Intergroup Contact Alleviates Loneliness: The Extensive Effect of Common Ingroup Identity

Abstract

Purpose

Previous studies show that intergroup contact, through common ingroup identity, has impact on intergroup processes such as reducing intergroup bias, improving intergroup relations, etc. The effect of intergroup contact on individual psychological process (through common ingroup identity), however, needs further exploration. Based on the positive effect of both intergroup contact and ingroup identification on mental health and well-being, this article proposes and tests a new model of individual loneliness reduction through intergroup contact by promoting common ingroup identity.

Methods

A total of 263 majority ethnic members and 275 minority ethnic members from China participated in the survey. Intergroup contact, common ingroup identity and loneliness were measured at three time-points (T1-T3) over an 8-month period. Longitudinal mediation analysis and parallel process Latent Growth Curve Model for mediation are used for the examination of the indirect effect of common ingroup identity.

Results

Longitudinal mediation analysis showed that intergroup contact quality at T1 positively predicted common ingroup identity at T2, and common ingroup identity at T2 alleviated loneliness at T3. Intergroup contact quality at T1 was indirectly linked to loneliness at T3 via common ingroup identity at T2. The parallel process latent growth curve model for mediation confirmed the robustness of the indirect effect of common ingroup identity. In addition, the growth rate of the quality of intergroup contact increased the growth rate of common ingroup identity, but reduced the growth rate of loneliness.

Conclusion

The current study revealed the protectiveness of intergroup contact and common ingroup identity on loneliness, viz., intergroup contact reduces individual loneliness by promoting common ingroup identity, the implication being that intergroup contact and common ingroup identity should be taken into account in intervening process of loneliness prevention so that an individual’s physical and mental health could be better safeguarded.

Introduction

Loneliness is defined as a negative emotional perception to the discrepancy between desired and actual quality or quantity of social relationships.^1,2 It induces various kinds of negative consequences of physical and mental health.³ In order to effectively reduce loneliness and protect physical and mental health of individuals, researchers proposed several interventions, such as group interventions, one-to-one interventions,^4,5 and found that group-based interventions are more effective than one-to-one interventions.⁶ However, how do groups protect from loneliness? Some recent work raised the effect of ingroup and intergroup contact on mental health,¹ and suggested that group membership, ie, the sense of social identification derived from membership, is a strong predictor of mental health,⁷ but few studies investigated the potential mechanism of intergroup contact on mental health and loneliness. In the present study, we proposed and tested a loneliness alleviation model by intergroup contact through common ingroup identity.

Loneliness

Almost all people have experienced loneliness at some point in their life.⁸ For most people, loneliness is transient, but for some, loneliness is chronic.⁹ Loneliness severely impacts individuals’ well-being^10,11 and impairs their physical and mental health. The past studies found an association between increased levels of loneliness and multimorbidity, such as elevated blood pressure, disrupted sleep and dysregulation of neuroendocrine and immune responses;¹² and it is also related to various negative psychological symptoms, for instance, cognitive performance and cognitive decline over time,^13,14 depression^15,16 and anxiety.¹⁷

The core feature of loneliness is lacking of belonging,¹⁸ and loneliness is regarded as an important indicator of threatened belongingness needs.¹⁹ Failure to have belongingness needs met may lead to feelings of social isolation, alienation and loneliness.²⁰ Evolutionary theory of loneliness suggests that the sense of loneliness motivates individuals to reconnect with others to meet belongingness needs, and then drives off the aversive and provides the positive.²¹ Thus, a sense of belonging is not only a precursor to social connectedness but also a buffer against loneliness. Indeed, while reconnection provides individuals ailed with loneliness opportunities to connect to the outside world, grouping provides a wider connectedness to others in society, and group affiliation is conceptualized as a psychological resource contributing to well-being and adjustment.²² For example, Haslam et al, to address the lack of belonging and social disconnection central to loneliness, examined the efficacy of a novel group intervention, GROUPS 4 HEALTH (G4H), and found that G4H was associated with a greater reduction in symptoms of loneliness and social anxiety.¹⁸

The Role of Intergroup Contact

Intergroup contact refers to interactions between members of different groups.²³ The development of affiliative ties across group boundaries provides an important vehicle for achieving relational diversity.²⁴ Intergroup contact and intergroup friendships – the latter being an intimate form of intergroup contact – have been proclaimed as the antidote for reducing many of the negative consequences associated with cross-race interactions,²⁵ for new relationships afforded by contact with other groups may counteract the negative effect of isolation and loneliness.²⁶ Specifically, intergroup contact, with its reductive effect on marginalization, perception of exclusion and discrimination, which cause the sense of loneliness and low well-being, helps individuals feel understood by out-group strangers²⁵ and facilitates psychological and physical well-being in one’s daily social interactions.^25,27

Research evidences of different paradigms showed that intergroup contact may positively relieve loneliness, as both longitudinal and horizontal studies have proved the positive effects of intergroup contact on individual problem-solving, cognitive flexibility, creativity, well-being and mental health in multiple groups, including cross-hukou-regional, ethnic, gender and country.^26,28–36 Specifically, an intergenerational technology program brings disparate age groups together to interact in meaningful activities, resulting in effectively reducing loneliness for older adults.²⁹ Our hypothesis thus is that intergroup contact may alleviate loneliness.

Common Ingroup Identity as a Mediator

Putnam (2000) referred to contact between dissimilar groups as “bridging social capital” and suggested that intergroup contact allowed for sharing of information, knowledge, and ideas and could generate broader identities.^26,36 According to common ingroup theory, the more positive intergroup contact is, the stronger is the common ingroup identity. Intergroup contact promotes common ingroup identity via three stages: decategorization, salient categorization, and recategorization.³⁷ During these processes, individuals transform previous single subgroup membership into a high, inclusive level of superordinate group membership, acquiring the sense of shared identity with others and the common ingroup membership as well.

The common ingroup identity may contribute to high value, control, happiness and self-esteem similar to ingroup identity, help individuals to be capable of relieving stress and get support from former outgroup member to improve mental health. Ingroup identity is considered the premise of social connection that provides individuals with psychological resources to cope with setbacks, changes and challenges³⁸ and reduces loneliness. Group memberships enable people to perceive connection with others through the network of membership relation,³⁹ enrich people by making them stronger and healthier with self-esteem, belongingness, meaningfulness, and well-being⁴⁰ and protect people against future depression, depression relapse, or alleviating depression symptoms.⁴¹ Furthermore, once individuals identify and learn adaptive behavior norms defined by ingroup, ingroup identification provides them with psychological benefits and supports them from other ingroup members against potential stressors.

Studies have found that when an inclusive social categorization was salient, people were more likely to provide help⁴² or take the advice of outgroup.^43,44 A typical common ingroup identity, national identification, was negatively correlated with depression and anxiety and positively correlated with well-being and mental health.^22,45 Additionally, increased community identification was linked to greater perceived social support, community-esteem, personal self-esteem, and self-efficacy. These psychological processes were, in turn, linked to increased resilience and well-being.⁴⁶ A study exploring the alleviating effect of community identity on loneliness showed that community identity negatively predicts loneliness and increases individual happiness by reducing loneliness.¹¹ Thus, we suggest that the effect of intergroup contact on loneliness is similar to the process of social integration that are both meaningful and valuable. Intergroup contact promotes common ingroup identity. Individuals thus can benefit from the superordinate identification, which leads to individual loneliness reduction. To put it in line with the present study, we hypothesize that common ingroup identity may mediate the effects of intergroup contact on loneliness.

Present Study

The goal of the present study is to test the mediating role of common ingroup identity between intergroup contact and loneliness. According to previous studies, estimation bias that exists in cross-sectional mediation analysis can be effectively avoided when mediation analysis of longitudinal data is conducted, and it provides strong evidence for revealing the causal direction and mediation mechanism among variables.⁴⁷ For the analyses, we collected the three-wave longitudinal data of freshmen in a Guangxi University, from the enrollment (Time 1) to the 8th month (Time 3).

Based on previous studies, quantity and quality are two aspects of intergroup contact,⁴⁸ but quality of intergroup contact has stronger effect than quantity in promoting intergroup help,⁴⁹ intergroup attitude,⁵⁰ intergroup support,⁵¹ and happiness.⁵² Therefore, the current study adopted intergroup contact quality as the measurement of intergroup contact. Above all, we assumed that intergroup contact quality promotes common ingroup identity, and common ingroup identity reduces loneliness. In other words, common ingroup identity may play a mediating role in the process in which intergroup contact quality reduces loneliness.

There are 55 ethnic minorities widely distributed in China. For the size of the population, 5 among the 55 ethnic minorities have their own National Autonomous Regions. The Zhuang is one of the 5 big-size ethnic groups, with a population approximate to 16.93 million, of which 92.88% lives in Guangxi, a southern National Autonomous Region. Zhuang has a long history and a glorious traditional culture, as manifested in the (spoken and written) language, myths, legends, folktales, mural, brocade, drama, ethnomedicine, etc. Zhuang’s traditional costume features embroidery, their ethnic jacket, for instance, is well known for the embroidered piping, embroidered waist and embroidered shoes, coming with silver jewellery.

Materials and Methods

Study Design

A longitudinal study was conducted to examine the indirect effect of intergroup contact on loneliness via common ingroup identity. The longitudinal mediation analysis was used to test the mediational model with a longitudinal design for a rigorous inference about the causal relations among intergroup contact, common ingroup identity and loneliness;⁵³ the Latent Growth Curve Model (LGCM) was used to test the trajectories of intergroup contact, common ingroup identity and loneliness over time in form of variability in the initial levels and the variability in the growth rates among individuals;⁵⁴ and the parallel process LGCM mediation analysis was utilized to test the model of growth rate mediation.⁵³

Participants

After the exclusion of the missing responses (N = 12), participants available were a total of 538 (51.67% females) Chinese youth between 17 and 22 years of age (mean age = 18.64 years, SD = 0.96) from a university in the Guangxi Zhuang Autonomous Region in China. Of these, 263 participants are of Han majority living in other provinces before entering the university; 275 participants are of ethnic minority including 123 Zhuang ethnic minority students and 151 other ethnic minority students (e.g. Yao, Dong, Miao) living in Guangxi before entering the university. To examine the potential impact of missing data, we compared the data collected at Time 1 from participants who did and did not participate in measurement at Time 3. The result shows that there were no significant differences in observed variables (p > 0.05) between participants who finished three measurements and those who were excluded.

Procedure

Freshmen in the university were invited to participate in the test. We randomly chose 46 classes of freshman year, and then randomly invited students who are of ethnic minority in Guangxi Autonomous Region, China, and those who are majority members but came from other provinces. The assessment was conducted using paper/pencil tests in classrooms. Before the test, students were told that we aimed at gaining a better understanding of their school lives and interpersonal relationships. We stressed the importance of the sincerity of their answers and informed them of their right to refuse to participate and withdraw from the test. Informed consent was obtained from all subjects simultaneously. Following this, participants completed a set of paper-and-pencil questionnaires.

This study involved assessment at three time points, time 1 was in the first week after enrollment of new students in the university, time 2 was in three months after the enrollment, time 3 was in eight months after the enrollment. Participants took about 20 minutes to complete the measurement per time in the presence of researchers, handing in to researchers right after their completion. All assessments were administered by well-trained professionals with psychology background.

Ethical Statements

The current research complies with the Declaration of Helsinki. It was also approved by the human research Scientific Review Committee of Southwest University of the People’s Republic of China (H20074).

Measures

The questionnaires used in the study came from published literature, and the two-way translation procedures were adopted to ensure the consistency of content and meaning in both the Chinese and English versions.

Intergroup Contact Quality

The measurement of intergroup contact quality used a five-item scale⁵⁵ (eg, for Han sample and the other ethnic minority sample, one of the questions is “When you meet with Zhuang students, in general, do you find the contact to be equal?”). Participants responded to these items on a 7-point Likert scale (from 1 = strongly disagree to 7 = strongly agree). Internal consistencies of the 5-item scale were satisfactory, with Cronbach’s alphas 0.95 at T1, 0.94 at T2, and 0.94 at T3, respectively. The results of CFA indicated that structural validity of the model was acceptable in the first wave (χ²/df = 1.52, CFI = 0.99, TLI = 0.98, RMSEA = 0.03, SRMR = 0.005), the second wave (χ²/df = 3.72, CFI = 0.99, TLI = 0.99, RMSEA = 0.07, SRMR = 0.007), and the third wave (χ²/df = 2.22, CFI = 0.99, TLI = 0.99, RMSEA = 0.05, SRMR = 0.006).

Common Ingroup Identity

We measured common ingroup identity with an item adapted from previous research.⁵⁶ Participants rated the extent of agreement with the statement: “our ethnic group and Zhuang share a common group” (1 = strongly disagree; 7 = strongly agree).

Loneliness

Loneliness was measured using a brief R-UCLA Loneliness Scale.⁵⁷ There were three items, such as “I feel left out” (from 1 = strongly disagree to 7 = strongly agree). The internal consistencies of three waves were 0.914, 0.921, and 0.937, respectively. The results of CFA indicated that structural validity of the model was saturated model in three wave, indicating that they perfectly fit the data.

Covariant

The number of friends is used as the control variable. There was one item to ask how many Zhuang friends do you have at the three points⁵² (1 = 0, 2 = 3, 3 = 3–5, 4 = 6–10, 5 = 10–30, 6 = more than 30).

Statistical Analysis

Descriptive statistics and correlation analysis were calculated for each variable. Next, we established the model with the intergroup contact quality at T1 as the independent variable, common ingroup identity at T2 as the mediator, loneliness at T3 as the dependent variable, and the number of friends as the control variable of independent variable and the dependent variable, the product of coefficients of two paths as the test of indirect effect. The significance of the indirect effects was evaluated using the bootstrap procedure (5000 draws). The expressions are following:

(1)

(2)

(3)

In (1), (2) and (3), X, M, Y refer to independent variable, mediated variable, and dependent variable, respectively, , and refer to autoregressive effect, a and b refer to lag effect, the product of a and b refers to indirect effect, refers to residual, and t, t+1 and t+2 refer to time point that repeat measured.

Although the longitudinal design approach is one of the analytical strategies for determining the directionality of effects between variables, repeated measurement of variables in a longitudinal design introduces additional within-person variability. In order to ensure methodological precision, we also used Latent Growth Curve Model (LGCM), which captures the collection of individual trajectories over time in the form of fixed and random effects⁵⁸ to estimate the individual variation and group variation in the development process at the same time (Figure 1). LGCM of every variable has an average initial status α and an average growth rate S. The variances of the intercept and the slope factors estimate the variability in the initial levels and the variability in the growth rates among individuals.⁵³

Figure 1.

Diagram of Latent growth curve model. V1, V2 and V3 are the values of three repeated measurements, E1, E2 and E3 are the random measurement errors at three time points. α and S enclosed in circles represent the unobserved Intercept and Slope factors respectively; M_i and D_i are the mean and variance of the unobserved Intercept factor, M_S and D_S are the mean and variance of the unobserved Slope factor.

As shown in Figure 1, V1, V2 and V3 are the values of three repeated measurements. E1, E2 and E3 are the random measurement errors at three time points. The two variables enclosed in circles represent the unobserved Intercept and Slope factors.⁵⁹ The two factors both have a mean (M) and variance (D) across the whole sample.⁵⁸ The expressions are following:

(4)

(5)

where α refers to intercept, S refers to slope, μ refers to mean, refers to variance.

Thus, we conducted the LGCM of every variable, and then built a parallel process LGCM based on the longitudinal mediation model. By setting regression equations among growth factors, we can better reveal the trajectory of intergroup contact quality, common ingroup identity and loneliness, and relationship among them.

Results

Correlation Analysis

Means, standard deviations, and intercorrelations for intergroup contact quality, common ingroup identity, loneliness at each wave of measurement are presented in Table 1. An inspection of the means and variability reveals highly stability. At the three points, the quality of intergroup contact was significantly positively correlated with common ingroup identity, and significantly negatively correlated with loneliness. Common ingroup identity was significantly negatively correlated with loneliness.

Table 1.

Descriptive Statistics and Correlation of Intergroup Contact Quality, Common Ingroup Identity and Loneliness

Variable	Mean	SD	1	2	3	4	5	6	7	8
1.T1 ICQ	5.89	1.02	–
2.T2 ICQ	5.98	0.93	0.60***	–
3.T3 ICQ	6.01	0.90	0.51***	0.58***	–
4.T1 CII	6.09	1.24	0.33***	0.26***	0.25***	–
5.T2 CII	5.98	1.13	0.28***	0.39***	0.29***	0.43***	–
6.T3 CII	5.99	1.13	0.28***	0.36***	0.41***	0.33***	0.45***	–
7.T1 Loneliness	3.01	1.36	−0.30***	−0.28***	−0.18***	−0.20***	−0.16***	−0.12**	–
8. T2 Loneliness	2.93	1.36	−0.34***	−0.40***	−0.30***	−0.18***	−0.19***	−0.17***	0.56***	–
9.T3 Loneliness	2.72	1.32	−0.30***	−0.35***	−0.38***	−0.16***	−0.23***	−0.20***	0.46***	0.61***

Notes: **p<0.01, ***p<0.001. T1, T2, T3 are points-in-time that variables are measured; SD, standard deviation.

Abbreviations: ICQ, intergroup contact quality; CII, common ingroup identity.

Longitudinal Mediation Analysis

Taking the number of friends as the covariate of loneliness and intergroup contact quality, we tested the indirect effect of common ingroup identity between intergroup contact quality and loneliness (Figure 2). The results showed that the model fitted well, χ² = 136.99, df = 39, CFI = 0.94, TLI = 0.90, RMSEA = 0.07, SRMR = 0.07. The quality of intergroup contact significantly positively predicted common ingroup identity at T2, and common ingroup identity at T2 significantly negatively predicted loneliness at T3. The indirect effect of intergroup contact quality at T1 on loneliness at T3 through common ingroup identity at T2 was significant as the bias corrected 95% confidence intervals was [−0.046, −0.006], the indirect effect did not include zero.

Figure 2.

Longitudinal mediation model.

Notes: All parameter estimates in the figure are unstandardized results. The control variable is not shown in the figure.

Parallel Process LGCM for Mediation Analysis

In order to test trajectories of intergroup contact, common ingroup identity and loneliness over time, we constructed no-growth model and unconditional linear growth model for every variable. The goodness of fit of the models for each variable is shown in Table 2. The results showed that the goodness of linear growth models was better for all three variables. The initial value and rate of change for each variable are shown in Table 3.

Table 2.

Comparisons of Fitted Growth Curve Models for the Variables

Variables	Models	χ2	df	CFI	TLI	RMSEA	SRMR
Intergroup contact quality	No-growth model	32.66	6	0.95	0.97	0.09	0.23
	Linear growth model	2.96	1	1.00	0.99	0.05	0.01
Common ingroup identity	No-growth model	10.31	3	0.97	0.97	0.07	0.06
	Linear growth model	2.26	1	1.00	0.98	0.05	0.02
Loneliness	No-growth model	13.46	6	0.97	0.98	0.07	0.06
	Linear growth model	0.24	1	1.00	1.00	0.00	0.01

Abbreviations: χ², chi-square test; df, degree of freedom; CFI, comparative fit index; TLI, Tucker-Lewis index; RMSEA, root-mean-square error of approximation; SRMR, standardized root mean square residual.

Table 3.

Estimate of Univariate LGCM

Variables	Intercept		Slope		Covariances
	Mean	Variances	Mean	Variances
Intergroup contact quality	5.91	0.63	0.01	0.003	−0.46
Common ingroup identity	5.65	0.67	0.02	0.01	−0.36
Loneliness	3.02	1.14	−0.04	0.02	−0.29

The model fit of intergroup contact quality was good, χ² = 2.96, df = 1, CFI = 1.00, TLI = 0.99, RMSEA = 0.05, SRMR = 0.01. The mean of intercept factor was 5.91, SE = 0.04, p < 0.001, the mean of slope factor was 0.01, SE = 0.01, p < 0.01. The variation of intercept factor was greater than zero (σ² = 0.63, SE = 0.06, p < 0.001), indicating that there was significant individual difference on the initial level of contact quality, and the variation of slope factor was marginal significant (σ² = 0.003, SE = 0.002, p > 0.05). The correlation between intercept and slope was significant (r = −0.46, p < 0.001), indicating that the higher the initial contact quality of individuals, the slower the growth speed of their contact quality.

There was no significant change except for mean slope and variation of slope factor when controlled the number of friends. Specially, the mean of intercept factor of contact quality was 5.91, SE = 0.074, p < 0.001, the mean of slope factor was 0.01, SE = 0.005, p < 0.01. The variation of intercept was greater than zero (σ² = 0.62, SE = 0.06, p < 0.001), the variation of slope was insignificant (σ² = 0.003, SE = 0.002, p > 0.05). The correlation between intercept and slope is significant (r = −0.02, p < 0.01).

The model fit of unconditional LGCM for common ingroup identity was good, χ² = 2.26, df = 1, CFI = 1.00, TLI = 0.98, RMSEA = 0.05, SRMR = 0.02. The mean of intercept factor of common ingroup identity was 6.06, SE = 0.51, p < 0.001. The mean of slope factor was −0.01, SE = 0.09, p = 0.17. The variation of intercept factor was greater than zero (σ² = 0.67, SE = 0.09, p < 0.001), indicating that there was significant individual difference at the initial level of common ingroup identity. The variation of slope factor was significant (σ² = 0.01, SE = 0.01, p<0.05). The correlation between intercept and slope was significant (r = −0.36, p < 0.001), indicating that the higher the initial common ingroup identity of individuals is, the slower is the growth speed of common ingroup identity.

Finally, we tested the developmental trajectories of loneliness. The model fitted well, χ² = 0.24, df = 1, CFI = 1.00, TLI = 1.00, RMSEA = 0.00, SRMR = 0.01. The mean of intercept factor of loneliness was 3.02, SE = 0.15, p < 0.001, the mean of slope factor was −0.04, SE = 0.07, p < 0.001. Both of the variation of intercept (σ² = 1.14, SE = 0.17, p < 0.001) and slope factors (σ² = 0.02, SE = 0.01, p < 0.001) were significant. The correlation between intercept and slope was significant (r = −0.29, p < 0.001), indicating that the higher the initial loneliness of individuals is, the slower is the growth rates of loneliness. The mean and variation of intercept factor and slope factor remained significant when controlled for the number of friends.

The independent variable LGCM, mediator LGCM and outcome variable LGCM are combined to establish a parallel process latent growth model for mediation, including the initial level mediation (α_X→α_M→α_Y) and growth rate mediation (S_X → S_M → S_Y).⁴⁷ Based on the longitudinal mediation model, we conducted the parallel process LGCM mediation analysis. Because we were interested in relations among slope of three variables, we only test the model of growth rate mediation. To investigate indirect effects of growth rates not confounded by initial differences between subjects, associations between intercept factors and slope factors in variables were also added as controls. The model is shown in Figure 3. The model fitted well, χ² = 106.10, df = 45, CFI = 0.96, TLI = 0.94, RMSEA = 0.05, SRMR = 0.05. The path coefficients are shown in Table 4. The slope of intergroup contact quality significantly positively predicted the slope of group identity within the community, B = 1.10, 95% CI [0.37, 2.96], the slope of common ingroup identity negatively predicts the slope of loneliness, B = −1.37, 95% CI [−3.31, −0.11], the 95% confidence interval of indirect effect was [−3.83, −0.21], did not include zero.

Figure 3.

Parallel process LGCM for mediation.

Notes: X is independent variable (ingroup contact quality); M is mediator (common ingroup identity); Y is outcome (loneliness). T1, T2, T3 are points-in-time that variables are measured. αx is intercept (initial status) factor of independent variable, αM is intercept (initial status) factor of mediator; αY is intercept (initial status) factor of outcome; SX is slope (growth rate) factor of independent variable; SM is slope (growth rate) factor of mediator; SY is slope (growth rate) factor of outcome. a is the effect of SX on SM, b is the effect of SM on SY. The dotted lines in the figure indicate that associations between intercept and intercept, slope and intercept were controlled.

Table 4.

Standardized Beta Coefficients, and Indirect Effect with 95% Bootstrap Confidence for Parallel Process LGCM

Paths	Estimator	95% CI
ICQ slope→CII slope(a)	1.10	[0.37, 2.96]
CII slope→Loneliness slope(b)	−1.37	[−3.31,-0.11]
ICQ slope→CII slope→Loneliness slope(a×b)	−1.50	[−3.83,-0.21]

Note: 95% CI is 95% confidence intervals.

Abbreviation: ICQ is intergroup contact quality; CII is common ingroup identity.

Additional Analysis

Based on the longitudinal mediation analysis, we explored differences between minority ethnic students and majority ethnic students and focused on the differences of paths that from contact quality at T1 to common ingroup identity at T2 and from common ingroup identity at T2 to loneliness at T3 and longitudinal mediation. The results showed (Table 5) that the difference of path coefficients and indirect effect between two groups were not significant.

Table 5.

Longitudinal Mediation Comparison of Two Groups

Paths	Minority		Majority		Comparison
	Estimator	95% CI	Estimator	95% CI
T1 ICQ→T2 CII(a)	0.19	[0.05, 0.28]	0.17	[0.05, 0.25]	0.02
T2 CII→T3 loneliness(b)	−0.14	[−0.21, −0.02]	−0.10	[−0.18, 0.00]	−0.04
T1 CQ→T2 CII→T3 loneliness(a×b)	−0.03	[−0.07, −0.01]	−0.02	[−0.04, −0.002]	−0.01

Notes: T1, T2, T3 are points-in-time that variables are measured. 95% CI is 95% confidence intervals.

Abbreviations: ICQ, intergroup contact quality; CII, common ingroup identity.

Discussion

In the present study, we conducted a longitudinal study to examine the relation among intergroup contact, common ingroup identity and loneliness. The results showed that intergroup contact could alleviate individual loneliness via promoting common ingroup identity. Furthermore, the growth rate of the quality of intergroup contact can increase the growth rate of common ingroup identity, resulting in reducing the growth rate of loneliness.

Satisfying social relationships are vital for one’s mental and physical health. At the individual level, satisfying interpersonal relationships with others can promote mental and physical health.⁶⁰ At the group level, positive intergroup contact can also promote mental health, reduce the impact of stress on individuals and eliminate negative emotions, because positive intergroup contact can create similarity and then weaken the salience of group boundaries,⁶¹ generate broader identities,⁶² reduce intergroup anxiety,⁶³ provide the knowledge and information,⁶⁴ provide social support^64,65 and a sense of social integration and belongingness;⁶⁶ longitudinally predict less social distance and more prosocial behavior.⁶⁷ Lonely individuals are more prosocial in the public situation that they can gain appreciation and a reputation, because they have a motivation to reconnect,⁶⁸ and prosocial behaviors can establish or enhance one’s relational value or attachment with others, which may resolve loneliness.¹⁹ Similarly, close social distance and social touch provide an opportunity for them to reconnect with our social network, and thus reduce loneliness.⁶⁹ In addition, the higher the quality of group contact, the more likely are the individuals to form friendships with previous outgroup members. Cross-group friendships promote common ingroup identity through social integration,⁷⁰ and common ingroup identity provides group membership and promotes the sense of belonging, which helps receiving greater numbers of ingroup support,⁴⁶ thereby reducing loneliness.⁴⁵ The positive effect of common ingroup identity is to be expected because individuals are more likely to provide social support to ingroup than outgroup members and are more likely to accept help from ingroup members.⁷¹ At the same time, increasing social activity, specifically through acquiring group memberships, may be a vital, curative ingredient in the treatment of loneliness.

The present study expands our knowledge about the individual-level effect of intergroup contact. As a group process, intergroup contact acts on various group psychology and behavior, such as the reduction of intergroup prejudice, and the improvement of intergroup relation and attitude.⁷² Nevertheless, a small amount of study found individual-level effect of intergroup contact, such as its positive effects on psychological capital for individuals, its provision of social support, and increasing effect on well-being.³⁶ However, how intergroup contact works on reducing negative emotions at individual level is not fully understood. We found the impact of intergroup contact on loneliness, which expands our understanding about the individual-level effect of intergroup contact except for well-being, expanding intergroup contact theory, providing evidence for relation between intergroup and individual psychological process, and also contributing to further understanding the connection between group psychological and individual psychological processes.

The findings of the present study provide new evidence for social cure of social identity. Group is a valued factor working on individual mental health. Based on the group identity approach, group identity affects individual mental well-being via self-esteem,⁷³ social support,⁷⁴ control perception and attribution style. The social cure of social identity puts forward the positive meaning of group membership⁴¹ and multiple group identification^71,75 for individual mental health and social adaptation, because different groups provide different types of social support. The study found that a higher, more inclusive superordinate group identification has a positive effect on relieving loneliness.

The results of the parallel process LGCM for mediation analysis showed that the growth rate of the quality of intergroup contact increased the growth rate of common ingroup identity and then reduced the growth rate of loneliness. This dynamic result pattern suggests that we should pay more attention to the growth rate of contact quality in the process of group-based interventions on loneliness and the pattern provides a fresh perspective that we should be more active and sincere when contact with others. In the future, we could explore other effective methods to improve the quality of intergroup contact and the effectiveness of these methods to reduce loneliness of mobile groups, such as cross-group friendship, intergroup attitude, amongst others. In addition, ERP and fMRI can also be used to investigate the neural relationship between intergroup contact and loneliness.

Contributions and Limitations

The current study has some contributions as following. It extends the individual-level effect of intergroup contact, as well as the individual-level effect of common ingroup identity. They jointly contribute to the alleviation of individual loneliness. This increases our knowledge about the “social cure” and the health benefits that originally arise from ingroup processes, with the indication that this benefit can be expanded by intergroup contact including former outgroup members. Meanwhile, our knowledge is also expanded to the dynamic development process among intergroup contact, common ingroup identity and loneliness, ie, the faster intergroup contact quality develops, the more common ingroup identity generates and the lower loneliness people would experience.

Despite these contributions, there are several limitations to be addressed for future studies. Although the parallel process LGCM for mediation is based on measurements at three different time points, the time durations are relatively short; therefore, it provides suggestions for short-term interventions only and cannot provide further detailed evidences for the developmental trajectories of intergroup contact for a longer time. A study on the developmental trajectories of loneliness by Liu et al, followed up for three years, found a nonlinear growth pattern of loneliness.⁷⁶ Therefore, future researches may establish a nonlinear LGCM by extending the tracking time so as to comprehensively explore the long-term relationships among the three. With previous studies indicating that contact quantity is also effective in improving intergroup relations,⁷⁷ for a better understanding of the relationship between intergroup contact and loneliness, more research efforts need to be made on how intergroup contact quantity alleviates loneliness. In addition, individual factors such as “The Big Five”, political orientations and social dominance orientation should also be taken account of in our future studies, because according to the concepts of situational evocation, social investment theory, and self-perception theory, intergroup contact and individual traits are bidirectionally associated.⁷⁸ What worth mentioning is that this study used self-report questionnaires, which are subject to social desirability bias. In future research, more and various measuring tools need to be employed; peer nominations, for instance, are to be employed to examine the role played by intergroup contact in alleviating loneliness via common ingroup identity. Last, but not the least, loneliness is also found to be related to suicidal behavior,⁷⁹ which is in turn frequently related to depressive symptoms,⁸⁰ and explorations on the effect of intergroup contact on those two are thus especially needed in our future studies.

Implications and Conclusion

The findings in the present study suggest that intergroup contact quality alleviates loneliness via common ingroup identity. Furthermore, the growth rate of the quality of intergroup contact can increase the growth rate of common ingroup identity, resulting in reducing the growth rate of loneliness. It has implications for prevention and intervention for cross-cultural transition, communication in multi-ethnic or multi-racial countries. For individuals, it is beneficial to reducing loneliness by intergroup contact and then constructing common identity with former outgroup members. One could join certain groups to get memberships, take part in intergroup activities with a positive attitude and full sincerity. For groups, building platforms for group members to alleviate loneliness and get health and well-being by organizing intergroup events and activating common ingroup identity between ingroup and outgroup in the process would be an important contribution for constructing harmonious society.

Funding Statement

This research was supported by the Major Project for Key Research Institutes of Humanities and Social Science by the Ministry of Education of China (Number 16JJD190007), the Key Project of Guangxi Education Science Plan (2023A013), the basic scientific research ability improvement project for the young and middle-aged teachers in Guangxi universities (2021KY0609).

Abbreviation

LGCM, latent growth curve model.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors report no conflicts of interest in this work.