Loneliness among Chinese international and local students in Germany: the role of student status, gender, and emotional support

ABSTRACT

For students, studying abroad has its advantages but might also have unwanted adverse effects, such as social isolation and loneliness. We comparatively analyse the role of emotional support in understanding loneliness among Chinese international students and local students in Germany from a gender perspective. Based on a unique sample, our findings suggest that Chinese international students experience higher levels of loneliness than local students. Female Chinese international students are less lonely compared to male ones, whereas the opposite is true for local students. Surprisingly, having a higher number of emotionally supportive ties is associated with higher levels of loneliness for all students. Our subgroup analysis further revealed that the relationship between emotional support and loneliness varies by gender and student status. Female local students benefit from having more emotionally supportive ties, while the opposite is true for female Chinese international students. We recommend conducting a longitudinal study for examining the causality of the relationship between networks and loneliness in the future. Our findings have implications for universities and higher education research in addressing loneliness among students.

KEYWORDS:

• Loneliness
• personal networks
• international student mobility
• gender
• emotional support

Introduction

This article examines the experience of loneliness among international Chinese and local students in Germany from a gender perspective. Chinese students in Germany form the largest body of international students. In 2019, there were 40,111 Chinese students enrolled at higher education institutions in Germany, corresponding to 13.3 percent of overall international students in the country (Wissenschaft Weltoffen 2022). Drawing on a large-scale countrywide novel survey conducted in 2016 and 2017 with students in their early 20s, we find that international students feel lonelier in comparison to local students. However, the relationship between emotionally supportive ties and loneliness yields mixed results when considering male and female local and international students separately. This new insight highlights the need for further research to uncover the similarities and differences between local and international students’ experiences of loneliness, with a focus on gender.

Loneliness is a negative emotional state characterised by feelings of sadness or emptiness resulting from a perceived lack of desired social relationships (Peplau and Perlman 1982). It is detrimental to mental and physical health, and well-being, and yet, very common across societies (Hawkley and Cacioppo 2010; Yang 2019). One group that is particularly at high risk of experiencing loneliness is international students (Chen and Chung 2007; Diehl et al. 2018; Wawera and McCamley 2020). Research indicates that being separated from friends and family, cultural differences, language barriers, and difficulties adjusting to a new environment can all contribute to feelings of loneliness among international students (Diehl et al. 2018; Sawir et al. 2008; Yan and Berliner 2009; Wawera and McCamley 2020). Moreover, international students may lack social networks or support systems in their new country of education, making it challenging for them to form meaningful relationships and contributing to feelings of homesickness and loneliness (Beech 2015; Hendrickson et al. 2011). Earlier research found that international students – in particular from Asia – tend to experience loneliness significantly more than other student groups in the U.S (Poyrazli et al. 2004; Yeh and Inose 2003). Chinese international students are found to be under stress and tend to experience loneliness, depression, and anxiety due to cultural distance, language barriers, and lack of social support in different country contexts (Tsai et al. 2017; Yeh and Inose, 2002; Wei et al. 2007). Much of this literature has findings on the contributing factors to loneliness and its impact on the health and well-being of international Chinese students only. However, focusing solely on the experiences of international students might distort the actual similarities between international and local students (Jones 2017). To address this gap, our study compares levels of loneliness between international and local students using a unique dataset collected in Germany before the COVID-19 pandemic. We argue that only through such a comparison we can gain a better understanding of the magnitude of international students’ experiences of loneliness and its predictors that are specific to the international student population.

Previous research indicates that the number of personal ties (network size), but also socially supportive networks (indicating their quality), are important protectors against loneliness (Hawkley and Cacioppo 2010; Kawachi and Berkman 2001; Small 2017). Emotional support is particularly crucial in reducing feelings of loneliness (Cotterell 2007; Morelli et al. 2015). When individuals feel supported and connected to others, they are less likely to experience feelings of loneliness and social isolation. International students benefit greatly from supportive personal relationships, which have been found to positively influence their well-being and academic performance (Beech 2015; Diehl et al., 2018). Against this background, we focus on the role of emotional support in explaining loneliness among international students in comparison to local ones.

Gender and loneliness have a complex and often ambiguous relationship. Previous studies using general population groups have suggested potential gender differences in the prevalence of loneliness, although with mixed evidence (Hawkley et al. 2008; Maes et al. 2019; Nicolaisen and Thorsen 2014). For the age cohort we focus on in this article (students mainly in their early 20s), a meta-analysis by Maes et al. (2019) shows slightly higher levels of loneliness for males than females. In the context of international student mobility, there is some evidence of gender differences in international students’ decision-making (Salisbury et al. 2010; Van Mol 2022). However, research on the well-being of international students either completely neglects gender (cf. Sondhi and King 2017) or provides small and inconsistent evidence (Sawir et al. 2008). Bringing together different streams of literature on social networks, loneliness, and international student mobility, our article addresses this second gap by examining the prevalence of loneliness and gender differences among international Chinese and local students in Germany.

Our study aims to contribute to the literature on the German context by examining the experiences of Chinese international students in Germany, which has received relatively less attention compared to English-speaking countries (e.g. Poyrazli et al., 2004; Yeh and Inose, 2003; Sawir et al., 2008). The German case is particularly interesting for three reasons. Firstly, Germany has historically attracted a large number of international students, making it a significant destination for those seeking higher education (Guruz 2011). For example, in 2019 it ranked the 4th most popular country of education for international students worldwide (Wissenschaft Weltoffen, 2022). However, there are limited studies examining the experiences of Chinese international students in non-English-speaking European countries such as Germany (Jiang and Kosar Altinyelken 2020). Secondly, Germany is known for its high-quality education system and a strong tradition of scientific research, making it an attractive destination for international students to pursue their academic goals (Altbach and Teichler, 2001; Bilecen 2014). Finally, while educational degree programs are offered in both English and German, the latter is the dominant language spoken outside of the campuses, which may pose an additional challenge for international students who need to excel simultaneously in two languages to study and live in the country.

Theoretical background and hypotheses

This article brings together different streams of literature on loneliness, gender, and emotional support as a social network construct concerning international student mobility.

International student mobility and loneliness

Loneliness refers to ‘the unpleasant experience that occurs when a person’s network of social relationships is deficient in some important way, either qualitatively or quantitatively’ (Perlman and Peplau 1981, 31). This conceptualisation goes beyond the concept of social isolation, which simply means a lack of relationships. It encompasses the notion of loneliness and emphasises the disparity between one's desired and actual social connections, as well as their quality rather than just quantity. This mismatch accumulates into the negative feeling of loneliness, which is generally associated with adverse consequences both on mental and physical health (Cacioppo and Patrick 2008; Hawkley and Cacioppo 2010). Loneliness is acknowledged to be a common experience among international students, and Chinese international students are no exception (Lin and Kingminghae 2014; Tsai et al. 2017). However, because earlier studies only focused on international students in their research designs and analyses, they have not provided an accurate understanding of the extent of their loneliness. Through a comparative approach, this study seeks to gain a more nuanced comprehension of loneliness by moving beyond the previous finding that international students feel lonely and by demonstrating to what degree they are truly affected.

Regardless of the country of education, the transition to university is an important life event for both local and international students, as it often involves a separation from supportive personal ties. Recent studies indicate that university students across different countries experience increased levels of loneliness (Peltzer and Pengpid 2017). Nonetheless, local students may have the advantage of being able to visit their parents and significant others more frequently than international students. Local students may not experience the additional stressors that their international peers do because they are in a familiar setting in terms of language and culture (Hechanova-Alampay et al. 2002). Although international students keep in touch with their families and friends via the Internet and communication technologies, they might not be able to visit or spend virtual time with them as much as they would like to. This separation from personal ties can be considered an additional stress factor increasing loneliness among international students (Diehl et al. 2018). Moreover, international students usually express a willingness to create new friendships, but this can be challenging and time-consuming, particularly when trying to connect with local students (Bilecen, 2014, 2021; Spencer-Oatey et al. 2017). Research has shown that many international students in the US do not have close American friends, despite their desire to do so, and are dissatisfied with the number and quality of their friendships (e.g. Wei et al. 2007). Thus, in addition to the challenge of separating from old ties, the process of forming new ties can add to feelings of loneliness that is the discrepancy between desired and actual social ties. Furthermore, earlier studies pinpoint Chinese culture-specific stressors to succeed in order not to lose face and bring shame to their families (Ge et al. 2019). Given the challenges of adjusting to a new culture, language barriers, and scattered social networks, it is common for international students, including Chinese international students, to experience feelings of loneliness and social isolation, which may be more pronounced than in their local peers (e.g. Grayson 2008; Yeh and Inose 2003). Against this background, we hypothesise:

(H1): Chinese international students experience higher levels of loneliness than local students in Germany.

Gender and loneliness among international students

The topic of gender and loneliness has received significant attention from scholars, but there is still a lack of consistent evidence and theoretical framework to fully understand the relationship (Nicolaisen and Thorsen 2014). According to a recent meta-analysis on gender differences and loneliness based on 600 studies across a variety of contexts, Maes et al. (2019) found that on average there were no gender differences in loneliness over the life span. However, they found evidence for children, adolescents, and young adults – the latter group being our focus in this study – that males tend to be lonelier than females. It has also been argued that in general, women tend to have higher intimacy expectations from their social ties, and when their network size decreases, they may experience more loneliness (Pinquart and Sörensen 2001). However, despite having more intimate connections, women often experience the negative consequences of other individuals’ issues within their networks (Kawachi and Berkman 2001). In addition, loneliness is often stigmatised for men, while women may be more likely to express negative feelings and be attentive to the feelings of others (Cacioppo et al. 2009). Taking into consideration potential measurement biases, previous research suggests that men experience more loneliness, but women tend to report more as they can acknowledge such feelings with fewer social consequences than men (De Jong Gierveld et al. 2006). In addition, men tend to disclose their loneliness when asked multiple items in survey studies, whereas women usually do the opposite (Pinquart and Sörensen 2001).

A few studies existing in the international student mobility literature have provided different findings regarding the relationship between gender and loneliness. For instance, in their study of undergraduate students in Taiwan, Chen and Chung (2007) found that female students tended to experience less loneliness compared to their male peers. Moreover, they found that the extent to which their social connectedness significantly predicted their loneliness did not correlate with their gender. When examining only international students from a range of backgrounds in Australia, Sawir et al. (2008) found that 67 percent of female and 62 percent of male students reported experiencing loneliness and social isolation. In a qualitative study conducted with Chinese international students in Canada, Ge et al. (2019) found that female students often encounter additional stressors due to their family obligations and gender-specific cultural expectations, which may put them at a higher risk of feeling lonely especially when they lack sufficient support structures. Given the number of limited studies with mixed-evidence on both the general population and the international Chinese student population in various countries, and the fact that our survey employs a direct measure of loneliness that may lead to more frequent responses from female participants, the results should be considered with caution. Thus, we hypothesise:

(H2): Female students report higher levels of loneliness than male students do.

Given the mixed findings on the relationship between gender and loneliness, as well as the diversity of contexts and conditions in the framework of student mobility, it is necessary to focus on the interaction effects of student status and gender to test H1 and H2. In a recent study, Barreto et al. (2021) investigated different predictors of loneliness in 237 countries worldwide and emphasised the importance of examining the interplay between age, gender, and culture when predicting loneliness. In line with this large-scale study, we also investigate the interaction of student status (i.e. international or local) and gender in our sample to gain insight into their levels of loneliness.

Emotional support and loneliness

Emotional support refers to the provision of comfort, encouragement, and understanding to individuals in the form of expressions including caring, showing concern, empathy, and sympathy (Fischer 1982; Small 2017). It creates a safe and nurturing environment where individuals can express their feelings, thoughts, and concerns. Emotional support is essential in managing stress, anxiety, depression, and other emotional challenges, and can improve overall well-being. It plays a critical role in coping with and reducing loneliness and social isolation by providing individuals a sense of comfort, being cared about, being valued, and a feeling of belonging (Cotterell 2007; Morelli et al. 2015). A lack of emotional support has been theorised as a major determinant of loneliness (Weiss 1974), whereas having more emotionally supportive ties is linked to lower levels of loneliness, nonetheless with some mixed evidence regarding gender (Kawachi and Berkman 2001).

Emotional support is often provided by strong ties of family members and friends (Wellman and Wortley 1990) and can take various forms, such as active listening, validation of feelings, and offering encouragement. Surveys often inquire whether respondents have someone they can confide in about their personal and private lives (Fischer 1982; Wellman and Wortley 1990; Small 2017). In that sense, emotional support also highlights the importance of trust in a given relationship. Nonetheless, in a recent study, Small (2017) argues that individuals can also choose to trust and confide in their weak ties. He also adds that researchers should be sceptical about the strength of strong ties, as these relationships can be complex and prone to information leakage to unintended recipients (Small 2017). According to Kawachi and Berkman (2001), women typically have larger networks that provide greater emotional support. However, women tend to be more impacted when someone within their network experiences stress or other issues.

International student mobility literature paid ample attention to students’ social support and well-being (Bilecen 2014; Diehl et al 2018; Hechanova-Alampay et al. 2002; Sawir et al. 2008). Through a network analysis, Bilecen (2014) investigated the experiences of international doctoral students in Germany and found that students tended to rely on emotional support from co-ethnic friendships, which they had established prior to their studies in Germany. Additionally, Lin and Kingminghae (2014) discovered that Chinese international students who had romantic partners residing in the country of education and co-ethnic ties that made them feel belonging to a cohesive culture experienced lower levels of loneliness. However, according to Sawir et al. (2008), although social networks play a crucial role in reducing loneliness for some international students, they are not a universal solution for all students, suggesting that there may be cultural variations in the support and students’ expectations from their social ties.

Our final hypothesis is based on the assumption that students who have emotionally supportive ties they can confide in when faced with personal issues or concerns are better equipped to cope with such challenges and, as a result, experience lower levels of loneliness. Therefore, our hypothesis is:

(H3): Those students who have networks with a high number of emotionally supportive ties experience less loneliness.

Research design

The data used in this article was collected for the Bright Futures Project, an international large-scale collaborative mixed-methods project with three teams located in China, Germany, and the UK. The extensive dataset that we collected for this international project is highly suitable for our analysis, providing us with a unique opportunity to scrutinise the loneliness levels of Chinese international students pursuing their Bachelor's or Master's degree at a German university and compare these findings with those of local students enrolled in German universities. Our data encompasses a pre-pandemic period when loneliness among (international) students was prevalent, yet not as extensively discussed as it has been since 2020.

A two-stage stratified sample design was used to collect the data in Germany. The universities in Germany that had more than 30 international students from China were stratified according to their status based on the QS World Ranking because Chinese international students stated during the initial qualitative interviews that they primarily referred to QS university rankings when deciding where to study. Universities from each stratum were contacted randomly. For the 24 universities that agreed to participate, Chinese international students were contacted by the university administrators via email with our survey link in Qualtrics. In some universities, a random sample of German students was also selected to allow for a comparison between Chinese international and local students in Germany. The research was approved by the Ethical Committee of Bielefeld University. The survey instrument for local and Chinese international students was generated in English by all the team members. Later, it was translated by multiple professional translators into German and Chinese. For the German translation, a native speaker team decided on the final wording, while for the Chinese version, a professional linguist was hired to merge all different translations to ensure that the meaning of the questions was maintained. The team in Germany also conducted cognitive interviews with Chinese international students to ensure that the survey questions were well understood as initially intended. The fieldwork in Germany took place between March 2017 and April 2018.

Study sample

The final sample (n = 792) contains international students from China and local students in Bachelor’s and Master’s level programs in Germany. We included only respondents who also filled in an additional network section (see below). Following the listwise deletion process, every observation with a missing value for the dependent variable or at least one of the independent variables was removed from the sample. For the four subgroups, the sample consisted of 259 female Chinese international students, 197 male Chinese international students, 198 female local students, and 138 male local students.

Operationalisation of the variables

Loneliness

Loneliness is the dependent variable and is measured via self-assessment using an item asking the respondents how often they felt lonely during the past 30 days. This common way of operationalising the extent of different feelings can also be found in the German Socio-Economic Panel (GSOEP). Response options on a fully verbalised five-point scale are ‘none of the time (1),’ ‘a little of the time (2),’ ‘some of the time (3),’ ‘most of the time (4),’ and ‘all of the time (5).’ This variable is treated as interval scaled (quasi-interval scale).

Gender and student status

Variables of both gender (1 = female, 0 = male) and student status (being an international student from China and being a local student from Germany) (1 = Chinese international students, 0 = local students) are dichotomous.

At the end of the main survey, we incorporated a network section that was further incentivised when students continued the survey. The respondents were asked the following name-generator question: ‘Other than your parents, who are the people you would consider important to you?’ By asking this question, we aimed to elicit strong ties that are perceived to be important by the respondents. The respondents could name a minimum of three to a maximum of eight persons (known as alters). Later, they were asked detailed questions about their alters including their gender, age, educational level, nationality, geographical location, type, and duration of their relationship, frequency of contact, mobility experience, and their supportive functions (emotional support, education information).

Emotional support

Once alters were elicited by the name-generator question, we asked about each alter: ‘Have you talked about your personal and private matters with this person in the past year?’ The answer was dichotomous (1 = yes, 0 = no). We then added up the number of alters who were emotionally supportive, a minimum of three to a maximum of eight emotionally supportive alters for each respondent.

Control variables

We control for the location of alters because we expected that Chinese international students’ networks would be more geographically scattered than those of the local students. This variable is dichotomised to distinguish between alters living in Germany and alters living in countries other than Germany. Again, we added up the total number of alters living in Germany (3–8). We control for being in a relationship (1 = in a relationship and 0 = not in a relationship) as one strong intimate tie, and overall life satisfaction (from very dissatisfied [1] to very satisfied [5]), because lonely individuals report lower levels of life satisfaction (Goodwin et al. 2001). Moreover, we control for network size because we do not have a fixed network size design.

Analytical approach

We used STATA 14 to analyse our data. First, we ran four different linear regression models, stepwise adding our predictor variables and the aforementioned interaction effect. Models 1–3 aim to test our hypotheses 1–3 on the relationship between loneliness and student status (H1, Model 1), gender (H2, Model 2), and emotional support (H3, Model 3). In Model 4, we added the interaction effect of student status and gender to depict the complexity of contexts and conditions. We adjusted for heteroscedasticity by specifying robust standard errors using Stata’s vce(robust) option as suggested by Bittmann (2019).

After analysing the initial findings, we proceeded to a more exploratory second step by running linear regression models to investigate the impact of emotional support on loneliness for each group separately. This included Chinese international male and female students, as well as male and female local students. All these models contain the main predictor emotional support and all control variables.

Results

Descriptives

Table 1 provides an overview of the descriptive results for each of the four subgroups separately (female Chinese international students, male Chinese international students, female local students, and male local students). We observe that male Chinese international students are the loneliest and male local students are the least lonely subgroup. The difference between these two subgroups is highly significant (p < .001). In addition, female local students are significantly less lonely compared to their Chinese international counterparts (p < .001). Among Chinese international students, women are significantly less lonely than men (p < .05), whereas there is no significant difference between male and female local students.

Table 1. Descriptives of all variables by four subgroups.

	Female Chinese international students	Male Chinese international students	Female local students	Male local students
	N = 259	N = 197	N = 198	N = 138
Loneliness (range 1–5), M (SE)	2.42 (0.08)	2.72 (0.09)	2.07 (0.07)	2.00 (0.10)
% none of the time	28.19	22.84	36.36	45.65
% a little of the time	30.12	22.84	27.78	23.19
% some of the time	21.24	24.87	30.81	18.84
% most of the time	12.36	18.78	3.03	10.14
% all of the time	8.11	10.66	2.02	2.17
% in a relationship	43.24	46.70	62.12	50.72
Life satisfaction (range 1–5), M (SE)	3.92 (0.05)	3.84 (0.05)	4.17 (0.07)	3.93 (0.09)
% very satisfied	21.62	19.29	43.43	36.23
% satisfied	51.35	48.22	39.90	36.96
% neither satisfied nor dissatisfied	24.71	29.44	8.08	13.04
% dissatisfied	1.93	3.05	7.07	11.59
% very dissatisfied	0.39	0.00	1.52	2.17
Network size (range 3–5), M (SE)	6.88 (0.12)	6.70 (0.15)	6.80 (0.13)	6.39 (0.17)
3 alters	13.51	19.29	11.62	16.67
4 alters	5.79	5.58	6.57	10.14
5 alters	4.25	2.03	6.57	8.70
6 alters	2.70	2.54	5.05	3.62
7 alters	2.70	0.51	6.06	3.62
8 alters	71.04	70.05	64.14	57.25
Number of alters living in Germany (range 0–8), M (SE)	2.17 (0.13)	1.34 (0.11)	6.16 (0.15)	5.98 (0.18)
Number of emotionally supportive alters (range 0–8), M (SE)	3.51 (0.13)	3.15 (0.16)	6.05 (0.14)	5.07 (0.20)

In terms of romantic relationships, female local students have the highest likelihood of being in one compared to female Chinese international students, who are less likely to be in a romantic relationship. Female Chinese international and male local students on average report the same level of life satisfaction, although male local students have more equally distributed responses. Female local students are more satisfied with their overall life, while male Chinese international students are less satisfied. Chinese international students in our sample, both men and women, have significantly lower numbers of connections living in Germany than local students, who have a network oriented towards local connections. The average number of connections living in Germany is similar between the male and female local students, while female Chinese international students have significantly higher average number of connections living in Germany than male Chinese international students. Chinese international students report a lower average number of connections with whom they discussed personal matters during the past year, and, therefore, receive less emotional support compared to local students. Interestingly, there is no significant difference in the number of emotionally supportive ties between female and male Chinese international students, but among local students, female students have more emotionally supportive ties than male students.

Effects of student status (international/local), gender, and emotional support on loneliness

We hypothesised a higher level of loneliness for Chinese international students compared to local students (H1). To test our hypothesis, we computed a linear regression model using student status as the independent variable (Table 2, Model 1). In our model, we control for relationship status, overall life satisfaction, network size, and the number of alters living in Germany. In support of our first hypothesis, we find that Chinese international students are significantly lonelier than local students (b = 0.33, p < 0.05). Despite controlling for network-related variables such as being in a relationship or the number of alters living in Germany, Chinese international students still experience higher levels of loneliness compared to local students. This suggests that a student’s status (international or local) still plays a role in their level of loneliness, even if they have the same number of ties living in Germany. The quality of personal connections appears to be more crucial than the quantity of connections in explaining loneliness among Chinese international students.

Table 2. Linear regression models.

	Model 1 (N = 792)		Model 2 (N = 792)		Model 3 (N = 792)		Model 4 (N = 792)
	b	SE	B	SE	B	SE	B	SE
Constant	4.37***	.25	4.38***	.25	4.36***	0.25	4.22***	.26
Student status (1 = int.)	.33*	.14	.33*	.14	.39**	.15	.71***	.18
Gender (1 = fem.)			–.05	.08	–.07	.08	.22*	.11
Emotional support					.05*	.02	.04	.02
Life satisfaction	–.45***	.05	–.45***	.05	–.45***	.05	–.46***	.05
In a relationship (1 = yes)	–.57***	.08	–.57***	.08	–.57***	.08	–.59***	.08
Network size	–.02	.02	–.02	.02	–.04	.03	–.04	.03
Number of alters living in Germany	–.01	.03	–.01	.03	–.02	.03	–.01	.03
Gender*Student status							–.50**	.16
R^2	0.21		0.21		0.22		.23

*p < 0.05; **p < 0.01; ***p < 0.00.

Previous studies examining the relationship between gender and loneliness have presented mixed evidence. Based on our loneliness measurement, which involved using a single direct question, we hypothesised that female students would report experiencing more loneliness than male students (H2). We computed a linear regression model using gender as the main independent variable and included all of the aforementioned control variables and student status (Table 2, Model 2). However, our findings indicate that the variable gender has no significant effect on loneliness, leading us to reject H2.

Our third hypothesis refers to the impact of emotional support on loneliness. We hypothesised that a greater number of emotional supportive ties would be correlated with lower levels of loneliness. To test this, our model contains the control variables, student status, gender, and the number of emotionally supportive ties as a predictor for loneliness (Table 2, Model 3). We find a weak but significant effect of emotional support on loneliness. However, the direction of this effect is contradictory to our hypothesis. Surprisingly, a higher number of emotionally supportive ties correlates with a higher level of loneliness (b = 0.05, p < 0.05). This unexpected finding might be explained by the idea that students who were already lonely may seek more emotional support and therefore talk more about their personal and private matters within their networks. Since our dataset is cross-sectional, we can neither claim the direction of this association nor make causality assumptions. Therefore, we can only conclude that there is no support for H3 based on our data.

Disentangling the role of gender and student status

Thus far, in our analyses, we find significant effects of student status that are in line with our initial assumption, while the significant effects of emotional support contradict our assumption. We do not find a significant effect of gender on loneliness. In order to investigate a potential interaction effect, we calculated an additional model containing all predictors, all control variables, and, in addition, the interaction effect of gender and student status (Table 2, Model 4). Model 4 shows no significant effect of emotional support but we find a significant interaction effect of student status and gender on loneliness (p < 0.01). Regarding their loneliness levels, being an international student compared to being a local student has a significantly stronger effect on male (b = 0.71, p < 0.001) than on female students (b = 0.21) (not shown in Table 1). Furthermore, being female has different effects on the loneliness levels of international and local students: being female is associated with lower loneliness levels for international students (b = −0.28), while for local students, being female is associated with increased loneliness (b = 0.22, p < 0.05). In other words, female local students are significantly lonelier than male local students, while male Chinese international students are significantly lonelier than female Chinese international students. After controlling for life satisfaction, relationship status, network size, and the number of alters living in Germany, we observed that Chinese international female students are less lonely compared to their male counterparts, whereas the opposite is true for local students – male students are less lonely than female students. In conclusion, there is a significant gender effect when considering gender in interaction with student status for both Chinese international and local students, although this effect is reversed.

Separate models for the four subgroups

In order to gain a deeper understanding of the impact of emotional support on the loneliness levels of both Chinese international and local students, we computed a separate model for each subgroup (Table 3). Each model contains all control variables and emotional support as predictors of loneliness. Initially, we observed great discrepancies in the R² values. Specifically, our model explains 29 percent of variance in male local students, while it accounts for only 18 percent of the variance in male Chinese international students. The R² values for female students range from 24 to 25 percent. Therefore, we can conclude that our model is best suited to explain loneliness levels among male local students compared to the other three groups.

Table 3. Linear regression models, separated models of the four subgroups.

	Loneliness
	Female Chinese international students Model A (N = 259)		Male Chinese international students Model B (N = 197)		Female local students Model C (N = 198)		Male local students Model D (N = 138)
	b	SE	B	SE	B	SE	B	SE
Constant	5.84***	0.43	5.09***	0.54	3.85***	0.38	3.44	0.44
Emotional support	0.09*	0.04	0.03	0.04	−.14**	0.06	0.10	0.05
Life satisfaction	−0.76***	0.09	−0.51***	0.13	−0.26***	0.07	−0.34***	0.09
In a relationship (1 = yes)	−0.44**	0.14	−0.71***	0.18	−0.49***	0.13	−0.78***	0.17
Network size	−0.08	0.04	−0.03	0.04	0.16*	0.07	−0.07	0.08
Number of alters living in Germany	−0.02	0.04	0.04	0.07	−0.10	0.05	0.04	0.08
R²	0.24		0.18		0.23		0.29

*p < 0.05; **p < 0.01; ***p < 0.001.

In our analysis, we find an unexpected association between emotional support and loneliness for female Chinese international students. As we had already noted in the initial step of our research, high levels of emotional support are significantly associated with high levels of loneliness among female Chinese students. However, the opposite is true for female local students, where lower levels of loneliness were linked to high levels of emotional support, aligning with our expectations based on previous research. This pattern does not hold for female Chinese students, possibly due to the idea that discussing personal and private matters with personal connections may be more burdensome for them than providing relief. Such conversations might cause homesickness, and demands from ties in China might add to the burden. In contrast, we do not find a significant relationship between emotional support and loneliness for male students, whether local or Chinese international. Therefore, we only find support for H3 among female local students.

Conclusion and discussion

This study is the first to examine the relationship between emotional support and loneliness among Chinese international and local students in Germany from a gender perspective. Our study makes three key contributions to the literature by integrating diverse theoretical frameworks. First, we not only investigate gender differences but also examine network explanations for differences in loneliness levels between Chinese international and local students. We use a novel dataset from the Bright Futures Project and conduct linear regression analysis to gain detailed insights into the reasons behind the variation in loneliness levels between these two student groups. In addition, we examine within-group differences by analysing the relationship between emotional support and loneliness separately for male and female students. Second, we contribute to the literature on international student mobility by investigating loneliness among Chinese international and local students in a non-English-speaking study context. This is significant because previous research has primarily focused on international students located in English-speaking countries (e.g. Poyrazli et al. 2004; Yeh and Inose 2003; Sawir et al. 2008). Our research in Germany is novel and highlights the high levels of loneliness for Chinese international students. Third, our study adds to the existing literature by demonstrating the significance of network explanations for loneliness among international students. Our findings suggest that emotional support from family, friends, and peers plays a critical role in mitigating loneliness levels among Chinese international students, highlighting the importance of social networks in promoting their well-being, while also pinpointing gender differences.

Our study’s results supported our initial hypothesis, indicating a higher level of loneliness among Chinese international students compared to local students aligning with the earlier studies (e.g. Grayson 2008; Sawir et al. 2008). Further research might benefit from exploring the composition of such personal ties as well as expectations regarding friendships and emotional support separately for international and local students. As to the practical implications of studies in this field, such research could provide universities with insights on how to tailor and diversify their efforts to combat loneliness and design interventions to alleviate it for different student bodies.

We found that gender alone does not predict the level of loneliness; however, when analysed together with student status, gender becomes a meaningful explanation. Only through this interaction, gender differences among students became visible. Among Chinese international students, male students reported higher levels of loneliness, whereas among local students, females did. This finding aligns with previous research conducted in Thailand with Chinese international students, which found that male students were significantly lonelier than their female counterparts (Lin and Kingminghae 2014).

However, an important question still remains: why do we see an opposite trend of gender between local and international student groups in explaining loneliness? One potential explanation could be related to gender-specific parental expectations pertaining to Chinese culture. Gender norms still effectively pressure young people, particularly related to marriage and having children. Young women might feel more pressure than young men to conform to these roles because they are often expected to form a family at a certain age. Studying abroad could function as a period when such expectations and norms are put on hold, allowing students to distance themselves both geographically and mentally from parental pressures to conform to gender norms. Female international Chinese students have a greater geographical distance from their parents, who may exert this pressure, than local female students. Thus, for female international Chinese students, studying abroad means escaping the pressure of conforming to gender norms related to marriage and having children, at least temporarily. ‘Their migration status serves as important resources for these women to resist fixed gender norms expected from them at home’ (Tu and Xie 2020), resulting in lower feelings of homesickness among female Chinese international students than among male Chinese international students. However, the extent to which different gender norms and family-related expectations influence feelings of loneliness among international students is still speculative. Future research can examine the effects of parental pressures and gender norms on Chinese international students in greater depth. Our results also have significant practical implications for universities that aim to tackle loneliness among their (international) student populations. Even in addressing loneliness, higher education policies can have the most enduring and impactful outcomes when they are designed to be gender-sensitive. Of course, this could also be a crucial consideration for trans and queer students, who often suffer from discrimination since homophobia on campus is still a significant problem (Ellis 2009). Unfortunately, we were unable to investigate this group in our study.

Our mixed findings regarding gender, students’ status, and loneliness can also be explained by the reporting effects in surveys. First, as mentioned earlier, in our sample, men might have reported their loneliness less than women (De Jong Gierveld et al. 2006), which can explain our findings only for the local students, but not for the Chinese internationals. Second, in our survey, we asked directly about feelings of loneliness, with one item, which might have affected responses. In previous research measuring loneliness with one or multiple survey items, women tend to disclose such feelings when asked directly, whereas men respond only with multiple questions (Pinquart and Sörensen 2001). Regarding methodological implications of our study, we support a more nuanced measurement of loneliness. Further studies could address these issues by incorporating multiple items on measuring loneliness and being culturally sensitive in the wording of questions for both local and international students.

Our assumption that having more emotionally supportive ties would explain lower levels of loneliness was not fully supported by our analysis. While we did see a confirmation of this assumption among female local students, the opposite was the case for female Chinese international students. Furthermore, for male students, both Chinese internationals and locals, we found no evidence of a relationship between emotional support and loneliness. There are three potential explanations for these different associations between emotional support and loneliness. First, it is important to consider how the survey question on emotional support was phrased and interpreted. Respondents were asked whether they discussed their personal and private matters with their designated personal ties. However, some may have interpreted this as a reciprocal exchange, where they also listened to others’ private matters, rather than receiving unidirectional emotional support. Some respondents may have even provided emotional support without receiving any in return. If these individuals considered themselves the primary confidantes for their personal ties, this could have increased their emotional burden and contributed to their loneliness. Moreover, as stated by earlier research some women tend to be affected more by their social ties who are going through tough times having problems (Kawachi and Berkman 2001). Future studies must investigate not only gender differences in loneliness and networks but also potential cultural differences as our findings suggest.

Second, the observed association between emotional support and loneliness can also be explained by the idea that students with higher levels of loneliness might have been talking about such issues with their personal ties, resulting in more emotional support from them. Feelings of loneliness, in this case, require more emotional support. Future studies with longitudinal designs could help to identify causal mechanisms. Our results highlight the need for further research on the relationship between emotional support and loneliness among different groups of international students.

Third, on the one hand, having interpersonal ties that students can confide in when they have personal issues or concerns can assist them to deal with such issues and, therefore, decrease their loneliness. However, on the other hand, ‘[l]onely individuals are not inclined to provide intimate disclosures […] because they believe that friends will respond critically and not maintain confidentiality of disclosure’ (Hamid and Lok 2000, 48). This suggests that loneliness might be perceived as a stigma or a taboo, and individuals tend to disclose such feelings less often to their interpersonal ties. Each of our potential explanations highlights the considerable uncertainty regarding the relationship between emotional support and loneliness as well as the need for more research.

Because our data was gathered prior to the COVID-19 pandemic, our findings illuminate the pre-pandemic situation. However, the pandemic has brought discussions around social distancing, social isolation, and loneliness to the forefront. Both international and local students have been impacted by the pandemic, although their experiences may differ (Bilecen 2020). Because the outbreak of the COVID-19 pandemic started in China, Chinese international students have faced multiple challenges in different country contexts (Ma and Zhan 2020; Xu et al. 2021). Such challenges included accessing timely information, healthcare services, living in quarantine that limited opportunities for social interaction, and increased incidents of racism and discrimination. It is therefore important for future research to examine the relationship between student status, gender, emotional support, and loneliness in the post-pandemic era compared to pre-pandemic times to determine whether and to what extent this relationship has changed. Additional questions that emerged from our study that require further exploration include socio-cultural differences between Germany and China that may impact this relationship, as well as potential generational factors, given that our respondents were born in the early 1990s.

The main methodological limitation of our study is that the network section was conducted immediately after the main questionnaire, which may have led to inconsistencies in understanding and respondent fatigue. Additionally, we used a single name-generator question and we asked support questions later including the one about the emotional support respondents actually received from their named alters. This question contained a discussion of personal and private matters, which could have also been understood as referring to the alters’ personal and private matters. This misunderstanding of the direction of the emotional support question could explain the unexpected finding of the inverse association between having a large emotionally supportive network and lower levels of loneliness. In addition, our study is based on cross-sectional data, which makes it impossible for us to deduce any causal mechanisms. Nevertheless, our study is pioneering in its investigation of personal networks to understand loneliness levels of local and Chinese international students in Germany from a gender perspective. Our study reveals the complex relationship between student status, gender, emotional support, and loneliness, raising awareness for loneliness of different groups of students.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Deutsche Forschungsgemeinschaft.

Notes on contributors

Isabell Diekmann

Isabell Diekmann is a Research Fellow at the Center on Migration, Citizenship and Development (Sociology of Transnationalization) at Bielefeld University. She earned her PhD in Sociology in 2022. Her research interests focus on prejudice, discrimination, (anti-Muslim) racism, migration and social inequalities.

Thomas Faist

Thomas Faist is Professor of Sociology at Bielefeld University. He has contributed to debates about migration, citizenship, and transnationality. His latest book publications include ‘The Transnationalized Social Question: Migration and the Politics of Social Inequalities in the Twenty-First Century’ (Oxford University Press, 2019); ‘Mobility instead of exodus: Migration and Flight in and from Africa’ (Springer Nature, 2023), and ‘Exit. Warum Menschen aufbrechen. Globale Migration im 21. Jahrhundert’ (C.H. Beck, 2022). Thomas Faist is a member of the North Rhine-Westphalian Academy of Sciences, Humanities and the Arts.