Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Aug 1.
Published in final edited form as: J Adolesc. 2024 May 3;96(6):1171–1181. doi: 10.1002/jad.12326

Daily links between Objective Smartphone Use and Sleep among Adolescents

Kaitlyn Burnell 1, Shedrick L Garrett 1, Benjamin W Nelson 1, Mitchell J Prinstein 1, Eva H Telzer 1
PMCID: PMC11303118  NIHMSID: NIHMS1983767  PMID: 38698757

Abstract

Introduction.

Concerns abound on how digital technology such as smartphone use may impair adolescent sleep. Although these linkages are supported in cross-sectional studies, research involving intensive longitudinal assessments and objective measures has called into question the robustness of associations.

Methods.

In this study, a sample of ethnically diverse U.S. adolescents (N=71; Mage=16.49; 56% girls) wore Fitbit devices and submitted screenshots of their smartphone screen time, pickups, and notifications over a fourteen-day period in 2021. The Fitbits recorded nightly sleep quality and sleep onset. Adolescents also completed daily diaries reporting the previous night’s sleep onset time and sleep quality.

Results.

On days when adolescents engaged in greater nighttime screen time and, to some extent, pickups relative to their own average, they also had poorer sleep outcomes that night. Greater screen time was associated with later self-reported and Fitbit-recorded sleep onset and poorer self-reported sleep quality. Greater pickups was associated with later self-reported and Fitbit-recorded sleep onset. Smartphone use during the day did not relate to sleep outcomes, indicating the importance of distinguishing nighttime from daytime use.

Conclusions.

Parents and clinicians should help adolescents develop healthy digital skills to avoid exacerbating sleep problems that are known to occur during this developmental period.

Keywords: Sleep, Adolescence, Smartphones

Greater digital technology use has been linked with poorer sleep (Brautsch et al., 2023). These linkages are particularly concerning for adolescents. Adolescents are heavy consumers of digital technology and are often at the forefront of adopting new media such as smartphones (Valkenburg & Piotrowski, 2017). Adolescent sleep health is also notoriously poor, with many contemporary adolescents getting insufficient sleep (Keyes et al., 2015). These recent declines coincide with the development and popularity surge of new media devices, including smartphones. Because of this, digital technology may be a cause for insufficient sleep. Most past work has relied on cross-sectional and self-report data (Alonzo et al., 2021). Some studies applying more rigorous methods have raised questions about the robustness of the linkages between digital technology use and sleep outcomes, arguing that associations may not replicate when using objective sleep measures or parsing out between- from within-person differences (Burnell et al., 2022; Mac Cárthaigh et al., 2022). Additional research is sorely needed to fully understand within-person associations between objective digital technology use and sleep. This study utilized objective measures of smartphone use and sleep to examine these associations.

Smartphone Use and Sleep During Adolescence

Nearly all (98%) of U.S. adolescents aged 15 – 17 have access to a smartphone (Vogels et al., 2022). The portability of these devices allows users to engage with them anytime, anywhere. This includes in the bedroom, with almost 70% of U.S. adolescents keeping their mobile device in bed with them or within reach while they sleep (Robb, 2019). Most adolescents use their devices before going to sleep and first thing after waking up, and some report waking up due to a notification (Robb, 2019). Because of these behaviors, there is concern that smartphones and other mobile devices may be interfering with sleep health through these displacement-related mechanisms (LeBourgeois et al., 2017).

The link between smartphone use and sleep among late adolescents may be particularly developmentally important. Late adolescents (ages 15 to 17) are more likely to have smartphones than younger adolescents (between ages 13 and 14; Vogels et al., 2022), and insufficient sleep tends to be higher (Gradisar et al., 2011). Importantly, the behaviors that late adolescents engage in may follow them into emerging adulthood, where they typically gain more independence (Arnett, 2000). With greater freedom over smartphone use and sleep schedules, some individuals may struggle with establishing healthy smartphone habits that avoid disrupting sleep. Thus, the late adolescent period may be particularly important to study.

Numerous studies have demonstrated how smartphone use relates to sleep outcomes, although conclusions are mixed. On the one hand, there is limited support for linkages between adolescent mobile phone use and both self-reported and objectively-recorded sleep outcomes (Cabré-Riera et al., 2019). Similar mixed associations were observed in a study using objective measures of sleep onset and sleep efficiency (Fobian et al., 2016). On the other hand, an ecological momentary assessment study with adolescents found within-person evidence in that those who texted more than their own average on a given day logged shorter sleep hours that night (Tavernier et al., 2017). Greater frequency of texting has been found to also be associated with later sleep onset (Charmaraman et al., 2021). Additionally, nighttime cell phone notifications and texting frequency are associated with poorer sleep quality among college students, although not when controlling for desire to check nighttime notifications (Murdock et al., 2017).

How Smartphone Use May Affect Sleep

A greater understanding of when smartphone use occurs may be needed to reconcile these mixed findings. Prior research typically has not contrasted daytime and nighttime digital technology use (Brautsch et al., 2023). Nighttime use may be particularly relevant due in part to delayed melatonin release and increases in overstimulation and difficulty disengaging from online social interactions (LeBourgeois et al., 2017). Evidence from intensive longitudinal designs suggests that nighttime digital technology use is generally associated with poorer sleep outcomes (Kubiszewski et al., 2014; Lüscher & Radtke, 2022), and experimental evidence indicates that bedtime phone restriction increases sleep duration and quality (He et al., 2020). However, a rigorous 14-day ecological momentary assessment study among college students did not find that greater bedtime social media use linked with objectively-measured poorer sleep (Das-Friebel et al., 2020). Moreover, the few studies that have directly contrasted daytime from nighttime digital technology use have found that associations with sleep outcomes are generally comparable, with one finding both types of use associated with poorer sleep outcomes (Hysing et al., 2014) and another study generally finding null associations for both daytime and nighttime use (Lee et al., 2021). Thus, more research is needed to explore these potential differences.

The current research examined smartphone screen time, pickups, and notifications. Overall screen time may disrupt sleep through three proposed mechanisms explaining digital technology and sleep relations: displacement, exposure to cognitively arousing content, and blue light emissions (LeBourgeois et al., 2017). The more an adolescent uses their phone (particularly at night), the less time they are spending sleeping, the greater exposure they have to sleep-disrupting light and the greater chance they have to encounter arousing content. Pickups are defined by engagement with one’s smartphone such as through activating or unlocking the device (Mac Cárthaigh et al., 2022). Unlike screen time, which can aggregate both active (e.g., sending messages) and passive (e.g., watching Netflix in the background) use, pickups can capture use that is potentially more intentional and active. Pickups may also affect sleep through displacement, in that an adolescent who picks up their phone during the night cannot be sleeping. Pickups may also prolong typical wake events during a sleep session. For example, if an adolescent wakes up normally in the middle of the night, checks their phone, and finds that they received a text message from their romantic interest, this may trigger cognitive arousal that further disrupts sleep, even if the original pickup intention was to simply check the time. This may also result in additional blue light exposure. Finally, notifications may disrupt sleep in that audible pings may wake up adolescents (Robb, 2019).

We examined both self-reported and Fitbit-recorded sleep onset and sleep quality, with Fitbit-recorded sleep efficiency used as a metric of sleep quality. We examined self-reported sleep domains in addition to collecting objectively-recorded data because self-reports may capture subjective self-evaluations of one’s sleep (e.g., George et al., 2019) and because past research demonstrates potential differences in associations between objective and self-reported sleep outcomes with technology use (Burnell et al., 2022). Sleep onset and sleep quality may be particularly relevant to the proposed mechanisms explaining relations between technology use and sleep. Specifically, displacement, light exposure, and exposure to cognitive arousing content may all delay the time in which an adolescent falls asleep, highlighting the need to examine sleep onset. Similarly, nighttime smartphone use may disrupt sleep quality, such as through receiving sleep-interrupting notifications (Robb, 2019) or engaging with cognitive arousing content that may render it difficult to have a restful sleep.

Utilizing Objective Measures to Determine Within- versus Between-Person Associations

Group-level associations often cannot be generalized to the individual-level (Fisher et al., 2018; Molenaar et al., 2009). As most studies examining digital technology use and sleep rely on cross-sectional designs (Alonzo et al., 2021), it is unclear the extent to which observed group-level associations occur at the individual-level. Intensive repeated assessment designs allow for the distinction of within- and between-person associations. Whereas between-person associations can capture if an adolescent who has greater smartphone use on average relative to their peers has poorer sleep outcomes, within-person associations can capture if an adolescent with greater smartphone use on a given day, relative to their own average, has poorer sleep outcomes that night. This method is particularly rigorous in that it controls for between-person differences and potential confounding factors.

An increasing number of studies have utilized ecological momentary assessment (EMA) designs to distinguish within- from between-person associations. These studies have found between-person associations between digital technology use and poorer sleep outcomes, with little evidence supporting within-person associations (Burnell et al., 2022; Hamilton et al., 2022; but also see Tavernier et al., 2017, which found inconsistent support for both between- and within-person associations). In other words, although there is evidence that adolescents who use greater digital technology relative to their peers also have poorer sleep outcomes (between-person), strong support is currently lacking that indicates that, on a given day, an adolescent who uses more digital technology relative to their own average also has poorer sleep outcomes that night (within-person).

An additional methodological issue is measurement. A growing number of studies have utilized objective measures of sleep (e.g., commercial wearable devices; George et al., 2019), but few have paired these assessments with objective measures of digital technology use. Self-reports of digital technology use are notoriously inaccurate (Parry et al., 2021) and may reflect subjective experiences rather than actual use. In the few studies that have employed objective measures of digital technology use, associations with sleep were not robust, inconsistent, or trivial (Foerster et al., 2019; Lee et al., 2021; Mac Cárthaigh et al., 2022).

The Current Research

This study employed a 14-day EMA design in which adolescents were provided with Fitbit devices which logged their nightly sleep (sleep quality and sleep onset) and completed morning diaries of subjective sleep experiences (previous night sleep onset and sleep quality). Participants also uploaded daily screenshots of their previous day’s smartphone total screen time, number of pickups, and number of notifications. We examined within- and between-person associations between daily smartphone screen time, pickups, and notifications and the sleep outcomes. We examined these associations for smartphone use aggregated across a full day, and separately for daytime and nighttime use. We expected that daytime smartphone use would not be linked with sleep outcomes at the between- or within-person level. In contrast, we predicted that nighttime smartphone use would be associated with self-reported sleep outcomes at the between-person level; given that past research is inconsistent in terms of associations 1) at the within-person level and 2) with objective sleep outcomes, these analyses were treated as exploratory. Additionally, it was expected that associations would be more robust for smartphone screen time and pickups, as these behaviors (as measured in the current research) may be more directly related to sleep-disrupting mechanisms such as displacement, blue light exposure, and cognitively arousing content exposure than notifications.

Method

Participants

Participants included 71 adolescents recruited from the southeastern United States (aged 15 to 18; Mage=16.49; SDage=0.63). The sample was 56% girls, 42% boys, and 1% non-binary, and 34% Hispanic, 30% White, 26% Black/African American, and 10% Multiracial/Other. Average parent-reported income was $45,000 to $59,999. Participants were recruited from a larger longitudinal study examining adolescent digital technology use, with the analytic sample part of Wave 5 of the study. Wave 5 took place primarily in 2021 and involved an initial sample of 103 adolescents. All included participants had at least two consecutive days of smartphone use data and at least one day of self-reported sleep data; 64 had Fitbit data. Participants were excluded if they were Android users (n=16; as smartphone data were pulled from the iOS Screen Time app), did not complete the 14-day EMA period (n=8), or were iOS users but did not have smartphone use data (n=8; all participants with smartphone use data had at least two consecutive days). The analytic sample of 71 did not differ from non-participating Wave 5 adolescents on age, gender, race/ethnicity, and income (ps>.105).

Procedure

Participants were delivered daily surveys for fourteen days using the ExpiWell app. Self-report measures relevant to the current study were collected during the morning survey, which was delivered at a random time between 9:00am and 12:59pm. Participants had two hours to complete the survey after receiving the survey prompt. Participants uploaded screenshots of yesterday’s smartphone use during a survey delivered at 1:00pm and had until 11:59pm to upload. Assent and consent from all participants and their parents (for those under 18 years) was obtained. Study procedures were approved by the University of North Carolina at Chapel Hill Institutional Review Board.

Measures

Smartphone Use

Once a day, adolescents uploaded screenshots of the previous day’s screen time (ICC=.58), pickups (ICC=.74), and notifications (ICC=.77), as logged by the iOS Screen Time app. Screenshots included a figure showing amount of use (for each variable) each day, starting at the 12:00am hour and ending at (and including) the 11:00pm hour. The values corresponding to each hour were extracted using a vector graphics editor, Adobe Photoshop, by trained research assistants. Extracted hourly values were summed across days and compared to the values from the iOS Screen Time app output. Days with a discrepancy greater than +/−10 minutes of screen time, pickups, or notifications were identified and re-extracted. The final agreement between the extracted and outputted screen time, pickup, and notification values was over 99.9%.

Screen time, pickups, and notifications were summed to reflect daily levels of each variable. We anticipated (and found support in our data) that many adolescents would go to bed after midnight. Because of this, smartphone use within a standard 24-hour day (with use ranging from 12:00am and 11:59pm) may not serve as the most accurate predictor for sleep outcomes, as smartphone use may occur after midnight and before sleep onset. We defined a day as ranging from 6:00am to 5:59am the following day, as 6:00am was usually the latest time that participants went to bed (only 3% of cases occurred after 6:00am) and the earliest time that participants woke up (only 8% of cases occurred before 6:00am; of these, half occurred between 5:45am and 6:00am). Because two consecutive days of smartphone screen shots were required for use to be properly aggregated, missingness was moderate (39% missing for screen time, 39% for pickups, 41% for notifications). Participants had on average 7.75 days of smartphone data (SD=4.21). Data uploads were generally not significantly correlated with demographics or study variables (ps>.110), except that girls (M=8.68) uploaded more screenshots than boys (M=6.33, p=.044).

Self-Reported Sleep

Participants responded to items, “Approximately what time did you fall asleep last night?” (self-reported sleep onset; ICC=.53) and “How well did you sleep last night?” (self-reported sleep quality; ICC=.30). Self-reported sleep onset ranged on a 1 (earlier than 8 pm) to 9 (later than 2 am) scale. Self-reported sleep quality ranged on a 1 (extremely well) to 4 (extremely bad) scale; the item was reverse coded so that higher values reflected better sleep quality. Missingness was present for 15% of cases for both variables in primary analyses. Participants had on average 9.55 days of self-report data (SD=3.91). Number of days with self-report data was generally not significantly correlated with demographics or study variables (ps>.060), except that girls (M=10.40) completed more surveys than boys (M=8.53, p=.047).

Fitbit-Recorded Sleep

Participants wore Fitbit Inspire 2 devices over the course of the EMA period. Past research has indicated that wearable devices perform similarly to research actigraphy methods (Lee et al., 2019). These devices provided participants’ total sleep quality (as measured by sleep efficiency, an assessment of restful and uninterrupted sleep, ICC=.36) and time of sleep onset (ICC=.46) on a given day. Missingness was present on 29% of cases for sleep onset and 19% of cases for sleep quality1. Participants averaged 11.45 days of Fitbit data (SD = 4.62). Days of data did not correlate with key study variables or demographics (ps > .070). We focus on sleep quality and sleep onset as these variables parallel our self-reported sleep domain variables. Data were also available on sleep duration, with within- and between-person associations presented in the supplement (Table S1).

Analytic Plan

Descriptive statistics and correlations (Table 1) were run in SPSS Version 28 (IBM, 2022). Main analyses were run in MPlus version 8.8 (Muthén & Muthén, 1998–2017). Random intercept multilevel models with maximum likelihood estimation were used to test how each smartphone use variable related to each sleep outcome. Because of a high correlation between pickups and notifications, each smartphone use variable was tested in separate models. The four sleep outcomes were also tested in separate models, resulting in twelve models (three smartphone use predictors x four sleep outcomes). These models were run in three sets: first, by examining associations when smartphone use is aggregated across a day (defined as between 6:00am and 5:59am the following calendar day), second, by examining associations when smartphone use is limited to daytime use (defined as use that occurred between 6:00am and 7:59pm2), and third, by examining associations when smartphone use is limited to nighttime use (defined as use that occurred between 8:00pm and 5:59am). In all models, the smartphone use variable of interest was person-centered by subtracting the aggregated person-level variable (in which the smartphone variable of interest was averaged across the 14-day EMA) from the raw daily-level variable (Curran & Bauer, 2011). This variable was included as a within-person predictor. The aggregated person-level variable was included as a between-person predictor. Therefore, each model included a single within-person predictor (e.g., nighttime screen time), a single between-person predictor (e.g., a participant’s averaged nighttime screen time over the EMA), and a single outcome (e.g., Fitbit-recorded sleep onset). Due to the large number of tests in which the twelve models were examined three times (36 total tests), Benjamini-Hochberg corrections were applied (Benjamini & Hochberg, 1995).

Table 1.

Daily Correlations, Means, and Standard Deviations for Key Study Variables

1 2 3 4 5 6 7 8 9 10 11 12 13
1. ST N --
2. ST D .32 --
3. ST Tot .83 .80 --
4. PU N −.11 −.07 −.11 --
5. PU D −.02 .10 .05 .38 --
6. PU Tot −.05 .06 .00 .66 .95 --
7. NOT N .12 .02 .09 .51 .33 .44 --
8. NOT D .00 .13 .08 .35 .62 .62 .65 --
9. NOT Tot .04 .10 .09 .44 .56 .61 .84 .96 --
10. SR SO .23 −.15 .06 .26 −.25 −.11 .21 −.04 .05 --
11. SR Qual −.02 .03 .00 .08 .18 .17 .01 .00 .01 −.21 --
12. FB SO .36 −.04 .19 .18 −.24 −.12 .04 −.11 −.07 .51 −.16 --
13. FB Qual .10 .02 .07 −.06 −.07 −.08 −.03 .03 .01 .07 −.02 .10 --
Mean 229.91 314.24 544.15 32.89 92.75 125.64 57.29 117.90 175.20 6.21 a 3.00 1:29am 93.08
SD 150.71 139.37 235.74 26.89 61.06 75.44 47.87 90.11 126.72 1.95 0.66 -- 3.66
Range 0–600 0–793 92–1356 0–221 0–383 1–497 0–247 0–687 0–928 1–9 1–4 8:18pm – 6:21pmb 73 – 100
Open in a new tab

Note. Bolded correlations are significant at p < .05. ST = Screen Time. PU = Pickups. NOT = Notifications. N = Nighttime. D = Daytime. Tot = Total Use (Summed nighttime and daytime). SR = Self-report. FB = Fitbit. SO = Sleep Onset.

a

Value falls between 12:00am (6) and 1:00am (7) response options.

b

8:00pm was used as a threshold for earliest sleep onset (in that 8:18pm was the earliest sleep onset recorded) and 7:59pm was used as a threshold for latest sleep onset (in that 6:21pm was the latest sleep onset recorded). Nearly all sleep onset data points occurred before 7:30am; one occurred at 9:15am and nine occurred at 10:42am or later. These nine cases were at least three standard deviations from the mean and removed in sensitivity analyses (see Tables 2 and 3).

Participants were included in analyses so long as they had at least one day of data on the outcome of interest. Models with self-reported sleep outcomes were run with the full analytic sample (N=71). Models with Fitbit sleep outcomes were run only for those with Fitbit data (n=64). Missingness occurring within-subject (e.g., if a participant was missing data on Day 6 of the study) was handled with Full Information Maximum Likelihood. Sensitivity analyses were conducted to confirm the robustness of the results for the multilevel models, in that models were run including covariates (school attendance (within-person) and age, gender, income, race/ethnicity and proportion of (completed) study days in which school was attended (between-person)), with outliers recoded to missing, and when limiting objective analyses to participants with three or more days of Fitbit data. Results generally remained the same (Tables 2 and 3).

Table 2.

Within-Person Associations between Smartphone Use and Sleep Outcomes

Total Day Night
b [95% CI] SE p β b [95% CI] SE p β b [95% CI] SE p β
Self-Report Sleep Onset
 ST 0.002 [0.001, 0.003] .00 <.001 .25 0.00 [−0.001, 0.001] .001 .946 .00 0.006 [0.005, 0.007] .001 <.001 .41
 PU 0.002 [−0.002, 0.01] .002 .243 .06 −0.002 [−0.01, 0.002] .002 .258 −.06 0.02 [0.01, 0.02] .004 <.001 .21
 NOT 0.00 [−0.003, 0.002] .001 .712 −.02 −0.003 [−0.01, 0.00] .001 .065 −.09 0.01 [0.002, 0.01] .003 .005 .15
Self-Report Sleep Quality
 ST −0.00 [−0.001, 0.00] .00 .035 −.11 0.00 [−0.001, 0.00] .00 .637 −.02 −0.001 [−0.001, 0.00] .00 .005 −.14
 PU 0.00 [−0.001, 0.002] .001 .659 .02 0.001 [−0.001, 0.002] .001 .525 .03 0.00 [−0.003, 0.003] .001 .803 −.01
 NOT 0.00 [−0.001, 0.001] .00 .774 −.01 0.00 [−0.001, 0.001] .001 .685 −.02 0.00 [−0.002, 0.002] .001 .925 .01
Fitbit Sleep Onset
 ST 0.10 [0.00, 0.20] a .05 .049 .12 −0.07 [−0.20, 0.06] .07 .311 −.06 0.41 [0.26, 0.57] .08 <.001 .30
 PU 0.06 [−0.37, 0.50] .22 .781 .02 −0.38 [−0.89, 0.13] .26 .144 −.10 1.34 [0.50, 2.18] .43 .002 .18
 NOT −0.09 [−0.32, 0.14] .12 .441 −.04 −0.20 [−0.49, 0.09] .15 .176 −.08 0.21 [−0.35, 0.78] .29 .458 .05
Fibit Sleep Quality
 ST 0.002 [0.00, 0.004] .001 .121 .09 0.001 [−0.002, 0.004] .001 .355 .05 0.003 [−0.001, 0.01] .002 .152 .08
 PU 0.002 [−0.01, 0.01] .01 .734 .02 0.001 [−0.01, 0.01] .01 .918 .01 0.01 [−0.01, 0.03] .01 .598 .03
 NOT 0.002 [−0.003, 0.01] .003 .432 .04 0.004 [−0.003, 0.01] .003 .298 .06 −0.001 [−0.01, 0.01] .01 .936 −.01
Open in a new tab

Note. ST = Screen Time. PU = Pickups. NOT = Notifications. Models were run separately by smartphone use variable. Unadjusted model results shown. Bolded associations survived Benjamini-Hochberg correction, with corrections applied by grouping p−values by within- and between-person associations. Results remained the same across sensitivity analyses unless otherwise noted with superscripts: when recoding outliers +/- 3 SD from the mean to missing for objective sleep onset (9 cases) and sleep quality (3 cases), when limiting objective analyses to participants with 3 or more days of FitBit data (n = 61), and when including covariates (school attendance (within-person) and age, gender, income, race/ethnicity and proportion of (completed) study days in which school was attended (between-person)).

a

Association was significant with outliers removed (p < .001).

Table 3.

Between-Person Associations between Smartphone Use and Sleep Outcomes

Total Day Night
b [95% CI] SE p β b [95% CI] SE p β b [95% CI] SE p β
Self-Report Sleep Onset
 ST −0.001 [−0.002, 0.001] .001 .546 −.08 −0.004 [−0.01, -0.001] .002 .007 −.33 0.002 [−0.001, 0.01] .001 .192 .16
 PU −0.01 [−0.01, 0.00] .003 .044 −.25 −0.01 [−0.02, -0.003] .003 .002 −.37 0.01 [−0.004, 0.03] .01 .157 .18
 NOT −0.001 [−0.004, 0.002] .002 .517 −.09 −0.003 [−0.01, 0.001] .002 .180 −.18 0.004 [−0.01, 0.01] .01 .409 .11
Self-Report Sleep Quality
 ST 0.00 [0.00, 0.001] .00 .697 .05 0.00 [−0.001, 0.001] .00 .506 .09 0.00 [−0.001, 0.001] .00 .971 .01
 PU 0.001 [−0.001, 0.002] .001 .331 .13 0.001 [−0.001, 0.003] .001 .223 .16 0.00 [−0.01, 0.004] .002 .929 −.01
 NOT 0.00 [−0.001, 0.001] .00 .811 −.03 0.00 [−0.001, 0.001] .001 .808 −.04 0.00 [−0.003, 0.002] .001 .851 −.03
Fitbit Sleep Onset
 ST 0.08 [−0.07, 0.22] .07 .296 .14 −0.10 [−0.38, 0.17] .14 .465 −.10 0.28 [0.06, 0.50]b .11 .013 .32
 PU −0.45 [−0.91, 0.01] .24 .057 −.25 −0.80 [−1.35, −0.25]a .28 .004 −.37 0.61 [−0.81, 2.04] .73 .398 .11
 NOT −0.16 [−0.44, 0.11] .14 .243 −.16 −0.33 [−0.72, 0.07] .20 .111 −.21 −0.08 [−0.82, 0.67] .38 .841 −.03
Fibit Sleep Quality
 ST −0.001 [−0.004, 0.002] .001 .673 −.06 −0.003 [−0.01, 0.003] .003 .344 −.13 0.00 [−0.004, 0.01] .002 .915 .01
 PU −0.01 [−0.01, 0.01] .01 .322 −.13 −0.01 [−0.02, 0.004] .01 .226 −.16 −0.001 [−0.03, 0.03] .01 .950 −.01
 NOT 0.00 [−0.01, 0.01] .003 .984 .00 0.00 [−0.01, 0.01] .004 .980 .00 0.001 [−0.01, 0.02] .01 .919 .01
Open in a new tab

Note. ST = Screen Time. PU = Pickups. NOT = Notifications. Models were run separately by smartphone use variable. Unadjusted model results shown. No associations survived Benjamini-Hochberg correction, with corrections applied by grouping p−values by within- and between-person associations. Results remained the same across sensitivity analyses unless otherwise noted with superscripts: when recoding outliers +/− 3 SD from the mean to missing for objective sleep onset (9 cases) and sleep quality (3 cases), when limiting objective analyses to participants with 3 or more days of FitBit data (n = 61), and when including covariates (school attendance (within-person) and age, gender, income, race/ethnicity and proportion of (completed) study days in which school was attended (between-person)).

a

Association was significant with outliers removed (p = .001) and when limiting analysis to those with three or more days of data (p = .001)

b

Association was significant with outliers removed (p = .003) and when limiting analysis to those with three or more days of data (p = .002)

Results

Correlations and descriptive statistics are in Table 1. Of note, daytime and nighttime use were only moderately correlated with each other for screen time and pickups. According to their Fitbits, participants’ average sleep onset was 1:29am. Main results are in Tables 2 and 3. When examining smartphone use across a full day, only one robust association emerged. When adolescents engaged in more screen time relative to their own average, they also self-reported later sleep onset. No within-person associations emerged for pickups and notifications. Between-person differences were not significant or not robust.

When distinguishing smartphone use between daytime and nighttime, important differences emerged. No within-person associations occurred for daytime use, and between-person differences were not significant or not robust. In contrast, numerous within-person associations were observed for nighttime use. Adolescents who had greater nighttime screen time, pickups, or notifications3 than their own average self-reported later sleep onset. Adolescents with greater nighttime screen time relative to their own average also self-reported poorer sleep quality and had later Fitbit-recorded sleep onset. Adolescents with greater nighttime pickups relative to their own average had later Fitbit-recorded sleep onset. Between-person associations for nighttime use were not significant or not robust.

Discussion

Concerns about digital technology use disrupting adolescent sleep permeate both scholarly and public discourse; however, many past studies rely on cross-sectional designs and self-report assessments (Alonzo et al., 2021). This study adds to the growing body of literature employing intensive repeated assessments to thoroughly explore linkages, and advances research by utilizing objective measurements of both smartphone use and sleep. Results suggest that when adolescents engage in greater smartphone use during the night (but not the day), they also have poorer sleep outcomes that night.

We considered three types of smartphone use (screen time, pickups, notifications), and when smartphone use occurred (across 24 hours, daytime use, nighttime use). When examining total use aggregated across the day, only one robust within-person association emerged for screen time: when adolescents engaged in more minutes of screen time relative to their own average, they self-reported later sleep onset, with a moderate effect size. The lack of consistent, robust linkages between the three types of smartphone use and sleep outcomes reflects other EMA studies that similarly found inconsistent evidence of within-person associations (Burnell et al., 2022; Hamilton et al., 2022; Tavernier et al., 2017).

Smartphone use that occurs midday may be less theoretically applicable to sleep outcomes that night, as this use is not clearly delaying sleep onset or interrupting sleep. Indeed, several significant associations emerged for nighttime use, whereas no within-person associations emerged for daytime use. Nighttime associations were most robust for screen time, in that more recorded screen time than one’s own average was associated with later self-reported and Fitbit-recorded sleep onset, and poorer self-reported sleep quality. Notably, although associations with sleep onset were observed for both self- and objective-reports, the association with sleep quality was only observed for the self-report measure. Self-reported and Fitbit-recorded sleep quality were not significantly correlated, in line with past research finding inconsistencies on how these objective and subjective reports relate to each other (Cudney et al., 2022). Subjective perceptions of sleep quality may be more critical for understanding linkages with screen time, in line with past research finding more consistent evidence of digital technology use linked with self-reports of sleep (Burnell et al., 2022). Additionally, adolescents who engaged in more pickups than their own average on a given day self-reported later sleep onset that night and had later Fitbit-recorded sleep onset. Effects sizes for screen time were moderate, whereas effect sizes for pickups were small. Receipt of notifications was not robustly associated with sleep outcomes.

These results showcase the necessity of distinguishing the time of day that use occurs. Past research (mainly cross-sectional) has supported linkages between nighttime digital technology use and poorer sleep outcomes (Kubiszewski et al., 2014; Lüscher & Radtke, 2022); however, studies employing more rigorous designs have not found robust support (Das-Friebel et al., 2020), and other studies found comparable associations for daytime and nighttime use (Hysing et al., 2014; Lee et al., 2021). The more robust associations for screen time could be because this metric may be more relevant for proposed mechanisms between greater digital technology use and poorer sleep. Longer screen time duration may result in delayed sleep onset (via displacement or light exposure) or increase the chance of encountering stimulating content that may challenge an adolescent’s ability to fall asleep or stay asleep (e.g., exposure to interpersonal conflict). Pickups and notifications may not map onto these mechanisms as cleanly. For example, pickups may be quick, fleeting, and habitual, and thus have a smaller impact. Notifications may have a reduced effect if the notifications are not audible or not of interest. These null results are notable as they are counter to previous research (e.g., Fobian et al., 2016; Johansson et al., 2016; Murdock et al., 2017) and despite some adolescents reporting being woken up by notifications (Robb, 2019). Subjective reports on the audibility of notifications is likely key in elucidating these associations.

In contrast to past studies (e.g., Burnell et al., 2022; Hamilton et al., 2022), this study found more consistent evidence for within-person associations compared to between-person associations. Indeed, no between-person results were significant after statistical correction. This may be a result of a smaller sample size that reduced power to detect between-person effects relative to the other studies (N = 388 in Burnell et al., 2022; N = 93 in Hamilton et al., 2022). Although resource intensive, future research should attempt to recruit larger samples in which objective digital technology use and sleep metrics can be gathered. These findings highlight the importance of employing objective assessments of digital technology use, which, to date, few studies have done in the sleep context. Inaccuracies in self-reports of digital technology use may be especially problematic when studying sleep outcomes. For example, studies employing evening reports of the day’s digital technology use may miss out on use that occurs after the assessment. Studies using morning reports of the previous day’s digital technology use may be hurt by accurate recall when use could have occurred during “groggy” late night periods. Objective measures of digital technology use can capture the key late-night and overnight window that may be especially vulnerable to recall problems.

Limitations and Future Directions

Study conclusions must be made in the context of several limitations. First, the overall sample size was relatively small with reduced power to detect between-person associations. Second, the objective smartphone metrics were quite broad and did not capture specific content. Additionally, the exact algorithms that are used to log smartphone metrics are proprietary to Apple and not publicly disclosed. Ideally, future research using objective measurement will partner with technology companies that allows for the transparent disclosure regarding the capture of such objective metrics. Moreover, analyses were hurt by missingness on the smartphone variables as cases could only be included in the presence of two calendar days of data. Future studies could involve an end-of-study laboratory-based session in which participants upload screenshots of past smartphone use, and researchers can verify completion of screenshots. Third, although Fitbit devices can reliably distinguish sleep from wake states (Haghayegh et al., 2019), they still fall short from gold standard polysomnography methods, and validation is limited by firmware updates (Goldsack et al., 2020). The Fitbits also did not collect or reliably collect data on other relevant sleep domains, such as sleep latency. Fourth, associations with notifications may have been underestimated as we did not collect data on if notifications were audible. Fifth, our data can only attest to general links between smartphone use and sleep outcomes. For example, it is unclear if and how specific types of smartphone content may relate to sleep outcomes. Future research directly testing the proposed mechanisms explaining the relations between digital technology use and sleep (i.e., displacement, blue light emission, cognitively arousing content) is needed.

Sixth, instructing participants to upload screenshots of the iOS Screen Time app may have drawn attention to smartphone habits, potentially altering phone engagement. There was no evidence of reactivity for included smartphone metrics (ps > .298), in that how participants engaged with their smartphones (in terms of minutes spent and number of pickups and notifications) did not change over the course of the EMA period (for example, there was no difference in smartphone use on Day 1 compared to Day 14). Nonetheless, it is possible that prompting participants to examine their smartphone use each day could have altered perceptions in how an adolescent is using their phone, such as by drawing attention to high usage of certain apps or which apps are most immediately used after pickup. Although this possible attention increase did not correspond to alterations in general screen time, pickups, and notifications, it could have altered patterns of app usage (e.g., using one app more and another app less). Seventh, to reduce participant burden and to avoid difficulties with third party app data collection on iOS devices (Cornet & Holden, 2018), a default app (iOS Screen Time) was used to collect smartphone data rather than instructing participants to download and utilize a new app, which could reduce compliance. This resulted in the exclusion of 16 Android users and eight iOS users who never uploaded smartphone data. These 24 participants did not differ from the analytic sample on key demographics, including race/ethnicity and income (ps>.094), but Android users were disproportionately boys (69%; p=.012). Although the gender distribution of the sample was only slightly skewed (56% girls, 42% boys), omission of Android users reduces generalizability. Finally, we did not obtain in-depth assessments of sleep health, such as what can be obtained through larger gold standard measures such as the Pittsburgh Sleep Quality Index.

Future research should develop methodological strategies to address these limitations. Perhaps most importantly, observational and machine learning methods should capture what types of content adolescents are interacting with, particularly before bed, and how types of engagement may relate to sleep. Interactive screen activities may be more consequential for sleep outcomes (Reichenberger et al., 2023), indicating that content that triggers cognitive arousal may be particularly key in understanding associations with sleep. Indeed, experimental research suggests that the benefits of filtering out blue light exposure are washed out when participants viewed Facebook content assumed to be cognitively arousing (Bowler & Bourke, 2019). Additionally, future studies can thoroughly test possible individual differences in how smartphone use may relate to sleep. For example, in line with the need to examine how specific types of digital content may relate to sleep outcomes, studies can explore how individual differences relevant to the consumption of this digital content may moderate associations. For example, adolescents’ emotional investment in social media content (e.g., feedback received) or the frequency in which they engage in maladaptive digital behaviors (e.g., doomscrolling) may be especially relevant for sleep disturbances.

Implications and Conclusions

Findings provide evidence that when an adolescent engages in greater nighttime screen time use and, to some extent, pickups relative to their own average, they also have poorer self-reported and Fitbit-recorded sleep outcomes. Guidance for adolescents, parents, and clinicians is needed to avoid further exacerbating these existing problems with the increasing use of digital technologies. Interventions encouraging individuals to structure digital technology use in a regimented manner (e.g., designating “phone free” times during the day) have had some success in improving sleep outcomes (Barber & Cucalon, 2017). Restricting use prior to bedtime may be particularly important (Hale & Guan, 2015), especially given the current study’s findings and in light of experimental evidence suggesting the benefits of restriction for sleep outcomes among college students (He et al., 2020). Utilizing built-in smartphone features including changing the interface to grayscale (Holte et al., 2023; Zimmerman & Sobolev, 2023) and batching notifications (Fitz et al., 2019) can be effective in reducing smartphone use and improving one’s sense of control over phone use, although rigorous testing in how these strategies affect sleep is still needed. Importantly, tactics to engage in healthier smartphone use may rely in tapping into adolescents’ existing drives for autonomy and independence, rather than implementing sweeping and homogenized restrictions (Galla et al., 2021).

To that end, although strategies such as parental monitoring may be beneficial in restrictions earlier in adolescence, adolescents’ growing bids for autonomy may conflict with parent rules and regulations and result in covert use. Indeed, simply having rules has been shown to have mixed associations with improved sleep outcomes (Giovanelli et al., 2022). Involving adolescents themselves in discussions about bedtime digital technology use restriction may be fruitful in helping them understand healthy digital technology use and when use should be limited. In turn, these discussions may help adolescents develop the skills needed to maintain healthy use patterns, especially when parental monitoring may be impossible or developmentally inappropriate (e.g., entry into college). Overall, adolescents should be mindful of their nighttime smartphone use, and how this use may be disrupting sleep outcomes.

Ethics and Integrity Policy Statements:

Raw data are currently unavailable due the sensitive nature of the study. This research was supported by the National Institutes of Health grant R01DA039923 and the Winston Family Foundation. The authors have no conflicts of interest to disclose. All study procedures were approved by the University of North Carolina at Chapel Hill Institutional Review Board.

Supplementary Material

Supinfo

Footnotes

1

Missingness was higher for sleep onset as sleep onset was computed via sleep stage data. Sleep stage data was not always present as it requires longer periods of uninterrupted sleep to compute.

2

This cutoff was data-driven, as the 8:00pm hour was the time when participants started to go to bed (with the exception of <1% of cases) and because the earliest anchor for self-reported sleep onset was 8:00pm and earlier.

3

The association with notifications may in part be explained by its high correlation with pickups; indeed, when controlling for pickups, the association was no longer significant (p = .160).

References

  1. Alonzo R, Hussain J, Stranges S, & Anderson KK (2021). Interplay between social media use, sleep quality, and mental health in youth: A systematic review. Sleep Medicine Reviews, 101414. 10.1016/j.smrv.2020.101414 [DOI] [PubMed] [Google Scholar]
  2. Arnett JJ (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55, 469–480. 10.1037//0003-066X.55.5.469 [DOI] [PubMed] [Google Scholar]
  3. Barber LK, & Cucalon MS (2017). Modifying the sleep treatment education program for students to include technology use (STEPS-TECH): Intervention effects on objective and subjective sleep outcomes. Stress and Health, 33, 684–690. 10.1002/smi.2746 [DOI] [PubMed] [Google Scholar]
  4. Brautsch LAS, Lund L, Andersen MM, Jennum PJ, Folker AP, & Andersen S (2023). Digital media use and sleep in late adolescence and young adulthood: A systematic review. Sleep Medicine Reviews, 68, 101742. 10.1016/j.smrv.2022.101742 [DOI] [PubMed] [Google Scholar]
  5. Bowler J, & Bourke P (2019). Facebook use and sleep quality: Light interacts with socially induced alertness. British Journal of Psychology, 110, 519–529. 10.1111/bjop.12351 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Benjamini Y, & Hochberg Y (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B: Methodological, 57, 289–300. [Google Scholar]
  7. Burnell K, George MJ, Jensen M, Hoyle RH, & Odgers CL (2022). Associations between adolescents’ daily digital technology use and sleep. Journal of Adolescent Health, 70, 450–456. 10.1016/j.jadohealth.2021.09.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cabré-Riera A, Torrent M, Donaire-Gonzalez D, Vrijheid M, Cardis E, & Guxens M (2019). Telecommunication devices use, screen time and sleep in adolescents. Environmental Research, 171, 341–347. 10.1016/j.envres.2018.10.036 [DOI] [PubMed] [Google Scholar]
  9. Charmaraman L, Richer AM, Ben-Joseph EP, & Klerman EB (2021). Quantity, content, and context matter: Associations among social technology use and sleep habits in early adolescents. Journal of Adolescent Health, 69, 162–165. 10.1016/j.jadohealth.2020.09.035 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cornet VP, & Holden RJ (2018). Systematic review of smartphone-based passive sensing for health and wellbeing. Journal of Biomedical Informatics, 77, 120–132. 10.1016/j.jbi.2017.12.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cudney LE, Frey BN, McCabe RE, & Green SM (2022). Investigating the relationship between objective measures of sleep and self-report sleep quality in healthy adults: a review. Journal of Clinical Sleep Medicine, 18, 927–936. 10.5664/jcsm.9708 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Curran PJ, & Bauer DJ (2011). The disaggregation of within-person and between-person effects in longitudinal models. Annual Review of Psychology, 62, 583–619. 10.1146/annurev.psych.093008.100356 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Das-Friebel A, Lenneis A, Realo A, Sanborn A, Tang NKY, Wolke D, von Mühlenen A, & Lemola S (2020). Bedtime social media use, sleep, and affective wellbeing in young adults: an experience sampling study. Journal of Child Psychology and Psychiatry, 61, 1138–1149. 10.1111/jcpp.13326 [DOI] [PubMed] [Google Scholar]
  14. Fisher AJ, Medaglia JD, Jeronimus BF (2018). Lack of group-to-individual eneralizability is a threat to human subjects research. PNAS, 115, E5105–6115. 10.1073/pnas.1711978115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fitz N, Kushlev K, Jagannathan R, Lewis T, Paliwal D, & Ariely D (2019). Batching smartphone notifications can improve well-being. Computers in Human Behavior, 101, 84–94. 10.1016/j.chb.2019.07.016 [DOI] [Google Scholar]
  16. Fobian AD, Avis K, & Schwebel DC (2016). The impact of media use on adolescent sleep efficiency. Journal of Developmental & Behavioral Pediatrics, 37, 9–14. 10.1097/DBP.0000000000000239 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Foerster M, Henneke A, Chetty-Mhlanga S, & Röösli M (2019). Impact of adolescents’ screentime and nocturnal mobile phone-related awakenings on sleep and general health symptoms: A prospective cohort study. International Journal of Environmental Research and Public Health, 16, 518. 10.3390/ijerph16030518 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Galla BM, Choukas-Bradley S, Fiore HM, & Esposito MV (2021). Values-alignment messaging boosts adolescents’ motivation to control social media use. Child Development, 92, 1717–1734. 10.1111/cdev.13553 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. George MJ, Rivenbark JG, Russell MA, Ng’eno L, Hoyle RH, & Odgers CL (2019). Evaluating the use of commercially available wearable wristbands to capture adolescents’ daily sleep duration. Journal of Research on Adolescence, 29, 613–626. 10.1111/jora.12467 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Giovanelli A, Ozer EJ, Adams SH, Park MJ, & Ozer EM (2022). Adolescent technology-use rules and sleep in a large representative sample. Journal of Adolescent Health, 70, 682–685. 10.1016/j.jadohealth.2021.10.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Goldsack JC, Coravos A, Bakker JP, Bent B, Dowling AV, Fitzer-Attas C, Godfrey A, Godino JG, Gujar N Izmailova E, Manta C, Peterson B, Vandendriessche B, Wood WA, Wang KW, & Dunn J (2020). Verification, analytical validation, and clinical validation (V3): the foundation of determining fit-for-purpose for Biometric Monitoring Technologies (BioMeTs). npj Digital Medicine, 3, 55. 10.1038/s41746-020-0260-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Haghayegh S, Khoshnevis S, Smolensky MH, Diller KR, Castriotta RJ (2019). Accuracy of wristband Fitbit models in assessing sleep: Systematic review and meta-analysis. Journal of Medical Internet Research, 21, e16273. 10.2196/16273 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hale L, & Guan S (2015). Screen time and sleep among school-aged children and adolescents: A systematic literature review. Sleep Medicine Reviews, 21, 50–58. 10.1016/j.smrv.2014.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hamilton JL, Chang S, Reinhardt L, Ladouceur CD, Silk JS, Moreno M, Franzen PL, & Bylsma LM (2020). Social media use predicts later sleep timing and greater sleep variability: An ecological momentary assessment study of youth at high and low familial risk for depression. Journal of Adolescence, 83, 122–130. 10.1016/j.adolescence.2020.07.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hamilton JL, Hutchinson E, Evankovich MR, Ladouceur CD, & Silk JS (2022). Daily and average associations of physical activity, social media use, and sleep among adolescent girls during the COVID-19 pandemic. Journal of Sleep Research. e13611. 10.1111/jsr.13611 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. He J-W, Tu Z-H, Ziao L, Su T, & Tang Y-X (2020). Effect of restricting bedtime mobile phone use on sleep, arousal, mood, and working memory: A randomized pilot trial. PLoS ONE, 15, e0228756. 10.1371/journal.pone.0228756 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Holte AJ, Giesen DT, & Ferraro FR (2023). Color me calm: Grayscale phone setting reduces anxiety and problematic smartphone use. Current Psychology, 42, 6778–6790. 10.1007/s12144-021-02020-y [DOI] [Google Scholar]
  28. Hysing M, Pallesen S, Stormark KM, Jakobsen R, Lundervold AJ, Sivertsen B (2015). Sleep and use of electronic devices in adolescence: results from a large population-based study. BMJ Open, e006748. 10.1136/bmjopen-2014-006748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. IBM Corp. Released 2015. IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp. [Google Scholar]
  30. Johansson AEE, Petrisko MA, & Chasens ER (2016). Adolescent sleep and the impact of technology use before sleep on daytime function. Journal of Pediatric Nursing, 31, 498–504. 10.1016/j.pedn.2016.04.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Keyes KM, Maslowsky J, Hamilton A, & Schulenberg J (2014). The great sleep recession: Changes in sleep duration among US adolescents, 1991–2012. Pediatrics, 135, 460–468. doi/ 10.1542/peds.2014-2707 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kubiszewski V, Fontaine R, Rusch E, & Hazouard E (2014). Association between electronic media use and sleep habits: an eight-day follow-up study. International Journal of Adolescence and Youth, 19, 395–407. 10.1080/02673843.2012.751039 [DOI] [Google Scholar]
  33. LeBourgeois MK, Hale L, Chang A-M, Akacem L, Montgomery-Downs HE, Buxton OM (2017). Digital media and sleep in childhood and adolescence. Pediatrics, 140, e20161758. 10.1542/peds.2016-1758J [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lee XK, Chee NIYN, Ong JL, Teo TB, Van Rijn E, Lo JC, & Chee MWL (2019). Validation of a consumer sleep wearable device with actigraphy and polysomnography in adolescents across sleep opportunity manipulations. Journal of Clinical Sleep Medicine, 15, 1337–1346. 10.5664/jcsm.7932 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lee PH, Tse ACY, Wu CST, Mak YW, & Lee U (2021). Temporal association between objectively measured smartphone usage, sleep quality and physical activity among Chinese adolescents and young adults. Journal of Sleep Research, 30, e13213. 10.1111/jsr.13213 [DOI] [PubMed] [Google Scholar]
  36. Lüscher J, & Radtke T (2022). The impact of adolescent’s daily electronic media use on sleep: Insights from adolescent’s and their mother’s point of view. Applied Psychology: Health and Well-being. Advance Online Publication. 10.1111/aphw.12397 [DOI] [PubMed] [Google Scholar]
  37. Mac Cárthaigh S, Perry J, & Griffin C (2022). Do objective data support the claim that problematic smartphone use has a clinically meaningful impact upon adolescent sleep duration? Behaviour & Information Technology. Advance Online Publication. 10.1080/0144929X.2022.2136533 [DOI] [Google Scholar]
  38. Molenaar PCM, & Campbell CG (2009). The new person-specific paradigm in psychology. Current Directions in Psychological Science, 18, 112–117. [Google Scholar]
  39. Murdock KK, Horissian M, & Crichlow-Ball C (2017). Emerging adults’ text message use and sleep characteristics: A multimethod, naturalistic study. Behavioral Sleep Medicine, 15, 228–241. 10.1080/15402002.2015.1120203 [DOI] [PubMed] [Google Scholar]
  40. Muthén LK, & Muthén BO (1998–2017). Mplus user’s guide. Eighth Edition. Los Angeles, A: Muthén & Muthén. [Google Scholar]
  41. Parry DA, Davidson BI, Sewall CJR, Fisher JT, Mieczkowski H, & Quintana DS (2021). A systematic review and meta-analysis of discrepancies between logged and self-reported digital media use. Nature Human Behavior, 5, 1535–1547. 10.1038/s41562-021-01117-5 [DOI] [PubMed] [Google Scholar]
  42. Reichenberger DA, Master L, Mathew GM, Snyder CK, Buxton OM, Hale L, & Chang A-M. (2023). Interactive screen-based activities predict worse actigraphic sleep health that night among adolescents. Journal of Adolescent Health. Advance Online Publication. 10.1016/j.jadohealth.2023.10.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Robb MB (2019). The new normal: Parents, teens, screens, and sleep in the United States. San Francisco, CA: Common Sense Media. [Google Scholar]
  44. Short MA, Booth SA, Omar O, Ostlundh L, & Arora T (2020). The relationship between sleep duration and mood in adolescents: A systematic review and meta-analysis. Sleep Medicine Reviews, 52, 101311. 10.1016/j.smrv.2020.101311 [DOI] [PubMed] [Google Scholar]
  45. Tavernier R, Heissel JA, Sladek MR, Grant KE, & Adam EK (2017). Adolescents’ technology and face-to-face time use predict objective sleep outcomes. Sleep Health, 3, 276–283. 10.1016/j.sleh.2017.04.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Twenge JM, Krizan Z, & Hisler G (2017). Decreases in self-reported sleep duration among U.S. adolescents 2009–2015 and association with new media screen time. Sleep Medicine, 39, 47–53. 10.1016/j.sleep.2017.08.013 [DOI] [PubMed] [Google Scholar]
  47. Valkenburg PM, & Piotrowski JT (2017). Plugged in: How media attract and affect youth. Yale University Press. [Google Scholar]
  48. Vogels EA, Gelles-Watnick R, & Massarat N (2022, August 10). Teens, Social Media and Technology 2022. Pew Research Center. Retrieved from https://www.pewresearch.org/internet/2022/08/10/teens-social-media-and-technology-2022/ [Google Scholar]
  49. Zimmerman L, & Sobolev M (2023). Digital strategies for screen time reduction: A randomized field experiment. Cyberpsychology, Behavior, and Social Networking, 26, 42–49. 10.1089/cyber.2022.0027 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supinfo
NIHMS1983767-supplement-Supinfo.docx (16.5KB, docx)

RESOURCES