Research on Language Learning During Study Abroad
According to Freed (1998), early research on language learning in study abroad contexts primarily used criterion-referenced tests to measure language growth. While these studies suggested a positive relationship between time spent abroad and second language acquisition, many lacked control groups and used measures that were unable to draw fine distinctions in language gains or conclusions about individual variation in results. Freed noted that later studies moved beyond exclusively test-based data to investigate the relationships between language development, student characteristics, and specific experiences abroad.
Kinginger (2011) enumerated three research trends that grew from the results of early outcomes studies that showed great differences in individual achievement after periods abroad. First, studies attempted to correlate language gains with quantitative accounting of student activities and target language use; second, ethnographies and case studies examined student perceptions of the study abroad sojourn; and most recently, researchers have pursued mixed-methods studies incorporating qualitative analysis of student behaviors and perspectives with assessment of language learning outcomes. Still, little of this in-depth analysis of the study abroad experience has included concurrent investigation of the host perspective (Kinginger, 2013b). Most previous study abroad research has also focused on learners of one language, most commonly French or Spanish (Llanes, 2011).
DuFon and Churchill's (2006) review of research findings indicated that learner engagement with the host community is a key factor in language acquisition during study abroad because the opportunities for and quality of interaction vary greatly and are mediated by both learner approaches and host culture practices. Recent studies examining study abroad outcomes have identified the need for interventions to support language development by encouraging students to increase their engagement with native speakers (Back, 2013; Cadd, 2012; Du, 2013; Goldoni, 2013; Kinginger, 2011), including homestay hosts (Knight & Schmidt-Rinehart, 2010; Martinsen, 2010; Vande Berg et al., 2009).
Oral Proficiency Gains From Study Abroad
The following studies cited involve U.S. university students unless otherwise noted. Researchers investigating oral proficiency development as a result of study abroad have frequently used such measures as the ACTFL Oral Proficiency Interview (OPI) and the Simulated Oral Proficiency Interview (SOPI). These assessments are rated according to the ACTFL Proficiency Guidelines—Speaking, a scale used to evaluate functional language ability that consists of five major levels, the lower three of which are divided into Low, Mid, and High sublevels (ACTFL, 1999). Gains in ACTFL proficiency ratings have been documented for groups of learners of French (Magnan & Back, 2007), German (Lindseth, 2010), and Spanish (Isabelli-García, 2006; Mendelson, 2004) after one semester abroad, learners of Portuguese after a six-week summer program (Milleret, 1991), learners of Japanese after an eight-week summer program (Hardison & Okuno, 2013), and high school and gap-year learners of Swedish over a year abroad (Spenader, 2011). Davidson (2010) found that OPI gains of more than 5,000 undergraduate and graduate students of Russian who participated in study abroad programs of various durations between 1994 and 2009 were strongly correlated with longer lengths of stay and showed a wide range of individual variation. Watson, Siska, and Wolfel (2013) found that 295 of 369 learners of seven target languages made gains on the OPI after semester programs in various countries.
Several studies comparing oral proficiency results for study abroad learners and control groups of similar learners at a home institution over one semester have observed differences between the groups. Freed (1995) found that greater numbers of study abroad than at-home learners of French moved more than one sublevel and crossed level boundaries in the ACTFL Guidelines. Segalowitz and Freed (2004) found significant gains in OPI ratings for a group of study abroad Spanish learners in contrast to a group at the home institution that did not show significant improvement. Hernández (2010a) also found that study abroad learners of Spanish made significantly greater gains in SOPI ratings than on-campus learners. In Vande Berg et al.'s (2009) study of more than 800 learners of seven target languages in programs of varied length, SOPI gains were significantly greater for the study abroad group; on average, these learners improved one ACTFL sublevel, while control students improved half as much.
Although the OPI and similar instruments have been widely used to investigate study abroad outcomes, critics have questioned the application of the ACTFL Guidelines to measure language learning during study abroad. Researchers noted that the Guidelines may not be sensitive enough to measure the incremental progress made by learners during their time abroad, especially for those with higher proficiency levels and in shorter-term programs (Freed, 1998; Llanes, 2011). Many studies have demonstrated that proficiency gains as measured by the Guidelines are more common for students who enter study abroad programs with lower proficiency levels (Davidson, 2010; Lindseth, 2010; Magnan & Back, 2007; Mendelson, 2004; Milleret, 1991), which may be due to the construction of the scale in which the amount of language control increases exponentially, rather than in a linear fashion, at each subsequent proficiency level.
The Homestay Experience
Research examining the relationship between the type of housing students experience during study abroad and their oral proficiency outcomes has produced mixed results. Rivers's (1998) analysis of proficiency scores from more than 1,000 undergraduate and graduate learners of Russian over 20 years found that homestay participants were less likely than those who lived in dormitories to gain in oral proficiency. Magnan and Back (2007) did not find a difference in OPI gains between French learners living with native speakers and those living with nonnatives in a semester program. While Vande Berg et al. (2009) did not find a correlation between type of housing and SOPI gains in their large-scale study, they reported an association approaching significance between homestays and greater oral proficiency gains for students of less commonly taught languages. Hernández (2010b) noted that 15 of 16 Spanish learners who made gains on the SOPI after one semester abroad lived in a homestay, while three of four who did not improve lived in apartments with nonnative speakers.
Studies investigating contact in the host home and language growth have shown surprising results that, like the mixed findings regarding housing type and oral proficiency outcomes, do not support assumptions that a homestay provides a linguistic advantage. Segalowitz and Freed (2004) found a negative correlation between time speaking with the host family and gains in length of longest turn for learners of Spanish in a semester program and suggested that interactions with members of the host family during the homestay may have been mostly short and formulaic. Working with learners in the same program, Lafford (2004) found a significant negative correlation between time speaking with the host family and use of strategies to fill communication gaps, pointing to a focus on meaning rather than on form. Martinsen (2010) found no relationship between Spanish learners' evaluations of relationships with their host families and gains on an oral skills test after a six-week summer program. Evidence that the homestay does not always provide a source of rich and pragmatically appropriate target language input can be seen in Iino's (2006) recordings of interactions at home, which demonstrated that family members used simplified language and provided limited corrective feedback to learners of Japanese in an eight-week summer program. Schmidt-Rinehart and Knight (2004) found, however, that for learners of Spanish in summer and semester programs, time spent with the host family was significantly correlated with students' belief that they had learned as much language as they had anticipated learning during the time abroad. Vande Berg et al. (2009) also reported a significant relationship between time spent with host families and SOPI gains for learners of French, German, and Spanish.
In addition to measures relating language growth to target language contact in the homestay, student perspectives on the homestay experience have been extensively reported, with a trend toward positive affective outcomes. Large-scale studies have reported that, at the conclusion of their programs, 85% of students felt comfortable with their host families and more than 90% would recommend a homestay to others (Schmidt-Rinehart & Knight, 2004, pp. 257, 260), 80% of learners thought that living with a host family was “very important” or “essential” to the improvement of their language skills (Gutel, 2007–2008, p. 177), and 64% held unequivocally positive views of the contributions of the homestay setting to their language learning (Diao et al., 2011, p. 122). Evaluations of homestay relationships and learning outcomes were not, of course, uniformly positive across study abroad participants and additionally fluctuated over time as students navigated the study abroad experience (Diao et al., 2011). Ethnographic and case studies have documented individuals' negative attitudes toward the homestay placement, including feelings of discomfort, isolation, and disappointment with limited interaction (Allen, 2010; Kinginger, 2008; O'Donnell, 2004; Pellegrino Aveni, 2005; Wilkinson, 1998) as well as positive feelings of integration and comfort (Spenader, 2011) and appreciation of the host family as a key point of access to language practice and social networks (Castañeda & Zirger, 2011).
The host family perspective on the homestay experience has been much less discussed in the literature. Knight and Schmidt-Rinehart (2002) interviewed 24 host families in Spain and Mexico and found that, while all considered the family to be a valuable linguistic resource for students, many mentioned individual student characteristics as factors that limited interaction and reported that they would not push students who were reluctant to participate in family activities. Comparing student and host family perspectives from the same sites, Schmidt-Rinehart and Knight (2004) discovered a marked discrepancy in assumptions about who was responsible for encouraging participation in family activities—students thought that hosts should issue invitations, and families felt that students needed to take initiative. Stephenson (1999) surveyed and interviewed 56 Chilean host families and found that they most commonly cited cultural exchange as a reason for hosting American students, followed by social reasons such as companionship, and believed that hosting changed their own appreciation of Chilean culture and their views of people from the United States, but not specific political beliefs and social values. Engel (2011) conducted interviews with members of 15 host families in Spain and found that they (1) viewed themselves as local guides, teachers, and cultural mediators for students, as well as surrogate family figures; (2) identified cultural transmissions from American students in the areas of lifestyle, food practices, values, religion, politics, and language; and (3) reported that hosting students benefited them economically and socially by augmenting income and providing companionship and a sense of purpose.
From the wealth of research on study abroad, it is clear that learner experiences and language development vary greatly and at times fail to meet expectations of the assumed homestay advantage. As Knight and Schmidt-Rinehart (2010) found in implementing assignments to increase student-family interactions in programs for Spanish, there can be a discrepancy between what students say they want to accomplish during a study abroad homestay experience and what they actually do while abroad. The current study was designed to further explore student and host family perspectives and their relationship to students' oral proficiency gains during study abroad across three target languages (Spanish, Mandarin, and Russian). The study addressed the following research questions:
- How did students perceive their relationship with their host family?
- To what extent were students' perceptions related to both their satisfaction with their language learning experiences and their actual language gains?
- How did host families perceive the disposition of the student they hosted?
- To what extent were host families' perceptions related to student language gains?