Using Machine Learning to Explore the Risk Factors of Problematic Smartphone Use among Canadian Adolescents during COVID-19: The Important Role of Fear of Missing Out (FoMO)

Abstract

The goal of the present study was to use machine learning to identify how gender, age, ethnicity, screen time, internalizing problems, self-regulation, and FoMO were related to problematic smartphone use in a sample of Canadian adolescents during the COVID-19 pandemic. Participants were N = 2527 (1269 boys; M_age = 15.17 years, SD = 1.48 years) high school students from the Lower Mainland of British Columbia, Canada. Data on problematic smartphone use, screen time, internalizing problems (e.g., depression, anxiety, and stress), self-regulation, and FoMO were collected via an online questionnaire. Several different machine learning algorithms were used to train the statistical model of predictive variables in predicting problematic smartphone use. The results indicated that Shrinkage algorithms (lasso, ridge, and elastic net regression) performed better than other algorithms. Moreover, FoMO, emotional, and cognitive self-regulation made the largest relative contribution to predicting problematic smartphone use. These findings highlight the importance of FoMO and self-regulation in understanding problematic smartphone use.

1. Introduction

There is no doubt that the daily lives of adolescents were and, in many ways, continue to be drastically altered by the COVID-19 global pandemic. In Canada, government efforts to limit the spread of COVID-19 included the implementation of a series of social distancing policies, which resulted in reduced face-to-face interactions. Restrictions such as these are likely to have major implications for adolescents for whom, developmentally, connecting with others is of the utmost importance [1]. In fact, in response to their reduced ability to connect with others face-to-face in the time of the pandemic, adolescents increasingly relied on their smartphones to stay connected [2,3,4]. While this adaptive use of smartphones likely helped adolescents meet their developmental social goals, increased smartphone use during the pandemic may have put adolescents at a greater risk for problematic smartphone use [5,6]. As such, there is a dire need for research investigating the risk factors of problematic smartphone use among Canadian adolescents in the pandemic context. Accordingly, the goal of the current study was to examine the relative influence of predictive factors related to problematic smartphone use using supervised machine learning in a sample of Canadian adolescents during the COVID-19 pandemic.

1.1. Problematic Smartphone Use among Adolescents during the Pandemic

Problematic smartphone use refers to a compulsive and problematic pattern of smartphone use that disrupts daily life and results in negative consequences [7,8,9,10]. Despite the many positive benefits of using smartphones, problematic smartphone use has been found to be associated with adjustment difficulties, such as academic failures and physical and mental health problems [11,12,13,14,15]. For example, researchers found that problematic smartphone use was associated with children’s low self-esteem [16].

While problematic smartphone use is generally conceptualized as a subset of behavioral addictions, similar to drug addiction, where the addiction is to the behavior or to the feelings experienced when acting out the behavior instead of to a substance [17,18], its conceptualization within the medical-addiction model remains a controversial topic among researchers. Indeed, many have argued that this conceptualization may be an oversimplification of psychological processes [19])—emphasizing the need for more research into the motivational, affective, and social processes underlying these problematic patterns of behavior.

During the COVID-19 pandemic, the use of the Internet on smartphones became an important tool for adolescents to connect with others, entertain themselves, pass the time, and engage with learning [20]. Despite this adaptive and arguably beneficial use of smartphones, researchers warned that an increase in smartphone use might result in an increase in problematic smartphone use during the COVID-19 pandemic. In fact, according to a recent systematic review, exposure to the COVID-19 pandemic did result in an increase in Internet-based addictive behaviors, including problematic smartphone use [5].

1.2. Risk Factors of Problematic Smartphone Use

Recent research has increasingly focused on examining the risk factors of problematic smartphone use. While this work has predominantly investigated demographic variables and smartphone usage [21,22,23,24,25] recent work has also examined the role of other psychopathology-related variables (e.g., mental health problems, personality traits) in explaining problematic smartphone use [26]. These studies have employed a variety of theoretical models to understand problematic smartphone use (e.g., Uses and Gratifications Theory, Theories of Behavioural Addiction, and Smartphone Attachment). While these models have been useful for understanding some of the underlying motivations guiding problematic smartphone use, they tend to be somewhat narrow in focus. Specifically, these studies have only focused on examining single risk factors. Contrarily, the Interaction of the Person-Affect-Cognition Execution model (I-PACE) [27] provides a more holistic model within which to conceptualize problematic adolescent smartphone use.

Specifically, this theoretical framework proposes that various factors interact to contribute to internet addiction, including individual characteristics (such as personality traits), affective processes (such as mood and emotions), cognitive processes (such as decision-making and attentional biases), and executive functions (such as self-regulation and impulse control) [27]. According to the I-PACE model, these factors work together in a dynamic manner and can lead to a vicious cycle of problematic smartphone use. For example, individuals with low levels of executive function (e.g., self-regulation) and psychopathology (such as anxiety or depression) may be more likely to use the internet excessively to cope with negative emotions [27]. Thus, drawing on the I-PACE model, we proposed several potential risk factors for problematic smartphone use. These factors include gender as an individual characteristic, internalizing problems as affective processes, self-regulation and fear of missing out as cognitive processes and executive functions, and screen time. Our study utilized these factors to guide our analysis and predictions of problematic smartphone use.

1.2.1. Internalizing Problems

As a result of social isolation during the COVID-19 pandemic, adolescents may have experienced increased internalizing problems, such as anxiety and depression [28,29]. For example, in a recent meta-analysis, researchers found that depression and anxiety were higher in children during pandemics overall, specifically during the COVID-19 pandemic [30]. In terms of problematic smartphone use, this is particularly concerning since a recent systematic review identified mental health as the primary reason for problematic smartphone use [26]. Moreover, this review also identified adolescents as being more likely to use smartphones to cope with negative emotions, which places them at an even higher risk for problematic smartphone use. In the context of the COVID-19 pandemic, during which social distancing policies have led to increases in levels of depression, anxiety, and experiences of negative emotions among adolescents [28,29,31,32], many may seek emotional comfort through smartphones as a potential coping mechanism. This, in turn, could lead to problematic smartphone use. Indeed, previous studies have consistently demonstrated that internalizing problems, such as depression, stress, and anxiety, are positively associated with problematic smartphone use [33,34].

1.2.2. Self-Regulation

Broadly speaking, self-regulation refers to an individual’s ability to control their cognitive, behavioral, and emotional impulses in order to achieve their goals [35]. Previous work has focused on behavioral regulation and found that individuals with deficiencies in their ability for behavioral regulation are more likely to develop problematic habitual behaviors, such as problematic smartphone use over time [36,37], as they lack the ability to control their impulses. For example, individuals who lack self-control are more likely to respond to messages as soon as they receive them, which could lead them to develop a habitual reliance on their smartphone which, in turn, could develop into problematic smartphone use. Contrarily, previous research has found that highly self-regulated people were more effective at regulating their problematic behaviors and reducing their smartphone use [33,38]. From an emotional regulation perspective, maladaptive emotional regulation strategies have been identified as a predictor of problematic smartphone use [39], such that difficulties with regulating one’s emotions leads to more engagement with the smartphone as a way of coping, which leads to the development of problematic smartphone use. That is, as adolescents experienced increased mental health problems and negative emotions during the pandemic; those with low emotional regulation ability were more likely to develop maladaptive mechanisms in response to these negative emotions and, thus, develop problematic smartphone use.

Despite evidence of the link between self-regulation (e.g., emotional and behavior regulation) deficits and problematic smartphone use, Researcher recently argued that, since self-regulation is goal-oriented, it could also be positively associated with smartphone engagement [40]. Specifically, as connecting with others is critical for children’s development, especially during adolescence [41], smartphone use can provide an important means for adolescents to meet their developmental social goals. As a result, reducing smartphone use may not be a relevant goal for adolescents, especially during the pandemic when many relied heavily on their devices to connect with others. Thus, in the present study, we examined whether self-regulation (e.g., emotional, behavioral, and cognitive regulation) could predict problematic smartphone use among Canadian adolescents.

1.2.3. Fear of Missing Out (FoMO)

FoMO is a personality trait characterized by a desire to maintain constant social interaction while also being concerned about missing out on important social experiences [42]. FoMO has been linked with negative affect [39], feelings of social disconnectedness [43] and, most commonly, internalizing problems, including depression and anxiety symptoms [44,45]. Extant research has conceptualized FoMO within notions of Self-Determination Theory [46], wherein human behaviors are motivated by the satisfaction of our basic psychological needs for competence, autonomy, and relatedness. Specifically, this work has identified FoMO as a major driver of the excessive use of digital technology to meet individuals’ basic needs for relatedness.

Indeed, previous studies have found that FoMO was an important predictor of problematic smartphone use [46,47,48,49,50,51]. For example, a longitudinal study found that FoMO was associated with problematic smartphone use over one year among adolescents [52]. Of particular relevance to the current study, researcher investigating the risk factors of problematic smartphone use among Chinese undergraduate students using machine learning algorithms found that FoMO made the largest relative contribution in predicting problematic smartphone use [46]. This work, in addition to the understanding that social distancing and associated decreases in socializing time during the pandemic can lead to higher levels of FoMO [53], suggests that FoMO may be a particularly salient risk factor for problematic smartphone use in the time of COVID-19. Specifically, it is possible that the reduction in face-to-face social interaction during COVID-19 may have increased FoMO for those already at risk, making them more likely to turn to their smartphones to decrease their anxiety about missing out [54]. This, in turn, could have contributed to their development of problematic smartphone use as they became preoccupied with checking up on what others were doing.

1.2.4. Screen Time

Screen time refers to the amount of time (minutes) that one spends actively using one’s device daily [55]. Historically, problematic smartphone use was thought to be a behavioral problem, so screen time was proposed as a key indicator of problematic smartphone use [56,57]. Indeed, an extensive body of previous work has indicated that screen time is associated with problematic smartphone use [25,57]. Therefore, in the current study, we also examined the predictive role of screen time.

1.2.5. Gender Differences

We explored potential gender differences, given that recent studies have found that girls report a higher level of problematic smartphone use [58,59]. For example, researchers found that while boys were more engaged in instrumental apps that were focused on activities (e.g., playing games), girls were more likely to be engaged with social networking services (SNS) that were focused on maintain bonding relationships—the latter of which was more likely to be linked with problematic smartphone use [60].

1.3. The Present Study

The current study sought to investigate the most prominent statistical predictors of problematic smartphone use among Canadian adolescents during the COVID-19 pandemic. Drawing upon the conceptual and empirical evidence, we hypothesized that screen time, internalizing problems, and FoMO would be positively associated with problematic smartphone use. Self-regulation (e.g., emotional, cognitive, and behavioral regulation), on the other hand, would be negatively associated with problematic smartphone use. Furthermore, girls would show higher levels of problematic smartphone use than boys. At last, given our relatively large dataset, we hypothesized that applying supervised machine learning algorithms to our data would produce a regression model that predicts problematic smartphone use among adolescents. Machine learning is a sub-area of artificial intelligence that is increasingly being used in psychology research to construct models with high predictive power [61,62]. In comparison to statistical methods that emphasize inference, machine learning focuses on prediction, with analysis performed autonomously or with minimal human guidance in large data sets [61]. Machine learning has been evidenced as an effective tool for identifying risk factors for problematic smartphone use in previous work by previous study [47]—the present study builds on this work by identifying risk factors in a sample of Canadian adolescents.

2. Materials and Methods

2.1. Procedures

The present study obtained ethical approval from the institutional review board of The University of British Columbia and the participating school districts. To assist with obtaining a representative sample within each school, this study used a passive parental consent procedure [63,64], and all participants provided active, informed assent prior to their participation. As a recruitment incentive, each participating school’s students were entered into a drawing for a chance to win a smartphone. Data was collected separately at each school between the months of October 2020 and May 2021. All self-report questionnaires were completed online through Qualtrics (www.qualtrics.com, accessed on 11 April 2021) during school hours.

2.2. Participants

Initially, 2527 adolescents participated in the study. However, after applying our exclusion criteria (which involved removing adolescents who only completed demographic information, such as gender, age, and ethnicity), 427 individuals were excluded from the data file. As a result, the final analysis was conducted on a sample of 2100 adolescents. Participants were N = 2100 (1053 girls; M_age = 15.29 years, SD = 1.48 years) high school students between the ages of 11 and 18 (grades 8–12) recruited from five secondary schools across three public school districts in British Columbia, Canada. For all the variables, missing data rates ranged from 0.2% to 6.2%. Little’s (1988) MCAR test was not significant, χ2 (50) = 59.04, p = 0.18, suggesting that the pattern of missingness was not systematic. Expectation maximization was used to handle the missing data. The sample was ethnically diverse, with approximately 39.0% self-identifying as East Asian; 24.1% White/Caucasian; 16.2% South Asian; 11.2% Southeast Asian; 4.8% West Asian; 3.3% Latin American; 3.1% Indigenous, and 1.3% Other (e.g., African American, Caribbean).

2.3. Measures

2.3.1. Smartphone Addiction

Problematic smartphone use was measured using the Smartphone Addiction Scale-Short Version (SAS-SV; [65]). The measure is comprised of 10 items that are rated on a 6-point Likert scale. Sample items included “I use my phone longer than intended”. A composite variable was created by calculating the average score across all items. This measure has been shown to have good psychometric properties [66].

2.3.2. Internalizing Problems

To assess their mental health, adolescents completed the Depression, Anxiety, and Stress Scale (DASS-21, [67]). The DASS-21 consists of 3 subscales: depression (7 items; e.g., “I couldn’t seem to experience any positive feeling at all”), anxiety (7 items; e.g., “I was aware of dryness of my mouth”), and stress (7 items; e.g., “I found it hard to wind down”), that are rated on a 4-point Likert scale, ranging from 0 (Did not apply to me at all) to 3 (Applied to me very much or most of the time). Sum scores are computed by adding up the scores on the items per (sub)scale and multiplying them by 2. These cut-off scores are derived from a set of severity ratings proposed by researchers [67]. This measure has been shown to have reliable psychometric properties [68].

2.3.3. Screen Time

To measure screen time, participants were asked to report their “average daily screen time” on their primary mobile device by accessing the preinstalled Screen Time application on iOS devices (i.e., iPhone/iPad) or the comparable Digital Wellbeing application on Android devices in hours and minutes. This measure captures participants’ daily average screen time over the previous seven days.

2.3.4. Fear of Missing Out (FoMO)

FoMO was measured using the Fear of Missing Out scale (FOMO; [42]). The scale is comprised of 10 items (e.g., “I get worried when I find out my friends are having fun without me”) rated on a 5-point Likert scale. A composite variable was created by calculating the average score across all items. This measure has demonstrated evidence of reliability and validity among Canadian adolescents [43].

2.3.5. Self-Regulation

Participants completed the Self-Regulation Questionnaire [35]. The scale contained 13 items on a 4-point Likert scale, which comprised three subscales: emotional self-regulation (5 items, e.g., “I have a hard time controlling my temper”), behavioral self-regulation (4 items, e.g., “As soon as I see things that are not working, I do something about it”), and cognitive self-regulation (3 items, e.g., “I develop a plan for all my important goals”). This measure has demonstrated evidence of reliability and validity among Western adolescents [35].

2.4. Statistical Analytic Strategy

SPSS 22.0 and R software, version 3.6.1, were used to conduct preliminary analysis and machine learning (caret R package). We first shuffled the effective sample’s participant rows using a fixed random seed (for subsequent, consistent replication). Then 70% (n = 1382) of the dataset was randomly selected as a training sample and the remaining 30% (n = 592) as a test sample. To simulate validation data for each analysis, we used k-folds repeated cross-validation (RCV), splitting the training sample into ten folds of sub-samples, training nine of the ten sub-samples, and testing the tenth sub-subsample. In these simulations, we used a fixed random seed to ensure consistent replication. We then validated the performance of the aggregated final predictive model for each analysis separately on the external test sample.

Next, regression-based symptom forecasting methods with supervised machine learning were conducted with problematic smartphone use as the continuous dependent variable. The predictor variables were gender, age, ethnicity, screen time, internalizing problems, self-regulation, and FoMO). Five separate machine learning algorithms were compared (e.g., “shrinkage” methods-ridge, lasso, and elastic net regression algorithms, extreme gradient boosting, and random forest). Those algorithms were selected because of their resistance to overfitting and prediction of nonlinear associations [69]. We compared the algorithms using root mean squared error (RMSE), mean absolute error (MAE), R-squared values, and statistical tests.

3. Results

3.1. Preliminary Analysis

Means and standard deviations for all study variables are presented in Table 1, and correlations among study variables are presented in Table 2. Overall, the pattern of associations was consistent with theoretical expectations. Adolescents’ problematic smartphone use was significantly and positively associated with screen time, depression, anxiety, stress, and FoMO but was negatively associated with emotional, cognitive, and behavioral self-regulation.

3.2. Machine Learning Results

To begin with, we conducted a comparison of various machine learning algorithms on both training and testing datasets. The findings indicated that the “shrinkage” techniques, specifically ridge, lasso, and elastic net regression algorithms, performed marginally better than the other algorithms in terms of the smallest RMSE and MAE values, as well as the largest R-squared values (see Table 2). Among these three algorithms, lasso regression was the top-performing one. Regarding the explained variance, ridge and lasso algorithms resulted in 36% in the training set and around 34% in the test set.

Subsequently, we utilized the models created by the “shrinkage” algorithms to evaluate the relative significance of the predictor variables in forecasting problematic smartphone use. Our findings demonstrated that, compared to other predictors, Fear of Missing Out (FoMO) was the most influential explanatory variable in forecasting problematic smartphone use in Canadian adolescents. Additionally, after FoMO, emotional and cognitive self-regulation was the next most robust predictor of problematic smartphone use (see Table 3).

4. Discussion

The COVID-19 pandemic reduced opportunities for adolescents to interact face-to-face, and many turned to their smartphones as a means of connecting with others [70]. While this strategy was an adaptive and often effective means of fostering social connection for many adolescents [43], this increased reliance on smartphones contributed to an increased risk of problematic smartphone use among this demographic [5]. Although many previous studies have investigated risk factors related to problematic smartphone use, these studies have largely focused on the examination of single risk factors (or a small set of risk factors) [26]. Furthermore, building on previous research [46], we included self-regulation (e.g., emotion, behavior, and cognitive regulation) to predict smartphone use. Our results of the simulation were comparable across the various machine learning statistical algorithms used, but shrinkage algorithms outperformed the others. Notably, our findings indicated that FoMO was the most predictive of problematic smartphone use among Canadian adolescents, followed by emotional and cognitive self-regulation.

4.1. Fear of Missing Out (FoMO)

The most important finding from this study was that, when compared to other risk factors, FoMO played a relatively large role in predicting problematic smartphone use among Canadian adolescents during COVID-19. This finding not only supports our hypothesis but also adds to the growing body of literature demonstrating that FoMO is strongly associated with smartphone engagement among adolescents [46,50]. Indeed, adolescents higher in FoMO may be more likely to turn to their smartphones to decrease their anxiety about missing out and check up on what others are doing. However, this can backfire as they may become hyperaware (e.g., through engaging with social media) of events and social interactions from which they may be excluded or left out—instead amplifying their existing feelings of FoMO [46]. These heightened feelings of FoMO can then propel adolescents to once again turn to their smartphones to alleviate some of their anxiety around missing out and, thus, propagate a vicious cycle of compulsive smartphone use and subsequent problematic smartphone use. Moreover, during the COVID-19 pandemic, adolescents’ limited opportunities for face-to-face interactions may have amplified this cycle, as many relied on their smartphones for social connection (i.e., a particularly important developmental need in adolescence) [41]. Thus, the risk of FoMO contributing to problematic smartphone use may have been heightened during the COVID-19 pandemic, as adolescents’ opportunities for face-to-face social interaction were limited. Moreover, our findings have important implications for intervention and prevention efforts targeting problematic adolescent smartphone use in Canada. Specifically, these efforts should aim to identify and target adolescents who have high levels of FoMO and help them break the vicious cycle of compulsive smartphone use as a means of coping with feelings of anxiety about being left out or excluded.

4.2. Self-Regulation

Following FoMO, emotional and cognitive self-regulation was the next strongest predictor of problematic smartphone use among Canadian adolescents. Our results supported researcher’s speculation that, as compared to social goals (e.g., FoMO), self-regulation was less likely (but still important) to predict problematic smartphone use among adolescents during the pandemic. In other words, while self-regulation was an important predictor of problematic smartphone use, adolescents’ developmentally rooted need to feel socially connected and included (as demonstrated by high levels of FoMO) was the most important driver of problematic smartphone use [40]. Moreover, our findings indicated that emotional and cognitive regulation were more strongly related to problematic smartphone use than behavioral regulation during the COVID-19 pandemic. As previously discussed, it is possible that adolescents with poor emotional and cognitive self-regulation were more likely to develop maladaptive cognitive mechanisms in response to the negative emotions they may have experienced during the pandemic and, thus, were more prone to developing problematic smartphone use. Taken together, our findings and greatly contribute to our understanding of problematic smartphone use [10]. Specifically, while problematic smartphone use was previously believed to be a behavioral problem, our findings suggest that it may also be related to emotional or cognitive problems, particularly in times of limited face-to-face social interactions, such as during the COVID-19 pandemic. In further support of this, our findings also confirmed that participants’ reported screen time was not associated with problematic smartphone use when other factors were considered—suggesting that smartphone use per se is not a significant risk factor for developing problematic smartphone use in adolescence. This finding should help appease some of the moral concerns around screens amongst parents, teachers, and practitioners, which have emerged alongside the increased reliance on technology during the pandemic. Rather, our findings suggest that those aiming to reduce the risk of problematic smartphone use among adolescents should strive to address their maladaptive cognitive mechanisms rather than attempting to restrict their screen time. Moreover, during the COVID-19 pandemic, specifically, assisting adolescents in developing self-regulation skills, particularly emotional and cognitive self-regulation skills, may be beneficial in reducing their risk of problematic smartphone use.

4.3. Internalizing Problems

Although relatively less important in predicting adolescents’ problematic smartphone use during COVID-19, internalizing problems (i.e., depression, anxiety, and stress) were positively associated with problematic smartphone use in our study (see Table 4). This is in line with previous studies that have consistently demonstrated that internalizing problems were associated with problematic smartphone use among adolescents [33,34]. In this way, as social distancing policies led to increases in internalizing problems among adolescents during the COVID-19 pandemic, many may have turned to their smartphones for emotional comfort—thereby heightening their risk of problematic smartphone use. Moreover, our results suggest that, as compared to other factors (self-regulation and FoMO), internalizing problems play a less important role in determining problematic smartphone use. Thus, it is possible that not all adolescents who have internalizing problems will develop problematic smartphone use, but rather that only those with higher FoMO and poorer self-regulation skills are at risk. In support of this notion, using machine learning, researcher found that relative to other risk factors, depression and anxiety were not particularly important in their contribution to problematic smartphone use [46]. Additionally, it is important to note that internalizing problems and FoMO are highly correlated, and previous work has suggested that there may be a link between internalizing problems and FoMO. For example, researchers have found that individuals who experience higher levels of anxiety and depression may be more likely to experience FoMO as a result of feeling socially disconnected or isolated [46]. Similarly, individuals who experience FoMO may be more likely to experience anxiety or depression as a result of the constant comparison and pressure to keep up with others [44]. Therefore, future studies should keep investigating the complex relations among internalizing problems, FoMO, and problematic smartphone use.

4.4. Gender Differences

In terms of gender differences, our results are consistent with the findings of several studies, which have found that, as compared to boys, girls reported higher levels of problematic smartphone use [58,59]. Indeed, researchers have suggested that because girls are more socially oriented, they are more likely to use SNS than boys, which may make them more prone to developing problematic smartphone use. As such, future research should continue to investigate the potential gender differences in various types of smartphone-related addictions.

4.5. Limitations and Future Directions

The results from the current study contribute to our understanding of the risk factors associated with problematic smartphone use among Canadian adolescents during the COVID-19 pandemic. Despite the contributions of this research to the extant literature, some limitations should be noted. First, our study was cross-sectional, limiting our ability to investigate the direction of effects between constructs and make reliable causal inferences. For example, previous studies have found that early problematic smartphone use could predict later internalizing problems [71,72,73]. As a result, future research should use a longitudinal design to investigate these effects over time. Moreover, our study only assessed screen time and did not measure smartphone checking, which is another important aspect of smartphone usage. In fact, recent research indicates that smartphone screen time and checking may have distinct associations, with checking to have more negative consequences [10]. Therefore, future studies should investigate the relationship between smartphone checking and problematic smartphone use. Furthermore, while we only used “shrinkage” methods, extreme gradient boosting, and random forest machine learning algorithms in the current study to train the statistical model of predictive variables in predicting problematic smartphone use, future research should compare other popular deep learning algorithms, such as neural network and Support Vector Machine.

Additionally, screen time was measured by asking participants to report their daily average screen time from their smartphone screen time application. Thus, it is possible that there is some error in the reporting of this data. However, our large sample size likely helped reduce some of the risks around the influence of this measurement error. Future work, including objective measures of screen time, is needed. Furthermore, our participants were recruited from a specific region and age range and were not randomly selected. Therefore, the predictors of problematic smartphone use among young children remain unknown. At last, there might be other risk factors [74], such as parenting neglect [75] or low academic performance [76,77,78], that could also predict problematic smartphone use. Measuring these factors and including them in our regression models may have yielded different results. Thus, future studies should further investigate other potential risk factors of problematic smartphone use.

5. Conclusions

Despite these limitations, the findings of this study contribute to the existing body of knowledge about the risk factors associated with problematic smartphone use among Canadian adolescents during the COVID-19 pandemic. Specifically, findings from the current study suggest that FoMO, and emotional and cognitive self-regulation, may play important roles in contributing to adolescents’ problematic smartphone use in the time of the pandemic.

Author Contributions

All authors contributed to the study’s conception and design. Material preparation, data collection and analysis were performed by B.X., N.P. and J.S. L.R. conducted literature searches and provided summaries of previous research studies. The first draft of the manuscript was written by B.X. and N.P., and all authors commented on previous versions of the manuscript. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Approval was obtained from the ethics committee of The University of British Columbia. The procedures used in this study adhere to the tenets of the Declaration of Helsinki.

Informed Consent Statement

Informed consent was obtained from all individual participants included in the study.

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Tables

Table 1

Descriptive Statistics of Predictors.

Variable	M	SD	Min	Max	Cronbach’s α
Depression	3.88	0.91	1	5	0.91
Anxiety	4.44	0.76	1	5	0.83
Stress	3.14	1.28	1	5	0.86
Problematic smartphone use	2.40	0.81	1	6	0.82
FOMO	2.32	0.80	1	5	0.87
Emotional self-regulation	3.12	0.67	1	4	0.83
Behavioural self-regulation	2.44	0.81	1	4	0.77
Cognitive self-regulation	2.61	0.71	1	4	0.74
Screen time	369.09	175.82	60	1456	NA
Age	15.29	1.48	11	19	NA
Gender	0.73	0.72	0	3	NA
Ethnicity	2.34	0.85	1	3	NA

Note: Ethnicity was coded as East Asian (1), White (2), and others (3) (included participants who selected more than one ethnicity, Black, Indigenous, Latin American, South Asian, Southeast Asian, and West Asian). Gender was coded as girls (1), boys (0), and others (2); included participants who selected more than one gender, non-binary, transgender, not sure).

Table 2

Correlations among Main Study Variables.

	1	2	3	4	5	6	7	8	9
1. Problematic smartphone use	1
2. Screen time	0.35 **	1
3. Stress	0.40 **	0.21 **	1
4. Anxiety	0.30 **	0.21 **	0.63 **	1
5. Depression	0.34 **	0.22 **	0.67 **	0.51 **	1
6. FOMO	0.40 **	0.16 **	0.42 **	0.32 **	0.38 **	1
7. Emotional self-regulation	−0.36 **	−0.23 **	−0.53 **	−0.35 **	−0.44 **	−0.44 **	1
8. Behavioural self-regulation	−0.31 **	−0.18 **	−0.40 **	−0.32 **	−0.37 **	−0.31 **	0.42 **	1
9. Cognitive self-regulation	−0.19 **	−0.14 **	−0.12 **	−0.08 **	−0.16 **	−0.02	0.11 **	0.17 **	1

Note: ** p < 0.01.

Table 3

Comparison of Five Machine Learning-based Regression Algorithms.

	Mean Model Fit Findings over Repeated Cross-Validations in the Training Sample			Model Fit Findings in the Test Sample
	RMSE	MAE	R2	RMSE	MAE	R2
Lasso	0.663	0.523	0.36	0.656	0.519	0.34
Ridge	0.664	0.525	0.35	0.655	0.520	0.34
Elastic Net	0.663	0.524	0.35	0.655	0.519	0.34
Extreme Gradient Boosting	0.737	0.573	0.27	0.738	0.574	0.21
RandomForest	0.677	0.536	0.33	0.738	0.532	0.29

Note: RMSE = Root Mean Squared Error; MAE = Mean Absolute Error.

Table 4

Variable Importance Coefficients for the Lasso, Ridge and Elastic Net Machine Learning Algorithms in Predicting Problematic Smartphone Use.

Variable	Lasso Regression Method	Ridge Regression Method	Elastic Net Regression Method
Age	0.029	0.028	0.028
Gender	0.088	0.090	0.081
Ethnicity	0.030	0.032	0.030
Screen time	0.008	0.007	0.008
Depression	0.067	0.072	0.069
Anxiety	0.011	0.026	0.015
Stress	0.084	0.075	0.081
Emotional self-regulation	−0.038	−0.046	−0.042
Behavioural self-regulation	−0.112	−0.115	−0.114
Cognitive self-regulation	−0.155	−0.144	−0.151
FOMO	0.235	0.213	0.228

Note: Gender is coded “0 = boys,” and “1 = girls”; a positive coefficient indicates girls were associated with greater smartphone addiction.

References

1. Vannucci, A.; Simpson, E.G.; Gagnon, S.; Ohannessian, C.M. Social media use and risky behaviors in adolescents: A meta-analysis. J. Adolesc.; 2020; 79, pp. 258-274. [DOI: https://dx.doi.org/10.1016/j.adolescence.2020.01.014]

2. Michaelson, V.; Steeves, V. “I’ll use it differently now”: Using dual-systems theory to explore youth engagement with networked technologies. Can. J. Public Health; 2020; 111, pp. 1033-1040. [DOI: https://dx.doi.org/10.17269/s41997-020-00347-w] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32642970]

3. Shin, M.; Juventin, M.; Chu, J.T.W.; Manor, Y.; Kemps, E. Online media consumption and depression in young people: A systematic review and meta-analysis. Comput. Hum. Behav.; 2022; 128, pp. 107-129. [DOI: https://dx.doi.org/10.1016/j.chb.2021.107129]

4. Yam, F.C.; Korkmaz, O.; Griffiths, M.D. The association between fear of Covid-19 and smartphone addiction among individuals: The mediating and moderating role of cyberchondria severity. Curr. Psychol.; 2021; 42, pp. 2377-2390. [DOI: https://dx.doi.org/10.1007/s12144-021-02324-z] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34690474]

5. Masaeli, N.; Farhadi, H. Prevalence of Internet-based addictive behaviors during COVID-19 pandemic: A systematic review. J. Addict. Dis.; 2021; 39, pp. 468-488. [DOI: https://dx.doi.org/10.1080/10550887.2021.1895962]

6. Serra, G.; Scalzo, L.L.; Giuffrè, M.; Ferrara, P.; Corsello, G. Smartphone use and addiction during the coronavirus disease 2019 (COVID-19) pandemic: Cohort study on 184 Italian children and adolescents. Ital. J. Pediatr.; 2021; 47, 150. [DOI: https://dx.doi.org/10.1186/s13052-021-01102-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34215311]

7. Chotpitayasunondh, V.; Douglas, K.M. How “phubbing” becomes the norm: The antecedents and consequences of snubbing via smartphone. Comput. Hum. Behav.; 2016; 63, pp. 9-18. [DOI: https://dx.doi.org/10.1016/j.chb.2016.05.018]

8. Davazdahemami, B.; Hammer, B.; Soror, A. Addiction to mobile phone or addiction through mobile phone?. Proceedings of the 2016 49th Hawaii International Conference on System Sciences (HICSS); Washington, DC, USA, 5–8 January 2016; pp. 1467-1476. [DOI: https://dx.doi.org/10.1109/HICSS.2016.186]

9. de Freitas, B.H.B.M.; Gaíva, M.A.M.; Diogo, P.M.J.; Bortolini, J. Relationship between Lifestyle and Self-Reported Smartphone Addiction in adolescents in the COVID-19 pandemic: A mixed-methods study. J. Pediatr. Nurs.; 2022; 65, pp. 82-90. [DOI: https://dx.doi.org/10.1016/j.pedn.2022.03.001]

10. Horwood, S.; Anglim, J. Emotion regulation difficulties, personality, and problematic smartphone use. Cyberpsychology Behav. Soc. Netw.; 2021; 24, pp. 275-281. [DOI: https://dx.doi.org/10.1089/cyber.2020.0328]

11. Hughes, N.; Burke, J. Sleeping with the frenemy: How restricting ‘bedroom use’of smartphones impacts happiness and wellbeing. Comput. Hum. Behav.; 2018; 85, pp. 236-244. [DOI: https://dx.doi.org/10.1016/j.chb.2018.03.047]

12. Lee, C.; Lee, S.-J. Prevalence and predictors of smartphone addiction proneness among Korean adolescents. Child. Youth Serv. Rev.; 2017; 77, pp. 10-17. [DOI: https://dx.doi.org/10.1016/j.childyouth.2017.04.002]

13. Luo, T.; Chen, W.; Liao, Y. Social media use in China before and during COVID-19: Preliminary results from an online retrospective survey. J. Psychiatr. Res.; 2021; 140, pp. 35-38. [DOI: https://dx.doi.org/10.1016/j.jpsychires.2021.05.057] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34090101]

14. Ma, S.; Huang, Y.; Ma, Y. Childhood maltreatment and mobile phone addiction among Chinese adolescents: Loneliness as a mediator and self-control as a moderator. Front. Psychol.; 2020; 11, pp. 813-845. [DOI: https://dx.doi.org/10.3389/fpsyg.2020.00813] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32477211]

15. Sourander, A.; Klomek, A.B.; Ikonen, M.; Lindroos, J.; Helenius, H. Psychosocial risk factors associated with cyberbullying among adolescents: A population-based study. Arch. Gen. Psychiatry; 2010; 67, pp. 720-728. [DOI: https://dx.doi.org/10.1001/archgenpsychiatry.2010.79] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/20603453]

16. Emirtekin, E.; Balta, S.; Sural, İ.; Kircaburun, K.; Griffiths, M.D.; Billieux, J. The role of childhood emotional maltreatment and body image dissatisfaction in problematic smartphone use among adolescents. Psychiatry Res.; 2019; 271, pp. 634-639. [DOI: https://dx.doi.org/10.1016/j.psychres.2018.12.059]

17. Billieux, J. Problematic use of the mobile phone: A literature review and a pathways model. Curr. Psychiatry Rev.; 2012; 8, pp. 299-307. [DOI: https://dx.doi.org/10.2174/157340012803520522]

18. Contractor, A.A.; Weiss, N.H.; Tull, M.T.; Elhai, J.D. PTSD’s relation with problematic smartphone use: Mediating role of impulsivity. Comput. Hum. Behav.; 2017; 75, pp. 177-183. [DOI: https://dx.doi.org/10.1016/j.chb.2017.05.018]

19. Pivetta, E.; Harkin, L.; Billieux, J.; Kanjo, E.; Kuss, D.J. Problematic smartphone use: An empirically validated model. Comput. Hum. Behav.; 2019; 100, pp. 105-117. [DOI: https://dx.doi.org/10.1016/j.chb.2019.06.013]

20. Elhai, J.D.; Yang, H.; Levine, J.C. Applying fairness in labeling various types of internet use disorders.•: Commentary on How to overcome taxonomical problems in the study of internet use disorders and what to do with “smartphone addiction”?. J. Behav. Addict.; 2021; 9, pp. 924-927. [DOI: https://dx.doi.org/10.1556/2006.2020.00071]

21. Bian, M.; Leung, L. Linking loneliness, shyness, smartphone addiction symptoms, and patterns of smartphone use to social capital. Soc. Sci. Comput. Rev.; 2015; 33, pp. 61-79. [DOI: https://dx.doi.org/10.1177/0894439314528779]

22. Durak, H.Y. Investigation of nomophobia and smartphone addiction predictors among adolescents in Turkey: Demographic variables and academic performance. Soc. Sci. J.; 2019; 56, pp. 492-517. [DOI: https://dx.doi.org/10.1016/j.soscij.2018.09.003]

23. Fatkuriyah, L.; Sun-Mi, C. The Relationship among Parenting Style, Self-Regulation, and Smartphone Addiction Proneness in Indonesian Junior High School Students. IJNP (Indones. J. Nurs. Pract.); 2021; 5, pp. 51-59. [DOI: https://dx.doi.org/10.18196/ijnp.v5i1.11186]

24. Hosen, I.; Al Mamun, F.; Sikder, M.T.; Abbasi, A.Z.; Zou, L.; Guo, T.; Mamun, M.A. Prevalence and associated factors of problematic smartphone use during the COVID-19 pandemic: A Bangladeshi study. Risk Manag. Healthc. Policy; 2021; 14, pp. 3797-3805. [DOI: https://dx.doi.org/10.2147/RMHP.S325126]

25. Noë, B.; Turner, L.D.; Linden, D.E.; Allen, S.M.; Winkens, B.; Whitaker, R.M. Identifying indicators of smartphone addiction through user-app interaction. Comput. Hum. Behav.; 2019; 99, pp. 56-65. [DOI: https://dx.doi.org/10.1016/j.chb.2019.04.023] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31582873]

26. Busch, P.A.; McCarthy, S. Antecedents and consequences of problematic smartphone use: A systematic literature review of an emerging research area. Comput. Hum. Behav.; 2021; 114, 106414. [DOI: https://dx.doi.org/10.1016/j.chb.2020.106414]

27. Brand, M.; Wegmann, E.; Stark, R.; Müller, A.; Wölfling, K.; Robbins, T.W.; Potenza, M.N. The Interaction of Person-Affect-Cognition-Execution (I-PACE) model for addictive behaviors: Update, generalization to addictive behaviors beyond internet-use disorders, and specification of the process character of addictive behaviors. Neurosci. Biobehav. Rev.; 2019; 104, pp. 1-10. [DOI: https://dx.doi.org/10.1016/j.neubiorev.2019.06.032] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31247240]

28. Cohen, Z.P.; Cosgrove, K.T.; DeVille, D.C.; Akeman, E.; Singh, M.K.; White, E.; Stewart, J.L.; Aupperle, R.L.; Paulus, M.P.; Kirlic, N. The impact of COVID-19 on adolescent mental health: Preliminary findings from a longitudinal sample of healthy and at-risk adolescents. Front. Pediatr.; 2021; 9, pp. 440-451. [DOI: https://dx.doi.org/10.3389/fped.2021.622608]

29. Islam, M.S.; Ferdous, M.Z.; Potenza, M.N. Panic and generalized anxiety during the COVID-19 pandemic among Bangladeshi people: An online pilot survey early in the outbreak. J. Affect. Disord.; 2020; 276, pp. 30-37. [DOI: https://dx.doi.org/10.1016/j.jad.2020.06.049]

30. Meherali, S.; Adewale, B.; Ali, S.; Kennedy, M.; Salami, B.; Richter, S.; Okeke-Ihejirika, P.E.; Ali, P.; da Silva, K.L.; Adjorlolo, S. et al. Impact of the COVID-19 pandemic on adolescents’ sexual and reproductive health in low-and middle-income countries. Int. J. Environ. Res. Public Health; 2021; 18, 13221. [DOI: https://dx.doi.org/10.3390/ijerph182413221]

31. Jin, L.; Hao, Z.; Huang, J.; Akram, H.R.; Saeed, M.F.; Ma, H. Depression and anxiety symptoms are associated with problematic smartphone use under the COVID-19 epidemic: The mediation models. Child. Youth Serv. Rev.; 2021; 121, 105875. [DOI: https://dx.doi.org/10.1016/j.childyouth.2020.105875]

32. Kara, M.; Baytemir, K.; Inceman-Kara, F. Duration of daily smartphone usage as an antecedent of nomophobia: Exploring multiple mediation of loneliness and anxiety. Behav. Inf. Technol.; 2021; 40, pp. 85-98. [DOI: https://dx.doi.org/10.1080/0144929X.2019.1673485]

33. Gao, Q.; Sun, R.; Fu, E.; Jia, G.; Xiang, Y. Parent–child relationship and smartphone use disorder among Chinese adolescents: The mediating role of quality of life and the moderating role of educational level. Addict. Behav.; 2020; 101, 106065. [DOI: https://dx.doi.org/10.1016/j.addbeh.2019.106065] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31543219]

34. Király, O.; Potenza, M.N.; Stein, D.J.; King, D.L.; Hodgins, D.C.; Saunders, J.B.; Griffiths, M.D.; Gjoneska, B.; Billieux, J.; Brand, M. et al. Preventing problematic internet use during the COVID-19 pandemic: Consensus guidance. Compr. Psychiatry; 2020; 100, 152180. [DOI: https://dx.doi.org/10.1016/j.comppsych.2020.152180] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32422427]

35. Novak, S.P.; Clayton, R.R. The influence of school environment and self-regulation on transitions between stages of cigarette smoking: A multilevel analysis. Health Psychol.; 2001; 20, pp. 196-207. [DOI: https://dx.doi.org/10.1037/0278-6133.20.3.196] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11403217]

36. Chang, F.C.; Chiu, C.H.; Chen, P.H.; Chiang, J.T.; Miao, N.F.; Chuang, H.Y.; Liu, S. Children’s use of mobile devices, smartphone addiction and parental mediation in Taiwan. Comput. Hum. Behav.; 2019; 93, pp. 25-32. [DOI: https://dx.doi.org/10.1016/j.chb.2018.11.048]

37. KiKim, Y.; Jeong, J.-E.; Cho, H.; Jung, D.-J.; Kwak, M.; Rho, M.J.; Yu, H.; Kim, D.-J.; Choi, I.Y. Personality factors predicting smartphone addiction predisposition: Behavioral inhibition and activation systems, impulsivity, and self-control. PloS ONE; 2016; 11, e0159788. [DOI: https://dx.doi.org/10.1371/journal.pone.0159788]

38. Duckworth, A.L.; Kirby, T.A.; Tsukayama, E.; Berstein, H.; Ericsson, K.A. Deliberate practice spells success: Why grittier competitors triumph at the National Spelling Bee. Soc. Psychol. Personal. Sci.; 2011; 2, pp. 174-181. [DOI: https://dx.doi.org/10.1177/1948550610385872]

39. Elhai, J.D.; Tiamiyu, M.F.; Weeks, J.W.; Levine, J.C.; Picard, K.J.; Hall, B.J. Depression and emotion regulation predict objective smartphone use measured over one week. Personal. Individ. Differ.; 2018; 133, pp. 21-28. [DOI: https://dx.doi.org/10.1016/j.paid.2017.04.051]

40. Shapka, J.D. Adolescent technology engagement: It is more complicated than a lack of self-control. Hum. Behav. Emerg. Technol.; 2019; 1, pp. 103-110. [DOI: https://dx.doi.org/10.1002/hbe2.144]

41. Baumeister, R.F.; Leary, M.R. The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychol. Bull.; 1995; 117, pp. 497-529. [DOI: https://dx.doi.org/10.1037/0033-2909.117.3.497]

42. Przybylski, A.K.; Murayama, K.; DeHaan, C.R.; Gladwell, V. Motivational, emotional, and behavioral correlates of fear of missing out. Comput. Hum. Behav.; 2013; 29, pp. 1841-1848. [DOI: https://dx.doi.org/10.1016/j.chb.2013.02.014]

43. Parent, N.; Dadgar, K.; Xiao, B.; Hesse, C.; Shapka, J.D. Social disconnection during COVID-19: The role of attachment, fear of missing out, and smartphone use. J. Res. Adolesc.; 2021; 31, pp. 748-763. [DOI: https://dx.doi.org/10.1111/jora.12658] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34448308]

44. Fabris, M.A.; Marengo, D.; Longobardi, C.; Settanni, M. Investigating the links between fear of missing out, social media addiction, and emotional symptoms in adolescence: The role of stress associated with neglect and negative reactions on social media. Addict. Behav.; 2020; 106, 106364. [DOI: https://dx.doi.org/10.1016/j.addbeh.2020.106364] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32145495]

45. Swan, A.J.; Kendall, P.C. Fear and missing out: Youth anxiety and functional outcomes. Clin. Psychol. Sci. Pract.; 2016; 23, pp. 417-435. [DOI: https://dx.doi.org/10.1111/cpsp.12169]

46. Elhai, J.D.; Yang, H.; Rozgonjuk, D.; Montag, C. Using machine learning to model problematic smartphone use severity: The significant role of fear of missing out. Addict. Behav.; 2020; 103, 106261. [DOI: https://dx.doi.org/10.1016/j.addbeh.2019.106261] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31901886]

47. Elhai, J.D.; Gallinari, E.F.; Rozgonjuk, D.; Yang, H. Depression, anxiety and fear of missing out as correlates of social, non-social and problematic smartphone use. Addict. Behav.; 2020; 105, 106335. [DOI: https://dx.doi.org/10.1016/j.addbeh.2020.106335]

48. Elhai, J.D.; Levine, J.C.; Alghraibeh, A.M.; Alafnan, A.A.; Aldraiweesh, A.A.; Hall, B.J. Fear of missing out: Testing relationships with negative affectivity, online social engagement, and problematic smartphone use. Comput. Hum. Behav.; 2018; 89, pp. 289-298. [DOI: https://dx.doi.org/10.1016/j.chb.2018.08.020]

49. Servidio, R.; Sinatra, M.; Griffiths, M.D.; Monacis, L. Social comparison orientation and fear of missing out as mediators between self-concept clarity and problematic smartphone use. Addict. Behav.; 2021; 12291, 107014. [DOI: https://dx.doi.org/10.1016/j.addbeh.2021.107014]

50. Sha, P.; Sariyska, R.; Riedl, R.; Lachmann, B.; Montag, C. Linking internet communication and smartphone use disorder by taking a closer look at the Facebook and WhatsApp applications. Addict. Behav. Rep.; 2019; 9, 100148. [DOI: https://dx.doi.org/10.1016/j.abrep.2018.100148]

51. Yang, H.; Liu, B.; Fang, J. Stress and problematic smartphone use severity: Smartphone use frequency and fear of missing out as mediators. Front. Psychiatry; 2021; 12, pp. 594-623. [DOI: https://dx.doi.org/10.3389/fpsyt.2021.659288]

52. Coco, G.L.; Salerno, L.; Franchina, V.; La Tona, A.; Di Blasi, M.; Giordano, C. Examining bi-directionality between Fear of Missing Out and problematic smartphone use. A two-wave panel study among adolescents. Addict. Behav.; 2020; 106, 106360. [DOI: https://dx.doi.org/10.1016/j.addbeh.2020.106360] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32135397]

53. Casale, S.; Flett, G.L. Interpersonally-based fears during the COVID-19 pandemic: Reflections on the fear of missing out and the fear of not mattering constructs. Clin. Neuropsychiatry; 2020; 17, pp. 88-102. [DOI: https://dx.doi.org/10.36131/CN20200211] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34908975]

54. Alinejad, V.; Parizad, N.; Yarmohammadi, M.; Radfar, M. Loneliness and academic performance mediates the relationship between fear of missing out and smartphone addiction among Iranian university students. BMC Psychiatry; 2022; 22, 550. [DOI: https://dx.doi.org/10.1186/s12888-022-04186-6]

55. Hodes, L.N.; Thomas, K.G. Smartphone screen time: Inaccuracy of self-reports and influence of psychological and contextual factors. Comput. Hum. Behav.; 2021; 115, 106616. [DOI: https://dx.doi.org/10.1016/j.chb.2020.106616]

56. Bae, S.M. The relationship between the type of smartphone use and smartphone dependence of Korean adolescents: National survey study. Child. Youth Serv. Rev.; 2017; 81, pp. 207-211. [DOI: https://dx.doi.org/10.1016/j.childyouth.2017.08.012]

57. Rozgonjuk, D.; Levine, J.C.; Hall, B.J.; Elhai, J.D. The association between problematic smartphone use, depression and anxiety symptom severity, and objectively measured smartphone use over one week. Comput. Hum. Behav.; 2018; 87, pp. 10-17. [DOI: https://dx.doi.org/10.1016/j.chb.2018.05.019]

58. Anshari, M.; Alas, Y.; Sulaiman, E. Smartphone addictions and nomophobia among youth. Vulnerable Child. Youth Stud.; 2019; 14, pp. 242-247. [DOI: https://dx.doi.org/10.1080/17450128.2019.1614709]

59. Anshari, M.; Alas, Y.; Hardaker, G.; Jaidin, J.H.; Smith, M.; Ahad, A.D. Smartphone habit and behavior in Brunei: Personalization, gender, and generation gap. Comput. Hum. Behav.; 2016; 64, pp. 719-727. [DOI: https://dx.doi.org/10.1016/j.chb.2016.07.063]

60. Cocoradă, E.; Maican, C.I.; Cazan, A.M.; Maican, M.A. Assessing the smartphone addiction risk and its associations with personality traits among adolescents. Child. Youth Serv. Rev.; 2018; 93, pp. 345-354. [DOI: https://dx.doi.org/10.1016/j.childyouth.2018.08.006]

61. Alpaydin, E. Machine Learning: The New AI; MIT Press: Cambridge, MA, USA, 2016.

62. Kelleher, A.; Kelleher, A. Machine Learning in Production: Developing and Optimizing Data Science Workflows and Applications; Addison-Wesley Professional: Boston, MA, USA, 2019.

63. Pokorny, S.B.; Jason, L.A.; Schoeny, M.E.; Townsend, S.M.; Curie, C.J. Do participation rates change when active consent procedures replace passive con-sent. Eval. Rev.; 2001; 25, pp. 567-580. [DOI: https://dx.doi.org/10.1177/0193841X0102500504]

64. Shaw, T.; Cross, D.; Thomas, L.T.; Zubrick, S.R. Bias in student survey findings from active par- ental consent procedures. Br. Educ. Res. J.; 2015; 41, pp. 229-243. [DOI: https://dx.doi.org/10.1002/berj.3137]

65. Kwon, M.; Kim, D.J.; Cho, H.; Yang, S. The smartphone addiction scale: Development and validation of a short version for adolescents. PloS ONE; 2013; 8, e83558. [DOI: https://dx.doi.org/10.1371/journal.pone.0083558]

66. De Pasquale, C.; Sciacca, F.; Hichy, Z. Italian validation of smartphone addiction scale short version for adolescents and young adults (SAS-SV). Psychology; 2017; 8, pp. 1513-1518. [DOI: https://dx.doi.org/10.4236/psych.2017.810100]

67. Lovibond, P.F.; Lovibond, S.H. The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav. Res. Ther.; 1995; 33, pp. 335-343. [DOI: https://dx.doi.org/10.1016/0005-7967(94)00075-U] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/7726811]

68. Zanon, C.; Brenner, R.E.; Baptista, M.N.; Vogel, D.L.; Rubin, M.; Al-Darmaki, F.R.; Gonçalves, M.; Heath, P.J.; Liao, H.-Y.; MacKenzie, C.S. et al. Examining the dimensionality, reliability, and invariance of the Depression, Anxiety, and Stress Scale–21 (DASS-21) across eight countries. Assessment; 2021; 28, pp. 1531-1544. [DOI: https://dx.doi.org/10.1177/1073191119887449]

69. Hastie, T.; Tibshirani, R.; Friedman, J.H.; Friedman, J.H. The Elements of Statistical Learning: Data Mining, Inference, and Prediction; Springer: New York, NY, USA, 2009; Volume 2, pp. 1-758.

70. Hamilton, J.L.; Nesi, J.; Choukas-Bradley, S. Teens and social media during the COVID-19 pandemic: Staying socially connected while physically distant. Perspect. Psychol. Sci.; 2020; in press

71. Baek, H.W.; Shin, Y.M.; Shin, K.M. Emotional and behavioral problems related to smartphone overuse in elementary school children. J. Korean Neuropsychiatr. Assoc.; 2014; 53, pp. 320-326. [DOI: https://dx.doi.org/10.4306/jknpa.2014.53.5.320]

72. Enez Darcin, A.; Kose, S.; Noyan, C.O.; Nurmedov, S.; Yılmaz, O.; Dilbaz, N. Smartphone addiction and its relationship with social anxiety and loneliness. Behav. Inf. Technol.; 2016; 35, pp. 520-525. [DOI: https://dx.doi.org/10.1080/0144929X.2016.1158319]

73. Poulain, T.; Vogel, M.; Kliesener, T.; Kiess, W. Associations between changes in behavioral difficulties and levels of problematic smartphone use in adolescents over a 1-year period. Eur. Child Adolesc. Psychiatry; 2021; 32, pp. 533-536. [DOI: https://dx.doi.org/10.1007/s00787-021-01874-8]

74. Roberts, J.A.; Pullig, C.; Manolis, C. I need my smartphone: A hierarchical model of personality and cell-phone addiction. Personal. Individ. Differ.; 2015; 79, pp. 13-19. [DOI: https://dx.doi.org/10.1016/j.paid.2015.01.049]

75. Kwak, J.Y.; Kim, J.Y.; Yoon, Y.W. Effect of parental neglect on smartphone addiction in adolescents in South Korea. Child Abus. Negl.; 2018; 77, pp. 75-84. [DOI: https://dx.doi.org/10.1016/j.chiabu.2017.12.008]

76. Lee, Y.K.; Chang, C.T.; Lin, Y.; Cheng, Z.H. The dark side of smartphone usage: Psychological traits, compulsive behavior and technostress. Comput. Hum. Behav.; 2014; 31, pp. 373-383. [DOI: https://dx.doi.org/10.1016/j.chb.2013.10.047]

77. Lee, Z.H.; Chen, I.H. The association between problematic internet use, psychological distress, and sleep problems during COVID-19. Sleep Epidemiol.; 2021; 1, 100005. [DOI: https://dx.doi.org/10.1016/j.sleepe.2021.100005] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35673626]

78. Samaha, M.; Hawi, N.S. Relationships among smartphone addiction, stress, academic performance, and satisfaction with life. Comput. Hum. Behav.; 2016; 57, pp. 321-325. [DOI: https://dx.doi.org/10.1016/j.chb.2015.12.045]