Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study

Laura Marciano; Anne-Linda Camerini

doi:10.1371/journal.pone.0263815

. 2022 Feb 18;17(2):e0263815. doi: 10.1371/journal.pone.0263815

Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study

Laura Marciano ^1,^*, Anne-Linda Camerini ¹

Editor: Michael Kaess²

PMCID: PMC8856513 PMID: 35180248

Abstract

Problematic smartphone use (PSU) during adolescence has been associated with negative short- and long-term consequences for personal well-being and development. Valid and reliable predictors and indicators of PSU are urgently needed, and digital trace data can add valuable information beyond self-report data. The present study aimed to investigate whether trace data (duration and frequency of smartphone use), recorded via an app installed on participants’ smartphone, are correlated with self-report data on smartphone use. Additionally, the present study aimed to explore which usage indicators, i.e., duration, frequency, and time distortion of smartphone use, better predict PSU levels cross-sectionally and longitudinally, one year later. Results from a sample of 84 adolescents showed that adolescents tend to rely on the frequency of smartphone use when reporting on the time they spent with the smartphone. Traced duration of smartphone use as well as time distortion, i.e., over-estimation, are significant predictors of PSU. Methodological issues and theoretical implications related to predictors and indicators of PSU are discussed.

Introduction

The rise of digital media has sparked numerous studies on adolescents’ use of these media, frequently focusing on the associations with developmental and well-being outcomes [1–7]. A large body of previous studies is concerned with the negative aspects of too much media use, especially adolescents’ dependency on digital devices like smartphones. In fact, the smartphone has become an indispensable part of most adolescents’ life, and its use is driven by the core motivations and needs of that age [8]. According to a recent review, one out of four adolescents shows symptoms of Problematic Smartphone Use (PSU) [5], which negatively affects social and emotional well-being [9] and academic outcomes [10]. However, until today, most of the studies focusing on PSU relied on self-report data making it hard to draw valid conclusions on which type of smartphone activity can be described as problematic [11]. Yet, technological advancements have enabled the use of digital trace data–defined as "records of activity undertaken through an online information system—thus, digital" [12] ^p767 –in addition to or even as a substitute for self-report data. To date, few studies made use of trace data to investigate PSU, focusing on adult populations [for a review, see 13]. The incorporation of digital trace data in research with adolescents is still scarce [14–16]. Therefore, the present study wanted to fill the gap and investigated adolescents’ PSU by considering both self-report and digital trace data of smartphone use at the cross-sectional and longitudinal level. This approach shed light on the behavioral patterns that lead to the development of PSU, thus informing educational interventions that promote a smart use of the smartphone in adolescence.

Literature review

Problematic smartphone use

Smartphones have become the preferred digital media device among adolescents. In the U.S., 95% of teens have access to a smartphone, through which they mostly pass time, connect with others, learn new things, and avoid face-to-face interactions [17]. In Western Europe, prevalence rates are similar. In Switzerland, 99% of 12 to 19-year-olds own a smartphone, on which they spend, on average, almost 4 hours during a weekday and more than 5 hours during a weekend day [13]. Early adolescents (12 to 13 years) spend more time playing online games, while they make less use of the smartphone for instant messaging, social media, online searches, and activities like listening to music and creating multimedia material. On the contrary, older adolescents use the smartphone more often for social media, photo and video editing, time management, e-mail, navigation, and vocal assistance [13]. In 2020, the amount of time spent on the device reached the highest increment in the last ten years, also due to the COVID-19 pandemic and associated social distancing and confinement measures [13], thus making it even more imperative to comprehend PSU objectively. The easiness of its portability, the ubiquitous accessibility, and the multitude of personalized Internet-based applications on the smartphone potentially facilitate the development of PSU. Real-time data showed that youth pick up their smartphone every five minutes, thus making smartphone use a deeply internalized, unconscious, and reinforced behavior [18]. As reported by Sohn and colleagues [5], one in four children and young people shows signs of PSU. According to the same review, PSU is consistently associated with higher levels of depression, anxiety, stress, sleep problems, and daily functional impairment, including poor academic achievement. Similar results have been found in other studies, where PSU was linked to internalizing symptoms, risk of cyberbullying behaviors, higher distractibility, and body image concerns, to name a few [10,19,20].

Although a standard agreement on what constitutes PSU does not yet exist, it has been compared to the similar and better-researched concept of Internet Addiction (IA). PSU and IA share similarities with the more general category of "behavioral addictions", introduced in 2010 by the DSM-5 Working Group after decades of debate [21,22]. Behavioral conducts carried out in an extreme way generate substantial problems for a person’s well-being, independently of the nature of the specific activity. The repetitive and compulsive engagement in smartphone-related activities shares common characteristics with substance use disorders and gambling disorder (the first disorder included as a non-substance behavioral addiction in the DSM-5). In the case of PSU, these characteristics comprise cognitive salience, spending a lot of time using the smartphone, unsuccessful attempts to reduce smartphone use, positive mood when using the device and symptoms like irritability and distress when one is unable to reach it, difficulty in regulating the use, unsuccessful attempts to stop, and interpersonal problems (with family or friends, at work or school) due to excessive smartphone use [23–28]. PSU is also closely linked to "nomophobia", which refers to discomfort, anxiety, nervousness or anguish caused by being without a mobile phone [29]. To summarize, although diverse terminologies and measures exist [30], PSU has been generally defined akin a behavioral or non-substance related addiction. However, it is important to consider that smartphones are linked to different (social) rewards, such as positive feedback and validation by peers [8]. This makes adolescents use the smartphone many times a day without necessarily reflecting an addictive behavior [31–33]. That said, self-report measures can only partially depict different features of PSU.

Overcoming self-report biases

It is commonly acknowledged that self-report is subject to systematic biases, which include, among others, recall, estimation, and social desirability bias [34–36]. Recall bias, for example, is the result of cognitive burden and occurs when respondents use heuristic shortcuts to recall the duration and frequency of everyday behaviors [37]. Furthermore, problems with time estimation are frequent, especially among younger populations who still have to develop a sense of time and the ability to quantify the time they engaged in different activities and to report on it [38]. On the other hand, social desirability bias is a systematic bias across all age. It describes the tendency of respondents to inaccurately report on sensitive topics to present themselves in the best possible light [39]. Self-report biases are a recognized methodological problem in different domains, including health sciences [40], research on organisations [41], consumer marketing [42], and media studies [43]. Concerning smartphone use, a recent review and meta-analysis underlined that, although several self-report scales correlate with objectively logged data, the strength of the relationship is far from convincing [44]. To this regard, the inclusion of objective trace data would complement the assessment of PSU.

Thanks to technological advancements, automated recording tools, such as tracking applications [45], are now available to overcome these biases [for an introduction, see 46], and they represent an excellent potential in incorporating digital trace data in survey research. However, the use of tracking applications to assess smartphone use is still at the beginning [46–48], and researchers pursuing the collection and analysis of trace data face different challenges. In their overview paper on integrating survey and digital trace data, Stier and colleagues [47] highlighted three potential issues: Firstly, as with any study, the collection of trace data requires informed consent from individuals who may have objections about automated tracking of their digital media behaviors. Secondly, technical and methodological issues can cause problems in the collection and scientific analysis of digital trace data alone and in combination with survey data. Thirdly, researchers need to improve the conceptual and theoretical frameworks for dealing with such data’s richness. A fourth challenge can be added to the list, namely, legal restrictions. For example, the European Union does not allow collecting shared digital media content with identifiable information about depicted persons without the active consent of all parties represented in this content (i.e., third-party consent) [49].

Despite these challenges in collecting digital trace data, researchers have started to trace smartphone use as an increasing phenomenon among younger populations across different societies [50–52]. Through smartphones, automated recording can happen in very different ways, for example, by tracking users via dedicated applications installed on the device [47] or via a software-modified smartphone to record the number and duration of calls [53]. For example, Mireku and colleagues [14] examined the similarity between self-reported mobile phone use data and objective mobile operator traffic data in 11–12 years old participants, focusing on calls and text messages. The number and duration of calls done by adolescents were also assessed by Inyang and colleagues [15] and Aydin and colleagues [16]. In addition, Reeves and colleagues [51,54] developed a technology based on a series of screenshots of the user’s mobile device to create what the authors call a ’screenome’ of digital media use.

Investigating PSU using objective trace data

Since the smartphone’s excessive use is an indicator of PSU [23], researchers have started to investigate the frequency and duration of smartphone use by incorporating objective trace data in their statistical analyses [55]. When it comes to the exploration of behavioral patterns of PSU, Lin and colleagues proposed diagnostic criteria for what the authors called "smartphone addiction" based on digital trace data and clinical interviews in a sample of adult participants [50]. Additionally, in another study [56], they found that the presence of underestimation (i.e., the discrepancy between traced duration of device use and self-reported use) and the traced frequency of smartphone use were better predictors of psychiatrists’ rating of smartphone addiction than the traced duration of device use. This result stands in partial contrast to a previous research conducted by Ko and colleagues [57], who found that self-report excessive time spent online (i.e. duration of use) was a risk factor of problematic Internet use as well as PSU. Furthermore, Tossell and colleagues [58] compared smartphone use indicators among young adults grouped into addicted and non-addicted smartphone users according to their self-assessment. They found that the self-declared addicts spent twice the time on their phones and started interacting with applications (especially those meant for social interactions) twice as often as the non-addicts. Likewise, a study by Noë and colleagues [59] revealed that user-interface interactions, especially if they involve lifestyle and social media applications, were related to higher PSU levels in a sample of young adults. To the best of our knowledge, and according to a recent review summarizing objective measures used to assess PSU [55], no study to date has combined self-report and digital trace data to predict PSU in mid adolescents cross-sectionally, let alone longitudinally.

Study aim

To fill this gap, the present study aimed to identify valid and reliable indicators of PSU in adolescence over time by using self-report and objective trace data on smartphone use. From a methodological point of view, the study wanted to explore to what extent self-report data are an accurate estimate of objective digital trace data. From a theoretical point of view, the study aimed to investigate which indicators among traced duration of smartphone use, traced frequency of use, and time distortion (i.e., the discrepancy between self-report and traced use), best predict PSU cross-sectionally and longitudinally, one year later. Based on the findings from previous studies with adult populations, we derived the following hypotheses:

H1: Longer duration of traced smartphone use is related to higher levels of PSU.
H2: Time distortion of smartphone use is related to higher levels of PSU.
H3: Higher frequency of traced smartphone use is related to higher levels of PSU.

Additionally, we added the following two research questions:

RQ1: How much does self-report smartphone use correlate with traced duration of smartphone use?
RQ2: Is there a difference in the correlation between self-report duration and traced duration when comparing smartphone use for weekdays and weekend days?

Methodology

The present study is part of the longitudinal MEDIATICINO panel study (www.mediaticino.usi.ch), focusing on digital media use and youth well-being. Since 2014, the study has followed a cohort of approximately 1’400 adolescents born in 2004/05 and residing in Canton Ticino, Italian-speaking Switzerland.

Data collection

Survey data

The larger panel study relies on an annual self-administered paper-and-pencil questionnaire to collect information about adolescents’ use of digital media (including duration and type of use), physical, psychological, and social well-being, as well as on different aspects of the parent-child relationship. Each year, survey data are collected in collaboration with schools, who distribute the questionnaire using a Unique Identifier (U-ID) and the associated student name, to which only school staff has access. The U-ID is used to match different waves and assure participants’ anonymity during data processing and analysis by the research team. Further information on the data collection procedure can be found elsewhere [60].

Digital trace data

In 2018 and 2019, the panel study was extended by introducing an automated way to collect data on smartphone use, i.e., by recording data of adolescents’ smartphone use via a dedicated application installed on their device. All the panel study participants were invited to participate through a letter forwarded by the collaborating schools to students’ families. Participants who provided parental consent received further information on how to download the application and register for the study. The application, called Ethica, was specially developed for public health research purposes for Android and iOS operating systems (ethicadata.com). Ethica automatically collects trace data, such as screen time and application usage. To match trace data with self-report questionnaire data, participants in the Ethica study received a generated login e-mail address matched to their U-ID. Schools distributed the instruction material with information on how to download Ethica and register for the study within the application by using the generated login e-mail address and a personal password. This procedure allowed the combination of digital trace data and questionnaire data while guaranteeing students’ anonymity. Students provided their active consent directly in the Ethica application upon enrolment.

Ethical considerations

The Cantonal education administration of Ticino approved the annual panel study based on self-administered questionnaires. The embedded Ethica study received additional approvals from the Ethics Committee of the Università della Svizzera italiana and from the Cantonal Data Protection Officer of Ticino.

Sample

In 2018, the larger panel included 1419 students, of which 1374 (96.8%) completed the paper-and-pencil questionnaire at school. For 264 (18.6%) students, parents provided informed consent to invite their children to the Ethica study. Despite parental consent, 169 (64%) students did not download the Ethica application. The remaining 95 students (6.7% of the initial sample) eventually participated in the Ethica study in 2018. Compared to the students who did not join the Ethica study, included participants did not differ in gender (p = .205), perceived socioeconomic status (p = .229), or self-reported daily smartphone use (p = .114) [61]. Due to technical problems and some participant dropouts, reliable trace data were available for 89 participants. When combining digital trace data and self-report data, we eliminated other five cases with missing values on self-report data. The analytical sample was composed of 84 participants with complete and matched data for T1 (M_age = 13.56, SD_age = .52, 46.4% males), of which 80 participants also had complete survey data for T2.

Measures

Perceived duration of smartphone use

Self-report smartphone use was measured in the annual survey with two questions: "How much time do you usually use the smartphone on a typical school day/weekend day?". For each question, students estimated their daily smartphone use by choosing one option on a scale with nine-time interval: 0 "never", 1 "up to 0.5 hours", 2 "between 0.5 and 1 hour", 3 "between 1 and 1.5 hours", 4 "between 1.5 and 2 hours", 5 "between 2 and 3 hours", 6, "between 3 and 4 hours", 7 "between 4 and 5 hours", and 8 "5 or more hours". To allow the comparison with trace data, results were converted into hours by using the midpoint for each category of the original interval scale: (0 = 0) (1 = 0.25) (2 = 0.75) (3 = 1.25) (4 = 1.75) (5 = 2.5) (6 = 3.5) (7 = 4.5) (8 = 5.5). For the highest interval (“5 or more hours”), we used 5.5 hours as a proxy of time spent with the smartphone [62]. The weighted mean between a typical school day and a typical weekend day was used as an approximate measure of smartphone use in terms of hours per day [(estimation of a typical weekday*5+estimation of a typical weekend day*2)/7] (M = 1.88, SD = 1.41, r between weekday and weekend day use = .90, p < .001).

Trace data for the duration of smartphone use

The traced duration was divided into hours on weekdays and weekend days. All weekdays were averaged to obtain an aggregate measure for typical weekday use, and all weekend days were averaged to obtain an aggregate measure for typical weekend day use. Once again, the weighted mean between a typical weekday and a typical weekend day was used as an approximate measure of smartphone use in terms of hours per day [(traced duration for a typical weekday*5+traced duration for a typical weekend day*2)/7] (M = 1.89, SD = 1.52, r between weekday and weekend day use = .855, p < .001). Missing data were handled with Hidden Markov Models [for more details, see 63].

Trace data for frequency of smartphone use

The frequency of smartphone use, i.e., checking behavior, was assessed by automatically counting how many times participants activated their smartphone screen during a weekday and a weekend day. Applying the abovementioned formula, we obtained an average measure of checking behavior per day weighted for weekdays and weekend days [(traced frequency for a typical weekday*5+traced frequency for a typical weekend day*2)/7] (M = 57.46, SD = 38.07, r between weekday and weekend day use = .888, p < .001)

Problematic smartphone use

Perceived PSU was assessed with the short version of the Smartphone Addiction Scale for adolescents (SAS-SV; 62) at both T1 in spring 2018 and T2 in spring 2019. The scale consists of ten items measured on a scale from 1 "strongly disagree" to 6 "strongly agree". All items were averaged to obtain an overall measure for each time point with higher values indicating higher levels of PSU (M_T1 = 1.86, SD_T1 = .85, M_T2 = 1.94, SD_T2 = .86). In line with the work by Kwon and colleagues [64], the SAS-SV showed good levels of reliability at both waves (α_T1 = .89 and α_T2 = .89). The original items and the Italian translation of the scale can be found in the S3 Table.

Social desirability

Given that PSU was assessed through self-report, we accounted for adolescents’ tendency to provide socially desirable answers in our multivariate analyses. Social desirability was measured at T1 with eight items from the Italian version of the Children’s Social Desirability Short Scale [60]. Items were assessed on a 5-point scale ranging from 1 "never" to 5 "always". All items were reverse coded and averaged to obtain an overall measure of social desirability with higher values indicating higher social desirability levels (M = 2.64, SD = .70, α = .84).

Analytical plan

After computing descriptive statistics for each smartphone use indicator, we, first, calculated the correlations between self-report data and trace data by considering both duration and frequency of smartphone use for a general day, weekdays, and weekend days. Since the data were not normally distributed, variables were log-transformed before they were entered in the analyses. A Bonferroni correction was applied when self-report smartphone use was compared to traced duration and frequency of use (p ≤ .017).

Second, to obtain a measure of time distortion, we calculated a difference index Δ by subtracting self-report duration of use from automatically recorded use. Hence, negative values represent an over-estimation and positive values an under-estimation of smartphone use.

Third, we calculated Pearson’s correlation coefficient for all the variables included in the subsequent regression models (i.e., for a general day, a weekday, and a weekend day). Additionally, we computed Spearman’s rank correlations (by keeping each item in the original ordinal scale format) to better investigate how traced duration, frequency, and time distortion, were related to each item of the SAS-SV scale. The results were corrected for multiple comparisons (p ≤ .005).

Eventually, we ran regression models with PSU at T1 and T2 as the outcome variables predicted by traced duration and frequency of smartphone use as well as time distortion. Gender and social desirability were included as control variables. In the longitudinal model, we also included PSU at T1 to control for autoregressive effects. In particular, six regression models were run to predict PSU levels cross-sectionally (at T1) and one year later (at T2). Due to the small sample size, we decided to highlight marginally significant effects (p < .10) when reporting the regression results [65]. The dataset and the survey instrument are available in a repository at the following link: https://osf.io/hwr2u.

Results

Descriptive results

Duration of smartphone use was automatically collected through the Ethica application on participants’ smartphones for 45 consecutive days from the enrollment date. Results of digital trace data showed that adolescents used the smartphone, on average, for one hour and fifty minutes per day, which was the average of all 45 recorded days irrespective of weekdays and weekend days (M = 1.89, SD = 1.52). The minimum was five minutes, and the maximum was six hours and fifty minutes. When looking at weekdays and weekend days separately, the recorded amount of time on a typical weekday was one hour and forty minutes (M = 1.68, SD = 1.51). In contrast, the average amount of time spent on the smartphone on a typical weekend day increased to two hours (M = 2.01, SD = 1.65). Trace data for frequency of use showed that adolescents activated the screen of their smartphones, on average, 57 times (median = 51) during a typical day, ranging from 7 to 222 times. The frequency of use tended to be lower on a weekday, with 55 times (median = 48, ranging from 7 to 196), on average, and higher on a weekend day (M = 66, median = 51, ranging from 1 to 286).

Regarding self-report measures, we found that participants indicated to spend approximately one hour and 53 minutes (M = 1.88, SD = 1.41) on their smartphone during a general day. However, they estimated to spend approximately one hour and forty minutes during a weekday (M = 1.67, SD = 1.40) and two hours and a half during a weekend day (M = 2.43, SD = 1.56). Also, participants reported somewhat higher levels of PSU at T2 compared to T1 (M_T1 = 1.86, SD_T1 = .85, M_T2 = 1.94, SD_T2 = .86). Yet, the increment was not statistically significant (t = -1.49, p = .140).

Additionally, based on the computed difference index (Δ), we estimated the extent of participants’ time distortion when reporting on their smartphone use. The results showed that, in general, adolescents tended to equally over- and under-estimate, with the mean of the difference index being close to zero (M = .006, SD = 1.83). In particular, 44% of the sample tended to over-estimate their smartphone use on a general day, 43% tended to under-estimate, and 13% provided a correct smartphone use estimate (with an error ranging from -30 minutes to +30 minutes). On weekdays, participants tended to under-estimate the time spent on the smartphone (M = .152, SD = 1.83; 38% over-estimators, 45% under-estimators, and 17% correct estimators). Conversely, during weekend days, adolescents tended to over-estimate smartphone use (M = -.358, SD = 2.02, with 54% over-estimators, 46% under-estimators, and 0% correct estimators). To note, the variability in time distortion was larger on a weekend day compared to a weekday. An overview of the results is shown in Fig 1.

Accuracy of self-report measures of smartphone use

For a general day, self-report duration of smartphone use positively and significantly correlated with traced frequency of use (r = .276, p = .012). Similarly, for a typical weekday, self-report duration was correlated with traced frequency of use (r = .244, p = .027). After applying a correction for multiple comparisons, the result was no longer significant. However, looking at a typical weekend day, the self-report measure significantly correlated with traced frequency of use (r = .325, p = .014). No significant relationship was found between self-reports and traced duration of smartphone use for a general day, a typical weekday, and a weekend day. Thus, to answer RQ1, there was a small-to-medium relationship between self-report smartphone use and trace data, but only when the traced frequency of smartphone use was considered. With regards to RQ2, no noteworthy differences in the correlation coefficients were evident for self-report and traced smartphone use, neither on weekdays nor on weekend days.

Preliminary correlations with problematic smartphone use

For a general day, Pearson’s bivariate correlations showed that PSU at T1 correlated only with social desirability (r = .478, p < .001), whereas PSU at T2 correlated with time distortion, with higher PSU being associated with over-estimation (r = -.236, p = .035), and social desirability (r = .326, p < .001). For weekdays, PSU correlated only with social desirability at both T1 (r = .478, p < .001) and T2 (r = .326, p < .001). Whereas, for weekend days, time distortion significantly and negatively correlated with PSU at both T1 (r = -.243, p < .001) and T2 (r = -.278, p < .001), in addition to social desirability (r = .478, p < -001 at T1 and r = .326, p < .001 at T2). Table 1 summarizes the results for a general day (see S1 and S2 Tables in the Supplement for weekdays and weekend days results).

Table 1. Bivariate correlations among predictor and outcome variables for a general day.

	1.	2.	3.	4.	5.	6.	7.
1. PSU at T1	1
2. PSU at T2	.575^**	1
3. Trace duration of smartphone use	.150	.102	1
4. Trace frequency of smartphone use	.199	.179	.580^**	1
5. Time distortion (Δ index)	-.176	-.236^*	.520^**	.079	1
6. Gender	.001	-.095	-.062	.137	-.192	1
7. Social desirability	.478^**	.326^**	.071	.323^**	-.190	-.122	1

Open in a new tab

Δ index represents traced duration minus self-report duration

*p < .05

** p < .001.

Since the SAS-SV covers different facets of the concept of PSU (e.g., excessive use, withdrawal symptoms, conflict), we correlated each item of the SAS-SV scale with traced smartphone use and the difference index as an indicator of time distortion. We found that item 9 "Using my smartphone longer than I had intended" was the only item that was positively and significantly correlated with traced frequency of use (rho = .322, p = .003). Frequency of smartphone use also correlated with item 3 "Feeling pain in the wrists or at the back of the neck while using a smartphone" (rho = .297, p = .007) and item 5 "Feeling impatient and fretful when I am not holding my smartphone" (rho = .260, p = .017). However, after applying Bonferroni’s correction, these latter results were no longer significant. Result are reported in S3 Table.

Furthermore, correlations between measures of smartphone use and PSU at T2 showed that time distortion negatively correlated with item 5 (rho = -.357, p = .001), meaning that over-estimators reported higher levels for this item. Time distortion also correlated with item 8 “Constantly checking my smartphone so as not to miss conversations between other people on Twitter or Facebook” (rho = -.221, p = .050). Traced duration of use was significantly related to item 10 “The people around me tell me that I use my smartphone too much” (rho = .226, p = .045), however the latter two were no longer significant after applying the correction for multiple comparisons. Results are reported in S4 Table.

Regression results

Including gender, social desirability and measures of smartphone use, i.e. traced duration and frequency of smartphone use, and time distortion as predictors of PSU (see Table 2), regression results showed that social desirability (β = -.455, p < .001) and traced duration of smartphone use (β = .307, p = .038) were significantly associated with PSU at T1. Time distortion, i.e., the tendency to over-estimate smartphone use, showed only a marginally significant association with PSU at T1 (β = -.232, p = .062). In the longitudinal model, the tendency to over-estimate predicted significantly higher PSU levels one year later (β = -.267, p = .029), even after controlling for baseline levels of PSU.

Table 2. Regression results for predicting PSU at T1 and T2 from traced smartphone use on a general day.

	Outcomes
	Problematic smartphone use at T1		Problematic smartphone use at T2
Predictor variables	B (S.E.)	β	B (S.E.)	β
1.Gender	.038 (.085)	.046	-.097 (.081)	-.116
2.Social desirability	-.269 (.064)	-.455 ^**	-.021 (.067)	-.035
3.Traced duration of smartphone use	.116 (.055)	.307 ^*	.050 (.055)	.132
4.Traced frequency of smartphone use	-.068 (.079)	-.114	.036 (.076)	.060
5. Δ index	-.053 (.028)	-.232 ^†	-.062 (.028)	-.267 ^*
6.PSU at T1			.520 (.110)	.494 ^**
Intercept	.108 (.28)		.170 (.277)
Adjusted-R²	.235		.344
F	6.112		7.89
p-value	< .001		< .001

Open in a new tab

Δ index represents trace duration minus self-report duration

†p < .1

*p < .05

**p < .01.

When we considered the same predictor variables in the context of weekday estimates (see S5 Table), social desirability was significantly associated with PSU at T1 (β = -.262, p < .001). Traced duration of smartphone use was also related to PSU at T1 (β = .269, p = .060), but the association was marginally significant. In addition, PSU at T1 predicted the outcome at T2 (β = .508, p < .001). Time distortion also had a partially significant effect (β = -.233, p = .054).

Eventually, when changing the context to weekend days (see S6 Table), PSU at T1 was significantly associated with social desirability at T1 (β = -.436, p < .001), traced duration of smartphone use (β = .337, p = .044), and time distortion, in particular the tendency to over-estimate (β = -.375, p = .005). Also, time distortion (β = -.302, p = .025) was a significant predictor of PSU at T2, controlling for PSU at T1 (β = .447, p < .001).

To conclude, H1 and H3 were generally sustained. Participants with a longer traced duration and an over-estimation of smartphone use also reported higher levels of PSU. Additionally, H2 was partially sustained, but only at the item level of PSU, i.e. for traced frequency of smartphone use and selected items of the SAS-SV.

Discussion

The pervasiveness of digital media devices, especially smartphones, has raised concerns about adolescents’ problematic use of these devices. Previous studies highlighted that PSU negatively impacts well-being [5,19,66,67]. However, the studies relied on self-report data to assess adolescents’ (problematic) smartphone use, which is subject to several biases like recall, estimation, and social desirability bias [34]. Objectively recorded smartphone use overcomes limitations of self-reports [68], though the use of trace data poses challenges, too [47]. To date, few studies with adult samples combined self-report and digital trace data to study PSU [55]. However, data collected from underage populations are still scarce [14–16,53] and focused mainly on the frequency of calls and text messages, leaving out overall smartphone use like duration and frequency and other user behaviors such as social media use, streaming, gaming, or Internet use in general. Furthermore, past studies did not relate trace data to PSU.

To the best of our knowledge, this is the first study using digital trace data to explore predictors of PSU in adolescence. Using a dedicated application installed on participants’ devices, we collected trace data from a sample of 84 adolescents aged 13 to 14 years and compared objective trace data to self-reported estimates. We, furthermore, used both trace data and self-report data to identify predictors of higher PSU levels, both cross-sectionally and longitudinally. Our results provided valuable insights into smartphone users’ dynamics and disentangled predictors of PSU in adolescence. While some results were in line with past research on adult populations, others were not.

In contrast to previous literature on the adult population, which showed inconsistent results on the accordance between self-report and trace data [37,66–68], we found that trace and self-report data showed similar values for smartphone use. Thus, adolescents tended to report a relatively reliable estimate of their time spent with the smartphone per day, which, in our sample, amounted to approximately two hours. However, looking at the correlations between trace and self-report data, coefficients were significant only for self-reported duration of smartphone use and traced frequency of smartphone use, not traced duration. This result was consistent for estimates referring to a general day, a weekday, and a weekend day. In other words, when adolescents were asked to estimate the total time (duration) they spent with their smartphones on a typical weekday and weekend day, they relied more on the frequency of use, i.e. the frequency of their checking behaviors. This result highlights an important point to consider when adolescents report on their online behaviors: The cognitive process involved in estimating the duration of smartphone use is determined by the interference (approximated by the frequency of checking behaviors) of the smartphone in adolescents’ everyday activities, and less by the duration of the use. In fact, checking the smartphone, especially in repeated short time intervals, is intrusive and distracting and, thus, more salient when asked to report on the duration of smartphone use.

The interference of the smartphone with everyday activities is also an indicator of PSU. As such, we found that the traced frequency of smartphone use in the present study correlated with item 9 of the SAS-SV scale, which tackles a subdimension of PSU related to the perception of excessive time spent on the smartphone. Without correcting for multiple comparisons, the frequency of smartphone use also correlated with the other two items of the same scale, thus further underlying the link between the two variables. Conversely, traced duration of smartphone use did not correlate with any item of the PSU scale.

The predictive role of the frequency of smartphone use is not surprising, considering the nature of checking behaviors. Indeed, checking habits are specific to smartphone use, rather than the use of other devices like, for example, laptops and tablets [69]. Smartphones offer quick and easy access to rewards and work as a gateway to other applications and activities, thus increasing the overall time spent on the device and interfering with daily life. As reported by Heitmayer and Lahlou [18], young people described picking up their phones as "feeling natural or automatic, and even unconscious like when you cough and put your hand over your mouth" (p. 5). Interestingly, the authors also found that almost all study participants picked up, unlocked, used, and put back the phone without a particular purpose in mind—an activity named "fidgeting". Additionally, when unlocked, young people often fall into a loop. For example, the engagement in one social media app also triggers the use of other social media platforms, thus making participants spend more time on the smartphone than initially intended. Considering our results, it is possible that adolescents with higher PSU levels also unlock and use the phone without a particular intention multiple times a day, thus entering into a loop.

However, when predicting PSU, the traced duration of smartphone use, rather than the traced frequency of use, showed a significant relationship. In other words, traced frequency is associated with self-reported duration of smartphone use, whereas traced duration is related to self-reported PSU.

This result is partially in line with findings from tracking studies in adult populations. For example, Montag and colleagues [70] found that both phone use recorded in hours per week and the number of calls and text messages were related to problematic mobile phone use. Also, Noë and colleagues [59] showed that overall usage and scrolling events were positively associated with PSU (with social media apps like Instagram being a primary source of activity for smartphone addicts). Tossell and colleagues [58] underlined how smartphone addict users spent twice as much time on the phone and started interaction with applications twice as often than non-addict users. Our results also align with a recent meta-analysis showing a small correlation between PSU and traced smartphone use [71]. On the contrary, Lin and colleagues [50] found that the frequency of checking behaviors, and not the duration of smartphone use, was related to PSU. Our findings are also consistent with findings from research relying on self-reports only. Several studies showed that a longer duration of smartphone use was associated with higher levels of PSU [72–74].

In general, smartphone use may obstruct everyday routines and specific activities such as studying/learning, working, and offline social interactions [75]. Scales measuring PSU pick up this aspect and include items such as "Having a hard time concentrating in class, while doing assignments, or while working due to smartphone use" or "Constantly checking my smartphone so as not to miss conversations between other people on Twitter or Facebook" [23]. The interference due to smartphone use creates functional impairments, which are similar to the ones caused by other behavioral addictions, but, at the same time, also specific to PSU [50]. In accordance with this hypothesis, using the smartphone in an absent-minded way has been found to have a pervasive and strong positive link with various measures of inattention [76]. Furthermore, prior research has identified impulsivity and the urgency to immediately check incoming notifications as two risk factors for PSU [77,78]. In addition, fear of missing out, i.e., the "pervasive apprehension that others might be having rewarding experiences from which one is absent" [79, p. 1841] has been related to PSU [80–82].

A second major finding is that adolescents tend to both over- and under-estimate their smartphone use. This result is in line with a recent meta-analysis [71] showing that only three out of the 49 included comparisons showed that self-reported media use was close to the logged mean, whereas an equal number of studies showed that participants either over- (k = 23) or under- (k = 23) estimated their time spent using digital media devices. To note, in our study, time distortion was related to PSU both cross-sectionally and longitudinally. Additionally, over-estimation was related to item 5 of the SAS-SV scale, which tackles the experience of withdrawal symptoms (e.g., irritability) when one cannot use the smartphone. More precisely, our study revealed that over-estimation predicted higher PSU levels over time, even after controlling for social desirability and baseline levels of PSU. One can conclude from this finding that adolescents who excessively used their smartphones were well aware of their problematic use and reported a higher duration of smartphone use. However, based on what we just said, adolescents seemed to rely more on their perceptions of the frequency of use, i.e. how much they repeatedly check their smartphones, rather than of the duration of use when reporting on the latter. This explains, in part, the role of over-estimation of smartphone use in relation to PSU, as shown in the significant effect of the difference index in the model for a general day and a typical weekend day.

Although our findings of over-estimation contradict results from a prior study by Lin and colleagues [50], who found that under-estimators had higher PSU levels, they are in line with previous research on social media and online gaming. For example, Turel and colleagues [83] found an effect of time distortion on social media addiction. In particular, the at-risk group showed a significant over-estimation bias, which positively correlated with the Facebook addiction scale; in contrast, the no-risk group tended to report an under-estimation of social media use. Furthermore, our results aligned with a recent study investigating social media use in adolescents [84], which also found that participants tended to over-estimate the time spent on social media platforms, with a more accurate estimation for Instagram use rather than the use of WhatsApp or Snapchat.

Consistent with different perception theories [85,86], time distortion can be a cognitive marker that may help classify youth in ’at-risk’ and ’no-risk’ groups concerning media-related addictions. From a cognitive perspective, over-estimation can be related to impaired attention mechanisms since people are engaging in highly rewarding behaviors, e.g., social networking or online gaming [83,87]. The involvement in this kind of activities may over-emphasize the perception of time spent doing them. Furthermore, over-estimation can be related to a cognitive distortion of short durations (which are often over-estimated) and long durations (which are often under-estimated), following Vierordt’s law [88–90]. Thus, frequent checking behaviors that last for short moments may be estimated, overall, as a longer (daily) use of the device than what it actually is.

A third important finding is that, in contrast to past studies that considered general smartphone use, our results added valuable insights on the differences and similarities in self-reported and traced smartphone use on weekdays and weekend days, and their role in PSU. However, future research should further investigate smartphone use in different days of the week characterized by more or less structured daily routines (e.g., school on weekdays, free time on weekend days). That would shed light on how such routines play a role in estimating smartphone use and predicting a problematic use of the device.

Eventually, it should be noted that social desirability bias plays an important role in predicting higher levels of PSU in adolescents. More precisely, we found that participants who were more inclined to provide socially desirable answers reported lower PSU levels. This finding may not be surprising, considering that social desirability describes respondents’ tendency to inaccurately report on sensitive topics, in order to present themselves in the best possible light [39]. When sensitive topics have a negative connotation, as in the case of PSU, adolescents with a higher tendency to provide socially desirable answers adapted their responses to what they think the other person (i.e. researcher) considers as socially acceptable. Thus, they tended to under-report when asked about (problematic) smartphone use, for example, how often they have a hard time concentrating in class or how often they miss planned work due to their smartphone use, or how often they use the smartphone longer than intended. While digital trace data allow accounting for a potential estimation bias, as demonstrated in the present study, the reliance on automatically recorded data is not sufficient to account for a possible social desirability bias inherent to self-reported measures such as those typically applied to assess PSU. Critics have repeatedly argued that participants with higher levels of social desirability propensity may adapt their habitual use of the smartphone when being aware that they are traced. The present study objectively collected data on adolescents’ smartphone use for 45 consecutive days to avoid biases due to short term adaptations of habitual smartphone use. Yet, other studies with adult populations used much shorter timeframes, arguing that a few days are enough to predict repetitive behaviors reliably [83].

Limitations and recommendations for future research

Our study comes with some limitations that should be acknowledged. In particular, due to the small sample size, our statistical analyses should be interpreted with caution, and future studies should replicate our findings with larger samples. Additionally, samples should represent the underlying population, which is a noteworthy challenge because many adolescents see digital trace data as a strong invasion of their privacy [91]. Privacy concerns are also a challenge for participants’ adherence over time, which is lower than in traditional survey-based studies, and only a few people are generally predisposed to donate log data [92].

Furthermore, we faced some technical problems due to the Ethica application malfunctions or incompatibility with older smartphone models. Although the application provider partially resolved these problems, they still resulted in dropouts, reducing even further our analytical sample. Besides, trace data collected as part of this study was limited to the overall duration and frequency of smartphone use. Thus, we have no information on what activities adolescents engaged in on their smartphones based on objectively recorded data. Although ethical and legal restrictions limit the possibility to obtain objective data on specific contents, application usage can be integrated into tracking studies. Future studies should thus focus on specific application usage, with a particular emphasis on social networking applications [93] and online gaming, given that both are popular activities among younger populations [94]. Finally, by investigating frequency, duration, and type of applications used, future research should rely more on objective criteria or expert ratings of PSU, as partially done in research on adult populations [50], to overcome the social desirability bias problem, thus providing a starting point for (smartphone-based) interventions to reduce PSU. This may further allow enhancing our knowledge of drivers and indicators of PSU.

Additionally, PSU was the only self-report measure of well-being related to trace smartphone use in this study. However, several other factors could impact the study’s findings, for example, the presence of comorbid mental disorders like subthreshold symptoms of depression and anxiety or personality traits like neuroticism and impulsivity [95]. Since the present study did not consider these additional factors in explaining the link between trace and self-report smartphone use, future research should further identify vulnerable adolescent populations characterized by mental disorders or dysfunctional traits to investigate differential relationships between predictors of PSU based on digital trace and self-report data.

Finally, on a side note, although the SAS-SV used in the present study has been found to predict PSU levels based on experts’ diagnoses [23], its validation has not been carried out in clinical settings. Hence, the SAS-SV can be only used to detect potentially high-risk groups of PSU in the general population. To overcome these limitations, future studies should complement self-assessments of PSU with clinical interviews to further validate the use of this scale in assessing symptoms of PSU.

Conclusions

PSU remains a critical concept to be studied in adolescents due to the identified negative short- and long-term consequences for personal well-being and development. Valid and reliable predictors and PSU indicators are urgently needed, and digital trace data proved to be an informative data source that requires further investigations to eventually develop diagnostic tools and indices of PSU that incorporate trace data.

Supporting information

S1 Table. Bivariate Pearson’s correlations among predictor and outcome variables for a weekday.

(DOCX)

Click here for additional data file.^{(15.4KB, docx)}

S2 Table. Bivariate Pearson’s correlations among predictor and outcome variables for a weekend day.

(DOCX)

Click here for additional data file.^{(15.3KB, docx)}

S3 Table. Spearman’s rank correlations between items of the SAS-SV at T1, traced duration and frequency of smartphone use, and difference index.

(DOCX)

Click here for additional data file.^{(16.7KB, docx)}

S4 Table. Spearman’s rank correlations between items of the SAS-SV at T2, trace duration and frequency of smartphone use, and difference index.

(DOCX)

Click here for additional data file.^{(16.8KB, docx)}

S5 Table. Regression results for PSU at T1 and T2 including smartphone use on a weekday.

(DOCX)

Click here for additional data file.^{(14.4KB, docx)}

S6 Table. Regression results for PSU at T1 and T2 including smartphone use on a weekend day.

(DOCX)

Click here for additional data file.^{(14.4KB, docx)}

Acknowledgments

We would like to thank the collaborating schools for their help in student recruitment and data collection. We also thank Mohammad Hashemian and Chiara Antonietti for their help during the preparation and data collection phase as well as Nathaniel Osgood and Teena Thomas during the data analysis phase.

Data Availability

The study data and questionnaire are available here: https://osf.io/hwr2u

Funding Statement

This research has received funding from the Swiss National Science Foundation under the Grant no. 10001C_175874. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Choi A. Emotional well-being of children and adolescents: Recent trends and relevant factors. 2018. Feb 9 [cited 2021 Mar 17]; Available from: https://www.oecd-ilibrary.org/education/emotional-well-being-of-children-and-adolescents_41576fb2-en. [Google Scholar]
2.Fardouly J, Vartanian LR. Social Media and Body Image Concerns: Current Research and Future Directions. Curr Opin Psychol. 2016. Jun 1;9:1–5. [Google Scholar]
3.Griffiths LJ, Dowda M, Dezateux C, Pate R. Associations between sport and screen-entertainment with mental health problems in 5-year-old children. Int J Behav Nutr Phys Act 2010. Apr 21;7:30. doi: 10.1186/1479-5868-7-30 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Hale L, Guan S. Screen time and sleep among school-aged children and adolescents: a systematic literature review. Sleep Med Rev. 2015. Jun;21:50–8. doi: 10.1016/j.smrv.2014.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Sohn S, Rees P, Wildridge B, Kalk NJ, Carter B. Prevalence of problematic smartphone usage and associated mental health outcomes amongst children and young people: a systematic review, meta-analysis and GRADE of the evidence. BMC Psychiatry. 2019. Nov 29;19(1):356. doi: 10.1186/s12888-019-2350-x [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Stiglic N, Viner RM. Effects of screentime on the health and well-being of children and adolescents: a systematic review of reviews. BMJ Open. 2019. Jan 1;9(1):e023191. doi: 10.1136/bmjopen-2018-023191 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Walsh JJ, Barnes JD, Cameron JD, Goldfield GS, Chaput J-P, Gunnell KE, et al. Associations between 24 hour movement behaviours and global cognition in US children: a cross-sectional observational study. Lancet Child Adolesc Health. 2018. Nov;2(11):783–91. doi: 10.1016/S2352-4642(18)30278-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Throuvala MA, Griffiths MD, Rennoldson M, Kuss DJ. Motivational processes and dysfunctional mechanisms of social media use among adolescents: a qualitative focus group study. Comput Hum Behav. 2019. Apr;93:164–75. [Google Scholar]
9.Mascia ML, Agus M, Penna MP. Emotional Intelligence, Self-Regulation, Smartphone Addiction: Which Relationship With Student Well-Being and Quality of Life? Front Psychol [Internet]. 2020. [cited 2021 Mar 17];11. Available from: https://www.readcube.com/articles/10.3389%2Ffpsyg.2020.00375. doi: 10.3389/fpsyg.2020.00375 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Camerini A-L, Marciano L. Media use and academic achievement. In: The International Encyclopedia of Media Psychology [Internet]. J. Van den Bulck. J. VWiley-Blackwell; 2020. [cited 2020 Aug 25]. Available from: https://blackwells.co.uk/bookshop/product/The-International-Encyclopedia-of-Media-Psychology-by-Jan-Van-den-Bulck-editor/9781119011064. [Google Scholar]
11.Verbeij T, Pouwels JL, Beyens I, Valkenburg PM. The accuracy and validity of self-reported social media use measures among adolescents. Comput Hum Behav Rep. 2021. Jan 1;3:100090. [Google Scholar]
12.Howison J, Wiggins A, Crowston K. Validity issues in the use of social network analysis with digital trace data. J Assoc Inf Syst. 2011;12(12):767–97. [Google Scholar]
13.Süss DD, Waller G, Jael B, Lilian S, Gregor W, Céline K, et al. Rapporto sui risultati dello studio JAMES 2020. 2020;76. [Google Scholar]
14.Mireku MO, Mueller W, Fleming C, Chang I, Dumontheil I, Thomas MSC, et al. Total recall in the SCAMP cohort: Validation of self-reported mobile phone use in the smartphone era. Environ Res. 2018. Feb;161:1–8. doi: 10.1016/j.envres.2017.10.034 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Inyang I, Benke G, Morrissey J, McKenzie R, Abramson M. How well do adolescents recall use of mobile telephones? Results of a validation study. BMC Med Res Methodol. 2009. Jun 12;9:36. doi: 10.1186/1471-2288-9-36 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Aydin D, Feychting M, Schüz J, Andersen TV, Poulsen AH, Prochazka M, et al. Predictors and overestimation of recalled mobile phone use among children and adolescents. Prog Biophys Mol Biol. 2011. Dec 1;107(3):356–61. doi: 10.1016/j.pbiomolbio.2011.08.013 [DOI] [PubMed] [Google Scholar]
17.Pew Research Center. Teens, Social Media & Technology 2018 [Internet]. Pew Research Center: Internet, Science & Tech. 2018 [cited 2021 Mar 17]. Available from: https://www.pewresearch.org/internet/2018/05/31/teens-social-media-technology-2018/.
18.Heitmayer M, Lahlou S. Why are smartphones disruptive? An empirical study of smartphone use in real-life contexts. Comput Hum Behav. 2021. Mar;116:106637. [Google Scholar]
19.Elhai JD, Dvorak RD, Levine JC, Hall BJ. Problematic smartphone use: A conceptual overview and systematic review of relations with anxiety and depression psychopathology. J Affect Disord. 2017. Jan;207:251–9. doi: 10.1016/j.jad.2016.08.030 [DOI] [PubMed] [Google Scholar]
20.OECD. Children & Young People’s Mental Health in the Digital Age from: shaping the future [Internet]. 2018. Available from: https://www.oecd.org/health/mental-health.htm.
21.American Psychiatric Association, editor. Diagnostic and statistical manual of mental disorders: DSM-5. 5th ed. Washington, D.C: American Psychiatric Association; 2013. 947 p. [Google Scholar]
22.Rosenberg KP. Behavioral addictions: Criteria, evidence, and treatment. Academic Press; 2014. [Google Scholar]
23.Kwon M, Lee J-Y, Won W-Y, Park J-W, Min J-A, Hahn C, et al. Development and Validation of a Smartphone Addiction Scale (SAS). PLOS ONE. 2013. Feb 27;8(2):e56936. doi: 10.1371/journal.pone.0056936 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Lin Y-H, Chang L-R, Lee Y-H, Tseng H-W, Kuo TBJ, Chen S-H. Development and Validation of the Smartphone Addiction Inventory (SPAI). PLOS ONE. 2014. Jun 4;9(6):e98312. doi: 10.1371/journal.pone.0098312 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Shaw M, Black DW. Internet addiction: definition, assessment, epidemiology and clinical management. CNS Drugs. 2008;22(5):353–65. doi: 10.2165/00023210-200822050-00001 [DOI] [PubMed] [Google Scholar]
26.Škařupová K, Ólafsson K, Blinka L. The effect of smartphone use on trends in European adolescents’ excessive Internet use. Behav Inf Technol. 2016. Jan 2;35(1):68–74. [Google Scholar]
27.Jo YS, Bhang SY, Choi JS, Lee HK, Lee SY, Kweon Y-S. Clinical Characteristics of Diagnosis for Internet Gaming Disorder: Comparison of DSM-5 IGD and ICD-11 GD Diagnosis. J Clin Med [Internet]. 2019. Jun 28 [cited 2020 Aug 4];8(7). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6678371/. doi: 10.3390/jcm8070945 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Petry DN. Behavioral Addictions: DSM-5® and Beyond. Oxford University Press; 2015. 257 p. [Google Scholar]
29.Bragazzi NL, Del Puente G. A proposal for including nomophobia in the new DSM-V. Psychol Res Behav Manag. 2014. May 16;7:155–60. doi: 10.2147/PRBM.S41386 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Harris B, Regan T, Schueler J, Fields SA. Problematic Mobile Phone and Smartphone Use Scales: A Systematic Review. Front Psychol [Internet]. 2020. May 5 [cited 2020 Sep 7];11. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7214716/. doi: 10.3389/fpsyg.2020.00672 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Busch P, McCarthy S. Antecedents and consequences of problematic smartphone use: A systematic literature review of an emerging research area. Comput Hum Behav. 2021. [Google Scholar]
32.Billieux J, Maurage P, Lopez-Fernandez O, Kuss DJ, Griffiths MD. Can Disordered Mobile Phone Use Be Considered a Behavioral Addiction? An Update on Current Evidence and a Comprehensive Model for Future Research. Curr Addict Rep. 2015. Jun 1;2(2):156–62. [Google Scholar]
33.Panova T, Carbonell X. Is smartphone addiction really an addiction? J Behav Addict. 2018. Jun 1;7(2):252–9. doi: 10.1556/2006.7.2018.49 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.de Reuver M, Bouwman H. Dealing with self-report bias in mobile Internet acceptance and usage studies. Inf Manage. 2015. Apr 1;52(3):287–94. [Google Scholar]
35.Krumpal I. Determinants of social desirability bias in sensitive surveys: a literature review. Qual Quant. 2013. Jun 1;47(4):2025–47. [Google Scholar]
36.Moorman RH, Podsakoff PM. A meta-analytic review and empirical test of the potential confounding effects of social desirability response sets in organizational behaviour research. J Occup Organ Psychol. 1992;65(2):131–49. [Google Scholar]
37.Boase J, Ling R. Measuring Mobile Phone Use: Self-Report Versus Log Data. J Comput-Mediat Commun. 2013;18(4):508–19. [Google Scholar]
38.Friedman ER. Judgments of Time Intervals by Young Children. Percept Mot Skills. 1977. Dec 1;45(3):715–20. doi: 10.2466/pms.1977.45.3.715 [DOI] [PubMed] [Google Scholar]
39.Ford LH. A social desirability questionnaire for young children. J Consult Clin Psychol. 1970;35(2):195–204. [Google Scholar]
40.Adams AS, Soumerai SB, Lomas J, Ross-Degnan D. Evidence of self-report bias in assessing adherence to guidelines. Int J Qual Health Care J Int Soc Qual Health Care. 1999. Jun;11(3):187–92. doi: 10.1093/intqhc/11.3.187 [DOI] [PubMed] [Google Scholar]
41.Donaldson SI, Grant-Vallone EJ. Understanding Self-Report Bias in Organizational Behavior Research. J Bus Psychol. 2002. Dec 1;17(2):245–60. [Google Scholar]
42.Lee E, Hu MY, Toh RS. Are Consumer Survey Results Distorted? Systematic Impact of Behavioral Frequency and Duration on Survey Response Errors. J Mark Res. 2000. Feb 1;37(1):125–33. [Google Scholar]
43.Slater MD. Operationalizing and Analyzing Exposure: The Foundation of Media Effects Research. Journal Mass Commun Q. 2004. Mar 1;81(1):168–83. [Google Scholar]
44.Parry DA, Davidson BI, Sewall C, Fisher JT, Mieczkowski H, Quintana D. A Systematic Review and Meta-Analysis of Discrepancies Between Logged and Self- Reported Digital Media Use [Internet]. PsyArXiv; 2020. [cited 2021 Mar 22]. Available from: https://psyarxiv.com/f6xvz/. [DOI] [PubMed] [Google Scholar]
45.Harari GM, Müller SR, Aung MS, Rentfrow PJ. Smartphone sensing methods for studying behavior in everyday life. Curr Opin Behav Sci. 2017. Dec 1;18:83–90. [Google Scholar]
46.Lauricella AR, Roberts M, Curtis P. Using LENA to understand children’s media use: Challenges and opportunities. In San Diego, CA, US; 2017.
47.Stier S, Breuer J, Siegers P, Thorson K. Integrating Survey Data and Digital Trace Data: Key Issues in Developing an Emerging Field. Soc Sci Comput Rev. 2020. Oct 1;38(5):503–16. [Google Scholar]
48.Brickhman D, Scandurra L, Powell K, Rich MO. Assessing the agreement of multiple methods of measuring children’s media use in U.S. and Mexican samples. In San Diego, CA, US; 2017.
49.GDPR. General Data Protection Regulation (GDPR)–Official Legal Text [Internet]. General Data Protection Regulation (GDPR). 2018 [cited 2021 Mar 17]. Available from: https://gdpr-info.eu/.
50.Lin Y-H, Chiang C-L, Lin P-H, Chang L-R, Ko C-H, Lee Y-H, et al. Proposed Diagnostic Criteria for Smartphone Addiction. Weinstein AM, editor. PLOS ONE. 2016. Nov 15;11(11):e0163010. doi: 10.1371/journal.pone.0163010 [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Reeves B, Ram N, Robinson TN, Cummings JJ, Giles CL, Pan J, et al. Screenomics: A Framework to Capture and Analyze Personal Life Experiences and the Ways that Technology Shapes Them. Human–Computer Interact. 2021. Mar 4;36(2):150–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Radesky JS. Mobile device passive sensing to measure parent and child smartphone and tablet usage. In San Diego, CA, US; 2017.
53.Goedhart G, Vrijheid M, Wiart J, Hours M, Kromhout H, Cardis E, et al. Using software-modified smartphones to validate self-reported mobile phone use in young people: A pilot study. Bioelectromagnetics. 2015. Oct;36(7):538–43. doi: 10.1002/bem.21931 [DOI] [PubMed] [Google Scholar]
54.Brinberg M, Ram N, Yang X, Cho M-J, Sundar SS, Robinson TN, et al. The idiosyncrasies of everyday digital lives: Using the Human Screenome Project to study user behavior on smartphones. Comput Hum Behav. 2021. Jan 1;114:106570. doi: 10.1016/j.chb.2020.106570 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Ryding FC, Kuss DJ. Passive objective measures in the assessment of problematic smartphone use: A systematic review. Addict Behav Rep. 2020. Jun 1;11:100257. doi: 10.1016/j.abrep.2020.100257 [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Lin Y-H, Lin Y-C, Lee Y-H, Lin P-H, Lin S-H, Chang L-R, et al. Time distortion associated with smartphone addiction: Identifying smartphone addiction via a mobile application (App). J Psychiatr Res. 2015. Jun;65:139–45. doi: 10.1016/j.jpsychires.2015.04.003 [DOI] [PubMed] [Google Scholar]
57.Ko C-H, Yen J-Y, Yen C-F, Lin H-C, Yang M-J. Factors predictive for incidence and remission of internet addiction in young adolescents: a prospective study. Cyberpsychology Behav Impact Internet Multimed Virtual Real Behav Soc. 2007. Aug;10(4):545–51. doi: 10.1089/cpb.2007.9992 [DOI] [PubMed] [Google Scholar]
58.Tossell C, Kortum P, Shepard C, Rahmati A, Zhong L. Exploring Smartphone Addiction: Insights from Long-Term Telemetric Behavioral Measures. Int J Interact Mob Technol IJIM. 2015. Mar 23;9(2):37–43. [Google Scholar]
59.Noe B, Turner LD, Linden DEJ, Allen SM, Winkens B, Whitaker RM. Identifying Indicators of Smartphone Addiction Through User-App Interaction. Comput Hum Behav. 2019. Oct;99:56–65. doi: 10.1016/j.chb.2019.04.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Camerini A-L, Schulz PJ, Jeannet A-M. The social inequalities of Internet access, its use, and the impact on children’s academic performance: Evidence from a longitudinal study in Switzerland. NEW MEDIA Soc. 2018. Jul;20(7):2489–508. [Google Scholar]
61.Camerini A-L, Marciano L. Self-selection bias in research including ecological momentary assessment and digital trace data. In: Passive Smartphone Data Collection and Additional Tasks in Mobile Web Surveys: Willingness, Non-Participation, Consent, and Ethics I. Zagreb; 2019.
62.Mößle T. dick, dumm, abhängig, gewalttätig? Nomos Verlagsgesellschaft mbH & Co. KG; 2012. doi: 10.1002/mrm.24177 [DOI] [Google Scholar]
63.Thomas Vattukalathil T. Application of data analytics and machine learning on data collected by smartphones to understand human behavioural patterns. 2019. [Google Scholar]
64.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. Dec 31;8(12):e83558. doi: 10.1371/journal.pone.0083558 [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Pritschet L, Powell D, Horne Z. Marginally Significant Effects as Evidence for Hypotheses: Changing Attitudes Over Four Decades. Psychol Sci. 2016. Jul 1;27(7):1036–42. doi: 10.1177/0956797616645672 [DOI] [PubMed] [Google Scholar]
66.Odgers C. Smartphones are bad for some teens, not all. Nature. 2018. 22;554(7693):432–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Mascia ML, Agus M, Penna MP. Emotional Intelligence, Self-Regulation, Smartphone Addiction: Which Relationship With Student Well-Being and Quality of Life? Front Psychol [Internet]. 2020. Mar 6 [cited 2021 Mar 17];11. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7068808/. doi: 10.3389/fpsyg.2020.00375 [DOI] [PMC free article] [PubMed] [Google Scholar]
68.van Velthoven MH, Powell J, Powell G. Problematic smartphone use: Digital approaches to an emerging public health problem. Digit Health. 2018. Jan 1;4:2055207618759167. [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Oulasvirta A, Rattenbury T, Ma L, Raita E. Habits make smartphone use more pervasive. Pers Ubiquitous Comput. 2012. Jan 1;16(1):105–14. [Google Scholar]
70.Montag C, Błaszkiewicz K, Lachmann B, Sariyska R, Andone I, Trendafilov B, et al. Recorded Behavior as a Valuable Resource for Diagnostics in Mobile Phone Addiction: Evidence from Psychoinformatics. Behav Sci. 2015. Dec;5(4):434–42. doi: 10.3390/bs5040434 [DOI] [PMC free article] [PubMed] [Google Scholar]
71.Parry DA, Davidson BI, Sewall CJR, Fisher JT, Mieczkowski H, Quintana DS. A systematic review and meta-analysis of discrepancies between logged and self-reported digital media use. Nat Hum Behav. 2021. May 17;1–13. doi: 10.1038/s41562-021-01049-0 [DOI] [PubMed] [Google Scholar]
72.Al-Barashdi HS, Bouazza A, Jabur NH. Smartphone Addiction among University Undergraduates: A Literature Review. J Sci Res Rep. 2015;210–25. [Google Scholar]
73.De-Sola Gutiérrez J, Rodríguez de Fonseca F, Rubio G. Cell-Phone Addiction: A Review. Front Psychiatry [Internet]. 2016. [cited 2021 Jan 15];7. Available from: https://www.frontiersin.org/articles/10.3389/fpsyt.2016.00175/full. doi: 10.3389/fpsyt.2016.00175 [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Haug S, Castro RP, Kwon M, Filler A, Kowatsch T, Schaub MP. Smartphone use and smartphone addiction among young people in Switzerland. J Behav Addict. 2015. Dec 1;4(4):299–307. doi: 10.1556/2006.4.2015.037 [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Martini M, Heinz A, Hinterholzer J, Martini C, Sachse P. Effects of wakeful resting versus social media usage after learning on the retention of new memories. Appl Cogn Psychol. 2020;34(2):551–8. [Google Scholar]
76.Marty-Dugas J, Ralph BCW, Oakman JM, Smilek D. The relation between smartphone use and everyday inattention. Psychol Conscious Theory Res Pract. 2018;5(1):46–62. [Google Scholar]
77.Contractor AA, Weiss NH, Tull MT, Elhai JD. PTSD’s relation with problematic smartphone use: Mediating role of impulsivity. Comput Hum Behav. 2017. Oct 1;75:177–83. [Google Scholar]
78.Kim Y, Jeong J-E, Cho H, Jung D-J, Kwak M, Rho MJ, et al. Personality Factors Predicting Smartphone Addiction Predisposition: Behavioral Inhibition and Activation Systems, Impulsivity, and Self-Control. PLOS ONE. 2016. Aug 17;11(8):e0159788. doi: 10.1371/journal.pone.0159788 [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Przybylski AK, Murayama K, DeHaan CR, Gladwell V. Motivational, emotional, and behavioral correlates of fear of missing out. Comput Hum Behav. 2013. Jul 1;29(4):1841–8. [Google Scholar]
80.Gezgin DM. Understanding Patterns for Smartphone Addiction: Age, Sleep Duration, Social Network Use and Fear of Missing Out. Cypriot J Educ Sci. 2018;13(2):166–77. [Google Scholar]
81.Wang J, Wang P, Yang X, Zhang G, Wang X, Zhao F, et al. Fear of Missing Out and Procrastination as Mediators Between Sensation Seeking and Adolescent Smartphone Addiction. Int J Ment Health Addict. 2019. Aug 1;17(4):1049–62. [Google Scholar]
82.Elhai JD, Levine JC, Dvorak RD, Hall BJ. Fear of missing out, need for touch, anxiety and depression are related to problematic smartphone use. Comput Hum Behav. 2016;63:509–16. [Google Scholar]
83.Turel O, Brevers D, Bechara A. Time distortion when users at-risk for social media addiction engage in non-social media tasks. J Psychiatr Res. 2018. Feb 1;97:84–8. doi: 10.1016/j.jpsychires.2017.11.014 [DOI] [PubMed] [Google Scholar]
84.Verbeij T, Pouwels JL, Beyens I, Valkenburg PM. The Accuracy and Validity of Self-Reported Social Media Use Measures Among Adolescents [Internet]. PsyArXiv; 2021. [cited 2021 Mar 17]. Available from: https://psyarxiv.com/p4yb2/. [Google Scholar]
85.Takahashi T, Oono H, Radford MHB. Psychophysics of time perception and intertemporal choice models. Phys Stat Mech Its Appl. 2008. Mar 15;387(8):2066–74. [Google Scholar]
86.Wittmann M, Paulus MP. Decision making, impulsivity and time perception. Trends Cogn Sci. 2008. Jan;12(1):7–12. doi: 10.1016/j.tics.2007.10.004 [DOI] [PubMed] [Google Scholar]
87.Rau P-LP, Peng S-Y, Yang C-C. Time distortion for expert and novice online game players. Cyberpsychology Behav Impact Internet Multimed Virtual Real Behav Soc. 2006. Aug;9(4):396–403. doi: 10.1089/cpb.2006.9.396 [DOI] [PubMed] [Google Scholar]
88.Block RA, Gruber RP. Time perception, attention, and memory: A selective review. Acta Psychol (Amst). 2014. Jun 1;149:129–33. [DOI] [PubMed] [Google Scholar]
89.Lejeune H, Wearden JH. Vierordt’s The Experimental Study of the Time Sense (1868) and its legacy. Eur J Cogn Psychol. 2009. Sep 1;21(6):941–60. [Google Scholar]
90.Tobin S, Grondin S. Video games and the perception of very long durations by adolescents. Comput Hum Behav. 2009. Mar 1;25(2):554–9. [Google Scholar]
91.Youn S, Hall K. Gender and online privacy among teens: risk perception, privacy concerns, and protection behaviors. Cyberpsychology Behav Impact Internet Multimed Virtual Real Behav Soc. 2008. Dec;11(6):763–5. [DOI] [PubMed] [Google Scholar]
92.Ohme J, Araujo T, de Vreese CH, Piotrowski JT. Mobile data donations: Assessing self-report accuracy and sample biases with the iOS Screen Time function. Mob Media Commun. 2020. Sep 30;2050157920959106. [Google Scholar]
93.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 [Internet]. 2019. Jun;9. Available from: http://search.ebscohost.com/login.aspx?direct=true&db=psyh&AN=2019-09430-001&site=ehost-live. [DOI] [PMC free article] [PubMed] [Google Scholar]
94.Süss DD, Waller G, Lilian S, Gregor W, Jael B, Céline K, et al. Findings of the JAMES Study 2018. 2018;82. [Google Scholar]
95.Carvalho LF, Sette CP, Ferrari BL. Problematic smartphone use relationship with pathological personality traits: Systematic review and meta-analysis. Cyberpsychology J Psychosoc Res Cyberspace [Internet]. 2018. Dec 31 [cited 2021 Aug 16];12(3). Available from: https://cyberpsychology.eu/article/view/11423. [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0263815.r001

Decision Letter 0

Michael Kaess

5 Jul 2021

PONE-D-21-09423

Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study

PLOS ONE

Dear Dr. Marciano,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

In particular, one reviewer raised serious concerns regarding the methology and statistical analyses of your manuscript. Please pay specific attention to these comments since methodological concerns should be targeted and alleviated during a revision.

Please submit your revised manuscript by Aug 19 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Michael Kaess, M. D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

3. Thank you for including your ethics statement: "The Cantonal education administration of Ticino approved the annual panel study based on self-administered questionnaires. The embedded Ethica study received approval from the Ethics Committee of the university where the research was carried out and the Cantonal Data Protection Officer. Participation in the Ethica study required active consent by parents who had to send a signed consent form to the Cantonal education administration, which, in turn, provided the research team with the associated U-ID. Students provided their active consent directly in the Ethica application upon enrolment."

a) Please amend your current ethics statement to include the full name of the ethics committee/institutional review board(s) that approved your specific study.

Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research.

4. Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: No

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: A timely and relevant topic, innovative methods, interesting findings, and overall, a well-written manuscript that I enjoyed reading. Nevertheless, some issues deserve further attention. There are three major hypotheses tob e tested and two further research questions. And some further approaches, such as dividing the sample into „light vs heavy users“. I’d encourage the authors to specify the signficance threshold for their hypotheses and to state whether correction for multiple comparisons was performed or not. Also, further approaches, as referred to above, need to be specified in the methods and not in the results section. Several times in the manuscript the paper refer tot he exploratory nature of their analysis. While I appreciate this transparency, there are nevertheless hypotheses which were explicitly tested, and several other parameters that were subject to exploratory analyses. I’d encourage the authors to differentiate between both approaches, to state whether a result survived testing for multiple comparisons or not, and to be very cautious in their approach to „also highlight marginally significant effects“ (p. 13), I’m afraid that this could be at times misleading when truly robust effects should be identified from a plethora of analyses performed in this study. Also, the limitations section needs some further attention. Only relationships with the SAS scale were used, and no comparisons with other instruments were made. Several factors that could have an impact on this study’s findings were not considered, e.g. the impact of comorbid mental disorders (if any in this sample, yet I couldn’t find a statement regarding this issue), subthreshold depression or anxiety levels or personality traits, e.g. impulsivity. Finally, although I’m aware that this may further increase the number of analyses, regression models (at least for a specific limited number of variables) could be considered for the SAS subscales reported by Kwon and colleagues in 2013. This information may beneficially complement the paper as supplementary data.

Reviewer #2: PONE-D-21-09423_reviewer

Review on “Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study”. The authors used subjective and objective measures of smartphone use at two assessments with a one year interval and examined the association between the subjective and objective measures, as well as the associations between these measures and the problematic smartphone use score. The study was conducted in a rather small sample of adolescents (n=84). In my opinion, the study has some major methodological issues, which make it impossible to interpret the data. Most importantly, the scale of the self-report measure and the conversion of the objective measures to match this scale, and the statistical analyses renders it hard to interpret the results. Moreover, in my opinion a Bland-Altman tests is the method to use when trying to compare two assessments (and not multiple correlation coefficients). In light of these major critics and because the problem with the scale is baked into the data, all most of my other comments listed below are minor issues.

Lines 49 – 51: However, until today, most of the studies focusing on PSU relied on self-report data, making it hard to draw valid conclusions on which type of smartphone activity can be described as problematic. – to make a stronger argument I would suggest to add references indicating that the validity of self-reports of use are questionable

Line 52: Could you please provide a definition of “digital trace data”? A short definition is sufficient.

Lines 63 – 68: Those are large percentages – however, the data stems from a diverse group. I would suggest that the very high numbers in the 90s are due to the large span of years of age of the participants (12 to 19 years). Could you provide more details on differences between the age groups? I.e., how the percentage is associated with years of age?

There are some redundancies because of duplicates in the paragraphs „Digital trace data“ and „Ethical considerations“; please stream line by deleting these redundancies

The „Sample“ paragraph clearly describes the reduction of the sample and the causes why only 84 out of the original panel of 1419 persons. Could you please provide some information whether there are differences between those who participated and those who did not? Please discuss

Measures: The scale of the self-rated use of the smartphone is not suitable to calculate a mean since it is not measured on an interval scale (the intervals between two ratings are not the same)!

Line 251: I think the results assessed by the scale should be presented in the results section and not in the method section. The same applies to the other two trace data measures. Moreover, what does is the correlation coefficient presented for each measure? The correlation between weekday use and weekend use? It is not addressed and not obvious to the readers.

Lines 274 and following: Again, I would refrain from presenting the results in the method section. Moreover, I would suggest to point out the psychometrics of the Smartphone Addiction Scale for adolescents (the alpha is presented within the brackets showing the means and SDs…).

Analytic plan: I think it is problematic to report marginally significant results and, on top of that, calculate multiple tests without correcting for the multiple testing.

Lines 319 and following: In my opinion, it is not possible to interpret the average scores of the subjective rating and you shouldn’t report or interpret the scores in the way you did. Moreover, which correlation coefficient did you calculate?

Table 1: You report a lot (a lot!) of correlation coefficients here. I would suggest to use a Bonferroni correction or another method accounting for the multiple tests that you conducted (and omit the marginal significant results!). Most important: If you converted the trace data into the same scale as the self-report measure, the data is not suitable for calculating means and SDs!

Lines 331 through 342: I think the statistical procedure is not adequate and therefore it is not possible to the interpret the results (and review this section).

Lines 344 and following: Duration of smartphone use and PSU – Did you use the trace data of the first assessment to divide the sample into light and heavy users? Please describe your procedure in more detail.

Besides the problems with the scale, I think categorizing the data can often be misleading. I follow your rational on the differentiation between those with more than 2 hr screen time and those with less – but for the other categorizations I would use the full information available and do not just compare those categories.

Regrading the regression analyses: have you checked for multicollinearity in your models? Can you please provide more information on that?

Table 3. Did you use the categorized variables in your regression analyses? I’m wondering because you list being part of the light users group/the under-estimators group/the low checkers group as references.

Minor issues:

Line 80: Please correct the references and delete the duplicate (10,20,20,21)

Please correct typos, e.g. Internet (line 156), line 455 (64,65,38,66),…

Reviewer #3: The paper is concisely written and addresses an important topic – the validity of indiciators of problematic smartphone use – assessed via digital trace data. Particularly the addressed sample of adolescents is very interesting for early interventions and thus, data derived from this population group are valuable. Unfortunately, only a very small subsample (89 out of 1419 students) could be recruited for the current study.

• To assess the representativeness of the sample and the results, it would be interesting to compare the reached subsample with the total sample of the Ethica study.

• The authors argue that self-report data on PSU might not be reliable but validate their digitally traced data also with data from a self-report measure on Smartphone Use (SAS-SV). What about the validity of this instrument? The limitations of this approach should be discussed. Why did you not conduct clinical interviews to assess PSU.

• The results on the indicator "frequency of checking" are of high interest, I wonder why the authors did not include a measure analoguous to "Perceived duration of smartphone use" to assess "perceived frequency of checking".

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2022 Feb 18;17(2):e0263815. doi: 10.1371/journal.pone.0263815.r002

Author response to Decision Letter 0

16 Aug 2021

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Partly

Reviewer #2: No

Reviewer #3: Yes

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

5. Review Comments to the Author

Reviewer #1: A timely and relevant topic, innovative methods, interesting findings, and overall, a well-written manuscript that I enjoyed reading.

A: Many thanks for the comment. We appreciate that you enjoyed the manuscript.

Nevertheless, some issues deserve further attention. There are three major hypotheses tob e tested and two further research questions. And some further approaches, such as dividing the sample into „light vs heavy users“.

I’d encourage the authors to specify the signficance threshold for their hypotheses and to state whether correction for multiple comparisons was performed or not.

A: Many thanks for the comment. In the current version, as also suggested by Reviewer 2, we removed the categorizations of high versus low users resp. checkers, and the categories of over- and under-estimation. Instead, we used the continuous scales. Additionally, we corrected the significance level of correlations by using Bonferroni’s post-doc test for multiple comparisons.

Also, further approaches, as referred to above, need to be specified in the methods and not in the results section.

A: Many thanks. We reformulated the Analytical plan subsection in the methods section and the Result section (which now includes the following sub-sections: “Descriptive results”, “Accuracy of self-report measures of smartphone use”, “Preliminary correlations with problematic smartphone use”, and “Regression results”).

Several times in the manuscript the paper refer tot he exploratory nature of their analysis. While I appreciate this transparency, there are nevertheless hypotheses which were explicitly tested, and several other parameters that were subject to exploratory analyses. I’d encourage the authors to differentiate between both approaches, to state whether a result survived testing for multiple comparisons or not, and to be very cautious in their approach to „also highlight marginally significant effects“ (p. 13), I’m afraid that this could be at times misleading when truly robust effects should be identified from a plethora of analyses performed in this study.

A: Many thanks for the comment. We now applied a Bonferroni correction for multiple comparisons when needed. We highlighted marginally significant p-values only for the regression results. We think that marginally significant regression results deserve mentioning, since our analyses are of explorative nature and no other study exists on the relationship between trace smartphone use and PSU in adolescence, thus we cannot ignore a significant trend which can be informative for future research.

Also, the limitations section needs some further attention. Only relationships with the SAS scale were used, and no comparisons with other instruments were made. Several factors that could have an impact on this study’s findings were not considered, e.g. the impact of comorbid mental disorders (if any in this sample, yet I couldn’t find a statement regarding this issue), subthreshold depression or anxiety levels or personality traits, e.g. impulsivity.

A: Many thanks for the comment, which we valued. However, we could not consider other instruments as they were not assessed as part of this study. Thus, we added the following sentence in the Limitations section: “Additionally, SP addiction was the only self-report measure of well-being related to trace SP use in this study. However, several other factors could impact the study’s findings, e.g. the presence of comorbid mental disorders like subthreshold symptoms of depression and anxiety or personality traits like neuroticism and impulsivity. Although the present paper did not consider these additional factors in explaining the link between traced and self-reported data, future research should further identify vulnerable adolescent populations…”

Finally, although I’m aware that this may further increase the number of analyses, regression models (at least for a specific limited number of variables) could be considered for the SAS subscales reported by Kwon and colleagues in 2013. This information may beneficially complement the paper as supplementary data.

A: Many thanks for the comment. We added Spearman’s correlation coefficient between trace data and each item of the SAS-SV scale at both T1 and T2, controlling for multiple comparisons, as reported in Table 3 and 4 of the Appendix. We also reported on the main significant results in the Results section of the manuscript (p. 17) and discussed them in the Discussion section (p. 22 and p.24).

We hope that our revisions are satisfactory and are available for any further questions you may have.

Reviewer #2: PONE-D-21-09423_reviewer

Most importantly, the scale of the self-report measure and the conversion of the objective measures to match this scale, and the statistical analyses renders it hard to interpret the results. Moreover, in my opinion a Bland-Altman tests is the method to use when trying to compare two assessments (and not multiple correlation coefficients). In light of these major critics and because the problem with the scale is baked into the data, all most of my other comments listed below are minor issues.

A: Many thanks for the comment. We acknowledge that using Bland-Altman tests would be beneficial when comparing two measures of the same constructs. In our case, that would be possible only for the duration of smartphone use but not for the frequency of use. Hence, we thank the reviewer for the suggestion, but we decided to rely on correlation analyses as done in comparable studies (see for example Verbeij, T., Pouwels, J. L., Beyens, I., & Valkenburg, P. M. (2021). The accuracy and validity of self-reported social media use measures among adolescents. Computers in Human Behavior Reports, 3, 100090. https://doi.org/10.1016/j.chbr.2021.100090), but changed the conversion of measures. More specifically, we now converted the self-reported measure of smartphone use by using the midpoint for each interval as a proxy of smartphone use, (e.g., 15 minutes for “0 to 30 minutes”. For the highest interval (“5 or more hours”), we used 5.5 hours as a proxy of time spent with the smartphone. This approach allowed us to consider the continuous scale of traced smartphone use and study its relation to self-report use and PSU.

A: Many thanks for the suggestion. We added the following citation to the sentence: Verbeij T, Pouwels JL, Beyens I, Valkenburg PM. The accuracy and validity of self-reported social media use measures among adolescents. Comput Hum Behav Rep [Internet]. 2021 Jan 1 [cited 2021 Jun 30];3:100090. Available from: https://www.sciencedirect.com/science/article/pii/S2451958821000385

Line 52: Could you please provide a definition of “digital trace data”? A short definition is sufficient.

A: Thanks for the comment, we added the requested definition (lines 54-55).

A: Many thanks for the comment. We added more information with respect to the type of smartphone use for different age ranges as available in the original report (lines 70-75).

There are some redundancies because of duplicates in the paragraphs „Digital trace data“ and „Ethical considerations“; please stream line by deleting these redundancies

A: Many thanks for the comment. We removed the redundant parts.

A: Many thanks for the comment. We added the following sentence (lines 239-242):

“With respect to students who did not join the Ethica study, participants did not report any difference in gender (p=.205), perceived socioeconomic status (p=.229), or self-reported daily smartphone use (p=.114).”

More information can be found here: (Camerini, A.-L., & Marciano, L. (2019, July 15). Self-selection bias in research including ecological momentary assessment and digital trace data. Passive Smartphone Data Collection and Additional Tasks in Mobile Web Surveys: Willingness, Non-Participation, Consent, and Ethics I. 8th ESRA Conference, Zagreb.)

Measures: The scale of the self-rated use of the smartphone is not suitable to calculate a mean since it is not measured on an interval scale (the intervals between two ratings are not the same)!

A: Thanks for the observation. We decided to convert the original interval scale into hours. We outlined this decision in the text under the description of Perceived duration of smartphone use (lines 255-262):

“To allow the comparison with trace data, results were converted into hours by using the midpoint for each category of the original interval scale: (0=0) (1=0.25) (2=0.75) (3=1.25) (4=1.75) (5=2.5) (6=3.5) (7=4.5) (8=5.5). For the highest interval (“5 or more hours”), we used 5.5 hours as a proxy of time spent with the smartphone.” (see Mößle, T. (2012). Dick, dumm, abhängig, gewalttätig? Nomos Verlagsgesellschaft mbH & Co. KG.)

By doing so, the comparison with the trace data of duration is now more straightforward and precise since both are expressed in the numeric equivalent of hours.

Line 251: I think the results assessed by the scale should be presented in the results section and not in the method section. The same applies to the other two trace data measures.

A: Many thanks for the comment. We reformulated the Analytical plan in the Methods section and the Result section (which now includes the following sub-sections: “Descriptive results”, “Accuracy of self-report measures of smartphone use”, “Preliminary correlations with problematic smartphone use”, and “Regression results”).

Moreover, what does is the correlation coefficient presented for each measure? The correlation between weekday use and weekend use? It is not addressed and not obvious to the readers.

A: We agree that reporting many correlations coefficients can be misleading. Hence, we removed the correlation table and reported only the Pearson’s correlations between smartphone use on a general day, weekdays, and weekend days, without mixing the results of weekdays and weekend days. We also applied a Bonferroni correction for multiple comparisons. All the variables were log-transformed before the analyses.

We reported the results in the text under the new section “Accuracy of self-report measures”.

A: Many thanks for the suggestion. We pointed out the values of Cronbach’s alpha in the text (lines 286-288). We also added a paired-sample t-test to assess if PSU changed one year later (line 344).

Analytic plan: I think it is problematic to report marginally significant results and, on top of that, calculate multiple tests without correcting for the multiple testing.

A: We now applied Bonferroni correction when correlating trace and self-report data (both self-report smartphone use and PSU). Additionally, regarding the choice to comment on marginally significant results of the regression analyses, we would like to emphasize that trends in social sciences like psychology reported an increase in the commentary of marginal effects, since sometimes “A marginal result might be interpreted as a caution against “accepting” the null hypothesis, a promising preliminary result, or sufficient evidence for some noncentral hypothesis, or it might even be interpreted as equivalent to a significant result” and the .05 cut-off is essentially arbitrary (Pritschet, L., Powell, D., & Horne, Z. (2016). Marginally Significant Effects as Evidence for Hypotheses: Changing Attitudes Over Four Decades. Psychological Science, 27(7), 1036–1042. https://doi.org/10.1177/0956797616645672). Since there are no official guidelines on how to report marginally significant results, but only many different opinions, we would prefer to highlight and not dismiss them, since they might be useful for future research on the same topic.

A: Many thanks for the comment. As stated before, to allow the comparison with trace data, self-report data were converted into hours by using the midpoint for each interval as a proxy of smartphone use, (e.g., 15 minutes for “0 to 30 minutes”. For the highest interval (“5 or more hours”), we used 5.5 hours as a proxy of time spent with the smartphone. This approach allowed us to consider smartphone use as a continuous scale and calculate Pearson’s correlation coefficients for self-report and trace data.

A: Many thanks for the comment. We agree that reporting many correlation coefficients can be misleading. Hence, we removed the correlation table and reported only the Pearson’s correlations between smartphone use on a general day, weekday and weekend day, without mixing the results of weekdays and weekend days.

Most important: If you converted the trace data into the same scale as the self-report measure, the data is not suitable for calculating means and SDs!

A: Many thanks for the comment. As stated above, to allow the comparison with trace data, self-report data for smartphone use were now converted into hours using the midpoint of each interval.

Lines 331 through 342: I think the statistical procedure is not adequate and, herefore, it is not possible to interpret the results (and review this section).

A: Thanks to the new approach to convert self-report smartphone use into hours, we now changed this part and reported on the correlation coefficients, corrected for multiple comparisons (see lines 360 to 371).

A: Many thanks for the comment. Trace data were collected only at T1. However, we now decided to avoid categorizing variables and we used the continuous variables for our analyses.

A: Many thanks for the comment, which we valued. Hence, we decided to use the full information available for all continuous variables and to report the results (i.e., correlation and regression results) for a general, a weekday, and a weekend day. The tables for a general day are placed in the main manuscript, whereas the tables for a weekday and a weekend day are reported in the supplement file. All the variables were log-transformed before the analyses to avoid problems of linearity.

Regrading the regression analyses: have you checked for multicollinearity in your models? Can you please provide more information on that?

A: many thanks for the comment. We added the correlation table for each regression analysis (in which it is possible to see that the correlations among trace data were still lower than r=.80 as a cut-off proposed by Berry WD, Feldman S. Multiple Regression in Practice (Quantitative Applications in the Social Sciences) SAGE Publications; Thousand Oaks. CA: 1985).

Additionally, we tested for multicollinearity in the regression analyses, and all VIF indices were under the value of 5.

A: Many thanks for the comment. We decided to remove the categorization of the variables in the main analyses, and we relied only on the continuous log-transformed variables for the analyses.

Minor issues:

Line 80: Please correct the references and delete the duplicate (10,20,20,21)

Please correct typos, e.g. Internet (line 156), line 455 (64,65,38,66),…

A: Many thanks. We checked the manuscript thoroughly and made the requested corrections.

Reviewer #3: The paper is concisely written and addresses an important topic – the validity of indicators of problematic smartphone use – assessed via digital trace data. Particularly the addressed sample of adolescents is very interesting for early interventions, and thus, data derived from this population group are valuable. Unfortunately, only a very small subsample (89 out of 1419 students) could be recruited for the current study.

A: Many thanks for the comment.

• To assess the representativeness of the sample and the results, it would be interesting to compare the reached subsample with the total sample of the Ethica study.

A: This is a good point and we now added information from difference tests showing that the final on subsample did not differ from the original sample. See Samples section lines 239 to 242.

A: Many thanks for the comment. We included the following sentence in the Limitations section to acknowledge the limitation of the SAS-SV scale when it comes to the clinical setting:

“Finally, although the SAS-SV has been found to predict PSU levels based on the experts' diagnoses, its validation was not carried out for the clinical setting. Hence, the SAS-SV can be used only to detect potentially high-risk groups of PSU, especially in the general population. Future investigations should complement these data with clinical interviews to further validate the use of this scale in assessing symptoms of PSU.”

A: Many thanks for the comment. We agree that this result would be of great interest and complete the picutre. However, when the larger panel study with annual paper-and-pencil questionnaires was designed, we did not include any questions about the frequency of smartphone use. This should be done in future studies interested in the comparison of self-report and digital trace data.

________________________________________

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(32.6KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0263815.r003

Decision Letter 1

Michael Kaess

15 Oct 2021

PONE-D-21-09423R1Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data studyPLOS ONE

Dear Dr. Marciano,

Please submit your revised manuscript by Nov 29 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Michael Kaess, M. D.

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #1: (No Response)

Reviewer #2: Review on the revised manuscript “Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study”.

The authors used subjective and objective measures of smartphone use at two assessments with a one-year interval and examined the association between the subjective and objective measures, as well as the associations between these measures and the problematic smartphone use score. I have seen the paper before and the authors have been very responsive to the reviewers’ comments. I appreciate the thought and energy that went into this revision. I have only minor comments left:

Could you please clarify the meaning of the reported rs on pages 11 and 12 in the measures section – it is not obvious to the readers

Concerning the descriptions of the questionnaire measures: please keep the descriptions consistent – for some questionnaires (e.g., Children's Social Desirability Short Scale), you report the mean and SD, whereas for others you don’t (e.g., SAS-SV).

Language: the manuscript should be carefully proof-read before publication (maybe by a native speaker)

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Philip Santangelo

Reviewer #3: No

PLoS One. 2022 Feb 18;17(2):e0263815. doi: 10.1371/journal.pone.0263815.r004

Author response to Decision Letter 1

1 Dec 2021

PONE-D-21-09423R1

Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study

PLOS ONE

Dear Dr. Marciano,

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Michael Kaess, M. D.

Academic Editor

PLOS ONE

Journal Requirements:

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

________________________________________

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

6. Review Comments to the Author

Reviewer #1: (No Response)

Reviewer #2: Review on the revised manuscript “Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study”.

Could you please clarify the meaning of the reported rs on pages 11 and 12 in the measures section – it is not obvious to the readers

Authors: We have added a clarification that the r concerns the correlation between weekday and weekend day use.

Authors: Many thanks for this observation. We provided the mean and SD for the SAS-SV in the results section, but, for consistency, now added this information also in the measures section.

Language: the manuscript should be carefully proof-read before publication (maybe by a native speaker)

Authors: We performed another careful language check of the manuscript.

Reviewer #3: (No Response)

________________________________________

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Philip Santangelo

Reviewer #3: No

________________________________________

In compliance with data protection regulations, you may request that we remove your personal registration details at any time. (Remove my information/details). Please contact the publication office if you have any questions.

Attachment

Submitted filename: REVISION_2.docx

Click here for additional data file.^{(19.2KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0263815.r005

Decision Letter 2

Michael Kaess

28 Jan 2022

Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study

PONE-D-21-09423R2

Dear Dr. Marciano,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Michael Kaess, M. D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0263815.r006

Acceptance letter

Michael Kaess

3 Feb 2022

PONE-D-21-09423R2

Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study

Dear Dr. Marciano:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Dr. Michael Kaess

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Bivariate Pearson’s correlations among predictor and outcome variables for a weekday.

(DOCX)

Click here for additional data file.^{(15.4KB, docx)}

S2 Table. Bivariate Pearson’s correlations among predictor and outcome variables for a weekend day.

(DOCX)

Click here for additional data file.^{(15.3KB, docx)}

S3 Table. Spearman’s rank correlations between items of the SAS-SV at T1, traced duration and frequency of smartphone use, and difference index.

(DOCX)

Click here for additional data file.^{(16.7KB, docx)}

S4 Table. Spearman’s rank correlations between items of the SAS-SV at T2, trace duration and frequency of smartphone use, and difference index.

(DOCX)

Click here for additional data file.^{(16.8KB, docx)}

S5 Table. Regression results for PSU at T1 and T2 including smartphone use on a weekday.

(DOCX)

Click here for additional data file.^{(14.4KB, docx)}

S6 Table. Regression results for PSU at T1 and T2 including smartphone use on a weekend day.

(DOCX)

Click here for additional data file.^{(14.4KB, docx)}

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(32.6KB, docx)}

Attachment

Submitted filename: REVISION_2.docx

Click here for additional data file.^{(19.2KB, docx)}

Data Availability Statement

The study data and questionnaire are available here: https://osf.io/hwr2u

[pone.0263815.ref001] 1.Choi A. Emotional well-being of children and adolescents: Recent trends and relevant factors. 2018. Feb 9 [cited 2021 Mar 17]; Available from: https://www.oecd-ilibrary.org/education/emotional-well-being-of-children-and-adolescents_41576fb2-en. [Google Scholar]

[pone.0263815.ref002] 2.Fardouly J, Vartanian LR. Social Media and Body Image Concerns: Current Research and Future Directions. Curr Opin Psychol. 2016. Jun 1;9:1–5. [Google Scholar]

[pone.0263815.ref003] 3.Griffiths LJ, Dowda M, Dezateux C, Pate R. Associations between sport and screen-entertainment with mental health problems in 5-year-old children. Int J Behav Nutr Phys Act 2010. Apr 21;7:30. doi: 10.1186/1479-5868-7-30 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref004] 4.Hale L, Guan S. Screen time and sleep among school-aged children and adolescents: a systematic literature review. Sleep Med Rev. 2015. Jun;21:50–8. doi: 10.1016/j.smrv.2014.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref005] 5.Sohn S, Rees P, Wildridge B, Kalk NJ, Carter B. Prevalence of problematic smartphone usage and associated mental health outcomes amongst children and young people: a systematic review, meta-analysis and GRADE of the evidence. BMC Psychiatry. 2019. Nov 29;19(1):356. doi: 10.1186/s12888-019-2350-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref006] 6.Stiglic N, Viner RM. Effects of screentime on the health and well-being of children and adolescents: a systematic review of reviews. BMJ Open. 2019. Jan 1;9(1):e023191. doi: 10.1136/bmjopen-2018-023191 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref007] 7.Walsh JJ, Barnes JD, Cameron JD, Goldfield GS, Chaput J-P, Gunnell KE, et al. Associations between 24 hour movement behaviours and global cognition in US children: a cross-sectional observational study. Lancet Child Adolesc Health. 2018. Nov;2(11):783–91. doi: 10.1016/S2352-4642(18)30278-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref008] 8.Throuvala MA, Griffiths MD, Rennoldson M, Kuss DJ. Motivational processes and dysfunctional mechanisms of social media use among adolescents: a qualitative focus group study. Comput Hum Behav. 2019. Apr;93:164–75. [Google Scholar]

[pone.0263815.ref009] 9.Mascia ML, Agus M, Penna MP. Emotional Intelligence, Self-Regulation, Smartphone Addiction: Which Relationship With Student Well-Being and Quality of Life? Front Psychol [Internet]. 2020. [cited 2021 Mar 17];11. Available from: https://www.readcube.com/articles/10.3389%2Ffpsyg.2020.00375. doi: 10.3389/fpsyg.2020.00375 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref010] 10.Camerini A-L, Marciano L. Media use and academic achievement. In: The International Encyclopedia of Media Psychology [Internet]. J. Van den Bulck. J. VWiley-Blackwell; 2020. [cited 2020 Aug 25]. Available from: https://blackwells.co.uk/bookshop/product/The-International-Encyclopedia-of-Media-Psychology-by-Jan-Van-den-Bulck-editor/9781119011064. [Google Scholar]

[pone.0263815.ref011] 11.Verbeij T, Pouwels JL, Beyens I, Valkenburg PM. The accuracy and validity of self-reported social media use measures among adolescents. Comput Hum Behav Rep. 2021. Jan 1;3:100090. [Google Scholar]

[pone.0263815.ref012] 12.Howison J, Wiggins A, Crowston K. Validity issues in the use of social network analysis with digital trace data. J Assoc Inf Syst. 2011;12(12):767–97. [Google Scholar]

[pone.0263815.ref013] 13.Süss DD, Waller G, Jael B, Lilian S, Gregor W, Céline K, et al. Rapporto sui risultati dello studio JAMES 2020. 2020;76. [Google Scholar]

[pone.0263815.ref014] 14.Mireku MO, Mueller W, Fleming C, Chang I, Dumontheil I, Thomas MSC, et al. Total recall in the SCAMP cohort: Validation of self-reported mobile phone use in the smartphone era. Environ Res. 2018. Feb;161:1–8. doi: 10.1016/j.envres.2017.10.034 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref015] 15.Inyang I, Benke G, Morrissey J, McKenzie R, Abramson M. How well do adolescents recall use of mobile telephones? Results of a validation study. BMC Med Res Methodol. 2009. Jun 12;9:36. doi: 10.1186/1471-2288-9-36 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref016] 16.Aydin D, Feychting M, Schüz J, Andersen TV, Poulsen AH, Prochazka M, et al. Predictors and overestimation of recalled mobile phone use among children and adolescents. Prog Biophys Mol Biol. 2011. Dec 1;107(3):356–61. doi: 10.1016/j.pbiomolbio.2011.08.013 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref017] 17.Pew Research Center. Teens, Social Media & Technology 2018 [Internet]. Pew Research Center: Internet, Science & Tech. 2018 [cited 2021 Mar 17]. Available from: https://www.pewresearch.org/internet/2018/05/31/teens-social-media-technology-2018/.

[pone.0263815.ref018] 18.Heitmayer M, Lahlou S. Why are smartphones disruptive? An empirical study of smartphone use in real-life contexts. Comput Hum Behav. 2021. Mar;116:106637. [Google Scholar]

[pone.0263815.ref019] 19.Elhai JD, Dvorak RD, Levine JC, Hall BJ. Problematic smartphone use: A conceptual overview and systematic review of relations with anxiety and depression psychopathology. J Affect Disord. 2017. Jan;207:251–9. doi: 10.1016/j.jad.2016.08.030 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref020] 20.OECD. Children & Young People’s Mental Health in the Digital Age from: shaping the future [Internet]. 2018. Available from: https://www.oecd.org/health/mental-health.htm.

[pone.0263815.ref021] 21.American Psychiatric Association, editor. Diagnostic and statistical manual of mental disorders: DSM-5. 5th ed. Washington, D.C: American Psychiatric Association; 2013. 947 p. [Google Scholar]

[pone.0263815.ref022] 22.Rosenberg KP. Behavioral addictions: Criteria, evidence, and treatment. Academic Press; 2014. [Google Scholar]

[pone.0263815.ref023] 23.Kwon M, Lee J-Y, Won W-Y, Park J-W, Min J-A, Hahn C, et al. Development and Validation of a Smartphone Addiction Scale (SAS). PLOS ONE. 2013. Feb 27;8(2):e56936. doi: 10.1371/journal.pone.0056936 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref024] 24.Lin Y-H, Chang L-R, Lee Y-H, Tseng H-W, Kuo TBJ, Chen S-H. Development and Validation of the Smartphone Addiction Inventory (SPAI). PLOS ONE. 2014. Jun 4;9(6):e98312. doi: 10.1371/journal.pone.0098312 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref025] 25.Shaw M, Black DW. Internet addiction: definition, assessment, epidemiology and clinical management. CNS Drugs. 2008;22(5):353–65. doi: 10.2165/00023210-200822050-00001 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref026] 26.Škařupová K, Ólafsson K, Blinka L. The effect of smartphone use on trends in European adolescents’ excessive Internet use. Behav Inf Technol. 2016. Jan 2;35(1):68–74. [Google Scholar]

[pone.0263815.ref027] 27.Jo YS, Bhang SY, Choi JS, Lee HK, Lee SY, Kweon Y-S. Clinical Characteristics of Diagnosis for Internet Gaming Disorder: Comparison of DSM-5 IGD and ICD-11 GD Diagnosis. J Clin Med [Internet]. 2019. Jun 28 [cited 2020 Aug 4];8(7). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6678371/. doi: 10.3390/jcm8070945 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref028] 28.Petry DN. Behavioral Addictions: DSM-5® and Beyond. Oxford University Press; 2015. 257 p. [Google Scholar]

[pone.0263815.ref029] 29.Bragazzi NL, Del Puente G. A proposal for including nomophobia in the new DSM-V. Psychol Res Behav Manag. 2014. May 16;7:155–60. doi: 10.2147/PRBM.S41386 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref030] 30.Harris B, Regan T, Schueler J, Fields SA. Problematic Mobile Phone and Smartphone Use Scales: A Systematic Review. Front Psychol [Internet]. 2020. May 5 [cited 2020 Sep 7];11. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7214716/. doi: 10.3389/fpsyg.2020.00672 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref031] 31.Busch P, McCarthy S. Antecedents and consequences of problematic smartphone use: A systematic literature review of an emerging research area. Comput Hum Behav. 2021. [Google Scholar]

[pone.0263815.ref032] 32.Billieux J, Maurage P, Lopez-Fernandez O, Kuss DJ, Griffiths MD. Can Disordered Mobile Phone Use Be Considered a Behavioral Addiction? An Update on Current Evidence and a Comprehensive Model for Future Research. Curr Addict Rep. 2015. Jun 1;2(2):156–62. [Google Scholar]

[pone.0263815.ref033] 33.Panova T, Carbonell X. Is smartphone addiction really an addiction? J Behav Addict. 2018. Jun 1;7(2):252–9. doi: 10.1556/2006.7.2018.49 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref034] 34.de Reuver M, Bouwman H. Dealing with self-report bias in mobile Internet acceptance and usage studies. Inf Manage. 2015. Apr 1;52(3):287–94. [Google Scholar]

[pone.0263815.ref035] 35.Krumpal I. Determinants of social desirability bias in sensitive surveys: a literature review. Qual Quant. 2013. Jun 1;47(4):2025–47. [Google Scholar]

[pone.0263815.ref036] 36.Moorman RH, Podsakoff PM. A meta-analytic review and empirical test of the potential confounding effects of social desirability response sets in organizational behaviour research. J Occup Organ Psychol. 1992;65(2):131–49. [Google Scholar]

[pone.0263815.ref037] 37.Boase J, Ling R. Measuring Mobile Phone Use: Self-Report Versus Log Data. J Comput-Mediat Commun. 2013;18(4):508–19. [Google Scholar]

[pone.0263815.ref038] 38.Friedman ER. Judgments of Time Intervals by Young Children. Percept Mot Skills. 1977. Dec 1;45(3):715–20. doi: 10.2466/pms.1977.45.3.715 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref039] 39.Ford LH. A social desirability questionnaire for young children. J Consult Clin Psychol. 1970;35(2):195–204. [Google Scholar]

[pone.0263815.ref040] 40.Adams AS, Soumerai SB, Lomas J, Ross-Degnan D. Evidence of self-report bias in assessing adherence to guidelines. Int J Qual Health Care J Int Soc Qual Health Care. 1999. Jun;11(3):187–92. doi: 10.1093/intqhc/11.3.187 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref041] 41.Donaldson SI, Grant-Vallone EJ. Understanding Self-Report Bias in Organizational Behavior Research. J Bus Psychol. 2002. Dec 1;17(2):245–60. [Google Scholar]

[pone.0263815.ref042] 42.Lee E, Hu MY, Toh RS. Are Consumer Survey Results Distorted? Systematic Impact of Behavioral Frequency and Duration on Survey Response Errors. J Mark Res. 2000. Feb 1;37(1):125–33. [Google Scholar]

[pone.0263815.ref043] 43.Slater MD. Operationalizing and Analyzing Exposure: The Foundation of Media Effects Research. Journal Mass Commun Q. 2004. Mar 1;81(1):168–83. [Google Scholar]

[pone.0263815.ref044] 44.Parry DA, Davidson BI, Sewall C, Fisher JT, Mieczkowski H, Quintana D. A Systematic Review and Meta-Analysis of Discrepancies Between Logged and Self- Reported Digital Media Use [Internet]. PsyArXiv; 2020. [cited 2021 Mar 22]. Available from: https://psyarxiv.com/f6xvz/. [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref045] 45.Harari GM, Müller SR, Aung MS, Rentfrow PJ. Smartphone sensing methods for studying behavior in everyday life. Curr Opin Behav Sci. 2017. Dec 1;18:83–90. [Google Scholar]

[pone.0263815.ref046] 46.Lauricella AR, Roberts M, Curtis P. Using LENA to understand children’s media use: Challenges and opportunities. In San Diego, CA, US; 2017.

[pone.0263815.ref047] 47.Stier S, Breuer J, Siegers P, Thorson K. Integrating Survey Data and Digital Trace Data: Key Issues in Developing an Emerging Field. Soc Sci Comput Rev. 2020. Oct 1;38(5):503–16. [Google Scholar]

[pone.0263815.ref048] 48.Brickhman D, Scandurra L, Powell K, Rich MO. Assessing the agreement of multiple methods of measuring children’s media use in U.S. and Mexican samples. In San Diego, CA, US; 2017.

[pone.0263815.ref049] 49.GDPR. General Data Protection Regulation (GDPR)–Official Legal Text [Internet]. General Data Protection Regulation (GDPR). 2018 [cited 2021 Mar 17]. Available from: https://gdpr-info.eu/.

[pone.0263815.ref050] 50.Lin Y-H, Chiang C-L, Lin P-H, Chang L-R, Ko C-H, Lee Y-H, et al. Proposed Diagnostic Criteria for Smartphone Addiction. Weinstein AM, editor. PLOS ONE. 2016. Nov 15;11(11):e0163010. doi: 10.1371/journal.pone.0163010 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref051] 51.Reeves B, Ram N, Robinson TN, Cummings JJ, Giles CL, Pan J, et al. Screenomics: A Framework to Capture and Analyze Personal Life Experiences and the Ways that Technology Shapes Them. Human–Computer Interact. 2021. Mar 4;36(2):150–201. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref052] 52.Radesky JS. Mobile device passive sensing to measure parent and child smartphone and tablet usage. In San Diego, CA, US; 2017.

[pone.0263815.ref053] 53.Goedhart G, Vrijheid M, Wiart J, Hours M, Kromhout H, Cardis E, et al. Using software-modified smartphones to validate self-reported mobile phone use in young people: A pilot study. Bioelectromagnetics. 2015. Oct;36(7):538–43. doi: 10.1002/bem.21931 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref054] 54.Brinberg M, Ram N, Yang X, Cho M-J, Sundar SS, Robinson TN, et al. The idiosyncrasies of everyday digital lives: Using the Human Screenome Project to study user behavior on smartphones. Comput Hum Behav. 2021. Jan 1;114:106570. doi: 10.1016/j.chb.2020.106570 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref055] 55.Ryding FC, Kuss DJ. Passive objective measures in the assessment of problematic smartphone use: A systematic review. Addict Behav Rep. 2020. Jun 1;11:100257. doi: 10.1016/j.abrep.2020.100257 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref056] 56.Lin Y-H, Lin Y-C, Lee Y-H, Lin P-H, Lin S-H, Chang L-R, et al. Time distortion associated with smartphone addiction: Identifying smartphone addiction via a mobile application (App). J Psychiatr Res. 2015. Jun;65:139–45. doi: 10.1016/j.jpsychires.2015.04.003 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref057] 57.Ko C-H, Yen J-Y, Yen C-F, Lin H-C, Yang M-J. Factors predictive for incidence and remission of internet addiction in young adolescents: a prospective study. Cyberpsychology Behav Impact Internet Multimed Virtual Real Behav Soc. 2007. Aug;10(4):545–51. doi: 10.1089/cpb.2007.9992 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref058] 58.Tossell C, Kortum P, Shepard C, Rahmati A, Zhong L. Exploring Smartphone Addiction: Insights from Long-Term Telemetric Behavioral Measures. Int J Interact Mob Technol IJIM. 2015. Mar 23;9(2):37–43. [Google Scholar]

[pone.0263815.ref059] 59.Noe B, Turner LD, Linden DEJ, Allen SM, Winkens B, Whitaker RM. Identifying Indicators of Smartphone Addiction Through User-App Interaction. Comput Hum Behav. 2019. Oct;99:56–65. doi: 10.1016/j.chb.2019.04.023 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref060] 60.Camerini A-L, Schulz PJ, Jeannet A-M. The social inequalities of Internet access, its use, and the impact on children’s academic performance: Evidence from a longitudinal study in Switzerland. NEW MEDIA Soc. 2018. Jul;20(7):2489–508. [Google Scholar]

[pone.0263815.ref061] 61.Camerini A-L, Marciano L. Self-selection bias in research including ecological momentary assessment and digital trace data. In: Passive Smartphone Data Collection and Additional Tasks in Mobile Web Surveys: Willingness, Non-Participation, Consent, and Ethics I. Zagreb; 2019.

[pone.0263815.ref062] 62.Mößle T. dick, dumm, abhängig, gewalttätig? Nomos Verlagsgesellschaft mbH & Co. KG; 2012. doi: 10.1002/mrm.24177 [DOI] [Google Scholar]

[pone.0263815.ref063] 63.Thomas Vattukalathil T. Application of data analytics and machine learning on data collected by smartphones to understand human behavioural patterns. 2019. [Google Scholar]

[pone.0263815.ref064] 64.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. Dec 31;8(12):e83558. doi: 10.1371/journal.pone.0083558 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref065] 65.Pritschet L, Powell D, Horne Z. Marginally Significant Effects as Evidence for Hypotheses: Changing Attitudes Over Four Decades. Psychol Sci. 2016. Jul 1;27(7):1036–42. doi: 10.1177/0956797616645672 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref066] 66.Odgers C. Smartphones are bad for some teens, not all. Nature. 2018. 22;554(7693):432–4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref067] 67.Mascia ML, Agus M, Penna MP. Emotional Intelligence, Self-Regulation, Smartphone Addiction: Which Relationship With Student Well-Being and Quality of Life? Front Psychol [Internet]. 2020. Mar 6 [cited 2021 Mar 17];11. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7068808/. doi: 10.3389/fpsyg.2020.00375 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref068] 68.van Velthoven MH, Powell J, Powell G. Problematic smartphone use: Digital approaches to an emerging public health problem. Digit Health. 2018. Jan 1;4:2055207618759167. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref069] 69.Oulasvirta A, Rattenbury T, Ma L, Raita E. Habits make smartphone use more pervasive. Pers Ubiquitous Comput. 2012. Jan 1;16(1):105–14. [Google Scholar]

[pone.0263815.ref070] 70.Montag C, Błaszkiewicz K, Lachmann B, Sariyska R, Andone I, Trendafilov B, et al. Recorded Behavior as a Valuable Resource for Diagnostics in Mobile Phone Addiction: Evidence from Psychoinformatics. Behav Sci. 2015. Dec;5(4):434–42. doi: 10.3390/bs5040434 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref071] 71.Parry DA, Davidson BI, Sewall CJR, Fisher JT, Mieczkowski H, Quintana DS. A systematic review and meta-analysis of discrepancies between logged and self-reported digital media use. Nat Hum Behav. 2021. May 17;1–13. doi: 10.1038/s41562-021-01049-0 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref072] 72.Al-Barashdi HS, Bouazza A, Jabur NH. Smartphone Addiction among University Undergraduates: A Literature Review. J Sci Res Rep. 2015;210–25. [Google Scholar]

[pone.0263815.ref073] 73.De-Sola Gutiérrez J, Rodríguez de Fonseca F, Rubio G. Cell-Phone Addiction: A Review. Front Psychiatry [Internet]. 2016. [cited 2021 Jan 15];7. Available from: https://www.frontiersin.org/articles/10.3389/fpsyt.2016.00175/full. doi: 10.3389/fpsyt.2016.00175 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref074] 74.Haug S, Castro RP, Kwon M, Filler A, Kowatsch T, Schaub MP. Smartphone use and smartphone addiction among young people in Switzerland. J Behav Addict. 2015. Dec 1;4(4):299–307. doi: 10.1556/2006.4.2015.037 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref075] 75.Martini M, Heinz A, Hinterholzer J, Martini C, Sachse P. Effects of wakeful resting versus social media usage after learning on the retention of new memories. Appl Cogn Psychol. 2020;34(2):551–8. [Google Scholar]

[pone.0263815.ref076] 76.Marty-Dugas J, Ralph BCW, Oakman JM, Smilek D. The relation between smartphone use and everyday inattention. Psychol Conscious Theory Res Pract. 2018;5(1):46–62. [Google Scholar]

[pone.0263815.ref077] 77.Contractor AA, Weiss NH, Tull MT, Elhai JD. PTSD’s relation with problematic smartphone use: Mediating role of impulsivity. Comput Hum Behav. 2017. Oct 1;75:177–83. [Google Scholar]

[pone.0263815.ref078] 78.Kim Y, Jeong J-E, Cho H, Jung D-J, Kwak M, Rho MJ, et al. Personality Factors Predicting Smartphone Addiction Predisposition: Behavioral Inhibition and Activation Systems, Impulsivity, and Self-Control. PLOS ONE. 2016. Aug 17;11(8):e0159788. doi: 10.1371/journal.pone.0159788 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref079] 79.Przybylski AK, Murayama K, DeHaan CR, Gladwell V. Motivational, emotional, and behavioral correlates of fear of missing out. Comput Hum Behav. 2013. Jul 1;29(4):1841–8. [Google Scholar]

[pone.0263815.ref080] 80.Gezgin DM. Understanding Patterns for Smartphone Addiction: Age, Sleep Duration, Social Network Use and Fear of Missing Out. Cypriot J Educ Sci. 2018;13(2):166–77. [Google Scholar]

[pone.0263815.ref081] 81.Wang J, Wang P, Yang X, Zhang G, Wang X, Zhao F, et al. Fear of Missing Out and Procrastination as Mediators Between Sensation Seeking and Adolescent Smartphone Addiction. Int J Ment Health Addict. 2019. Aug 1;17(4):1049–62. [Google Scholar]

[pone.0263815.ref082] 82.Elhai JD, Levine JC, Dvorak RD, Hall BJ. Fear of missing out, need for touch, anxiety and depression are related to problematic smartphone use. Comput Hum Behav. 2016;63:509–16. [Google Scholar]

[pone.0263815.ref083] 83.Turel O, Brevers D, Bechara A. Time distortion when users at-risk for social media addiction engage in non-social media tasks. J Psychiatr Res. 2018. Feb 1;97:84–8. doi: 10.1016/j.jpsychires.2017.11.014 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref084] 84.Verbeij T, Pouwels JL, Beyens I, Valkenburg PM. The Accuracy and Validity of Self-Reported Social Media Use Measures Among Adolescents [Internet]. PsyArXiv; 2021. [cited 2021 Mar 17]. Available from: https://psyarxiv.com/p4yb2/. [Google Scholar]

[pone.0263815.ref085] 85.Takahashi T, Oono H, Radford MHB. Psychophysics of time perception and intertemporal choice models. Phys Stat Mech Its Appl. 2008. Mar 15;387(8):2066–74. [Google Scholar]

[pone.0263815.ref086] 86.Wittmann M, Paulus MP. Decision making, impulsivity and time perception. Trends Cogn Sci. 2008. Jan;12(1):7–12. doi: 10.1016/j.tics.2007.10.004 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref087] 87.Rau P-LP, Peng S-Y, Yang C-C. Time distortion for expert and novice online game players. Cyberpsychology Behav Impact Internet Multimed Virtual Real Behav Soc. 2006. Aug;9(4):396–403. doi: 10.1089/cpb.2006.9.396 [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref088] 88.Block RA, Gruber RP. Time perception, attention, and memory: A selective review. Acta Psychol (Amst). 2014. Jun 1;149:129–33. [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref089] 89.Lejeune H, Wearden JH. Vierordt’s The Experimental Study of the Time Sense (1868) and its legacy. Eur J Cogn Psychol. 2009. Sep 1;21(6):941–60. [Google Scholar]

[pone.0263815.ref090] 90.Tobin S, Grondin S. Video games and the perception of very long durations by adolescents. Comput Hum Behav. 2009. Mar 1;25(2):554–9. [Google Scholar]

[pone.0263815.ref091] 91.Youn S, Hall K. Gender and online privacy among teens: risk perception, privacy concerns, and protection behaviors. Cyberpsychology Behav Impact Internet Multimed Virtual Real Behav Soc. 2008. Dec;11(6):763–5. [DOI] [PubMed] [Google Scholar]

[pone.0263815.ref092] 92.Ohme J, Araujo T, de Vreese CH, Piotrowski JT. Mobile data donations: Assessing self-report accuracy and sample biases with the iOS Screen Time function. Mob Media Commun. 2020. Sep 30;2050157920959106. [Google Scholar]

[pone.0263815.ref093] 93.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 [Internet]. 2019. Jun;9. Available from: http://search.ebscohost.com/login.aspx?direct=true&db=psyh&AN=2019-09430-001&site=ehost-live. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0263815.ref094] 94.Süss DD, Waller G, Lilian S, Gregor W, Jael B, Céline K, et al. Findings of the JAMES Study 2018. 2018;82. [Google Scholar]

[pone.0263815.ref095] 95.Carvalho LF, Sette CP, Ferrari BL. Problematic smartphone use relationship with pathological personality traits: Systematic review and meta-analysis. Cyberpsychology J Psychosoc Res Cyberspace [Internet]. 2018. Dec 31 [cited 2021 Aug 16];12(3). Available from: https://cyberpsychology.eu/article/view/11423. [Google Scholar]

PERMALINK

Duration, frequency, and time distortion: Which is the best predictor of problematic smartphone use in adolescents? A trace data study

Laura Marciano

Anne-Linda Camerini

Roles

Abstract

Introduction

Literature review

Problematic smartphone use

Overcoming self-report biases

Investigating PSU using objective trace data

Study aim

Methodology

Data collection

Survey data

Digital trace data

Ethical considerations

Sample

Measures

Perceived duration of smartphone use

Trace data for the duration of smartphone use

Trace data for frequency of smartphone use

Problematic smartphone use

Social desirability

Analytical plan

Results

Descriptive results

Fig 1. Representation of traced and self-reported estimates (in hours per day) of the duration of smartphone use for a general day, a weekday, and a weekend day.

Accuracy of self-report measures of smartphone use

Preliminary correlations with problematic smartphone use

Table 1. Bivariate correlations among predictor and outcome variables for a general day.

Regression results

Table 2. Regression results for predicting PSU at T1 and T2 from traced smartphone use on a general day.

Discussion

Limitations and recommendations for future research

Conclusions

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Michael Kaess

Roles

Author response to Decision Letter 0

Decision Letter 1

Michael Kaess

Roles

Author response to Decision Letter 1

Decision Letter 2

Michael Kaess

Roles

Acceptance letter

Michael Kaess

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases