©Journal of International Students

Volume 12, Issue 2 (2022), pp. 889-908

ISSN: 2162-3104 (Print), 2166-3750 (Online)

ojed.org/jis

Korean universities have shown a dramatic change in international student enrolment over the last 20 years. While a notable increase in enrolment is undeniable, factors related to international students at Korean universities are not well known or are poorly understood. In this exploratory correlational study, we investigated the relationship between gender and study level using the push–pull model among four pull factor dimensions: (a) Appeal of Korea, (b) Experiential Motivations, (c) Social Network Influences, and (d) Institutional Appeal. Short-term exchange students (N = 601) showed that Experiential Motivations was the most salient pull factor dimension in general. A 2x3 analysis of variance indicated statistically significant differences by gender and study level among the four pull factor dimensions. We conclude by discussing mobility programs and the need to account for the different motivations of potential students typologically in order to design policies and programs more effectively.

Keywords: exchange students, Korean learning, pull factors, student mobility

Student mobility paradigms have largely been characterized by movement from East to West and/or South to North (Habib et al., 2014; D. Kim et al., 2018; S. W. Lee, 2017; Park, 2019). However, the 21st century has seen changes to these directional trends, particularly in Asia (Chan, 2012). The Republic of Korea (hereafter Korea) is unique in this sense as international student enrollment has changed dramatically over the previous two decades (Jon, 2009; T. Kim, 2011; S. W. Lee, 2017). In 2000, only 3,963 international students were enrolled at Korean institutions nationwide (S. W. Lee, 2017). Yet by December 2019, Fall semester enrollment alone had grown to 111,858 (Higher Education in Korea, n.d.). As of late 2019, annual international student enrollment numbers started reaching over 140,000 (National Institute for International Education, n.d.). Reasons for this growth, unsurprisingly, are multifaceted.

At the governmental level, initiatives such as Brain Korea 21, Study Korea, and the World Class University project have been implemented to recruit and attract foreign students (Byun et al., 2013; Green, 2015; T. Kim, 2017). These programs often coincide with governmental financial assistance through the Global Korea Scholarship (GKS) and the Korean Government Scholarship Program (KGSP) (Krechetnikov & Pestereva, 2017). Other initiatives include the development of regional student multilateral mobility consortiums such as the Collective Action for Mobility Program of University Students in Asia (CAMPUS Asia) or University Mobility in Asia and the Pacific (UMAP) to increase interinstitutional study pathways (Hou et al., 2017; S. J. Kim, 2017). Moreover, it is not just international students who have been targeted; transnational branch campuses of foreign universities have been established at an education hub in Songdo, Incheon, which is currently home to one Belgian and four American universities (Jon et al., 2014). Although these branch campuses typically attract local Korean students at present, part of the educational hub’s larger mission is to attract students from the Asian region (Jon et al., 2014). At the university level, recruiting targets have been increased in response to local enrollment shortfalls as a result of Korea’s declining birth rate (Alemu & Cordier, 2017). The number of classes offered in English has similarly been expanded to attract and be more accessible to a more diverse student body (Byun & Kim, 2011; Chun et al., 2017). Nevertheless, while the increase in enrollment is undeniable, factors in general related to international students enrolled at Korean universities are not well known or are poorly understood (Alemu & Cordier, 2017).

While differences in motivation have been attributed to differing countries of origin in prior research (see S. W. Lee, 2017), complicating our understanding of international student enrollment in general is that international students are a heterogeneous group beyond just national origins. Further, different levels of study (noncredit, undergraduate, graduate) and mobility type (short- or long-term) have different motivating factors. Moreover, student decisions to study abroad result from a complex web of interactions between multiple sociocultural and socioeconomic dimensions in both the home (e.g., study abroad desire) and host country such as destination appeal (Altbach, 2015). As a result, the subsequent decisions leading to enrollment can be markedly different across numerous typological dimensions. In prior research, examples of conflated or overlooked student types are relatively easy to find (see Madge et al., 2015; Rensimer, 2016; Stewart, 2019). These complications are similarly present in international student research in the Korean context. While push and pull factors associated with degree-seeking students are better investigated (e.g., Alemu & Cordier, 2017), push factors are not necessarily applicable to exchange students given their relatively quick and intended return to their home countries. At the same time, it is not known what attracts short-term exchange students to study abroad specifically in Korea in the first place.

Although international students as a student category are heterogeneous, distinctions in the literature have been lacking (Madge et al., 2015; Rensimer, 2016), obfuscating research findings. Other student types are simply overlooked due to similar yet subtly different situational characteristics such as classifying expatriates (i.e., long-term labor/marriage migrants [and potentially their dependents]) as international students (Rensimer, 2016). Further, there are numerous ways to achieve international education through conventional movement where students physically move to the location of the university (Beech, 2015), as well as when the university moves to the location of the students (transnational education; Francois, 2016). While transnational education has often been viewed as an enterprise for local students, this is not always the case due to immigration (Dobos, 2011); the lines between international and transnational education can blur (Rensimer, 2016). The lines can become even more ambiguous when distance education can eliminate physical movement and borders altogether (Stewart, 2019). In this regard, virtual academic exchanges have long been possible yet the practice is relatively uncommon (Jager et al., 2019). While cross-border distance education in terms of short-term study abroad may seem paradoxical, COVID-19 has prompted new discussions (e.g., Altbach & de Witt, 2020) as mobility programs adapt and/or reinvent themselves, at least in the short term. In any case, one-way student mobility that can be investigated more clearly is by sampling short- or long-term international students separately, or by clearly identifying this difference as a demographic variable. Short-term mobility is characterized by temporary sojourns enabled by interuniversity or multilateral consortium agreements and subsequent credit transfer (DeLoach et al., 2019; Perez-Encinas & Ammigan, 2016). Long-term mobility, by contrast, typically involves directly enrolling at an institution in degree programs. Nevertheless, characterizing mobility by sojourn length is only one piece of the mobility puzzle; there are also different motivations related to student type (Rensimer, 2016; Wilkins et al., 2012). The theoretical lens of push–pull theory is one way of conceptualizing these influences (Altbach, 2015).

The push–pull model of international student mobility describes external forces that act on students. Push factors are often an environmental pressure causing students to seek education abroad (Li & Bray, 2007). Typical push factors can often be an adverse condition in one’s home country such as the lack of certain classes or programs. By contrast, pull factors are ones that attract students such as financial incentives (e.g., full or partial scholarships) or benefits (perceived or real)—for example, degree prestige in one’s home country (Altbach & Knight, 2007; Mazzarol & Soutar, 2002; Nghia, 2019). While this conceptualization is useful, push–pull theory is not without its own limitations. For example, the push–pull model typically only considers external forces and does not explicitly take students’ personal attributes or individual socioeconomic contexts into account (Li & Bray, 2007). In other words, internal attributes can also influence students’ decisions to study abroad, or some students may have easier access to another country via heritage visas (Greenholtz & Kim, 2009) that facilitate the study, which is not accounted for in an external-only point of view. Moreover, the model is youth-centric and/or traditional-student oriented (Iloh, 2018); the simple dichotomy of push and pull likely does not adequately capture nontraditional student motivations or more subtle scenarios that surface through globalization, global nomadism, cultural hybridity, and immigration/expatriation (Greenholtz & Kim, 2009; Rensimer, 2016).

Current research (e.g., Altbach & Knight, 2007; Enright & Newton, 2005; Li & Bray, 2007; Mazzarol & Soutar, 2002; Nghia, 2019) has often explored push–pull factors from the perspective of degree-seeking (or long-term) international students, which also limits the applicability of findings to a certain degree when considering different typological student populations. For example, in the case of short-term mobility, different internal dynamics can be seen more clearly through the tourist/entertainment-like nature of academic exchanges (see Lam et al., 2011; Llewellyn-Smith & McCabe, 2008) and individual goals such as personal growth (Nilsson, 2015) compared to traditional international degree-seeking student counterparts who are pulled by the prestige of a degree or the name of a particular institution (Nghia, 2019). Further, pull factors, arguably, have much more relevance than push factors do in the context of short-term study abroad.

The combinations of external determinants (push and/or pull), sojourn types (short or long), mobility methods (international or transnational), and motivations (intrinsic and/or extrinsic) produce complex and distinct educational scenarios that are not only bound to both a place and time, but relatively between home and destination countries (Enright & Newton, 2005). In Australia, for example, Mazzarol and Soutar (2002) investigated the push and pull factors of predominantly long-term international undergraduate degree students, finding that pull factors included positive perception of the degree, as well as the university and its faculty. Lam et al. (2011) produced similar results among graduate students in Malaysia who viewed the academic reputation of the university, along with its research reputation, to be the most salient pull factors. Nghia (2019) also noted that degree-seeking students often factor in the possibility of staying in the host country after graduation for work or residence, which is another potential pull factor that does not apply to short-term students. Weirs-Jennsen (2020) highlighted the absence of tuition fees in Norway as a strong pull factor for degree-seeking students, which similarly would not apply to exchange students whose tuition is already waived. In the case of Korea, long-term Chinese international students have been attracted by lower entrance standards (Park, 2019) as well as geographical proximity to home (Alemu & Cordier, 2017).

Such pull factors, arguably, are not relevant for short-term exchange students given their lack of official admission status (i.e., they are not degree students at the host university) and relatively quick returns to their home countries. For example, short-term exchange students in Australia were attracted by desirable tourist destinations and considered the characteristics of the destination country to be more relevant than the appeal of the institution (Llewellyn-Smith & McCabe, 2008). Lesjak et al. (2015) found differences by gender among exchange students in the European Union with respect to their motivations for conducting an exchange, as well as reasons for selecting a particular exchange destination. Recent research also suggests that K-pop and Hallyu (the Korean Wave) are also pull factors to some degree for international students, though popular media is more likely the avenue through which students have become aware of Korea rather than it being a determinant for studying abroad in the nation itself (S. W. Lee, 2017). Given the rather recent increase in international student enrollment in Korea (see S. W. Lee, 2017), related research is very much still developing.

When it comes to prior international education research in Korea, much has focused on students engaged in long-term mobility (i.e., degree-seeking international students), the majority of whom come from China or East Asia (e.g., Bae & Song, 2017; Jon et al., 2014; S. W. Lee, 2017; J. Lee et al., 2017; Park, 2019). In this sense, international student research in Korea is perhaps more accurately characterized as regional. When sampling has been diverse by nationality or region of origin, it has still predominantly sampled long-term degree students. For example, Alemu and Cordier’s (2017) multi-institutional survey investigating international student satisfaction in Korea only received 20% of responses from exchange students. In other words, there is a gap in extant literature regarding other types of international students, and with international students originating from other regions in the world. Further, there are growing calls for research with foreign residents in Korea (Shin & Moon, 2019) and exchange students are a part of this broader population, even if only residing short-term. As Korea’s presence on the global mobility landscape is comparatively new, prior studies are not only timely and valuable, but ongoing empirical research is needed.

Since differences in pull factors have been documented in prior literature by student type (i.e., Llewellyn-Smith & McCabe, 2008; Mazzarol & Soutar, 2002), and other exchange student research in Korea has been situated in different mobility program contexts (i.e., short 4–5 week summer programs) investigating local students’ perspectives of interacting with exchange students (e.g., Jon, 2009) or primarily with long-term degree students (e.g., Alemu & Cordier, 2017), there is a gap to fill regarding short-term students.

Since there is little known about the phenomenon of short-term exchange students in Korea, we used an exploratory correlational approach to investigate the factors related to exchange students in Korea. Further, since exchange students are engaged in temporary and comparatively short educational sojourns, we focused on identifying the specific pull factors and their larger pull dimensions that were influential in their decisions to participate in an exchange. To determine appropriate pull factors, we consulted relevant literature on international student destination choice (e.g., Ahmad & Buchanan, 2016; Ahmad et al., 2016; S. W. Lee, 2017; Li & Bray, 2007; Llewellyn-Smith & McCabe, 2008; Mazzarol & Soutar, 2002; Park, 2019; Wilkins et al., 2012). Then, we aggregated pull items and refined them based on the Korean context. This process occurred in conjunction professionals working in the Office of International Affairs, and the Office of International Admissions and Management, leading to eight demographic questions and 30 pull factor statements to measure the following four pull dimensions: (a) Appeal of Korea (AK; 10 items), (b) Experiential Motivations (EM; five items), (c) Social Network Influences (SNI; five items), and (d) Institutional Appeal (IA; 10 items). The questionnaire (see Appendix A) was written in both English/Korean and piloted in a private social media group (which is managed by the Office of International Affairs) for formative evaluation and content validity prior to implementation (Archer, 2008; Bennett & Nair, 2010; Burford et al., 2009; Edwards et al., 2009). Based on the resulting four pull dimensions and exploratory nature of the study, we sought to answer the following research questions:

We conducted the study at a large private research university in northern Seoul, which has enrollment of around 20,000 students, and 3,300 of which are international. To capture a current snapshot of exchange student perceptions, we recruited five semesters worth of students from 2018 to 2020, resulting in a survey population of N = 1,423. The reason for the specific timeframe was due the overlapping of exchange periods with various students from 2018 completing their exchanges in 2019, while others were ending soon/had recently begun in 2019, or who were about to start in Spring 2020 when the study was conducted. The Office of International Affairs granted permission and assistance to create a mailing list from their internal database. Total enrollment of both new and continuing exchange students ranged from 300–500 each semester, often representing 50–100 nationalities. Moreover, exchange students are able to enroll across almost all colleges (with three exceptions), in addition to the university’s language institute, meaning they have diverse academic backgrounds, language abilities, and thus possibly diverse motivations.

We collected data from late Fall 2019 through early Spring 2020 for 2 months. We emailed notice of the questionnaire to students 2 weeks in advance of data collection. The email included information about the study, the principal investigator, and the approximate length of time needed (5 minutes) to complete it. Respondents were asked to rate pull factor statements on a five-point scale (i.e., 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree). Reminder emails to nonrespondents were automated in Survey Monkey at various intervals to promote participation (Waclawski, 2012). Students reviewed an informed consent page and submitted an “I agree” response to participate. They did not receive any compensation for participation.

The survey population yielded 1,406 valid email addresses because 17 were either invalid or had bounced. We received 611 complete responses for a 43.4% complete response rate. Out of 55, 24 nations represented roughly 88% of all respondents, and students’ home university regions typically varied between 3%–4% with the population total. Response percentages by region were proportional to the student body demographics with Europe at 46.7%, Asia at 33.3%, and the Americas at 20%. Thus, around 67% of respondents originated from outside of Asia (this percentage increases to about 85% if only compared to East Asia only). Such diversity is a stark contrast to international degree-seeking students in Korea (see Bae & Song, 2017; Jon et al., 2014; J. Lee et al., 2017; S. W. Lee, 2017; Park, 2019). Respondent characteristics (female [83%], aged 18–34 [M = 22.2], taking undergraduate [66.1%] courses for 4–6 months [76%] at the university’s main campus [95.3%]) were confirmed to be consistent with the program as a whole by the university’s Office of International Affairs.

For the four dimensions on the questionnaire, Cronbach’s Alpha (a scale reliability test) was calculated for each dimension with the AK at .731, EM at .784, SNI at .751, and IA at .712. The internal consistency of the questionnaire’s four pull factor dimensions can be considered reliable since they are ≥ .7.

The distribution of students by gender and level of study is presented in Table 1.

Table 1: CrossTab for Gender ´ Level of Study

In order to answer the study’s three research questions, a 2´3 ANOVA procedure was applied with two levels of gender (male and female) and the three levels of study (language institute, undergraduate, and graduate). Where significant differences in study levels appeared, Sheffe’s post hoc test was conducted to determine the differences between groups.

A comparison of pull factor dimension scores is illustrated in Figure 1. Notable is that EM (4.55) was the most salient pull dimension among students, particularly when compared with IA (3.67) and SNI (3.0). Pearson’s correlation analysis resulted in significantly positive correlations among pull factor dimensions. Specifically, AK was positively correlated with EM, SNI, and IA. These dimensions’ Pearson’s correlation coefficients were .423, .260, and .426, respectively. The EM dimension also had positive correlations (.136) with SNI and with IA, which was .309. Lastly, SNI had a positive correlation with IA, with a correlation coefficient of .358.

A 2x3 ANOVA was conducted to see if there were differences by gender and study level across the four pull factor dimensions, but no statistically significant differences were found. However, there were differences found by gender and study level across individual pull factor dimensions. An analysis of each of the four dimensions is presented below within the context of the study’s three research questions.

To see if there were any gender and study level differences, as well as an interaction effect for the factor AK, we applied a 2x3 ANOVA. The statistical analysis produced significant differences in terms of participants’ gender (Type III Sum of Squares = 96.667, df = 1, MS = 96.667, F = 5.337, p = .021) and study level (Type III Sum of Squares = 147.894, df = 2, MS = 73.947, F = 4.083, p = .017); however, no significant interaction effect between gender and study level was found.

Female participants’ mean scored 39.57 on the AK dimension, whereas their male counterparts scored 37.91. In terms of study level, graduate students had a score of 37.07, which was lower than regular undergraduate students or those attending the university’s language center. Sheffe’s post hoc analysis revealed the nature of the significant differences where graduate students showed lower scores than participants in the language and undergraduate study by 2.91 (p = .000) and 2.18 (p = .006), respectively.

To see if there were gender and study level differences and an interaction effect for the EM pull factor dimension, a 2x3 ANOVA procedure was applied. The statistical analysis showed significant differences in terms of participants’ study levels (Type III Sum of Squares = 155.093, df = 2, MS = 77.546, F = 14.797, p = .000), though no significant differences were found by gender. Further, no significant interaction effect between gender and study level was found.

Participants attending the language school EM had a mean score of 23.83, which is higher than participants in undergraduate and graduate levels by 1.31 (p = .000) and 2.53 (p = .000), respectively, which turned out to have significant differences by Sheffe’s post hoc analysis. In addition, undergraduate students scored higher than graduate students by 1.22 (p = .004). In simpler terms, the higher level of study, the less pull there was by EM.

To see if there were gender and study level differences and an interaction effect for the SNI pull dimension, a 2x3 ANOVA procedure was applied. The results showed significant differences among participants’ gender (Type III Sum of Squares = 122.405, df = 1, MS = 122.405, F = 16.869, p = .009), but there were no significant differences in study level. Further, no significant interaction effect between gender and study level was found. The mean score for male participants on the SNI pull factor dimension was 16.05, higher than female participants by 1.23. The Sheffe’s post hoc analysis indicated a significant difference between participants in the language school with a score of 1.13 (p = .017).

A 2x3 ANOVA procedure was applied to check differences in the IA dimension by participants’ gender and study level. The result showed no significant differences.

Research Question 1 asked if there were differences by gender among the four pull factor dimensions, and differences were found in two dimensions: AK and SNI. Female participants’ score on the AK dimension was 39.57 (male participants’ score was 37.91). On the SNI dimension, male participants’ score was 16.05 (higher than female participants by 1.23). Research Question 2 asked if the participants’ levels of study were related to differences in pull factor dimensions, and results were positive for AK and EM. In the AK dimension, graduate students had lower scores than undergraduate or language institute students. In the EM dimension, participants in the language institute showed higher scores than both undergraduate and graduate students, while graduate students had a higher score than undergraduate students. Lastly, to answer Research Question 3, there were no significant interaction effects between gender and study level in any of the four pull factor dimensions.

In this study, language students were pulled more by EM, and female students were more likely to be studying at the university’s language institute. Thus given that EM was the greatest pull dimension, it would seem a relatively easy return on investment to develop or increase cultural activities, programs, or experiences for both male and female students. Further, given that the vast majority of exchange students in this study were female (and often disproportionately are in exchange programs, see Lesjak et al., 2015; Li & Bray, 2007; Llewellyn-Smith & McCabe, 2008; Mazzarol & Soutar, 2002; Nghia, 2019), it would also be beneficial to develop and/or integrate experiential programming into language programs since female students were more likely to be enrolled in them. Nevertheless, this need not be limited to language programs or female students; experiential programming can be integrated into undergraduate and graduate programs through extracurricular clubs and activities that center on intercultural and international experiences for students of both genders at different levels of study since their pull factors were correlated differently. This type of strategic programming might manifest through student-driven organizations such as an International Student Organization, or in conjunction with government-supported community programs, in addition to university International Affairs offices. We recognize, however, that this requires staffing, funding, and expertise that may not be readily available, or easily acquired for all universities.

At present in Korea, most universities likely need to invest resources to develop such programming, especially outside of the capital metropolitan area. In simpler terms, one size does not fit all, but with limited resources, budgets, and staffing, strategic implementation of experiential programming may be very effective for any university and lead to increased interest and subsequent enrollment, fostering a positive feedback loop. In this study, exchange students were attracted to destination characteristics more than institutional ones, which confirms previous research findings (e.g., Llewellyn-Smith & McCabe, 2008). Ultimately, rather than homogenize international students as a singular entity, mobility programs should take the different motivations of potential students by type into account, in addition to other characteristics such as national/regional origins in order to design policies and programs more effectively. Thus, institutions or departments may want to focus efforts first and foremost on experiences for short-term exchange students as a practical starting point; this would be particularly pragmatic if only limited financial or human resources are available. Related research findings regarding positive perceptions of personal development by means of short-term academic exchanges have also been found in prior research (see Nilsson, 2015). One the one hand, other study abroad research has shown that gender influences the decision to study abroad as well as the study destination, with a stronger relationship among female students (see Lesjak et al., 2015; Nghia, 2019), and the results from this study support this finding from the perspective of pull factors and pull factor dimensions. However, the same study did not find differences by study level and exchange motivations, suggesting that students found exposure to other countries and cultures to be more valuable and important. While this is both plausible and likely accurate for many students (see Llewellyn-Smith & McCabe, 2008; Nilsson, 2015), the findings from our study are semi contradictory. While we also found EM to be the most salient pull-dimension overall, the results regarding level of study were negatively correlated to the AK as well as EM for graduate students. We posit that this may be related to age (or at least a proxy for age) as graduate students are typically older than their undergraduate counterparts. Such students may have more established academic or career goals in mind when compared with undergraduate or language school students. Nevertheless, no significant differences were found in regard to study level and IA, which is paradoxical since we might expect older students, or students studying at the graduate level, to be more attracted by the institutional qualities (see Lam et al., 2011). However, in the case of older (or even nontraditional) short-term exchange students in Korea, the reasons for this remain unclear and is an open area of research.

We recognize that findings from this study have limitations. First, respondents were not only pulled from a single institution, but one that is only similar to a handful of large metropolitan universities in Korea at present. Moreover, pull factors or pull factor dimensions are a moving target to some degree; their applicability should also be considered in the context of both the time and place they originate. Increasing the sample size to include multiple institutions would also strengthen inferences from pull dimensions among exchange students. Moreover, as a correlational study, it is possible that gender may be a proxy for another variable that is directly related to our results. Given these limitations, future research should be conducted in mixed method designs, with more rigorous quantitative approaches, as well as with qualitative approaches such as case studies, to provide reference points for the underlying perspectives that such students have, and which influence their decisions to come to Korea temporarily for study. For example, graduate students could share their insights about the appeal of Korea as a study destination, as well as the specific appeal of their chosen host university given the paradoxical results we found with this subcategory of exchange student.

We conclude by reiterating that not all international students are the same typologically, and consequently the factors that attract them to study abroad are different. While this is no different in the context of Korea, this study contributes to the literature in several ways. First, it provides a conceptual contribution by specifically examining not only a clearly delineated subtype (i.e., exchange) of international student and their specific pull factors, but granularly so by level of study and gender. Prior research, by contrast, has focused largely on degree-seeking international students. Second, it provides an empirical contribution with a data point for the Korean context where ongoing research is needed to better understand the growing inflow of international students. In turn, this can assist Korean universities (and other universities in similar positions elsewhere in the world) to not only expand their programs or partnerships by developing programs and policies to attract students in strategic ways, but to further refine them based on the most salient pull factor dimensions that students have.

Appendix

KYUNGSUK LEE, PhD, is an Adjunct Professor of Korean language at Boise State University where she teaches beginner’s and intermediate Korean. She has been teaching Korean in distance and in-person as well. She specializes in Korean language learning in informal settings where language intersects with culture. Email: kyungsuklee@boisestate.edu

WILLIAM H. STEWART, EdD, is the Inbound Exchange Student Program Manager at Hankuk University of Foreign Studies, where he coordinates all aspects of inbound exchanges. He specializes in transnational and international education, particularly where these fields intersect with distance education. His research focuses on student motivations for, and experiences with distance education in cross-border settings with a focus on the Korean context. He earned a doctorate in Educational Technology from Boise State University. Email: wstewart@hufs.ac.kr ORCid: https://orcid.org/0000-0002-8227-849X