How Does the Social Support Affect Refugees’ Life Satisfaction in Turkey? Stress as a Mediator, Social Aids and Coronavirus Anxiety as Moderators

Abstract

The number of refugees has increased exponentially due to international crises, wars, and political pressures in recent years worldwide. Turkey hosts the largest refugee population in the world with 3,672,646 Syrian refugees. This study aimed to examine the relationship among refugees’ stress, life satisfaction, social support, coronavirus anxiety, and social aids they get during the COVID-19 process. We hypothesized that stress, social aids, and coronavirus anxiety could play a role in the relationship between life satisfaction, stress, and social support. A cross-sectional survey was used to collect data from 628 Syrian refugees via an online questionnaire in Turkey. Confirmatory factor, correlation, and multiple regression analyses were conducted. In addition, the moderator and mediator role of variables tested using the 95% bias-corrected confidence interval from 5000 resamples was generated by the bias-corrected bootstrapping method. The study results show that stress partially mediated the relationship between family support and life satisfaction and between friend support and life satisfaction. In addition, coronavirus anxiety had a moderating effect on the relationship between family support and stress and friend support and stress. Finally, social aids moderated the relationship between stress and life satisfaction. In conclusion, the role of social aids, stress, and coronavirus anxiety in the relationship between social support, stress, and life satisfaction has been revealed.

1. Introduction

Migration is defined as the temporary or permanent leaving of individuals or communities from their places of residence for some reason [1]. As a result of the anti-government protests in the Arab geography in late 2010 due to the events in the Middle East in recent years, a humanitarian crisis arose in Syria in March 2011 [2]. This crisis first turned into a civil war in Syria and then an international issue affecting the region [3]. With the effects of war in Syria, almost half of Syria’s total population migrated to neighboring countries, especially Turkey. Turkey had an open-door policy in this process. For this reason, the number of refugees has increased rapidly every day. According to June 2021 data of the Directorate General of Migration Management, there are 3,672,646 registered Syrians in Turkey [4].

Due to migration, refugees face some difficulties. The most important of these difficulties are economic problems, health problems, and access to healthcare that refugees experience due to legal and physical barriers, adaptation problems that refugees experience due to language and cultural differences, and education problems [5]. Another difficulty encountered by refugees is psychological problems [6]. All difficulties refugees face in the migration process might cause stress on them [7], and stress negatively affects individuals’ lives, reducing their life satisfaction. Life satisfaction is a concept related to individuals’ evaluation of how well they live according to their expectations [8]. Therefore, individuals need coping mechanisms to reduce stress’ negative effects and increase their life satisfaction. One of the important mechanisms that refugees use to cope with their stressors during the migration process is social support networks [9].

Social support networks consist of family, friends, relatives, and a special person who can support individuals and reduce their stress [10]. Social support is too important, especially for refugees who migrated due to forced reasons such as war, political pressure, and fear of death [11]. Since refugees who had to migrate have left behind many material and moral losses. After the psycho-social and economic difficulties experienced in the migration process, it is challenging to try to have a new life in a different country, but social support networks empower individuals in such situations and make them feel safe [10]. In this context, social support networks often become prominent. However, refugees experiencing forced migration need not only social support but also social aids due to the financial losses they experience during the migration process [12]. In the literature, social aids, which can be an important source of support for refugees, has not been examined as much as a social support network. For this reason, it is important to fill this gap in the literature to discuss the life satisfaction of refugees comprehensively.

Many factors that may create stress and anxiety in refugees were mentioned above. However, in addition to these, the COVID-19 pandemic must be mentioned. The COVID-19 pandemic affected many fields, such as medical, social, political, economic, etc., and has many outputs such as anxiety, stress, and depression [13]. Therefore, it is important to discuss the coronavirus anxiety, which might affect refugees.

We started our study with a research question. This research question is: How do coronavirus anxiety and social aids affect the indirect effect of stress in the relationship between social support and life satisfaction? The research tried to understand the effect of coronavirus anxiety and social aids on the indirect effect of perceived stress in the relationship between social support and life satisfaction. Even though refugees use the social support mechanism to cope with stress, coronavirus anxiety is an important variable that may affect stress during the pandemic period. In addition, although it is known that stress reduces the life satisfaction of refugees, the role of getting social aid in the pandemic period has not been sufficiently investigated. For this reason, it is crucial to examine coronavirus anxiety and the role of social aids. In addition, investigating coronavirus anxiety and the role of social aids will contribute to understanding the refugees’ stress and life satisfaction during the COVID-19 pandemic.

2. Literature Review and Hypothesis Development

Migration is a challenging and multidimensional process. In this process, refugees might face psycho-social, economic, education, and health-related problems [14]. These problems may create stress on refugees, and they may suffer a lot from the negativities of the process [15]. The COVID-19 pandemic process might be given as an example of this situation. The COVID-19 crisis is a process that creates psychological effects on individuals [16]. This process affects refugees negatively in psycho-social aspects. Stress, which is one of the negative effects, disrupts the harmony of individuals in the communication between people and the environment [17]. Individuals encounter different sources of stress throughout their lives and develop other coping mechanisms. One of these coping mechanisms is social support. Social support consists of an individual’s social resources such as family, friends, and a special person [18,19]. Cobb [20] mentions that social support makes individuals feel respected and valued. In addition, it helps individuals feel better and try different ways to cope with stress [21]. In this context, the following hypotheses were developed to examine the effect of social support on perceived stress:

Hypothesis 1 (H₁).

Family (a), friend (b), and special person (c) support have a negative effect on perceived stress.

According to the ecological system theory, one of the environments in which the individual interacts is the micro-environment. The micro-environment consists of groups such as an individual’s family, relatives, friends, and peers. There is a constant interaction between the individual and individual’s environment, and this interaction ensures the system’s sustainability by being in balance. If this system cannot ensure a state of balance, it is not possible to mention the well-being and social functionality of the individual. One of the variables of this study is life satisfaction, which is a concept related to how people evaluate their own lives [8]. Some determinants of life satisfaction are psychological well-being, having functionality, initiating and maintaining social relationships, and having social connections [22]. If the balance of the micro-system of the individual is disturbed, the interaction with an individual’s close environment such as family, friends, and peers in the micro-system is damaged. Therefore, this situation may affect the individual’s life satisfaction by damaging the individual’s well-being, social functionality, and social relations. Situations that harm the individual’s well-being, functionality, and relationships may also create stress in the individual. The situations that cause stress in the individual are not only related to the balance of the micro-system but also include the situations that occur in the meso and macro environment of the individual. Any situation that causes stress in the individual may affect the life satisfaction of the individual. In this context, the following hypotheses were developed to examine the effect of perceived stress and social support on life satisfaction [23,24]:

Hypothesis 2 (H₂).

Perceived stress has a negative effect on life satisfaction.

Hypothesis 3 (H₃).

Family (a), friend (b), and special person (c) support have a positive effect on life satisfaction.

Social support sources, which are likely to contribute to life satisfaction, might also affect individuals’ stress states. In addition, stress sources that disrupt individuals’ adaptation might also reduce the quality of life of individuals [25]. In this context, the hypotheses created to test the role of perceived stress in the relationship between social support and life satisfaction are as follows:

Hypothesis 4 (H₄).

Perceived stress has a mediator role on the effect of family (a), friend (b), and special person (c) support on life satisfaction.

Another factor that may affect the life satisfaction of individuals is the economic status of individuals. Frijters et al. [26] mention the effect of economic status on life satisfaction in their study. Furthermore, it is well-recognized that refugees who suffer many financial losses and need help experience stress [27]. For this reason, it is thought that social aids, which are related to individuals’ economic status, may take a role in the relationship between perceived stress and life satisfaction. In this context, the following hypothesis has been created:

Hypothesis 5 (H₅).

Getting social aid has a moderation role in the effect of perceived stress on life satisfaction.

As mentioned above, social support is one of the mechanisms for coping with stress. The individual, a social being, continues life by integrating with the social environment in every sphere of life. The support of family, friends, and a special person makes individuals feel stronger, improves their positive thoughts, and facilitates their problems. Therefore, individuals’ social support from their environment might affect the stress experienced by individuals [28]. In addition to the issues brought about by migration, refugees are also exposed to various psycho-social effects of the COVID-19 process. COVID-19 significantly affects all segments of society and changes everyday life.

In addition, it causes some uncertainties by affecting everyday life and creates anxiety in individuals [29]. In this context, the following hypotheses were developed to determine the role of coronavirus anxiety in the relationship between social support and perceived stress:

Hypothesis 6 (H₆).

Coronavirus anxiety has a moderation role in the effect of family (a), friend (b), special person (c) support on perceived stress.

In Figure 1, the research model is shown in accordance with the hypotheses formed. The research model shows social support’s (family, special person, and friend support) direct effect on life satisfaction. Additionally, the mediator role of perceived stress between social support and life satisfaction relationship is also illustrated. The moderator role of COVID-19 anxiety in the relationship between social support and perceived stress is demonstrated. In addition, the moderator role of social aids in the relationship between perceived stress and life satisfaction is shown.

3. Materials and Methods

3.1. Participants and Procedure

According to the Directorate General of Migration Management, 3,672,646 registered Syrians live in Turkey [4]. This cross-sectional study was conducted with 628 Syrian refugees living in different provinces of Turkey between September 2020 and January 2021. Due to the COVID-19 pandemic, the online survey method was preferred in collecting research data. In this context, it aimed to reach Syrian refugees residing in different provinces of Turkey. Firstly, authors living in different provinces reached Syrian refugees through their environmental resources (neighbors, colleagues, etc.). With the reached refugees’ help, information about the study was transmitted to other Syrian refugees, and the contact information of the Syrian refugees who were considering participating in the study was obtained with their permission. Afterward, they were contacted through this contact information and asked whether they would participate in the study. The questionnaire form was sent to 708 people who agreed to participate in the study. In total, 673 of 708 (95.1%) participants responded. Therefore, the number of valid questionnaires within the participants’ responses is 628 (93.3%). Before data collection, a general information form regarding the study was submitted to the participants. With this form, the participants were informed about the purpose of the study, the scales to be used, and the confidentiality of information. Citizens of Syrian, aged 18 and over, and who agree to participate in the study were included in this study.

3.2. Measures

A questionnaire form consisting of five parts was created as a data collection tool in this study. In the first part of the form, there are demographic questions about the participants and a question about getting social aid. The second part of the form includes the “Life Satisfaction Scale”, the third part includes the “Multidimensional Perceived Social Support Scale”, the fourth part includes the “COVID 19 Anxiety Scale”, and the last part includes the “DASS-21 Scale” stress sub-dimension. The questionnaire form was provided to the participants in Arabic. The Coronavirus Anxiety Scale’s and Multidimensional Scale of Perceived Social Support’s questions were translated and back-translated to Arabic from Turkish by two native Arabic interpreters. In addition, The Satisfaction with Life Scale’s and Depression Anxiety Stress Scale-21 (DASS-21)-Stress Dimension’s questions were translated and back-translated to Arabic from English by two native Arabic interpreters.

Demographics: Several questions were asked to get information about the participants. In this part, the participants stated their age, gender, education level, household income, and whether they got social aid or not. Information about the social aid variable, which is the moderator variable, was obtained with the question: “Which of the following do you get social aid from?” The options for the question are as follows: (a) The Red Crescent, (b) Support to Life Association, (c) ASAM (Asylum Seekers and Immigrants Solidarity Association), (d) Municipalities, (e) Local Associations, (f) Other (…), and (g) I don’t get any social aid. The responses were categorized as yes and no. Options a, b, c, d, e, and f were accepted as “yes (I get social aid)” and option g was accepted as “no (I don’t get any social aid)”.

The satisfaction with life scale (SWLS): This scale was designed to determine individuals’ satisfaction from their lives. The scale was developed by Diener et al. [30]. In the calculation of the scale, a 5-point Likert-type scale was used. The scale consists of 5 items, and the scale’s internal consistency coefficient (Cronbach’s α) was calculated to be 0.88. In addition, the criterion validity of the scale was provided [30].

Multidimensional Scale of Perceived Social Support (MSPSS): This scale was designed to measure the social support individuals perceive from their environment. The scale was reviewed and redesigned by Eker et al. [31]. The scale is a 7-point Likert type. The scale consists of 12 items. The internal consistency coefficient (Cronbach’s α) of the scale was found to be 0.89. In addition, the construct validity of the scale was provided. The scale consists of three sub-dimensions: family support, friend support, and special person support [31].

Depression Anxiety Stress Scale-21 (DASS-21)-Stress Dimension: This scale was designed to measure the depression, anxiety, and stress individuals perceive from their environment. Lovibond and Lovibond [32] developed the scale. A 4-point Likert-type scale was used in the calculation of the scale. There are 21 items on the scale, and the internal consistency coefficient was calculated to be 0.81. The scale consists of three sub-dimensions: depression, anxiety, and stress, each containing seven items. The stress sub-dimension of which validity and reliability were determined was used in this study [32].

Coronavirus Anxiety Scale: This scale was developed by Lee [33] to identify possible cases of dysfunctional anxiety associated with the COVID-19 crisis. The scale was adapted to Turkish by Biçer et al. [34], and the adapted form was used in this study. The scale consists of five items and one dimension in total. The internal consistency coefficient was determined as 0.832, and the construct validity was also provided [34].

3.3. Statistical Analyses

Firstly, descriptive statistics and correlations were analyzed using SPSS software (version 23.0). Second, multiple regression analysis was used to test the first seven hypotheses (H_1a, H_1b, H_1c, H₂, H_3a, H_3b, and H_3c). Third, we conducted a mediation analysis using Model 4 of Hayes’ PROCESS macro to examine the mediation role of perceived stress [35]. Finally, we conducted a moderation analysis using Model 21 of the PROCESS macro to determine the moderating effect of coronavirus anxiety and getting social aid. Moderating and mediating effects in the model were tested using the 95% bias-corrected confidence interval from 5000 resamples generated by the bias-corrected bootstrapping method.

4. Results

4.1. Frequencies of Demographic Variables

The characteristics of the participants are presented in

Table 1. Sample characteristics.

Variables	n	%
Gender
Male	280	44.6
Female	348	55.4
Total	628	100
Age
18–25	229	36.5
26–35	245	39.0
36–45	105	16.7
46–65	49	7.8
Total	628	100
Education status
Primary School Graduate	66	10.5
Secondary School Graduate	101	16.1
High School Graduate	151	24.0
University and Post Graduate	310	49.4
Total	628	100
Household income
2000 TL1 ($229) and less	268	42.7
2001–4000 TL ($230–$460)	218	34.7
4001–6000 TL ($461–$692)	86	13.7
6001–9000 TL ($693–$1038)	34	5.4
9001 TL ($1039) and more	22	3.5
Total	628	100
Do you get social aid?
Yes	180	28.7
No	448	71.3
Total	628	100

¹ Turkish Lira.

. As seen in Table 1, more than half of the participants (348 people) are women, 39.0% of the participants are between the ages of 26 and 35 (245 people), and 36.5% of them are between the ages of 18 and 25 (229 people). Nearly half (49.4%) of the participants were university graduates (310 people), and almost half of them (268 people) had an income of 2000 Turkish Lira ($229) or less. In addition, 71.3% of the participants stated that they did not get social aid.

4.2. Confirmatory Factor Analysis Results

Table 2. Confirmatory factor analysis.

Indexes	Values	Acceptable Values
CMIN/DF	2.525	<5
GFI	0.904	>0.90
CFI	0.939	>0.90
NFI	0.903	>0.90
RMSEA	0.049	<0.08

CMIN/DF = chi-square/degree of freedom; GFI = goodness of fit index; CFI = comparative fit index; NFI = normed fit index; RMSEA = root mean square error of approximation.

shows the results of the confirmatory factor analysis.

When Table 2 is examined, it is seen that the goodness of fit indices have acceptable values. Therefore, these results show that the model is compatible with the data and acceptable [36].

4.3. Correlation Analysis Results

The relationships of the main variables of this study with each other were examined by correlation analysis, and descriptive statistics for these variables were calculated. These results are presented in

Table 3. Means, standard deviations, and correlations of the main study variables.

Variables	1	2	3	4	5	6
1. Family support	–
2. Friend support	0.550 **	–
3. Special person support	0.385 **	0.554 **	–
4. Perceived stress	−0.130 **	−0.088 *	0.049	–
5. Life satisfaction	0.344 **	0.299 **	0.158 **	−0.239 **	–
6. Coronavirus anxiety	−0.036	−0.018	0.077	0.269 **	−0.014	–
M	5.13	4.40	4.05	2.17	2.17	1.41
SD	1.42	1.52	1.64	0.70	0.88	0.66

* p < 0.05; ** p < 0.01.

.

As a result of the correlation analysis, it was aimed to reveal the relationships between the main variables of the study. The results show a negative relationship between family and friend support and perceived stress (p < 0.05). However, there is no significant relationship between special person support and perceived stress (p > 0.05). There is also a weak negative correlation between perceived stress and life satisfaction (r = −0.24, p < 0.01).

4.4. Multiple Regression Analysis Results

Multiple regression analysis was conducted to determine the effect of social support (family, friend, and special person support) on perceived stress and life satisfaction and the effect of perceived stress on life satisfaction. In this analysis, age, gender, education status, and household income were used as control variables. Multiple regression analysis results are shown in

Table 4. Main effects on perceived stress and life satisfaction.

Variables	Model 1: PS		Model 2: LS
	B	SE	B	SE
Constant	2.481 ***	0.189	1.463 ***	0.249
FAS	−0.065 **	0.023	0.144 ***	0.027
FRS	−0.047	0.024	0.095 ***	0.028
SPS	0.067 **	0.020	−0.006	0.024
PS	-	-	−0.241 ***	0.046
Control Variables
Age	−0.028	0.030	0.074 *	0.035
Gender	−0.004	0.056	0.034	0.066
Education status	0.006	0.027	−0.013	0.032
Household income	0.002	0.027	−0.030	0.031
R2	0.035		0.181
F	3.255		17.07
p	<0.01		<0.001

FAS: Family support, FRS: Friend support, SPS: Special person support, PS: Perceived stress, LS: Life satisfaction, * p < 0.05; ** p < 0.01; *** p < 0.001.

.

In Table 4–Model 1, the effect of family, friend, and special person support on perceived stress was evaluated. The results first revealed that family support negatively predicted perceived stress (B = −0.065, p < 0.01). Secondly, friend support negatively predicted perceived stress but was not statistically significant (p > 0.5). Thirdly, special person support positively predicted perceived stress (B = 0.067, p < 0.01). As a result, hypothesis H_1a was accepted; however, hypotheses H_1b and H_1c were rejected.

In Table 4–Model 2, the effect of family support, friend support, special person support, and perceived stress on life satisfaction was evaluated. The results showed that family and friend support positively predicted life satisfaction (B = 0.144; B = 0.095, p < 0.001). Special person support negatively predicted life satisfaction but was not statistically significant (p > 0.05). In addition, perceived stress negatively predicted life satisfaction (B = −0.241, p < 0.001). As a result, hypotheses H₂, H_3a, and H_3b were accepted, and hypothesis H_3c was rejected.

4.5. Mediation Analysis Results

Mediation analysis was conducted using Model 4 of Hayes’ [35] PROCESS macro to test indirect effects shown in the research model. In this analysis, age, gender, education status, and household income were used as control variables.

The mediator role of perceived stress in the relationship between family support and life satisfaction was tested. The results first revealed that family support positively predicted life satisfaction (B = 0.160, 95% CI = (0.10, 0.21)) (

Table 5. Total, direct, and indirect effects on life satisfaction.

					B	SE	LLCI	ULCI
Total effect of FAS on LS					0.160	0.028	0.105	0.214
Direct effect of FAS on LS					0.144	0.027	0.090	0.198
Total effect of FRS on LS					0.107	0.029	0.050	0.163
Direct effect of FRS on LS					0.095	0.028	0.040	0.151
Total effect of SPS on LS					−0.022	0.024	−0.069	0.025
Direct effect of SPS on LS					−0.006	0.024	−0.053	0.041
Indirect Path					B	SE	LLCI	ULCI
Independent		Mediator		Dependent
FAS	>	PS	>	LS	0.015	0.007	0.004	0.030
FRS	>	PS	>	LS	0.011	0.006	0.001	0.024
SPS	>	PS	>	LS	−0.016	0.006	−0.028	−0.006

FAS: Family support, FRS: Friend support, SPS: Special person support, PS: Perceived stress, LS: Life satisfaction, LLCI: Lower level confidence interval, ULCI: Upper level confidence interval.

). Secondly, family support negatively predicted perceived stress (B = −0.065, p < 0.01) (Table 4–Model 1). Thirdly, perceived stress negatively predicted life satisfaction (B = −0.241, p < 0.001) (Table 4–Model 2). Finally, the bias-corrected bootstrapping mediation analysis indicated that family support predicting life satisfaction through perceived stress was significant (B = 0.015, SE = 0.01, 95% CI = (0.01, 0.30)). Consequently, perceived stress partially mediated the relation between family support and life satisfaction. As a result, hypothesis H_4a was accepted.

The mediator role of perceived stress in the relationship between friend support and life satisfaction was tested. The results first revealed that friend support positively predicted life satisfaction (B = 0.107, 95% CI = (0.05, 0.16)) (Table 5). Secondly, friend support negatively predicted perceived stress but was not statistically significant (p > 0.05) (Table 4–Model 1). Thirdly, perceived stress negatively predicted life satisfaction (B = −0.241, p < 0.001) (Table 4–Model 2). Finally, the bias-corrected bootstrapping mediation analysis indicated that friend support predicting life satisfaction through perceived stress was significant (B = 0.011, SE = 0.01, 95% CI = (0.001, 0.024)). Consequently, perceived stress partially mediated the relation between friend support and life satisfaction. As a result, the H_4b hypothesis was accepted.

The mediator role of perceived stress in the relationship between special person support and life satisfaction was tested. The results first revealed that special person support negatively predicted life satisfaction but was not statistically significant (95% CI = (−0.07, 0.02)) (Table 5). Secondly, special person support positively predicted perceived stress (B = 0,067, p < 0.01) (Table 4–Model 1). Thirdly, perceived stress negatively predicted life satisfaction (B = −0.241, p < 0.001) (Table 4–Model 2). Finally, the bias-corrected bootstrapping mediation analysis indicated that special person support predicting life satisfaction through perceived stress was significant (B = −0.016, SE = 0.01, 95% CI = (−0.028, −0.006)). Consequently, perceived stress fully mediated the relation between friend support and life satisfaction. As a result, hypothesis H_4c was accepted.

4.6. Moderation Analysis Results

Moderation analysis was conducted using Model 21 of Hayes’ PROCESS macro to test moderating effects shown in the research model [35]. In this analysis, age, gender, education status, and household income were used as control variables. Moderation analysis results are shown in

Table 6. Moderation analysis results.

Variables	Model 1: PS		Model 2: PS		Model 3: PS		Model 4: LS
	B	SE	B	SE	B	SE	B	SE
Constant	−0.021	0.179	0.099	0.201	0.470 *	0.189	1.688 ***	0.21
Age	−0.041	0.029	−0.039	0.030	−0.037	0.030	0.076*	0.035
Gender	0.018	0.054	0.017	0.054	0.020	0.054	0.030	0.066
ES	0.004	0.026	0.006	0.026	0.004	0.026	−0.014	0.032
HI	0.014	0.026	0.015	0.026	0.016	0.026	−0.034	0.315
FAS	−0.055 *	0.022	−0.059 **	0.022	−0.059 **	0.022	0.143 ***	0.273
FRS	−0.038	0.023	−0.033	0.023	−0.039	0.023	0.095 ***	0.029
SPS	0.050*	0.020	0.050 *	0.020	0.052 **	0.020	−0.006	0.023
CA	0.286 ***	0.041	0.275 ***	0.040	0.266 ***	0.041	-	-
PS	-	-	-	-	-	-	−0.311 ***	0.055
SA	-	-	-	-	-	-	0.006	0.074
PS X SA	-	-	-	-	-	-	0.241 *	0.101
FAS X CA	0.063 *	0.027	-	-	-	-	-	-
FRS X CA	-	-	0.073 *	0.029	-	-	-	-
SPS X CA	-	-	-	-	0.031	0.025	-	-
R2	0.108		0.101		0.103		0.188
F	8.322		8.458		7.848		14.30
p	<0.001		<0.001		<0.001		<0.001

* p < 0.05; ** p < 0.01; *** p < 0.001, ES: Education status; HI: Household income; FAS: Family support; FRS: Friend support; SPS: Special person support; CA: Coronavirus anxiety; SA: Social aid; PS: Perceived stress; LS: Life satisfaction.

.

The interaction of family support and coronavirus anxiety (FAS X CA) had a significant effect on perceived stress, as shown in Table 6–Model 1 (B = 0.063, p < 0.05, 95% CI = (0.001, 0.116)). According to this result, hypothesis H_6a was accepted. In addition, the effect of family support on perceived stress varies according to the low, medium, and high levels of coronavirus anxiety, as seen in Figure 2. Perceived stress decreases as the family support increases at all levels of coronavirus anxiety. However, this decrease among refugees with high coronavirus anxiety is slower than among refugees with a medium or low level of coronavirus anxiety.

When the results in Table 6–Model 2 are examined, it is seen that the interaction of friend support and coronavirus anxiety (FRS X CA) had a significant effect on perceived stress (B = 0.073, p < 0.05, 95% CI = (0.016, 0.123)). As a result, hypothesis H_6b was accepted. As seen in Figure 3, the effect of friend support on perceived stress varies according to the low, medium, and high levels of coronavirus anxiety. While the coronavirus anxiety level is low, friend support reduces perceived stress. Friend support reduces perceived stress more slowly when the coronavirus anxiety level is medium. When the coronavirus anxiety level is high, friend support increases perceived stress. However, the effects are not statistically significant when the coronavirus anxiety is low and medium.

In Table 6–Model 3, it was determined that the interaction effect of special person support and coronavirus anxiety (SPS X CA) on perceived stress was not statistically significant (p > 0.05, 95% CI = (−0.019, 0.080)). Therefore, hypothesis H_6c was rejected.

According to the analysis results presented in Table 6–Model 4, the interaction of perceived stress and social aid (PS X SA) had a significant effect on life satisfaction (B = 0.241, p < 0.05, 95% CI = (0.042, 0.439)). Therefore, hypothesis H₅ was accepted. As seen in Figure 4, the effect of perceived stress on life satisfaction varies according to whether people get social aid or not. Life satisfaction decreases as the perceived stress increases regardless of to what extent refugees get social aid. However, this decrease is smaller among refugees getting social aid than not getting social aid.

4.7. Results of the Research Model

Multiple regression analysis, moderator analysis, and mediator analysis were performed to test the research hypotheses. The direct and indirect effects obtained as a result of these analyses are presented in Figure 5.

5. Discussion

In this study, the relationships among perceived social support, stress, life satisfaction, coronavirus anxiety, and social aids received by Syrian refugees living in Turkey were examined. Within the scope of these relationships, the roles of coronavirus anxiety, social aids, and perceived stress were investigated. Overall, the results showed that perceived stress mediated the relationship between social support sub-dimensions and life satisfaction. On the other hand, it was observed that coronavirus anxiety had a moderator role in the relationship between the sub-dimensions of social support (except special person support) and perceived stress, and social aids had a moderator role on the relationship between perceived stress and life satisfaction.

This study found that family support reduced stress. Family is vital for refugees, who often deal with extraordinary situations. A study carried out with North Korean refugees highlights that family support is a coping mechanism for negative situations [37]. In another study conducted with Congolese refugees living in Australia, most participants reported that family support in coping with stress is essential [38]. A study with Syrian refugees in Sweden found that higher social support was associated with lower post-traumatic stress disorder [39]. The results of the studies in the literature support this finding of this study. However, contrary to expectations in this study, we discovered that special person support increases stress. We also found that the effect of friend support on stress was insignificant. These findings do not match with the results of other studies in the literature [40,41].

This study found that family and friend support increases life satisfaction. A study conducted with refugees in Turkey revealed that when the social support increases, refugees’ life satisfaction increases as well [42]. In another study on social support and life satisfaction with refugee women in South Korea, women who received higher levels of social support reported higher levels of life satisfaction [43]. In addition, this study found that stress reduces life satisfaction. There is a negative relationship between individuals’ stress and their life satisfaction [44]. A study on North Korean refugees discovered that stress negatively affects life satisfaction [45].

Another study conducted with 590 Syrian refugees in Turkey reported that life satisfaction decreases as the level of post-trauma stress increases [46]. The results of the studies in the literature support our findings. In addition, we discovered that stress has a mediator role in the relationship between family, friend, special person support, and life satisfaction. Refugees face many stressors due to the migration process. Therefore, it is essential to consider stress in the relationship between refugees’ social support and life satisfaction. However, while reviewing the literature, we could not find any other study in which stress has a mediator role between family, friend, special person support, and life satisfaction. This study contributes to the literature by addressing the mediator role of stress in the relationship between social support and life satisfaction, which has not been addressed before.

We found that coronavirus anxiety had a moderating effect on the relationship between family support and perceived stress. Table 4 shows that family support reduces the stress of refugees. However, as seen in Figure 2, if refugees have high coronavirus anxiety, family support is almost ineffective in reducing stress. This situation reveals that coronavirus anxiety is very important. In a study conducted with 739 Palestinians, some of whom live in the camp, 65% of the participants stated that their anxiety had increased since the beginning of the pandemic [47]. Another study conducted with 18,171 people in 35 countries discovered that coronavirus anxiety increased people’s stress [48]. These findings support the findings of our study. Therefore, although refugees use social support networks such as family support to cope with stress, the increasing coronavirus anxiety with the pandemic almost eliminates the effect of social support networks in reducing stress. For this reason, various precautions should be taken to minimize or even eliminate the impact of coronavirus anxiety. One of them might be the provision of mental health services. However, cultural and language barriers and increasing stigma prevent refugees from receiving these services [49]. Therefore, to reduce or eliminate the coronavirus anxiety refugees, states should take measures to remove these barriers. Many considerations, such as virtual care and service delivery with interpreters, can be among these measures. We also found that coronavirus anxiety had a moderating effect on the relationship between friend support and perceived stress. However, while the level of coronavirus anxiety is high and moderate, the moderator effect is not significant. One unanticipated finding was that the moderating effect of coronavirus anxiety on the relationship between special person support and perceived stress was not statistically significant. Many studies in the literature examine the relationship between social support and perceived stress [50,51]. However, this study, carried out during the COVID-19 period, contributes to the literature by examining the role of coronavirus anxiety in the relationship between social support and perceived stress.

Finally, we also found that social aids had a moderating effect on the relationship between perceived stress and life satisfaction. As seen in Table 4, stress reduces the life satisfaction of refugees. However, in Figure 4, the decrease in life satisfaction of refugees who get social aids is much smaller than those who do not get social aids. Therefore, this situation shows how important social aids are in reducing the negative effect of stress (reducing the life satisfaction of refugees). Couch [52], in his study, found that young refugees in Australia are often unemployed during the pandemic. In addition, young refugees stated that they were responsible for providing economic support to their families, but they could not do it when they lost their jobs. For this reason, young refugees also need economic support. Another study in Australia shows that refugees need social aids during the pandemic but cannot reach social aids due to misinformation and various barriers. It also stated that refugees could not be protected from the negative effects of COVID-19 if they did not get social aid [53]. Therefore, the findings of these studies support our study. This finding of our study constitutes another important contribution to the literature by revealing that getting social aid during the pandemic period is very important for refugees.

6. Strengths and Limitations

One strength of this study is the sample size. Although refugees constitute a hard-to-reach population, the study is based on a large sample of Syrian refugees selected from a comprehensive sample frame. However, this study has some limitations. First, the study was carried out online due to the COVID-19 pandemic. For this reason, only people who have the technological device to answer the online questionnaire were reached. This reveals the limitations of the study in terms of inclusiveness. Secondly, the coronavirus anxiety scale items used in the study “e.g., When I read or listened to the news about the coronavirus, I felt dizzy or as if I was going to faint” are intended to measure anxiety caused by the physical symptoms of the coronavirus. However, coronavirus anxiety may occur from other variables besides physical symptoms. Finally, this study was carried out with participants from various provinces of Turkey. Participants were reached with the environmental resources of the researchers. Therefore, the generalizability of the results to Turkey is limited. It is essential to evaluate the results of the study within the scope of the limitations specified above.

7. Conclusions

This study is one of the few studies to focus on coronavirus anxiety and social aids’ effects on mediator role of stress in the relationship between social support and life satisfaction. Stress is an important component to refugees’ life satisfaction assessment and improvement. In addition, the study highlighted coronavirus anxiety and social aids’ effects and revealed that these factors have a great importance in refugees’ life satisfaction. Further examination of these factors in future studies may shed light on the preparation of more comprehensive interventions and policies for refugees. In addition, although the majority of refugees in our study are relatively low income, most do not get social aids. Therefore, the reasons for not getting social aid should be examined in detail in future studies.