Content uploaded by Bart Rienties
Author content
All content in this area was uploaded by Bart Rienties
Content may be subject to copyright.
1
Draft paper (accepted 07-09-2012):
Full paper published at: http://jsi.sagepub.com/content/17/4/332
Please cite as:
Rienties, B., Hernandez Nanclares, N., Jindal-Snape, D., Alcott, P. (2013). The role of cultural
background and team divisions in developing social learning relations in the classroom.
Journal of Studies in International Education. 17(4), 322-353. DOI:
10.1177/1028315312463826. Impact factor: 1.000.
The role of cultural background and team divisions in developing
social learning relations in the classroom
Bart Rienties*
Centre for Educational and Academic Development, University of Surrey,
Guildford, UK
Nuria Hernández Nanclares
Faculty of Economics, University of Oviedo, Oviedo, Spain
Divya Jindal-Snape
School of Education, School work and Community Education, Dundee University,
Dundee, United Kingdom
Peter Alcott
Faculty of Business, Economics and Law, University of Surrey, Guildford, UK
*Corresponding author: B.rienties@surrey.ac.uk
2
The role of cultural background and team
divisions in developing social learning
relations in the classroom
Abstract
A common assumption is that students prefer to work together with students from
similar cultural backgrounds. In a group-work context, students from different
cultural backgrounds are “forced” to work together. This might lead to stress and
anxiety, but at the same time may allow students to learn from different
perspectives. The prime goal of this article is to understand how international and
home students from different cultural backgrounds build learning and work-
relations with other students in- and outside their classroom using an innovative
quantitative method of Social Network Analysis in a pre-post test manner.
In Study 1, 50 Spanish and seven Erasmus economics students worked in self-
selected teams. In Study 2, 69 primarily international students in a post-graduate
management program in the United Kingdom worked in randomised teams. The
results indicate that in Study 1 learning ties after 14 weeks were significantly
predicted by the initial team division and friendship ties. The seven international
students integrated well. In Study 2, learning ties after 14 weeks were primarily
predicted by the team division, followed by initial friendship ties, and co-national
friendships. Although international students developed strong (multi-nationality)
team learning relations, international students also kept strong links with students
with the same cultural background. As the initial team division had an eight times
stronger effect on learning ties than cultural backgrounds, these results indicate
that the instructional design of team work has a strong influence on how
international and home students work and learn together.
Introduction
An increasing number of students prefer to study at a university abroad (Russell,
Rosenthal, & Thomson, 2010; Van der Wende, 2003). In many “Western” universities,
teachers and institutes place a lot of responsibilities on students to self-determine their
learning (Hofstede, 1986; Tempelaar, Rienties, Giesbers, & Schim van der Loeff, 2012).
International students may experience a culture shock when the higher educational
organisation, behaviours and expectations of the host university are different from those
of the students’ culture (Zepke & Leach, 2005; Zhou, Jindal-Snape, Topping, & Todman,
2008). De Vita (2001, p. 167) refers to this as cultural learning style, “which re-proposes
learning as a culturally-based phenomenon may then explain why teaching methods,
3
learning tasks and environments which promote learning in some cultures may be
ineffective in others”.
While a large body of research on internationalisation has focussed on
determining how individual characteristics, such as academic integration (Rienties,
Beausaert, Grohnert, Niemantsverdriet, & Kommers, 2012; Zepke & Leach, 2005),
learning styles (De Vita, 2001; Joy & Kolb, 2009; Tempelaar, et al., 2012), personal-
emotional adjustment, stress and anxiety (Rienties, et al., 2012; Russell, et al., 2010;
Ward, Okura, Kennedy, & Kojima, 1998), influence how international students learn and
adjust to the host-institute, to our knowledge a limited amount of studies have focussed
on how social (learning) relations of international and home students influence how
students learn in- and outside the classroom. The degree to which students are able to
develop friendship relations has an influence on how students cope with the complex
demands of higher education (Furnham & Alibhai, 1985; Hendrickson, Rosen, & Aune,
2011; Rienties, et al., 2012). A common assumption by many teachers in large
international classrooms in higher education is that most students seem to prefer to
develop friendship relations and work together with students from similar cultural
backgrounds (Hendrickson, et al., 2011; Montgomery, 2009; Volet & Ang, 1998).
In a student-centred learning environment, whereby students are given more
responsibilities in self-determining their learning, collaborating with fellow-students,
and/or providing feedback to each other (Decuyper, Dochy, & Van den Bossche, 2010;
Katz, Lazer, Arrow, & Contractor, 2004), one would expect that cultural differences and
cultural learning styles amongst students will become more visible than in a teacher-
centred environment, where the teacher sets the pace of the learning activities and is
primarily responsible for assessment and feedback-provision (Montgomery, 2009;
Rienties, Willis, Alcott, & Medland, In Press). In the last 20 years, there has been a rapid
growth in the use of team-learning in higher education to engage students in active
learning (Decuyper, et al., 2010; Springer, Stanne, & Donovan, 1999). By implementing a
learning structure based in teams, teachers aim to convert their classroom in a learning
4
environment where students learn from and together with their fellow team members
(Hernandez Nanclares, Rienties, & Van den Bossche, 2012; Katz, et al., 2004).
As a result, in a student-centred environment students from different cultural
backgrounds are “forced” to work together with other international students and with
home students (Eringa & Huei-Ling, 2009; Montgomery & McDowell, 2009). This might
lead to stress and anxiety for some, but at the same time would allow for an opportunity
to learn from different perspectives and cultural backgrounds (Hendrickson, et al., 2011;
Kim, 2001), and enhance international students abilities to adapt their learning style to the
host-institute. This is particularly important to foster as Volet and Ang (1998) found that
international and Australian (home) students preferred to work in teams with their own
people. Furthermore, in a qualitative study amongst 60 British students at two universities
Peacock and Harrison (2009) found that most British students preferred to work and
develop friendships with co-national students, which Peacock and Harrison (2009)
describe as a form of “passive xenophobia”.
Although an increasing number of studies have recently looked at whether and
how international and home students develop learning and friendship relations (see next
section), most studies have used either qualitative methods, such as interviews or focus
groups (e.g. Montgomery, 2009; Peacock & Harrison, 2009; Volet & Ang, 1998), or used
ex-post questionnaires amongst home or international students to reflect upon the extent
to which they developed those relations (e.g. Hendrickson, et al., 2011; Neri & Ville,
2008) in a wider university-context. Also, in most of these studies the focus was on either
international students or home students, and/or mostly only a subsample of the entire
cohort following a particular program or class were taken into consideration, thereby
limiting our understanding how learning and friendship relations in the classroom actually
develop (for those who did not take part in the research).
The prime goal of this article is to understand the extent to which students from
different cultural backgrounds build friendship, learning and work-relationships with
other students in their class. Therefore, in this study we will contrast two studies that
5
differed in their degree of internationalisation to understand how international students
over time build and develop learning relations with other students using a technique
called Social Network Analysis (Hernandez Nanclares, et al., 2012; Katz, et al., 2004;
Rienties & Veermans, 2012). SNA can be considered as a wide-ranging strategy to
explore and predict social structures to uncover the existence of social positions of
(sub)groups within a network (Curşeu, Janssen, & Raab, 2012; Katz, et al., 2004;
Krackhardt & Stern, 1988; Rienties & Veermans, 2012). Although SNA techniques are
increasingly used in educational psychology to identify social learning patterns in the
classroom, only a couple of researchers have used (in a limited manner) SNA by counting
the number of friends of international students (e.g. Furnham & Alibhai, 1985;
Hendrickson, et al., 2011; Neri & Ville, 2008). Not a single article in this journal has thus
far used this technique to understand the extent to which international students and home
students learn from each other and build friendship relations over time.
The two case-studies were selected for the purpose of illustrating two relatively
“extreme” cases of internationalisation, as research by Ward et al. (2005) indicates that a
critical mass of international students may tip the balance of social interaction and the
learning climate in the class from positive to negative. Study 1 represents a case-study of
limited Erasmus program, whereby most students were local, with only a small minority
of international students following a half-year Erasmus program in Spanish. In contrast,
Study 2 represented an extreme form of intense internationalisation at a university in the
UK, whereby 18 different nationalities were present and only 4% were home students.
Please note that it is not our intention to compare these two studies, as the context of the
studies (team structure, tasks, language, country) are completely different. However, we
are interested in understanding whether these contexts might have played a part in the the
development friendship and learning networks.
Friendships, cultural background and team learning
Current research indicates that institutes and the social networks of students have a large
influence on how international students adjust (Rienties, et al., 2012; Rienties, Grohnert,
6
Kommers, Niemantsverdriet, & Nijhuis, 2011; Tinto, 1998; Zepke & Leach, 2005; Zhou,
et al., 2008). For example, the social life outside the academic environment has a strong
influence on academic and social integration. Having a sufficient number of friends from
the same culture as well as host-culture (Bochner, McLeod, & Lin, 1977; Furnham &
Alibhai, 1985; Montgomery & McDowell, 2009), sharing accommodation with other
students (Ward, et al., 1998), being member of a study association, student fraternity or
joining a sports club can influence social integration and finally increase academic
performance (Rienties, et al., 2012; Russell, et al., 2010). This allows students to establish
a social life that is closely attached to the university setting (Tinto, 1998).
In recent research by Hendrickson et al. (2011) on social friendship networks of
84 international students at the University of Hawaii, a distinction was made between co-
national, home-national and multi-national friendships. Most studies on friendship have
focussed on co-national friendship networks (i.e. friends from the same country).
Although co-national friendship networks provide (short-term) support through social
interaction with students who are experiencing similar emotions, Kim (2001) argues that
it will hinder adaptation processes in the long-run. Hendrickson et al. (2011) found that
students with relatively more co-national friends were less satisfied with their lives.
Having more relations with home-national students in contrast was positively correlated
with satisfaction and connectivity. Multi-national friendships, a third type of friendship
that commonly is developed by international students from different cultural
backgrounds, are often build because international students share a similar experience and
are open to learn from other cultures (Hendrickson, et al., 2011; Montgomery &
McDowell, 2009). Therefore, our first two hypotheses test whether the social learning and
social friendship networks at the beginning of the two Studies were determined by
cultural background.
H1: The social friendship networks of international students are different from those of
home students at the start of the module.
7
H2: International students’ social friendship networks at the start of the module are built
on the same cultural background.
Montgomery and McDowell (2009) found amongst 70 business, engineering and
design students that international students build strong multi-national networks that
provide them with a supportive environment, whereby students make a deliberate choice
of whom to become friends with to provide academic and social support. Although these
results are encouraging, Montgomery (2009) unfortunately does not report how students
were selected for these focus groups, whether there was an under- or over-representation
of particular groups of students, and whether these results are also applicable to non-Post
1992 UK institutes.
Previous research has shown that establishing friendship relations with home-
national students is difficult for international students, due to language issues
(Montgomery & McDowell, 2009; Rienties, et al., 2012), perceived discrimination
(Russell, et al., 2010), and the fact that most home-national students already have well-
established friendship networks (Hendrickson, et al., 2011; Rienties, et al., 2012;
Rienties, et al., 2011). Furthermore, according to Peacock and Harrison (2009, p. 494)
amongst British students there was a “perceived threat that an international student could
bring the marks of the group down through his or her lack of language ability, lack of
knowledge of the United Kingdom or understanding of British pedagogy”. Both
Montgomery and McDowell (2009), using a fine-grained analysis of learning networks
amongst seven international students, and our research (Rienties, et al., 2011) amongst
871 business students at five Dutch business schools, showed that the social worlds of
home and international students are strongly segregated. Also Neri and Ville (2008)
found that international students have a tendency to develop relations with co-national
students, while Hendrickson et al. (2011) found that international students develop both
co-national and home-national friendships. Therefore, the third hypothesis is:
8
H3: The development of social learning networks over time is related to similarity of
cultural backgrounds.
Higher educational institutes have a key role and responsibility in creating a powerful
learning experience for both home and international students (De Vita, 2001; Russell, et
al., 2010; Van der Wende, 2003). In particular, how teachers design their module and
how students are encouraged to work together in small-groups has been found to have a
strong influence on academic integration (Eringa & Huei-Ling, 2009; Hernandez
Nanclares, et al., 2012). The positive effects of collaborative learning over individual
learning has been shown in various studies and meta-analyses (Decuyper, et al., 2010;
Michaelsen & Richards, 2005; Springer, et al., 1999): enhanced cognition; higher
achievement; higher-level reasoning; better transfer of knowledge; more frequent
generation of new ideas or solutions; more positive attitudes of students towards the
subject matter; and increased motivation to do difficult tasks. However, at the same time
there is an increased recognition that the introduction of teams as basic learning units in
the classroom also questions the value of the classroom as learning space; a space in
which the different agents in the learning process - teachers and students - are together
(Hernandez Nanclares, et al., 2012; Hommes et al., 2012; Michaelsen & Richards, 2005).
Within educational psychology and internationalisation literature in particular,
limited research has been conducted in order to assess whether international and home
students also learn from the experiences of other teams in their class and what the
underlying mechanisms for creating this learning space are (Hernandez Nanclares, et al.,
2012; Montgomery, 2009; Rienties & Veermans, 2012). However, in a ground-breaking
study in medical science using Social Network Analysis, Hommes et al. (2012) found that
the primary predictor for academic performance was the social learning network medical
students were engaging in, rather than more “classical” educational concepts like
academic motivation, prior performance or academic integration. Hernandez Nanclares
et al.(2012) found that students over time primarily developed strong relations with
9
students within their team, with a limited amount of links with students from other teams.
Therefore, the final hypothesis is:
H4: The development of social learning networks over time is related to team division.
Methods
Study 1 Limited Erasmus program
Participants and settings
Study 1 took place in an elective third-year course of International Economic Relations at
a Spanish university. The students met twice a week during two-hour class session in a 14
weeks period. 57 (26 males, 31 females) students were divided into eleven teams, which
consisted of four to seven members per team, who self-selected their members. Apart
from seven international students (3x German, 2x Polish, 1x Irish, 1x French), 50 Spanish
students came from the same geographical area. Therefore, we label this study as a
typical example of “Limited Erasmus program”, whereby an institute has provided access
to higher education to international exchange students, but the program is basically taught
in the same manner as before (i.e. in Spanish). The reason for the self-selection of team
members rather than random formation of teams was that most Spanish students were not
familiar with active learning methods such as team work. The seven international students
were assigned to seven separate teams by the teacher. As a result, four teams had only
Spanish students. Research has highlighted when (novice) teams are formed, having a
couple of members within a team that are familiar or even friends is beneficial for social
interaction (Hernandez Nanclares, et al., 2012; Krackhardt & Stern, 1988). During the
fourteen weeks, the eleven teams had to solve five authentic tasks related to international
economics that were highly inter-related. These activities include the creation of a
conceptual map of globalization, writing a reflection on an economics blog by a famous
economist or organisation, and preparing and participating in a final conference about
globalization. The assignments were designed in such a way that they require a broad
10
range of concepts, abilities and skills from each teams A detailed description of the
design has been published elsewhere (Hernandez Nanclares, et al., 2012).
Study 2 International Classroom
Participants and Setting
Study 2 took place in a post-graduate program of Event management at a research-
intensive British university. In contrast to Study 1, whereby most of the students were
from the same cultural background as the institute, in this setting only three students were
from the UK. Therefore, we refer to this Study as an “International classroom”. 72% of
the students were from Confucian Asian and Southern Asian countries, primarily China,
Thailand and India. The third largest group of international students came from Eastern
Europe. 84% of participants were female.
Nine small working teams were formed at random by the tutor after Week 1.
Fifteen students from a “food management” specialisation were divided into Teams 8-9.
The remaining 54 students from the hospitality management program were divided into
Teams 1 – 7. The 69 students had worked together in different small teams within their
specialisation in Semester 1 and had known each other for four months. During the 14
week course period, students met formally once a week during three-hour interactive
class session. At the same time, students were expected to meet with the peers of their
team during the week in order to work on three team processes/products, one of which
was organising a profitable event, all of which were assessed on a group level by the
teacher. A detailed description of the design of the module has been published elsewhere
(Rienties, et al., In Press).
Measuring friendship and learning networks
For ascertaining how international and home students from different cultural backgrounds
learned together over time during the two modules, we employed a method developed
within the field of Social Network Analyses. Numerous researchers have found that
11
SNA networks provide robust and accurate depictions of actual learning processes
and social networks (Curşeu, et al., 2012; Hommes, et al., 2012; Katz, et al., 2004;
Russo & Koesten, 2005), and recent research highlights that social networks are
the key determinant for learning (Hommes, et al., 2012; Russo & Koesten, 2005).
That is, the evolution of the social friendship and learning networks was analysed as
follows. First, the (possible) influence of pre-existing friendship relations was taken into
consideration by using so-called “closed-network” analysis (Hernandez Nanclares, et al.,
2012; Krackhardt & Stern, 1988). The 57 students in Study 1 answered the Social
Network question stem “I am a friend of ...” in Spanish, while the 69 students in Study 2
answered the same question in English in a check-box manner in order to minimise
questionnaire-fatigue. That is, given that we were primarily interested in how networks
developed over time and students had to fill in the questionnaire 2-3 times, a check-box
manner was adopted rather than a rating for each student, which requires more time from
students and might lead to socially deserved answering. A list with all respective names
of the students was provided as is commonly done in SNA (Haythornthwaite & Wellman,
1998). This approach is different from the open-network approach used by Hendrickson
et al. (2011), where students could freely list the names of students they considered as
friends.
Second, (perceived) learning from team members and other members was
measured using SNA in Week 4 for Study 1. In Study 2, given that most students had
already worked together before, we measured the initial working network in Week 1.
Third, in both studies we measured the social learning networks at the end of the modules
at Week 14 in order to analyse whether the dynamics of inter- and intra-team learning and
international and home students had changed. For all three measurements a 100%
response rate was established for Study 1, while for Study 2 a response rate of 71% and
84% was established. The relatively lower response rate of Study 2 at the beginning of
the module can be explained by the fact that some of the students were still in their home
country in the first week of the module. During the post-measurement some of the
12
participants were not present at the debriefing as they were working on their final thesis
and were collecting data.
Data analysis
First, a graphical analysis of the friendship and learning networks was conducted in order
to identify the overall social network structure and identify possible patterns of sub-group
development, as recommended by Wassermann and Faust (1994). Afterwards, a
quantitative analysis was conducted in order to determine the dynamics of social
friendship and learning networks at the beginning and end of each module. For both
studies, as a proxy for cultural backgrounds a multi-national vs. home national matrix was
constructed for Study 1 in order to control for differences in nationality and allowed us to
test H1. Given that more international students with the same nationality were present in
Study 2, a separate co-nationality matrix was constructed in order to test H2, a procedure
similar to creating a dummy-variable for each person with the same nationality in
“classical” statistical analyses. Furthermore, given that there were 37 Chinese students
present in Study 2, and Montgomery (2009) found that some students had a prejudice
against working with Chinese students, we constructed a final Chinese vs. non-Chinese
matrix. Finally, a team division matrix was constructed in order to control the influence
of the team division on the social learning network in order to test H4.
Follow-up quadratic assignment procedure Pearson correlations (Hanneman &
Riddle, 2005) were conducted in order to compare similarity measures between the
friendship and learning networks. Finally, multiple regression quadratic assignment
procedures (MRQAP) were used to test whether pre-existing friendship and learning
relations amongst international and home students predicted social learning networks
after 14 weeks using 2000 random permutations. Basically, MRQAP tests are
permutation tests for multiple linear regression model coefficients for data organized in
square matrices of relatedness of friendship and learning, and the interpretation of the
standardised betas is similar to more OLS regression analyses (Krackhardt, 1988). Data
13
were analysed on a network level using UCINET version 6.350. Although SNA data can
be transformed and exported to “classical” statistical programs, such as Stata or SPSS as
done by Hendrickson et al. (2011), analysis in UCINET is superior given that the specific
learning relations between international and home students (i.e. our primary research
interest) remain intact.
Results
Study 1 Limited Erasmus program
Descriptive statistics
In order to illustrate the power of SNA in understanding initial friendship networks of
international and home students at the start of the module and how the social learning
network after fourteen weeks developed, the social friendship network at Day 1 (Figure
1) as well the social learning network after Week 14 (Figure 2) are presented. Four
aspects can be distinguished from these figures. First of all, Figure 1 illustrates who
students considered as their friends and what the direction of the friendship was.
Furthermore, the label attached to each node represents the respective team number.
Finally, the colour and shape of the node represents the nationality of the respective
student. For example, as indicated by the black arrow in Figure 1, a female German
student of team 11 (white, diamond) indicated that she was friends with a Spanish male
student from Team 8, a Spanish female student from Team 11, and a fellow-German
female student from Team 4, which is indicated by the direction of the arrow.
Insert Figure 1 about here
Second, the respective German female student from Team 11 had no so-called “reciprocal
links” with the three class mates. However, her Spanish friends of team 8 and 11 did have
a reciprocal friendship relation, indicating that these students “acknowledged” each
others’ friendship. A crucial point to remember is that SNA is not based upon the
perception of one participant how he or she perceives the social network. That is,
14
although the German female student indicated that she was only friends with two
students, the other 57 students independently “confirmed” that they also did not have a
friendship relation with this student. In other words, SNA measures the (perceived)
network interactions amongst all 59 participants simultaneously, which verifies and/or
provides counter perceptions from all participants. Third, the social network graph shows
the respective position of individual students as well as of each of the 11 teams. In Figure
1, some learners and teams were on the outer fringe of the friendship network and were
not well-connected to other members or teams. This seems to be in particular for
international students, thereby providing initial support for H1 and H2. Eight Spanish,
one German and one French student (as illustrated by the two grey arrows) were not
connected at all to any of the students in the module in Figure 1. This is not an
unexpected result, as most students in this interdisciplinary elective module were from
separate disciplines and specialisations.
Insert Figure 2 about here
Finally, when comparing Figure 1 with Figure 2, the number of (learning) links between
students and teams alike increased substantially (See also Table 1). In Figure 2, most
teams (e.g. Team 2, 4, 5, and 10) had strong links to their respective team members and
were positioned distinctively as “separate” teams. Of particular interest to this study, all
international students developed strong reciprocal links within their respective team. For
example, as highlighted by the black arrow, the German female student from Team 11
had four links to her fellow Team 11 members. However, at the same time she had eight
links to other members outside her team, four of which were with other international
students and four with other Spanish students. Furthermore, the French and German
international students who had no friends at the beginning of the module developed
strong (reciprocal) links with their respective team members and other students (as
highlighted by the two grey arrows).
15
Learning ties and prior friendship relations over time
In Table 1, the multi-national vs. home national network, initial friendship and learning
ties after four and fourteen weeks and the team division are illustrated, as well as the
density scores for the entire classroom and the correlations between the five social
networks using UCINET QAP correlations. Density compares to the number of ties
present in the social network divided by the total number of possible ties. The overall
density of learning increased from 6% after four weeks to 9% after fourteen weeks, which
implied that only 9% of all possible network links were used for learning.
Insert Table 1 about here
The dichotomised network measure of multi-national vs. home national network was
significantly correlated to the initial friendship social network, thereby providing further
support for H1. Furthermore, the multi-national vs. home national network was
significantly correlated with the learning network after fourteen weeks. International
students developed significantly fewer links with Spanish students in comparison to
Spanish students (M
INT
= 4.86, M
SP
= 6.84, T = 3.95, p < .01), while they developed
significantly more links to international students (M
INT
= 2.29, M
SP
= .68, T = -5.74, p <
.01), providing initial support to H3. However, one has to be cautious to over-interpret
this result, given that the sample size of Spanish and international students was unequal.
The size of the multi-national vs. home national network correlations were lower than the
size of correlations between the initial friendship and two learning networks, indicating
that new learning links were established over time, irrespective of the cultural background
of the students. The team division was strongly correlated to the learning network after
four and fourteen weeks.
Finally, using multiple regression quadratic assignment procedures in order to
estimate which of the four matrices had the strongest influence on our dependent variable,
learning ties after fourteen weeks were significantly predicted by the initial team division
(β = .62; p < .01), followed by friendship ties (β = .08; p < .01), and multi-national vs.
16
home national network (β = .02; p < .05), with an adjusted R-square of 0.42, whereby β
represent standardised betas. Adding the learning ties after four weeks further improved
the fit of the model. That is, learning ties after fourteen weeks was primarily predicted by
the initial team division ties (β = .54; p < .01), followed by learning ties after four weeks
(β = .12; p < .01), friendship ties (β = .07; p < .01), and multi-national vs. home national
network (β = .04; p < .05), with an adjusted R-square of 0.43.
In sum, in Study 1 over time international students built strong learning relations
with both international and Spanish students. Although the multi-national vs. home
national network was a significant predictor for social learning, the effect size was small.
All international students developed learning relations with members within their team,
but also were brokers and bridge builders between teams, as illustrated by Figure 4.
Therefore, we found limited support for H3, but strong support for H4 that most learning
relations are a result of team-divisions.
Study 2 International classroom
Descriptive statistics
In Figure 3, the initial friendships at the beginning of the module of Study 2 are
illustrated, whereby three aspects are visually present. First, as expected there were two
clusters of students, whereby Teams 8-9 formed different sub-groups on the left side of
Figure 3, which was due to the fact that these two teams followed a separate
specialisation in food management before participating in this module. Second, a large
group of Chinese students (blue, square) formed a highly linked subgroup in the Event
Management specialisation on the top right hand side of Figure 3. As students were
randomised in teams, students from various friendship networks were “forced” to work
together. For example as highlighted by the grey circles, Team 5 had five students
relatively closely clustered, while three Chinese students were relatively far away from
the other members at the beginning of this module. Finally, a highly culturally diverse
17
group of international students and British home students (Black, triangle up) was
situated on the bottom right of Figure 3.
Insert Figure 3 about here
Insert Figure 4 about here
As illustrated in Figure 4, after fourteen weeks students in Study 2 developed substantial
learning links with their respective team members, which was similar to our findings
from Study 1. As expected, the food-specialisation group continued to learn primarily
from students of their own specialisation, and as a result formed a relatively separate
subgroup. For most teams, a relatively clear “team structure” could be visually identified
in Figure 4 (e.g. Team 2, 8, 7), in that students from the same team were closely located
together in the social learning network. This is visually also illustrated by team 5,
whereby almost all members were closely situated together. Although almost all students
developed learning relationships with team members irrespective of their cultural
backgrounds, a central group of Chinese students remained visible in the middle of Figure
4, while the other international and English students were more situated on the right side
and outer fringe of the network.
In Table 2, the co-nationality matrix and Chinese vs. non-Chinese were positively
correlated with initial friendships, thus providing initial support for H1 and H2. The
learning networks after fourteen weeks were most strongly correlated with team divisions
and initial work and friendship relations. As the teams were mixed at random, no
significant correlations were found with respect to nationality or Chinese students.
However, a significant correlation of team division was found with initial friendships and
work relations, which could be explained by the fact that most students were already
familiar with each other.
Finally, again using multiple regression quadratic assignment procedures to
estimate which of the four matrices explained most of the variance of the dependent
social learning network variable, learning ties after fourteen weeks were again primarily
predicted by the team division (β = .43; p < .01), followed by initial friendship ties (β =
18
.18; p < .01), and the co-nationality matrix (β = .06; p < .05), with an adjusted R-square
of 0.25. The separate dummy matrix of Chinese vs. non Chinese was not a significant
predictor when the same culture matrix was included. A separate MRQAP without the
same culture matrix did however show that the Chinese vs. non Chinese matrix was a
significant predictor for social learning. In sum, in Study 2 over time international
students from different cultural backgrounds built strong learning relations with both
international and English students. All international students developed learning relations
with members within their team, but at the same time the same culture matrix was a
significant predictor for social learning. Therefore, we conclude that although
international students developed strong mixed-nationality team learning relations,
international students also kept strong links with students with the same cultural
background.
Discussion and Conclusion
Studies 1 and 2 have provided a unique insight using in-depth dynamic social network
analyses into how international students develop learning relations with co-national,
multi-national and home-national students over time in two different team-learning
settings. The purpose of selecting these two Studies was to illustrate the power of using
Social Network Analysis techniques in understanding the complexities of learning in
class between international and home students. We explicitly want to remind the reader
that it is not our intention to contrast and compare the two studies. Instead both Studies
provide two examples of how social learning processes in institutes with a limited
Erasmus program and institutes with an extensive international and diverse classroom
may take place, and we strongly encourage other researchers to use SNA to determine
whether similar or different patterns will emerge over time.
It is clear from the data from Study 1, and with some students in Study 2, that
active learning methods such as team work were effective in crossing cultural boundaries,
in line with expectations raised by Hendrickson et al. (2011). That is, for both studies the
19
best predictor for explaining the extent to which learning ties after fourteen weeks were
constructed was the initial team division, which had a eight to ten times larger
standardised beta size than our proxies for cultural background. This is an encouraging
finding, which seems to contradict the findings of Volet and Ang (1998), and Peacock
and Harrison (2009) that most international and home students prefer to work with co-
national students. That is, when “forced” to work together in multi-national teams for a
substantial period of fourteen weeks on several authentic and complex team products,
students seem to be able to overcome some of the initial cultural barriers that prevent
students to learn together in multi-national teams.
Nonetheless, the two studies also highlight some complex and subtle transitional
processes that some international students seem to go through. That is, both the positions
international and home students take in the social networks figures and the relative sizes
of the cultural background proxies in the MRQAPs indicate that cultural backgrounds had
a marginally stronger impact on learning ties in Study 2. It seems that the motivators for
studying or working together might over time have an impact on how students interact
with students from other cultures. Some of the drivers for Erasmus students in Study 1 to
move to Spain was to learn Spanish language and culture. For this to happen, they needed
to work closely and interact well with the Spanish home students, both in the context of
academic and social interaction. Further, it can be argued that since there were only seven
international students and they were allocated to separate groups by the teacher, they had
no choice but to interact with others from a different culture.
In the case of international students in Study 2, when there was a large group of
Chinese students, they seemed to form closer networks with them, supporting Volet and
Ang’s (1998) assertion and Ward et al. (2005) “tipping point” theory. However, when the
international students came from smaller groups, they were seen to integrate well with
home students or international students from other countries, as was illustrated in Figure
3 and Figure 4, as the need to develop links outside one's culture probably was stronger
20
for these students. Also Montgomery and McDowell (2009) found that international
students built friendship and learning relationships by actively interacting with each
other, irrespective of cultural backgrounds. Therefore, our research would suggest that
although there was a significant but small-in-size tendency amongst international students
to network with others from the same culture, it was not dependent on cultural similarities
alone.
The visualisation of social networks in Study 2 are complex, as existing
relationships have definitely played a big part and it is not clear whether existing
relationships were initially formed based on factors related to the same culture or
academic motivation. This might also suggest that there might be a generalisation and
maintenance of networks from one setting to another as can be seen in the case of
students who came from the same specialisation staying together in teams. However, the
primary predictor for learning after fourteen weeks was the team division initiated by the
teacher in Study 2, thereby implying that although cultural differences played a part, most
students were able to effectively establish learning links over time irrespective of cultural
backgrounds.
As can be seen by the substantial learning links formed by students in Studies 1
and 2, team work was effective when the international students were able to form a
stronger module/task team identity as compared to cultural identity. It is suggested that it
might not be a simple case of some cultures not finding team work effective; other factors
such as language fluency (Peacock & Harrison, 2009; Ward, et al., 2005), the ability to
develop trust in a team (Decuyper, et al., 2010), or learning styles (De Vita, 2001; Joy &
Kolb, 2009; Tempelaar, et al., 2012) might be playing a part which need to be taken into
account when working with international students. It would seem that instead of
randomly allocating students to teams, it is better for the teacher and students to be clear
of the purpose of team work. As highlighted by Peacock and Harrison (2009), in their
study most of the British students seemed to shift the responsibilities for stimulating
21
interaction with international students to the teacher and the institute in general. If one of
the purposes is better interaction and formation of cross-cultural networks, then higher
educational institutes and teachers in particular need to rethink how active learning
methods, such as small-group or team-learning, seem to have a significant and strong
impact on learning and friendship relations between international and home students.
Constraints and Limitations
Although both studies were developed and designed with the highest care, there are
several limitations. A first limitation of this research is that both social network analysis
of learning networks and friendship networks were self-survey instruments, whereby
socially desirable behaviour might influence the results. However, a large body of
research (Borgatti & Cross, 2003; Curşeu, et al., 2012; Hommes, et al., 2012; Katz, et al.,
2004; Wassermann & Faust, 1994) has found that SNA techniques provide a robust
predictor for actual social networks and learning outcomes, in particular given the high
response rates in both our studies and the longitudinal research design. Furthermore,
given that we used multiple regression quadratic assignment procedures to predict the
social learning networks after fourteen weeks, which is a conservative technique
(Krackhardt, 1988) given that 2000 random permutations of alternative models were
conducted, in both Studies we found strong and robust findings (with adjusted R-squares
explaining 25-43% of variance) that primarily team-divisions, initial friendships and co-
national relations are predicting learning.
A second limitation is that we did not conduct a fine-grained analysis of the
actual learning interactions between students, such as done by Montgomery and
McDowell (2009). Although these methods provide an in-depth insight of interactions,
the amount of time and effort to follow just a couple of students is prohibitive for larger
class sizes as reported here. Students in Study 2 could possibly separate their two
motivators (academic and social) in the course of this study. Therefore, strong in-team
connections at university were possible but they might have strong social outside-
22
university connections with students from the same culture as well, supporting the
findings of Tinto (1998). It is possible that they did this to create two distinct networks as
that would give them two networks to support them during their integration in the home
country. It is possible that these students might have been more resilient as, if one
network or part of their life was not working well they could rely on the other network.
A particular contribution of our studies that is relevant for teachers and
researchers is that SNA programs are increasingly intuitive to use, so by conducting a
similar analysis as we have done in the first weeks of a module, teachers and researchers
alike will be able to better understand the complexities in the classroom, and possibly
intervene where necessary if a particular (group of) student(s) is consistently excluded
from social learning interactions. Alternatively, when first asking students to fill in the
friendship network, teachers can create multi-cultural groups with a mix of friends from
different cultural backgrounds, thereby balancing the opportunities to learn from different
perspectives while at the same time ensuring that at least one or two friends are present
for each student in each team, as recommended by Krackhardt & Stern (1988) and
common transition practice across educational stages. For researchers, in particular, the
dynamic use of SNA by measuring social learning and friendship interactions over time
allows them many new angles in understanding internationalisation and social interaction
processes. For example, experimenting with different compositions of teams based upon
cultural backgrounds and friendships, different task-structures, or different assessment
methods would allow a deeper insight into how institutes can actively encourage learning
across cultural borders. Furthermore, triangulation with qualitative research methods
would further strengthen our understanding of the underlying mechanisms of why some
international students develop strong links with co-nationals, while others develop strong
links with home students.
23
References
Bochner, S., McLeod, B. M., & Lin, A. (1977). Friendship patterns of overseas students:
A functional model. International Journal of Psychology, 12(4), 277-294.
Borgatti, S. P., & Cross, R. (2003). A Relational View of Information Seeking and
Learning in Social Networks. Management Science, 49(4), 432-445.
Curşeu, P., Janssen, S., & Raab, J. (2012). Connecting the dots: social network structure,
conflict, and group cognitive complexity. Higher Education, 63(5), 621-629.
De Vita, G. (2001). Learning Styles, Culture and Inclusive Instruction in the Multicultural
Classroom: A Business and Management Perspective. Innovations in Education
and Teaching International, 38(2), 165-174.
Decuyper, S., Dochy, F., & Van den Bossche, P. (2010). Grasping the dynamic
complexity of team learning: An integrative model for effective team learning in
organisations. Educational Research Review, 5(2), 111-133.
Eringa, K., & Huei-Ling, Y. (2009). Chinese Students’ Perceptions of the Intercultural
Competence of Their Tutors in PBL. In D. Gijbels & P. Daly (Eds.), Real
Learning Opportunities at Business School and Beyond (Vol. 2, pp. 17-37):
Springer Netherlands.
Furnham, A., & Alibhai, N. (1985). The friendship networks of foreign students: a
replication and extension of the functional model. International Journal of
Psychology, 20(6), 709.
Hanneman, R. A., & Riddle, M. (2005). Introduction to social network methods.
Riverside, CA: University of California.
Haythornthwaite, C., & Wellman, B. (1998). Work, friendship, and media use for
information exchange in a networked organization. Journal of the American
Society for Information Science, 49(12), 1101-1114.
Hendrickson, B., Rosen, D., & Aune, R. K. (2011). An analysis of friendship networks,
social connectedness, homesickness, and satisfaction levels of international
students. International Journal of Intercultural Relations, 35(3), 281-295.
Hernandez Nanclares, N., Rienties, B., & Van den Bossche, P. (2012). Longitudinal
analysis of knowledge spillovers in the classroom. In P. Van den Bossche, W. H.
Gijselaers & R. G. Milter (Eds.), Learning at the Crossroads of Theory and
Practice (Vol. 4, pp. 157-175). Dordrecht: Springer.
Hofstede, G. (1986). Cultural differences in teaching and learning. International Journal
of Intercultural Relations, 10(3), 301-320.
Hommes, J., Rienties, B., de Grave, W., Bos, G., Schuwirth, L., & Scherpbier, A. (2012).
Visualising the invisible: a network approach to reveal the informal social side of
student learning. Advances in Health Sciences Education.
24
Joy, S., & Kolb, D. A. (2009). Are there cultural differences in learning style?
International Journal of Intercultural Relations, 33(1), 69-85.
Katz, N., Lazer, D., Arrow, H., & Contractor, N. (2004). Network Theory and Small
Groups. Small Group Research, 35(3), 307-332.
Kim, Y. Y. (2001). Becoming intercultural: an integrative theory of communication and
cross-cultural adaption. Thousand Oaks, California: Sage.
Krackhardt, D. (1988). Predicting with networks: Nonparametric multiple regression
analysis of dyadic data. Social Networks, 10(4), 359-381.
Krackhardt, D., & Stern, R. N. (1988). Informal networks and organizational crises: An
experimental simulation. Social Psychology Quarterly, 51(2), 123-140.
Michaelsen, L. K., & Richards, B. (2005). Drawing Conclusions from the Team-Learning
Literature in Health-Sciences Education: A Commentary. Teaching and learning
in medicine, 17(1), 85-88.
Montgomery, C. (2009). A Decade of Internationalisation. Journal of Studies in
International Education, 13(2), 256-270.
Montgomery, C., & McDowell, L. (2009). Social Networks and the International Student
Experience. Journal of Studies in International Education, 13(4), 455-466.
Neri, F., & Ville, S. (2008). Social capital renewal and the academic performance of
international students in Australia. Journal of Socio-Economics, 37(4), 1515-
1538.
Peacock, N., & Harrison, N. (2009). “It’s So Much Easier to Go with What’s Easy”.
Journal of Studies in International Education, 13(4), 487-508.
Rienties, B., Beausaert, S., Grohnert, T., Niemantsverdriet, S., & Kommers, P. (2012).
Understanding academic performance of international students: the role of
ethnicity, academic and social integration. Higher Education, 63(6), 685-700.
Rienties, B., Grohnert, T., Kommers, P., Niemantsverdriet, S., & Nijhuis, J. (2011).
Academic and social integration of international and local students at five
business schools, a cross-institutional comparison. In P. Van den Bossche, W. H.
Gijselaers & R. G. Milter (Eds.), Building Learning Experiences in a Changing
World (Vol. 3, pp. 121-137): Springer Netherlands.
Rienties, B., & Veermans, K. (2012). Symposium: Understanding emerging knowledge
spillovers in small-group learning settings, a social network perspective. In V.
Hodgson, J. C., M. de Laat, D. McConnell, T. Ryberg & P. Sloep (Eds.),
Proceedings of the 8th International Conference on Networked Learning 2012
(pp. 522-524). Maastricht.
Rienties, B., Willis, A., Alcott, P., & Medland, E. (In Press). Student experiences of self-
reflection and peer assessment in providing authentic project based learning to
25
large class-sizes. In P. Van den Bossche, W. H. Gijselaers & R. G. Milter (Eds.),
Advances in Business Education and Training (Vol. 5, pp. XX-XX). Dordrecht:
Springer.
Russell, J., Rosenthal, D., & Thomson, G. (2010). The international student experience:
three styles of adaptation. Higher Education, 60(2), 235-249.
Russo, T. C., & Koesten, J. (2005). Prestige, Centrality, and Learning: A Social Network
Analysis of an Online Class. Communication Education, 54(3), 254-261.
Springer, L., Stanne, M. E., & Donovan, S. S. (1999). Effects of Small-Group Learning
on Undergraduates in Science, Mathematics, Engineering, and Technology: A
Meta-Analysis. Review of educational research, 69(1), 21-51.
Tempelaar, D. T., Rienties, B., Giesbers, B., & Schim van der Loeff, S. (2012). How
cultural and learning style differences impact students’ learning preferences in
blended learning. In E. J. Francois (Ed.), Transcultural Blended Learning and
Teaching in Postsecondary Education (pp. 30-51). Hershey PA: IGI-Global.
Tinto, V. (1998). Colleges as Communities: Taking Research on Student Persistence
Seriously. The Review of Higher Education, 21(2), 167-177.
Van der Wende, M. C. (2003). Globalisation and Access to Higher Education. Journal of
Studies in International Education, 7(2), 193-206.
Volet, S. E., & Ang, G. (1998). Culturally Mixed Groups on International Campuses: an
Opportunity for Inter-cultural Learning. Higher Education Research &
Development, 17(1), 5-23.
Ward, C., Masgoret, A.-M., Newton, J., & Crabbe, D. (2005). Interactions with
international students: Report prepared for education New Zealand. Wellington,
New Zealand: Center for Applied Cross-cultural Research, Victoria University of
Wellington.
Ward, C., Okura, Y., Kennedy, A., & Kojima, T. (1998). The U-Curve on trial: a
longitudinal study of psychological and sociocultural adjustment during Cross-
Cultural transition. International Journal of Intercultural Relations, 22(3), 277-
291.
Wassermann, S., & Faust, K. (1994). Social Network Analysis: methods and applications.
Cambridge: Cambrdige University Press.
Zepke, N., & Leach, L. (2005). Integration and adaptation. Active Learning in Higher
Education, 6(1), 46-59.
Zhou, Y., Jindal-Snape, D., Topping, K., & Todman, J. (2008). Theoretical models of
culture shock and adaptation in international students in higher education. Studies
in Higher Education, 33(1), 63-75.
26
Table 1 Multi-national vs. home national, friendship and learning ties, density and
correlations for Study 1.
M
SD
Density
(%)
Multi-
national
vs. home
national
network
Initial
Friendshi
p
Learning
after 4
weeks
Learning
after 14
weeks
Multi-national vs.
home national
network
43.17
14.13
78
Initial Friendship
2.26
2.36
4
.10**
Learning after 4
weeks
3.51
1.81
6
.03
.25**
Learning after 14
weeks
5.07
3.21
9
.04*
.24**
.51**
Team Division
4.26
0.57
8
-.02
.24**
.69**
.62**
*p < .05 . **p <.01.
Table 2 Multi-national vs. home national networks, friendships and learning ties for
Study 2.
M
SD
Densit
y (%)
Co-
Nationality
Chinese
vs. Non
Chinese
Initial
Friendship
Learning
after 14
weeks
Co-nationality
20.03
16.29
29
Chinese vs. Non
Chinese
33.61
2.49
50
.65**
Initial Friendship
13.71
11.54
20
.06
.15**
Learning after 14
weeks
6.17
5.51
9
.08**
.07**
.25**
Team division
6.96
0.46
10
.01
.02
.15**
.46**
*p < .05 . **p <.01
27
Figure 1 Social friendship network at the start of Study 1
28
Figure 2 Social learning network of Study 1 after fourteen week
29
Figure 3 Social friendship network at the start of Study 2
30
Figure 4 Social learning network of Study 2 after fourteen weeks
