Media and Communication

2024 • Volume 12 • Article 8597

https://doi.org/10.17645/mac.8597

ARTICLE Open Access Journal

Exploring Adolescents’ Social Media Connection and

Disconnection: A Latent Class Approach
Lise‐Marie Nassen 1 , Kathrin Karsay 2 , Laura Vandenbosch 1 , and Lara Schreurs 1
1
Media Psychology Lab, KU Leuven, Belgium
2
Department of Communication, University of Vienna, Austria

Correspondence: Lise‐Marie Nassen (lise‐[email protected])

Submitted: 30 April 2024 Accepted: 3 October 2024 Published: 25 November 2024

Issue: This article is part of the issue “Disconnectivity in a Changing Media and Political Landscape” edited
by Qinfeng Zhu (University of Groningen) and Çiğdem Bozdağ (University of Groningen), fully open access at
https://doi.org/10.17645/mac.i456

Abstract
In industrialized societies characterized by ubiquitous connectivity, many individuals disconnect from their
phones or social media to break patterns of habitual use, reduce information overload, alleviate stress, and
avoid distractions. Although research has predominantly focused on (young) adults, information on digital
disconnection among adolescents remains limited. In response, in the present study, we analyzed survey
data from 956 Flemish adolescents in Belgium (𝑀 age = 15.10, 𝑆𝐷 = 1.61, and 64.8% girls) and, using latent
class analysis, identified two subgroups: Class 1 labeled as “low adoption of social media apps and
disconnection tools,” and Class 2 labeled as “high adoption of social media apps and disconnection tools.”
Adolescents in Class 2 were more likely to use social media, specifically social networking, instant messaging,
and entertainment apps, and to adopt disconnection tools (e.g., iOS Screentime or the Forest app). Next,
drawing on the media practice model, we investigated how sociodemographic and sociopsychological
factors contribute to those usage patterns and found that girls, older adolescents, and adolescents with
higher scores for depressive symptoms and flourishing were more likely to belong to Class 2. Those
findings offer new insights into adolescents’ social media connection and disconnection that can inform
interventions to promote healthier smartphone use among adolescents.

Keywords
adolescents; disconnection; latent class analysis; social media use; voluntary disconnection

1. Introduction

Concerning social media use among adolescents, there is a shared view that they constantly use the phone
to seek gratification from social media platforms. Nevertheless, not all adolescents behave that way.

© 2024 by the author(s), licensed under a Creative Commons Attribution 4.0 International License (CC BY). 1
The emerging sense of an inescapable online presence has triggered social media users, both young and old,
to reflect on their use and that of others. Even so, no study has examined such disconnection practices
among adolescents. That is surprising because adolescents’ developmental sensitivities (e.g., heightened
peer focus and identity exploration) make them more susceptible to becoming dependent on and affected
by social media use (Heffer et al., 2019). In turn, teachers, parents, and policymakers have attempted to
restrict adolescents’ social media screen time by, for instance, banning smartphones in high schools
(Anderson et al., 2024). However, adolescents do not readily accept forms of disconnection imposed upon
them (Valkenburg & Piotrowski, 2017).

Part of fostering a healthy relationship with social media platforms lies in providing the right tools to build a
sustainable digital balance and exercise self‐control when using the platforms. Although such digital balance
(Vanden Abeele, 2021) can be obtained by voluntary disconnection, self‐imposed disconnection among
adolescents has rarely been investigated. In particular, adolescents’ adoption of disconnection tools (e.g.,
iOS Screentime or the Forest app) to limit social media use has yet to be mapped.

Against that background, in the present study, we performed latent class analysis (LCA), which allowed us to
identify subgroups of adolescents (i.e., classes) based on their social media use and adoption of disconnection
tools. We also drew on the media practice model (Steele & Brown, 1995), which posits that individuals interpret
and respond to media with reference to their lived experiences. According to this model, sociodemographic
factors (i.e., age, gender, and socioeconomic status [SES]), as well as sociopsychological factors (i.e., depressive
symptoms, loneliness, flourishing, and satisfaction with life) that shape adolescents’ lived experiences, may be
associated with the likelihood of belonging to a specific social media (dis)connection group. Those associations
formed the focus of the current research.

2. Adolescents’ Developmental Tasks and Social Media Use

Research on social media use and its effects has often specifically examined adolescents (Valkenburg et al.,
2022). Adolescence is a developmental period marked by significant social, biological, and cognitive changes
(Nesi et al., 2018) that occur while performing developmental tasks such as navigating peer relationships,
developing self‐identity, and establishing autonomy from adults (Valkenburg & Piotrowski, 2017). For
contemporary adolescents, many of those tasks are partly undertaken on social media (Nesi et al., 2018).
Adolescents receive their first smartphones at increasingly younger ages, with a recent report showing that
the average age has dropped to 8 years among Flemish adolescents (Vanwynsberghe et al., 2022).

Smartphones give access to social media platforms, including TikTok, WhatsApp, and Snapchat, which are
adopted by 78%, 74%, and 66% of children before their 12th birthdays (Vanwynsberghe et al., 2022). In turn,
those platforms have become essential to adolescents’ lives by affording constant opportunities for interacting
with peers and exploring identity (Masur et al., 2022; Nesi et al., 2018). However, due to the ever‐increasing
use of social media, many scholars, parents, caretakers, and policymakers have voiced concerns about social
media’s potential negative effects on mental and physical health given harmful social media interactions and
time displacement effects (Masur et al., 2022). One factor that might help adolescents strike a healthy balance
in their digital media use is restricting their use of social media.

Media and Communication • 2024 • Volume 12 • Article 8597 2

3. Adolescents’ Disconnection From Social Media

Policy reports (Vanwynsberghe et al., 2022) and academic research on adolescents’ social media use
(Schmuck et al., 2023) that investigated the rules parents and schools impose on children highlighted that
smartphones and social media restrictions are part of the upbringing of minors in contemporary society.
Adolescents may need rules for social media use because they are believed to have limited to no self‐control
over it (Meinert & Reinecke, 2018). That reasoning may explain why voluntary disconnection has rarely been
studied and why the literature on the topic mostly focuses on (young) adults. Voluntary digital disconnection
can be defined as:

A deliberate form of non‐use of devices, platforms, features, interactions, and/or messages that occurs
with higher or lower frequencies, and for shorter or longer periods, after the initial adoption of these
technologies, and with the aim of restoring or improving one’s perceived overuse, social interactions,
psychological well‐being, productivity, privacy and/or perceived usefulness. (Nassen et al., 2023, p. 13)

However, there is good reason to believe contemporary adolescents care about digital balance (Jorge et al.,
2023). In their qualitative research, Neves et al. (2015) explored adolescents’ motivations for disconnecting
from social media and identified a sense of uselessness (e.g., perceived waste of time), unwanted social
practices (e.g., online gossip), and unsatisfactory self‐presentation (e.g., reluctance to partake in online
impression management) among the top reasons. Beyond that, van der Wal et al.’s (2024) focus group study
revealed that adolescents do engage in voluntary disconnection from social media as they found that
adolescents try to disconnect from social media practices, including watching TikTok videos, even if they
often fail in their attempts.

Among adults, various strategies for disconnecting from social media have been identified, including outright
quitting, taking periodic breaks, reducing use, switching platforms, and adopting disconnection tools. These
strategies can be applied at different levels, including the level of a device, (branded) application, feature,
interaction, and message (Meier & Reinecke, 2021). Although most research on disconnection has focused
on disconnecting from devices (Nassen et al., 2023), examples of more specific strategies for disconnecting
are, for instance, restricting oneself to one hour on TikTok per day or silencing group messages. Another
increasingly common way to disconnect is by adopting mobile disconnection tools (Nassen & Karsay, 2024;
Nguyen, 2021), which provides a nuanced strategy by allowing disconnection from certain aspects of the
smartphone without obstructing connections that the user wants to maintain. On that count, we distinguish
two types of nuanced strategies for disconnecting from social media use (Nassen & Karsay, 2024). The first
involves using built‐in device settings, including iOS Screentime, to limit use, while the second involves using
external detox apps, for example, the Forest app, which are non‐default apps designed to set limits.

Such nuanced strategies for disconnecting from the mobile phone may be especially appealing to adolescents
rather than invasive ones (e.g., a week without Instagram or a full day without using the phone), which are at
odds in a time of constant connectedness (Nassen et al., 2023). Indeed, young adults (18–24 years old) are
prevalent users of such nuanced disconnection tools (Schmuck, 2020). Adolescents may also fulfill their need
for digital disconnection by using those tools and, in doing so, balance disconnection with their needs for online
peer connections. After all, not all adolescents prefer unlimited social media use. Similar to adult users, they
likely seek a trade‐off between social media connection and disconnection (Rosič et al., 2024). Vanden Abeele

Media and Communication • 2024 • Volume 12 • Article 8597 3

and Nguyen (2024) identified four types of digital well‐being experiences among adults and confirmed the
existence of distinguishable patterns of connection and disconnection. Meanwhile, other studies have even
identified meaningful subgroups of adolescents based on their social media use (Foerster & Röösli, 2017).
Nevertheless, no study thus far has simultaneously considered both adolescents’ use of and disconnection
from social media platforms, much less disconnection involving the adoption of disconnection tools. Therefore,
our first research question (RQ) was:

RQ1: What are the distinct subgroups of adolescents based on their use of social media apps and
adoption of disconnection tools?

4. Individual Differences: Adolescents’ Lived Experiences

Different sociodemographic and sociopsychological factors could play a role in whether someone adopts
disconnection tools. The media practice model (Steele & Brown, 1995) suggests that lived experiences shape
how individuals interpret and react to media. Adolescents are active (social) media users who, based on their
identities and lived experiences, select, interact, and apply media. Following that reasoning, adolescents’
adoption of disconnection tools is likely informed by the factors that shape their lived experiences, including
sociodemographic and sociopsychological ones.

4.1. Sociodemographic Factors

4.1.1. Age

Social media use increases throughout adolescence due to more access to mobile devices and an increased
desire to interact with peers (Coyne et al., 2019). Although it seems evident that social media connection
increases, it is unclear whether adolescents also disconnect more from social media as they mature. In the
early stages of adolescence, smartphone and social media use is primarily defined by rules and restrictions
imposed by parents (Anderson et al., 2024). However, as adolescents grow older, they gain independence in
their usage choices and develop self‐control (Meinert & Reinecke, 2018; Siebers et al., 2021). As such, it is
plausible that the desire to disconnect voluntarily becomes more prevalent. However, that possibility has yet
to be explored.

4.1.2. Gender

Regarding gender, research has shown that girls use social media more often than boys and are more likely
to use it for social purposes, for example, posting pictures (Nesi & Prinstein, 2015). For that reason, girls may
find it more challenging than boys to disconnect from social media. On that topic, studies have indicated that
voluntary disconnection could also be more difficult for adult women due to differences in social expectations
(Van Bruyssel et al., 2023). Care work is often mentioned as an important expectation ascribed to women that
carries over into the online context, where women are expected to display more social etiquette (e.g., sending
messages on every birthday) and engagement (e.g., making sure that the family gets together), which causes
them to have less opportunities to disconnect (Baumer, 2018). Such social expectations may be valid for young
girls as well. Jorge et al. (2023) found that voluntary disconnection seems more difficult for girls than boys
because their gendered position makes them feel more connected to their friends on social media. Even so,

Media and Communication • 2024 • Volume 12 • Article 8597 4

they might also be more willing to disconnect due to a stronger need to spend time with their families and
without their phones.

4.1.3. SES

Contextual factors may also influence adolescents’ social media use and adoption of disconnection tools.
Among adolescents, inequalities exist, for instance, in digital skills, digital literacy, and (the stability of)
internet access (Nguyen & Hargittai, 2024). Thus, recent scholarship on digital inequality has not only
considered differences in access to media and technology but also differences in taking breaks from media
when needed. For instance, during the Covid‐19 pandemic, some adolescents had more options to spend
their free time in terms of spatial resources (e.g., their rooms to retreat to) and material resources (e.g.,
various materials for hobbies available at home). By contrast, others may have had fewer options, with their
only means of self‐expression and engaging in hobbies being via their mobile devices. Because engaging in
practices of disconnection can therefore be a privilege (Treré, 2021), exploring the role of SES in the
adoption of social media connection and digital disconnection tools is important.

Adolescents’ identities can be attributed to forms of socioeconomic inequality and gendered socialization
(Steele & Brown, 1995). Growing older also naturally involves gaining lived experiences. Thus, according to
the media practice model, those sociodemographic characteristics can impact adolescents’ social media use
and adoption of disconnection tools (Steele & Brown, 1995). For that reason, our second RQ was:

RQ2: Do sociodemographic characteristics (i.e., age, gender, and SES) vary across different classes of
adolescents in terms of social media use and the adoption of disconnection tools?

4.2. Sociopsychological Factors

Adolescence is characterized by stage‐salient developmental tasks and continuous fluctuations in well‐being

(Valkenburg et al., 2022). Sociopsychological factors, including levels of perceived loneliness, depressive
symptoms, satisfaction with life, and flourishing, shape and are shaped by young peoples’ lived experiences
throughout adolescence (Steele & Brown, 1995). As such, those sociopsychological factors may also
determine adolescents’ social media use and adoption of disconnection tools.

4.2.1. Loneliness

The relationship between loneliness and social media use in adolescents is multifaceted and complex. Social
media interactions can reduce loneliness among adolescents because meaningful interactions occur on
social media (Yang & Brown, 2013). As such, feeling lonely may prevent adolescents from pursuing voluntary
disconnection due to fear of feeling excluded. Loneliness has also been identified as a negative consequence
that can result from disconnection (Nassen et al., 2023). However, some studies have suggested that limiting
social media use can decrease loneliness (Hunt et al., 2018) by freeing up time for meaningful face‐to‐face
connections (Nguyen, 2023). Loneliness may therefore play a role in adolescents’ social media use and
adoption of disconnection tools but remains underexamined.

Media and Communication • 2024 • Volume 12 • Article 8597 5

4.2.2. Depressive Symptoms

Social media use may also relate to ill‐being indicators such as depressive symptoms. On that count,
adolescents who depend on social media feedback for their self‐worth can feel more depressed (Schreurs
et al., 2024). This can imply that adolescents who feel depressed engage in social media disconnection in an
attempt to alleviate those feelings. Moreover, given the prevalence of negative discourse on social media
use and ill‐being (Valkenburg et al., 2022), adolescents could embrace the same mindset and attribute their
negative feelings to their social media use. However, no research has investigated factors of ill‐being,
including depressive symptoms, and adolescents’ voluntary disconnection from smartphones.

4.2.3. Flourishing and Satisfaction With Life

Adolescents’ well‐being may inform their social media use and adoption of disconnection tools. In the
present study, we opted to investigate two distinct constructs of well‐being—flourishing, and satisfaction
with life—because they are subdimensions of two types of well‐being: hedonic well‐being and eudaimonic
well‐being. On the one hand, hedonic well‐being focuses on the subjective experience of contentment and
pleasure (Huta & Waterman, 2013) and includes, for instance, satisfaction with one’s life (Diener et al., 1985).
On the other, eudaimonic well‐being focuses on another dimension commonly described as “the good life,”
which is characterized by experiences of meaningfulness, authenticity, and self‐actualization (i.e., knowing
one’s true self and expressing it; Martela & Sheldon, 2019). One way to operationalize eudaimonic well‐being
is to measure flourishing (Diener et al., 2010). Overall, social media use research has consistently investigated
its links with hedonic well‐being. Also, research on adolescents’ well‐being has tended to focus on hedonia,
while the quest for meaning, self‐realization, and flourishing (i.e., eudaimonic well‐being) among adolescents
remains underexplored, even though those constructs are deemed essential to achieve optimal development.
Additionally, since advocates of self‐help initiatives such as digital disconnecting claim that disconnection
practices can uniquely influence eudaimonia (Syvertsen & Enli, 2020), eudaimonic well‐being is particularly
interesting to consider when studying disconnection. From a purely hedonic perspective, engaging in
voluntary disconnection is redundant if social media is used to provide immediate gratification. However,
from an eudaimonic perspective, there are instances where maximizing pleasure might become problematic,
which also applies to the context of social media use (Meier & Reinecke, 2021). For example, spending so
much time on social media that one neglects other activities that bring joy or begins phubbing loved ones,
might not result in the highest levels of well‐being in the long run. However, literature on voluntary
disconnection and eudaimonic well‐being is scarce and is, among adolescents, virtually non‐existent.

Therefore, in this study, we aimed to examine subdimensions of both types of well‐being and investigate
levels of satisfaction with life and flourishing among adolescents. In addition to well‐being, we also sought
an understanding of how adolescent levels of other sociopsychological factors (i.e., loneliness and depressive
symptoms) relate to adolescents’ social media app use and adoption of disconnection tools. Thus, our final
RQ was:

RQ3: Do sociopsychological characteristics (i.e., loneliness, depressive symptoms, flourishing, and

satisfaction with life) vary across different classes of adolescents in terms of the use of social media
apps and the adoption of disconnection tools?

Media and Communication • 2024 • Volume 12 • Article 8597 6

5. Methods

5.1. Participants and Procedure

In this study, we used data from a larger multi‐wave panel study conducted among Flemish adolescents in
Belgium. Data from the third wave, in particular, conducted in June 2020 with 966 adolescents, served as
our variables of interest. Adolescents were invited to participate in our study through their school but
completed the online questionnaire at home. In the context of the larger research project, we collaborated
with 24 high schools randomly selected from a list provided by the Flemish Education Department.
However, because participation required schools’ consent, the sampling method can be regarded as
convenience sampling. Active consent was obtained from all participating adolescents, while parents
provided passive consent. Confidentiality was assured, and ethical approval was obtained through the KU
Leuven ethics committee (SMEC, G‐2018 031187). Adolescents received a €5 voucher for completing the
survey. During data collection, which coincided with the Covid‐19 pandemic, schools were partly reopened,
and several restrictive social and economic measures were lifted following a national lockdown. A more
detailed description of the sampling method and procedure appears in Schreurs et al. (2023).

Ten adolescents failed the survey’s attention check and were removed from the analytical sample, which thus
consisted of 956 adolescents. Their mean age was 15.10 years (𝑆𝐷 = 1.599), 64.9% were girls, and 86% were
born in Belgium. By comparison, in the overall population in 2020, 49.2% of high school students were girls,
and 90% were born in Belgium (Onderwijs Vlaanderen, 2020).

5.2. Measures

5.2.1. Nuanced Mobile Disconnection Tools

After participants were given a definition of disconnection tools (i.e., tools used to control how much time
a person spends on their smartphone and that enable them to limit the time spent using social media apps),
they were instructed to indicate which disconnection tools they used. The questionnaire presented them with
six external detox apps that were among the most popular at the time—i.e., Moment, Forest, QualityTime,
Space, OFFTIME, and RealizeD—and two built‐in device settings—i.e., iOS Screentime and Digital Well‐Being
for Android. They had to mark every app that they had previously used. Regarding the use of built‐in device
settings, 34.5% of the participants used iOS Screentime, while 7.6% used Android Digital Well‐Being. As for
external detox apps, 5.6% of the sample used the Forest app, while other apps (e.g., Moment and RealizeD),
were used by less than 1%. Moreover, 3.6% reported using a different detox app not listed on the questionnaire.
Lastly, 43.2% indicated that they were not using any disconnection tools. If the participants indicated using
at least one of the tools before, then they were identified as users of disconnection tools. We distinguished
two variables for the adoption of disconnection tools: the use of external detox apps and the use of built‐in
device settings.

5.2.2. Social Media Use

Adolescents reported how frequently they used 10 social media platforms—Facebook, Instagram, Snapchat,
Messenger, WhatsApp, YouTube, TikTok, Reddit, X, and Tumblr—on a 6‐point Likert scale (1 = never, 2 = seldom,

Media and Communication • 2024 • Volume 12 • Article 8597 7

3 = every few weeks, 4 = a few times a week, 5 = every day, and 6 = multiple times a day). We aggregated
those platforms into four categories of social media use: instant messaging (i.e., WhatsApp, Messenger, and
Snapchat; 𝑀 = 4.05, 𝑆𝐷 = 1.043), social networking (i.e., Facebook and Instagram; 𝑀 = 4.01, 𝑆𝐷 = 1.229),
microblogging and news (i.e., Reddit, X, and Tumblr; 𝑀 = 1.36, 𝑆𝐷 = 0.724), and entertainment (i.e., YouTube
and TikTok; 𝑀 = 4.67, 𝑆𝐷 = 1.169). This categorization is based on affordances shown by literature that
Snapchat, for instance, is used mainly as an instant messaging platform due to the ephemerality and briefness
of the messages, while also taking into account the specific platform affordances at the time of the study
(i.e., Bayer et al., 2020). The categorization was also conceptualized based on literature describing how those
social media platforms are used and what motivations for their use played a role in 2020 (Bayer et al., 2020;
Rhee et al., 2021) when the data were collected.

5.2.3. Sociodemographic Factors

Age was assessed by subtracting adolescents’ birth year from 2020, the year of the study, while gender was
indicated as either boy (=1) or girl (=2). Regarding SES, participants answered a validated measure of how well
off they think their family is compared with other families (𝑀 = 7.67, 𝑆𝐷 = 1.397, range from 1–10; Goodman
et al., 2001).

5.2.4. Sociopsychological Factors

Loneliness was measured with the UCLA loneliness scale (RULS‐8) validated among Dutch‐speaking
adolescents (Goossens et al., 2013). Participants indicated how true eight statements were for them on a
5‐point Likert scale ranging from 1 (completely untrue) to 5 (completely true). Examples were “I miss company”
and “I feel excluded” (𝑀 = 2.15, 𝑆𝐷 = 0.719, 𝛼 = 0.83).

Depressive symptoms were measured with a 12‐item version of the Center for Epidemiologic Studies
depression scale (Poulin et al., 2005). Adolescents had to indicate how often they experience symptoms
such as lack of appetite (e.g., “I did not feel like eating, my appetite was gone”) and tiredness (e.g., “I felt too
tired to do things”) in the past week on a 4‐point Likert scale ranging from 1 (never/rarely) to 4 (always;
𝑀 = 1.84, 𝑆𝐷 = 0.495, 𝛼 = 0.86).

Flourishing was captured with four items on the flourishing scale (Diener et al., 2010). Adolescents indicated
their agreement to items such as “I am optimistic about my future” on a 7‐point Likert scale ranging from 1
(completely disagree) to 7 (completely agree; 𝑀 = 4.95, 𝑆𝐷 = 1.070, 𝛼 = 0.78).

Last, satisfaction with life was measured using one item on the satisfaction with life scale (Diener et al., 1985),
for example, “I am satisfied with my life” rated on a 7‐point Likert scale ranging from 1 (strongly disagree) to 7
(strongly agree; 𝑀 = 5.61, 𝑆𝐷 = 1.338).

5.3. Analysis

Before analysis, confirmatory factor analysis supported the hypothesized one‐factor structure of the
validated scales for loneliness, depressive symptoms, and flourishing. Mean scores were created and
transformed into positive integers. We followed a three‐step approach (Asparouhov & Muthén, 2014),

Media and Communication • 2024 • Volume 12 • Article 8597 8

beginning with an LCA in Mplus to identify subgroups for social media use and the adoption of
disconnection tools. We employed latent class indicators (i.e., social media use, built‐in device settings, and
external detox apps) to estimate the latent class model. Second, based on the most optimal class model, we
created the variable of most likely class membership for each adolescent using the latent class posterior
distribution. Third, logistic regressions were conducted to investigate the associations between the
independent variables (i.e., sociodemographic and sociopsychological factors) and the dependent variable
(i.e., most likely class membership) while accounting for misclassification in the second step (Asparouhov &
Muthén, 2014). Please see OSF for anonymized data, syntaxes, and complete materials (https://osf.io/69drn/
?view_only=17697ea2b53f4d369eb5a4a7b735acb5).

5.3.1. Identifying the Optimal Class Solution

One to five class solutions were considered. The overall fit of each model was compared via the following tests:
the Akaike information criterion (AIC), the Bayesian information criterion (BIC), and the Lo‐Mendell‐Rubin
likelihood ratio test (LMR‐LRT). While a lower AIC and BIC indicate a better fit, the LMR‐LRT provides a 𝑝‐value
that denotes whether a class solution is statistically better than the previous solution with one fewer class
(𝑝 < 0.05). Furthermore, we evaluated the classification’s accuracy based on the classification probabilities for
most likely latent class membership, entropy, and minimum and maximum class sizes. Probabilities closer to
1 indicated a good classification, with acceptable values ranging between 0.80 and 0.90. The entropy value
needed to be greater than 0.60, and a class needed to have a minimum class size of 5% of the total sample or
at least 50 participants (Weller et al., 2020).

5.3.2. Predictors of Class Membership

To predict class membership, the variable of most likely class membership was employed as a latent class
indicator variable, with uncertainty rates prefixed at the probabilities. We included the independent variables
(i.e., sociodemographic and sociopsychological factors) as auxiliary variables (Asparouhov & Muthén, 2014)
by following the R3STEP method. The specified auxiliary variable was treated in the three‐step method as a
latent class predictor. Lastly, we examined the observed means and proportions of the predictive variables for
each class of the optimal class solution using SPSS.

6. Results

Entertainment apps were the most commonly used social media platforms, which more than half of the
adolescents used at least daily. By contrast, microblogging and news apps were rarely used. Less than half of
the participants (43.2%) did not use any disconnection tools. However, 42.1% used at least one of the
built‐in device settings, 11.7% used at least one external detox app, and 4.5% used both disconnection tools.

6.1. Model Selection

Table 1 presents the model fit indices and classification accuracy indices of five LCA models. As shown, the
AIC decreased from the second to the fifth model, and for all solutions, class membership probability and
entropy estimates were appropriate. Entropy was the highest in the five‐class solution, however this solution’s
minimal class size was less than 5%. The LMR‐LRT 𝑝 value was non‐significant in the three‐class solution,

Media and Communication • 2024 • Volume 12 • Article 8597 9

Table 1. Model fit indices and classification accuracy indices LCA models (𝑛 = 956).
Par. Class BIC AIC Entropy LMR‐LRT LMR‐LRT Min. Max. Min. Max.
value p‐value class class probability probability
size size
21 1 11563.642 11461.524 — — — 956 956 1 1
a
43 2 11338.727 11129.629 0.704 373.422 < 0.001 236 720 0.777 0.959
65 3 11343.929 11027.850 0.669 144.820 0.8407 138 561 0.767 0.894
87 4 11440.719 11017.659 0.755 53.834 0.8058 48 405 0.733 0.954
109 5 11543.744 11013.704 0.758 47.640 0.1318 48 397 0.679 0.862
Notes: Par. = number of free parameters; BIC = Bayesian information criterion; AIC = Akaike’s information criterion;
LMR‐LTR = Lo‐Mendell‐Rubin likelihood ratio test; a = elected as the final model.

which indicates that adding classes after the two‐class solution did not improve the model’s fit compared
with a model with one fewer class (𝑝 < 0.001). Beyond that, the BIC was lowest in the two‐class solution.
Accordingly, we selected the two‐class solution, which was also the most parsimonious model, to further
interpret. This parsimonious model balances model complexity (i.e., number of latent classes) and model fit
(i.e., how well the model explains the observed data).

6.2. Model Interpretation

Table 2 displays the prevalence and item‐response probabilities for the variables of social media use and the
adoption of disconnection tools. Class 1 comprised 24.7% of the participants, while Class 2 comprised 75.3%.
Class 1 had a high probability of being moderate to low social media users. For example, the adolescents in that
class had a 0.461 probability of using instant messaging only every few weeks. However, they also showed
a higher use of entertainment media apps. Regarding disconnection tools, adolescents in Class 1 showed low
probabilities for using built‐in device settings (0.266) and external detox apps (0.133). Based on those results,
we labeled Class 1 as “low adoption of social media apps and disconnection tools.”

Class 2 consisted of adolescents who scored high for social media use. For instance, the probability of using
instant messaging daily in Class 2 was 0.408. Concerning social networking, the adolescents in the group
showed moderate to high use. However, microblogging and news apps, as in the other class, were not
frequently used. Regarding disconnection tools, the adolescents in Class 2 showed a 0.459 probability of
using built‐in device settings and a 0.098 probability of using external detox apps. Based on those results,
we labeled Class 2 as “high adoption of social media apps and disconnection tools.” We further explored the
significant differences in social media use and use of disconnection tools between the two classes with an
independent samples 𝑡‐test, we found significant differences for instant messaging, 𝑡(907) = 31.899,
𝑝 < 0.001; entertainment media, 𝑡(907) = −12.432, 𝑝 < 0.001; social networking, 𝑡(363.020) = −21.442,
𝑝 < 0.001; and built‐in device settings, 𝑡(444.615) = −5.336, 𝑝 < 0.001; but not for microblogging,
𝑡(359.120) = 0.852, 𝑝 = 0.395, or external detox apps, 𝑡(349.317) = 1.899, 𝑝 = 0.058.

Media and Communication • 2024 • Volume 12 • Article 8597 10

Table 2. Prevalence and item‐response probabilities for disconnection tool use and social media use (𝑛 = 956).
Latent classes
Latent class prevalences Class 1 (24.7%) Class 2 (75.3%)
Indicators Sample proportions Inter‐response probabilities
Total sample Class 1 Class 2

Disconnection tools
Use of built‐in device settings 0.405 0.266 0.459
Use of external detox tools 0.108 0.133 0.098

Instant messaging
Never 0.020 0.072 0
Seldom 0.058 0.211 0
Every few weeks 0.173 0.461 0.062
A few times a week 0.402 0.253 0.458
Daily 0.295 0 0.408
Multiple times a day 0.053 0.004 0.072

Entertainment media
Never 0.010 0.036 0
Seldom 0.031 0.062 0.019
Every few weeks 0.113 0.213 0.075
A few times a week 0.276 0.411 0.225
Daily 0.263 0.128 0.315
Multiple times a day 0.307 0.150 0.367

Microblogging and news

Never 0.756 0.768 0.751
Seldom 0.152 0.131 0.160
Every few weeks 0.074 0.070 0.075
A few times a week 0.013 0.025 0.009
Daily 0.006 0.007 0.005
Multiple times a day 0 0 0

Social networking
Never 0.055 0.180 0.007
Seldom 0.045 0.141 0.008
Every few weeks 0.171 0.337 0.107
A few times a week 0.410 0.341 0.437
Daily 0.205 0 0.283
Multiple times a day 0.114 0 0.158

6.3. Associations Between Sociodemographic and Sociopsychological Factors and Latent

Class Membership

Last, using logistic regressions, we investigated whether different sociodemographic and sociopsychological
factors predicted the probability of belonging to Class 1 or Class 2. Table 3 presents the observed means
and proportions of the variables by class, while Table 4 shows the results of the logistic regressions on Class 2
membership probability for each predicting variable. Being older and being a girl predicted a higher probability
of belonging to Class 2 than to Class 1. Moreover, higher scores for depressive symptoms and flourishing
predicted a higher probability of belonging to Class 2 than to Class 1.

Media and Communication • 2024 • Volume 12 • Article 8597 11

Table 3. Observed means and proportions of study variables by class (𝑛 = 956).
Class 1 Class 2
M/% SD M/% SD
Age 14.42 1.339 15.32 1.917
Gender (% girl) 47.5 — 70.6 —
SES 7.57 1.478 7.70 1.369
Depressive symptoms 1.81 0.602 1.91 0.564
Loneliness 2.12 0.801 2.16 0.740
Life satisfaction 5.63 1.492 5.61 1.283
Flourishing 4.83 1.217 5.00 1.084

Table 4. Results of logistic regression, membership Class 2 (𝑛 = 956).

Class 2 (reference category = Class 1)
B SE OR 95% CI
Age 0.565*** 0.091 1.760 [1.473/2.104]
Gender 1.233*** 0.199 3.431 [2.323/5.066]
SES 0.085 0.071 1.089 [0.948/1.251]
Depressive symptoms 0.402* 0.182 1.494 [1.045/2.136]
Loneliness 0.101 0.134 1.106 [0.851/1.439]
Life satisfaction −0.015 0.078 0.985 [0.845/1.147]
Flourishing 0.172* 0.085 1.188 [0.851/1.439]
Notes: 𝐵 = logit coefficient; SE = standard error; OR = odds ratio; CI = confidence interval; * 𝑝 < 0.05, ** 𝑝 < 0.01,
*** 𝑝 < 0.001.

7. Discussion

Through LCA, we aimed to investigate whether classes of adolescents significantly differ in their use of
social media and adoption of disconnection tools (RQ1). Examining social media use and voluntary practices
of disconnection simultaneously is crucial when studying different groups of adolescents. Previous LCA on
adolescents’ social media use has primarily focused on usage (e.g., Foerster & Röösli, 2017), while research
on disconnection has primarily investigated involuntary disconnection, including parental control measures
or school smartphone bans (e.g., Schmuck et al., 2023). However, initial research has indicated that today’s
adolescents are indeed concerned about maintaining a healthy digital balance (Jorge et al., 2023).

In our sample of 956 Flemish adolescents, we identified heterogeneity in how adolescents (dis)connect from
social media and found two distinct classes. Adolescents in Class 1 can be labeled as “low adoption of social
media apps and disconnection tools.” In Class 2, labeled as “high adoption of social media apps and
disconnection tools,” adolescents reported higher use of instant messaging apps, entertainment media apps,
and social networking apps than ones in Class 1. Moreover, adolescents in Class 1 showed low probabilities
of using disconnection tools, specifically built‐in device settings, while ones in Class 2 had higher
probabilities of using those settings.

Media and Communication • 2024 • Volume 12 • Article 8597 12

The different patterns in app usage between the two classes suggest that adolescents in Class 1 maintain
a more detached relationship with social media. They do not use social media intensely and do not want
assistance in disconnecting from the platforms that they use. A quarter of our sample belonged to Class 1.
Conversely, adolescents in Class 2 showed a higher use of social media but were also more likely to search
for balance by adopting disconnection tools. That finding indicates that most adolescents are frequent social
media users, but they seek balance in their use. Adolescents adopting disconnection tools mostly used built‐in
device settings (e.g., iOS Screentime), while only 11.7% of the sample indicated using external detox apps
(e.g., the Forest app). While previous studies have tended to merge those types of strategies (e.g., Schmuck,
2020), our findings suggest that it is important to distinguish types of disconnection tools because they seem
to be adopted differently. One explanation could be that such tools might be adopted hierarchically by first
attempting to disconnect without them (e.g., by putting the phone away) and, if that fails, by using the readily
available settings (e.g., built‐in device settings). In turn, external detox apps may be viewed as a last resort, for
they also require the most effort to adopt.

Furthermore, we investigated whether sociodemographic (RQ2) and sociopsychological factors (RQ3) varied
across the identified classes. Regarding sociodemographic variables, being older predicted a higher
probability of belonging to Class 2. That finding aligns with previous results (Coyne et al., 2019) and
literature on social media use among Flemish adolescents (Vanwynsberghe et al., 2022), which have shown
that, throughout adolescence, the use of social media platforms increases. In that regard, age can be seen as
a proxy for developmental change (Berk, 2014). Beyond the fact that age tends to determine whether and to
what extent adolescents have access to mobile devices, a social developmental argument also applies, as the
importance of peer relationships and staying connected increases during this time (Nesi et al., 2018).

Age predicting disconnection may be explained by higher social media use in the first place, but also by
maturing and being able to reflect on their use (Meinert & Reinecke, 2018; Siebers et al., 2021). In line with
research on developing self‐control throughout adolescence, older adolescents will be more likely to reflect
on their use, perceive it as too much, feel guilty about it, and want to change it (Coyne et al., 2019). Those
feelings of perceived overuse can motivate voluntary disconnection (Nassen et al., 2023).

Regarding gender, girls were more likely to belong to Class 2 than boys. Research has indeed shown that
girls use social media more often than boys and that it aligns with their tendency to spend more time on
social relationships, dyadic friendships, and popularity, especially during adolescence (Twenge & Martin,
2020). Girls also seem more likely to engage in voluntary disconnection. Turel and Vaghefi (2019) found in
their study among young adults that female social networking users were more likely to fail at self‐imposed
abstinence, which can be viewed as a risk for isolation, boredom, and fear of missing out. While those
findings focus on failing at voluntary disconnection, they might indicate why adolescent girls are more likely
than boys to use disconnection tools. The adoption of such tools can also be seen as a more drastic step in
wanting to disconnect after offline mechanisms of self‐control fail. Moreover, multiple studies have
indicated that voluntary disconnection for women can be more difficult than for men due to gender
differences in social expectations, including care work (e.g., Van Bruyssel et al., 2023). Those expectations
may also apply to adolescent girls and make voluntary disconnection more difficult for them (Jorge et al.,
2023). Girls might feel more obliged to stay connected at all times with close ones through social media
(Baumer, 2018). In that regard, our findings confirm the possibility of the described gender difference, as
previously suggested by Jorge et al. (2023).

Media and Communication • 2024 • Volume 12 • Article 8597 13

SES did not predict membership to either of the classes. This could be due to our samples’ overall high score
for SES. Only 10.4% of respondents rated themselves 5 or less on a 10‐point scale regarding their family’s
SES compared with other families. Because research has often emphasized socioeconomic inequalities in
practices of disconnection (Gui & Büchi, 2021; Nguyen, 2021), future studies should pay close attention to
digital inequalities among adolescents in terms of their connection and disconnection and be sure to include
adolescents in their samples who represent all levels of SES. Especially in contexts such as the Covid‐19
pandemic, the family context can influence the use and possibility for non‐use of social media (Nguyen &
Hargittai, 2024).

Regarding sociopsychological factors, higher reported levels of depressive symptoms predicted a higher
probability of belonging to Class 2 (“high adoption of social media apps and disconnection tools”) than to
Class 1 (“low adoption of social media apps and disconnection tools”). This result is in line with past findings
of positive associations between social media use and depression (Liu et al., 2019). Individuals with
depression might perceive social media use as a potential driver in gaining social connectedness and peer
support and, therefore, want to use it more (Dolev‐Cohen & Barak, 2013). However, other research has
suggested that using social media can heighten depressive symptoms among adolescents via media effects
mechanisms such as availability stress, social comparison, and approval anxiety (e.g., Schreurs et al., 2023).
As such, adolescents with depression may attribute their perceived ill‐being to their high level of social
media use and disconnect to alleviate those feelings. In any case, future qualitative research is needed to
gain a more in‐depth understanding of the motivations for social media (dis)connection among adolescents
with depression.

Surprisingly, scoring higher on flourishing (Diener et al., 2010) also increased the probability of belonging to
Class 2, whereas being satisfied with one’s life (Diener et al., 1985) did not predict membership to that class.
It is plausible that people with higher eudaimonic well‐being are also ones who use disconnection tools, for
those strategies can be seen as acts of self‐help that are mostly adopted for improving eudaimonia (Syvertsen
& Enli, 2020). In that context, it is puzzling that depressive symptoms also predict a higher chance of belonging
to Class 2. A possible explanation for the result may be the more state‐like measure used for depressive
symptoms (e.g., “in the past week…I did not feel like eating, my appetite was gone”) and the more trait‐like
measure used for eudaimonic well‐being (e.g., “I lead a purposeful and meaningful life”). The difference could
suggest that disconnecting from social media may come in response to differing needs; adolescents who
struggle with ill‐being (e.g., depressive symptoms) in a day‐to‐day context may use disconnection to alleviate
those feelings, whereas ones who perceive their lives overall as being purposeful and fulfilling may also engage
in disconnection to maintain their eudaimonia. Research has indeed shown that digital detox apps can be
valuable in preventing the effects of social media use that harms well‐being (Schmuck, 2020). Individuals who
score high for eudaimonia will also struggle in life from time to time but, in those periods, have the highest
chances of finding strategies to restore well‐being (e.g., by adopting disconnection tools). Future research
is needed, however, to disentangle other traits and state well‐being dynamics among adolescents in their
disconnection strategies.

7.1. Practical Implications

The findings of our study present some practical implications. We found that about half of the adolescents in
our sample voluntarily use disconnection tools to balance their social media use. So far, the focus of research

Media and Communication • 2024 • Volume 12 • Article 8597 14

on managing adolescents’ use has been on parental control and smartphone bans at schools. Current
interventions for digital literacy increasingly focus on helping adolescents gain the knowledge and skills to
exercise self‐control over their use of digital media. However, the efficacy of those initiatives has been
questioned, because self‐control is not fully developed in adolescents (Meinert & Reinecke, 2018). Our study
highlights the importance of assigning responsibility to adolescents regarding their social media use because
voluntary disconnection seems to be an initiative that they can and are willing to undertake themselves. It is
also important to address equal gender socialization concerning social media to ensure that boys and girls
face equal opportunities to engage in voluntary disconnection. Moreover, initiatives to promote voluntary
disconnection and healthy social media diets should target adolescents who frequently use social media,
because this subgroup seems to experience an imbalance and seeks both connection and disconnection.

7.2. Limitations and Directions for Future Research

Some limitations of our study should be addressed. First, we selected the two‐class solution over the
five‐class solution primarily because one of the five classes was smaller than our minimal class size criterion
of 5% or more than 50 participants. This small class size might be due to our sample’s lack of diversity.
Second, we assessed social media use and the use of disconnection tools with self‐report measures. Future
research could alternatively collect log data from adolescents’ phones to gather an objective measure for
those usage variables. Third, our data was collected six months into the Covid‐19 pandemic when
restrictions had already been partly lifted. That specific context might have resulted in altered or heightened
(social) media use and, as a consequence, a greater need for disconnection. In response, future research
should replicate our findings in a more recent, non‐Covid‐19 context. Fourth, as for the sociopsychological
variables, we used self‐report measures, meaning that only a perceived version of the concepts was
assessed. However, these subjective assessments are also important in accounting for adolescents’
perspectives (Valkenburg & Piotrowski, 2017). Another interesting avenue for future research could be
latent class growth analysis, which requires longitudinal data, and would allow us to determine changes in
the trajectories of social media use and disconnection tool use throughout adolescence. Such research could
be particularly relevant given our finding that age was associated with class membership. Future research
should also investigate more recent types of nuanced mobile disconnection tools, including in‐app features
for well‐being (e.g., Instagram’s time spend feature).

8. Conclusion

This study was the first to reveal different subgroups based on social media use and the adoption of
disconnection tools among adolescents. A quarter of the adolescents exhibited patterns of low social media
use and low interest in using tools to disconnect from social media. Most adolescents used social media very
frequently and were more likely to seek out tools to disconnect. Girls, older adolescents, and adolescents
who scored higher for depressive symptoms and flourishing were more likely to seek out a balance between
connecting and disconnecting from social media by adopting those disconnection tools. Thus, along with
examining social media use among adolescents, our study has highlighted the importance of future research
to equally prioritize the exploration of disconnection strategies, recognizing their significant role in fostering
a comprehensive understanding of adolescents’ social media behaviors. For adolescents, mastering skills for
maintaining a balanced social media diet and proficient self‐control will be essential for thriving on social
media later in life.

Media and Communication • 2024 • Volume 12 • Article 8597 15

Funding
This work was supported by the Research Foundation Flanders (FWO‐Vlaanderen) under Grant 11D3119N.

Conflict of Interests
The authors declare no conflict of interests.

Data Availability
Please see OSF for additional information on the study and the sample, anonymized data, syntaxes, and full
materials (https://osf.io/69drn/?view_only=17697ea2b53f4d369eb5a4a7b735acb5).

References
Anderson, M., Faverio, M., & Park, E. (2024). How teens and parents approach screen time. Pew Research
Center. https://www.pewresearch.org/internet/2024/03/11/how‐teens‐and‐parents‐approach‐screen‐
time/#how‐teens‐feel‐when‐they‐don‐t‐have‐their‐phone
Asparouhov, T., & Muthén, B. (2014). Auxiliary variables in mixture modeling: Three‐step approaches using
Mplus. Structural Equation Modeling, 21(3), 329–341. https://doi.org/10.1080/10705511.2014.915181
Baumer, E. P. (2018). Socioeconomic inequalities in the non use of Facebook. In Proceedings of the 2018
CHI Conference on Human Factors in Computing Systems (pp. 1–14). Association for Computing Machinery.
https://doi.org/10.1145/3173574.3174190
Bayer, J. B., Triệu, P., & Ellison, N. B. (2020). Social media elements, ecologies, and effects. Annual Review Of
Psychology, 71, 471–497. https://doi.org/10.1146/annurev‐psych‐010419‐050944
Berk, L. E. (2014). Development through the lifespan (6th ed.). Pearson.
Coyne, S. M., Padilla‐Walker, L. M., Holmgren, H. G., & Stockdale, L. A. (2019). Instagrowth: A longitudinal
growth mixture model of social media time use across adolescence. Journal of Research on Adolescence,
29(4), 897–907. https://doi.org/10.1111/jora.12424
Diener, E. D., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of
Personality Assessment, 49(1), 71–75. https://doi.org/10.1207/s15327752jpa4901_13
Diener, E. D., Wirtz, D., Tov, W., Kim‐Prieto, C., Choi, D. W., Oishi, S., & Biswas‐Diener, R. (2010). New
well‐being measures: Short scales to assess flourishing and positive and negative feelings. Social Indicators
Research, 97, 143–156. https://doi.org/10.1007/s11205‐009‐9493‐y
Dolev‐Cohen, M., & Barak, A. (2013). Adolescents’ use of instant messaging as a means of emotional relief.
Computers in Human Behavior, 29(1), 58–63. https://doi.org/10.1016/j.chb.2012.07.016
Foerster, M., & Röösli, M. (2017). A latent class analysis on adolescent’s media use and associations with health
related quality of life. Computers in Human Behavior, 71, 266–274. https://doi.org/10.1016/j.chb.2017.02.
015
Goodman, E., Adler, N. E., Kawachi, I., Frazier, A. L., Huang, B., & Colditz, G. A. (2001). Adolescents’ perceptions
of social status: Development and evaluation of a new indicator. Pediatrics, 108(2), Article e31. https://
doi.org/10.1542/peds.108.2.e31
Goossens, L., Klimstra, T., Luyckx, K., Vanhalst, J., & Teppers, E. (2013). Reliability and validity of the Roberts
UCLA loneliness scale (RULS‐8) with Dutch‐speaking adolescents in Belgium. Psychologica Belgica, 54(1),
5–18. http://doi.org/10.5334/pb.ae
Gui, M., & Büchi, M. (2021). From use to overuse: Digital inequality in the age of communication abundance.
Social Science Computer Review, 39(1), 3–19. https://doi.org/10.1177/0894439319851163
Heffer, T., Good, M., Daly, O., MacDonell, E., & Willoughby, T. (2019). The longitudinal association

Media and Communication • 2024 • Volume 12 • Article 8597 16

 between social‐media use and depressive symptoms among adolescents and young adults: An empirical
reply to Twenge et al.(2018). Clinical Psychological Science, 7(3), 462–470. https://doi.org/10.1177/
2167702618812727
Hunt, M. G., Marx, R., Lipson, C., & Young, J. (2018). No more FOMO: Limiting social media decreases
loneliness and depression. Journal of Social and Clinical Psychology, 37(10), 751–768. https://doi.org/
10.1521/jscp.2018.37.10.751
Huta, V., & Waterman, A. S. (2013). Eudaimonia and its distinction from hedonia: Developing a classification
and terminology for understanding conceptual and operational definitions. Journal of Happiness Studies, 15,
1425–1456. https://doi.org/10.1007/s10902‐013‐9485‐0
Jorge, A., Agai, M., Dias, P., & Martinho, L. C.‐V. (2023). Growing out of overconnection: The process of
dis/connecting among Norwegian and Portuguese teenagers. New Media & Society, 26(11), 6779–6795.
https://doi.org/10.1177/14614448231159308
Liu, D., Baumeister, R. F., Yang, C. C., & Hu, B. (2019). Digital communication media use and psychological
well‐being: A meta‐analysis. Journal of Computer‐Mediated Communication, 24(5), 259–273. https://doi.
org/10.1093/jcmc/zmz013
Martela, F., & Sheldon, K. M. (2019). Clarifying the concept of well‐being: Psychological need satisfaction as
the common core connecting eudaimonic and subjective well‐being. Review of General Psychology, 23(4),
458–474. https://doi.org/10.1177/1089268019880886
Masur, P. K., Veldhuis, J., & de Vaate, N. B. (2022). There is no easy answer: How the interaction of content,
situation, and person shapes the effects of social media use on well‐being. In D. Rosen (Ed.), The social
media debate (pp. 187–202). Routledge. https://doi.org/10.4324/9781003171270‐12
Meier, A., & Reinecke, L. (2021). Computer‐mediated communication, social media, and mental health:
A conceptual and empirical meta‐review. Communication Research, 48(8), 1182–1209. https://doi.org/
10.1177/0093650220958224
Meinert, J., & Reinecke, J. (2018). Self‐control during adolescence: Examining the stability of low self‐control
and the effects of parental social controls. European Journal of Criminology, 15(5), 523–543. https://doi.org/
10.1177/1477370817749180
Nassen, L. M., & Karsay, K. (2024, February). Nuanced mobile disconnection strategies: A way to mitigate the
negative effects of excessive phone use? Etmaal 2024, Rotterdam, The Netherlands.
Nassen, L. M., Vandebosch, H., Poels, K., & Karsay, K. (2023). Opt‐out, abstain, unplug. A systematic review
of the voluntary digital disconnection literature. Telematics and Informatics, 81, Article 101980. https://
doi.org/10.1016/j.tele.2023.101980
Nesi, J., Choukas‐Bradley, S., & Prinstein, M. J. (2018). Transformation of adolescent peer relations in the social
media context: Part 1—A theoretical framework and application to dyadic peer relationships. Clinical Child
and Family Psychology Review, 21, 267–294. https://doi.org/10.1007/s10567‐018‐0261‐x
Nesi, J., & Prinstein, M. J. (2015). Using social media for social comparison and feedback‐seeking: Gender
and popularity moderate associations with depressive symptoms. Journal of Abnormal Child Psychology, 43,
1427–1438. https://doi.org/10.1007/s10802‐015‐0020‐0
Neves, B. B., de Matos, J. M., Rente, R., & Martins, S. L. (2015). The “non‐aligned” young people’s narratives of
rejection of social networking sites. Young, 23(2), 116–135. https://doi.org/10.1177/1103308815569393
Nguyen, M. H. (2021). Managing social media use in an “always‐on” society: Exploring digital wellbeing
strategies that people use to disconnect. Mass Communication and Society, 24(6), 795–817. https://doi.org/
10.1080/15205436.2021.1979045
Nguyen, M. H. (2023). “Maybe I should get rid of it for a while…”: Examining motivations and challenges

Media and Communication • 2024 • Volume 12 • Article 8597 17

 for social media disconnection. The Communication Review, 26(2), 125–150. https://doi.org/10.1080/
10714421.2023.2195795
Nguyen, M. H., & Hargittai, E. (2024). Digital disconnection, digital inequality, and subjective well‐being:
A mobile experience sampling study. Journal of Computer‐Mediated Communication, 29(1). https://doi.org/
10.1093/jcmc/zmad044
Onderwijs Vlaanderen. (2020). Statistisch jaarboek van het Vlaams onderwijs 2019–2020. https://onderwijs.
vlaanderen.be/nl/onderwijsstatistieken/statistisch‐jaarboek‐van‐het‐vlaams‐onderwijs‐2019‐2020#pdf
Poulin, C., Hand, D., & Boudreau, B. (2005). Validity of a 12‐item version of the CES‐D used in the National
Longitudinal Study of Children and Youth. Health Promotion and Chronic Disease Prevention in Canada,
26(2/3), 65–72.
Rhee, L., Bayer, J. B., Lee, D. S., & Kuru, O. (2021). Social by definition: How users define social platforms
and why it matters. Telematics and Informatics, 59, Article 101538. https://doi.org/10.1016/j.tele.2020.
101538
Rosič, J., Carbone, L., Vanden Abeele, M. M., Lobe, B., & Vandenbosch, L. (2024). Measuring digital well‐being
in everyday life among Slovenian adolescents: The perceived digital well‐being in adolescence scale. Journal
of Children and Media, 18(1), 99–119. https://doi.org/10.1080/17482798.2023.2272651
Schmuck, D. (2020). Does digital detox work? Exploring the role of digital detox applications for problematic
smartphone use and well‐being of young adults using multigroup analysis. Cyberpsychology, Behavior, and
Social Networking, 23(8), 526–532. https://doi.org/10.1089/cyber.2019.0578
Schmuck, D., Stevic, A., Matthes, J., & Karsay, K. (2023). Out of control? How parents’ perceived lack of
control over children’s smartphone use affects children’s self‐esteem over time. New Media & Society, 25(1),
199–219. https://doi.org/10.1177/14614448211011452
Schreurs, L., Lee, A. Y., Liu, X. S., & Hancock, J. T. (2024). When adolescents’ self‐worth depends on their
social media feedback: A longitudinal investigation with depressive symptoms. Communication Research,
51(6), 631–659. https://doi.org/10.1177/00936502241233787
Schreurs, L., Meier, A., & Vandenbosch, L. (2023). Exposure to the positivity bias and adolescents’ differential
longitudinal links with social comparison, inspiration and envy depending on social media literacy. Current
Psychology, 42, 28221–28241. https://doi.org/10.1007/s12144‐022‐03893‐3
Siebers, T., Beyens, I., Pouwels, J. L., & Valkenburg, P. M. (2021). Social media and distraction: An
experience sampling study among adolescents. Media Psychology, 25(3), 343–366. https://doi.org/
10.1080/15213269.2021.1959350
Steele, J. R., & Brown, J. D. (1995). Adolescent room culture: Studying media in the context of everyday life.
Journal of Youth and Adolescence, 24, 551–576. https://doi.org/10.1007/BF01537056
Syvertsen, T., & Enli, G. (2020). Digital detox: Media resistance and the promise of authenticity. Convergence,
26(5/6), 1269–1283. https://doi.org/10.1177/1354856519847325
Treré, E. (2021). Intensification, discovery and abandonment: Unearthing global ecologies of dis/connection
in pandemic times. Convergence, 27(6), 1663–1677. https://doi.org/10.1177/13548565211036804
Turel, O., & Vaghefi, I. (2019). Social media detox: Relapse predictors. Psychiatry Research, 284, Article 112488.
https://doi.org/10.1016/j.psychres.2019.112488
Twenge, J. M., & Martin, G. N. (2020). Gender differences in associations between digital media use and
psychological well‐being: Evidence from three large datasets. Journal of Adolescence, 79(1), 91–102.
https://doi.org/10.1016/j.adolescence.2019.12.018
Valkenburg, P. M., Meier, A., & Beyens, I. (2022). Social media use and its impact on adolescent mental health:
An umbrella review of the evidence. Current Opinion in Psychology, 44, 58–68. https://doi.org/10.1016/
j.copsyc.2021.08.017

Media and Communication • 2024 • Volume 12 • Article 8597 18

Valkenburg, P. M., & Piotrowski, J. T. (2017). Plugged in: How media attract and affect youth. Yale University
Press. https://drupal.yalebooks.yale.edu/sites/default/files/files/Media/9780300228090_UPDF.pdf
Van Bruyssel, S., De Wolf, R., & Vanden Abeele, M. (2023). Who cares about digital disconnection? Exploring
commodified digital disconnection discourse through a relational lens. Convergence. Advance online
publication. https://doi.org/10.1177/13548565231206504
Vanden Abeele, M. M. P. (2021). Digital wellbeing as a dynamic construct. Communication Theory, 31(4),
932–955. https://doi.org/10.1093/ct/qtaa024
Vanden Abeele, M. M. P., & Nguyen, M. H. (2024). Digital media as ambiguous goods: Examining the digital
well‐being experiences and disconnection practices of Belgian adults. European Journal of Communication,
39(2), 122–144. https://doi.org/10.1177/02673231231201487
van der Wal, A., Valkenburg, P. M., & van Driel, I. I. (2024). In their own words: How adolescents use social media
and how it affects them. Social Media + Society, 10(2). https://doi.org/10.1177/20563051241248591
Vanwynsberghe, H., Joris, G., Waeterloos, C., Anrijs, S., Vanden Abeele, M., Ponnet, K., De Wolf, R., Van
Ouytsel, J., Van Damme, K., Vissenberg, J., D’Haenens, L., Zenner, E., Peters, E., De Pauw, S., Frissen, L.,
& Schreuer, C. (2022). Onderzoeksrapport apestaartjaren : De digitale leefwereld van kinderen en jongeren.
Mediaraven.
Weller, B. E., Bowen, N. K., & Faubert, S. J. (2020). Latent class analysis: A guide to best practice. Journal of
Black Psychology, 46(4), 287–311. https://doi.org/10.1177/0095798420930932
Yang, C. C., & Brown, B. B. (2013). Motives for using Facebook, patterns of Facebook activities, and late
adolescents’ social adjustment to college. Journal of Youth and Adolescence, 42, 403–416. https://doi.org/
10.1007/s10964‐012‐9836‐x

About the Authors

Lise‐Marie Nassen is a PhD candidate at the Media Psychology Lab, KU Leuven, Belgium.
Her research focuses on voluntary disconnection practices from smartphones and social
media, specifically looking at the associations between disconnecting and psychological
well‐being.

Kathrin Karsay (PhD, 2018) is an assistant professor in the Department of Communication

at the University of Vienna. Her research examines the interrelationships between digital
media use and well‐being in daily life, focusing on mental, physical, and social health among
youth. She has explored the effects of both connection and disconnection from social media
and smartphones, shedding light on their impacts on well‐being.

Laura Vandenbosch (BOF‐ZAP research professorship grant) is director of the Media

Psychology Lab, Department of Communication Science, KU Leuven, Belgium. The
relationship between media and well‐being is the core subject of her research, leading to
contributions in several fields including developmental psychology, sexology, body image,
social relationships, and communication theory.

Media and Communication • 2024 • Volume 12 • Article 8597 19

 Lara Schreurs (PhD) is an assistant professor at the Media Psychology Lab, KU Leuven,
Belgium. Lara’s research focuses on social media, adolescents, well‐being, social media
literacy, and interventions. She has received fellowships from the Research Foundation
(FWO), Flanders, and the internal research funds of the KU Leuven.

Media and Communication • 2024 • Volume 12 • Article 8597 20