32

Learning through immersion
during study abroad
*

Sally Sieloff Magnan and Barbara A. Lafford

Historical discussion

Input and interaction are recognized as essential for language learning (see Mackey, Abbuhl, and Gass, Chapter 1, this volume). Because it can provide large amounts of authentic input and interaction with native speakers, immersion in the target country has been considered the ideal context for language learning. This branch of second language acquisition (SLA) research has focused on the effects of a study abroad (SA) experience on foreign language learning. Research on SA thus fits into studies that investigate the context of learning. The SA context is often compared with learning in foreign language classrooms or in immersion programs in the native country, in an attempt to demonstrate advantages of the abroad experience or, more recently, to consider whether this context is more conducive to language learning than others.

For several decades research in SA has asked two basic questions: What do sojourners learn? What factors facilitate that learning? The way researchers have approached these questions relates to their SLA theoretical perspectives.

In line with interactionist theories of language learning, early SA research focused on linguistic outcomes to demonstrate the advantage of the abroad immersion setting for language acquisition. Often it used quantitative analysis to measure language gain and relate it to specific programs or behaviors of learners abroad. As this line of research progressed, the definition of what constitutes language broadened from grammar, vocabulary, and pronunciation features to include pragmatic and certain sociolinguistic features, which has led to a change in focus for SA research.

Issues surrounding factors that facilitate learning have become the focus of a second wave in SA research, dating from the mid-1990s. Looking primarily at individual and social factors, this research responds to the “social turn” in second language (L2) studies (Block, 2003), which has stimulated interest in investigating issues of identity, gender, and the effects of social networks on SLA. This body of research, which increasingly uses qualitative techniques such as diary studies and interviews, has presented case studies that cast light on curious results from language gain studies done in quantitative frameworks. For example, insights into the individual nature of family stays helps explain differences in language gain among learners who all participate in homestay programs.

Certain early aspects of this research (e.g., interaction studies) relate to cognitive psychology (psycholinguistics). Other more recent aspects (e.g., work on identity, gender, and learner choice of behavior) relate to more socially based theories, including especially Vygotskian-based work in sociocultural and activity theory. The trajectory of SA research has thus followed the development of SLA theoretical approaches, with each contributing to our understanding of how input and interaction enable language learning.

Core issues

The primary issue in SA research today is to determine relationships between language gain and social factors. Research looking at these two areas has rarely connected them in systematic ways. Language gain studies have typically worked with many learners in order to enable the use of statistical analysis, whereas the studies of social factors which seek to examine the depth of experience of an individual, have needed to use case studies on small numbers of learners. The different foci and scope of studies in these two paradigms has made it difficult to consider findings together and thus to come to a comprehensive picture of learning in a SA setting.

A most compelling issue then is how individual and social factors affect linguistic gain outcomes, that is, to make the connection between the first question, “What do sojourners learn?” and the second question “What factors facilitate that learning?” Once we understand this crucial connection, we can address the practical issue of how programs, instructors, and individuals might enhance the immersion experience abroad. We can only then also address the related questions that have occupied much research to date: Who should go abroad, and when in the learning trajectories? What length and type of programs are best? During the programs what should learners do abroad? What is the role of native hosts in the learners’ SA experience? What goals are achievable during a SA experience? How do individual differences affect learners’ outcomes and satisfaction with SA?

Findings to date have provided partial answers that point toward the following notions: language acquisition is best fostered by longer study before the SA experience, by longer stays abroad, by intense social contacts with native hosts and native social networks and by the avoidance of native language (L1) peer social networks. Yet, language gain is no longer seen as the only goal of SA: personal growth and learning strategy development, identity shifts, and broadened motivations and attitudes are also now viewed as positive outcomes. The crucial nature of the interaction between individual and situational variables is becoming evident as research demonstrates repeatedly how different students shape experiences in unique and personal ways.

More nuanced examinations of these outcomes and the factors fostering them will help researchers compare language learning gains in immersion settings abroad with those in country, in both US immersion settings and classrooms. The SLA field needs to go in the direction of making ties between different theoretical perspectives and their diverse data sets, between the questions of what learners acquire and what facilitates their learning. These ties need to be made through both investigations of large data sets using statistical analysis as well as closer examinations using qualitative methods.

In this chapter, we will focus primarily on the individual and social factors that facilitate learning during SA. However, to situate that discussion, we will also consider briefly the work on linguistic gain. Respecting the chapter's length limitations, we have selected certain studies as touchstones to demonstrate the major issues, research methodologies, and findings. We look exclusively at languages other than English, which stands apart in its global use today, and focus mostly on learners in European countries, consistent with the dominance of Europe as a SA location (Gore, 2005). Beyond Europe, we include some studies from Asia, especially Japan, where the number of SA students has been growing (Gore, 2005).

Data and common elicitation measures

Research on SA has used a variety of data types, research designs, data collection instruments, and qualitative, quantitative, or mixed methods analytical approaches. The choices scholars make limit what we know about the data, and determine what other kinds of data are needed in future research.

Elicitation measures

Types of data. The types of data elicited in SA research are reviewed under the Instruments section below.

Research designs. Both experimental and qualitative/descriptive research designs are common in SA research. Experimental designs incorporate both single (SA) group pre-post test designs (O1 X O2) (O1= first observation [pre-test], X=treatment [SA experience], O2= second observation [post-test]), and quasi-experimental designs, which use pre- and post-tests with experimental (SA) and control (AH=at home) groups. These designs have been used to study changes over time in linguistic outcomes (e.g., Brecht et al., 1995; Ife et al., 2000; Kinginger, 2008; Segalowitz and Freed, 2004), sociolinguistic and intercultural competence (e.g., Allen et al., 2006; Elola and Oskoz, 2008; Jones and Bond, 2000; Kinginger, 2008), learning and communication strategies (e.g., Adams, 2006; Lafford, 1995, 2004), and attitude and motivation (e.g., Chieffo and Griffiths, 2004; Isabelli-García, 2006). Although the size of these studies varies considerably with some being quite large (e.g., Brecht et al., 1995; Davidson, 2010), a general weakness of these experimental studies relates to sample size. In addition, these studies cannot use random samples because they work with the limited number of students who self-select to study abroad, which must also be considered when interpreting statistical results (noted much earlier by Freed, 1995).

Qualitative/descriptive SA research includes case studies of individual learners as well as group studies using agglomerated data from several learners (outside of a pre-post test format). This type of research has complemented experimental designs in its investigation of attitudes and motivation (e.g., Isabelli-García, 2006; Kinginger, 2008), intercultural competence (Elola and Oskoz, 2008), and learning strategies (Paige et al., 2004). Such contextualized case studies have also provided unique insights into changes in self-perceptions of identity (e.g., Iino, 2006; Kinginger, 2008; Miyahira and Petrucci, 2006; Pearson-Evans, 2006; Pellegrino-Aveni, 2005), the role of gender and race in language acquisition (e.g., Polanyi, 1995; Siegal, 1995; Talburt and Stewart, 1999), learners’ attitudes and motivations (e.g., Brecht and Robinson, 1995; Isabelli-García, 2006; Kinginger, 2008), and information about the social networks in which learners operate while abroad (e.g., Isabelli-García, 2006; Kinginger, 2008; Knight and Schmidt-Rinehart, 2002; Magnan and Back, 2007; Papatsiba, 2006; Pearson-Evans, 2006). What is typically missing here, however, is a quantitative component on the larger data set from which the case studies were drawn (see Kinginger, 2008, as a noted exception), which would allow researchers to combine a broader view of the SA experience to complement the restricted lens of the case study.

Standardized and other objective tests used to study language gain. The research on SA includes the use of large-scale standardized tests. Such tests have been used for diverse languages to measure growth in different linguistic skills: listening and reading abilities in Russian (Educational Testing Service Listening/Reading, Brecht et al., 1995); general proficiency in Japanese (Japanese Proficiency Test, Huebner, 1995); and general proficiency in French (Test de français international, Kinginger, 2008). Objective tests have also been used in a pre-post test format to measure gains in vocabulary (French: Language Awareness Interview, Kinginger, 2008; Spanish: Lexical Association Test, Ife et al., 2000), and in grammar (French: Test de français international, Kinginger, 2008; Spanish: cloze tests [Schell, 2001] and grammaticality judgment tests [Isabelli, 2004]).

A common instrument used in SA research to collect oral data has been the American Council on the Teaching of Foreign Languages (ACTFL) Oral Proficiency Interview (OPI). Using this tool, SA scholars have investigated changes in oral proficiency (Brecht et al., 1995; Guntermann, 1992a, b; Kinginger, 2008; Magnan and Back, 2007; Segalowitz and Freed, 2004), grammatical knowledge (Collentine, 2004; Isabelli, 2004), pronunciation (Simoes, 1996); fluency (Segalowitz and Freed, 2004), communication strategies (DeKeyser, 1991; Lafford, 1995, 2004), and lexical abilities (Guntermann, 1992a, b; Lafford and Ryan, 1995). Although the OPI has been used quite often in SA research to study these factors , criticisms against the OPI as not representing a natural form of interaction (e.g., Johnson and Tyler, 1998) need to be kept in mind when relying on this measure in SA research. In addition, more SA studies should use multiple measures to document linguistic abilities before and after a sojourn abroad in order to triangulate results and gain a greater understanding of the effects of the SA experience.

Instruments used to study individual and social factors. In contrast to SA studies investigating linguistic gains (products), instruments used to gather data about factors influencing the learning process are more open-ended and personalized in nature. They range from written surveys and personal diaries to in-person interviews and direct observations.

Surveys and questionnaires. Surveys and questionnaires have gathered valuable information on several variables that might influence the trajectory of target language acquisition. The Language Contact Profile (LCP, Segalowitz and Freed, 2004) and other demographic questionnaires have been used to record learners’ personal information (age, sex), linguistic backgrounds, and language contact with the native and target languages through media and hours spent with host families (Spanish: Lafford, 2004; Segalowitz and Freed, 2004; French: Kinginger, 2008; Magnan and Back, 2007; French/Italian/Spanish: Allen et al., 2006; Japanese: Miyahira and Petrucci, 2006; Russian: Pellegrino-Aveni, 2005). In these studies, the LCP and other such measures are sometimes used retrospectively (e.g., Magnan and Back, 2007), which leads scholars to question the accuracy of data they present.

Surveys and questionnaires have also been used to collect data on learners’ attitudes and motivations, and how they might change over time in a SA context (e.g., Chieffo and Griffiths, 2004; Isabelli-García, 2006; Pellegrino-Aveni, 2005). Other surveys and questionnaires have been used to study learning strategies (e.g., Strategic Inventory for Language Learning, SILL [Adams, 2006; Huebner, 1995] and a languages strategy survey [Paige et al., 2004]), perceptions of identity and personality (Allen et al., 2006; Bakalis and Joiner, 2004; Miyahira and Petrucci, 2006), adaptation or acculturation to the target culture (e.g., Strategic Inventory for Learning Culture, SILC [Paige et al., 2002] and the Intercultural Development Inventory, IDI [Allen et al., 2006; Elola and Oskoz, 2008]) and the impact of SA on learners years after their sojourn [Dwyer, 2004].

Discourse completion task questionnaires, in which learners complete a dialog with appropriate rejoinders, have been used to gauge gains in pragmatics (Speech Act Measure of Language Gain [Cohen and Shively, 2007]). An extended version of this task (Free Discourse Completion Task [Barron, 2006]), in which the learner has to write both sides of a dialog, has been used to study the acquisition of pronouns of address. These tasks can be criticized for artificiality, but are widely used because they can be administered relatively easily in a fairly controlled manner. Lafford (2010) points out how methodological variations in interlanguage pragmatics studies (e.g., research designs; instruments [oral and written discourse completion tasks (DCTs), roleplays, verbal reports, naturalistic data] for data collection; learner characteristics [pre-treatment language levels, living conditions]; pragmatic instructional interventions; data analysis) used in SA studies to date could directly affect our (mis) understanding of how L2 pragmatic competence is acquired.

Journals/diary studies. This method of data solicitation has been quite prominent in the study of learners’ perceptions of identity, race, and gender (e.g., Russian: Pellegrino-Aveni, 2005; Polanyi, 1995; French: Kinginger, 2008; Japanese: Iino, 2006; Miyahira and Petrucci, 2006; Pearson-Evans, 2006; Siegal, 1995; Spanish: Isabelli-García, 2006; Talburt and Stewart, 1999). Journals and diaries have also been used to study learners’ attitudes and motivation (Russian: Polanyi, 1995; French: Douglass, 2007; Kinginger, 2008; Spanish: Isabelli-García, 2006), social networks (Pearson-Evans, 2006), learning strategies (Adams, 2006), and pragmatic abilities (Siegal, 1995). The intimate, personal nature of diary and journal entries provide a unique window to an individual learners’ thought processes and attitudes and how these may change over time in reaction to “rich points” (Agar, 1996)—meaningful experiences that promote new understandings of the target culture.

Social network logs. The number and depth of social relationships that learners establish and maintain abroad has been studied with social network logs. Researchers have used these logs to record both contact with natives and with compatriots from the home country, other foreigners, and (virtually) with people back home. For example, Isabelli-García's (2006) groundbreaking study of the establishment of first- and second-degree social networks chronicled learners’ interactions with various people in the SA environment.

Learner corpora and blogs. Written data from SA students have often taken the form of learner corpora and blogs. Researchers have analyzed corpora of student reports and texts to study identity issues (Miyahira and Petrucci, 2006; Papatsiba, 2006; Russian: Pellegrino-Aveni, 2005). Other SA scholars have leveraged learner engagement with internet-based social technologies by using their blogs to study both identity (Japanese: Miyahira and Petrucci, 2006) and intercultural competence (Spanish: Elola and Oskoz, 2008). Like diaries and journals, learner blogs often provide access to learners’ perceptions of themselves and members of the target culture as well as their reactions to experiences in the SA setting. However, the public nature of blogs often invites commentary and interaction with those who read them (e.g., native speakers of the target language or other learners).

Oral interviews and roleplays. In addition to measuring linguistic gain, the ACTFL OPI has been used frequently to gather information about changes in learners’ self-perceptions of identity, as well as adjustments in their attitudes and motivation during the SA experience (Huebner, 1995; Kinginger, 2008; Pellegrino-Aveni, 2005; Polanyi, 1995). A modified version of the OPI, the Simulated Oral Proficiency Interview (SOPI), was used by Isabelli-García (2006) to investigate the influence of social networks on L2 acquisition in an SA context.

In addition to the OPI, several SA scholars have used ethnographic, sociolinguistic, and open interviews to study learner reactions to issues of gender, race, identity, attitudes, motivation, and sociolinguistic/pragmatic competence (Brecht and Robinson, 1995; Douglass, 2007; Iino, 2006; Isabelli-García, 2006; Kinginger, 2008; Miyahira and Petrucci, 2006; Pellegrino-Aveni, 2005; Shively, 2008; Siegal, 1995; Talburt and Stewart, 1999; Twombly, 1995). Focus group interviews involving the interchange of ideas among SA students have sometimes been used to complement individual interviews in the study of identity, race, and gender (Miyahira and Petrucci, 2006; Talburt and Stewart, 1999). Alred and Byram (2006) used semi-structured oral interviews with British L2 French-speaking SA alumni to gauge the long-term effects of SA ten years after the experience. To get a better picture of the context in which SA learners operate, scholars have interviewed faculty, host families, and program coordinators (Iino, 2006; Knight and Schmidt-Rinehart, 2002; Shively, 2008; Talburt and Stewart, 1999). Roleplays have also been used to study sociolinguistic competence (Kinginger, 2008) and communication strategies (Lafford, 1995, 2004).

Observations and field notes. Several SA studies complemented the use of other instruments with observations and field notes. Joining the learners in various activities, some researchers used a participant observation approach (Shively, 2008) whereas other investigators served as outside observers (Kinginger, 2008; Talburt and Stewart, 1999). Studies of L2 learners in Japan have also videotaped or audiotaped SA learners in conversations with host families and friends to study the joint construction of folk beliefs (Cook, 2006), norms of interaction with homestay families (Iino, 2006), and gender roles (Siegal, 1995).

To get a broader picture of the language learning experiences, scholars have observed students as they interact in classroom situations in the SA context. For instance, investigators have examined the value of formal instruction (Brecht and Robinson, 1995) and the nature of native speaker-non-native speaker (NS-NNS) interactions (McMeekin, 2006) using data collected in SA classroom settings. Pellegrino-Aveni (2005) observed SA classroom interaction to gain insights into the strategies learners used to manage their language use and to study the effects of the observation process on learners’ interactions. Wilkinson (2002) noted that learners brought SA classroom-type interaction into their conversation with host families.

Data analysis

Quantitative, qualitative, and a mixed method approach have all been used in SA research. In SA research quantitative approaches have been used mostly to investigate possible gains in linguistic outcomes (pronunciation [Simoes, 1996], fluency [Segalowitz and Freed, 2004], grammar [Collentine, 2004], lexical abilities [Collentine, 2004; Kinginger, 2008; Lafford and Ryan, 1995], and communication strategies [DeKeyser, 1991; Lafford, 1995, 2004]). These studies often involve a substantial number (25–100) of university undergraduates (mostly aged 18–21) in order to make generalizations from the analysis. In addition, scholars using a quantitative approach normally use parametric measures such as ANOVAs, Repeated Measures, and correlations (Pearson's r) to test for significant changes in SA learners’ linguistic performance over time and for significant correlations of linguistic gain with learner attitudes or amount of language contact. Although there are several advantages of quantitative research (systematicity, generalizability, relatively short time for statistical analysis, prestige factor), this decontextualized mode of inquiry averages responses from the entire group of subjects, thus obscuring individual differences and eliminating possible explanatory social and environmental factors that could be captured using a qualitative case-study approach.

Even though the first in-depth qualitative case studies of SLA in a SA context appeared in the early to mid-1990s (e.g., DeKeyser, 1991; Polanyi, 1995; Siegal, 1995) SA qualitative research has gained notable momentum only recently. Most recent case studies of the social aspects of SLA in a SA setting have been carried out on data from learners of Russian (Pellegrino-Aveni, 2005; Polanyi, 1995), Japanese (Cook, 2006; Miyahira and Petrucci, 2006; Siegal, 1995), French (Kinginger, 2008), and Spanish (DeKeyser, 1991; Isabelli-García, 2006). These studies typically investigate data from a small number of subjects (four–six), which limits the generalizability of their findings, but allows more opportunity for scholars to carry out “thick” descriptions of the complex interrelationships of various social, cognitive, and situational factors.

Qualitative analyses provide insights into an individual's situated experience in a given environment. To understand the complexity of language learning in SA environments, qualitative SA studies almost always triangulate data from various instruments (interviews, observations/field notes, journals/ diaries), and occasionally also with quantitative data. For instance, Kinginger (2008) was the first SA study done in a sociocultural theoretical framework that tied quantitatively-measured linguistic outcomes (e.g., TFI scores, Language Awareness Interview) to social and contextual factors (e.g., attitudes, self-perceptions, social networks) investigated with the use of qualitative data (e.g., interviews, journals/diaries, observations, roleplays). The triangulation of these different types of data provides insights into the diverse factors that can lead to notable differences in individual linguistic achievement during a sojourn abroad.

This type of mixed methods approach, in which several data sources are analyzed and the results are triangulated to strengthen the validity of the findings, has been used primarily to investigate social factors including identity, race, and gender motivation, attitude, learning styles and strategies, and intercultural competence (e.g., Allen et al., 2006; DeKeyser, 1990; Douglass, 2007; Elola and Oskoz, 2008; Hokanson, 2000; Iino, 2006; Isabelli-García, 2006; Jones and Bond, 2000; Kinginger, 2008; Miyahira and Petrucci, 2006; Pellegrino-Aveni, 2005; Siegal, 1995; Talburt and Stewart, 1999; Twombly, 1995; Wilkinson, 1998).

Empirical verification: Language gain

There is now substantial empirical evidence that learners improve their linguistic and socio-linguistic abilities through studying abroad. In recent reviews, Collentine and Freed (2004) and Churchill and DuFon (2006) emphasized the breadth of learning demonstrated: improvement in pronunciation and fluency; gain in vocabulary and grammatical accuracy; enhancement in interaction skills, including listening, speaking, and narrative ability; increases in literacy skills; development of pragmatic abilities, including politeness strategies, routines, terms of address, and manipulation of speech acts and registers; intercultural sensitivity; and the development of learning and communicative strategies. These gains vary by language, learning situation, and the learners themselves. Gains often appear in a U-shaped progression, where learners make marked improvement early, appear to stop progressing for a while, and then resume a rapid trajectory of improvement (Pearson-Evans, 2006).

Juxtaposing this language gain against learning in the classroom, research generally shows an advantage for the abroad experience; however, this advantage does not always hold when SA learners are compared with students in domestic immersion settings (Collentine and Freed, 2004, special issue Studies in Second Language Acquisition). The editors explained that these studies “provide no evidence that one context of learning is uniformly superior to another for all students, at all levels of language learning, and for all language skills” (p. 164).

Most of these studies, however, do not examine individual or social factors in tandem with linguistic gain. Thus, the disconnect is striking between the two questions, What do learners gain? and What factors influence that gain?

Empirical verification: Individual and social factors

More recent studies associated with the social turn in SLA, which have relied heavily on qualitative methodological approaches, have provided insight into the individual nature of learning abroad, and the complexity of the social situations in which it occurs. These findings help researchers nuance empirical findings on linguistic gain and probe into long-held assumptions regarding programmatic and individual variables.

Programmatic variables. As Churchill and DuFon (2006) pointed out, the design of SA programs has an explicit—and sometimes unintended—impact on the learner's experience. Of particular interest are the effects of program length, living conditions, and social networks, both with the host community and with compatriots on the program or from home.

Length of program. Although “the research of the effect of program length is relatively scare and inconclusive” (Churchill and DuFon, 2006, p. 23), studies that compared gain over time revealed greater gains for longer stays (for Spanish grammar and/or vocabulary: Ife et al., 2000; Isabelli, 2004; for French pragmatics: Hoffman-Hicks, 1999). Dwyer's (2004) 50-year retrospective of SA reported gains in increased confidence and commitment to language study that favored the year and even the summer program, over the semester program. Other studies have shown learners improving aspects of their language skills in programs of a few weeks (for listening: Cubillos et al., 2008; for pronunciation: Simoes, 1996; for self-perceived ability: Cubillos et al., 2008; for intercultural and personal growth: Jones and Bond, 2000). SA studies typically make conclusions about the effect of length of stay using a cross-sectional approach to compare the outcomes of different groups of students participating in programs of varying lengths. In order to understand more fully the effects of time on linguistic abilities in a SA context, more longitudinal research is needed to measure the effects of time spent abroad on the linguistic abilities of given groups at various time intervals in specific contexts.

Participants’ language learning backgrounds. The relationship between program length and language gain may be sensitive to the language learning background or the initial proficiency level of the learner. The level of linguistic background needed for success has been studied as the threshold hypothesis (cf., Lafford, 2006). Regan (2003) reviewed findings to suggest that lower-level students make the most advances because they have the most to gain, a finding recognized by Churchill and DuFon (2006) with the caveat that advanced learners may enter more readily than beginning learners into contact situations that facilitate language development. Providing evidence to support this contention, Magnan and Back (2007) demonstrated that students with more course work, but the same ACTFL level, made greater gains in speaking proficiency than students with less coursework. They speculated that the increased coursework might have prepared the learners to benefit from aural and written input and given them more confidence, all of which drove their social networking, leading to their linguistic gain (cf., Churchill and DuFon, 2006). This suggestion recalls the finding of Brecht et al. (1995) that preprogram reading and grammar skills could predict success abroad.

Initial proficiency level also influences how students learn. Students at lower proficiency levels appear to benefit more from social, especially oral, interaction (Freed, 1990) than from other types of input. In contrast, students at upper levels may profit more than students at lower levels from involvement with written and aural media (Freed, 1990) and from sociolinguistic (Kasper, 1996) and cultural (Pearson-Evans, 2006) aspects of the language.

Living conditions. For years, the home stay was considered the sine quo non recommendation because, according to the Contact Hypothesis, it provided the social and cultural interaction conducive to language learning (Allen et al., 2006; Knight and Schmidt-Rinehart, 2002; Rivers, 1998). In Schmidt-Rinehart and Knight (2004) and Knight and Schmidt-Rinehart (2002) host mothers assisted American students linguistically, culturally, and psychologically. Other studies have found that interaction with host families has helped increase cultural understanding (Cook, 2006) and learner confidence (Magnan and Back, 2007). Iino (2006) illustrated how conversation in the home can be particularly useful in cultures where different linguistic norms are expected of foreigners (e.g., Japan); however, as Iino pointed out, the home stay also complicates learning foreigner patterns because there is typically no foreigner linguistic model in the household other than the learner him or herself.

Recent ethnographic research, however, points to problems with home stays, such as those described poignantly by Kinginger (2008) and Pellegrino-Aveni (2005). According to Knight and Schmidt-Rinehart (2002), the major problem areas were incompatibility of personalities, lack of interaction, interaction patterns that are not beneficial to language acquisition, or highly stressful living situations.

Program length may also play a role in students’ relationships with host families. For example, Schmidt-Rinehart and Knight (2004) and Knight and Schmidt-Rinehart (2002) revealed that students had fewer problems with home stay when they were on longer programs, which might be explained by both the commitment made between students and host families and the time available to work through cultural differences.

Quantitative results also question the superiority of home stays. In a large study, Rivers (1998) found that dormitory students gained more in their Russian language proficiency than home stay students. Magnan and Back (2007) found no significant difference between American students in France who lived with French families or apartment-mates and those who lived in international dormitories or apartments alone or with other Americans. In contrast to these findings, Allen et al. (2006) found that their home stay participants reported significantly higher target language linguistic abilities than their non-home-stay peers. These findings may be influenced by the learners’ initial proficiency level, however. In all three studies, the group making the greatest gain had either higher initial proficiency levels or more language course work, a variable that might be more influential than living condition.

Social networks and language contact. Beyond the family environment, research has recently considered other social networks, which include three types of communities: those composed primarily of host culture natives, those composed of fellow nationals on the program, and those made virtually with friends and family at home. Lybeck (2002) explained that the first is vital to the SA experience because learners who engage in exchange networks with native speakers feel connected in the host culture. Providing examples of students participating in sports clubs, church groups, community service, and other curricular activities, Kinginger (2008) and Isabelli-Garcia (2006) attested to how students with such networks made the most gain in language ability and cultural insight.

The opposite picture occurs when the social network is with members of the home culture. Magnan and Back (2007) found a negative correlation between language gain and contact with fellow Americans. Research demonstrates how learners often regret they did not interact more with members of the host country because they spent time with the compatriot group (Kinginger, 2008; Wilkinson, 1998).

A social network with people from home is formed increasingly via technology (Pearson-Evans, 2006). In several studies, students who remained connected with friends and family at home through email, English music on iPods, blogs, cell phones, IMs, and internet social networks such as Facebook (eg., Kinginger, 2008; Knight and Schmidt-Rinehart, 2002) demonstrated fewer linguistic gains and less satisfaction with their abroad experience than others who relied less on these virtual contacts. In addition, people from home visit and travel with learners, creating a space where the native language is used, and generally taking the learner from the target culture (e.g., Kinginger's Liza).

A few researchers have pointed to positive effects of technological connections. Pearson-Evans (2006) and Papatsiba (2006) explained how virtual attachments help learners through initial adjustments abroad, build confidence, and reduce loneliness. Elola and Oskoz (2008) demonstrated how blog exchanges between SA students in Spain and classroom students in the USA fostered intercultural competence in both groups.

Contact through media. Many learners mention how they used media (e.g., radio, television, newspapers) to improve their language abilities and gain cultural insights (Kinginger, 2008; Pellegrino-Aveni, 2005). Because media provides authentic input, which should feed language learning, it is surprising that Magnan and Back (2007), using the LCP, did not find a significant relationship between improvement in speaking proficiency and amount of contact with aural and written media. Also using the LCP, Segalowitz and Freed (2004) suggested that learners’ initial oral abilities played a role in determining the learners’ extracurricular L2 contact. Individual factors (initial language ability and individual learner difference) interact with situational factors (access to media) to affect language gain.

Individual variables. Learners’ personal backgrounds and preferences influence how they approach the SA experience, and are also amplified by it (Kinginger, 2008). The most common individual variables treated in research are identity and gender, motivation, attitude, and learning and communicative styles and strategies.

Identity and gender. Identity has many dimensions related to internal factors of an individual, such as race and physical characteristics, gender, age, social class, personal motivations, goals and attitudes, personal history, and even attractiveness of personality, all of which affect a learner's access and reception in foreign communities (Churchill and DuFon, 2006; Iino, 2006; PellegrinoAveni, 2005). For example, Miyahira and Petrucci (2006) demonstrated how Okinawan heritage students from Brazil and Peru were expected to speak and act like host nationals and met with disapproval when they did not.

Identity is responsive to external factors, such as expectations of the learner by members of the host society. Taking a post-structural perspective, many studies (e.g., Kinginger, 2008; PellegrinoAveni, 2005) examined how perceived power relationships evolve between learners, peers, and natives, and how these relationships influence learners’ willingness, and ability, to communicate in the L2.

The role of gender seems particularly influential for women during SA. Siegal's(1995)American women studying in Japan resisted learning honorifics because they perceived a conflict between their self-images as powerful women and the Japanese view of humility. Polanyi (1995), Twombly (1995), and Talburt and Stewart (1999) recounted the alienating experiences of American women in Russia, Costa Rica, and Spain, respectively, who expressed discomfort with ritual flirtation and sexual remarks made by host culture men. Gender issues also affect males, who have been recorded as assuming a protector role for their female compatriots (e.g., Kinginger's [2008] Bill) or as avoiding social male relationships (Isabelli-García, 2006) because they did not want to be associated with male target culture behavior.

The dynamics of gender and cultural behavior may help explain a discrepancy between the linguistic achievement of males and females. Brecht et al. (1995) and Polanyi (1995) suggested that men may make greater linguistics gains abroad than women, because, in line with the Contact Hypothesis, women have less access to meaningful social interaction (Kinginger, 2008).

Pellegrino-Aveni (2005) studied the interplay of internal and external identity factors in terms of learners’ perceptions of their ideal self and their actual self. When there is too great a contrast between the ideal self and the actual self the learner's self-image is destabilized. Researchers have found explanations for this destabilization in discourses of power (Gore, 2005) and sociocultural approaches and activity theory (Kinginger, 2008).

Differences between images of the ideal self and the actual self are strongly related to learners’ motivation and goals for SA. Learners with a history of doing well in language classes and whose academic success is important to their self-esteem may be frustrated or even depressed when expectations are too high (e.g., Douglass’ [2007] Claire; Pellegrino-Aveni's [2005] Rebeccah). Others may feel infantilized, such as Pellegrino-Aveni's Bob, who objected to classroom games, or Jim, when his host father persisted in explaining to him how to boil water.

As a social mirror of self, identity is not monolithic. Kinginger (2008) demonstrated how learners and hosts co-created new identities daily: changes in social situations led to evolution in learners’ self concepts, which, in turn, led to more L2 use, and to a further change in their identities. For example, the behavior of Pellegrino-Aveni's (2005) Bob was influenced by his self-comparison to compatriot peers. Because they did not understand social rituals and hierarchies, Pearson-Evans’ (2006) Lucy rejected initiatives of her host family, and Kinginger's (2008) Beatrice missed rich points in cultural interaction.

Goals and motivation. Learners go through stages of adjustment leading toward the convergence of the ideal and actual self. Part of this adjustment is often a shift in motivation for language study and in goals for the SA experience. For example, Pellegrino-Aveni's (2005) Rebeccah, who first did not want to be perceived as a foreigner, later used the foreigner personae to foster interaction with native hosts.

Gore (2005) classified goals for SA into two types of discourse: (a) a dominant Grand Tour discourse that is particularly prevalent among women who view SA as a finishing step in a liberal education, and (b) an alternate discourse, which covers a wide variety of goals including inter-cultural competence and academic advantage. Research (e.g., Kinginger, 2008) has shown that, although most students express their goals in terms of the alternate discourse as they begin their SA programs, many of them change to a Grand Tour approach when they become frustrated with language learning and cultural differences that threaten the self. Students’ goals also relate to the personal backgrounds they bring to the SA experience. For example, Kinginger's Deidre's lack of travel experience and goal of finishing off her French studies encouraged her to take a Grand Tour approach and to not invest in language learning, whereas Louis's previous reading of the French author Céline and family travel fostered his alternate discourse approach and heavy investment in class work and community service.

Attitude. Attitude change appears to be an inherent part of the abroad experience. Although research has demonstrated that SA does not always bring tolerance or understanding of other cultures (Papatsiba, 2006), Allen et al. (2006) demonstrated how, after only an eight-week summer session in France, Italy, or Spain, high school students significantly decreased their identification with their native cultures and significantly increased their identification with the target culture.

Learners also increase their appreciation of their home country, sometimes to the point of becoming its defender (Alred and Byram, 2006; Papatsiba, 2006). In the case of heritage learners returning to the ancestral homeland, SA affects how they view their identity in relation to the diaspora (Miyahira and Petrucci, 2006).

It may be the attitude toward others and the self that evolves the most during the SA experience. Learners have reported less fear of new situations, greater comfort with inabilities, increased self-awareness, and more confidence (Alred and Byram, 2006; Magnan and Back, 2007; Pellegrino-Aveni, 2005), and a better ability to manage a variety of sociocultural tasks (Allen et al., 2006). In fact, Kinginger (2008) questioned the common assumption that SA is about language learning and even that all participants are language learners. Gore (2005) contended that identity development is now a major outcome of the SA experience.

Personalities, cognitive styles, and strategies. Research in SA, as well as in SLA generally, is increasing focus on personality and learning styles and particularly on learning and communication strategies. Bakalis and Joiner (2004) argued that students who have personalities characterized by receptivity to diversity and change are more likely to participate in a SA program than other students. Once abroad, Knight and Schmidt-Rinehart (2002) discussed how personality type was related to problems with home stays, noting difficulties when students were shy, hesitant to interact, or were perceived as non-communicators. In fact, most host families in their study felt that personality was more important in adjusting to the home stay environment than language ability.

Hokanson (2000) used cognitive styles to relate the activity preferences of American students in Guatemala to language acquisition. She found that different types of learners gravitated toward different activities: extroverts spent time with instructors exploring markets, museums, and cafes, while introverts studied mostly at school. After one month, the extroverts were more fluent in oral production, and the introverts were more prolific and expressive in writing. Such activity choices can become purposeful learning strategies, as demonstrated by Kinginger (2008) and PellegrinoAveni (2005). In addition to social choices, research has considered learners’ use of metalinguistic strategies, such as self-monitoring and repair (Lafford, 1995), as ways to build language proficiency.

Personal background can also affect the choice and use of learning strategies. Adams (2006) found that previous instruction in mnemonic devices had a significant effect on the use of memory strategies at the beginning of the SA period, a difference that was not maintained throughout the abroad experience. Preferences for different learning strategies are related to other factors of identity, especially to gender, as shown by Adams (2006) who found that males used more cognitive strategies than females throughout the abroad program. Lafford (1995) showed that the SA experience broadens the repertoire of communicative strategies of L2 learners. Dekeyser (1990) and Guntermann (1992a, b) found that language learners in Spain and Latin America, respectively, simply avoided using grammatical structures they did not control.

Many authors have suggested instructing students in learning and communicative strategies as a way to enhance their language acquisition while abroad (e.g., Adams, 2006; Magnan and Back, 2007). Exploring curricular interventions with a training manual (Paige et al., 2002), Paige et al. (2004) and Cohen and Shively (2007) found that SA students increased their strategy use in many areas (speaking, listening, pragmatic performance in making requests and apologies, non-verbal communication strategies, interpreting culture, and coping with cultural shock), but were unable to relate this growth statistically to the curricular intervention.

In summary, empirical findings have verified common thinking—the SA environment is rich with opportunity for linguistic and cultural learning and for self-growth. More important, however, is that this empirical verification has demonstrated the complexity of the experience, and how it essentially depends on individual and social factors that shape interaction. What continues to elude researchers is a neat relationship between outcome gains and individual experience. Experience responds to individual volition as well as to situational constraints, and, as such, is highly mutable and fluid.

Applications

It is thus no small challenge for practitioners to draw applications from SA research, either for individuals or for programs at home or abroad. The SA research reviewed here could imply the following to facilitate student linguistic success abroad: (a) require a basic knowledge of the target language before studying abroad including the use of authentic oral and written materials to build language and intercultural competence (threshold hypothesis); (b) encourage students to study abroad for longer periods of time; (c) set up email exchanges and opportunities for prospective SA students to engage in social computing networks with their future host families to facilitate the establishment of first- and second-level social networks before they arrive on site; (d) provide extensive pre-departure training for SA learners in which they learn about the target language culture, L2 pragmatics, and strategies for learning and communicating in the L2 abroad; (e) require students to live in an interactive home stay situation (without other English speakers) or with native speakers of the target language in other situations; (f) encourage participation in service learning and internships in which learners can use their L2 naturally with NSs of the target language; (g) discourage contacts among co-nationals such as those made through in-country orientations, group trips, and common courses; and (h) discourage virtual connections and extended visits with people back home.

These recommendations are primarily to foster language learning and intercultural competence. However, recent research on SA students (Gore, 2005) has shown that most students who spend time abroad are not language majors and may not even have language learning as their primary goal. The recognition that different learners have different personal goals should make programmatic decision-makers wary of a “one program fits all” approach. As identity growth becomes a more recognized, and valued, outcome of SA, and as individual differences are better understood, it will become increasingly possible to match students with programs developed toward a range of learning goals.

Future directions

Traditional SA research in the 1990s focused heavily on linguistic gains in a SA context (What do sojourners learn?). Recent trends in SA research have shown a strong interest in exploring how social and contextual variables (e.g., length of program, living situation, social networks) interact with individual learner factors (e.g., identity, attitudes, gender, and race issues) during SA. We now need to explore systematically the impact of the interaction of these sociocontextual factors on linguistic gains in a SA context (What factors facilitate learning?). Using as a model such studies as Kinginger (2008) with its quantitative and qualitative dimensions investigating the experience of the same SA students, this impact needs to be front and center on the SA research agenda in both the short and long term. It is only when this tie is made between linguistic abilities and individual and contextual factors that the field of SA research can make a more solid contribution to the field of SLA as a whole.

The preceding review of SA research has pointed to the need for more studies that would give investigators insight into the complex factors contributing to changes in learner performance and perceptions during a SA experience. Research questions could include the following:

  • •    How does the establishment and maintenance of various types of first- and second-degree social networks with native speakers by SA learners affect their linguistic progress and attitudes toward the target culture?
  • •    In homestay situations, how does the composition of the family, the number of students in a home, the orientation for both students and host families, and family interaction affect learners’ linguistic gains, attitudes, and perception of self?
  • •    What role do electronic communication pathways (e.g., Skyping, texting, emailing, blogging, Facebooking, Twittering) play in determining or facilitating the amount and type of contact SA learners have with NSs of the target language abroad and with speakers of the learners’ native language at home or in the SA environment?
  • •    What types of interactions and feedback do learners experience in formal classroom environments in the host country taught by local or home institution faculty? How do these interactions compare to those in classroom settings in the learners’ home institutions?
  • •   What factors in the individual predispose a learner to a successful SA experience? To what degree are those factors responsive to instructional intervention either before or during the SA experience?

In addition to these areas of research, there remain pressing questions about what different methodologies can bring to our understanding.

  • •   What types of insights can be provided by observations of interactions between learners and host families, instructors, and program administrators, or by interviews with these same individuals?
  • •   What new insights can be gained by using concordance programs to analyze learner corpora of oral texts in order to understand interaction patterns between NS-NNS and NS-NS, and also between NS-NNS and NNS-NNS individuals?

The current social turn in SLA research encourages investigation of these questions, especially within qualitative research frameworks. As we move forward, we need to blend the interactionist focus on linguistic outcomes, with its frequent quantitative analyses capable of making generalizations, with the in-depth qualitative examination of individual experience. These two routes to understanding the SA experience may appear incompatible: the individual nature may always defy group analysis. Nonetheless, by investigating the relationship between the questions “What do sojourners learn?” and “What factors influence that learning?” we are enriching our understanding of the complexity of the immersion experience in a SA setting.

Notes

* We would like to recognize and thank Seungyeon Lee and Michelle Petersen for their work with Barbara Lafford and Jacques Arceneaux and Alice Astarita and for their work with Sally Magnan. These four graduate students offered an essential contribution to this article through their careful bibliographic work.

References

Adams, R. (2006). Language learning strategies in the SA context. In M. A. DuFon and E. Churchill (Eds.), Language learners in SA contexts (pp. 259–92). Clevedon, UK: Multilingual Matters.

Agar, M. (1996). The professional stranger: An informal introduction to ethnography (Second Edition). New York: Academic Press.

Allen, H., Dristas, V., and Mills, N. (2006). Cultural learning outcomes and summer SA. In M. Mantero (Ed.), Identity and second language learning: Culture, inquiry, and dialogic activity in educational contexts (pp. 187–214). Charlotte, NC: Information Age Publishing.

Alred, G. and Byram, M. (2006). British students in France: 10 years on. In M. Byram and A. Feng (Eds.), Living and studying abroad (pp. 210–231). Clevedon, UK: Multilingual Matters.

Bakalis, S. and Joiner, T. (2004). Participation in tertiary SA programs: The role of personality. International Journal of Educational Management, 18, 286–291.

Barron, A. (2006). Learning to say “you” in German: The acquisition of sociolinguistic competence in a SA context. In M. A. DuFon and E. Churchill (Eds.), Language learners in SA contexts (pp. 59–88). Clevedon, UK: Multilingual Matters.

Block, D. (2003). The social turn in second language acquisition. Edinburgh: Edinburgh University Press.

Brecht, R., Davidson, D., and Ginsberg, R. (1995). Predictors of foreign language gain during SA. In B. F. Freed (Ed.), Second language acquisition in a SA context (pp. 37–66). Amsterdam: Benjamins.

Brecht, R. and Robinson, J. L. (1995). On the value of formal instruction in SA: Student reactions in context. In B. Freed (Ed.), Second language acquisition in a SA context (pp. 317–334). Philadelphia, PA: John Benjamins.

Chieffo, L. and Griffiths, L. (2004). Large scale assessment of student attitudes after a short-term study-abroad program. Frontiers, 10, 165–178.

Churchill, E. and DuFon, M. (2006). Evolving threads in SA research. In M. DuFon and E. Churchill (Eds.), Language learners in SA contexts (pp. 1–27). Clevedon, UK: Multilingual Matters.

Cohen, A. D., and Shively, R. L. (2007). Acquisition of requests and apologies in Spanish and French: Impact of SA and strategy-building intervention. The Modern Language Journal, 91(2), 189–212.

Collentine, J. (2004). The effects of learning contexts on morphosyntactic and lexical development. Studies in Second Language Acquisition, 26, 227–248.

Collentine, J. and Freed, B. (Ed.) (2004). Studies in Second Language Acquisition, 26, 2 [Special issue].

Cook, H. (2006). Joint construction of folk beliefs by JFL learners and Japanese host families. In M. A. Dufon and E. Churchill (Eds.), Language learners in SA contexts (pp. 120–173). Clevedon, UK: Multilingual matters.

Cubillos, J., Chieffo, L., and Fan, C. (2008). The impact of short-term SA programs on L2 listening comprehension. Foreign Language Annals, 41, 157–185.

Davidson, D. (2010). Study abroad: When, how long, and with what results? New data from the Russian front. Foreign Language Annals, 43, 6–26.

DeKeyser, R. M. (1990). From learning to acquisition? Monitoring in the classroom and abroad. Hispania, 70, 238–247.

DeKeyser, R. (1991). Foreign language development during a semester abroad. In B. F. Freed (Ed.), Foreign language acquisition research and the classroom (pp. 104–199). Lexington, MA: D.C. Heath.

Douglass, K. B. (2007). Climbing the Eiffel Tower: An activity theoretic analysis of motives in an individual learner of French. Unpublished doctoral dissertation. Pennsylvania State University, University Park.

Dwyer, M. (2004). More is better: The impact of SA program duration. Frontiers, 10, 151–164.

Elola, I. and Oskoz, A. (2008). Blogging: Fostering intercultural competence development in foreign language and study abroad contexts. Foreign Language Annals, 41, 454–477.

Freed, B. (1990). Language learning in a SA context: The effects of interactive and non-interactive out of class contact on grammatical achievement and oral proficiency. Linguistics, language teaching and language acquisition. Georgetown University Round Table on Languages and Linguistics.

Freed, B. (1995). Introduction. In B. Freed (Ed.), Second language acquisition in a study abroad context (pp. 3–34). Philadelphia: John Benjamins.

Gore, J. (2005). Dominant beliefs and alternate voices. New York: Routledge.

Guntermann, G. (1992a). An analysis of interlanguage development over time: Part I, por and para. Hispania, 75, 177–187. http://www.jstor.org.ezproxy1.lib.asu.edu/stable/pdfplus/344777.pdf.

Guntermann, G. (1992b). An analysis of interlanguage development over time: Part II. ser and estar. Hispania, 75, 1294–1303.

Hoffman-Hicks, S. (1999). The longitudinal development of French foreign language pragmatic competence: Evidence from SA. Unpublished doctoral dissertation. Indiana University, Bloomington.

Hokanson, S. (2000). Foreign language immersion homestays: Maximizing the accommodation of cognitive styles. Applied Language Learning, 11, 239–264.

Huebner, T. (1995). The effects of overseas language programs: Report of a case study of an intensive Japanese course. In B. Freed (Ed.), Second language acquisition in a SA context (pp. 171–193). Philadelphia: John Benjamins.

Ife, A., Vives Boix, G., and Meara, P. (2000). The impact of SA on the vocabulary development of different proficiency groups. Spanish Applied Linguistics, 4, 55–84.

Iino, M. (2006). Norms of interaction in a Japanese homestay setting: Toward a two-way flow of linguistic and cultural resources. In M. DuFon and E. Churchill (Eds.), Language Learners in SA Contexts (pp. 151–173). Clevedon, UK: Multilingual Matters.

Isabelli, C. A. (2004). The acquisition of null subject parameter properties in SLA: Some effects of positive evidence in a natural learning context. Hispania, 87(1), 150–162.

Isabelli-García, C. (2006). SA social networks, motivation, and attitudes: Implications for SLA. In M. DuFon and E. Churchill (Eds.), Language learners in SA contexts (pp. 231–258). Clevedon, UK: Multilingual Matters.

Johnson, M. and Tyler, A. (1998). Re-analyzing the OPI: How much does it look like natural conversation? In R. Young and A. He (Eds.), Talking and testing (pp. 28–47). Amsterdam: John Benjamins.

Jones, M. and Bond, M. (2000). Personal adjustment, language acquisition, and cultural learning in short-term cultural immersion. International Review, 10, 33–49.

Kasper, G. (1996). The development of pragmatic competence. In E. Weltens, and T. Bongaerts (Eds.), Eurosla (Vol. 6, pp. 103–120). Amsterdam: UV Uitgeverij.

Kinginger, C. (2008). Language learning in SA: Case studies of Americans in France. The Modern Language Journal, 92, Monograph. Oxford: Blackwell.

Knight, S. and Schmidt-Rinehart, B. (2002). Enhancing the homestay: SA from the host family's perspective. Foreign Language Annals, 35, 190–201.

Lafford, B. (1995). Getting into, through and out of a survival situation: A comparison of communicative strategies used by students studying Spanish abroad and “at home”. In B. Freed (Ed.), Second language acquisition in a SA context (pp. 97–121). Amsterdam: Benjamins.

Lafford, B. (2004). The effect of context of learning on the use of communication strategies by learners of Spanish as a second language. Studies in Second Language Acquisition, 26(2), 201–226.

Lafford, B. (2006). The effects of study abroad vs. classroom contexts on Spanish SLA: Old assumptions, new insights and future research directions. In C. Klee and T. Face (Eds.), Selected proceedings of the 7th conference on the acquisition of Spanish and Portuguese as first and second languages (pp. 1–25). Somerville, MA: Cascadilla Proceedings Project.

Lafford, B. (2010). Variation in the acquisition of Spanish L2 pragmatic competence “at home” and abroad: Context, methods, and outcomes. Presentation given at the annual meeting of the Hispanic Linguistics Syposium. Indiana University, October, 2010.

Lafford, B. A. and Ryan, J. M. (1995). The acquisition of lexical meaning in a SA context: The Spanish prepositions por and para. Hispania, 78(3), 528–547.

Lybeck, K. (2002). Cultural identification and second language pronunciation of Americans in Norway. The Modern Language Journal, 86, 174–191.

Magnan, S. and Back, M. (2007). Social interaction and linguistic gain during SA. Foreign Language Annals, 40, 43–61.

McMeekin, A. (2006). Negotiation in a Japanese SA setting. In M. A. DuFon and E. Churchill (Eds.), Language learners in SA contexts (pp. 177–202). Clevedon, UK: Multilingual Matters.

Miyahira, K. and Petrucci, P. (2006). Going home to Okinawa: Perspectives on heritage language speakers studying in the ancestral homeland. In M. Mantero (Ed.), Identity and second learning (pp. 257–282). United States: Information Age Publishing.

Paige, R., Cohen, A., Kappler, B., Chi, J., and Lassegard, J. (2002). Maximizing SA. Minneapolis, MN: CARLA.

Paige, R., Cohen, A., and Shively, R. (2004). Assessing the impact of a strategies-based curriculum on language and culture learning abroad. Frontiers, 10, 253–276.

Papatsiba, V. (2006). SA and experiences of cultural distance and proximity: French Erasmus students. In M. Byram and A. Feng (Eds.), Living and studying abroad (pp. 108–133). Clevedon, UK: Multilingual Matters.

Pearson-Evans, A. (2006). Recording the journey: Diaries of Irish students in Japan. In M. Byram and A. Feng (Eds.), Living and studying abroad (pp. 38–64). Clevedon, UK: Multilingual Matters.

Pellegrino-Aveni, V. (2005). SA and second language use: Constructing the self. Cambridge: Cambridge University Press.

Polanyi, L. (1995). Language learning and living abroad: Stories from the field. In B. Freed (Ed.), Second language acquisition in a SA context (pp. 271–291). Amsterdam: Benjamins.

Regan, V. (2003). Sociolinguistics and language learning in a SA context. Frontiers, 4, 61–90.

Rivers, W. (1998). Is being there enough? The effects of homestay placements on language gain during SA. Foreign Language Annals, 31, 492–500.

Schell, K. (2001). Functional categories and the acquisition of aspect in L2 Spanish: A longitudinal study. Unpublished doctoral dissertation. University of Washington, Seattle.

Schmidt-Rinehart, B. and Knight, S. (2004). The homestay component of SA: Three perspectives. Foreign Language Annals, 37, 254–262.

Segalowitz, N. and Freed, B. (2004). Context, contact, and cognition in oral fluency acquisition: Learning Spanish in at home and SA contexts. Studies in Second Language Acquisition, 26, 173–199.

Shively, R. (2008). Politeness and social interaction in SA: Service encounters in L2 Spanish. Unpublished doctoral dissertation. University of Minnesota, Twin Cities.

Siegal, M. (1995). Individual differences and SA: Women learning Japanese in Japan. In B. Freed (Ed.), Second language acquisition in a SA context (pp. 225–244). Amsterdam: Benjamins.

Simoes, A. (1996). Phonetics in second language acquisition: An acoustic study of fluency in adult learners of Spanish. Hispania, 79, 87–95.

Talburt, S. and Stewart, M. (1999). What's the subject of SA? Race, gender and “living culture”. The Modern Language Journal, 91, 163–175.

Twombly, S. (1995). Piropos and friendship: Gender and cultural clash in SA. Frontiers, 1, 1–27.

Wilkinson, S. (1998). SA from the participants’ perspective: A challenge to common beliefs. Foreign Language Annals, 31, 23–39.

Wilkinson, S. (2002). The omnipresent classroom during summer SA: American students in conversation with their French hosts. The Modern Language Journal, 86, 157–173.