key: cord-0960784-3lzg8dhi
authors: Peplak, Joanna; Klemfuss, J. Zoe; Yates, Tuppett M.
title: Parent-Adolescent Conversations About COVID-19 Influence Adolescents’ Empathic Concern and Adherence to Health Protective Behaviors
date: 2021-08-23
journal: J Adolesc Health
DOI: 10.1016/j.jadohealth.2021.08.013
sha: 008a8e004c7f145807c3d5311a237290129e7c2d
doc_id: 960784
cord_uid: 3lzg8dhi

PURPOSE: This longitudinal investigation assessed how the frequency of parent-adolescent conversations about COVID-19, moderated by adolescents’ stress, influenced adolescents’ empathic concern and adherence to health protective behaviors (HPBs) throughout the first year of the COVID-19 pandemic. METHODS: Participants were 181 adolescents (M(age) = 15.23 years; 51% female; 47% Latinx) and their parents. Frequency of parent-adolescent conversations about COVID-19 (i.e., pandemic-related symptoms, health behaviors, and social effects), empathic concern toward vulnerable others, and adolescent HPBs were assessed via surveys in the first months of the pandemic, and empathic concern and HPBs were assessed again nine months later. RESULTS: Results revealed that more frequent parent-adolescent conversations early in the pandemic predicted increased adherence to HPBs throughout the pandemic when adolescents reported low stress (direct effect), but conversation frequency predicted decreased adherence to HPBs via reduced empathic concern when adolescents reported high stress (indirect effect). CONCLUSIONS: Parents and other socialization agents, such as teachers, should be sensitive to adolescents’ stress prior to engaging them in frequent conversations about the pandemic to mitigate the potential negative impact these conversations may have on adolescents’ empathic concern and adherence to HPBs. Decreasing adolescents’ stress may be an initial step in promoting effective message transference. Collective action (including wearing masks and receiving the vaccination) remains critical to overcome COVID-19. The current contribution to understanding processes underlying adolescents’ adherence to recommended HPBs is critical pandemic fatigue and stress continue to rise.

This investigation demonstrated that and how the frequency of parent-adolescent conversations about COVID-19 influenced adolescents' adherence to COVID-19 health protective behaviors over the first pandemic year. Generating knowledge about psychosocial mechanisms underlying adolescents' health behaviors may inform the development of strategies to encourage social responsibility amongst youth during this pandemic and beyond.

Since the United States (U.S.) declared COVID-19 a national emergency in March of 2020, efforts to halt the human transmission of the virus have led to pervasive and prolonged disruptions to daily life, such as stay at home orders, school closures, and cancellation of activities [1] . These restrictions have contributed to increased family conflict, financial strain, and psychosocial distress among youth and parents alike [2, 3] . Nevertheless, consistent adherence to recommended health protective behaviors (HPBs; e.g., social distancing, masking) has been, and continues to be, critical in limiting transmission of the virus and preventing illness and death [4] .

One group that may be at particular risk of violating HPBs protocols is adolescents.

Adolescence is marked by decreasing family attachment, increasing peer attachment, and more frequent engagement in risk-taking behavior [5] . These normative developmental processes and behaviors were disrupted by pandemic-related restrictions that required adolescents to stay at home with their family, keep physical distance from friends, and avoid public spaces.

Adolescents may find it difficult to accept the social and emotional costs of COVID-19-related restrictions, and thus may be less likely to follow recommended health behaviors thereby elevating the risk for continued community transmission.

This investigation sought to advance our understanding of adolescents' COVID-19 HPBs across the first year of the U.S. COVID-19 pandemic. The Health Beliefs Model (HBM) explains change and maintenance of health-related behaviors and suggests that intrapersonal beliefs, such as self-efficacy, perceived threat and severity of the virus, and perceived benefits of health behavior adherence influence individuals' treatment seeking and adherence [6] . Recent research on adherence to COVID-19 HPBs suggests that interpersonal and affective factors also influence individuals' adherence to health behaviors [7] . Thus, drawing from the HBM and related research, we examined the individual and interactive influences of psychosocial (i.e., frequency of parent-adolescent conversations) and intrapersonal psychological factors (i.e., youths' perceived stress and empathic concern) on adolescents' HPBs. Generating knowledge about whether, how and in what contexts these factors influence adolescents' motivation to follow HPBs may inform the development of strategies to encourage social responsibility during this pandemic and beyond.

As primary socialization agents, parents are responsible for teaching their children HPBs that prevent the spread of COVID-19 and may be important sources of adolescents' information regarding the global impacts of the virus [8] . Research demonstrates that parent-child conversations influence children's prosocial behavior (i.e., voluntary behavior that is intended to benefit others) [9] and direct conversations about specific behaviors often translate into youths' engagement in those behaviors [10] . Adherence to HPBs in the context of COVID-19 may be considered a form of prosocial behavior, particularly given that some of these behaviors protect others as much or more than oneself (e.g., mask wearing) [11] . Thus, it is likely that frequent parent-adolescent conversations about COVID-19 may direct adolescents' attention to the J o u r n a l P r e -p r o o f pandemic and spur their motivation to engage in HPBs. Indeed, as the HBM posits [6, 7] , psychosocial factors, including conversations with family members, may effectively motivate individuals to seek medical interventions. Although there is evidence that parents engaged their children (birth to 18 years) in conversations about viral transmission suppression (e.g., personal hygiene and social distancing) during the early phase of the pandemic [8] , the current study was the first to assess whether and how the frequency of COVID-19-related conversations influenced changes in adolescents' engagement in COVID-19 HPBs across the first year of the pandemic.

Parenting practices influence behavior, in part, via changes in prosocial emotions. For instance, empathic concern-feelings of concern or sorrow for others in need [9] -is promoted through the facilitation of perspective taking and emotional sensitivity to others, which, in turn, motivate prosocial action [12, 13] . Thus, frequently discussing COVID-19-related topics, such as HPBs and the societal impacts of the pandemic, may heighten adolescents' attention to the physical and psychological states and needs of others-particularly those most vulnerable to being affected by COVID-19. In turn, this heightened empathic concern may encourage adolescents to practice HPBs with the goal of protecting others from harm. Indeed, emerging evidence on pandemic-specific empathic concern suggests that inducing empathic concern via emotional content relevant to COVID-19 (i.e., a vignette of a vulnerable individual being infected with the virus) was more effective in increasing adults' motivation to socially distance and wear a mask in public than an emotionally neutral presentation of information about COVID-19 guidelines [14] . Based on this work, we hypothesized that frequent parent-adolescent conversations about COVID-19 would direct adolescents' attention to the pandemic and J o u r n a l P r e -p r o o f encourage adolescents' ongoing or increased adherence to HPBs by activating their pandemicspecific empathic concern.

During epidemics and pandemics, individuals may experience heightened stress and anxiety [15, 16, 17] . In turn, pandemic-related stress and anxiety may interfere with how individuals communicate and process information pertinent to the stressor [8, 18] . Because stress and anxiety can cloud adolescents' ability to remember and retrieve information [19] , it is possible that parent-adolescent conversations about COVID-19 might be less effective in promoting HPBs when adolescents are stressed. Additionally, frequent conversations about COVID-19-an anxiety-provoking topic [18] -might overwhelm stressed adolescents, thwarting their ability to extend their concern toward others in need. Indeed, youths' ability to experience concern for others hinges on their capacity to effectively down-regulate their own distress [9] . If youth are overwhelmed, they may experience personal distress (i.e., feeling psychologically and physiologically distressed by another's state) [20] , which may lead to lower empathic concern over time and, in the case of COVID-19, prompt less engagement in other-oriented HPBs. In the current study, we tested the hypothesis that adolescents' perceived stress would reduce otherwise positive relations between the frequency of parents' conversations with their adolescents and adolescents' adherence to COVID-19 HPBs by interfering with adolescents' ability to experience pandemic-specific empathic concern.

The present study had two overarching aims: First, we sought to examine the influence of psychosocial factors on adolescents' HPBs [6] . Specifically, we examined how the frequency of parent-adolescent conversations about COVID-19 related to adolescents' adherence to COVID-J o u r n a l P r e -p r o o f 19 HPBs over time, and whether this relation was mediated by changes in adolescents' pandemic-specific empathic concern. We predicted that the more often parents and adolescents engaged in conversations about COVID-19 early in the pandemic, the more likely adolescents would experience increases in empathic concern, and, in turn, engage in HPBs 9 months later.

Second, we assessed how adolescents' stress might influence these associations. Based on prior studies of personal distress and information processing [19, 20] , we expected that adolescents' perceived stress would qualify expected relations among COVID-19 conversations, empathic concern and HPBs, such that frequent conversations between parents and adolescents about COVID-19 would decrease adolescents' empathic concern if adolescents concurrently experienced high stress and, by extension, undermine their HPBs. We employed a two-timepoint longitudinal design to assess our research questions using data collected at the beginning of the pandemic when cases of COVID-19 were low but steadily rising (April and May of 2020), and during the second surge of COVID-19 in the U.S. when COVID-19 cases and deaths were at an all-time high (November and December 2020) [4] .

The current sample of 181 adolescent-parent dyads (Mage of adolescents = 15.23 years; SD = 0.57; 51% girls; 93% biological mothers) were part of an ongoing longitudinal study of child development among 250 children who have been followed since the preschool period. In the spring of 2020 and again in the winter of 2020, continuing families (N = 235) received an invitation to complete an on-line survey regarding the COVID-19 pandemic. Invitations were issued via phone calls, texts, and emails. Adolescents in the current study were diverse with regard to ethnicity and race (47% Latinx, 18% Black, 10% white and 25% multi-ethnic/racial) J o u r n a l P r e -p r o o f and representative of the surrounding community from which the sample was recruited [21] .

Participating families were similarly diverse with regard to economic status (20% resided below 130% of the poverty line and qualified for government assistance, such as food stamps) and about half the parents had completed some college or technical school (48.3%). There were no significant sociodemographic differences between the current participants and the broader sample from which they were drawn.

Families were recruited to participate in a study of "children's early learning and development" via flyers distributed to community-based childcare centers. Participants were screened by phone to ensure the child was (1) not diagnosed with a developmental disability, (2) between 3.9 and 4.6 months of age, and (3) English proficient. Data for the current study were obtained via online surveys using the Qualtrics survey platform due to the COVID-19 restrictions on in-person data collection. Surveys were completed individually on participants' own devices during April and May of 2020 and again during November and December of 2020.

Informed consent and assent for the current study was obtained prior to survey completion and surveys took approximately 1.5 hours to complete. Adolescents and parents were compensated for their time and all procedures were approved by the human research review boards of the participating universities.

At T1 adolescents and parents reported how often they talked to each other about six topics relevant to COVID-19 during the past 2 weeks from 1 (not at all) to 5 (daily). Topics included (1) symptoms and signs of COVID-19, (2) things to do to prevent getting or spreading J o u r n a l P r e -p r o o f (3) what to do if someone in their home got sick with COVID-19, (4) social distancing, (5) healthcare problems due to COVID-19, and (6) discrimination against people because of the pandemic. Adolescent and parent reports were significantly correlated (r = .20, p = .018) and were aggregated to capture both adolescent and parent perceptions of the frequency of engagement in these conversations to mitigate informant bias. As Kraemer and colleagues [22] argue, it is necessary to acquire and aggregate multiple independent observations from different informants within the same context to remove extraneous variance due to perspective.

Particularly given the shared nature of conversational experiences, aggregating multiple perspectives yielded a more valid account of how often parent-adolescent conversations were happening than relying on a single informant. Cronbach's αs equaled .82 for both parent-and adolescent-reported conversations and remained strong (α = .81) when all parent and adolescent items loaded together.

Adolescents completed the Perceived Stress Scale at T1 [23] . Participants reported how much they experienced 10 stress items during the preceding two weeks (e.g., "how often have you felt nervous and stressed?") using a 5-point Likert scale (0 = never to 4 = very often;

Cronbach's α = .77).

Adolescents reported on their empathic concern for those affected by COVID-19 at T1 and T2 using four items from Pfattheicher and colleagues [14] . Participants rated each item (e.g., "I am very concerned about those who are most vulnerable to getting sick from COVID-19") on a scale from 1 (not true) to 3 (certainly true; Cronbach's αs at T1 and T2 equaled .87 and .85, respectively). A standardized residualized gain score (RGS) was calculated to measure change in J o u r n a l P r e -p r o o f empathic concern from T1 to T2 [see 24]. A RGS represents the deviation of scores at T2 from the regression line of T2 on T1 scores, which leaves only variability that is unexplained by T1 empathic concern-construed as the variability due to change [24] . In other words, positive scores indicate larger-than-expected increases in empathic concern and negative scores indicated smaller-than-expected increases in empathic concern based on the adolescent's initial empathic concern score at T1. Although it is argued that RGSs may be biased in non-experimental studies when there are suspected subpopulations in the data [24] , we sampled from a population of individuals from one geographical region and who are of a similar socio-demographic background. Thus, we have no reason to suspect a biased estimation of the RGS in our model.

Adolescents reported on how often in the past 2 weeks they engaged in HPBs to protect them and others against COVID-19 across 14 items at T1 and 21 items at T2 on a scale from 1

(not at all) to 5 (always). Items involved HPBs reflecting cleanliness (e.g., hand washing, avoiding touching face), avoiding public spaces/travel, social distancing, and staying at home (Cronbach's alpha T1 = .88; T2 = .94). Additional items were added at T2 to reflect changes to CDC recommendations between T1 and T2 (e.g., wearing a mask). We created a composite score at each time point because we were interested in frequency of adherence to all HPBs recommended by the CDC, thus accurately capturing variations in adherence.

All analyses controlled for adolescents' age, gender, ethnicity/race (i.e., Latinx vs. non-Latinx), and family income-to-needs based on the parent's reported household income (assessed at age 12), which included all financial contributions to the household (e.g., salary, child J o u r n a l P r e -p r o o f support), divided by the appropriate poverty threshold for the household size and number of children under 18 in the home.

All analyses were conducted in SPSS 26 and Mplus (version 8.5) [25] . Study variables were evaluated to ensure they met parametric statistics assumptions. Using G*Power analyses, we determined that the sample size was sufficiently powered to detect small to medium effects [26] .

Following descriptive and bivariate analyses, a path model evaluated whether the frequency of parent-adolescent conversations about COVID-19, adolescents' perceived stress, and the interaction between the two at T1 predicted changes in adolescents' pandemic-specific empathic concern and changes in their adherence to HPBs from T1 to T2. Following Cohen and colleagues [28] , all continuous study variables were mean centered. We also tested the indirect effect of frequency of conversations (and its interaction with stress) on changes in adherence to HPBs via residualized changes in empathic concern (see Figure 1 for conceptual model). Effects were estimated using bootstrapping at 10,000 resamples to control for Type I error, obtain confidence limits and standard errors for the indirect effect test that are preferable to the traditional Sobel test [29] , and mitigate power problems due to the asymmetric and non-normal sampling distribution of indirect effects [30] . We rejected the null hypothesis (i.e., no indirect effect) if the 95% confidence interval (CI) of an estimate did not include zero [31] .

Of the 181 parent-adolescent dyads who participated, 80% of dyads completed both T1 and T2 surveys, 10% of dyads completed either the T1 or T2 survey but not both, and 10% of dyads were missing either parent or adolescent reports for the T1 or T2 survey. Across specific J o u r n a l P r e -p r o o f variables, data were missing for parent-adolescent conversation frequency (n = 37 were missing data from either or both respondents; 20.4%), perceived stress (n = 25; 14%), empathic concern at T1 (n =26; 14%) and T2 (n = 24; 13%), residualized change in empathic concern (n = 42, 23 .3%), and HPBs at T1 (n =24; 13%) and T2 (n = 18; 10%). Missing data were handled using full-information maximum-likelihood (FIML) as supported by Little's [27] missing completely at random (MCAR) test, χ 2 (30) = 30.89, p = .42.

Means, standard deviations and zero-order correlations among the study variables are presented in Table 1 . On average, parents and adolescents engaged in ~two COVID-19

conversations per week during the early stages of the pandemic and, on average, adolescents reported relatively low levels of perceived stress. Adolescents reported high levels of pandemicspecific empathic concern at both time points, but no significant changes in empathic concern were found across time, t(138) = .63, p = .49. Adolescents reported engaging in COVID-19

HPBs often, but the frequency of HPBs adherence decreased across time, t(144) = 5.25, p < .001.

A MANOVA revealed a main effect of gender (Wilks' λ = 0.75, p < .001), but not of race/ethnicity (Wilks' λ = 0.96, p = .49) or their interaction (Wilks' λ = 0.98, p = .86) across the primary study variables. Relative to boys, girls expressed higher perceived stress (Mdiff = 0.47, p < .001), higher empathic concern at T1 (Mdiff = 0.36, p < .001) and T2 (Mdiff = 0.36, p < .001), greater increases in empathic concern across time (Mdiff = 0.46, p = .006) and more adherence to HPBs at T2 (Mdiff = 0.36, p = .01).

Standardized parameter estimates for the path model are shown in Table 2 . There was a significant interaction between parent-adolescent conversations and perceived stress in predicting changes in empathic concern from T1 to T2. As depicted in Figure 2 , parentadolescent conversations predicted smaller-than-expected changes in empathic concern over time at high levels of perceived stress (i.e., + [32] to graph the interactive cross-over between conversation frequency and perceived stress in predicting changes in empathic concern and adherence to COVID-19 HPBs. This technique identifies a region of significance at which the simple slope becomes statistically significant and shows confidence bands that specify the simple slope estimate [33] . Residualized gains in empathic concern predicted increased adherence to COVID-19 HPBs from T1 to T2. There was a significant direct effect of the interaction between the frequency of parent-adolescent conversations and adolescents' perceived stress in predicting adolescents' COVID-19 HPBs. As depicted in Figure 3 through decreased empathic concern, showing partial mediation. In other words, more frequent parent-adolescent conversations predicted increased HPBs across the first 9-months of the COVID-19 pandemic among adolescent who reported low levels of perceived stress, but decreased HPBs via reductions in empathic concern among adolescents who reported high stress at the start of the pandemic.

This study sheds new light on how family socialization processes and empathic concern in contexts of adolescent stress influenced adolescents' adherence to COVID-19 HPBs during the first year of the COVID-19 pandemic. The findings demonstrate that, depending on adolescents' level of perceived stress, frequent conversations about COVID-19 with one's parent could help or hinder changes in adolescents' empathic concern and adherence to HPBs across time. This study is one of the first to assess the influence of both inter-and intrapersonal factors on adolescents' COVID-19 HPBs using a longitudinal approach. Moreover, the current findings begin to clarify how and for whom parent-adolescent conversations about major stressors, such as the COVID-19 pandemic, may promote or undermine adolescent's positive health behaviors.

Conversations between parents and adolescents provide opportunities for the communication of values, norms, and rules [34] , which, in turn, influence behavior [35] . The current findings extend prior research by showing that the frequency of parent-adolescent conversations about COVID-19 played a specific role in socializing adolescents' pandemicrelated empathic concern and HPBs. Importantly, however, we found that adolescents' perceived stress qualified these relations. Specifically, among adolescents who reported relatively high levels of perceived stress, frequent conversations with their parents about COVID-19 decreased their empathic concern toward vulnerable others over time, which, in turn, was associated with decreases in their adherence to HPBs. If stressed adolescents were already psychologically overwhelmed, messages relayed by parents about COVID-19 (whether positive or negative) may have caused even more stress by focusing the adolescent's attention on the stressor, eventuating in an emotional blunting effect. In turn, this emotional blunting may have undermined adolescents' prosocial action (i.e., adhering to HPBs). This pattern is consistent with previous research showing that youth experiencing personal distress, particularly in contexts involving others' suffering, are more likely to flee or exhibit self-soothing behavior rather than prosocial behavior [36] .

It is important to note that engaging in frequent parent-adolescent conversations about COVID-19 when adolescents are experiencing high levels of stress may signal other features of the family context that could hinder HPBs. For example, it is possible that parents who engaged their adolescents in frequent conversations about COVID-19 despite their child's high levels of stress may not have recognized their child's distress due to their own feelings of anxiety or a broader insensitivity to their child's needs. This finding speaks to the need for efforts to promote sensitive parenting in the context of COVID-19 (and likely other major life stressors) so that parents can tailor their support and socialization efforts to the social-emotional needs of their adolescents.

Among adolescents who reported relatively low levels of perceived stress, higher frequencies of parent-adolescent conversations about COVID-19 increased adherence to HPBs.

Adolescents who were experiencing less stress may have had the necessary psychological resources to process and direct their attention to pandemic-related messages, which, in turn, may have allowed them to appreciate how their own actions might play a role in the pandemic.

Relatively frequent conversations about health restrictions and the socio-political repercussions of these restrictions may have prepared less-stressed adolescents for the effort that would be required to consistently and intensively combat the virus via adherence to HPBs. Further, because discussing stressful events in ways that facilitate the expression of emotions and experiences may supplement youths' regulatory capacities, it is possible that less-stressed adolescents may have engaged in more emotion-focused conversation about the various J o u r n a l P r e -p r o o f aspects-both positive and negative-of the pandemic. Parents' scaffolding of emotion-focused conversations might have encouraged adolescents to attend to their own emotional needs and the needs of others at the same time they may have alleviated adolescents' anxiety related to the pandemic [37] . More research is needed to understand if and how specific messages of varying content and valence may differentially affect adolescents' empathic concern and health behaviors.

This study has implications for parents, teachers, and public health officials who seek to encourage adolescents' HPBs in the context of the ongoing COVID-19 pandemic, and likely with regard to other health outcomes (e.g., obesity, substance use). Considering and addressing adolescents' stress before communicating with them about the pandemic may positively influence how much adolescents care about the effects of their behaviors on vulnerable others and the extent to which they help reduce the spread of the virus. Indeed, recent work has highlighted the negative impact of overexposure to COVID-19 in the media on youths' acute stress and depressive symptoms, suggesting that transferring messages about emotionally-salient topics may be beneficial only to the extent it does overwhelm adolescents and thwart their capacity to act on such messages [15] . It is important to note that, just as too much exposure can thwart positive development, too little communication with adolescents about stressful events may also negatively impact their mental health [see 38]. Thus, we encourage parents, teachers, and providers to adopt a sensitive approach to talking about the pandemic (and other major life stressors) with youth. Such conversations should be tailored to adolescents' socioemotional strengths and vulnerabilities in ways that optimize their capacity to attend to the stressor at hand, and act in a ways that protect and promote the health and wellbeing of themselves and others.

Further, our findings have implications for how public health officials may most effectively prompt adolescents' engagement in HPBs now and in the future. Consistent with Pfattheicher and colleagues' [14] findings, infusing parent-adolescent conversations about COVID-19 with messages that activate empathic concern for others, such as considering the perspectives of vulnerable individuals, may heighten adolescents' motivation to engage in HPBs over time.

Despite this study's strengths, there are some limitations worth noting. First, we only collected data from one sample of adolescents in the U.S., thus limiting the generalizability of these findings to adolescents from other national and international regions. Second, the limited waves of available data thwarted our ability to examine trajectories of change in HPBs. Third, we focused on one period of adolescence (middle adolescence) and thus additional research is needed to examine the potential for differential developmental effects of conversations on HPBs across adolescence. Fourth, the current focus on the frequency of COVID-19 conversations limited our ability to examine how specific conversation content (as well as valence and quality of content) may influence adolescents' empathic concern and HPBs. Researchers would benefit from gathering direct observations of parent-adolescent COVID-19-conversations to extend this research. Fifth, although aggregating across parent and adolescent informants limited the bias associated with any single perspective, future research would benefit from employing experience sampling methods to more objectively measure frequency of conversations across time. Finally, we had a relatively small sample and thus did not have the power to employ a more complex statistical analysis to assess change (i.e., a latent difference score). Note. Change in empathic concern is a standardized RGS. TI = time 1, T2 = time 2, HPBs = health protective behaviors. Empathic concern (T1) .45 *** .08 -4. Empathic concern (T2) .23 ** .08 .64 *** -5. Change in empathic concern -.05 .01 .00 .77 *** - 6. Adherence to HPBs (T1) .32 *** -.11 .38 *** .35 *** .14 - Note. Indirect effect is from interaction between parent-adolescent conversations and stress (X1*X2) to change in adherence to HPBs (Y) through change in empathic concern (M; see conceptual model in Figure 1 ). HPBs = Health protective behaviors, β = standardized regression coefficient, SE = standard error, CI = confidence interval. Change in empathic concern is a standardized residualized gain score. No p-values given for indirect effects, as indirect effects are known to be non-normal.

J o u r n a l P r e -p r o o f 

Stress and coping in the time of COVID-19: Pathways to resilience and recovery

Parenting in a time of COVID-19

Increased risk for family violence during the COVID-19 pandemic

Parent-child relationships during adolescence

The health belief model. Health Behavior and Health Education: Theory

Adherence to COVID-19 precautionary measures: Applying the health belief model and generalised social beliefs to a probability community sample

Thematic analysis of parent-child conversations about covid-19: "playing it safe

Prosocial development

Role-modeling and conversations about giving in the socialization of adolescent charitable giving and volunteering

An evidence review of face masks against COVID-19

Parenting styles or practices? parenting, sympathy, and prosocial behaviors among adolescents

The socialization of prosocial development

The emotional path to action: Empathy promotes physical distancing and wearing of face masks during the COVID-19 pandemic

The unfolding COVID-19 pandemic: A probability-based, nationally representative study of mental health in the United States

Depression and anxiety during the covid-19 pandemic in an urban, low-income public university sample

Reactions to COVID-19: Differential predictors of distress, avoidance, and disregard for social distancing

The protective role of parent resilience on mental health and the parent-child relationship during COVID-19

Physiological stress reactivity and episodic memory in children

Social neuroscience. The social neuroscience of empathy

Current population survey: Annual social and economic supplement

A new approach to integrating data from multiple informants in psychiatric assessment and research: Mixing and matching contexts and perspectives

Perceived stress in a probability sample of the United States

Using residualized change versus difference scores for longitudinal research

Mplus User's Guide. Eighth Edition. Muthén & Muthén

Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses

A test of missing completely at random for multivariate data with missing values

Applied multiple regression/correlation analysis for the behavioral sciences

Explanation of two anomalous results in statistical mediation analysis

Methods for integrating moderation and mediation: A general analytical framework using moderated path analysis

The Sage sourcebook of advanced data analysis methods for communication research

Tests of Certain Linear Hypotheses and Their Application to Some Educational Problems

Probing interactions in fixed and multilevel regression: Inferential and graphical techniques

Handbook of child psychology: Social, emotional, and personality development

A meta-analytical review of family communication patterns and their associations with information processing, behavioral, and psychosocial outcomes

Empathy as a "risky strength": A multilevel examination of empathy and risk for internalizing disorders

Family Narratives and the Development of Children's Emotional Well-Being

The authors thank the adolescents and parents who participated in this study, and the research assistants and graduate students at the Child Narratives Lab and the Adversity and Adaptation Lab who helped with data collection and data processing. The project described was supported by the UC Irvine COVID-19 Basic, Translational and Clinical Research Funding Opportunity.