listening skills are really important in everyday situations.

This episode reveals Ayumi’s cognitive agency as she monitors her ability to carry out a transaction. Ayumi’s message also suggests that she struggled to communicate with the receptionist. As she observes, the receptionist was not very accommodating to her, and Ayumi hesitantly “managed to apply”. Ayumi was also able to monitor her own listening skills, implying that her listening ability was not quite adequate for this task.

Ayumi recalled episodes like this in the PostSAQ and interview, expressing frustration when the staff acted in a hurried or impatient manner toward her. At times, the staff even seemed annoyed at her. However, indicating her affective agency, she reported that such episodes were ultimately positive experiences and motivated her to study English more.

4.2.2. Kenji

Kenji, the oldest of the four learners, wrote about several occasions where he enjoyed going out to enjoy drinks, conversation, and live music. These episodes enabled him to notice his abilities and consider ways that he could improve. We see an example of this in the following message, in which Kenji provides two examples of his social agency: having a conversation with a taxi driver, as well as ordering drinks.

Yesterday, I went to an English pub for the first time since I came to the U.K. It is the bar called EAGLE, where Dr. Watson and Dr. Click, who discovered the double helix of DNA, used to go. I went by taxi. I had a conversation with the taxi driver, and in the shop I was able to properly order a drink. Because I want to be able to speak more about various things, I want to do my best at studying English.

Demonstrating his cognitive agency, Kenji shares the goal-related statements of desiring to speak more about various topics and to do his best to study English.

After his study abroad experience, Kenji commented that he wanted to express himself more completely. In response to PostSAQ16, which is about changes in motivation before and after SA, Kenji said, “I feel that I want to study more to express how I feel and what I want to say in English. Also, I feel that I want to put effort towards vocabulary to get a variety of expressions.” This response reveals Kenji’s cognitive agency in relation to noticing his limitations in expressing himself, i.e., monitoring, as well as the goal-related comment of wanting to study more vocabulary. Reflecting on his SA experiences, Kenji was able to make concrete statements regarding how he would like to improve his English skills.

4.3. Classmates teaching or introducing each other to new things

As noted above, Maya reported many episodes involving other international students. Here, we consider her messages about classmates teaching each other things. Only Maya wrote of such episodes, and all of them involved examples of her social agency.

Maya

In Maya’s messages about classmates teaching each other things, she wrote of teaching international students about Japanese sweets and golf as well as learning about air hockey and Korean dance. Most of these experiences thus involved some physical activity, and gestures became a useful tool for her. We see this in the following message, which is about Maya teaching a Chinese friend to play golf.

In the afternoon. I played golf with my friend. I play golf in the class of Central Japan University, so I taught him how to play. I couldn’t come up with appreciate words. I think used gestures than English. But playing sports is good opportunity to improve communication skill.

Note. In the interview Maya confirmed that she meant to say “appropriate words” rather than “appreciate words”.

Maya also includes some comments related to her cognitive agency in this message. In her final comment about playing sports, she evaluates such activities as being helpful for her communication skills. In addition, regarding her monitoring, she notes that she used gestures when she could not find the appropriate words. In the interview, too, Maya identifies gestures as being helpful for her communication. In addition to helping her explain things, she says gestures helped her to get others’ attention when she wanted to speak. Gestures thus represent a strategy Maya employed to enhance and increase her participation in conversations, a tool that enables her social agency.

4.4. Reading Things Outside of the Classroom Such as Schedules, Signs, Menus, and Advertisements

Only Kenji wrote about reading things outside of the classroom, and we consider his messages below.

Kenji

Kenji’s messages about reading things outside of the classroom, which involved him reading street signs, menus, and exhibitions at a museum, highlight his behavioral agency. An example of this is presented below, as Kenji demonstrates his behavioral agency by reading the description of a painting in a museum:

I went sightseeing in London today, too. I saw van Gogh’s Sunflower at the National Gallery. I read the painting’s description, and I was able to deepen my understanding. Since I came to study abroad, not only English-speaking opportunities, but English reading opportunities have also increased, and little by little, I feel that my English ability has been improving.

Kenji says that in reading the description he was able to “deepen [his] understanding”, demonstrating not just a linguistic benefit but also one of gaining knowledge. In the second half of the message, he provides two examples of his cognitive agency. First, he notes that, in addition to speaking opportunities, reading opportunities have increased for him. This noticing of the affordances for learning is important when seeking out learning opportunities beyond the classroom in the study abroad setting. Furthermore, Kenji comments on improvements in his English, an example of his monitoring. This episode illustrates how Kenji benefitted from seeking out non-social affordances for learning English.

4.5. Summary of Results

During their sojourns, the four learners enacted their agency to capitalize on a range of affordances beyond the classroom for language learning. In terms of external agency, students frequently reported their social agency in extended conversations and brief exchanges. However, behavioral agency was also reported, particularly by Kenji. The learners’ messages also revealed important connections between their external and internal agency. For example, they frequently demonstrated their cognitive agency in monitoring their language use; and, while reporting both successful and unsuccessful experiences, affective agency sometimes accompanied cognitive agency in the expression of positive or negative emotions. Finally, learners expressed their cognitive agency in goal-related statements, and these often emerged from noticing gaps in their language abilities.

5. Discussion

As we saw above, the four learners were successful, at least to some degree, in capitalizing on the affordances of SA. It is therefore worthwhile to consider which factors seemed to facilitate this success. Below, we will discuss how the settings and other people facilitated learners in exercising their agency while abroad. We also consider the role of learners’ internal states, including their anxiety and motivation.

5.1. Proximal Settings and Social Networks

All of the learners in this study reported having social interactions during their sojourns, and many of these were connected with their living arrangements and school life. In terms of living arrangements, Yuya reports on his regular conversations with the porter’s lodge staff, who were on the campus to support the student residents. Yuya says that he sometimes talked to them at their office, whereas other times, he saw them around campus. Ayumi writes of her ongoing talks with her host family. Many of these took place in the home after dinner, a time that afforded opportunities for extended conversations. Another proximal location that afforded social network development was the school itself. Maya took an evening class that led to her sustained relationships with other international students. Due to these connections, she was able to enjoy a range of interactions with these new friends during such activities as shopping and eating out. The positive and sustained relationships that were reported by the four learners point to the importance of taking advantage of and capitalizing on social networks. Additionally, the proximity of these networks afforded learners ease in terms of incorporating English speakers into their daily routines.

Benson et al. [30] have demonstrated how opportunities for language use can be rather constrained based on the realities of one’s living situation. That the students in our study reported limited connections in their social networks is therefore understandable, and others have reported a similar trend [28,29]. As the four learners in our study only spent three to four weeks abroad, it was likely difficult to form new relationships. Despite this challenge, the learners were able to make sustained social connections, and these led to multiple opportunities for the learners to express their social agency while abroad.

Yet, it is not the case that proximity alone assures that social networks afford interaction in the target language. Wilkinson [45] and Allen [46] show that homestays, for example, do not always lead to extensive interaction in the target language. In our study, students characterized their frequent interlocutors as engaging and supportive. Ayumi said that her host family played games with her, and Yuya wrote of playful conversations with the porter’s lodge staff. Ayumi and Yuya also reported that these groups used English that was easy to understand, and they both wrote of conversations with teachers and teaching assistants who also demonstrated patience in listening to them. We also saw examples in interactions with international students that were likely to result in lower anxiety levels. Ayumi said that it was easier to speak with international students than with Australians. And Maya, who reported many episodes with international students, reported in the questionnaire that she had a positive and enjoyable experience speaking with them.

Thus, these positive and sustained relationships point to the importance of not only taking advantage of proximal social networks, but also of seeking out individuals who are supportive. While proximity affords learners ease in terms of incorporating English speakers into their daily routines, interlocutors’ supportive dispositions enable learners’ affective agency in making them more comfortable and reducing their anxiety.

5.2. Public Pedagogy and Intrinsic Motivation

In contrast to the other three students, Kenji expressed his agency through more solitary activities. While some of these involved interactions with other individuals in shops, a number of his episodes involved him listening to live performances and reading things outside of the classroom, including exhibits in museums and street signs. These episodes can be considered examples of public pedagogy. According to Sandlin, Schultz, and Burdick’s [32] handbook on the topic, public pedagogy refers to “informal spaces of learning such as popular culture, the internet, public spaces such as museums and parks, and other civic and commercial spaces” (p. 2). Kenji exercised his behavioral and social agency to explore these informal spaces for language learning.

While the notion of public pedagogies is helpful in understanding what students did beyond the classroom and how these activities relate to their learning, its connection with learner agency may not be evident. When learners explore their surroundings and engage in activities such as listening to live music, visiting museums, and exploring other public places, they are making connections between themselves and the outside world and often exploring their own interests. Self-determination theory (SDT), a theory of motivation, “begins with the assumption that people are by nature active, and that with an evolved tendency to engage the environment, assimilate new knowledge and skills, and integrate them into a coherent psychological structure” [47] (p. 225). SDT includes cognitive evaluation theory, which explains that people engage in activities because they find them interesting and enjoyable. This notion of intrinsic motivation can help us to understand why students take advantage of certain learning activities, such as those connected with the public pedagogies, beyond the classroom.

Kenji detailed a number of activities he found personally interesting and that centered around public pedagogies. Regarding his affective agency, he repeatedly referred to his enjoyment of listening to and discussing music, and he expressed his interest in art and in exploring public places by reading signs and advertisements. Thus, Kenji’s agency was facilitated by engaging with various public pedagogies that he found interesting.

Finally, it is important to highlight that a number of Kenji’s language activities were non-social ones. Benson et al. [30] argue that researchers should not only focus on the social aspects of LBC but the spatial ones as well. Kenji’s SA experience included non-social activities that impacted his learning, including reading menus, street signs, advertisements, and museum displays. This behavioral dimension of Kenji’s agency must be considered, as these experiences represent a significant part of Kenji’s study abroad experience.

6. Conclusions

We conclude this study with a brief discussion of the implications, limitations, and directions for future research. There are a few theoretical implications from the present study for research examining agency in study abroad settings. First, by adopting an ecological framework for examining students’ language learning beyond the classroom, we could perceive how agency and affordances are inextricably intertwined. Agency is enacted in relation to something, and opportunities and resources become affordances as a result of learners’ perceptions and actions [25]. Second, an ecological framework was helpful in better understanding learner agency in relation to non-social, material resources. As Benson et al. [30] note, not much research has looked at the physical spaces and material resources beyond the classroom. By adopting an ecological lens, this study was able to consider a wide range of learning experiences, including those in which learners reported reading and listening to various authentic texts outside of the classroom. Finally, by operationalizing agency as a multidimensional construct, we could examine the internal dimensions of agency along with the external ones. This perspective is critical in order to better understand the affective and cognitive processes that underlie learners’ external language activities.

Regarding the limitations of this study, we agree with two general recommendations from Kinginger [48] about ways to strengthen qualitative studies in SA research. Kinginger says that few studies have looked at the impact of SA experiences longitudinally. If data were collected at various time intervals after their SA experiences, then we could better understand the long-term impact of these programs on learners. Kinginger also suggests that qualitative researchers in SA should collect data from a greater number of stakeholders. A broader understanding of learners’ SA experiences can be gained by doing so.

Finally, we offer some suggestions for future research. While this study made use of a messaging application, other kinds of technology could be used to enhance the kinds of data that are collected. First, as suggested by [1], recordings of students’ LBC interactions could be made. This would give us actual data of learners’ interactions in the target language. In addition, photo or video recordings of the settings could provide more contextual details of where these episodes take place. Any of these forms of media could be used in tandem with stimulated recall procedures to elicit richer and more detailed accounts of students’ experiences. It may also be helpful to give learners the option to make video or audio recordings of their thoughts. Particularly, if they did this in their L1, such recordings would be less time consuming than composing a text message. Moreover, as the use of video conferencing apps such as Zoom has become much more prevalent, particularly as a result of the COVID-19 pandemic, it is now much easier for researchers to conduct interviews with students during their sojourns.

Study abroad provides learners with a wide range of authentic language learning experiences, and, by investigating learners’ abilities to draw on the potential benefits in the environment, we can more concretely understand how learners exercise their agency in these complex settings as well as provide more effective support for their learning beyond the language classroom. It is our hope that more studies will pursue this line of research in the future in order to provide greater insight into the interrelationships between internal and external learner agency and the affordances of study abroad.