Pooja Patidar, Dr. Sandeep Arya, Dr. Abhishek Shukla^1*

¹ Jaypee University of Engineering and Technology.

*Corresponding Author:Abhishek Shukla, Jaypee University of Engineering and Technology AB Road, Raghogarh, Guna, India 473226, Tel: +91-9098663067; Fax: +91-9098663067

Citation: Pooja Patidar, Dr. Sandeep Arya, Dr. Abhishek Shukla (2024) Social Media and Mental Health: The role of Emotional Investment and Self-esteem. Addict drug sensitiz 5: 129.

Received: December 21, 2024; Accepted: Jauary 10, 2024; Published: January 13, 2024.

Copyright: © 2024 Abhishek Shukla, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Aims: The consumption of social media has grown ingrained in the lives of today's youth. However, previous studies ignored the independent role of emotional perspective, self-esteem, and social media gratification with respect to social media usage and mental health. The purpose of the study is to explore the combination of crucial factors responsible for mental health.

Methods: The current study has integrated the uses and gratification theory (U&G), stimuli-organism-response (S-O-R), and sociometric theory in its earnest attempt to create a theoretical framework for the impact of social media usage on mental health through motives of social media usage (MSMU), emotional investment and self-esteem. A model was proposed to examine the relationship using a structured equation modeling method on data from 660 Indian participants.

Results: The results revealed that the most critical factors influencing mental health were the total time spent on social media (TTSSM), and passive and active non-social usage (ANSU). Furthermore, it was also found that some combination of MSMU, emotional investment, and self-esteem mediated the relationship between TTSSM, ANSU, and mental health.

Conclusion: The findings of the current study present empirical reasoning and information related to social media by representing that gratification, emotional investment, and self-esteem impact mental health. The present study's findings are strong enough to encourage parents, educators, and health professionals to counsel young people about their emotional involvement, building self-esteem, and use of social media.

Keywords: Social Media; Depression; Anxiety; Mental Health; Uses and Gratification Theory; S-O-R Theory, Emotional Investment

In the twenty-first century, the use of social platforms is a constantly increasing activity. Particularly, in the pandemic, there was a surge noticed in online and digital activities. Around 43% of users increase their time spent on social media (Global Webindex, 2020). Worldwide, 4.20 billion (53.6%) users were active by January 2021 (Chaffey, 2021). In India 448 million users were active on social media and 78 million (21%) users were increased during 2020-2021 (Kemp, 2021). Millennials and Generation Z are the major contributors to approximately 67% of social media usage in India (Keelery, 2020a). In the United States, almost 7 out of 10 people use social media to communicate with others, get news, exchange information, and enjoy themselves (Pew Research Center, 2018). Young individuals engage in social media activity for various purposes for example entertainment, identity creation, networking, and maintaining relationships (Ifinedo, 2016).

Studies on social media have shown mixed results as far as its impact is concerned, making it a “doubtful advantage” tool. Previous studies revealed that there is blooming interest in investigating and assessing the features and prevalence of problematic/excessive social media use (Kuss a, 2011).  Some studies showed that individuals who spend more time on social media platforms more tend to experience anxiety and depression symptoms (Banjanin et al., 2015). Likewise, the frequency of social media use is also associated with poor psychological well-being among young adults (Shakya & Christakis, 2017; Kross et al., 2013). Excessive usage of social media has negative consequences for users' personal, professional, and/or social, lives (Griffiths et al., 2014). Pantic and Damjanovic (2012) found a strong link between depressive symptoms and social media use. Facebook is really the most common and most frequently used social media platform and demonstrated that adolescents also use other social media platforms on a regular basis (Lenhart et al., 2015). However, to conclude that poor psychological is an outcome of total time spent on social media (TTSSM) and the frequency of social media use is excessively simplistic.  It’s important to narrow down the factors for poor mental health due to the usage of social media. Another important factor in the extant literature related to social media is the uses and gratification of social media. Although little is known about the uses and gratification of social media’s effect on mental health. The present study contributes to the literature in terms of specific uses and gratification of social media corroborate in the diminishing of mental health.

The Purpose of social media use is different according to various users, for example, for teens who used social media, makes them feel closer to friends (78 percent), makes them more aware (49 percent), and more linked to their family (42 percent), but they also say it makes them feel under pressure to always show their best oneself (15 percent), overloaded with data they get (10 percent), overwhelmed (9 percent), and/or as if they were really missing out (9 percent) (Tompson, 2017). Psychological well-being is defined as a combination of frequent happy feelings and relatively uncommon negative emotions (Diener and Larsen, 1993).

Previous studies on social media usage and mental health-focused frequency and duration of social media use. Considering the usage of social media by young adults, the frequency and duration of social media usage may not be enough to identify the underlying reason for diminishing mental health. Young adults’ usage of social media has increased enormously (Kemp, 2021). Young adults deeply care the social media resulting in the induction of emotions while using social media, for instance, users feel disappointed and disconnected when they were not able to use social media (Woods & Scott, 2016) termed emotional investment. This emotional investment of an individual in social media can be a crucial factor to investigate in diminished mental health. Due little known about the emotional investment and social media, the current study corroborates in literature by investigating which uses and gratification involve/increase the emotional investment leads to an impact on mental health i.e., depression and anxiety (Woods & Scott, 2016)

Many researchers focus on this topic because Well-being is commonly described as a general outcome of social media studies, with effects on psychological indicators such as happiness and life satisfaction. (Chou and Edge, 2012), stress, and quality of life (Bevan et al., 2014), and depression (Tandoc et al., 2015). Despite an increasing number of studies, the link between social media use and well-being remains a matter of debate (Pantic, 2014).

U & G Theory

The uses and gratification approach investigates why and how individuals use the media, while also what gratifications they acquire from it, in contrast to mass media ideas and theories that focus on media effect. U & G theory has its roots in the study of mass communication (Gan and Li, 2018). U & G’s theory states that there are social and psychological incentives to use all types of media that justify the use of certain media over others to meet a particular individual need (Katz and Blumler, 1974). According to U & G theory, different rewards influence user behavior patterns. U & G’s theory reflects the motivation to use specific media and the motivation to use common social media (Gan, 2017).

U&G theory has recently been applied to online photo sharing (Malik et al., 2016), Internet use (Dhir et al., 2017, 2016, 2015), online social media (Dhir et al., 2017), photo tagging (Dhir et al., 2017; Dhir and Torsheim, 2016), and various forms of new popular media, such as Instagram and Flickr (Liu et al., 2013; Bui, 2014; Zhang et al., 2016). U&G has also been utilized to find out the reasons for teen's consistent usage of social media (Mäntymäki and Riemer, 2014), as well as the serious consequences of specific behaviors (Mäntymäki and Islam, 2016).

S-O-R Theory

                The S-O-R theory developed by Woodworth (1918) states that stimuli (S) affect an individual's behavioral response (R) to an organism (O).  Furthermore, Mehrabian and Russell (1974) used this idea to develop the S-O-R framework, which that asserts, a stimulus (S) received from the environment causes humans to evaluate their internal states (organism, O), ending in positive or negative reactions (R) (Mehrabian and Russell, 1974). Environmental signals (stimuli) can activate a person's internal evaluation state (organisms), resulting in psychological and cognitive responses to the stimulus, according to the paradigm. Yang et al. (2021) investigated the impact of metacognitive beliefs and catastrophic misunderstandings on social media users' health concerns. Using the S-O-R model, another study investigated the impact of social media use on weariness and dependency on internalized disorders (depression and anxiety) (Cao et al., 2019). As a result, the S-O-R framework may be able to detect changes in mental health brought on by external stresses with greater accuracy. The S-O-R framework has also been applied to social media research (Wu & Feng, 2021). Furthermore, this theory has been evident in the domain of Marketing (Wang et al., 2007), Web Design (Parboteeah et al., 2009), Information Systems (Animesh et al., 2011), etc., studied the effects of various stimuli on the user's behavioral response. The current study integrates the S-O-R theory in the context of social media usage for example TTSSM and active-passive usage as a stimulus, self-esteem, and motive of social media usage (MSMU) as organism and mental health of an individual as the response.

Sociometer Theory

                Mark Leary and colleagues (1995) developed a sociometer theory to describe the facets of the self-esteem system. Self-esteem, or a person's sense of self-worth or value, maintains track of one's social approval and alerts the self to any potential threats to one's sense of belonging. The sociometer theory is one of the most used theories for self-esteem related studies. According to studies using a number of methods, acceptance leads to increases in state self-esteem, and rejection leads to decreases in self-esteem. These effects emerge when individuals are given false input about approval or rejection in the experiment when individuals envision being rejected or accepted, and when they recollect a past experience of acceptance or rejection (Leary et al., 1995). In the social media context, one’s self-esteem changes according to the approval of others on social media, for example, accepting the invitation, liking, commenting and sharing the post, etc.

Previous studies applied the U&G theory to different uses and gratification of social media in the context of addictive scenarios (Kircaburun et al., 2020). Various previous studies have examined user behavior with reference to a particular social media platform for example Facebook (Hsu et al., 2015), Twitter (Coursaris et al., 2013), and online social networking sites (Chaouali, 2016). However, to evaluate the social media (SM) behavioral response, emotions involved in the process of various activities on social media and users’ self-esteem cannot be ignored. For example, emotional investment and self-esteem negatively affect anxiety (Roberts, 2006) and depression (Watson et al., 2002). Therefore, the current study integrated sociometer, S-O-R, and U&G theory to examine whether the stimuli (social media usage) affect the response (mental health) through organism process (gratification of social media and self-esteem)? In the current proposed model, self-esteem and emotional investment are treated as organisms due to it was evident that self-esteem and emotional investment affect mental health (Nguyen et at., 2019) whereas emotional investment in social media also increases depression and anxiety (Alsunni & Latif 2021).

Social media and Mental health

                The concern of social media on mental health is often represented in unpleasant terms. A previous study on a large sample of adolescents and young adults reported the link between social media usage (SMU) and mental health complications (Royal Society for Public Health, 2022). The use of social media has impacted negatively on mental health, which includes the elements of emotional, physical, and social well-being (Galderisi et al., 2015). National Mental Health Survey (2016) revealed that 1 out of 20 people in India suffers from depression, the most impacted group is below 25 years and above 65 years, and the treatment, the gap varies between 70% to 92% (Duffy et al., 2020). Young individuals with mental health issues are at greater risk of poor mental health for their whole life (Kessler et al., 2007). The impact of social media is an outcome of various confounding factors in diminishing mental health. Previous studies examined various factors for example duration, frequency, active and passive usage, and motives of social media usage which affect mental health. Therefore, it is important to investigate the combination of factors that affect the mental health of individuals. Previous studies reported that extensive use of SM was linked with depression and anxiety (Yan et al., 2017). However, multiple studies have found a link between greater social media use and increased anxiety and depression symptoms (Banjanin et al., 2015; Woods & Scott, 2016; Shukla & Chouhan, 2020), particularly among females (Viner et al., 2019). The studies found that there was an inverse relationship between TTSSM and psychological distress among Australian samples (O’Dea and Campbell, 2011). In contrast, other studies revealed that the use of social media for at least 2 hours daily is connected with psychological distress (Sampasa-Kanyinga and Lewis, 2015). It was also found that most of the studies conducted on social media were self-reported studies in which error cannot be eliminated. Therefore, it is required to use some other techniques/technology to track the TTSSM and various social media platforms. Hence, the current study applied the android application technology to track the respondent’s TTSSM and various social media platforms on which the user spent their leisure time. Therefore, in light of these findings, we hypothesize that,

H1 (a): TTSSM will affect depression.

H1 (b): TTSSM will affect anxiety.

                Concentrating only on the duration spent on social media may not help us to identify the major cause of poor mental health. Therefore, it will be important to include active-passive usage of social media. Gerson et.al. (2017) found that the uses of Facebook can be classified into three subscales: active social use, active non-social use (ANSU), and passive use. According to Burke et al., (2010), users who interact on Facebook contact are less inclined to feel lonely and have better feelings about establishing social capital. Users who are actively involved with the social site, creating material, and chatting with friends are said to be active users. Using Facebook to build social capital (Ellison et al., 2007), call on friends for help (Liu & Yu, 2013), and create social connectedness (Grieve et al., 2013), all that has been related to a number of subjective well-being measures (Ellison et al., 2007). The second component comprises elements that reflect ANSU, in which only liking a post is considered as ANSU (Gerson et.al. 2017). The presence of ANSU was surprising, as it denotes a type of SM interaction that is halfway between active and passive. The ANSU components indicate a degree of SM involvement in which a user approves the content posted by others through liking the post but does not contact directly with friends. Because of its non-social nature, ANSU was likely lumped in with passive use or active use and thus may be disregarded as a distinct degree of engagement (Gerson et al. 2017), whereas scrolling, browsing, and consuming other people's content constitutes the passive mode of using social media.

Individuals using social media actively express their feelings and emotions with others which helps in positive mental health, whereas passive usage results in feelings of envy and inferiority leading to negative mental health. Both active and passive social media use seems to be a factor causing emotional distress and connected to poor mental health. A previous study analyzed that active and passive usage of social media is differentially connected to depression and anxiety (Thorisdottir et. al., 2019). Furthermore, some studies reported that being both actively and passively using social media – specifically related to Facebook - resulted in an increased number of depressive and anxiety symptoms (Escobar-Viera et.al., 2018; Thorisdottir et. al., 2019). In contrast, Banjanin et al. (2015), posited that social media activities and depressive symptoms were not related. One of the studies highlighted that passive usage increases depressive symptoms whereas active usage decreases depressive symptoms (Escobar-Viera et.al., 2018). Most of the studies were conducted on active and passive usage of social media whereas, there is one grey usage also available between active and passive usage i.e., ANSU of social media which may impact mental health. Expanding the strand of research, it is worthwhile exploring the influence of active, ANSU and passive usage of social media on mental health in light of the inconclusive and limited evidence of extant studies. On basis of the aforementioned findings, we hypothesize that,

H2 (a): Active usage of social media will affect depression.

H2 (b): ANSU usage of social media will not affect depression.

H2 (c): Passive usage of social media will affect depression.

H3 (a): Active usage of social media will affect anxiety.

H3 (b): ANSU usage of social media will not affect anxiety.

H3 (c): Passive usage of social media will affect anxiety.

Social Media, Emotional Investment, and Self-Esteem

Social media creates unique social pressure on individuals to be available all the time to respond to the messages and like the content instantaneously (Thomée et al., 2010). A previous study found that "friends typically build emotional closeness links which are unsurpassed in individuals' social lives," and while people may find it difficult to express emotion within their family relationships, they “can often reduce their guard, when one or more cohorts are present (Gaines et al, 1998). Furthermore, emotional experiences in personal relationships have been connected to relational aspects such as the quality of alternative partners, investment, and satisfaction, (Bevan, 2008). A previous study found that higher emotional investment is linked with depression and anxiety among university students (Alsunni & Latif, 2021). Very limited literature is available on the emotional investment of social media users (Woods & Scott, 2016). Therefore, more studies are needed to examine the effect of the emotional investment of social media users on mental health.

It was found that emotional investment plays a crucial role in influencing the self-esteem of social media users (Woods & Scott, 2016). Self-esteem can be defined as a positive or negative perception of the self which develops across time and various situations (Rosenberg, 1965; Brown & Marshall, 2006). As young age is important as individuals are developing their self-identity (Brewer & Kerslake, 2015) during this process feedback on the self possibly affects self-esteem (Valkenburg et al 2017; Erol & Orth, 2011). A previous study revealed a positive correlation between social media use and self-esteem (Valkenburg et al., 2017), whereas Malik and Khan (2015) discovered a negative link between self-esteem and excessive social media use. Previous studies investigated the positive and negative effects of social media on adolescents, and they got that feedback obtained from social media networks, whether negative or positive, has a significant influence on an individual’s self-esteem (Valkenburg et al., 2006). Adolescents and young adults were more vulnerable to low self-esteem (Orth et al., 2015) leads to depression and anxiety (McLaughlin & King, 2015).

Social media forums are considered to be crucial tools for maintaining relationships, entertainment, gathering information, passing time, and creating a social network (Quan-Haase and Young, 2010). As per the U&G theory, how and why of any particular media been used by various individuals? Entertainment as a factor for Facebook usage; and coolness, creativity, documentation, and surveillance as factors for Instagram usage, have been put forward by studies in the past (Alhabash and Ma, 2017). However, little is known about which gratification of usage of social media affects mental health. Social media usage has been reported both intrusive and emotional consequences for individuals.

 A previous study reported that social media use is related to physical, mental, and interpersonal consequences which include sleep loss, lack of exercise, effect on performance, depression, anxiety, and diminished interpersonal relationship with family and friends (Andreassen, 2015). Studies reported the effect of social media usage on mental health, still, more knowledge is needed to shed insights on which specific social media uses lead to, impair, or are of no relevance to mental health (Meier and Reinecke, 2020). The previous studies demonstrated motives of social media usage (MSMU) affect mental health, however, the effect of MSMU as a mediator is still missing. Hence, it is worth investigating how active, passive usage and most importantly non-active use can impact mental health through various motives of social media usage.

Filling the gap arising out of the inconclusiveness of the extant studies regarding the identification of factors responsible for diminishing mental health, the current study contributes its bit to the coffers of the academic literature by proposing a model in which MSMU, emotional investment, and self-esteem used as a mediator to investigate the link between social media usage (TTSSM and active-passive usage) and mental health (anxiety & depression).

Previous studies reported the direct effect of social media usage (TTSSM, active-passive, MSMU) on mental health, still, there were a series of activities conducted by the participant while using the social media for example time spent on specific use may majorly affect the mental health. Even, with the studies so far conducted on children and adolescents, some studies have been conducted on young adults (Stronge et.al., 2019) who are more vulnerable to poor mental health. To our best knowledge and belief, this is the first study for examining various factors (total time spend on social media, active and passive usage, the motive of social media usage, emotional investment, and self-esteem) which may contribute to mental illness with the same sample and use of the android application to track the total time spent on social media and various social media platforms. The current study proposed serial mediation models (SMM) with various mental health symptoms for example SMM 1 (Figure 2) for depression as mental health symptoms likewise, SMM2 (Figure 3) for anxiety, which includes MSMU (M1), emotional investment (M2) and self-esteem (M3) as mediators between social media usage and various mental health symptoms. Therefore, along with other hypotheses and on the basis of the above findings we hypothesized that:

Figure 2: Serial Mediation Model 1 (SMM 1)

Figure 3: Serial Mediation Model 2 (SMM 2)

H4 (a): MSMU, emotional investment, and self-esteem mediate the relationship between TTSSM and depression.

H4 (b): MSMU, emotional investment, and self-esteem mediate the relationship between TTSSM and anxiety.

H5 (a): MSMU, emotional investment, and self-esteem mediate the relationship between active- ANSU- passive usage and depression.

H5 (b): MSMU, emotional investment, and self-esteem mediate the relationship between active-ANSU-passive usage and anxiety.

Participants, design, and procedure

                The participants were selected through the snowball sampling method during the period to March 2021. The responses for the current study were collected from students aged 18 to 19 years. During this time restaurants, clubs, gyms, etc., opened with COVID-19 protocol except for schools and universities. Due to schools and universities being closed, the online method was adopted to conduct the study. The questionnaire comprised several questions that helped the researcher to determine the participant's eligibility to participate in the present study. Students’ selection was based on a few criteria, which were included in the screening questionnaire. Firstly, participants desired to be 18 years of age. Secondly, participants are capable of accessing the Internet. Thirdly, participants should have an account on different social media platforms, fourth, they had spent significant time on social media on a daily basis, and lastly, the students are using android smartphones. The current study was approved by the university’s ethical committee. Making use of the existing participants for getting newer ones in, the snowball sampling method was used as it is a non-probability referral chain-forming sampling method (Ghaljaie et al., 2017). The primary data was taken at the onset of the study and through the primary data, the respondents were encouraged to refer their friends, family, and acquaintances to take part in the study. The Snowball sampling technique is specifically used when the population is unknown. This sampling method has been used in this study as participants were not available physically in the university as universities were shut down due to COVID-19. Recently few studies have used this method successfully to investigate different mental health issues in the population (Arafa et.al., 2021).

                Initially, the students were contacted for participation in the survey. After receiving consent from the students, they have been encouraged to contact their interested friends as much as possible to participate in the study, henceforth the chain was created for the participants to take part in the study. To encourage the students to take part in the study and spread the word among their acquaintances, they have been given non-monetary incentives for example extra time to submit their assignments, etc. Keeping the participation voluntary, the respondents were anonymous and could withdraw at any point in time. Setting the context, the study objectives, and data confidentiality were explicitly taken care of by the ethical committee of the university on the first page of the form.  The survey was hosted on Google forms and an anonymous link was generated to be further shared with the target population. Participants signed the consent form for the study and responded to the questions related to socio-demographic characteristics, COVID-19, social media usage, emotional investment, self-esteem, motives of social media usage, and mental health. Subsequently, after receiving the consent form with a few eligibility questions and being selected for the study, participants were asked to install an android application named “Stay Free” on their respective android devices. The Stay Free app is a self-control and productive app that tracks the phone usage and time spent on various activities like social media (Facebook, Instagram, Twitter, WhatsApp, etc.), entertainment (YouTube, etc.,), education (Google classroom, online class app, etc.,), gaming (Ludo, etc.,), utility activity, etc., (Waddilove, 2020). The participants were instructed that they must keep the android app for at least seven days for tracking their social media usage. After seven days of usual social media usage, participants must submit the screenshots through the google form (Figure 1). Two-phased methodologies were adopted in the study. Firstly, the participants (N=50) were contacted over the phone and were requested to fill out the consent form along with a few eligibility questions, and afterward, participants were asked to send the same consent form and other staged instructions (for example eligibility questions, android application installation and reporting of the data) to their friends, family, and acquaintances. 

Figure1: Screenshot of android application.

Participants

The survey assessed on 908 participants of which 9 participants were found COVID-19 positive and 31 participants who were unhealthy from the chronic disease were excluded from further analysis. Out of these 868, 660 (76%) responded by completing the questionnaire and were considered for advanced statistical analyses. The sample consists of 52.1% male (n = 344) and 47.9% female (n = 316). Participants resided in 13 Indian states, with residing 44.8% (n = 296) in Uttar Pradesh and 29.1% (n = 192) in Madhya Pradesh state. Additionally, 87.3% (n = 576) have not experienced COVID-19 symptoms in the past by the participants (Table 1).

Table 1 Descriptive characteristics (N=660)

ANSU: Active non-social usage; MER: Maintain existing relationships; MNPS: Meet new people and socializing; MEPO: Make express, present or more popular oneself; PT: Pass time; TMT: As a task management tool; EMT: Entertainment; IAE: informational and educational.

Measure

The survey included ad-hoc sections and standardized questions. There were three sections, detailed below.

Section A contains the questions related to social-demographic characteristics and health status. Participants were asked about their gender (male and female), location (State), COVID-19 symptoms or tested positive (Yes or No), and chronic disease (Yes or No). The purpose of asking the COVID-19 and chronic disease-related questions is to ensure the participants should not be diseased with COVID-19 or any chronic disease at the time of the survey.

Total time spent on social media.

The TTSSM was calculated through an android application. The total time was calculated as the summation of time spent on every social networking platform. In this study, the total amount of time spent on Facebook and Instagram was considered for further analysis.

Active-Passive usage of social media.

Section B contains the questions related to active and passive social media use (Gerson et.al., 2017) and social media usage (Mehmet et.al.,2016). Passive active use measure (PAUM) is a Facebook use questionnaire with 13 items designed to identify the Facebook user’s engagement activity. Respondents were asked to report their engagement activities using a 5-point Likert scale (1-never, 5-very frequently). Gerson et.al. (2017) found that the uses of Facebook can be classified into three subscales: active social use, ANSU, and passive use.  When users and their friends got into direct communication in form of writing posts or comments, that constitute active social use. Merely liking a post in conversations between friends and users constituted active non-social use. And, finally, instances where the content was consumed but no interaction happened between the users and their friends on social media sites, constituted the passive social use. Showing fine reliability (internal as well as test-retest) and validity, these multi-item scales fit in well with requirements (Gerson et al., 2017). Originally, the PAUM was designed to measure the active and passive usage pattern of Facebook, but the current study used to identify the usage pattern of Instagram as well due to mostly the same characteristics found in Facebook and Instagram.

Uses and Gratification of social media.

Facebook Usage Aim (FAU) (Horzum, 2016) was originally designed to identify the aim of using Facebook. This scale was used to identify the user’s objective to use Facebook and Instagram due to almost the same characteristics found in both social media platforms named as motives for social media use (MSMU). The scale included 30 items which were classified into seven various subscales for example a. Maintain existing relationships [MER], b. Meet new people and socialize [MNPS], c. Make express, present, or more popular oneself [MEPO], d. Pass time [PT], e. As a task management tool [ATMT], f. Entertainment [ENT] g. informational and educational [IAE] in the form of a 5-point Likert scale. The scale has good internal consistency (0.91), test-retest reliability, and construct, discriminant, convergent, and divergent validity.

Emotional Investment in social media.

Next in the line, section B comprises questions related to the social media use integration Scale for emotional investment (Jenkins-Guarnieri et al., 2013). This scale includes a 10-item measure of self-reporting used to measure two-dimension termed Social Integration and Emotional Connection (SIEC) and Integration into Social Routines (ISR). Making use of a 5-point Likert scale, the respondents are asked to report their levels of emotional investment (1 = Strongly disagree, 5 = Strongly agree). The Social Media Use Integration Scale reported decent reliability with Cronbach's alpha of .89. The reliability of the current sample was reported with Cronbach's alpha of .866.

Self-Esteem

Self-esteem was measured through Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965). Participants rated their trait self-esteem levels on- a 4-point Likert scale from “Strongly disagree (1)” to “Strongly agree (4).  There is no cut-off and used as a continuous measure, a higher score depicted higher self-esteem. The measure indicates high reliability with Cronbach's alpha of .88 in the current study.

Section C includes the mental health-related questions. Mental health measured through various subscales includes depression &anxiety.

Depression

Depression symptoms were evaluated via the Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977). The CES-D is a self-report instrument that includes 20 items with excellent reliability and validity (Radloff, 1977). 

Anxiety

 Anxiety symptoms were measured via a generalized anxiety disorder questionnaire (GAD-7). GAD-7 is a 7-item self-reported questionnaire designed to identify possible causes for anxiety and its severity (Spitzer et al., 2006). Participants are asked to report the questions for example how often they have been upset for the last two weeks experiencing any of the seven main anxiety disorder symptoms. Participants report the anxiety (GAD) symptoms on a 4-point Likert scale ranging from 0-to 3 (1 = Not at all, 3 = Almost every day). Thus GAD-7 scores range from 0-to 21 representing scores of 5 for mild, score 10 for moderate, and 15 for severe symptoms. The scale has demonstrated excellent internal consistency (α = .92) and reliability (Spitzer et al., 2006).

Statistical Analyses and Results

The statistical analysis was conducted using SPSS version 26 and Jamovi 2.2.5 based on R statistical tool (IBM, Armonk, NY, USA; Jamovi Project, 2021) to perform descriptive statistics analyses, reliability analyses, and bivariate correlation. Additionally, IBM SPSS-AMOS version 26 (Byrne, 2016; Sharma et al., 2017) was used to analyze the relationships proposed in the model. The indirect effect was calculated using SPSS-AMOS version 26 indirect effect plugins (Gaskin et al., 2020).

The screenshot of the data reported by the participants was entered into the SPSS and Jamovi, raw scores of active, and passive social media usage, use of social media, depression, and GAD were calculated. The data of the proposed model were analyzed using structured equation modeling (SEM). SEM has been employed in the regression method of mediation analysis due to SEM provides a more suitable interpretation mediation framework for mediation analysis and various types of analysis (Kraemer et al., 2001). The maximum likelihood technique is adopted to analyze the hypothesized models, shown in Figure 2. The current study proposed social media referred to as the stimulus, various gratification, emotional investment, and self-esteem is referred to as an organism, and mental health is referred to as a response, inspired from S-O-R theory. The statistical analyses were conducted in two stages, firstly estimated the measurement model and second structured equation modeling were performed on the proposed hypothesized model. 

Measurement Model and Construct Reliability

                To validate the proposed model, the overall model fit was assessed, construct validity, and reliability. The overall model fit was evaluated using the following fit indices i.e., chi-square (X2), degree of freedom (df), normed-fit index (NFI), normed chi-square to degree-of-freedom (CMIN/DF), goodness-of-fit index (GFI) (Bonett, 1980), comparative fit index (CFI) (Bentler. 1990), adjusted goodness-of-fit index (AGFI), standardized root mean squared residual (SRMR) (Hu and Bentler,1999), and root means a square error of approximation (RMSEA) (Browne and Cudeck, 1992). The purpose of selecting these fit indices is to confirm about the data fit the various serial mediation proposed model (McDonald and Ho, 2002). Table 3 presented the various proposed model fit indices and their threshold values which showed a good fit.

                The current study used composite reliability (CR), discriminant validity, and convergent validity to assess measure construct reliability and validity. The average variance extraction (AVE) technique was utilized for convergent validity and most maximum shared variance (MSV) and √AVE were applied to estimate the divergent validity of measures. Also, Cronbach alpha (α) was determined to confirm the reliability of the instruments.

Table 2: Person correlation between various constructs.

Note∗ p ≤ .05, ∗∗p ≤ .01, ∗∗∗p ≤ .001.

ANSU: Active non-social usage; TTSSM: time spent on social media, LOC: location, MER: Maintain existing relationships; MNPS: Meet new people and socializing; MEPO: Make express, present or more popular oneself; PT: Pass time; TMT: As a task management tool; ENT: Entertainment; IAE: informational and educational, EI: emotional investment, SE: self-esteem, DEP: depression, GAD: anxiety.

Table 3: Fit indices for proposed model and their respective threshold.

DEP: depression, GAD: anxiety.

                Construct reliability, discriminant validity, and convergent validity were investigated in the current study. As shown in Table 4, the AVE values of all latent constructs were higher than the threshold level (0.5), implying convergent validity. The square root of each construct's average variance extraction values was higher than the inter-construct correlation (Table 2). In addition, the values for the maximum shared variance (MSV) were lower than the comparable AVE values. (Table 4). According to the findings, the current study data satisfy the condition of discriminant validity (Matemba & Li, 2018).

Table 4: Reliability and validity of the constructs.

ANSU: Active non-social usage; MER: Maintain existing relationships; MNPS: Meet new people and socializing; MEPO: Make express, present or more popular oneself; PT: Pass time; TMT: As a task management tool; ENT: Entertainment; IAE: informational and educational; EI: Emotional Investment.

Correlational Analysis

                Correlation analysis and descriptive statistics are shown in Table 2. TTSSM was found to be associated with the majority of the constructs in the correlational analysis. For example, TTSSM is most frequently linked to depression and anxiety (r =.633, p <0.01, r =.539, p< 0.01). TTSSM was also linked to active (r =.119, p< 0.01), active non-social (r =.258, p< 0.01), and passive (r =.278, p< 0.01) usages. Eventually, it was observed that TTSSM was related to a variety of MSMU besides informational and educational use (IAE). Surprisingly, TTSSM was positively related with emotional investment (r = .425, p< 0.01). When compared to ANSU and active social media usage, passive social media usage was found to be more strongly associated with depression and anxiety. Despite the fact that various uses and gratifications of social media have been linked to depression and anxiety, when compared to other social media motives, the most significant motive was entertainment with depression (r =.404, p< 0.01), anxiety (r =.387, p< 0.01).

However, the least depression found when an individual uses the social media for information and education (IAE) (r = .102, p< 0.01), anxiety in maintaining an existing relationship (MER) (r =.192, p< 0.01). Emotional investment is significant factor found to be positively associated with depression (r = .430, p< 0.01), anxiety (r = .279, p< 0.01), whereas self-esteem found to be correlated negatively with depression (r = -.419, p< 0.01), anxiety (r = -.325, p< 0.01). Squared multiple correlations of various latent constructs are also represented in Table 5.

Table 5: Squared multiple correlation of various latent constructs.

MER: Maintain existing relationships; MNPS: Meet new people and socializing; MEPO: Make express, present or more popular oneself; PT: Pass time; TMT: As a task management tool; EMT: Entertainment; IAE: informational and educational, EI: emotional investment, SE: self-esteem, DEP: depression, GAD: anxiety.

Structure Equation Modelling Analysis

                The current study data showed convergent and discriminant validity, we examined the hypothesized relationship of proposed model through structural equation modelling method (Figure 1). For SMM 1, X2 = 20.47, df = 18, CMID/DF = 1.13, GFI = 0.996, AGFI = 0.974, NFI = 0.993, CFI = 0.999, SRMR = 0.0169, RMSEA = 0.014. These values are considerably within the permissible range of indices (Table 3), thus showed a good model fit. Subsequently, found that statistical outcome accounted for of 41%, 47%,31%, 29%, 34%, 25%, 28%, 31%, 18%, 20%, 22% the variance in depression, anxiety, MER, MNPS, MEPO, PT, TMT, ENT, IAE, self-esteem and emotional investment respectively.

                As shown in Table 6, the direct positive effect was found between TTSSM and depression (β = .497, p<0.01), TTSSM and anxiety (β = .448, p<0.01). Therefore, H1(a) and H1(b) were accepted. Whereas ANSU effect anxiety (β = .081, p<0.05) and passive usage positively affect depression (β = .161, p<0.01). Hence, H3 (b) was rejected and H2 (c) was accepted.

Table 6: Standardized direct effects for the paths of various model.

ANSU: Active non-social usage; TTSSM: Total time spent on social media; MER: Maintain existing relationships; MNPS: Meet new people and socializing; MEPO: Make express, present or more popular oneself; PT: Pass time; TMT: As a task management tool; EMT: Entertainment; IAE: informational and educational, COG: cognitive distortion, DEP: depression

N = 660, * p < 0.05,** p < 0.01.

In the present study, the full path of significant mediation which includes MSMU, emotional Investment (EI) and self-esteem (SE) has been depicted in Table 7. The significant mediation was found between TTSSM and depression through IAE-EI-SE (β = -.024, p<0.05), MER-EI-SE (β = .034, p<0.05), MEPO-EI-SE (β = .032, p<0.05) and ENT-EI-SE (β = .421, p<0.05). Hence, H4 (a) was accepted. Eventually, mediation test found that anxiety was influenced by TTSSM through IAE-EI-SE (β = -.025, p<0.05), MER-EI-SE (β = .035, p<0.05), MEPO-EI-SE (β = .033, p<0.05) and ENT-EI-SE (β = .414, p<0.01); MER-EI-SE (β = .036, p<0.05), MEPO-EI-SE (β = .036, p<0.05) and ENT-EI-SE (β = .406, p<0.01), MER-EI-SE (β = .035, p<0.05), MEPO-EI-SE (β = .035, p<0.05),  and ENT-EI-SE (β = .410, p<0.01) (Table 8). Table 7 and 8 represents the significant path of structural model.  Therefore, H4 (b) was accepted. 

MSMU, emotional and self-esteem does not mediate the relationship between active usage and depression, whereas active non-social usage influence depression via MEPO-EI-SE (β = .017, p<0.05), ENT-EI-SE (β = .069, p<0.01) and IAE-EI-SE (β = -.009, p<0.05) and passive and depression via MER-EI-SE (β = .028, p<0.05), MEPO-EI-SE (β = .030, p<0.05), ENT-EI-SE (β = .124, p<0.05) and IAE-EI-SE (β = -.010, p<0.05). Hence, H5 (a) was partially accepted due to few MSMU factors, EI and SE mediates the relationship between social media usage and mental health. Furthermore, mediation test was examined for anxiety and found that passive usage effect anxiety through MER (β = .030, p<0.05), MEPO (β = -.031, p<0.05), ENT (β = .122, p<0.01) and IAE (β = -.011, p<0.05); active non-social through MEPO (β = .017, p<0.05), ENT (β = .068, p<0.01) and IAE (β = -.009, p<0.05) ; subsequently, active non-social usage via MEPO-EI-SE (β = .019, p<0.05), ENT-EI-SE (β = .067, p<0.01) and IAE-EI-SE (β = -.009, p<0.05) and passive and depression via MER-EI-SE (β = .031, p<0.05), MEPO-EI-SE (β = -.034, p<0.05), ENT-EI-SE (β = .119, p<0.01) and IAE-EI-SE (β = -.011, p<0.05)  and passive through MER-EI-SE (β = .030, p<0.05), MEPO-EI-SE (β = .034, p<0.05), ENT-EI-SE (β = .120, p<0.01) and IAE-EI-SE (β = -.011, p<0.05),. Therefore, H5(a) and H5 (b) partially accepted.

Table 7: Standardized Indirect effects for the paths of various model.

<t