Making the Good Better? Investigating the Long-Term Associations Between Capitalization on Social Media and Adolescents’ Life Satisfaction

ABSTRACT

Initial evidence suggests that sharing positive news with others on social media may have short-term emotional benefits for young people. However, little is currently known about whether these capitalization processes have long-term implications for adolescent well-being. Drawing upon the broaden-and-build theory of positive psychology, this study examined the longitudinal reciprocal relationship between capitalization on social media and adolescent life satisfaction. The moderating role of trait optimism was also determined. Three-wave panel data were collected from Belgian adolescents between January and October 2020 (N = 1,125; M age = 15.38, SD = 1.44; Girls = 48.4%), and data were analyzed using a random intercept cross-lagged panel model. No significant cross-lagged relationships were found between capitalization on social media and adolescent life satisfaction at the within-person level. Moreover, the relationships were not moderated by trait optimism. Null results suggest that the antecedents and effects of capitalization on social media may be momentary in nature rather than manifested in long-term associations. Findings therefore add to the limited literature regarding social media’s positive role in adolescent well-being.

KEYWORDS:

• Capitalization
• life satisfaction
• well-being
• social media
• adolescence

Introduction

When young people experience a positive life event, they often discuss their good news with others to help mark, share, and celebrate the event (Demir et al., 2016). This process is known as capitalization (Langston, 1994), and capitalizing on positive life events can have important implications for subjective well-being, including increased positive affect and life satisfaction (Gable et al., 2004). Previous scholarship has predominately examined the effects of capitalization in offline contexts. Yet, as adolescents often share positive content relating to their personal successes on social media (Schreurs & Vandenbosch, 2021a), capitalization processes are also likely to occur online. Initial evidence suggests that sharing positive events on social media can inform the well-being of young people (Bierstetel, 2021; Choi & Toma, 2014; Vermeulen et al., 2018). However, much of the existing time-sensitive research has relied on daily diary methodology, and thus, less is known about the long-term consequences of capitalization on social media. This long-term perspective is significant, as it is possible that sharing positive news with others on social media can produce emotional effects long after the sharing has taken place (Choi & Toma, 2014).

Furthermore, the existing literature regarding capitalization processes has primarily focused on their consequences, rather than their antecedents. Initial research does, however, suggest that subjective well-being can positively predict sharing positive news in offline contexts (e.g., Hershenberg et al., 2014). As dispositional factors can also guide what individuals disclose online (Luo & Hancock, 2020; P. M. Valkenburg & Peter, 2013), a positive reciprocal relationship between adolescent life satisfaction and capitalization on social media may exist. Whilst the broaden-and-build theory of positive emotions (Fredrickson, 2001) supports this reasoning, research regarding the bi-directional relationship between life satisfaction and capitalization on social media is currently lacking.

To address the limitations within the current knowledge base, this study examined the reciprocal relationship between adolescent life satisfaction and capitalization on social media using three-wave panel data. However, as the bi-directional association is unlikely to be the same for all young people (P. Valkenburg, 2017), we also explored whether individual differences moderated the reciprocal relationship. Specifically, as optimists expect more positive outcomes (Scheier & Carver, 1985) and have a greater tendency to benefit from positive information and life events (Scheier et al., 1994), we determined the moderator effect of trait optimism. We utilized a random intercept cross-lagged panel design to distinguish within- and between-person effects (Hamaker et al., 2015), thus enabling us to provide a more nuanced understanding of the longitudinal relationship between life satisfaction during adolescence and capitalization on social media.

By drawing upon theories of positive psychology (Fredrickson, 2001), this research provides novel insight into the reciprocal relationship between social media use and adolescent well-being. Indeed, as adolescence is a period of heightened vulnerability to mental health problems (Blakemore, 2019), much of the existing literature has explored how social media use can inform negative aspects of well-being during adolescence (e.g., depressive symptoms and anxiety) (Schønning et al., 2020). Less is therefore known about social media’s positive role in adolescent well-being. Despite this, emotional regulation capacities develop considerably during adolescence (Young et al., 2019), and initial evidence suggests that young people are increasingly turning to social media to upregulate their positive emotions (Blumberg et al., 2016). Further research is therefore required to determine the extent to which positive regulatory processes on social media, such as capitalization, relate with positive adolescent well-being. This study helps to address this gap in the literature, and sheds light on the extent to which social media represents a new context which can promote adaptive behavior and – potentially - an upward spiral of positive emotions.

Capitalization, Social Media, and Adolescent Life Satisfaction

Whilst adolescence is often associated with an increase in stressful life events (Larson & Ham, 1993), positive life events such as satisfying social interactions, enjoyable activities, and successful performances are also commonplace during the teenage years (Keyfitz, 2016). As adolescents are particularly sensitive to positive experiences and possible rewards, how they respond to positive events can have important consequences for subjective well-being (Chun et al., 2022; Gentzler et al., 2013; Nelis et al., 2018). Several adaptive (and maladaptive) responses have been identified in the literature (Gentzler et al., 2013). To help regulate emotions following positive life events, young people often turn to relational partners (e.g., peers and parents) to mark, share, and celebrate their good news (Demir et al., 2016). Interestingly, the social sharing of positive events can provide emotional benefits above and beyond those triggered by the positive event itself (Langston, 1994). Indeed, studies have found that on days individuals discuss positive life events with others, they report increased positive affect and life satisfaction (Gable et al., 2004). Furthermore, longitudinal survey research suggests that the tendency to share positive life events with others can predict increased well-being several weeks (Lambert et al., 2013) and even months later (Donato et al., 2014).

There are various possible explanations for why individuals can capitalize on positive events by sharing their good news with others. For instance, sharing positive events can help individuals relive the experience and even enhance the memorability and salience of the event (Gable et al., 2004; Langston, 1994). Capitalization effects are also dependent, in part, upon the response of the communication partner with whom the events are shared. As the targets of capitalization are typically close relational partners (Gable et al., 2006), supportive and enthusiastic feedback is commonplace, and positive feedback can amplify the meaning of the event and improve social relationships (Reis et al., 2010). When capitalization attempts are successful, then, they can contribute to a cycle of “broadening and building” (Fredrickson, 2001), whereby sharing good news with others can foster additional positive affect and the development of personal and social resources (Gable & Reis, 2010). It should be noted, however, that passive-destructive responses from communication partners can rebuff attempts to generate pride and dampen positive feelings about the event, thus negatively affecting well-being (Gable et al., 2004).

Previous research regarding capitalization has primarily studied behavior in offline settings. However, over the past decade, social media have become increasingly important contexts for adolescent peer communication, social support, and mood management (Nesi et al., 2018). It is therefore reasonable to assume that many young people are turning to social media to help share their good news with others and benefit from positive capitalization effects (Choi & Toma, 2014). Significantly, the sharing of positive content is commonplace on social media, to the extent that a positivity bias exists (Reinecke & Trepte, 2014). Supported by the affordance of asynchronous communication and content editing, social media users often strategically curate their content to emphasize their positive emotions and personal successes (Schreurs & Vandenbosch, 2021b; Spottswood & Hancock, 2016). Consistent with these norms, adolescents frequently post positive content and express their positive emotions online (Schreurs & Vandenbosch, 2021a; Waterloo et al., 2018), and research suggests that young people share around a third of all positive life events on social media (Choi & Toma, 2022).

Whilst capitalization processes are therefore likely to occur online, the particularities of the social media environment may serve to amplify the emotional effects associated with sharing positive news. P. Valkenburg’s (2017) model of social media self-effects, for instance, outlines how sharing social media content not only potentiates the internalization of optimized self-presentations, but also enables users to receive feedback from wider audiences, both of which may have implications for capitalization effects. Though empirical research regarding capitalization processes on social media is limited, initial evidence does indeed demonstrate the positive short-term emotional effects that sharing positive life events on social media can have for young people. In their week-long daily diary study, Choi and Toma (2014) found that young people report increased positive affect on days they share positive events on social media. Furthermore, cross-sectional and experimental research has identified that these positive effects are most likely when other social media users are responsive to young peoples’ capitalization attempts (Bierstetel, 2021). Qualitative research has reported similar findings, in that adolescents have discussed how receiving positive feedback on social media enabled them to capitalize on their positive emotions (Vermeulen et al., 2018). Importantly, then, as social media are often designed to stimulate positive feedback (e.g., “Like” buttons but no “Dislike” buttons) (Koutamanis et al., 2015), adolescents tend to respond promptly and positively to content shared by peers (P. M. Valkenburg et al., 2006; Winstone et al., 2021), and favorable feedback is particularly commonplace when young people share positive content on social media (Metzler & Scheithauer, 2017; Yang & Brown, 2016). Therefore, given that positive feedback elicits more positive capitalization effects, it is likely that adolescents’ capitalization attempts on social media typically have positive implications for life satisfaction.

Yet, as most time-sensitive studies regarding capitalization on social media have drawn upon daily diary methodology (e.g., Choi & Toma, 2014), less is known about the long-term consequences of such behavior. Initial research in offline contexts has, however, demonstrated that capitalization processes can have positive implications for life satisfaction several weeks or even months later, thus suggesting that effects may not be fleeting (Donato et al., 2014; Lambert et al., 2013). Furthermore, there is reason to believe that these enduring or accumulative effects may be particularly pronounced on social media due to message persistence, in that adolescents can return to their shared content and reminisce about the positive event long after it has occurred (Choi & Toma, 2014). Indeed, whilst sharing ephemeral social media content is increasingly commonplace (Israfilzade & Babayev, 2020; Zhang et al., 2021), young people still report to browse their own social media pages, and research has evidenced that viewing one’s own profile can enhance self-esteem, support self-affirmation, and remind individuals of their good qualities and positive relationships (Gonzales & Hancock, 2011; Toma & Hancock, 2013; Vogel & Rose, 2016). With this in mind, we predict that capitalization attempts on social media will have positive long-term implications for adolescent life satisfaction, and therefore propose the following hypothesis:

H1:

Adolescents’ higher capitalization on social media will predict higher life satisfaction five months later.

Reciprocal Relationship

Extensive research has explored the emotional consequences of capitalization, though fewer studies have considered the factors that inform adolescents’ tendency to engage in capitalization processes. Initial evidence does, however, suggest that individual differences can predict capitalization attempts in offline contexts (Palmer et al., 2016). When considering the predictive role of subjective well-being, research has found that depressive symptoms negatively associate with the sharing of good news (Hershenberg et al., 2014), whilst those with low life satisfaction (Chadwick, 2012) and high depressive symptoms (Feldman et al., 2008; Gentzler et al., 2016; Werner-Seidler et al., 2013) have a greater tendency to dampen – rather than amplify – their positive emotions. Cumulatively, these results support the notion that happier individuals react to events in more positive and adaptive ways than unhappy individuals (Lyubomirsky, 2001). These findings are therefore consistent with positive psychology’s broaden-and-build theory, which proposes that because positive emotions support the building of personal and social resources, they enhance individuals’ ability to manage their emotions in the future (Fredrickson, 2001). As such, positive emotions can trigger an upward spiral toward increased positive well-being (Fredrickson & Branigan, 2005; Fredrickson et al., 2004), whereby positive emotions support the building of personal and social resources, and personal and social resources predict increases in positive emotions (Fredrickson, 2013). In the context of capitalization processes, then, it is possible that sharing positive news with others can contribute to an upward spiral of positive emotions (Gable & Reis, 2010), and a positive reciprocal relationship between capitalization and subjective well-being may exist.

To the best of our knowledge, research is yet to determine the antecedents of capitalization on social media. Research has, however, explored the relationship between subjective well-being and the positivity of content shared on social media. For instance, studies have found that Facebook users with low self-esteem tend to express more negativity (e.g., sadness, frustration, and fear) and less positivity (e.g., happiness, excitement, and gratitude) in their status updates (Forest & Wood, 2012; Landauer, 2014). Concurrent negative correlations have also been found between depressive symptoms and the positivity of social media interactions (Davila et al., 2012). Whilst these studies considered post valance rather than sharing positive events per se, their findings align with previous research in offline contexts, in that those who experience more positive emotions report sharing more positive content. These results are therefore consistent with the social enhancement hypothesis (Kraut et al., 2002), which suggests that those with greater offline resources benefit from social media because online platforms provide additional channels to leverage and extend their existing assets. As such, guided by the extant empirical evidence and underpinned by broaden-and-build theory (Fredrickson, 2001), it is reasonable to assume that adolescents who are more satisfied with their lives are more likely to share good news on social media as a means of capitalizing on their positive emotions. We therefore predict that there will be a positive reciprocal relationship between adolescent life satisfaction and capitalization on social media, and propose the following hypothesis:

H2:

Adolescents’ higher life satisfaction will predict a higher capitalization on social media five months later.

Optimism

Although a positive reciprocal relationship between adolescent life satisfaction and capitalization on social media may exist, the bi-directional association is unlikely to be the same for all young people. The social media literature proposes that such relationships may vary due to intraindividual dispositions (e.g., P. Valkenburg, 2017). As such, we also explored how individual differences may moderate the reciprocal relationship, with a specific focus on the role of trait optimism. Optimism, a core concept within positive psychology, refers to the relatively stable tendency to have positive expectations regarding the future (Carver et al., 2010; Scheier et al., 1994). Expectations regarding the future can have important implications for adolescent mental health, and research has consistently found that optimism associates with better subjective well-being, more positive emotions, and fewer psychological problems for young people (Kennes et al., 2021; Murberg, 2012; Puskar et al., 1999).

Whilst several underlying mechanisms can help to explain the relationship between optimism and well-being (Kennes et al., 2021), researchers often point to the fact that optimists exhibit superior adjustment to stressful life events. Indeed, optimists are more inclined to believe that they can overcome negative events, and therefore have a greater tendency to draw upon proactive coping strategies which seek to eliminate, reduce, or manage stressors (Nes & Segerstrom, 2006). For instance, research regarding optimism and media use has found that when young people experience high levels of stress, those scoring higher in trait optimism exhibit more adaptive media-based coping strategies (Eden et al., 2020). It should also be noted that optimism associates with more positive and supportive peer relationships, and this social support can play an instrumental role in promoting adjustment in challenging situations (Brissette et al., 2002).

Whilst several studies have examined how optimists adjust to challenging life events, less research has explored whether optimism can inform the management of positive events. However, as expectations can influence how individuals approach both stressors and opportunities (Carver et al., 2010), optimism may also have implications for how adolescents negotiate positive events. Initial evidence supports this reasoning, and optimism has been found to positively correlate with savoring positive events (F. B. Bryant, 2003; F. Bryant & Veroff, 2007) and up-regulating positive emotions (Livingstone & Srivastava, 2012). Savoring positive events can also mediate the relationship between optimism and positive well-being (Jose et al., 2018), thus suggesting that optimists have a greater tendency to proactively engage in behavioral strategies that enhance and extend periods of positive mood following positive events. In this sense, it appears that individuals with more positive expectations for the future are more likely to take adaptive steps to achieve positive outcomes following both negative and positive events.

Despite this, and to the best of our knowledge, no research has directly examined whether trait optimism moderates capitalization relationships. Nevertheless, as optimists typically have greater social competence (Lemola et al., 2010) and better social relationships (Carver et al., 2010), they may be more likely to utilize and benefit from these resources through capitalization processes. Optimism may therefore have a contributory effect (Holbert & Park, 2020) on the reciprocal relationship between adolescent life satisfaction and capitalization on social media. More specifically, it is possible that adolescents armed with both positive emotions (i.e., high life satisfaction) and positive expectations (i.e., high optimism) have the greatest tendency to share good news with others on social media. Furthermore, as optimists have particularly supportive relationships (Carver et al., 2010) and interpret feedback more positively (Norem, 2001), they may be more likely to experience positive capitalization effects. Guided by this reasoning, we predict that the more optimistic adolescents are, the more likely it is that capitalization on social media will contribute to an upward spiral of positive emotions. We therefore propose the following hypothesis:

H3:

The reciprocal positive relationship between adolescents’ life satisfaction and capitalization on social media is stronger amongst adolescents with high (vs. low) levels of optimism.

Control Variables

To avoid potential omission bias, it is necessary to account for the confounding effects of other variables on adolescents’ life satisfaction and capitalization on social media. Research has consistently evidenced that both social media use and life satisfaction are determined by adolescents’ age (e.g., OFCOM, 2022; Orben, Lucas, et al., 2022), sex (e.g., Danielsen et al., 2010; Twenge & Martin, 2020), educational experiences (e.g., Crede et al., 2015; Salmela-Aro et al., 2017), and socio-economic status (Bannink et al., 2016; Skogen et al., 2022). Moreover, adolescents’ life satisfaction appears to be closely linked to general social media use (Orben, Przybylski, et al., 2022), and it is possible that young people who use social media more frequently are more likely to capitalize on the platforms. Therefore, age, sex, educational level, socio-economic status, and general social media use were included as control variables in our hypothesized model.

Method

Open Science

This research was part of the “Positive Body and Sex Project,” and our anonymized study data, pre-registration, syntaxes, and outputs are freely available on https://osf.io/ru8vn/?view_only=9989b47ba8b24c0189e8b429743cf1c4. More information regarding this project, together with other published studies that are part of this project, can be found on https://osf.io/c5g6p/?view_only=7f1a0125b9b34875a55c50f8f0d9c77d.

Sample and Procedure

Longitudinal survey data were collected using a three-wave panel design with five-month intervals between waves. Whilst social media use is part of the daily lives of many adolescents (Pouwels et al., 2021), adolescent life satisfaction is relatively stable (Macek et al., 2022). It would therefore be inappropriate to have short intervals between waves (Johannes et al., 2022), as the impact of capitalization on life satisfaction may be too small to observe in the short-term. However, long-term changes in life satisfaction can be observed during the adolescent years (Orben, Lucas, et al., 2022), and it is possible that communication activities (e.g., capitalization on social media) exert their influence on life satisfaction over several months (Dienlin et al., 2017). As such, five months was considered a sufficient interval to assess changes in capitalization processes and adolescent life satisfaction.

The study was conducted among Flemish adolescents between January and October 2020. The second author contacted 100 randomly selected Flemish schools via e-mail and/or phone and invited them to participate in the study. Of these 100 schools, 16 agreed to participate. Active parental consent was sought from parents whose adolescents were younger than 16, and passive parental consent was obtained from parents whose adolescents were older than 16. Respondents were offered a monetary incentive for their participation, and they received gift vouchers worth €5 following wave two, and €7 following wave three. The study was approved by the ethics committee of [BLINDED].

During the first wave (W1), 1,313 participants completed a paper-and-pencil survey during class hours (M age = 15.20, SD = 1.65; Girls = 47.1%). 1,125 respondents aged 12–18 years reported scores in all variables of interest and were therefore retained (M age = 15.38, SD = 1.44; Girls = 48.4%). Attrition bias was assessed using a multivariate logistic regression, with two groups of participants as dependent variables (0 = excluded from the study at W1, 1 = included in the study at W1), and age, sex, socio-economic status, and educational level as independent variables. Whilst younger adolescents were more likely to be excluded at W1 (exp(B) = .56, B = −.59, p < .001), no other differences emerged at the p < .05 level. In the second wave (W2), 534 respondents completed an online survey (M age = 15.17, SD = 1.33; Girls = 66.5%), and there was an attrition rate of 52.53%. The change in data collection methods due to the COVID-19 physical distancing measures that were operative in June may partially explain this drop-out rate. In the third wave (W3), 421 respondents completed the online survey (M age = 15.27, SD = 1.25; Girls = 67.5%), and there was an attrition rate of 21.17%.

Following the procedures of Erreygers et al. (2018), data were used from respondents who participated in at least one of wave of data collection. The final analytic sample therefore consisted of 1,125 respondents (M age = 15.38, SD = 1.44; Girls = 48.4%). Following the secondary school system in Belgium, 46.8% of respondents were in the first education level (i.e., where they were being prepared for higher education), 48.7% were in the middle education level (i.e., where they were being taught primarily technical skills), and 13.3% were in the third education level (i.e., leading to specific vocational careers, such as a baker).

Measures

General social media use

At each wave of data collection, participants reported how much time they spent on Facebook, Instagram, YouTube, Snapchat, and TikTok on a daily basis over the past five months (1 = Never, 2 = Less than 10 minutes per day, 3 = 10–30 minutes per day, 4 = 30–60 minutes per day, 5 = 1–2 hours per day, 6 = 2–3 hours per day, 7 = More than 3 hours per day). Mean scores were used and higher scores represented higher general social media use (W1: M = 3.70, SD = 1.11; W2: M = 4.06, SD = 1.16; W3: M = 4.00, SD = 1.15). The measure was stable across waves (r = .68–.74, p < .001), and it was used as a control variable in this study.

Capitalization on social media

Guided by the literature regarding capitalization processes during adolescence (e.g., Gentzler et al., 2013, 2014), we developed a four-item scale to explore capitalization on social media. Respondents were invited to think about positive events (e.g., a birthday, doing fun things with friends) that they shared with others online (e.g., via social media or instant messaging apps) during the past five months. They were then asked to rate the extent to which the following statements captured why they shared their positive news with others (1 = Completely disagree, 2 = Strongly disagree, 3 = Disagree, 4 = Neutral, 5 = Agree, 6 = Strongly agree, 7 = Completely agree): “ … to make the event more pleasant,” “ … to better remember the event,” “ … to remember how good I felt during this event,” “ … to enjoy the event even more.” A confirmatory factor analysis with maximum likelihood estimation supported the single factor structure of the items (χ² (51) = 414.28, p < .001, RMSEA = .08, CFI = .93, TLI = .91) (see OSF for results). The measure had moderate stability across the three waves (r = .40–.54, p < .001), and reported acceptable Cronbach’s alpha scores (W1: α = .90, W2: α = .90, W3: α = .90). This variable was treated as a latent variable, though for descriptive purposes, mean scores were generated, with higher scores representing higher capitalization motive (W1: M = 4.30, SD = 1.47; W2: M = 4.32, SD = 1.47; W3: M = 4.20, SD = 1.57).

Life satisfaction

The single item life satisfaction measure was used (Cheung & Lucas, 2014). More specifically, respondents were asked to indicate the degree to which they were satisfied with their lives on a seven-point Likert scale (1 = Completely unsatisfied, 2 = Strongly unsatisfied, 3 = Unsatisfied, 4 = Neutral, 5 = Satisfied, 6 = Strongly satisfied, 7 = Completely satisfied). The measure had a moderate stability across the three waves (r = .45–.61, p < .001), with higher scores representing higher life satisfaction (W1: M = 5.38, SD = 1.46; W2: M = 5.49, SD = 1.21; W3: M = 5.42, SD = 1.19).

Optimism

The optimism subscale of the Youth Life Orientation Test (Y-LOT; Ey et al., 2005) was used during W1. Respondents were asked to rate the extent to which they agreed with six items on a seven-point Likert scale (1 = Disagree completely, 2 = Strongly disagree, 3 = Disagree, 4 = Neutral, 5 = Agree, 6 = Strongly agree, 7 = Completely agree). Example items included “I usually expect to have a good day” and “I am a lucky person.” A confirmatory factor analysis with maximum likelihood estimation supported the one factor structure of the items (χ² (14) = 195.08, p < .001, RMSEA = .11, CFI = .92, TLI = .88) (see OSF for results), and an acceptable Cronbach’s alpha score was reported (α = .80). Mean scores were used, with higher scores representing a higher optimism (M = 4.44, SD = .88). For H3, respondents were categorized into “low optimism” (50.1%) and “high optimism” (48%) according to their optimism scores, with a cutoff at the 50^th percentile. 1.9% of responses were missing on the optimism scores.

Analytic Strategy

Descriptive statistics and bivariate correlations were calculated using SPSS 26.0. Intraclass Correlation Coefficients (ICCs) were then computed for focal variables. A high ICC (>.50) indicates proportionately higher between-person variance, whilst a low ICC (<.50) indicates proportionately higher within-person variance (Keijsers, 2016). The reciprocal relations between capitalization on social media and adolescent life satisfaction were tested using a multiple indicator random intercept cross-lagged panel model (RI-CLPM) in Mplus 8.7. Control variables age, sex, educational level, socioeconomic status, and general social media use were included in the model. To include all available observations and to avoid generalizability issues that could result from only using participants with complete data, full information maximum likelihood estimation (FIML) was used. The use of FIML to handle missing data is widely recommended, especially when the amount of missing data is moderate to large (Widaman, 2006).

To test the moderating effect of trait optimism, a multi-group analysis was conducted. Specifically, a model in which the cross-lagged and autoregressive paths were allowed to differ between adolescents with low and high optimism was compared to a model in which these paths were constrained to be equal. Should the chi-square difference test concerning these models be significant, subsequent tests would be conducted the determine the differences between parameters. Model fit was evaluated using three goodness-of-fit-indices: the root mean square error of approximation (RMSEA), the Comparative Fit Index (CFI), and the Tucker-Lewis Index (TLI) (Hu & Bentler, 1999). Model fit was considered acceptable when CFI and TLI scores were between .90 and .95, and the RMSEA value was between .05 and .08. Moreover, CFI and TLI scores larger than .95, and an RMSEA value smaller than .05, indicated good model fit (Kline, 2015).

Results

Zero-order correlations between the study variables can be found in Table 1. The ICCs were .45 for capitalization and .53 for life satisfaction. These results indicate that a substantial amount of variance in the subsequent measurements of the same variable were explained by within-person differences.

Table 1. Zero-order correlations for study variables.

	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Life satisfaction W1	1	.53***	.45***	−.01	−.06	−.05	.46***	−.11***	−.12***	−.06	.22***	−.07*	−.07	−.03
2. Life satisfaction W2		1	.61***	−.02	.03	−.10	.46***	−.10**	−.08	−.08	.20***	−.03	−.04	−.08
3. Life satisfaction W3			1	−.08	−.04	−.12*	.37***	−.01	−.10*	−.04	.26***	−.08	−.09	−.07
4. Capitalization W1				1	.47***	.40***	.06	.02	.13***	.07*	.01	.30***	.27***	.26***
5. Capitalization W2					1	.54***	.02	.15***	.17***	.10*	.03	.32***	.27***	.29***
6. Capitalization W3						1	.00	.12*	.19***	.16***	.02	.28***	.34***	.24***
7. Optimism (baseline)							1	−.11***	−.13***	−.06	.24***	−.04	−.07	−.02
8. Age (baseline)								1	−.02	.31***	−.12***	.13***	.15***	.01
9. Sex (baseline)									1	−.14***	−.03	.18***	.14**	.14*
10. Educational level (baseline)										1	−.07*	.16***	.14**	.19***
11. Socioeconomic status (baseline)											1	.05	.03	.05
12. General social media use W1												1	.70***	.68***
13. General social media use W2													1	.74***
14. General social media use W3														1

Note. *p < .05, **p < .01, ***p < .001.

RI-CLPM

The hypothesized RI-CLPM reported a poor model fit (χ² (166) = 503.46, p < .001, RMSEA = .08, CFI = .89, TLI = .86). Thus, we consulted the modification indices, and in instances where correlations were theoretically meaningful, we allowed error terms to covary. We correlated the errors of item 1 (“to make the event more pleasant”) and item 4 (“to enjoy the event even more”) of the capitalization measurement at W2 and W3. Following these modifications, the model reported an acceptable fit (χ2 (164) = 395.17, p < .001, RMSEA = .06, CFI = .92, TLI = .90).¹ Parameter estimates for each cross-lagged and autoregressive path can be found in Table 2. No between-person relations between capitalization on social media and adolescent life satisfaction emerged (β = −.24, p = .06). Furthermore, the only significant path at the within-person level was an autoregressive relation between life satisfaction at W1 and W2 (β = .21, p = .04). Thus, H1 and H2 were not supported.

Table 2. Coefficients for the hypothesized RI-CLPM model.

	Unstandardized coefficients (B)	Standardized coefficients (β)	Confidence interval (of β)	S.E.
Within-personal level
Life satisfaction W1 ➔ Life satisfaction W2	.15	.21*	[.05; .37]	.10
Capitalization W1 ➔ Life satisfaction W2	.13	.13	[−.04;.29]	.10
Life satisfaction W2 ➔ Life satisfaction W3	.12	.13	[−.17; .43]	.18
Capitalization W2 ➔ Life satisfaction W3	.11	.12	[−.12; .36]	.15
Capitalization W1 ➔ Capitalization W2	.12	.13	[−.14; .40]	.16
Life Satisfaction W1 ➔ Capitalization W2	−.02	−.03	[−.21; .15]	.11
Capitalization W2 ➔ Capitalization W3	.27	.22	[.01; .44]	.13
Life Satisfaction W2 ➔ Capitalization W3	−.03	−.03	[−.21; .15]	.11
Between-person level
Life Satisfaction WITH Capitalization	−.14	−.24	[−.46; −.03]	.13

Note. *p < .05; **p < .01; ***p < .001.

To test H3, a chi-square difference test determined whether model fit differed between a model in which the paths were allowed to vary between adolescents with low and high optimism, and a model in which the paths were constrained to be equal. The chi-square difference test was not significant (Δχ² (38) = 50.12, p = .09), indicating no significant differences in the hypothesized model between adolescents with low and high optimism. H3 was therefore not supported.

Discussion

This study sought to explore the longitudinal relationship between capitalization on social media and adolescent life satisfaction. Guided by the broaden-and-build theory of positive psychology (Fredrickson, 2001), we predicted that there would be a positive reciprocal relationship at the within-person level, whereby sharing positive news with others on social media would contribute to an upward spiral of positive emotions. Furthermore, we predicted that the bi-directional association between capitalization on social media and life satisfaction would be particularly strong for adolescents high in optimism. Despite this, we found no support for our hypotheses. Below, we discuss the results in more detail, and provide possible explanations for the null findings.

Cross-sectional, experimental, and daily diary studies have previously identified positive relationships between capitalization on social media and subjective well-being (Bierstetel, 2021; Choi & Toma, 2014; Vermeulen et al., 2018). However, less is known about the long-term emotional effects of sharing positive news with others on social media. Whilst research in offline contexts suggests that capitalization may produce positive effects weeks or even months later (Donato et al., 2014; Lambert et al., 2013), in this study, adolescents’ capitalization on social media did not significantly predict their life satisfaction five months later (H1). Similarly, although subjective well-being has been found to associate with sharing positive news in offline environments (Chadwick, 2012; Hershenberg et al., 2014), adolescents’ life satisfaction did not predict their capitalization on social media five months later (H2). In this sense, we found no evidence to support the hypothesized reciprocal long-term relationship between capitalization on social media and adolescent life satisfaction.

These null results may, in part, be resultant of the time interval between waves of data collection. Whilst long-term changes in life satisfaction are normative during the adolescent years (Orben, Lucas, et al., 2022), it has been argued that momentary ups and downs can have important implications for individuals’ satisfaction with life (Willroth et al., 2020). The interplay between capitalization on social media and adolescents’ life satisfaction may therefore be more transient in nature, and thus unobservable using long-term longitudinal designs. In fact, long-term longitudinal studies concerning capitalization processes are rare (Gable & Reis, 2010), and much of the existing evidence regarding capitalization derives from research utilizing daily diary methodology. It is therefore possible that the hypothesized reciprocal effects are typically short-lived, and thus best explored using daily diaries or short-term longitudinal designs.

Furthermore, the measurement tool used to determine scores for capitalization on social media may also have contributed to the null results. Guided by the existing literature regarding capitalization processes during adolescence, we developed a four-item scale to explore capitalization on social media. The scale measured capitalization motive, whereby adolescents reported on the extent to which they had shared positive news on social media over the past five months for capitalization purposes. The scale reported a single factor structure and good internal consistency, and it can therefore support researchers to explore what motivates young people to share positive content on social media. However, the hypothesized long-term reciprocal effects may be accumulative in nature, and thus, it is possible that frequency of capitalization may have had a stronger relationship with adolescent life satisfaction. To provide further insight into capitalization on social media and adolescent well-being, future research should explore how often young people share positive news on social media, rather than capitalization motive alone.

Finally, as optimists have a greater tendency to benefit from positive information and life events (Scheier et al., 1994), we also predicted that the bi-directional association between capitalization on social media and life satisfaction would differ in accordance with adolescents’ optimism. However, we found no support for this hypothesis (H3). True moderator effects may have been obscured by the aforementioned issues relating to length of interval between waves and measurement of capitalization motive rather than frequency. However, given that we tested the moderating effect of baseline trait optimism (i.e., “pre-pandemic” optimism), the COVID-19 lockdowns at W2 and W3 may also have had implications for this result. Whilst optimism is relatively stable, it can change over time and circumstance (Millstein et al., 2019), and adolescents’ optimism tended to reduce during the pandemic (Soest et al., 2022). This was particularly the case for young people who reported that the pandemic had high negative social consequences (e.g., less contact with friends) (Wieczorek et al., 2022). It is therefore possible that some adolescents fluctuated between relatively high and low levels of optimism across the study, thus further obscuring the true moderator effects. Given that social circumstances may inform media uses and effects (Vranken et al., 2023), it may be wise for researchers to revisit the moderating role of optimism at a point in time where optimism is less likely to fluctuate.

Limitations and Future Research

Whilst this is the first study to explore the long-term reciprocal relationship between capitalization on social media and adolescent life satisfaction, it comes with limitations that ought to be addressed in future research. First, using the powRICLPM package in R, we found that our sample had adequate power to detect between-person effects (power = .92–1), small reciprocal within-person relations (β = .10–.30, power = .96–1), and concurrent correlations (power = .99–1). Yet, when it comes to detecting small cross-lagged effects, power was inadequate (power ≈ .6). The possibility of a type-2 error can therefore not be ruled out, and future research with larger samples could revisit our study hypotheses. Furthermore, as previously outlined, shorter-term longitudinal studies may provide greater insight into the antecedents and effects of capitalization on social media during adolescence. Nevertheless, additional long-term longitudinal research would shed further light on whether capitalization effects are indeed fleeting.

To help provide more nuanced insight into capitalization on social media, future research should measure frequency of capitalization to assess accumulative effects over time. Given that capitalization effects are informed by the response of the communication partner with whom the events are shared, it may also be prudent to examine the moderating effect of peer feedback on social media. Although positive peer feedback is commonplace on social media, positive capitalization effects are likely to be heighted for young people who receive supportive and enthusiastic feedback (i.e., more “Likes” or positive comments) from close ties (P. M. Valkenburg et al., 2017). Researchers should also consider the effects of sharing inauthentic positive content on social media. The positivity bias on social media can prompt adolescents to share inauthentic content, and discrepancies between the actual self and the image young people portray on social media may have negative implications for subjective well-being (Twomey & O’Reilly, 2017). In this sense, authenticity is likely to affect how sharing positive content on social media informs adolescent well-being. Finally, it may be wise for researchers to differentiate between social media platforms, or even between different features within the same platform, to explore capitalization processes. It is, for instance, possible that curated “timeline” posts serve as a reminder of past achievements and positive experiences, whilst ephemeral content – such as Instagram Stories – allow for more immediate and authentic responses to current events. Whilst qualitatively different, it is currently unclear as to whether sharing positive news using a specific platform or feature leads to different capitalization effects.

Future research may also consider utilizing different analytic approaches. In this study, we drew upon the RI-CLPM to help explore the hypothesized relationships at a within-person level. However, recent critiques of the design have proposed that effects captured by the RI-CLPM are less persistent and more-short term than those captured by the traditional cross-lagged panel model (CLPM) (Orth et al., 2021). If slower to develop – but more durable – relations are more observable through the traditional CLPM, there may be some mismatch between our study data and analysis. To address this concern, we also conducted a traditional CLPM, and as with the RI-CLPM, no significant relations emerged (see OSF for full results). Whilst these findings rule out a mismatch between our study data and analysis, future long-term longitudinal research regarding capitalization processes on social media may consider drawing upon the traditional CLPM to avoid these concerns.

Conclusion

This study examined the longitudinal reciprocal relationship between capitalization on social media and adolescent life satisfaction. Null results were found, thus suggesting that the antecedents and effects of capitalization processes on social media may tend be momentary in nature. Further research is required to learn more about the conditions under which capitalization on social media may have positive emotional effects. Overall, this investigation sheds further light on the reciprocal relationship between social media use and adolescent well-being, and therefore adds to the limited literature regarding how social media may positively inform the well-being of young people.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

See “Open Science” section of manuscript for detail on how to access anonymized study data, pre-registration, syntaxes, and outputs on OSF https://osf.io/ru8vn/?view_only=9989b47ba8b24c0189e8b429743cf1c4.

Additional information

Funding

This work was supported by Bijzonder Onderzoeksfonds KU Leuven under grant number C14/18/017.

Notes

1. Two parsimonious models were also tested. In the first parsimonious model, only focal variables were included. In the second parsimonious model, all focal and control variables were included, except for daily social media use. The two parsimonious models reported similar results (first model: OSF, second model: OSF) as the hypothesized model.