Friends or Strangers? Modeling Types of In-game Relationship, Social Capital and Psychological Well-being
Shuang Feng
Uppsala University, Shuang.Feng.4464@student.uu.se
Online games are becoming more and more popular nowadays. Interacting with others in games has also become a channel for establishing or developing social relationships. In this article, the author conducted an online survey (N=165) to study the relationship between types of in-game relationships, social capital, and psychological well-being. In-game relationships mainly include two types: playing games with friends and playing games with strangers. The author used the framework of social capital, which includes bonding and bridging. Regarding psychological well-being, the author selected two indicators related to social aspects, namely loneliness and relatedness. The author constructs a structural equation model. The results show that playing with friends will enable bonding and bridging while playing with strangers will enable bridging. Second, two different social capitals can both increase players’ feelings of relatedness and reduce players’ feelings of loneliness. This shows that social relationships in online games have a certain impact on people's psychological well-being. This research also provides some information for game design and understanding of social relationships in games.
Additional Keywords and Phrases:Online game, In-game relationships, Social capital, Psychological well-being ACM Reference Format:
This work was submitted in partial fulfillment for a master’s degree in Human-Computer Interaction at Uppsala University, Sweden, on 14.06.2021. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author must be honored.
© 2021 Copyright is held by the owner/author(s).
1 INTRODUCTION
Nowadays, video games provide people with an interesting environment where they can play with many others without being restricted by region, language, gender, etc. The social aspect in some video games has also become an attractive part of the game. For example, in the massively multiplayer online game, people often have social interaction with friends or strangers in accomplishing common goals. These social interactions may be helpful for the building and maintenance of social relationships between the players in the game. Besides, the social relationships in the game may also give the players social support, which may have some impact on players' psychological well-being. The author used a quantitative survey method to study the difference between the social ties formed within strangers and the ones formed between friends in online games. Besides, the author also studied the impact of these social ties on psychological well-being.
In online games, people can choose the mode of PUG (pick up groups), which means the players can play with the people selected randomly. In this kind of mode, most of them are strangers. People can also play online games with friends. They can invite friends into the same group or room to play together. These two different modes may cause the social ties established between the players in the game to be different.
Meanwhile, these two modes are both important experiences for players so the research may contribute to future game theory or game design.
However, are all social relationships and social interaction in the game beneficial to people's psychological well-being? The author analyzes two aspects of psychological well-being which are loneliness and relatedness to delimit the work. Relatedness refers to the need to feel connected to or cared for by others [9].The reason I chose these two indicators is that they are social aspects of well-being[6]. Since the author wants to study the social aspects of the game and its influence on players, it is reasonable to choose these two indicators. It is meaningful to study the influence of social relationships formed between the players on their psychological well-being since online games become popular gradually.
Research questions:
1.In the game, how do the different relationship types (playing with friends vs playing with strangers) relate to social ties?
2.What is the relationship between the social ties formed in the game and the feeling of loneliness as well as the needs of relatedness of the players?
The author uses the quantitative method to study these two research questions. Participants in the online survey are people who often play games with others. The study asked participants to choose a game they often played with others and asked them to focus on this game when answering questions. This also characterizes the characteristics of the sample. The survey collects data about two different in-game relationships, social ties, relatedness and loneliness. Among them, the measurement of social ties and psychological well-being uses mature scales to ensure the reliability of the survey [5,7,9]. In the result part, the structural equation model is used to analyze the data. The introduction to the structural equation model is in the following section.
2 BACKGROUND
2.1 Relationship types in online game
Both personal factors and contextual factors affect the impact of online gaming on health [25]. The key factors that potentially affect well-being include not only the type of game and the time spent on the game, but also the relationships and social experience in the game. For example, playing games alone and playing games with others can affect well-being differently[25]. Some researchers have studied the importance of social experience in games. For example, when the player feels a human control the opponent, the gaming experience differs from the one when he or she feels that a computer controls the opponent [1,15,20].
Eklund has studied different game contexts, which are mainly divided into social contexts among friends, family, and strangers [10]. She used the separate regression model in the article. She found that the situation of gaming with family remains stable and is not affected by age, time spent on the game, and dedication to the game of the player [10]. Players spend more time playing with strangers. More people treat gaming with friends as social activities (instead of with family members or strangers) [10]. These findings may mean that playing
online games with different people may give them unique experiences and establish different social relationships.
2.2 Online game and social capital
Whether online or offline, social interaction and support are good for people's mental health and well-being [16].
Through social interaction, people feel they are not isolated from society. Mental health and well-being are essential parts for people to have a good life experience. Whether through online activities such as online games together or offline activities such as talking with friends face to face, social interaction plays a certain role in maintaining mental health and well-being. There are also some studies on the impact of the use of the Internet on social isolation among the elderly [3]. Their theme is to study the impact of loneliness on health.
The study found that participants who used to go online reported reduced loneliness and improved quality of communication with others [3]. This also shows that online social interaction has certain benefits for mental health and psychological well-being.
The study of social relationships in the online environment usually uses the framework of social capital, namely bridging and bonding [6]. Social capital exists in the social network formed among people and society as social ties. As a resource, it is meaningful to individuals and organizations [22]. The role of social capital is to provide people with channels for information exchange and support in daily life. Putnam used this concept to study the impact of isolating activity on people's lives in American society [22]. Later, this concept is often used to study the social relationships of people in the online or offline context.
Bridging ties may lack depth but have a large breadth [22]. This means that this type of relationship is diverse. Bonding ties refer to strong relationships, through which people will feel the emotion and social support [22]. These two social capitals play different roles in social life and both of them are important social ties.
Some studies use the framework of social capital to investigate game communities such as World of Warcraft [2,8] and second life [14]. These studies have successfully shown that relationships in games can enable social capital [2,14]. The consensus seems to be that relationships in games may support bridging between players and are unlikely to enable bonding [2,8,14]. However, some research shows bonding and bridging capital can both be enabled in the game. In one research, the authors studied the relationships between the properties of the game and social capital. They found that the interdependence of the game could enable both the bridging and bonding capital [6]. Interdependence is reflected as collaborative play in the game [6]. This shows that the bonding capital can also be enabled by using some kind of game mechanics. When the players interact and collaborate in the game, the bonding capital is established. In another research, the authors studied how the passion of the players affects the social capital in the game. They found that harmonious passion can positively predict both bonding and bridging [18]. Harmonious passion means that the players think that game is in harmony with other activities in their life [18]. If the players don’t spend too much time and are addicted to the game, the bonding capital can also be enabled through the gaming experience [18].
2.3 The relationship between social ties formed in the game and players' psychological well-being In the previous part, I discussed the social capital that may be formed in the game, but what is the meaning of this social capital for players? In the physical world, we usually associate social capital with a positive state of
mental health [7,22]. People can get social support through social interaction with others. If we change the context from offline to online, what impact will online social capital have on psychological well-being?
The debate over whether playing games is a potentially problematic behavior is still ongoing. There are some reports about the relationship between gaming and violence [27], the impact of gaming addiction on mental health [11,24], and whether gaming is a social isolation behavior [11,19]. These studies show playing games can be potentially problematic behavior. Some other studies have shown that players who spend time and energy to establish relationships in the game seem to spend less time managing their offline social relationships [4,11,19]. However, in other research, the conclusion is different. In the research of Trepte et al., they found that social capital in games is positively correlated with offline social support [23]. The research results show that the social capital formed in the game has a good impact on the players’ offline experience, showing the impact on psychological well-being [23]. Social capital bridging has also been reported to be positively correlated with the well-being of social players [28]. It means that the social interaction between players can form social ties, which is also beneficial for their psychological well-being. The results and conclusions of all these researches are different because they study in different fields and directions. The conclusion can be that playing games is harmful because of addiction, violence, and social isolation. The result can also be that playing games is beneficial because of the social support required from the gaming experience.
The author wants to study the relationship between the social capital formed in the game and the psychological well-being of players. The inspiration of this paper comes from the research of Trepte et al. [23], which is that social capital in the game positively affects players’ psychological well-being. The author intends to select two aspects of psychological well-being, which are loneliness and relatedness, to test the impact of social ties formed in the game. The reason I selected these two indicators was that they are social aspects of well-being [6]. Since I want to study the social relationship, social capital in the game, and its effect on psychological well-being, the choice of these two social indicators makes sense. To be more specific, I want to investigate whether and to what extent the social capital formed in the game reduces the player’s loneliness and makes them feel related to others.
3 METHODOLOGY AND METHODS 3.1 Methodology
The author wants to study the relationship between types of in-game relationships, social capital, and psychological well-being. Positivism paradigm fits best for the study. The positivism is mainly used to explore whether there is a correlation or a causal connection between the phenomena studied. If there is a certain connection between the phenomena, hypotheses can be established and we can construct a model to explain the relationship between variables [26]. Positivism allows me to use deductive approaches and quantitative methods to do the research.
3.2 Structural equation modeling
Structural equation modeling (SEM) is a comprehensive and flexible statistical approach to test the hypotheses and the relations between variables [13]. It is usually used to analyze data in social and behavioral science.
The SEM is becoming a standard approach for testing hypotheses gradually [13]. The relations among the observed variables and latent variables can also be tested by using the SEM [13].
If the researcher wants to study the relations between multiple variables, it is easier to use the SEM approach. Because by using traditional statistical approaches such as multiple regression, the researcher needs to test different paths in the model over one time. While using SEM, we can test all the paths in the model at one time. Besides, many user-friendly computer software such as LISREL and AMOS can analyze the data and test hypotheses by using the SEM approach.
How to build the structural equation model then? The standard process is the model specification, estimation, evaluation of fit, model modification, and interpretation [13]. By using software such as AMOS, the researcher can test and modify the model through the output of the software. There is also a standard about when the model can be accepted [17]. When the model can be accepted, the researcher can use the output to interpret the model and test the hypotheses.
Some researchers study relationships between online games and players by using the SEM approach. In one research, they study how the properties of online games affect the social capital of players by using a structural equation model [6]. In another paper, they also use the SEM approach to find how different passions affect social capital [18]. These all show that the structural equation model can be a suitable approach to test the model and hypotheses in this article.
As stated in the research questions, the author wants to study how the relationship types in the game (namely playing with friends vs playing with strangers) affect the social capital (bonding and bridging) and how the social capital affects the players’ psychological well-being (loneliness and relatedness). The following hypothesized path model (Figure 1) is formed. The paths show the relations between variables. The author collected data about the variables and used the SEM to test the relations among these variables.
Figure 1. Hypothesized path model.
3.3 Survey
The author conducted a survey to collect data about participants' types of social relationships, gaming experience, and psychological well-being. Social relationships include playing with friends and playing with strangers. The gaming experience is mainly about the social capital enabled in the game, which is measured with mature scales. The author also used two other scales to measure two aspects of psychological well-being (loneliness and relatedness). I will mention the usage and detailed design of the questionnaire later.
Play with friends
Play with strangers
Bridging Bonding
Relatedness Loneliness
3.4 Hypotheses
According to the literature review and hypothesized path model, the author proposed the following hypotheses which will be tested with the SEM approach.
H1: The social capital enabled by playing games with friends is different from the social capital enabled by playing games with strangers.
H2: Social capitals enabled in the game are negatively related to loneliness.
H3: Social capitals enabled in the game are positively related to the feeling of relatedness.
3.5 Participants and process
The data comes from the survey. The target group is the people who often play games with others. The way to recruit participants is to find game players from related forums, people the author knows, previous and current class lists to complete the survey. After collecting the data, statistical analysis software SPSS and AMOS are used to analyze the data. SPSS is used to calculate the mean, standard deviation, and correlation coefficients for variables. I use AMOS to construct the structural equation model.
3.6 Pre-screen
Before starting the main study, I screened the target group. The title mentioned by the author when the questionnaires were released was to find people who play games. The purpose of pre-screen is to screen out the target group (that is, people who often play games with others) for subsequent research. The pre-screen problems include 1. choose how often you play games (the options are daily, several times a week, once a week, less than once a week); 2. to what extent do you consider yourself as a gamer [21] (choose from 1 to 10);
3.The percentage of time you spent playing games with others (1 means playing alone, 10 means playing with others). The inspiration for setting these criteria comes from one study whose target group is also the people who often play games with others [6].
After the pre-screen, I finally received 282 questionnaires. Since the goal of the article is to study the impact of playing games with others on social capital and psychological well-being, the author believes that people who often play games with others should be selected. So the following standards were established: 1. People who play games at least a few times a week 2. Think of themselves as gamers to some extent (3/10 or higher) The reason I selected this standard was that the author wanted to select people who played games and admired their gamer identity. It’s helpful to give neutral answers. 3. Spend some time playing games with others (3/10 or higher). I invited 192 participants who met these criteria to take part in the main study.
3.7 Main study
After signing the consent form, the participant filled out the questionnaire. Among these invited people, 171 completed the questionnaire. The questionnaire comprises the basic information including gender, age, and the game which the participant usually plays with others, and the items of scales which will be mentioned later.
The scales are to collect the data about the in-game relationship, social capital, loneliness, and relatedness.
The data of 6 users were removed because of unqualified behaviors. One example is that the participant chose contradictory options in the designed reverse scoring question (for instance, in the scale for measuring relatedness, questions 3, 6, and 7 are contrary to the description of other questions, and some participants all choose the same score 5).
The average age of the remaining 165 people is 26.9 years old (SD=5.6, min=19, max=45). 114 (69%) are male and 51 (31%) are female. Most people (n=104) play games every day, and the remaining people play games several times a week. The participants consider themselves gamers (mean=7.67, SD=1.64). They play games with others more often than alone (mean=7.49, SD=1.97).
3.8 Measures
Several scales are used to collect the data about the relationship in the game, social capital, and psychological well-being.
3.8.1 Games involved.
At the beginning of the questionnaire, participants need to fill out a game they often play with others. The reason this question is raised is that the author wants to help participants answer the following scales more easily because they just need to think about the experience in just one game. I told the participants that they need to focus on the game they filled out when answering the following questions.
3.8.2 The type of relationship.
This measure shows whether the gamer is playing with strangers or with friends. In the survey, participants need to answer two questions:
1. To what extent do you play games with friends?
2. To what extent do you play games with people you don’t know?
7 point scale is used. 1 means not at all, 7 means all the time. The design of these two questions refers to one research where they also collect the data about the different in-game relationships by using a 7 point scale [18].
3.8.3 Social capital.
It shows the type of social ties established between the players. William’s Internet Social Capital Scales were used to measure bridging and bonding [7]. I changed some words to adapt to the context of the game. The reason I used this scale was that it is designed for both online and offline contexts. Since the author wants to study the online game, this scale is suitable. Besides, the question items are tested to be valid [7]. The researchers also used this scale to measure social capital and test their models [6,18].
3.8.4 Loneliness.
To test the loneliness of players, Russell et al.’s UCLA Loneliness scale was used [5]. I chose this scale because it has high internal consistency and validity criteria [5]. They improve the scale to be valid and trustworthy. By using this scale, the author can collect valid data about the feeling of loneliness of the participants.
3.8.5 Relatedness.
To test the players’ needs for relatedness, part of the Basic Psychological Need Satisfaction questionnaire was used [9]. The reason I selected this scale was that it is promising and has valid evidence [9]. By using this scale, I can collect the data about relatedness validly, which can also make the results of the article convincing.
3.9 Process
After the participants signed the consent form, they filled out the scales. In answering questions, the questionnaire reminds participants that when they answer the questions, the people or relationships they think of are the group of people they play with or the relationship they have established in the game. The first step for the participants to do is to fill out the basic problems such as age, gender, and so on. The following questions that participants need to answer belong to, in order, the type of relationship, social capital scale, loneliness, and relatedness scale. I group the problems into several parts to make the answering process less energy-consuming.
3.10Data analysis
The data is mainly composed of quantitative data, which are the scales used in the questionnaire. There is also a small amount of qualitative data, which is the name of a game that users often play with others collected at the beginning of the questionnaire. The author used SPSS to store the quantitative data and make the basic data analysis such as descriptive statistical analysis containing mean, the standard deviation of the variables, and so on.
3.10.1 Quantitative data.
The author used SPSS to analyze mean, standard deviation, and correlation coefficients for variables. The AMOS statistical package was used to build the structural equation model. The construction process used the maximum likelihood method. The author used the output of SEM to test the hypotheses and the relations among the variables.
3.10.2 Qualitative data.
The author collected the games that participants often play with others and made a simple descriptive statistical analysis. Here the descriptive statistical analysis means that the author analyzed how many times the game mentioned by the participants appears in the total sample. I will sum the popular games up in the next section.
3.11Characterize the sample
In addition to basic age and gender, the questionnaire also collects the games that participants play most often with others. These data are used to describe the characteristics of the sample.
3.11.1 Games involved.
Previous research on social ties in games mainly focused on a specific game, such as World of Warcraft [2,8,23]. By choosing just one game, the researchers can contribute to the study of this specific game but may lack generalization. The author wants to study the general impact of different relationships on the formation of social capital and the impact of social capital on psychological well-being, so the sample source is not limited to a specific game. At the beginning of the questionnaire, participants were asked to fill in a game that they often played with others. In the end, 13 games were collected. The top five popular games are Arena of Valor: 5v5 Arena Game (66), League of Legends (27), Game for Peace (21), World of Warcraft (9), Overwatch (6).
3.12Ethical considerations
Participants signed the consent form before filling out the questionnaire. The consent form includes the purpose of the research and how the data will be processed. All participants reported their acceptance of the participation. All data files only exist on my computer. The participant’s questionnaire filling process is anonymous. These measures are to ensure that the personal information of participants is not disclosed.
4 RESULTS
This section introduces the results of using SPSS for descriptive statistical analysis and correlation analysis of the data. The hypothesized path model(Figure 1) is tested with AMOS statistical package and the result of SEM are also presented.
4.1 The result of correlation analysis
Table 1: Mean, standard deviation, and correlation coefficients for variables (**=p<0.01).
A=play with friends, B=play with strangers
Mean SD A B Bonding Bridging Loneliness Relatedness
A 5.78 1.20 -
B 4.49 1.35 .321** -
Bonding 3.82 0.61 .439** 0.243** -
Bridging 4.20 0.57 .697** .404** .580** -
Loneliness 1.67 0.46 -.551** -0.21** -.685** -.661** -
Relatedness 5.58 0.96 .451** .313** .688** .672** -.835** -
The mean, standard deviation, and correlation coefficients of the variables included in the hypothesized path model are listed in Table 1.
4.2 Structural equation model
As can be seen from Table 1, A (play with friends) and B (play with strangers) are significantly correlated (r=0.32). Therefore, in the structural equation model, the two variables A and B may co-vary. In addition, bonding and bridging are also significantly correlated (r=0.58), so the error terms of these two variables are also allowed to co-vary in the structural equation model. In the same way, loneliness and relatedness are also significantly correlated (r=-0.84), so the error terms of these two variables are also allowed to co-vary in the structural equation model. I presented the final structural equation model in Figure 2.
The model fit indexes got from AMOS are shown in Table 2. The model fit reflects how well the model fits the data collected. According to the recommendation of the structural equation model’s model fit threshold, it can be seen that the model fit is acceptable [17]. It means that the author can use the structural equation model to test the hypotheses and find the relations among the variables.
Table 2: Model fit summary
Index Threshold
CMIN/DF 2.87 <3,good
P 0.001 <0.05
CFI 0.97 >0.95,great;>0.90,traditional
TLI 0.91 >0.95.great;>0.90,traditional
RMSEA 0.09 <0.05,great;<0.10,acceptable
The structural equation model is presented in Figure 2.
Figure 2: standardized coefficients for the hypothesized path model A=play with friends, B=play with strangers
Gray lines indicate the insignificant path
5 ANALYSIS
5.1 Correlation analysis
From the data in Table 1, we can see that playing with friends has positive correlations with playing with strangers. Playing with friends is also positively correlated with two social capitals. Playing with friends has a positive correlation with relatedness as well. However, it negatively correlated with loneliness. Playing with strangers is positively correlated with two social capitals. It is also positively correlated with relatedness. But it is negatively correlated with loneliness. However, compared with correlations between playing with friends and other variables, the correlations between playing with strangers and other variables are weaker. The biggest difference is that the correlation between playing with strangers and bonding is weak. Bonding is strongly positively correlated with bridging and relatedness, while it is strongly negatively correlated with loneliness.
Bridging is also positively correlated with relatedness and negatively correlated with loneliness. Relatedness is negatively correlated with loneliness.
5.2 Analysis of structural equation model
According to the structural equation model in Figure 2, the author tries to answer the research questions.
1. In the game, how do the different relationship types (playing with friends vs playing with strangers) relate to social ties?
We can see from the structural equation model that playing with friends can significantly predict bonding capital (β=0.40, p<0.01) and also significantly predict bridging capital (β=0.63, p<0.01). It means that playing games with friends can enable bonding and bridging. The interaction in the game helps to develop the social capital between the players. From the experience of gaming with friends, people can get social support as well as have the willingness to build more social relationships. Playing with strangers can not predict bonding capital (β=0.11, not significant), but it can predict bridging capital (β=0.20, p<0.01). It means that playing games with strangers can also enable bridging but the effect of enabling bonding is not significant. From the experience of gaming with strangers, people can know more people and develop diverse social relationships.
The path model can explain 20% of the variance of bonding capital (R2=0.20) and 52% of the variance of bridging capital (R2=0.52). It means that the paths between in-game relationships and social capital in the structural equation model are reasonable and can explain some data.
2. What is the relationship between the social ties formed in the game and the feeling of loneliness as well as the needs of relatedness of the players?
From the structural equation model, we can see that bonding capital can negatively predict the feeling of loneliness (β=-0.45, p<0.01) and it can positively predict the needs of relatedness (β=0.45, p<0.01). It means that the bonding capital formed in the game can be helpful to reduce the feeling of loneliness and increase the relatedness of players. Bridging capital can negatively predict the feeling of loneliness (β=-0.40, p<0.01). It can also predict the needs of relatedness (β=0.41, p<0.01). It means that the bridging capital formed in the game can also be beneficial to reduce loneliness and increase the feeling of relatedness of players. The path model can explain 57% of the variance of loneliness(R2=0.57) and 59% of the variance of relatedness(R2=0.59). It means that the paths between social capital and the indicators of psychological well-being in the model can be acceptable. These paths can explain some data.
6 DISCUSSION
The research shows that play with others (no matter friends or strangers) is related to the establishment of social capital. Social capitals are also different according to relationship, which is consistent with Putnam's social capital theory [22]: that is, in a given online/offline network, social capital reflects the different closeness of social relationships. In addition, the social capital enabled in the game has a positive correlation with psychological well-being, which also shows to a certain extent that playing video games with others is beneficial to players’ psychological well-being.
6.1 Summary of the result
The author used the structural equation model to test the hypotheses and summarized the result.
Hypothesis 1: The social capital enabled by playing games with friends is different from the social capital enabled by playing games with strangers.
The structural equation model shows that the relationships between playing games with friends and the two social capitals (bridging and bonding) are indeed different from the relationships between playing games with strangers and the two social capitals. Playing with friends can predict bridging and bonding capital. This may be because players have established certain social ties outside the game. The interaction in the game will reflect or strengthen these social ties. Playing with strangers can predict bridging capital, but cannot predict bonding. As mentioned before, bonding is a kind of close and strong social tie. People can get strong social support from bonding. Interacting with strangers in the game may not be enough to enable strong bonding capital. But by playing games with strangers, the player can enable bridging capital which helps them to build diverse social relationships. So the hypothesis can be accepted. The social capital enabled by playing games with friends differs from the social capital enabled by playing games with strangers.
Hypothesis 2: Social capitals enabled in the game are negatively related to loneliness.
We can see from the structural equation model that both bonding and bridging negatively predict loneliness.
This also shows to a certain extent that the social capital formed in the game is good for reducing loneliness.
So the hypothesis can be accepted. Social capitals enabled in the game are negatively related to loneliness.
Hypothesis 3: Social capitals enabled in the game are positively related to the feeling of relatedness.
From the structural equation model, we can see that both bonding and bridging positively predict the feeling of relatedness. This also explains to a certain extent that the social capital formed in the game can be helpful to make people feel related. So the hypothesis can be accepted. Social capitals enabled in the game are positively related to the feeling of relatedness.
In conclusion, depending on whether people play online games with friends or strangers, different social capitals can be enabled. A strong social tie, bonding, is often established between friends. Playing games with strangers is more likely to enable bridging capital. Social capital has certain benefits for the psychological well-being of players. Although it cannot be concluded from the research that the social capital enabled in the game directly leads to good psychological well-being, these variables are indeed significantly positively related.
6.2 Implication for design 6.2.1 Design for friendship.
The structural equation model shows that playing with friends can predict bonding and bridging. Playing online games with friends may become more and more popular as a new way of socializing. In some special periods, such as pandemic and travel ban periods, online gaming may become new and interesting social activities between friends. Even if friends cannot meet frequently for various reasons, their social interaction in the game may also enhance social capital. Social capital can provide people with necessary social and emotional support. Most game designs nowadays do not restrict whether the players are strangers or friends. Maybe games designed specifically for friends can also be full of fun. Normal social activities between friends, such as offline activities or online regular chats, play an important role in people’s social life. Playing online games with friends can also be a new and attractive possibility for developing friendships. Games designed specifically for friends may not only have the characteristics of normal games but also can have some additional features. For example, the game designer can design the game in a way that makes it easier for players to record their chat, collaboration, and the time spent with friends in the game. The interaction between friends in the game has already helped to enable social capital. Besides, the recorded chat, video, and so on can be the topics of the offline chat outside the game. In this way, friendship is maintained and developed through online games. If players want to get strong social support through the game, they can choose to play online games with their friends or just choose a game that is specifically designed for friends.
6.2.2 Play with strangers.
We can see from the structural equation model that playing with strangers can predict bridging but cannot predict bonding. So if the players pursue to play with more strangers and develop diverse social relationships, they can choose a game with a pick up group mechanism. Playing with strangers can provide the social capital of bridging, so players’ social life has a greater breadth. They can get to know more people and the wider world by playing online games. Pick up group mechanism is also a wonderful choice for players if they want to enable bridging capital from the online game. The players can play with more people and the social interaction between strangers in the game is also good for their social life.
6.3 Implication for theory 6.3.1 Generalization.
The author allows participants to choose the games they often play with others instead of restricting them to a specific game, which expands the scope of the research and is more comprehensive. The games involved are League of Legends, first-person shooting games, online chess, and so on, which contain many types. In the previous studies, the researchers focused on one game such as second life, World of Warcraft [2,8,14]. These studies can undoubtedly contribute to the understanding or design of the specific game. While we can only use the result in just one game and may lack some generalization. In this paper, the author intends to find the general effect of social interaction in the game on players and try to give an overall understanding of these in-game social interactions, social capital, and players’ psychological well-being. We can apply the results in many types of games.
The age range of the participants is also relatively wide, ranging from 19 to 45 years old. The study did not only select young people and ignore other age groups who play games. For those who take games as a leisure and entertainment activity, all of them can be the target group. The purpose of this way of choosing a target group is that I want to let the opinions of different people be heard. The author wants to exclude bias as much as possible since people of different ages can have different thoughts and attitudes towards the experience of online gaming.
The results of the study show the overall relationship between different in-game relationships, social capital, and psychological well-being. Since the game and ages of participants are diverse, we can apply the results to many contexts. The results can let the reader have an overall understanding about how in-game relationships, social capital, and psychological well-being are related. It can contribute to future studies and give other researchers some inspiration.
6.3.2 In-game social capital and psychological well-being.
With the popularity of online games, people spend more time on them gradually. The impact of social interaction in games on people's psychological well-being needs more attention. As mentioned in previous studies, social interaction in online games has a positive impact on the social capital construction of players [23]. This conclusion is also consistent with the results of my study, that is, playing games with others online can enable social capital. Social capital can provide people with necessary emotional and social support.
In addition, the two social capitals, whether bonding or bridging, can be helpful to reduce the player's loneliness and improve the player's feelings of relatedness. A caring relationship is very important to our well-being because it provides us with social support and satisfies our emotional needs [9]. Here the caring relationship means that people feel others care for them. The feeling of relatedness is part of a caring relationship, while the feeling of loneliness is harmful to the establishment of a caring relationship. The social capital enabled in the game can positively predict relatedness while negatively predicting loneliness. It shows that the social capital enabled in the game can be beneficial for the player's psychological well-being.
6.4 Limitations and future work
The study tested the relationship between the variables, which are in-game relationship, social capital, and psychological well-being, and made some explanations by using the structural equation model. The result was discussed and the author also gave some ideas about the implication of the result for design and theory. But the research also has some limitations. Some future work can be done for improvement.
First, there may be some other factors that influence the effect of in-game relationships on social capital. For example, in one research, they found that playing games with friends or strangers affects social capital through varying levels of passion as an intermediary [18]. In the model of this study, the direct influence of different in-game relationship types on social capital is considered. While other factors that may exist which affect the establishment of social capital are not considered. The future work can be that I consider some potential factors to build the model. In this way, I can explain how different in-game relationships affect the establishment of social capital. In this article I discussed the phenomenon but couldn’t give a solid explanation about why it happened. It would be an interesting research direction for future work.
Second, the model construction considers the overall social capital and psychological well-being. In future research, maybe I can study which specific aspects of these variables are useful to make better designs. The
variables are calculated with scales and the scales are composed of many items. The data collected in every item can be different. It is possible to select part items of scales to conduct research to draw more targeted conclusions. Of course, the complete scale should be used, but when the data collection is finished, it is possible to find which items strongly contributed to the variables. It is also interesting to select other indicators of psychological well-being in addition to loneliness and relatedness to study the impact of online games on the psychological well-being of people.
Finally, some qualitative data can be additional material for the research. The author may get a lot of information from qualitative data. With the qualitative data, the author can explain the results got by using the structural equation model from other angles. The author can ask the participants why their experience of playing online games with friends differs from the experience of playing with strangers. Qualitative data contains opinions from participants and may contribute to design or theory. Through interviews with participants, the researcher can understand the participants’ motivations for playing games with strangers or friends. Meanwhile, how the in-game social interaction affects their psychological well-being can be stated in words instead of scales. What are their expectations for the function of the game? These are all valuable data for theory and design.
7 CONCLUSION
As online games gradually become a new place for social activities, research about the impact of online games on people's social life and psychological well-being becomes meaningful. The findings of this article make contributions to some research directions. The first is about the relationship between different in-game relationship types and social capital in the online game. Playing with friends and playing with pick up groups are both very common ways of playing together in the game. In this article, the author states that different in-game social relationships, namely playing with friends and playing with strangers, have different effects on the establishment of social capital. Playing with friends can predict bridging and bonding, which is not surprising. Playing with strangers, however, can also positively predict bridging, which shows that even the interaction in pick up groups can also be beneficial for players’ social life. This means that online gaming with others, whether friends or strangers, is good for the establishment of social capital and will provide players with social support. Second, no matter what kind of social capital in the game can both positively predict relatedness while negatively predict the feeling of loneliness. This shows that social capital enabled through online games is good for people's psychological well-being.
The article also provides some insights into the design of social games and implications for theory. I also discussed the limitations of the work and the aspects that can be improved, which can be the future research direction. The social relationship formed in the game may differ from the offline relationship we have known before, but it indeed provides players with social support and has a positive impact on players’ psychological well-being. This may change people’s stereotypes about online games in the past. The in-game social relationship can also be beneficial and meaningful. In the future, social games using these positive effects may play a role that cannot be ignored in our social life.
REFERENCES
[1] Cairns, P., Cox, A. L., Day, M., Martin, H., & Perryman, T. 2013. Who but not where: The effect of social play on immersion in digital games. International Journal of Human-Computer Studies, 71, 1069e1077.
Journal of Computer-Mediated Communication 11, 4: 885–909.
[3] Cotten, S. R., Anderson, W. A., & McCullough, B. M. 2013. Impact of internet use on loneliness and contact with others among older adults: Cross-sectional analysis. Journal of Medical Internet Research, 15, e39.
[4] Cuihua Shen and Dmitri Williams. 2011. Unpacking time online: Connecting internet and massively multiplayer online game use with psychosocial wellbeing. Communication Research 38, 1: 123–149. https://doi.org/10.1177/0093650210377196
[5] Dan Russell, Letitia A. Peplau, and Carolyn E. Cutrona. 1980. The revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology 39, 3: 472–480. https://doi.org/10.1037/0022-3514.39.3.472
[6] Depping, A. E., Johanson, C., & Mandryk, R. L. 2018, October. Designing for friendship: Modeling properties of play, in-game social capital, and psychological well-being. In Proceedings of the 2018 Annual Symposium on Computer-Human Interaction in Play (pp.
87-100).
[7] Dmitri Williams. 2006. On and Off the ’Net: Scales for Social Capital in an Online Era. Journal of ComputerMediated Communication 11, 2: 593–628. https://doi.org/10.1111/j.1083-6101.2006.00029.x
[8] Dmitri Williams, Nicolas Ducheneaut, Li Xiong, Yuanyuan Zhang, Nick Yee, and Eric Nickell. 2006. From tree house to barracks: The social life of guilds in World of Warcraft. Games and culture 1, 4: 338–361.
[9] Edward L Deci and Richard M Ryan. 2000. The" what" and" why" of goal pursuits: Human needs and the self determination of behavior.
Psychological inquiry 11, 4: 227–268.
[10] Eklund, L. 2015. Playing video games together with others: Differences in gaming with family, friends and strangers. Journal of Gaming
& Virtual Worlds, 7, 259e277
[11] Emese Domahidi, Ruth Festl, and Thorsten Quandt. 2014. To dwell among gamers: Investigating the relationship between social online game use and gaming-related friendships. Computers in Human Behavior 35: 107–115.
[12] Helena Cole and Mark D Griffiths. 2007. Social interactions in massively multiplayer online roleplaying gamers. CyberPsychology &
Behavior 10, 4: 575–583.
[13] Hoyle, R. H. 1995. Structural equation modeling: Concepts, issues, and applications. Sage.
[14] Isto Huvila, Kim Holmberg, Stefan Ek, and Gunilla Widén-Wulff. 2010. Social capital in second life. Online Information Review 34, 2: 295 –316.
[15] Johnson, D., Wyeth, P., Clark, M., & Watling, C.2015. Cooperative game play with avatars and agents: Differences in brain Activity and the experience of play. In Proceedings of the 33rd annual ACM conference on human factors in computing systems (pp. 3721e3730) (Seoul, Korea).
[16] Kawachi, I., & Berkman, L. F. 2001. Social ties and mental health. Journal of Urban Health, 78, 458e467.
[17] Li-Tze Hu and Peter M Bentler. 1995. Evaluating model fit.
[18] Perry, R., Drachen, A., Kearney, A., Kriglstein, S., Nacke, L. E., Sifa, R., ... & Johnson, D. 2018. Online-only friends, real-life friends or strangers? Differential associations with passion and social capital in video game play. Computers in Human Behavior, 79, 202-210.
[19] Rachel Kowert, Emese Domahidi, Ruth Festl, and Thorsten Quandt. 2014. Social gaming, lonely life? The impact of digital game play on adolescents’ social circles. Computers in human behavior 36: 385–390.
[20] Ravaja, N., Saari, T., Turpeinen, M., Laarni, J., Salminen, M., & Kivikangas, M. 2005. Spatial presence and emotions during video game playing: Does it matter with whom you play? Presence: Teleoperators and Virtual Environments, 15, 327e333.
[21] Regan Lee Mandryk and Max Valentin Birk. 2017. Toward Game-Based Digital Mental Health Interventions: Player Habits and Preferences. Journal of medical Internet research 19, 4.
[22] Robert D Putnam. 2001. Bowling alone: The collapse and revival of American community. Simon and Schuster.
[23] Sabine Trepte, Leonard Reinecke, and Keno Juechems. 2012. The social side of gaming: How playing online computer games creates online and offline social support. Computers in Human Behavior 28, 3: 832–839.https://doi.org/10.1016/j.chb.2011.12.003
[24] Sheena Lewis, Jason B Ellis, and Wendy A Kellogg. 2010. Using virtual interactions to explore leadership and collaboration in globally distributed teams. In Proceedings of the 3rd international conference on Intercultural collaboration, 9–18.
[25] Shen, C., & Williams, D. 2011. Unpacking time online: Connecting internet and massively multiplayer online game use with psychosocial well-being. Communication Research, 38, 123e149.
[26] Smith, S. M., Smith, S., Booth, K., Smith, S. A., & Zalewski, M. (Eds.). 1996. International theory: positivism and beyond.
Cambridge University Press.
[27] S.M. M Grüsser, R. Thalemann, and M.D. D Griffiths. 2007. Excessive Computer Game Playing: Evidence for Addiction and Aggression?
CyberPsychology & Behavior 10, 2: 290–292.https://doi.org/10.1089/cpb.2006.9956
[28] Vella, K., Johnson, D., & Hides, L. 2015. Indicators of wellbeing in recreational video game players. In Proceedings of the annual meeting of the Australian special interest group for computer human interaction (pp. 613e617) (Adelaide SA, Australia).
