key: cord-0721571-4ez4wmwb authors: Eales, Lauren; Gillespie, Sarah; Alstat, Reece A.; Ferguson, Gail M.; Carlson, Stephanie M. title: Children’s screen and problematic media use in the United States before and during the COVID‐19 pandemic date: 2021-09-06 journal: Child Dev DOI: 10.1111/cdev.13652 sha: 6aff1d9ea792a92ad1a02e40d20dae87d7c7263e doc_id: 721571 cord_uid: 4ez4wmwb This mixed methods study examined parent‐reported child screen media use before and during the COVID‐19 pandemic by examining 2019–2020 changes in parent perceptions of media, screen media use (SMU), and problematic media use (PMU) in children aged 2–13 years (N = 129; 64 boys, 64 girls, 1 nonbinary; 90.7% White, 4.6% Hispanic/Latino, 0.8% Black, 8.5% multiethnic; primarily middle‐to‐high income). Quantitative analyses showed a significant SMU and PMU increase (medium effect size). There was a steeper increase in PMU among school‐age (older) children. Together, the qualitative and quantitative results suggest that the PMU and SMU increase were influenced by distal, proximal, and maintaining factors including the COVID‐19 pandemic, distance learning, child behaviors, other children, parental mediation, and positive media reinforcement. The COVID-19 pandemic has caused marked changes across all layers of children's social ecologies, including family routines, schooling, media habits, and the broader economy. Cross-national studies in the early months of the pandemic identified the patterns of selfreported lifestyle changes among adolescents and adults, including more time spent physically inactive and using screens (Dutta et al., 2020; Pišot et al., 2020) . The most striking change in children's daily lives and routines was the rapid closure of schools to reduce transmission of the virus. In the spring of 2020, nearly 93% of respondents in U.S. households with children reported that their children were engaging in some form of "distance learning" via online resources (80%) or paper materials (20%) sent home by the school (U.S. Census Bureau, 2020). Additionally, roughly 40% of Americans are working from home full time during the pandemic (Smith, 2020) , creating new demands on parents who are simultaneously working and monitoring their children during the day (Eales et al., in press) . Such cross-cutting changes in daily life have reinvigorated debates about screen media usage among children and adolescents, both within the academic literature (Nagata et al., 2020) and within the popular press (Kamenetz, 2020; Wartella, 2020) . Prior to the pandemic, 66% of all U.S. parents reported that they believe parenting is harder today than it was 20 years ago; most citing technology as the main reason for this change (Auxier et al., 2020) . Additionally, 71% of parents of children under 12 years old reported that they were somewhat or very concerned that their child spends too much time in front of screens, even though the majority also reported confidence in knowing appropriate limits (Auxier et al., 2020) . The disruptions to daily life and family routines necessitate efforts to understand how to minimize the risks and maximize the potential benefits of screen media use for children and families during this pandemic. Child screen media use has changed since the COVID-19 pandemic began, with research emerging on its correlates and new patterns of use. The World Health Organization (WHO) and American Academy of Pediatrics (AAP) generally recommended a maximum of one hour of screen time for children under 5 and consistent, family-specific limits for older children (AAP, 2016; WHO, 2019) . However, parents have felt many tensions of parenting in the age of screens since before the pandemic began, with qualitative work revealing cognitive and emotional tensions in parents of 0-to 8-year-old children . In light of the pandemic, parents and researchers have recognized the infeasibility for most families to meet the screen media guidelines offered by these entities, while others have highlighted the potential benefits that screen-based socialization and educational programming may have for children and adolescents during this time (Nagata et al., 2020) . Parents are seeking new guidance on children's screen time use in light of distance learning and changes to family routines and work situations (Kamenetz, 2020) . Preliminary, cross-sectional data from families around the world have documented concerning trends in media use and other relevant health behaviors. Online schooling has been associated with increased usage of electronic devices without parental supervision during class (Lau & Lee, 2021) . Among adolescents in India, an increase in screen exposure was associated with disruptions in sleep behavior and more sedentary time (Dutta et al., 2020) . In Portugal, parents retrospectively reported not only an increase in screen time and family activities but also a decrease in physical activity (Pombo et al., 2020) . Parents in Turkey reported that they instituted ground rules related to screen time by May 2020 and had observed an increase in screen time in their children (Eyimaya & Irmark, 2021) . The current study aims to extend the extant cross-sectional research on screen media use during the pandemic to examine screen media use and problematic media use at two time points (hereafter referred to as "pre-pandemic" [February-April, 2019] and "postonset" [May-June, 2020] ) in a group of U.S. families. Research on screen media use during COVID-19 stems from known risks and buffers of screen media use for child development. Weight gain, sleep disruption, inattention problems, and developmental delays are all associated with increased screen media use early in life (AAP, 2016; Tamana et al., 2019) . During the COVID-19 pandemic, children's social, intellectual, and self-regulatory development are potentially at risk due to home confinement (Goldschmidt, 2020) . Of course, while there is not a perfect association between media use and risks, specific types of media use are associated with these risks. For example, using screen media at night is associated with shorter sleep times for children because of the light emitted by the electronic devices close to bedtime (LeBourgeois et al., 2017) . Additionally, poor executive functioning skills are associated with non-PBS television content and a young age of beginning screen media in preschoolers (Nathanson et al., 2014) , though other researchers have found no association between screen time and executive functioning skills in preschoolers (Jusienė et al., 2020) . Other studies have found that poorer selfregulation abilities are cross-sectionally linked to more media use in children, and parents using media to regulate their child's distress are longitudinally linked to increased negative emotionality in some children (Linder et al., 2020; Gordon-Hacker & Gueron-Sela, 2020) . During the COVID-19 pandemic, the bidirectional associations between poorer child self-regulation and screen media use may have become more pronounced. While schools were closed, parents may have increasingly felt the need to occupy children, particularly those with selfregulation difficulties, with media use (Radesky, 2020) . While there are certainly risks of screen media use in childhood, screen media use can also serve as a buffer for adjustment, particularly during the COVID-19 pandemic. Screen media use can foster social development in online spaces and connections with distant family members, which is of particular relevance when children may not be able to see older family members like grandparents during the pandemic (Chassiakos et al., 2016; Eales et al., in press; Grose, 2021) . In a sample of Chinese children and adolescents, media use-above and beyond reading and physical activity-helped alleviate pandemic-related distress (Jiao et al., 2020) . Parents can also promote adaptive screen media behavior and regulate child media use during the pandemic by watching screen media with their children and explaining or guiding them through what they are seeing (Coyne et al., 2017; Király et al., 2020; Vanderloo et al., 2020) . A month after COVID-19 was declared a pandemic, a UNICEF article encouraged parents to "rethink" their assumptions on screen time, focusing on what it can do for their children instead of how it can harm them. The authors recommended parents have their children stay in touch with friends, engage with their children through video games and online experiences, and encourage their children to stay physically active in front of screens (Winther & Byrne, 2020) . Taken together, these findings suggest that although there are risks of screen media use for children, parents can find ways to use media to their child's advantage during the pandemic. from benign media use (Domoff et al., 2019 (Domoff et al., , 2020 . As mentioned previously, there is not a simple cause and effect relation between increased screen media use and unwanted outcomes in children. Rather, these negative effects and associations stem from various factors: how the media are used, what the media are, the characteristics of the child using the media, and so on. Domoff et al., (2020) recently elucidated an Interactional Theory of Childhood Problematic Media Use (IT-CPU), an extension of Bronfenbrenner's bioecological model that emphasizes the proximal, distal, and maintaining factors that can lead to the emergence of problematic media use. Distal factors, such as household chaos and digital environmental design, can exert their influence on proximal factors like a child's characteristics, parent media use and beliefs, and peer technology access. Maintaining factors keep problematic media use in play for the child through factors like positive reinforcement of media use for the child, using media to cope, and peer influences to play video games or engage online together. Domoff and colleagues' conceptualization of problematic media use also urges researchers to not consider simple screen time metrics as an indicator of problematic use, echoing calls from other researchers advocating to use more nuanced ways of assessing media use in children and adolescents (e.g., Kaye et al., 2020) . Given the changes the COVID-19 pandemic has exerted across many areas of family life, a rise in problematic media use could be a concern for families. When considering screen media use and problematic media use for children before and during the COVID-19 pandemic, what factors can moderate this trajectory? Given the well-established differences in media use across age groups in childhood (Rideout & Robb, 2020) , it stands to reason that older children in our sample are more likely to use more screen media and could have a larger increase in screen media use post-onset. Additionally, the proximal, distal, and maintaining factors of PMU in Domoff et al.'s (2020) IT-CPU are likely to change across the years (e.g., using more media to cope, more peer influence), potentially leading to higher problematic media use among older children than their younger peers. The ways in which parents perceive media use, engage with their child's media use, and use their own media use can also impact how a child uses screen media (e.g., Lauricella et al., 2015) . Policymakers and researchers advocate for parents to watch and engage in media with their child, which can teach them how to regulate their own media and potentially buffer a reliance on media use (e.g., Coyne et al., 2017; Kiraly et al., 2020) . Therefore, a parent's lower engagement with their child's media use pre-pandemic could lead to a steeper increase in screen media use and problematic media use post-onset, as the "proximal" and "maintaining" factors of the IT-CPU model could be present (i.e., the parent is not modeling or teaching their child how to use screen media adaptively). This engagement with media can also be seen as a form of parental mediation of media (active mediation [watching with discussion; also called instructive mediation]; Valkenburg et al., 1999) , which is linked with positive child outcomes and healthier screen media habits (e.g., Coyne et al., 2017; Griffiths et al., 2016; Mendoza, 2009) . Other forms of parental media mediation are coviewing, where a parent watches media with their child without discussion, and restrictive mediation, where a parent sets specific rules or prohibits certain media content for their child (Barkin et al., 2006; Valkenburg et al., 1999) . Relevant to the COVID-19 pandemic, increased parental stress pre-pandemic has also been linked to increased restrictive and active mediation, as well as coviewing (Warren & Aloia, 2019) . The ways in which a parent uses screen media as a regulation tool or a virtual "babysitter" could also certainly lead to steeper increases in benign or problematic screen media use post-onset, as these also contribute to the dyadic factors of maintaining problematic media use as per the IT-CPU. Giving a child a device for calming (regulator) or to keep them occupied while a parent has to work (babysitter) is theorized to strengthen the maintenance of problematic media use (Domoff et al., 2020) . Parents clearly play a large role in their child's screen media and problematic media use. From a young age, parents help decide what screen to turn on, when to turn it on, and when to turn it off (Nikken & Schols, 2015) . How parents mediate their child's media, parent's use of their own media, and their attitudes surrounding media are all related to a child's media use (Coyne et al., 2017) . Parents who use their own devices during parent-child interactions are less likely to respond to their child's bids for attention, and in response, children may engage in more attention-seeking behaviors (Kildare & Middlemiss, 2017) . A parent's attitudes about screen media can also influence how their family uses screen media: Families with media-focused parents are more likely to have children who use more media (Wartella et al., 2013) . Therefore, how parents perceive screen media use for their own children could moderate the change between pre-pandemic and post-onset for screen media use and problematic media use. The current study explores how children are using screen media differently pre-pandemic versus post-COVID-19 onset using data from U.S. parents collected at two time points (February-April 2019, May-July 2020). Our first aim was to describe how children were using screen media differently pre-pandemic versus post-onset and parent perceptions of their child's media use. This was a confirmatory aim based on international observations of changing screen media habits in children and adolescents (Dutta et al., 2020; Eyimaya & Irmark, 2021; Lau & Lee, 2021; Pombo et al., 2020) . We hypothesized that, in our qualitative data, parents would express more concern about their child's media use post-onset and there would be a shift in how they were talking about their child's media use. We also hypothesized that in our postonset quantitative data: (1) Children would use more screen media than pre-pandemic; (2) children would use screens in more problematic ways than pre-pandemic; and (3) parents would have more negative perceptions of media use. Our second aim (primarily exploratory) was to examine the parent and child factors that moderate the trajectories of change in screen media use and problematic media use pre-pandemic to post-onset. We hypothesized that being an older child, having a parent with more positive perceptions of media use, having a parent with lower participation with children during media activities, and greater parental use of screens as a regulation tool would lead to a steeper increase in non-school-related screen media use and problematic media use post-onset. We utilized a mixed method approach to capture strengths of qualitative and quantitative approaches by (1) qualitatively capturing differences in how parents perceived their child's media use pre-pandemic versus postonset and (2) quantitatively assessing change in screen media use and problematic media use over time as well as moderators of these changes (Creswell & Plano Clark, 2018) . This study collected the qualitative and quantitative data simultaneously, the analyses were conducted separately, and then the results were synthesized in the interpretation process (convergent design, questionnaire variant; Creswell & Plano Clark, 2018) . Participants were recruited via a university-managed participant pool of families in a major U.S. city. In February-April 2019 (T1 or pre-pandemic), parents of children aged 2-11 years were randomly selected from this pool to receive an email inviting them to complete an online survey. Two-hundred and forty-six of these participants indicated they would be willing to be contacted for a follow-up study. In May-July 2020 (T2 or post-onset), after the COVID-19 pandemic hit, these participants were contacted again via email with a link to a new survey. Altogether, 169 participants responded to the survey again, two of whom took the survey twice, and only 131 of those participants responded for the same child (deduced by date of birth and parent initials). One participant was removed because they lived in a different country and one was removed because they had over 35% missing data, resulting in a final sample size of 129. The original 246 participants, 169 re-responders, and final 129 sample did not differ in terms of parent education, family income, child age, or child gender. Only families meeting the age criteria were sent the e-mail link, and there were no exclusion criteria. Parents included 127 mothers and 2 fathers (M 2020 Age = 39.4, SD 2020 Age = 4.34, Range = 29-50) of children described in 2020 as 64 girls, 64 boys, and 1 nonbinary child (M 2020 Age = 6.14, SD 2020 Age = 2.21, Range = 2.33-12.75). One participant indicated a different gender for their child in 2020-given that gender is not an integral part of our analyses, they were kept in the sample. Descriptions of age ranges at both time points are provided in Table 1 . The average participant family income at T1 was $125,000-$149,999 (Min = <$25,000, Max = $200,000+), which is relevant to the types of devices available to children and stressors faced by the family during the pandemic (Cluver et al., 2020) . For 86.8% of the families, both parent and child were monoethnically White and non-Hispanic/Latino. Of the children, 90.7% were White; 4.6% were Hispanic/Latino; 0.8% were Black/African American; and 8.5% were multiethnic (combinations of Black/African American, White, and Asian ethnicities). At T1, there were between one and four children in the home (average = 2), and 95.3% of participating parents were married to the target child's other parent. At T1, 58.1% of the parents had at least some graduate school; 34.9% had a Bachelor's degree; 5.4% had some college; and 1.6% had "other." The sample demographic characteristics are largely due to convenience sampling and the demographics of the university-managed participant pool, which we note as a limitation of the study's generalizability. See Table 1 for full descriptive statistics of relevant study variables (and see Supporting Information B for correlations between all variables). Additionally, most children were enrolled in school at the time of data collection and their use of media for distance learning was assessed: in May or June 2020, six (4.7%) were not in virtual school; 30 (23.3%) participated in virtual schooling for less than an hour/day; 27 (20.9%) participated between 1 and 2 h/day; 23 (17.8%) participated for 3-4 h/day; five (3.9%) participated for 5-6 h per day; and one (0.8%) participated for 9+ h/day. Following Institutional Review Board approval, the researchers received a randomly selected participant list from the city-wide participant pool of parents of children aged 2-11 years old. This participant pool recruits participants from across Minnesota (United States). These participants received a survey link via email between February and April, 2019, where they were eligible to receive a $100 e-gift card through a raffle for their participation. At this time, participants also indicated if they would be willing to be recontacted for another survey in the future. After further IRB approval, these participants were recontacted via email in May-July, 2020 with an invitation to complete another 30-40 minute survey, shortly after the state Governor began to dial back stayat-home orders. Each participant could elect to receive a $10 e-gift card, informational resources, both, or neither for the 2020 data collection. Participants were identified via a unique participant-created ID (same ID at both time points). Qualitative: Parent perceptions of child media use Open-ended questions developed for this survey allowed parents to report on other elements of their child's media use or pandemic-related experiences that they found important or concerning. At T1, the prompt read "If you have any final thoughts regarding your child's media use, please write them out here. Is there anything we didn't ask that you think is important?" At T2, this prompt also invited them to comment on the impact of the COVID-19 pandemic and a second prompt asked: "Is there anything we didn't ask that you think is important such as other current events that are impacting you and/or your family right now?" Only responses pertaining to children's screen media use were included in analyses; "N/A" responses were excluded. Adapted versions of items from the Common Sense Census (CSC) of children's media use (Rideout, 2017) were used to measure children's total screen media use across common non-school-related activities at both T1 and T2. Parents reported the average time per day that the child spent using screen media for eight nonschool-related activities (e.g., watching shows or movies on a computer or laptop, watching shows or movies on a TV, playing games on a handheld game player, doing anything not school-related on a smartphone or tablet). Very small wording changes were made between T1 and T2 (see Supporting Information A). Responses were anchored to a 7-point Likert scale that included never, 15 min or less, 15-45 min, 45-60 min, 1-1.5 h, 1.5-2 h, 2 h or more. Quantities were given for the midpoint of each response: never (0), 15 min or less (7.5), 15-45 min (30), 45-60 min (52), 1-1.5 h (75), 1.5-2 h (105), 2 h or more (130). The original Rideout (2017) measure asks first about days spent on these activities, then hours/minutes. We adapted the questionnaire at T1 for length and made the time options multiple choice to ensure individuals did not erroneously type in the incorrect amount of time spent on an activity (e.g., typing in "20" when asked how many hours their child spent doing an activity, when they really meant 20 min). Any quantities that were over two standard deviations over the mean were restricted to two standard deviations over the mean (T1 n = 9; T2 n = 6). Quantitative: Problematic media use (PMU) The 9-item Problematic Media Use Measure -Short Form (Domoff et al., 2019) was used to capture screen media use that is disruptive to family functioning (e.g., my child's screen media use interferes with family activities) or obsessive in nature (e.g., screen media are all that my child seems to think about). The measure demonstrated convergent validity with total daily screen time and parent-rated concern about the child's media use and showed incremental validity for predicting the child's overall functioning above total daily screen time (Domoff et al., 2019) . Parents rated how true these statements were for their child on a 5-point Likert scale from "1 = Never" to "5 = Always." A mean score was computed at each time point, with higher scores indicating more problematic media use (possible range: 1-5). The scale showed acceptable internal consistency at both time points (Time 1 α = .91; Time 2 α = .94). Quantitative: Parental participation with child screen media Parents reported how often they participate in media with their child during four different activities (watching their TV shows, watching online videos, playing console video games, and using games or apps on a smartphone or tablet). All four items from the CSC scale were used (Rideout, 2017) . Responses were rated on a 4-point Likert scale from "1 = All of the time" to "4 = Never." A fifth option allowed them to respond that the item was not applicable if their child does not do an activity. Allowing for "not applicable," the measure had good reliability (α = .74). The mean of all four items was computed (allowing for N/As) to result in a score from 1 to 4, a higher score meaning they do activities with their child a small amount of the time. Quantitative: Parent perceptions of screen media Using items drawn from the CSC (Rideout, 2017) , parents reported their attitudes about the effect of children's media use across six different domains of their child's life: social skills, learning, ability to focus, behavior, physical activity, and creativity. For each domain, they responded to the question "Overall, do you think your child's media use helps, hurts, or makes no difference to his/her [domain]" using a 5-point Likert scale anchored to "1 = Helps a lot," "3 = Makes no difference," and "5 = Hurts a lot." The mean of each time point was calculated to yield two scores in year 1 and year 2, range = 1-5 (Time 1 α = .78; Time 2 α = .70). Quantitative: Screen media as regulation and babysitter Two items were created for this study to assess how parents use screen media to regulate children's emotions or behavior. On a 4-point Likert scale from "1 = Strongly agree" to "4 = Strongly disagree," parents reported their agreement with the statements: "When my child is upset, giving my child a device with a screen is the easiest way to have them calm down" (regulation tool) and "When I need to get work done at home, I often give my child a mobile device to play with, let them watch TV, or play video games to keep them occupied" (screen as babysitter). A fifth option allowed them to say that the items were not applicable to them. Quantitative: Social desirability (covariate) Social desirability bias in the parents' reporting was assessed with a brief version of the Marlowe-Crowne Social Desirability Scale that included true or false ratings of the three highest loading items of the scale (Reynolds, 1982) . Higher score indicated a more socially desirable response, and the average of the sum of these dichotomous items at each time point was used as a covariate in regression analyses (possible range = 0-3; α = .75). In line with Braun and Clarke's thematic analysis methodology (2006), coders read participant responses multiple times and independently generated initial codes. The third author served as the primary coder for T1 and the second author served as the primary coder for T2. These two coders read participant responses and independently generated initial codes using both analyst-driven values coding and emotion coding techniques (beliefs and values, attitudes including concerns, practices, and emotions), and data-driven coding (any other themes present in the data) (Braun & Clarke, 2006; Saldaña, 2015) . The first author then organized codes into potential themes and the entire coding team met on multiple occasions to resolve discrepancies, establish consensus, and agree on final themes, after which a thematic diagram was created and illustrative quotes were selected for each code (Braun & Clarke, 2006; Hill et al., 2005) . Aim 1 To address our first question of how SMU, PMU, and parent perceptions of media have changed from prepandemic to post-onset, we conducted Wilcoxon signed ranks tests to compare the means across both years, given the non-normal distribution of our data (assessed via visual inspection and the Shapiro-Wilk's test). The Wilcoxon test effect size r was used. To address our second question-what moderates the change of SMU and PMU pre-pandemic to postonset-we conducted two hierarchical regressions, using T1 SMU and PMU to predict T2 SMU and PMU, respectively. Covariates (social desirability and income) and main effects (T1 SMU and PMU, T1 child age, T1 participation with child media, and T1 perceptions of media) were entered into the first step, followed by twoway interactions between T1 SMU and PMU and the main effects. Because we still wanted to examine the associations of using media as a regulation tool or babysitter, while accounting for those that indicated "not applicable" (meaning they did not believe the item applied to their child), those two main effects and their interactions with T1 SMU and PMU were entered into a third step with the acknowledgment that this would be a smaller sample size. Additionally, post hoc analyses were conducted treating age as a dichotomous variable, with "1 = age 5 or older." This was to examine whether or not the child being school-aged during the pandemic (i.e., around 6 years or older) would be associated with their screen media use. In both models, variables included in the interaction analyses (i.e., all variables other than covariates) were mean centered to reduce collinearity (Aiken & West, 1991) . Main effects and moderations were considered significant if the p-value was less than 0.05; effect sizes were determined by the regression coefficients and ∆R 2 . The proportion of missing data was very minimal across variables: One participant was missing 1 item from the Problematic Media Use scale at T1, one participant was missing items across the participation with child media items, and one participant was missing one screen media use item from 2019. Given this very small amount of missingness, the PMU score for the participant whose problematic media use item was missing was simply averaged across all other eight items for that participant, and the participants missing participation with child media items and one SMU item were given the sample mean. Mean substitution is an appropriate method to address missing data at such low levels of missingness (Parent, 2013) . Analyses were run both with and without this strategy and the results stayed the same. Additionally, many participants chose "N/A" for items related to media as a regulation tool and as a babysitter. Given that we did not want to assume why they indicated "N/A" and imputation was not the best strategy, they were instead removed from the final step of the regression analysis. Aim 1: Qualitative and quantitative analyses (pre-pandemic to post-onset change) In total, 57% (73) of the sample responded to the openended questions at one or both time points (prompts are described above, "Qualitative: Parent Perceptions of Child Media Use"). Fifty-one participants responded to the pre-pandemic (T1 or 2019) prompt and 42 participants responded to a post-onset (T2 or 2020) prompt. Twenty participants responded to the prompts at both time points. Supporting Information C shows a diagram of the emotions in parents' response regarding their children's media in each year, with the overlapping section representing emotions present in both years. A wider array of emotions was observed post-onset, spanning a larger affective range than pre-pandemic. Pre-pandemic emotions were primarily negatively valenced, including unhappy, concerned, and confused (although there is one positively valenced emotion captured in both years: grateful). Post-onset emotions included both negatively valenced emotions and more neutral or positively valenced emotions like relieved, mindful, ambivalent, and bored. Additionally, the unique emotions seen post-onset also include loss, stressed, distracted, fear, sad, relieved, dislike, uncertain, dissatisfied, and isolated. Thematic analysis of parent responses revealed eight major themes related to screen media use before and during the COVID-19 pandemic. Themes and codes are displayed in Figure 1 , and themes are organized within three broad domains: beliefs and values (2 themes), practices (3), and attitudes (3). Additionally, illustrative participant quotes for each code can be found in Supporting Information D. All eight themes were present both pre-pandemic and post-onset (in italics below), with some identical codes across the time points (17 codes) and others codes unique to one time point (15 pre-pandemic, 10 post-onset). Within the theme of positives and negatives of screen media, across both years parents believed media can have F I G U R E 1 Thematic diagram of parent beliefs and values, practices, and attitudes about screen media use before and during the COVID-19 pandemic. Note. Themes (centered, bolded at the top of each box) and codes (phrases within each box) are organized within three broad domains (written out on the left: Beliefs and Values, Practices, and Attitudes). When codes were consistent across both 2019 and 2020 (both pre-pandemic and post-onset), they are centered and in gray. When codes were present only in 2019 or 2020, they are depicted on the appropriate side of the chart (left: 2019/pre-pandemic and right: 2020/post-onset) and are surrounded by white negative consequences for children (e.g., poor child behaviors). However, pre-pandemic only, parents also expressed that there is an educational value for media and that media skills are important for a child's future. As one parent of a 3-year-old stated, "it is important for kids to become digitally intelligent" (ID111). Only pre-pandemic parents saw a trade-off of media, including believing too much screen time takes away from other activities, like playing. Within another theme of monitoring screen time, parents across both years believed it is important for media to be monitored and limited, even mentioning specific software to be used to send materials safely to children. Parents across both years also believed that context plays a large role in how easy it is to manage screens, mentioning that screen time is easier to manage with one child, when it is not winter, and when both parents do not work full time. However, only pre-pandemic did parents believe media use was a privilege for their children, and that other families around them also struggle with device use. As one mother of a 5-year-old stated pre-pandemic, "I have a lot of 'mom guilt' about [not enforcing stricter screen time limits], and I'm certain other caregivers feel guilt around this issue as well" (ID53). Parental mediation of their child's media was a theme in parents' discussion of their practices. Across both years, parents were aware of how they were limiting and monitoring screen content, if they were using screens as a way to occupy their child when they were busy, and how they were participating in coviewing with their child (e.g., watching shows together as a family). However, prepandemic only, they were trying to use screen media as an educational tool, occasionally using it as a reward, and tended to view specific reasons and times for media use (e.g., using the Wii for exercise, or using media as a "brain break" [ID72]). Post-onset only, parents were using mediation in much more specific ways related to COVID-19. They were allowing more screen time for at-home distance learning. Also, because of increased screen time, post-onset parents were more mindful of their child's media use, including avoiding its use as a babysitter or attempting healthy media use. Post-onset parents also said they were less able to monitor their child's screen media during the pandemic and were less restrictive with screen time and content. For example, "My husband's and my remote workload increased dramatically and we were not able to support or monitor [screen time] almost not at all. Eventually we realized that our children got addicted to screen time" (ID46, parent of 9-year-old). Another theme was parents discussing their child's screen time practices. Across both years, parents mentioned their child's screen use differed based on context (e.g., less in the summer, more when they communicate with family). However, only pre-pandemic parents mentioned their child's screen time quantity in relation to when they purchased devices (e.g., purchase an Xbox for a child's birthday) and where they are in the home (e.g., not keeping a television in the main living space). Only post-onset parents mentioned their children were using more screen media since the pandemic began and explicitly mentioned increased iPad and tablet use daily and not just as a reward. As one parent of a 4-year-old stated, "Pre COVID-19 we only used ipads on plane rides… now they are used daily" (ID104). Finally, parents discussed their child's activities and behavior surrounding screen time. Parents across both years mentioned the ways in which their children were both influencing and responding to screen media limits (e.g., child is receptive to screen time limits, children find workaround of screen media limits, undesirable child behavior leads to screens getting taken away). Parents in both years also mentioned other children were influencing their child's screen media use, including friends and older siblings. For example, one parent of a 5-year-old at T1 stated, "My youngest is more interested in screen media because of his older brother's interest" (ID62). Parents at both years also described non-screen time activities their children participate in, such as family game nights and other activities with siblings. However, only pre-pandemic parents explicitly mentioned behavior changes they observe in their child due to screen media (e.g., increased swear words, becoming irritable, imitating what they see on the screen). Parents expressed attitudes about the factors influencing their child's media. Across both years, parents listed reasons they believed their children's media use fluctuated (e.g., being an only child, having a stay-at-home parent). Parents also thought siblings and peers influence their child's screen media use (e.g., thinks their child's media use is higher because their friend's screen media use is not monitored). However, pre-pandemic only, parents perceived that child age had an influence on media use, such that their child would use more media as they get older. However, post-onset only, parents tended to view their child's screen media use as influenced by distance learning (e.g., inevitable screen time increase due to virtual schooling). Additionally, post-onset parents said they thought their child's screen use was changing (or not changing) due to COVID-19. A parent of a 3-year-old said that they "have tried to keep media use similar to pre pandemic. Some increase was [inevitable]" (ID88). Within another theme, parents expressed different thoughts and strategies regarding media. Across both years, parents expressed opinions of screen time and challenges with media use and monitoring (e.g., finding it challenging to monitor child media use, not knowing how to monitor media use, thinking their child is getting too much screen time). In addition, pre-pandemic only, parents said they and their co-parent had different opinions regarding screen time and expressed more parental guilt around their child's screen media use. Pre-pandemic parents also thought their parental mediation strategies (e.g., being strict with limits, not placing a television in a living room) were effective. However, post-onset parents were concerned their child was getting too much screen time because of COVID-19: As one parent of a 4-year-old stated regarding the pandemic, "Screentime has SKYROCKETED, never in a million years would I imagine we would be using screens this much" (ID104). Post-onset, parents also expressed concerns about their child's expected screen media use after COVID-19 such as concerns about the high amount of screen time becoming a new norm and anticipation of difficulty reducing screen time after COVID-19 resolves. Finally, parents expressed their thoughts about outcomes of screen media use-and for this theme, all codes were present in both years. Parents in both years expressed concerns surrounding negative behavioral effects of media on their child (e.g., irritability or aggressiveness, rage, wetting themselves). For example, one parent of a 7-year-old stated their son would "love to have all the game time in the world -but his behavior gets off track when he gets too much screen time" (ID103). Parents in both years also thought their children could become addicted to or dependent on media, explicitly mentioning dependence on the iPad and television and noticing how their child's relationship to media had changed since the pandemic began. Parents also thought there were some positives of media, including helping keep their child quiet, being educational, and helping with hand-eye coordination. Quantitative analyses. Figure 2 shows the overlapping distributions of these three variables across the 2 years. T2 screen SMU and then PMU were regressed on their T1 counterparts, covariates, and potential moderators. When regressing T2 SMU on all main effects and interactions in step 2, T2 SMU was significantly associated with T1 SMU (β = 0.45, p < .001) and child age (β = 13.13, p < .01). Thus, a 1-year age increase was associated with a 13.13-min increase in T2 SMU. Parent perceptions of media use, parent participation with media, and covariates were not significant, nor were any interactions (Supporting Information E). Media as a regulation tool and babysitter were not significant as main effects nor interactions. When regressing T2 problematic media use (PMU) on all main effects and interactions in step 2, T2 PMU was significantly associated with T1 PMU (β = 0.66, p < .001), child age (β = 0.11, p < .01), and parent perceptions of media as hurting their child (β = 0.27, p < .05). A 1-year increase in child age was associated with a.11-unit higher T2 PMU, and a one-unit increase in negative parental perceptions of media was associated with a 27-unit higher T2 PMU. Parent participation with media and the covariates were not significant nor were any interactions (see Supporting Information F). Media as a regulation tool and a babysitter were not significant as main effects or interactions (Model 3). Given our hypothesis that these associations might differ for preschool and school-age children, we also ran the same analyses using a dichotomous age variable, such that 1 = 5 and older (Supporting Information G and H). With this, being 5 years or older at T1 was associated with a 65.16-min higher SMU at T2. Additionally, a PMU × age interaction was now present, such that for older children, a one-unit change in PMU at T1 was associated with a.41-unit higher difference in T2 PMU than for younger children (β = 0.41, p < .05). Screen media use has become an increasingly hot button issue for families, researchers, and educators across the United States and globally since the COVID-19 pandemic began (e.g., Király et al., 2020; Nagata et al., 2020) . The present study utilized a mixed methods approach to understand how and how much a sample of children in the United States were using media pre-pandemic versus post-onset, using parent report data collected at two time points (2019 and 2020). We utilized both qualitative analyses to investigate the nuances of how parents were discussing screen media, and quantitative analysis to investigate how parents were perceiving media and rating their child's general screen media use (SMU) and problematic media use (PMU) across years (and what might moderate that trajectory). By understanding how media use shifted prepandemic to post-onset, we can better know what to expect regarding media throughout the pandemic's course, through future COVID-19 and other pandemic waves, and how to talk to families about their child's media use. Below, we provide an integrative discussion of the quantitative and qualitative findings (themes are italicized). Both the quantitative and qualitative results demonstrated that there was a shift in non-school-related SMU prepandemic to post-onset; children went from using an average of 149 min per day (2.48 h) to an average of 199 min per day (3.32 h) post-onset (medium effect size change; see Figure 2a ). This increase held even after age, income, and social desirability were accounted for. Pre-pandemic SMU also significantly predicted post-onset SMU, suggesting stability in rank order compared to other families, even as all families tended to see a rise in SMU. Qualitative findings supported those conclusions as parents discussed their children using more screen media and having more access to screen media devices like iPads (see Figure 1 and Supporting Information D theme Screen time: "More screen media use during COVID-19" and "More access to iPad/tablet during COVID-19"). Post-onset, parents tended to attribute this rise in SMU to either the pandemic or a change in their child's learning situation (see Figure 1 and Supporting Information D theme Factors influencing screen media). While the quantitative results did not specifically take school-related media use into account, these qualitative findings clearly suggest parents were also seeing an inherent increase in screen time due to distance learning, as seen in other studies (Lau & Lee, 2021; Pombo et al., 2020) . PMU also increased from pre-pandemic to post-onset quantitatively. Although the median of PMU only increased by 0.22 on a scale from 1 to 5, this change was still a medium effect size (see Figure 2b for a graphic representation of the distribution). As with screen media use, the finding that pre-pandemic PMU predicted postonset PMU suggests that there was rank order stability between families, even as all families saw a rise in PMU. Qualitative results reflected parent concerns about their child's media use. Regarding child activities and behavior surrounding media that could theoretically include problematic behaviors, parents did not express much difference across the 2 years. In both years, parents observed that their children were influencing screen media limits, as were other children; interestingly, behavior changes due to screen media were only brought up prepandemic (see Figure 1 and Supporting Information D: Child activities and behavior). Additionally, in both years, parents perceived their child could be addicted to screen media use and there were negative behavioral effects related to it (see Figure 2 and Supporting Information D: When examining the distal, proximal, and maintaining factors of PMU in the IT-CPU model, qualitative analyses yielded many factors that could influence PMU in earlymid childhood. Distal factors, in this case, relate to the new digital learning space and challenges that COVID-19 were bringing to the family environment that could result in more PMU (see Figure 1 and Supporting Information D: Factors influencing screen media and Parent thoughts and strategies regarding media). This is also validated by other studies, showing that 93% of respondents in U.S. households had children engaging in some form of "distance learning," the majority via online materials (U.S. Census Bureau, 2020). However, as described in our Methods section, most of our sample that was in school was only using media for virtual learning less than 4 hours per day (93%). SES is also a distal influence on PMU; thus, family income was included as a covariate in all analyses. Other distal influences of PMU, including digital environmental design and parent's PMU, were not measured in the current study. Proximal factors in the IT-CPU model include challenging child behaviors, which we saw qualitatively in the Child activities and behavior and Outcomes of screen media use themes. Proximal and maintaining influences of problematic media use also include social factors, which were evident here with parents discussing how their child's media use is influenced by their peers or siblings, and how they think their child's media use increases because of their peers (Figure 1 and Supporting Information D: Child activities and behavior: "Other children influence media use"; Factors influencing screen media: "Siblings and peers"). Parent and family proximal factors can also influence PMU, including parental stress, parental media beliefs, and inconsistent or limited parenting practices, all of which were apparent in our analyses (Domoff et al., 2020; Lauricella et al., 2015) . Parental stress was seen in our emotional content analysis, where there was a wider array of emotions post-onset and the word "stress" was present post-onset, but not pre-pandemic (see Supporting Information C). Interestingly, "guilty" was only mentioned pre-pandemic. It is possible that while parents were expressing more concerns about their child's media, they were more accepting of it and no longer felt as much guilt given other pandemic-related stressors. Parental media beliefs were both qualitatively seen and quantitatively related to problematic media use. Parents expressed specific beliefs about screen media use both at both time points, seeing negative consequences at both time points and believing screen media use should be limited and monitored (see Figure 1 and Supporting Information D: Positives and negatives of screen media and Monitoring screen time). There were more codes expressing positives of screen media pre-pandemic than post-onset, though the quantitative analysis did not result in a significant quantitative difference in perceptions of child screen media use between years. Additionally, within attitudes, parents did see positive outcomes from media for their children at both time points (see Figure 1 and Supporting Information D: Outcomes of screen media use: "Media can be positive"). Parents also expressed specific concerns about screen time both during COVID-19 and after the pandemic ends, wondering how they would get back to a "normal" amount of screen time (see Figure 1 and Supporting Information D). A less positive view of child media use was quantitatively associated with more PMU post-onset, even after controlling for pre-pandemic PMU. Although we did not specifically hypothesize about the direction of the main effect, this is contrary to the hypothesis that a positive perception of media would be associated with a steeper increase in PMU. It is possible that these parents with more negative perceptions of media use pre-pandemic were also more likely to become overwhelmed by the increased media use when the pandemic hit, therefore leading to reduced parental mediation strategies. Additionally, perhaps children with parents who had a worse perception of media pre-pandemic were simply "catching up" with the other children's problematic media use post-onset. Future longitudinal research should aim to further tease out why negative perceptions of media could result in a steeper increase of problematic media use over time. Media parenting practices can also influence PMU, and this was certainly borne out in our qualitative analyses, though not in the regression analysis. When including parent participation with their child's media use in our models, it was not associated with SMU nor PMU. However, the parental mediation theme contained many codes relevant to media parenting practices at both time points. As defined previously, parental mediation strategies can encompass active mediation (discussing media content with children), coviewing (watching media together without any discussion), and restrictive mediation (prohibiting certain media content and setting specific rules; Barkin et al., 2006; Valkenburg et al., 1999) . Parents were engaging in restrictive mediation at both time points and coviewing; additionally, parents expressed that they were less able to restrict and monitor screen time during COVID-19, reflecting the limited media parenting practices Domoff et al. (2020) refer to (see Figure 1 and Supporting Information D). With the qualitative data, we cannot speak to whether these differing parental mediation strategies led to increased post-onset PMU; however, taken together, these results suggest that parents were concerned with their mediation strategies postonset and felt they were less able to engage in them than before. Our regression analyses differ from our qualitative analyses perhaps due to the nature of the questions we used for parent participation with media. These quantitative items mostly related to coviewing (e.g., watching television together), not necessarily engaging in active or restrictive mediation. Although quantitative data with parents in Turkey around the same time period as this data collection revealed most parents said they were instituting screen time rules (Eyimaya & Irmark, 2021) , our qualitative analyses show that parents do not feel as capable to do so as pre-pandemic. Additionally, these results show that, while parents are engaging in increased monitoring of their child's media use to avoid problematic behaviors during the pandemic (consistent with Király et al., 2020; Vanderloo et al., 2020) , they are struggling to meet this goal with the many other demands the pandemic is bringing to their lives. Finally, we also saw some maintaining factors of the PMU from the IT-PCU model (Domoff et al., 2020) , though these results were more mixed. As mentioned previously, another maintaining factor of problematic media use is screen media influence-at both time points, parents mentioned that a sibling or peer's higher SMU can result in an increase their child's SMU (see Figure 1 and Supporting Information D). Another maintaining factor of PMU could be positive reinforcement for the child, and in this study, we conceptualized this as parents using media as a regulation tool. However, when including this in our regression analyses, it was not related to problematic media use. Given the variety of ways in which screen media use relates to self-regulation (Gordon-Hacker & Gueron-Sela, 2020; Linder et al., 2020) , future research on media as a regulation tool during the pandemic should aim to use a larger sample and more nuanced measurement (i.e., not a one-item measure) to capture this potential association. Media as a babysitter were also not quantitatively associated with PMU nor SMU (potentially due to the smaller sample size used for that analysis at n = 106), but it was a prominent code in the parental mediation theme: Parents were using screens to keep their child occupied, and postonset parents said this was especially needed because of working and schooling from home (see Figure 1 and Supporting Information D). This is consistent with samples of parents before COVID-19 as well, where samples of Dutch and U.S. parents viewed media as a potential babysitter (Nikken, 2019; Wartella et al., 2013) . The small differences between the qualitative and quantitative analyses (e.g., parental mediation being prominent in the qualitative analyses but not the quantitative analyses) speak to both the added nuance qualitative analyses provided in this study and the fact that parents might be conceptualizing parental mediation in different ways than our brief quantitative items could measure. Together, the qualitative and quantitative results suggest that there was an increase in PMU between prepandemic and post-onset, and this increase was possibly influenced by a variety of distal, proximal, and maintaining factors including the COVID-19 pandemic, distance learning, other children, child behaviors, parental mediation, and positive media reinforcement. As hypothesized, age was significantly related to postonset PMU and SMU; additionally, when treated as a dichotomous variable, we saw a PMU X age interaction such that for children aged 5 and older, there was a steeper increase in PMU. A year increase in age pre-pandemic was associated with a 13-min increase in post-onset SMU and a 0.11-unit increase in PMU. With our dichotomous analysis, being a child 5 years or older pre-pandemic was associated with a 65.12-min higher SMU post-onset than their younger counterparts. These results speak to the importance of age and school-age categorization when considering the widespread challenges children face during the COVID-19 pandemic, and how media use might factor into their lives. Children at different developmental stages face different challenges during the pandemic, including different social demands. While a younger child might not necessarily need to engage with friends virtually, older children have more out-of-family connections that they might be trying to maintain (Masten & Motti-Stefanidi, 2020) . The items of the Problematic Media Use Measure -Short Form (Domoff et al., 2019) include screen media use interfering with family activities; screen media causing problems; child becoming frustrated when they do not use screens media; and it being difficult to stop the child from using screen media. It follows that this would be more difficult for older and school-age children versus their younger counterpartsespecially as parents are more pre-occupied with work and childcare demands at home. Older children, who are seen as more independent, are (quite literally) more likely to be left to their own devices. While our study has many strengths, including its mixed methods approach, data collection at two time points, and the inclusion of both screen media use and problematic media use, it is not without limitations. The first is that our sample was primarily White, middle to upper class, and highly educated due to the convenience sampling approach that was used. This could affect the type of devices these children had access to and the types of jobs their parents had during the pandemic: If their parents had a job that enabled them to work from home, they would have a different ability to monitor their child's media use than parents who had to go out and work. Additionally, our sample was primarily comprised of two-parent households, which would also affect the ways in which they could monitor media. Parents at different income levels were also facing different levels of stress during the pandemic, which could affect the family and media environments (Cluver et al., 2020) . One of the distal factors in Domoff et al.'s (2020) IT-CPU is SES/poverty, and so families at lower levels of SES could have much different experiences of problematic media use compared to the families in the current sample given higher economic stress during the pandemic. These factors, taken together, mean that the findings are limited in generalizability due to the demographics and convenience nature of the sample. Second, the two measures that assessed media as a regulation tool and media as a babysitter were one-item measures that were created for this sample, and many parents selected "N/A," meaning that they could not be included in the quantitative analysis. Additionally, there was a negative correlation between these items and problematic media use, indicating further psychometric analysis is needed. Future research should utilize more in-depth measures of how parents are using media (as a regulation tool, as a babysitter, etc.) with larger samples to uncover potential associations with screen media use and problematic media use. Third, due to the nature of this global pandemic, it was not possible to have a control group. Thus, it is possible our results could, in part, reflect normative increases in SMU and PMU with age. Our qualitative analyses revealed that pre-pandemic parents thought their child's SMU would increase with age, introducing an important confound. Previous research evidenced a 35-minute increase in daily SMU between children ages 2-4 years and 5-8 years (Rideout & Robb, 2020) ; our dichotomous regression analysis showed that for children 5 and older, there was a 65-min increase in SMU, much larger than previous normative research. Additionally, our qualitative analyses suggested that these changes were, in fact, COVID-19-specific. Importantly, this survey was parent report, meaning that we did not have objective measures of screen media use nor clinical observations of problematic media use. Although the PMUM-SF measure used for PMU in the study is meant for parent report, clinical observations of a child's PMU behaviors would be ideal. However, due to the nature of the pandemic, and the fact that the construct of child PMU is still in its nascent stages, parent report was the only available methodology for this sample. Parent's own PMU is another potential distal influence on child PMU that was not assessed in this study. Parents tend to underreport their child's media use (Wood et al., 2019 )-while we tried to take this into account by including social desirability in our models and still saw an increase in PMU and SMU, it is possible we underestimated actual SMU. We also limited the top item of the SMU scale at "two hours or more," which could have resulted in underestimated SMU. Future research could use more objective measures of screen media use (e.g., utilizing screenshots of app use, media device trackers; Kaye et al., 2020) . We also did not obtain data on what the children were watching, just how much. However, our inclusion of problematic media use as a focal construct in this study at least addresses how children are using media, as well as how much. Finally, our qualitative analyses were limited to thematic analysis of open-ended questions that were optional for parents to answer; as such, only a select subset of our sample provided responses we analyzed qualitatively. There are different and more comprehensive qualitative data collection methodologies that can be employed in the future to study this phenomenon, including observations, focus groups, and interviews . Future work should aim to analyze longitudinal trajectories of SMU and PMU through more in-depth qualitative methodologies. Our findings regarding increased problematic and screen media use, age differences, and increased emotionality of parents and concerns about their child's media use have several implications. First, as researchers are already doing (e.g., Vanderloo et al., 2020) , they can provide advice for parents about how to monitor their child's media use during a time when they are stuck at home. Additionally, researchers should consider the unique challenges the COVID-19 pandemic places on school-age children and their parents relative to younger children. Parents are also clearly experiencing many more emotions during the pandemic relative to pre-pandemic, including stress, loss, and isolation. A focus on parental mental health during and after the pandemic will be important moving forward, especially as their mental health has implications for their child's well-being during this time (e.g., Daks et al., 2020) . Finally, our thematic analyses clearly show that parents are concerned about their child's screen media use not only during the COVID-19 pandemic but also once the pandemic is over. As schools begin to open up more and vaccines are distributed, guiding parents in how to help their child adjust to "back to normal" routines will be of the utmost importance, especially as there were already high levels of stress around children's screen media use pre-pandemic . The COVID-19 pandemic provided a unique challenge for parents around the world, including in the United States, regarding their children's media use. There are increased concerns about too much screen use and its developmental implications (e.g., Nagata et al., 2020) . The findings of our mixed methods study show both an increase in screen media use and problematic screen media use between pre-pandemic and post-onset, as well as an age moderation such that school age children on average had a steeper increase in problematic media use than preschool age children. Additionally, the qualitative findings support the increase in screen media use and distal, proximal, and maintaining factors of problematic media use. Significant events and changes to family life are reflected in children's media use as well as parents' attitudes and behaviors surrounding it. Future research will be needed to trace the post-pandemic changes in the amount, content, and parental mediation of children's media use. Multiple regression: Testing and interpreting interactions Parenting children in the age of screens Parental media mediation styles for children aged 2 to 11 years Using thematic analysis in psychology Children and adolescents and digital media Parenting in a time of COVID-19 Parenting and digital media Research design: Qualitative, quantitative, and mixed methods approaches Designing and conducting mixed methods research Psychological flexibility and inflexibility as sources of resiliency and risk during a pandemic: Modeling the cascade of COVID-19 stress on family systems with a contextual behavioral science lens Interactional theory of childhood problematic media use Development and validation of the problematic media use measure: A parent report measure of screen media "addiction" in children Effect of COVID-19 lockdown on sleep behavior and screen exposure time: an observational study among Indian school children Family resilience and psychological distress in the COVID-19 pandemic: A mixed methods study Relationship between parenting practices and children's screen time during the COVID-19 Pandemic in Turkey The COVID-19 pandemic: Technology use to support the wellbeing of children Maternal use of media to regulate child distress: A double-edged sword? Longitudinal links to toddlers' negative emotionality Parental mediation and adolescent screen time: A brief overview. Education and Health The upside to screen time. The New York Times Consensual qualitative research: An update Behavioral and emotional disorders in children during the COVID-19 pandemic Executive function and screen-based media use in preschool children. Infant and Child Development Then the coronavirus happened. The New York Times The conceptual and methodological mayhem of "screen time Impact of parents mobile device use on parent-child interaction: A literature review Preventing problematic internet use during the COVID-19 pandemic: Consensus guidance Parents' views on young children's distance learning and screen time during COVID-19 class suspension in Hong Kong Young children's screen time: The complex role of parent and child factors Digital media and sleep in childhood and adolescence The moderating role of parental strain on the relationship between child media use and regulation Multisystem resilience for children and youth in disaster: Reflections in the context of COVID-19 Surveying parental mediation: Connections, challenges and questions for media literacy Screen time for children and adolescents during the coronavirus disease 2019 pandemic The relation between television exposure and executive function among preschoolers Parents' instrumental use of media in childrearing: Relationships with confidence in parenting, and health and conduct problems in children How and why parents guide the media use of young children Handling item-level missing data: Simpler is just as good Maintaining everyday life praxis in the time of COVID-19 pandemic measures (ELP-COVID-19 survey) Correlates of children's physical activity during the COVID-19 confinement in Portugal Smartphones and children: Relationships, regulation, and reasoning Parent perspectives on their mobile technology use: The excitement and exhaustion of parenting while connected Development of reliable and valid short forms of the marlowe-crowne social desirability scale The common sense census: Media use by kids age zero to eight The Common Sense census: Media use by kids age zero to eight The coding manual for qualitative researchers Roughly 40% of Americans are working from home full time because of the pandemic Screen-time is associated with inattention problems in preschoolers: Results from the CHILD birth cohort study Developing a scale to assess three styles of television mediation: "Instructive mediation", "restrictive mediation", and "social coviewing Applying harm reduction principles to address screen time in young children amidst the COVID-19 pandemic Parenting style, parental stress, and mediation of children's media use As kids' screen time surges during the pandemic, here's what research suggests Parenting in the age of Digital Technology: A National Survey Rethinking screen-time in the time of COVID-19 Concordance of child and parent reports of children's screen media use Guidelines on physical activity, sedentary behaviour and sleep Children's screen and problematic media use in the United States before and during the COVID-19 pandemic The first author was supported by a pre-doctoral fellowship from the National Institute of Mental Health of the National Institutes of Health under Award Number T32 MH015755. Data collection was supported by funds awarded to G. M. Ferguson and S. M. Carlson by the Institute of Child Development, University of Minnesota. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.We thank the participating parents for their time both before and during the COVID-19 pandemic.