Social Media’s Negative Impact on Mental Health Subjugated by the Advantage on Young Adults

The present research aims to investigate whether presumed negative effects of social media use on mental health have been outweighed by positive factors that contribute influential advantages. Previous research has demonstrated that social media has only negative effects on a person by emphasizing that social media is a major contributor to poor mental health and thus this research examines the relationships between social media impact on learning new knowledge and increased productivity, learning new knowledge and increased productivity on task performance, and direct social media impact on task performance. Researchers used a quantitative approach, collecting questionnaire from 100 respondents that was distributed to young adults (18-26 age), who are currently located in Jakarta, Indonesia. Contrary to what has often been assumed, social media is linked with task performance outcomes. Correlation and data analysis concluded that social media were significantly related with positive associations. Adoption of social media that is used correctly is the main driving force of the beneficial outcomes, which will be explained by the hypotheses centered on direct relationship of social media impact and task performance. The future implications of this study will be discussed.


INTRODUCTION
Is social media responsible for the excessive usage of social media that results in negative effects such as body image issues and mental health? Ever since the emergence of social media till present times, it has dominated over people's center of attention and is presumed to have a negative impact that contributes to the decline of their mental health and exacerbating mental illnesses (Karim et al., 2020). While it is commonly known that all internet users have access to smartphones or other devices, young adults and adolescents are exposed to a vast amount of unfiltered information, available by just one click away. Recent research reiterated the frequent use of social media and technology directly relates to adolescence, within ages 15-18, who experienced body image concerns, decreased sleep quality, risk of social comparison, and depression (Nesi, 2020).
According to Nesi (2020), 40% of adolescents use their smartphones before retiring to bed and 36% of them check their phones at least once throughout the night. Ostic et al. (2021) stressed that the excessive use of social media has a direct impact on adolescents and young adults' psychological well-being. In this vein, users are correlated with smartphone addiction due to the fear of missing out, which may inadvertently cause a decline in mental health and other issues. Beyens et al. (2021) recorded an interesting finding that passive public use of social media has negative effects such as "positivity bias" on users' psychological well-being. Positivity bias is an inclination to post positive information than negative information online (Beyens et al., 2021). As evidenced by Hu & Liu (2020), the social comparison theory claims that people compare themselves with others, and passive social media use is linked with social comparison, implying that the use of social media lowers users' self-esteem.
Despite the negative connotations that persist with social media impact, the advantage of social media consumption is underexplored, and it can otherwise provide as a beneficial source of information as well as providing an advantage for people in many aspects of life, such as staying informed of current news and an exchange of new knowledge and skills through the daily contents that contain values (Intyaswati et al., 2021). As Intyaswati et al. (2021) stated, social media has enabled users worldwide the ease of accessibility to sharing and learning new information, in this case, political knowledge. Thus, it can be inferred that even though it was deemed negative on people's mental health, people are also very reliant and depend on social media whether it is for personal or work matters since it increases productivity. Hence, social media's nature of fast accessibility enables users of all ages to maximize productivity and task performance (Khan et al., 2021).
A recent study by Vaingankar et al. (2022) from Singapore have recorded youth users, aged 15-24, to have increased social media use from 65% in 2013 to 80% in 2016 and their life satisfaction and happiness remained fairly constant despite increased social media consumption. This notably explains how the 28% gap of increase did not contribute wholly to negative impact on youth users. Vaingankar et al. (2022) published the findings of their study using 95 participants who had a history of psychological distress. The findings showed that despite having access to several social media platforms, such as Instagram, TikTok, WhatsApp, and YouTube, these participants used them for networking and sharing online information about "news, fashion, hobbies, health, and employment." With regard to negative stigmatization of social media consumption, this study asserted that it is actually associated with positive attributes including happiness, positive growth and attitude, and mental health awareness (Vaingankar et al., 2022). One of the participants from this study shared how she felt sad and posted it on Twitter and her close friend would immediately send her countless memes to cheer her up (Vaingankar et al., 2022). It suggests that social media can be an outlet of expression and it can be a safe place, if used correctly, for these young users to also receive help.

Problem Background
Social media is to blame for exacerbating mental health issues. This study investigates the impact of social networking on mental health. It discovers that people who use social media frequently have better mental health than those who do not use social media. This study's findings, as shown in Figure 1, were divided into two categories of mental health: anxiety and depression. Time spent on social media making a positive impact in the field of mental health. The impact of social media on mental health should be investigated further through qualitative research and cohort studies (Karim et al., 2020).
Mental health is defined as a state of well-being in which people understand their abilities, solve daily problems, function well, and contribute significantly to the life of their communities. The benefits and drawbacks of social media on mental health are currently being debated. Social networks play an important role in preserving our mental health. Mental health, health behaviors, physical health, and mortality risk are all influenced by the quantity and quality of social relationships. Several studies on the effects of social media have been conducted, and it has been proposed that long-term use of social media platforms such as Facebook may be associated with negative signs and symptoms of depression, anxiety, and moodiness. Furthermore, social media may place a great deal of pressure on people to conform to stereotypes that others want to see and to be as popular as others (Karim et al., 2020).
Based on the discussion, the research questions of this study are: (1) Does social media impact young adults to learn new knowledge? (2) Does social media impact young adults to be more productive? (3) Does learning new knowledge have a positive impact on task performance? (4) Does being more productive have a positive impact on young adults' task performance? Figure 1. Gender distribution of social media audiences worldwide as of January 2020 (Karim et al., 2020) The purpose of this research is to determine whether using social media provides advantages or benefits that may outweigh the disadvantages. In doing so, the main goal of this research is to raise awareness of the unrealized benefits of social media, as opposed to its drawbacks by focusing on young adults whose attention spans may be short-term, as well as its age range is recognized for its prime years.

Theoretical background on social media
• Internet users online are up to date with the latest devices and technologies, allowing them easy and wide access to a vast source of information-attainable within seconds such as smartphones that heightened the complexity and dimensionality of internet and social media usage (Krairit, 2018). Aichner et al. (2021) have gathered an informational evaluation of how social media was generally defined over the span of 25 years such as the term computer-supported social networks (CSSNs) that emerged in the 1996 since it connects people with computer networks and modern times since 2019, the term has changed to social media (SM) which contains a vast spectrum of information and communication online for internet users worldwide.
• Social networking services (SNS), which offer a platform for interaction with user-generated material online, communication, sharing, and content exchange, are another sub-domain of social media (Alhabash & Ma, 2017). Young adults (18-26 years old) make up the majority of the target audience, according to data statistics from Alhabash and Ma (2017), with social media use increasing by 90%. In extension to this is the fact that more than half of these young adults use Instagram, making it as one of the leading social media platforms (Alhabash & Ma, 2017).
• As previously mentioned before regarding mobile technologies through the devices-cellphones, smartphones, tablets-that provide easy and fast access to the wide source of social media, these mobile technologies also prove to be an excellent source of education and learning for students in higher education institutions (Ansari & Khan, 2020). This study also revealed about 67% students, who were subject to the study, concurred that social media and mobile devices helped them succeed academically and grow as individuals (Ansari & Khan, 2020). Another finding using Chi analysis, which involved school students (aged 12-19 years old), proves that social media did not have any negative association or impact on students' academic performance and skills, regardless of the time spent on social media apps (El-Badawy & Hashem, 2015).
• Another study with philosophical underpinnings similarly links some of the daily uses of social media to the idea that people use them to leave an impression on others, as discussed in Goffman's book The Presentation of Self in Everyday Life (Qi et al., 2018). This study also discussed the relationship between Goffman's quote and how people utilize social media or SNS (social networking services) to present a slightly more sophisticated version of themselves (Qi et al., 2018). In doing so, Qi et al. (2018) reveal that people use social media to present a more positive picture of their current lives, as though publishing to a specific audience.

The influence of social media on human behavior
Social isolation may be lessened by using social media since it fosters a sense of connection with relevant people (Twenge & Campbell, 2019). Indeed, social media enables people of all ages to engage with both strong ties, such as coworkers, acquaintances, and strangers, and tight links, such as family, friends, and relatives, using their sense of belonging in various communities (Chen & Li, 2017;Roberts & David, 2020). As a result, despite concerns about the potential adverse effects of social media use on one's well-being, there is an increasing number of studies highlighting social media as a new communication channel (Twenge & Campbell, 2019;Barbosa et al., 2020), emphasizing that it can play an important role in developing one's presence, identity, and reputation, facilitating social interaction, forming and maintaining relationships, and sharing ideas (Carlson et al., 2016), which may be significantly correlated with social support (Chen & Li, 2017;Holliman et al., 2021).
Recent research (e.g., David et al., 2018;Bano et al., 2019;Barbosa et al., 2020) suggested that the impact of smartphone usage on psychological well-being is dependent on the amount of time spent on each type of application and the activities that users engage in. Putnam (1995) succeeded in defining social capital as a social life containing a variety of factors including networks, norms, and trust, that enable people to act more efficiently to achieve shared goals. According to Li and Chen (2014), social capital consists of "resources embedded in one's social network that can be assessed and used for instrumental or expressive returns such as mutual support, reciprocity, and cooperation." Putnam (1995,2000) described social capital as having two dimensions: bridging and bonding, taking into account the various norms and networks in which it occurs. Bridging social capital refers to the inclusive nature of social interaction, which occurs when people from different backgrounds connect through social networks. As a result, heterogeneous weak ties are typically used to provide bridging social capital (Li & Chen, 2014). This dimension broadens people's social horizons and perspectives, as well as their access to resources and information. The social and emotional support that each individual receives from his or her social networks, particularly close ties, is referred to as bonding social capital (e.g., family and friends).
Ultimately, it is anticipated that social capital and psychological health are associated (Bano et al., 2019). Williams (2006) stressed that the importance of interaction in fostering effective connections, which result in advantageous outcomes including emotional support. The social capital theory is then used to examine the relationship of how social media use and psychological health are correlated in the following subsections.

Research Model and Hypothesis
This research paper is divided into five parts mainly based on the impact of social media. Figure 2 illustrates the conceptual framework model where it examines the relationship between social media impact from daily use and learning new knowledge, increased productivity, and positive impact on task performance.

Social media impact to learn new knowledge
Many corporate companies these days use modern technologies such as information technology (Marbun et al., 2020). There's no denying that the internet and its different services are a valuable source of knowledge and information (Mauroner, 2016). One example of a recent breakthrough in information technology is social media (Marbun et al., 2020). One of the studies by Cao et al. (2016) concluded that social media improves knowledge transfer and job performance. Therefore, based on the discussion, the first hypothesis of this study is: H1: Social media has an impact on young adults to learn new knowledge.

Social media impact to increase more productivity
Alternatively, while some may view social media as unproductive if it is solely used for personal networking matters such as socializing, social media plays a critical role in boosting productivity and creativity in the workplace, particularly for employees (Babu et al., 2020). In terms of work productivity, social media can directly impact people in the workplace to be more productive due to the elevated communication skills that social media provide-social networking (Babu et al., 2020). Another study has also confirmed that social media has elevated and increased productivity through its easy accessibility to collect and share information quickly such as improved recruitment processes that decreases hiring expenses for both big and small businesses (Ganga, 2017). Therefore, based on the discussion, the second hypothesis of this study is: H2: Social media has an impact on young people to be more productive.

Learning new knowledge impact on task performance
One uses what they have previously known and perhaps use it to learn something new, commonly identified as prior knowledge. Earlier research claimed that prior knowledge has a significant impact on learning engagement (Rodrigues, 2007;Pecore et al., 2017). Prior knowledge may reduce cognitive load, resulting in greater learning engagement (Van Riesen et al., 2019). According to Yang et al. (2018), self-regulated learning improved the relationship between prior knowledge and learning. Therefore, based on the discussion, the third hypothesis of this study is: H3: Learning new knowledge has a positive impact on task performance.

Being productive impact on task performance
Self-regulated learning can improve the relationship between prior knowledge and learning engagement (Yang et al., 2018). Even more importantly, instructional support is especially effective in situations with low cognitive load and high prior knowledge (Seufert et al., 2007). Furthermore, prior knowledge has a significant impact on students' self-regulation. When prior knowledge is lacking, self-regulated learning can help to improve learning performance (Yang et al., 2018). Students receive explanations that connect prior knowledge to new concepts from teachers or peers (Williams & Lombrozo, 2013). Furthermore, help-seeking behaviors are selfregulated learning strategies that determine how the quality of help influences learning engagement (Ryan et al., 2005). As a result, when evaluating the impact of prior knowledge on learning engagement, we should consider prior knowledge as a variable that interacts with both cognitive load and help obtained through self-regulated learning to influence learning engagement. Therefore, based on the discussion, the fourth hypothesis of this study is: H4: Being more productive has a positive impact on task performance.

Social media impact (SMI) has a direct relationship on task performance
One notion worth investigating is the media-richness theory, which claims that increased exposure to a communication medium will improve information processing and task performance (Suh, 1999). Since people are wired to process complicated level information much like a computer, Daft and Lengel (1986) proposed that particular media (text, audio, video, and face-to-face) will help businesses to process information on a more advanced and efficient manner. Another study in the same vein claimed that communication results in more advanced task performance and enhances an employee's capacity to perform successfully for the business (Marcotte et al., 2020). Motowidlo et al. (2003) also supplemented how a job performance consist of two subdivisions, task performance and contextual performance. Whereas task performance is linked with an employee's primary responsibilities, contextual performance refers to tasks or activities outside of the job responsibility (Motowidlo et al., 2003). With respect to task performance, an employee would therefore contribute to the company by performing well and Ndung'u et al. (2021) disclosed how job satisfaction is linked to social media use and job performance. As a result, the use of social media involving communication for social or work purposes contributes to higher task performance since task performance is a subdivision of job performance. In addition to this rationale, communication would entail media that bridges information processing between people, thus this hypothesis was formed. Therefore, based on the discussion, the fifth hypothesis of this study is: H5: There is a direct relationship between SMI and task performance.

METHODS
This research proceeded using quantitative method that will examine as well as analyze whether social media has a positive impact that implies a positive correlation with people's mental health which possibly influences hypothesized relationships, the specific target of the respondents is young adults in Indonesia. For the quantitative data, researchers distributed questionnaires to respondents in order to gather information from their responses. In the same vein, the questionnaires would be classified as quantitative due to the close-ended questions with multiple choice answers using the 5-point Likert scale for researchers to be able to draw a conclusion based on the sample of the data. The requirements of the respondents would be people aged from 18-26 years old who currently live in Jakarta, Indonesia. Other than that, they must also be familiar and actively use social media on a daily basis whether it is for social or work purposes.
The questionnaires are comprised of two sections, the first section includes the general questions for the respondents' profile as demonstrated in Table 1. There are four close-ended questions for each variable (social media impact and new knowledge, productivity, task performance, productivity impact and task performance, social media impact and task performance). In line with Roscoe's (1975) guidelines for research, they proposed the sample size to be greater than 30, but less than 500 thus this research has determined the sample should be gathered for at least 100 (Memon et al., 2020).

Demographic Analysis
This research used a demographic analysis that involved high school students, college students, and workers or rather classified as young adults-males and females from the ages of 18 -26, in Jakarta, Indonesia. The data was collected anonymously with experience of correlation of direct activity with social media platforms such as Instagram, Twitter, YouTube, Facebook, TikTok, and WhatsApp, with daily consumption use of at least 1 hour per day. This research was conducted by researchers who distributed questionnaires using Google forms to 100 young adults in Jakarta. Table 1 illustrates the demographic characteristics of the respondents.

Scale Measurement
The survey was conducted using a questionnaire to test the hypothesized models in Figure 2. The questionnaire was designed according to the research framework model in Figure 2, referring to the five hypothesized models. The questions from the questionnaire were divided into two sections. The purpose of the first section is to identify the respondents' profiles such as age, gender, education level, and current occupation. The second section examines whether social media has an influential impact on positive factors (learning new knowledge, increased productivity, and positive impact on task performance). All of the questions from both sections used a 5-point Likert scale of (1) strongly disagree; (2) disagree; (3) neutral; (4) agree; (5) strongly agree.
With the application of SmartPLS (Partial Least Square-PLS-SEM) version 3 as the primary software to analyze descriptive data and statistics. As Reinartz et al., (2009) suggested how PLS can be used and applied for research purposes if the sample size is relatively small, this research is line since our sample is low (n = 100). The descriptive analysis from Table 2 shows the calculation of the mean and standard deviation.

Indicator Loadings
Indicators with outer loadings ranging from 0.4 to 0.7 may also be retained. While dropping the indicators, check to see if the AVE and composite reliability increase after each removal. If neither of the two increases, keep the indicator in this range. Items below must be dropped 0.4 (Hair et al, 2011). As shown in Table 3, all factor loadings values are greater than the minimum of 0.71, with many exceeding the acceptable level of 0.9. Griethuijsen et al. (2014) reported a cross-national study on student interests in science in which several of the Cronbach's alpha values calculated are below the acceptable values of 0.7 or 0.6 (Taber, 2017). Typically, the minimum threshold for Average Variance Extracted is 0.50. AVE less than 0.50 is sometimes acceptable. A CR of 0.70 or higher is acceptable (Fornell and Larcker, 1981). All the AVE values in this analysis are higher than the acceptable value of 0.5. Table 3 shows that the remaining constructs are reliable and valid based on the values of indicator loadings, Cronbach's alpha, and AVE.

Reliability and Validity Tests
Following the data collection stage, an initial round of confirmatory factor analysis (CFA) in SmartPLS 3 was performed to assess the data's validity and reliability and to identify potentially problematic items. Cicchetti (1994) defined fair as 0.4 to 0.59, good as 0.60 to 0.74, and excellent as greater than 0.75 (Matheson, 2019).

Multicollinearity Test and Model Summary (R 2 )
Variance inflation factors (VIFs) and tolerance indices (TIs) are two relevant and frequently used quantities that can be used to investigate individual predictors for potentially strong contributions to (near) multicollinearity (e.g., Wooldridge, 2015). In addition, we calculated the R-Square for dependent variables to assess the levels of variety of the independent variables to the dependent variables, as shown in Table 4. An R-square of 0 to 0.09 (or 0% to 9%) is too low for an empirical model in social science research. This R-square range is unacceptable. It must be rejected. A R-square of 0.10 to 0.50 (or 10% to 50% when expressed in percentage) is acceptable in social science research only when some or all of the explanatory variables are statistically significant. In social science research, an R-square of 0.50 to 0.99 is acceptable, especially when the majority of the explanatory variables are statistically significant.

Hypothesis Testing
The hypothesis was tested using multiple regression analysis in Smart-PLS (Partial Least Square-PLS-SEM) version 3 and using bootstrapping analysis with a 97.5% significance level to obtain the beta, t-statistics, and p-values for each hypothesis. Table 5 shows the outcome of the hypothesis testing. The first hypothesis indicates that the path coefficient score of the social media impact to learn new knowledge has a positive effect (beta = 0.207, t = 2.023, p = 0.044), indicating that social media impact has a significant positive influence on learn new knowledge. The second hypothesis shows that the path coefficient score of the social media impact to more productivity has a negative effect (beta = 0.083, t = 0.982, p = 0.326), indicating that online learning has a negative significant influence on more productivity. Third, the path coefficient score of learn new knowledge to positive impact on task performance has a positive effect (beta = 0.444, t = 4.012, p = 0.000), indicating that learn new knowledge has a positive significant influence on positive impact on task performance. The fourth hypothesis shows that the path coefficient score of more productivity to positive impact on task performance has a positive effect (beta = 0.690, t = 9.618, p = 0.000), indicating that more productivity has a positive significant influence on positive impact on task performance. The last hypothesis shows that the path coefficient score of social media impact to positive impact on task performance has a positive effect (beta = 0.543, t = 9.025, p = 0.000), indicating that social media impact has a positive significant influence on positive impact on task performance. The findings of this study, with all significant results supported by data analysis, suggest that students with social media impact can achieve benefits for learning outcomes. The current study intends to integrate five variables-social media impact and learn new knowledge, social media impact and more productivity, learn new knowledge and task performance, more productivity and task performance, and social media impact and task performance-as well as examine these hypothesized relationships. By conducting the quantitative analysis using PLS-SEM, the objective is to examine whether social media can have a positive influence or impact on young adults, who are active users of the social media platforms for personal or work matters. To confirm our hypothesized relationships between our five variables, all our hypotheses are accepted, implying that social media use has positive both direct and indirect influence. The results of our findings that are supported by data analysis would then reveal how social media impact leads to positive factors, including attaining new knowledge, increased productivity, and positive impact on task performance. In other words, the negative impacts of social media use are outweighed by its positive advantages that are beneficial.

CONCLUSION
The findings of this research paper is in line with previous research since the result of hypothesis 1 is validated which affirms how social media impact positively influences users to learn new knowledge (Mauroner, 2016). The result of hypothesis 2 is aligned with Ganga's (2017) study which states how social media impact users to be more productive. Based on hypothesis 3, Yang et al. (2018) stated that self-regulated learning improved the relationship between prior knowledge and learning. For the hypothesis 4, help-seeking behaviors are selfregulated learning strategies that determine how the quality of help influences learning engagement (Ryan et al., 2005). The last result of hypothesis 5 shows a significant impact since it states that social media impact has a direct relationship on task performance as it was substantiated by previous research by Ndung 'u et al. (2021). This study discovered how communication from social media use can positively influence the learning of new knowledge, increased productivity, and task performance of internet users despite the amount of social media consumption.