The Influence of Culture on Collabora�ve Learning Prac�ces in Higher Educa�on

Objec�ve: This study systema�cally iden�fies the various per�nent cultural dimensions and the group processes involved in collabora�ve peer learning before empirically exploring their associa�ons among a sample of university students. Methodology: Twenty statements encompassing various cultural dimensions that could possibly influence students’ preferences towards collabora�ve peer learning prac�ces were incorporated into a survey. The data were collected from 147 mul�cultural students studying applied sciences in Finland. Chi-squared test of independence was used to test the associa�on between the cultural dimension scores for students and their stated preferences for collabora�ve learning prac�ces. Findings: A major finding of this study is that learners from diverse cultural backgrounds have differing modes of engaging in the eight separate collabora�ve learning processes. The findings clearly reveal that students from cultures that are hierarchical, collec�vist, less bound by rules, and tradi�onal are less likely to have any prior familiarity with peer-learning methods. It seems rela�vely clear that the power distance and collec�vism dimensions have the greatest impact on shaping the preferences of students for collabora�ve peer-learning methods. Value Added: Very few studies have looked at how the cultural backgrounds of students can influence their use of, and preferences towards, collabora�ve peer-learning methods. This study iden�fies key processes in collabora�ve learning prac�ces, which are shaped by culture as communica�on, decision-making, leadership, evalua�on, trust building, the expression of disagreement, scheduling, and persuasion within a peer group. Recommenda�ons: This study found that Hofstede’s framework might be too constraining when understanding how culture shapes a student’s preferences towards collabora�ve peer learning in the educa�onal context. It would be even more frui�ul to develop an altogether endogenous framework that is more suitable for exploring the influence of culture on learning and educa�on. Such a model should iden�fy the various dimensions of culture beyond those of na�onal iden�ty and consider how they jointly influence a�tudes towards collabora�ve learning rather than considering them in isola�on. This model should also take a more dynamic approach towards both culture and learning.

of independence was used to test the associa�on between the cultural dimension scores for students and their stated preferences for collabora�ve learning prac�ces.

Findings:
A major finding of this study is that learners from diverse cultural backgrounds have differing modes of engaging in the eight separate collabora�ve learning processes. The findings clearly reveal that students from cultures that are hierarchical, collec�vist, less bound by rules, and tradi�onal are less likely to have any prior familiarity with peer-learning methods. It seems rela�vely clear that the power distance and collec�vism dimensions have the greatest impact on shaping the preferences of students for collabora�ve peer-learning methods.
Value Added: Very few studies have looked at how the cultural backgrounds of students can influence their use of, and preferences towards, collabora�ve peerlearning methods. This study iden�fies key processes in collabora�ve learning prac�ces, which are shaped by culture as communica�on, decision-making, leadership, evalua�on, trust building, the expression of disagreement, scheduling, and persuasion within a peer group.
Recommenda�ons: This study found that Hofstede's framework might be too constraining when understanding how culture shapes a student's preferences towards collabora�ve peer learning in the educa�onal context. It would be even more frui�ul to develop an altogether endogenous framework that is more suitable for exploring the influence of culture on learning and educa�on. Such a model should iden�fy the various dimensions of culture beyond those of na-�onal iden�ty and consider how they jointly influence a�tudes towards collabora�ve learning rather than considering them in isola�on. This model should also take a more dynamic approach towards both culture and learning. the resul�ng student migra�on has increased the presence of intercultural contacts among interna�onal students (Dunne, 2013). On the other hand, collabora�ve peer-learning methods, which involve various structures where students work in small groups, have become increasingly common prac�ces in higher educa�on. However, previous research (Popov et al., 2012;Dennehy, 2015;Cagiltay, Bichelmeyer, & Akilli, 2015) has provided conflic�ng evidence about the influence of culture on shaping a student's preferences towards collabora�ve learning prac�ces.
Some studies suggest that intercultural contact can be poten-�ally beneficial to students par�cipa�ng in collabora�ve peer learning through enhanced academic and social adjustment (Wang, 2012), intercultural development, and reciprocal tolerance (Volet, 2004). Others, meanwhile, suggest that cultural diversity in collabora�ve peer-learning groups fosters posi�ve learning outcomes, such as a well-developed social awareness, divergent perspec�ves, and problem-solving skills (Cantor, 2004).
There are also several challenges involved in arranging mul�cultural collabora�ve learning, and these can lead to nega�ve learning outcomes (Dunne, 2013). These include prejudices and stereotypical a�tudes towards culturally distant students, among others. Unresolved cultural conflicts can have an adverse effect on collabora�ve learning, because they arouse nega�ve emo�ons and divert the peer group's a�en�on away from achievable tasks (Ayoko, Callan, & Härtel, 2008). Troubled interpersonal group dynamics, differences in the working and interac�onal styles of group members, and dysfunc�onal communica�on can all result in nega�ve emo�ons emerging in a collabora�ve learning group (Barron 2003).
To ensure posi�ve learning outcomes for the students and the host ins�tu�on, it is vital to understand the factors that shape the preferences of mul�cultural students towards collabora�ve peer learning. This study seeks to explore how the cultural backgrounds of students influence their orienta�on towards collabora�ve peerlearning arrangements. This study focusses on group dynamics and learning processes and how they are shaped by the par�cipants' cultural backgrounds, because it is important to understand these issues (Cohen, 1994;Cagiltay, Bichelmeyer, & Akilli, 2015;Rozkwitalska, Sułkowski, & Magala, 2017;. Although some previous studies have already dealt with aspects of group dynamics in a mul�cultural group (Popov, et al., 2012), very few studies specifically deal with how cultural dimensions can influence the preferences of par�cipants with regards to collabora�ve peer learning and the processes therein.
This study aims to contribute to this area by first iden�fying the cultural dimensions that are likely to influence collabora�ve peerlearning preferences and prac�ces. Exis�ng studies in this area focus excessively on business management issues (Zhou & Shi, 2011; and disregard the collabora�ve learning perspec�ve in the context of higher educa�on. This study system-a�cally iden�fies the various per�nent cultural dimensions and the group processes involved in collabora�ve peer learning before empirically exploring their associa�ons among a sample of university students. A wide range of similar studies have tended to apply Hofstede's largely accepted 6D model of culture (Hofstede, Hofstede, & Minkov, 2010) while trying to understand the rela�onship between culture and group processes, albeit without sufficient cri�cal reflec�on. This study also establishes how empirically valid Hofstede's 6D model is within the educa�onal context. Given the widespread adop�on of Hofstede's 6D model in contexts ranging from management to educa�on, this issue also clearly warrants further inves�ga�on.

Literature review Key concepts
Collabora�ve peer learning is defined as "learning which occurs through social interac�on between peers, directed towards the accomplishment of a common task" (Magin, 1982, p. 108), and this defini�on has remained largely unchanged in the years since. For example, Cronise (2016) considers it as an approach where the experiences of individual learners are shared to collec�vely construct solu�ons. Collabora�ve peer learning can therefore be considered an interac�on between each student and his or her peers through structured protocols to achieve a common goal by developing a shared mental model and a system of accountability (Newell & Bain, 2018).
Culture, meanwhile, is defined as "the collec�ve programming of the mind that dis�nguishes the members of one category of people from those of another" (Hofstede, Hofstede, & Minkov, 2010, p. 6). Although there are several other significant frameworks for analysing the different dimensions of culture, this study uses Hofstede's 6D model to understand collabora�ve peer learning processes for several reasons. Indeed, Hofstede himself pointed out the important implica�ons of his framework within the context of teaching and learning (Hofstede, Hofstede, & Minkov, 2010), and it is probably the most-applied model in compara�ve cross-cultural research (Stahl, Maznevski, Voigt, & Jonsen, 2010). The six dimensions in Hofstede's model cover the dimensions used in other comparable frameworks, such as Schwartz's seven value types (Schwartz, 1992), Hall's classifica�on of culture (Hall, 1959), and the GlOBE project (House, Javidan, Hanges, & Dorfmand, 2002). More importantly, Hofstede's model is used as the basis for iden�fying several group processes in Erin Meyer's 'cultural map' (Meyer, 2014), which is used in this study. What is more, a further aim of this study is to evaluate the applicability of Hofstede's 6D model within the context of various collabora�ve learning processes.
Cross-cultural differences can be understood using the six different dimensions within Hofstede's 6D model. The 'Individualism vs. Collec�vism' (IDV) dimension refers to the degree of interdependence that a society maintains among its members. The 'Power Distance Index' (PDI) is defined as the degree to which power imbalances and inequality are accepted as normal in a society. 'Masculinity' (MAS) refers to the degree to which individuals in a society prefer to be the best (masculine) or do what they like best (feminine). The 'Uncertainty Avoidance Index' (UAI) relates to the extent to which a society feels uncomfortable with unpredictable situa�ons and thus tries to avoid such situa�ons. 'Time orienta�on' (LTO) is the degree to which a culture maintains its link with the past when dealing with the present and the future. The final dimension, 'Indulgence vs. Restraint' (IVR), deals with whether a society places importance on curbing desire to achieve long-term gains (restraint) or regards immediate gra�fica�on as a cultural norm (indulgence) (Hofstede, Hofstede, & Minkov, 2010).
Despite the success and applicability of Hofstede's 6D model, there are also ample cri�ques available. The most obvious one is the assump�on that human values are explained solely by the na-�onal culture (Chiang, 2005) or that all members of a culture exhibit the same cultural values (Williamson, 2002). Even if we allow for the existence and importance of na�onal culture, the model does not adequately demonstrate that na�onal culture determines individual ac�ons. The impact on individual behaviour is dependent upon several other contextual factors and situa�onal variability (McSweeney, 2002). The most dominant cri�que of the model further suggests that the model conflates country and cultural effects, provides indica�on of existence but not the magnitude of cultural effects and ul�mately, is not comprehensive in its inclusion of cultural values (Kirkman, Lowe, & Gibson, 2006;Beugelsdijk, Kostova, & Roth, 2017). Moreover, from a more cri�cal perspec�ve, it has been claimed that this model -instead of being representa�ve of societal cultures -privileges the western masculine managerial viewpoint (Ailon, 2008). Despite these well-known flaws of Hofstede's 6D model, several empirical studies, including this, use the model as it provides quan�ta�ve and approximate measures of rich cultural variety (Williamson, 2002).

Cultural influences on collabora�ve learning preferences
If learning occurs through social interac�on, it must also depend on the cultural context in which it occurs (La�uca, 2016;Maniku�y, Anuradha, & Hansen, 2007). Although peer collabora-�on is becoming increasingly common in higher educa�on, it may not be equally preferred across various cultures. For example, col-labora�ve peer learning is a diffused form of learning with minimal hierarchical imposi�on from the instructor, so it may not be commonly found in hierarchical cultures where there is a greater power distance between the teacher and students (Maniku�y, Anuradha, & Hansen, 2007). On the other hand, in collec�vist cultures, didac�c methods with a focus on coopera�ve learning are more common (Kennedy, 2002;Ngwainmbi, 2004).
Previous research (Stahl, Maznevski, Voigt, & Jonsen, 2010) has broadly dealt with the impact of heterogeneous group members on group dynamics, but few studies have dealt specifically with the aspects of group dynamics in a mul�cultural group (Popov, Brinkman, Biemans, Mulder, & Kuznetsov, 2012). Based on these studies, important collabora�ve group learning processes, which are influenced by culture, can be separated into eight dis�nct processes within a group: communica�ng, evalua�ng, leading, persuading, deciding, trus�ng, disagreeing, and scheduling (Meyer, 2014). According to Meyer (2014), these eight processes can comprehensively describe the cultural orienta�ons of individuals towards group processes (Meyer, 2014). Thus far, however, group processes have been largely studied within the context of global business. In contrast, this study shows how Hofstede's cultural dimensions affect the orienta�on of par�cipants towards these processes within the context of collabora�ve peer learning.
The literature suggests that in low-context cultures (Hall, 1959), people are trained to communicate literally and explicitly (Meyer, 2014), while in the high-context cultures, communica�on is subtler and relies upon subconscious assump�ons of common reference points and shared knowledge (Hall, 1959). This suggests that students from low-context cultures may prefer very explicit, wri�en, and codified forms of communica�on modes when compared to students from high-context cultures.
Individualis�c cultures, meanwhile, are more direct, and they emphasize task-related informa�on in their communica�on (House, Javidan, Hanges, & Dorfmand, 2002). Task-focused indi-vidualis�c people also tend to prefer the explicit documenta�on of group processes and outcomes, so they want precise objec�ves, structured learning, detailed assignments, strict �metables, and an unambiguous assessment (Hofstede, Hofstede, & Minkov, 2010). In collec�vist cultures, meanwhile, the focus on preserving group harmony suggests that individuals from these cultures are less prone to strongly communica�ng their individual viewpoints.
It has also been suggested that individualists engage in peer col-labora�on largely for calcula�ve reasons, and they seek an equitable division of tasks, assessments, and learning outcomes (Hofstede, Hofstede, & Minkov, 2010). In collabora�ve peer groups, individualists may also not contribute fully, or they may even ac�vely undermine group progress, which is a phenomenon known as 'social loafing' (Latane, Williams, & Harkins, 1979), when there is no appropriate system for individual reward. In collec�vist cultures, in contrast, the outcome is considered a collec�ve effort.
One of the fundamental tenets of collabora�ve peer learning is consensual decision-making (Cronise, 2016). It is common in some cultures for authorita�ve figures to make decisions unilaterally (Meyer, 2014). In an educa�onal context, this means that instructors are expected to exert greater influence (Smith & Dugan, 1998). In more egalitarian cultures that prefer consensual decision-making, students are more inclined to resolve group issues among themselves.
The approaches to building trust and relationships also differ across cultures. In collaborative learning groups, relationship-building approaches among peers can either be task-or rela�onship-focused (Meyer, 2014). Individuals from hierarchical cultures are more likely to be task-oriented than rela�onship-oriented (Hofstede, Hofstede, & Minkov, 2010) because they are more concerned with establishing structures than seeking the opinions of subordinates (Bochner & Hesketh, 1994). In contrast, collec�vist individuals value stable rela�onships, so they consequently invest more �me and effort in developing them. Moreover, in cultures with a long-term mindset, the selec�on of peers in collabora�ve groups is mo�vated by a desire to develop long-las�ng rela�onships that will help achieve professional ambi-�ons (Maniku�y, Anuradha, & Hansen, 2007).
Cultural background can also influence individual tendencies to confront others and display disagreements emo�onally (Meyer, 2014). Indeed, in the cogni�vist view of knowledge, learning occurs through conflict (La�uca, 2016). How these disagreements are expressed, however, varies between cultures. In individualis�c cultures, disagreements are expressed directly, and this is viewed as being produc�ve and having no direct consequences for personal rela�onships. In contrast, due to the emphasis on group harmony in collec�vist cultures, highly collec�vist individuals prefer to avoid direct confronta�on and adopt media�on approaches instead (Stahl, Maznevski, Voigt, & Jonsen, 2010). Different cultures can also have different senses of �me (Hall, 1959). For example, those with a monochronic percep�on of �me perform tasks sequen�ally and without interrup�on, so s�cking to rou�nes and comple�ng tasks on �me are important scheduling principles. In contrast, in cultures with a polychromic percep�on of �me, tasks are performed in a fluid manner, with many ac�vi�es being undertaken at once (Meyer, 2014). Hence, there are vari-a�ons in how the strictness of schedules may be perceived by peer group members with different cultural backgrounds.
The 'Western' way consists of categorising objects and applying formal reasoning processes, whereas the 'Eastern' way thinks more in terms of informal rela�onships. In collec�vis�c, high-context socie�es, knowledge is intui�vely decoded from a holis�c context using mul�ple sources (Von Queis, 2005). The 'Eastern' method of cogni�on is therefore largely a consequence of col-lec�vism (Maniku�y, Anuradha, & Hansen, 2007). In summary, 'Western' individuals tend to favour analy�c cogni�on, while people from 'Eastern' cultures tend to engage in holis�c cogni�on. As a result, the culture can directly influence which argument is considered more persuasive. In some cultures, deduc�ve reasoning is priori�zed, so principles will be discussed before applica�on. In other cultures, induc�ve reasoning is priori�zed, and so applica-�on will be discussed before principles (Meyer, 2014).

Method
A range of possible preferences towards collabora�ve peer-learning prac�ces were discussed in the previous sec�on based on students' cultural dimensions. These preferences were incorporated into a survey as statements. The iden�fied processes related to communica�ng, evalua�ng, leading, persuading, deciding, trus�ng, disagreeing, and scheduling within a collabora�ve group (Meyer, 2014). Altogether, twenty statements encompassed various cultural dimensions that could possibly influence students' preferences towards collabora�ve peer learning prac�ces. These statements are summarized in table 1.
All the statements were Likert scale items with three choices: 'disagree', 'neither agree nor disagree', and 'agree'. In addi�on to these, the survey also solicited demographic informa�on about the respondents, including their familiarity with collabora�ve learning prac�ces. The ques�onnaire was set up using the Webropol system (Taimitarha, 2011), and a public link was sent to students a�ending various universi�es in Finland. There are currently 25 universi�es of applied sciences in Finland. The data were collected from students during the winter of 2020 and the early spring of 2021. Some 147 respondents completed the survey. Use and experience of collaborative peer learning I prefer to collaborate with peers rather than work alone (pref_group) Peer learning is not common in my culture (pref_notcommon) I have to depend upon others too much to complete group tasks (pref_toodependence)

Communicating
It is important to write down agreements to prevent future misunderstanding (com_recap) I only speak when I am invited to speak (com_speakinvited)

Evaluating
There is more free-riding in a multicultural group The na�onali�es of the students were used to assign scores to them based on various cultural dimensions using the 'Hofstede Insights' database (Hofstede Insights, 2020) . This database provides an extensive list of scores, ranged 1 to 100, over six different cultural dimensions for many different countries. Higher scores represent stronger characteris�cs. For this research, these numerical scores were converted to categorical scores by using the median as a cut-off point for each dimension. These categorical scores, namely 'low' or 'high', were then assigned to respondents according to their na�onali�es.
Subsequently, the chi-squared test of independence was used to test the associa�on between the cultural dimension scores for students and their stated preferences for collabora�ve learning prac�ces (table 1). This sta�s�cal method was appropriate for use because the variables are ordinal and non-normally distributed (Huberty & Morris, 1989). Further associa�on tests were also conducted for the various other demographic characteris�cs of the students and their preferences regarding collabora�ve learning prac�ces. The IBM SPSS Sta�s�cs version 24.0 so�ware for Windows was used to analyse the data.

Results
There were 147 (N = 147) respondents from different countries, with Finland, Russia, China, India, and Nepal being most represented. The number of respondents were split roughly equally in terms of gender (49.7% male and 50.3% female). Most respondents (57.1%) were between 21 and 25 years of age, with 42.9% of the respondents being in their second year of study and 30.6% being in their first year.
Some 41.89% of the student sample came from egalitarian cultures, compared to 58.11% from hierarchical cultures. Students who were iden�fied as collec�vist comprised 56.67% of the sample, with 43.24% coming from more individualis�c countries. Those students considered as coming from feminine cultures made up 78.38% of the sample, while 21.62% of the sample were deemed as origina�ng from masculine cultures. Students from cultures with a long-term orienta�on cons�tuted 58.10% of the sample, compared to 41.90% from cultures with a short-term orienta�on. Finally, 62.20% of the respondents came from cultures leaning more towards instant gra�fica�on, while 37.80% of the students came from more restrained cultures. The general char-acteris�cs of the respondents are summarised in more detail in table 2. The chi-squared test of independence was used to determine the strength and direc�on of any associa�ons between the respondents' scores for the six cultural dimensions and their statements about collabora�ve learning processes (see table 1). The results of this are provided in table 3. The results presented in table 3 only include the highly significant associa�ons. They reveal that students from power-distanced and collec�vist cultures both exhibit similar preferences towards collabora�ve learning methods. Such students have less prior experience with collabora�ve peer-learning methods, and they also prefer to not work in mul�cultural teams, with them suspec�ng that more free-riding occurs in such teams. They also tend to be suspicious of cultures other than their own and require �me to develop trust in others, and they prefer that agreements be formalised whenever they are made among peers.
Some other cultural dimensions also exert significant influences on the students' preferences for collabora�ve learning prac�ces. For example, students from feminine cultures are less suspicious of individuals from other cultures, so they need less �me to develop trus�ng rela�onships with them. Students from both conserva�ve cultures and those with a high degree of uncertainty avoidance tend to be less familiar with collabora�ve learning methods. They also prefer to work alone instead of collabora�ng with peers, with them fearing free-riding prac�ces within a mul�cultural collabor-a�ve group. What is more, students from more conserva�ve cultures express strong emo�ons when disagreeing with their peers. In terms of demographic characteris�cs, only the gender of the respondents had any impact on shaping their preferences for collabora�ve peer-learning prac�ces, with the results sugges�ng that males need more �me than females to develop trus�ng rela-�onships with their peers.

Discussion
The results clearly reveal that students from cultures that are hierarchical, collec�vist, less bound by rules, and tradi�onal are less likely to have any prior familiarity with peer-learning methods.
These four cultural dimensions are characteris�c of most South Asian countries -such as India, China, Nepal, and Bangladeshwhich were strongly represented in the sample. This clearly indicates that students from this region may not have sufficient previous experience of working in collabora�ve peer groups. They may therefore require an addi�onal orienta�on when beginning studies in a very different culture, as is the case with most 'Western' na�ons. This result is significant, because in theory, highly collec�vist cultures should be familiar with collabora�ve learning. However, the results suggest that educa�onal systems in South Asian countries tend to use it less frequently. This may be because there can be disjuncture between the culture and the educa�on system in a society (Signorini, Wiesemes, & Murphy, 2009). Nevertheless, given that these cultures represent the largest and fastest growing source of educa�on consumers, while collabora�ve learning methods are designed elsewhere, this seems to be important issue to consider (Edmundson, 2007).
It seems rela�vely clear that the power distance and collec�vism dimensions have the greatest impact on shaping the preferences of students for collabora�ve peer-learning methods. It may, however, be a misguided endeavour to consider the influence of individual cultural dimensions on shaping students' preferences towards collabora�ve learning. For example, when we consider South Asian students, their high degree of collec�vism suggests that they should be inclined towards collabora�ve approaches for learning, but their high degree of power distance suggests the opposite. In a culture with a strong power distance dimension, teachers assume an authorita�ve role in educa�on at the expense of collabora�ve learning. Hence, the interac�on between the dimensions of individualism vs. collec�vism (IDV) and power distance index (PDI) warrants further inves�ga�on. More broadly, however, this strongly suggests that we should consider the com-bined influence of the different cultural dimensions on a�tudes towards peer-learning processes rather than trying to isolate them individually.
Whereas most of the theore�cal expecta�ons derived from Hofstede's 6D model were confirmed in this study, some findings were contradictory. For instance, it suggests that students from collec�vist cultures should be more familiar with peer-learning methods, but the results indicate otherwise. Hofstede's model also suggests that individuals from collec�vist cultures tend to priori�ze group goals over individual gains, yet the results of this study suggest that students from collec�vist cultures o�en prefer to collaborate less with peers, with them suspec�ng unequal individual gains in learning outcomes and evalua�ons. In addi�on, and contrary to expecta�ons, those who scored lower according to the uncertainty avoidance index (UAI) preferred to have fixed schedules. Some of these contradictory findings suggest that individual scores for cultural dimensions based on na�onali�es are not as sta�c as Hofstede's 6D model implies. This further indicates that certain aspects of collabora�ve learning are adopted within the new environment, even though they may not be compa�ble with the preferences of the students' original culture.

Conclusions
This study was undertaken to comprehend how cultural dimensions influence the preferences of students for collabora�ve peer learning. A major contribu�on of this study was showing how Hofstede's cultural dimensions affect the orienta�on of par�cipants towards various group processes in the context of collabora�ve peer learning. A major finding of this study is that learners from diverse cultural backgrounds have differing modes of engaging in the eight separate collabora�ve learning processes. It is therefore vital to consider the cultural aspects of learners and support individual learners in enabling them to interact efficiently and achieve their goals in collabora�ve learning. This is even more important when there is a high degree of intercultural contact. Culture ma�ers in collabora�ve peer-learning arrangements, although not necessarily in the same way as suggested by previous research, most notably Hofstede's model.
The findings of this study also have several managerial implica-�ons. Given that several European higher school ins�tu�ons have gradually expanded interna�onal programs targeted to mul�cultural students, it is essen�al to design them in such way that they are compa�ble or tailored towards the cultural orienta�ons of foreign students. This study also reveals that a par�cular group of students with a certain kind of cultural orienta�on have less familiarity and preferences towards collabora�ve learning methods. Given the ubiquity of collabora�ve learning as a pedagogy tool in several higher educa�on ins�tu�ons, incoming students with such preferences or level of familiarity may need addi�onal training or orienta�on regarding collabora�ve learning methods. When forming groups to complete course tasks, administrators and instructors need to be aware of the possible conflicts emana�ng from differences in cultural orienta�on, preferences, and familiarity with collabora�ve learning methods. Instructors most definitely need to consider tailoring evalua�ng schemes to assess individual performances in a collabora�ve learning team. Higher educa�on marketers also should think about marke�ng their programs in a culturally compa�ble methods with associated marke�ng materials, adver�sements, and communica�on channels.
This study has several limita�ons, however. Firstly, the sample of respondents comprised only foreign students who were studying at Finnish universi�es of applied sciences, so the findings may not be generalizable to other contexts. The sample also over-repres-ents South Asian students, although this may reflect a general global trend where South Asian students are increasingly becoming the primary consumers of global educa�on services (Edmundson, 2007;Van Bouwel & Veugelers, 2013;OECD, 2021). Despite the availability of several alterna�ve frameworks for understanding cultural dimensions, this study focused on Hofstede's 6D model, so any shortcomings of Hofstede's model also apply to this study. Indeed, this is one of the more crucial findings of this study. However, as discussed earlier, Hofstede's 6D model also covers the dimensions iden�fied in the other common alterna�ve frameworks. Finally, although this study highlights many cultural preferences regarding collabora�ve learning prac�ces, it does not discuss how to evaluate the effec�veness of collabora�ve learning, so future research should inves�gate this issue.
Previous research has already cast some doubt on the empirical validity of Hofstede's cultural dimensions in the educa�onal context . Likewise, many of this study's findings also contradict the theore�cal expecta�ons, further suppor�ng this doubt. What is more, the applica�on of Hofstede's framework within the educa-�onal context has been accused of oversimplifying cultural differences, using inconsistent categories, and viewing culture as a sta�c construct . It is strongly debatable as to whether na�onality can, or should be, used as a proxy for individual scores for cultural dimensions. This study also supports the no�on that Hofstede's 6D model oversimplifies a culture by reducing it to standalone factors rather than considering their combined simultaneous effects. Furthermore, the model also conceptualizes culture in a rather sta�c way because it disregards the fact that several aspects of culture can be adapted to a new learning environment.
However, previous researchers have stopped short of sug-ges�ng steps towards building a new framework for cultural dimensions, one that is more a�uned with learning and educa�on.
Future research should therefore consider building a more appropriate framework for understanding the dimensions related to cross-cultural learning. It would be even more frui�ul to develop an altogether endogenous framework that is more suitable for exploring the influence of culture on learning and educa�on. Such a model should iden�fy the various dimensions of culture beyond those of na�onal iden�ty and consider how they jointly influence a�tudes towards collabora�ve learning rather than considering them in isola�on. This model should also take a more dynamic approach by accep�ng that many cultural aspects of collabora�ve learning can be learned within a new learning environment.