Corresponding author: So Yeon PARK, PhD, RN College of Nursing, Dong-A University, 32 Daesingongwon-ro, Seo-gu, Busan 49201, Korea INTRODUCTION
Adolescence is a transitional stage between childhood and adulthood, marked by rapid physical, cognitive, and psychosocial growth, making it a critical period for establishing long-term health habits [1,2]. In Korean society, adolescents spend long hours sitting due to heavy academic workloads, and sedentary behavior during non-academic time has also increased with growing digital device use [3,4]. These behavioral changes pose serious health risks, threatening both physical and mental health. Sedentary behavior, characterized by low energy expenditure, increases cardiovascular disease risk by >34% [5], causes hyperglycemia through insulin resistance, and elevates the risk of metabolic syndrome [5].
According to World Health Organization guidelines on physical activity and sedentary behavior [1], increased sedentary behavior is associated with reduced quality of life, depression, and lower prosocial behavior. Systematic reviews also report anxiety, stress, decreased self-esteem, and sleep disorders [6,7]. The United States Centers for Disease Control and Prevention recommends limiting sedentary behavior based on related studies [8]. In Korea, growing emphasis has been placed on the risks of sedentary behavior among children and adolescents and the need for corresponding guidelines [9]. Despite such demands [10], research on sedentary behavior in Korea remains insufficient, with only preliminary surveys on its prevalence among children and adolescents [11].
Recent studies have examined sedentary behavior as a health behavior influenced by various factors [12]. For adolescents, a multidimensional approach that considers such complex contexts is required. To date, most studies in Korea have focused on physical characteristics such as weight, obesity, and sex [13,14] or on the relationship between subjective health perception [4] or smartphone use time and anxiety levels [15]. Few have analyzed structural relationships where various factors interact simultaneously.
Although adolescents’ behavioral patterns may vary depending on differences in activity patterns by weekday vs. weekend, emotional state, and school or home environment, these contexts have not been comprehensively considered [6]. Gorely et al. [16] emphasized that sedentary behavior and physical activity patterns differ by day, time, and social context and with activity type or environmental factors among adolescents.
Accordingly, this study aimed to analyze the factors influencing sedentary behavior among adolescents in a more systematic and comprehensive manner, categorizing the main variables into three domains: physical, psychological, and social characteristics.
Physical characteristics included sex, age, and body mass index (BMI). Sex influences physical activity and media use [17,18], age reflects behavioral characteristics according to developmental stage [18], and BMI indirectly reflects the physical health status of adolescents, which is significantly associated with sedentary time [14] and can be used to predict the level of health risk in adolescents.
Psychological characteristics included perceived stress, anxiety, and smartphone overdependence. Stress and anxiety are common emotional problems during adolescence and closely related to sedentary behavior [4,15]. Smartphone overdependence has recently emerged as a major factor influencing adolescents’ daily lives and sedentary time [19]. However, while the relationship between smartphone use and sedentary behavior has been established, temporal patterns (weekdays vs. weekends) remain insufficiently elucidated. Based on the findings of Gorely et al. [16], which reported differences by day and time, it is necessary to examine these aspects in the present study as well.
Social characteristics included cohabitation with family members, which reflects social support and living environment. Living with guardians may influence daily routines and smartphone regulation, thereby affecting sedentary behavior patterns [20].
Therefore, this study aimed to identify the effects of physical (sex, age, and BMI), psychological (perceived stress, anxiety, and smartphone overdependence), and social (living with family) characteristics on non-academic sedentary behavior among Korean adolescents through multivariate analysis.
METHODS
Research design
This study was a secondary analysis on data from the 19th Korea Youth Risk Behavior Survey, conducted in 2023 [21], to identify factors influencing sedentary behavior among adolescents.
Participants
The Korea Youth Risk Behavior Survey is an anonymous, web-based survey conducted annually by the Korea Disease Control and Prevention Agency, targeting middle- and high-school students enrolled in Korea as of April each year. In this study, data from 52,880 adolescents who participated in the 2023 survey were analyzed.
Measurements
Sedentary behavior
Sedentary behavior was assessed using the item, “During the past 7 days, how many hours per day on average did you spend sitting?” Excluding time spent on learning, responses were collected separately for weekdays (Monday to Friday) and weekends (Saturday to Sunday) and converted into minutes. Accordingly, sedentary behavior was analyzed separately for weekdays and weekends.
Physical characteristics
Physical characteristics included sex, age, and BMI. BMI was calculated as kg/m2.
Psychological characteristics
Psychological characteristics consisted of perceived stress, anxiety, and smartphone overdependence. Stress responses of “very much” and “much” were recoded as “yes,” while those of “a little,” “not much,” and “not at all” were recoded as “no.” Anxiety was measured using the 7-item Generalized Anxiety Disorder Scale. Participants responded to the question, “Over the last 2 weeks, how often have you been bothered by the following problems?” on a 4-point scale ranging from “not at all” (0) to “nearly every day” (3). The total score ranged from 0 to 21, with higher scores indicating greater anxiety. Cronbach’s α in this study was 0.91, consistent with that reported previously [22]. Smartphone overdependence was assessed using the Smartphone Overdependence Scale, which consists of 10 items scored on a 4-point scale from “strongly disagree” (1) to “strongly agree” (4), with the total score ranging from 10 to 40. Higher scores indicate greater overdependence. Cronbach’s α was 0.84 at the time of development [23] and 0.91 in this study.
Data collection and ethical considerations
The Korea Youth Risk Behavior Survey is a nationally approved statistical survey conducted annually by the Korea Disease Control and Prevention Agency, officially approved by Statistics Korea. Until 2014, the survey underwent review by the Institutional Review Board (IRB) of the Korea Disease Control and Prevention Agency; however, since 2015, it has been exempted from IRB review under the Enforcement Rule of the Bioethics and Safety Act [24]. All participants received information on the purpose and content of the survey and voluntarily consented to participate. The survey was conducted anonymously using a self-administered, web-based questionnaire. Responses were collected in a form that could not identify individuals, and the raw data were provided to researchers by the Korea Disease Control and Prevention Agency. Approval for use of the data in this study was obtained according to procedures outlined on the official Korea Youth Risk Behavior Survey website (https://www.kdca.go.kr/yhs/). This study was exempted from approval by the IRB because it was a secondary data analysis using de-identified, publicly available data. Analyses were performed using secondary data that did not permit the identification of participants.
Data analysis
Data from the Korea Youth Risk Behavior Survey were analyzed using a complex sampling design. Stratification, clustering, weighting, and finite population correction coefficients included in the raw data provided by the Korea Disease Control and Prevention Agency were reflected in the analyses, which were conducted using IBM SPSS Statistics version 29.0 (IBM Corp.). The raw data of this survey contained some nonresponses due to missing value processing for logical errors and outliers. As the item nonresponse rate was low (within 2%), no imputation was performed, and the results were calculated using the available data. The handling of missing values for the variables used in this study was as follows: sedentary behavior data were treated as missing when the reported daily sitting time was 24 hours or 0 minutes. Age values were treated as missing if participants were younger than 12 years or older than 19 years. Height, weight, and BMI values were treated as missing when they were outside the mean ±3·standard deviations for each sex and school grade.
The analysis procedures were as follows: First, sedentary time was analyzed using complex sample descriptive statistics. Second, differences in sedentary behavior according to physical, psychological, and social characteristics were analyzed using complex sample descriptive statistics and a complex sample general linear model (GLM). Third, factors influencing sedentary behavior were analyzed using a complex sample GLM.
Diagnostic statistics for regression assumptions are not provided in the SPSS complex sample GLM procedure; therefore, separate general linear regression analyses were conducted with and without applying sampling weights to exploratorily examine the regression assumptions. Results of assumption testing for the independent variables indicated that tolerance values ranged from 0.71 to 0.98, exceeding the threshold of 0.1, and the variance inflation factor ranged from 1.02 to 1.42, below the cutoff value of 10, indicating no multicollinearity concerns. The Durbin-Watson test, based on the unweighted regression analysis, yielded values between 1.91 and 1.95, close to 2, indicating the absence of autocorrelation. Therefore, the basic assumptions of the regression analyses were considered to be satisfied.
RESULTS
Sedentary behavior among Korean adolescents
The sedentary behaviors of the participants are presented in Table 1. The average weekday sedentary time was 206.35 minutes, while the average weekend sedentary time was 321.96 minutes.
Differences in sedentary behavior according to physical, psychological, and social characteristics of Korean adolescents
Differences in weekday sedentary behavior according to physical, psychological, and social characteristics of the participants were as follows: among physical characteristics, sex (P<0.001), age (P=0.035), and BMI (P<0.001) were significantly associated with weekday sedentary behavior. Among psychological characteristics, stress, anxiety, and smartphone overdependence were significantly associated with weekday sedentary behaviors (P<0.001). The social characteristics of living with family showed no statistically significant differences (P=0.319) (Table 2).
Differences in weekend sedentary behavior according to physical, psychological, and social characteristics of the participants were as follows: among physical characteristics, sex (P<0.001), age (P=0.035), and BMI (P<0.001) were significantly associated with weekend sedentary behavior. Among psychological characteristics, stress, anxiety, and smartphone overdependence were significantly associated with weekend sedentary behavior (P<0.001). The social characteristics of living with family showed no statistically significant differences (P=0.065) (Table 2).
Factors influencing sedentary behavior among Korean adolescents
Factors influencing sedentary behavior were analyzed using a complex sample GLM (Table 3). The dependent variable was sedentary behavior, entered separately for weekdays and weekends, and the independent variables included seven factors: physical characteristics (sex, age, and BMI), psychological characteristics (stress, anxiety, and smartphone overdependence), and social characteristics (living with family). Although living with family did not show a significant association with sedentary behavior in the univariate analysis, it was included in the regression model considering its theoretical validity as an important social contextual factor.
The factors influencing weekday sedentary behavior were sex (P<0.001), age (P<0.001), BMI (P<0.001), anxiety (P=0.003), and smartphone overdependence (P<0.001), all of which showed statistically significant associations. However, stress (P=0.370) and living with family (P=0.196) were not significantly associated. Compared with female participants, male participants had 7.94 units more weekday sedentary behavior. For each 1-unit increase in age, weekday sedentary behavior decreased by 1.71 units. For each 1-unit increase in BMI, weekday sedentary behavior increased by 1.33 units. For each 1-unit increase in anxiety, weekday sedentary behavior increased by 0.56 units. For each 1-unit increase in smartphone overdependence, weekday sedentary behavior increased by 1.85 units.
The factors influencing weekend sedentary behavior were sex (P<0.001), age (P<0.001), BMI (P<0.001), stress (P=0.012), anxiety (P<0.001), and smartphone overdependence (P<0.001), all of which showed statistically significant associations. However, living with family was not significantly associated (P=0.901). Compared with female participants, male participants had 33.36 units more weekend sedentary behavior. For each 1-unit increase in age, weekend sedentary behavior decreased by 11.89 units. For each 1-unit increase in BMI, weekend sedentary behavior increased by 3.03 units. Adolescents with stress had 5.67 units more weekend sedentary behavior than those without stress. For each 1-unit increase in anxiety, weekend sedentary behavior increased by 1.37 units. For each 1-unit increase in smartphone overdependence, weekend sedentary behavior increased by 2.38 units.
DISCUSSION
This study analyzed the physical, psychological, and social characteristics influencing non-academic sedentary behavior among adolescents and compared differences between weekdays and weekends. The results showed that sex, age, BMI, anxiety, and smartphone overdependence were significant influencing factors on both weekdays and weekends, whereas stress was significantly associated only on weekends. In contrast, living with family was not significantly related to sedentary behavior.
The regression analysis results showed a relatively low explanatory power, with R2=0.008 for the weekday sedentary behavior model and R2=0.022 for the weekend model. However, this outcome should be interpreted within the context of social and behavioral research. According to Ozili [25], it is common and acceptable for R2 values to be relatively low in these fields, since human behaviors—such as adolescent sedentary behavior—are influenced by numerous unobserved factors, including genetic predispositions, environmental cues, social norms, and individual motivations, as well as measurement errors. Therefore, this study was primarily designed not for high predictive accuracy, but rather to identify the specific, statistically significant effects of physical, psychological, and social characteristics on adolescent sedentary behavior.
Among the physical characteristics, sex consistently had a significant effect, with male students reporting longer sedentary times than female students on both weekdays and weekends. This may be related to the tendency of male students to prefer digital device–based sedentary activities, such as gaming and video watching, compared with female students [17,18]. In this study, sedentary time decreased with increasing age, with a more pronounced change observed on weekends. This contrasts with previous studies reporting that sedentary time increases with age among adolescents [26,27] but is consistent with the findings of Lim [14], which indicated that as adolescents advance in grade level, their study time increases and available leisure time decreases. This result may reflect the distinctive academic environment of Korean adolescents, where older students face heavier academic workloads and spend less time in leisure-based sedentary activities. Ibabe et al. [28] found that as adolescents grow older, structured leisure time tends to decrease, and unstructured screen-based activities increase, predicting higher problematic technology use (assessed through indicators such as time spent using mobile devices, social networking, chatting or emailing, and eating while using smartphone). This pattern supports the interpretation that older adolescents’ sedentary behavior becomes more academically driven and less discretionary. BMI was positively associated with sedentary behavior on both weekdays and weekends, supporting previous research suggesting that low physical activity may lead to weight gain, while a higher BMI may result in avoidance of activity [16]. During weekdays, relatively structured schedules physically restrict sedentary time among adolescents, whereas on weekends, such restrictions are reduced, leading to increased leisure time involving screen use [29].
Among psychological characteristics, stress was not significantly associated with weekday sedentary behavior but was associated with weekend sedentary behavior. In contrast, anxiety showed a significant positive association with sedentary behavior on both weekdays and weekends, and smartphone overdependence was identified as a predictor of sedentary behavior regardless of the day of the week. Hoare et al. [6] reported a positive relationship between anxiety and media-based sedentary behavior in adolescents, and Pearson and Biddle [30] emphasized that adolescents often engage in screen-based activities as a means of emotional recovery when experiencing stress. Subiron-Valera et al. [31] further reported that in environments with less external control, stress can directly increase screen time. These findings suggest that in autonomous environments, emotional stress may easily translate into immediate pleasure-seeking behaviors.
When experiencing emotional distress, people tend to prioritize immediate mood recovery over long-term goals [32]. From this perspective, sedentary behavior can be interpreted as an emotion-regulation strategy. In particular, the significant increase in sedentary behavior on weekends compared with weekdays suggests that emotional responses are more likely to translate directly into behavior in environments with greater autonomy. Meanwhile, anxiety is not a temporary emotion but a chronic emotional state, and adolescents often cope with it through avoidance strategies such as task avoidance [32]. Sedentary activities often involve media use, which can offer immediate rewards and emotional stability, potentially serving as a coping mechanism for alleviating anxiety [19,29,30]. However, when sedentary activities become habitual, they may negatively affect both physical and emotional health. Therefore, reducing sedentary behavior among adolescents requires interventions that not only limit time but also address the context in which sedentary behavior is used as a coping strategy for stress or anxiety, along with providing emotional support.
Living with family was not significantly associated with sedentary behavior, suggesting that qualitative family dynamics may have a greater influence on adolescents’ behavior than cohabitation itself. Bounova et al. [33] emphasized that parental monitoring and communication are key factors in reducing adolescents’ screen-related sedentary behavior, highlighting the importance of the parental role. Similarly, Yang et al. [20] reported that parents’ excessive smartphone use weakened their ability to regulate their children’s media use, underscoring the significance of parental modeling. Taken together, these findings suggest that parenting attitudes and media-related communication are more influential than quantitative indicators such as living arrangement. As this study used secondary data without qualitative family variables, future research should examine these factors in more depth.
In this study, adolescents spent more sedentary time on weekends than on weekdays, with stress and smartphone overdependence exerting stronger effects during weekends. This indicates that in more autonomous environments, such as weekends, emotional factors and exposure to digital media directly influence behavior. These findings suggest that interventions to reduce sedentary behavior among adolescents should consider differences by day of the week and focus on managing emotional factors in highly autonomous contexts.
Through a multidimensional analysis, this study demonstrated that single behavioral guidelines or time-based standards alone are insufficient. Instead, integrated approaches addressing emotional regulation, digital literacy, and age-specific strategies are needed. Anxiety and smartphone overdependence were related to sedentary behavior, indicating that these factors merit careful attention in developing intervention strategies. These findings suggest that intervention strategies should not only focus on limiting screen time but also take into account adolescents’ emotional states, patterns of media use, and everyday living environments. To ensure practical applicability, interventions should be implemented through school-based education, parental guidance on balanced media use, and community programs that provide structured leisure activities.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
