Edited by: Joanna Sokolowska, University of Social Sciences and Humanities, Poland
Reviewed by: Lara Nikola Wolfers, Leibniz-Institut für Wissensmedien (IWM), Germany; Lisa Perks, Merrimack College, United States
This article was submitted to Emotion Science, a section of the journal Frontiers in Psychology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
In spring 2020, COVID-19 and the ensuing social distancing and stay-at-home orders instigated abrupt changes to employment and educational infrastructure, leading to uncertainty, concern, and stress among United States college students. The media consumption patterns of this and other social groups across the globe were affected, with early evidence suggesting viewers were seeking both pandemic-themed media and reassuring, familiar content. A general increase in media consumption, and increased consumption of specific types of content, may have been due to media use for coping strategies. This paper examines the relationship between the stress and anxiety of university students and their strategic use of media for coping during initial social distancing periods in March-April 2020 using data from a cross-sectional survey. We examine links between specific types of media use with psychological well-being concepts, and examine the moderating roles of traits (hope, optimism, and resilience) as buffers against negative relationships between stress and anxiety and psychological well-being. Our findings indicate that stress was linked to more hedonic and less eudaimonic media use, as well as more avoidant and escapist media-based coping. Anxiety, on the other hand, was linked to more media use in general, specifically more eudaimonic media use and a full range of media-based coping strategies. In turn, escapist media was linked to negative affect, while reframing media and eudaimonic media were linked to positive affect. Avoidant coping was tied to poorer mental health, and humor coping was tied to better mental health. Hedonic and need-satisfying media use were linked to more flourishing. Hope, optimism, and resilience were all predictive of media use, with the latter two traits moderating responses to stress and anxiety. The findings give a nuanced portrait of college students’ media use during a pandemic-induced shutdown, showing that media use is closely intertwined with well-being in both adaptive and maladaptive patterns.
In the spring of 2020, COVID-19 concerns drove American universities to cancel face-to-face classes, which resulted in millions of residential college students leaving campus mid-semester with no plan to return (
During this same period, video streaming increased sharply, especially during daytime hours (
Psychological stress is many-faceted, but usually stems from a disconnect (or disequilibrium) between one’s available resources and the demands they face (
Along with many others, one population suddenly facing unexpected stress due to COVID-19 countermeasures were the suddenly relocated (at least, moved online) United States university students. In March 2020, many American residential universities moved classes online, sent students away from residential facilities, and shut down or minimized capacity of residence halls to protect students, employees, and staff against COVID-19 (
In terms of problem-focused coping, specifically, adolescents who reported stress in specific domains (e.g., parents, peers, appearance) preferred to watch talk shows on these topics (
Emotion-focused forms of coping via media may be particularly relevant in the context of the COVID-19 crisis. The coping literature suggests that emotion-focused coping strategies are particularly effective and functional if the individual has low control over the situation and stressor, making problem-focused coping difficult or even impossible (
During social distancing, students were isolated from their friends and routine, as well as concerned about changes in the local pandemic status, and therefore we might expect that stress and anxiety would be heightened during social distancing, and that crucial coping resources, such as the availability of social support, will be largely absent or impaired. As such, if users are turning to media to cope with negative feelings, we may see overall increases in media use. At the same time, media can be used as part of various and even competing coping strategies: for some users, media may play a role in problem-focused coping, where they turn to media to keep monitoring the local situation or to learn about other pandemics. On the other hand, users may feel a need to distance themselves from the current situation, and focus instead on the emotional benefits of media. The first aim of the present study was to examine the relationships between stress and anxiety resulting from social distancing and the use of media exposure within a variety of well-established coping strategies (
One central mechanism that connects media use to psychological well-being is the mood-altering effects of media exposure. A large number of studies in the tradition of mood management theory (
Beyond mood valence, entertainment research distinguishes content based on hedonic versus eudaimonic motivations for media consumption (cf.
Eudaimonic motivations, on the other hand, are concerned with existential questions of purpose in life, meaning, or moral values. These motives often lead to more contemplative and emotionally complex media selections and experiences, and are often associated with exposure to somber or poignant media content. Previous research suggests that such forms of eudaimonic entertainment may provide important role models for dealing with critical life events (
Another avenue of media research demonstrates that entertainment media can satisfy intrinsic needs. Intrinsic needs are universal human drives which benefit individuals, such as being competent, having autonomy over one’s own life, and feeling a deep sense of connection in personal relationships (self-determination theory;
In sum, then, we predict that (H1) stress and (H2) anxiety will have positive associations with (a) quantity of media exposure, (b) using media to cope, (c) hedonic media use, (d) eudaimonic media use, and (e) intrinsically satisfying media use.
While stress and coping may shift patterns of media consumption and gratifications, we also sought to explore how media use may be influencing users’ self-reports of psychological well-being more generally. All forms of media use discussed above (and addressed in H1 and H2) have been linked to psychological well-being in previous research (
The present study examines the association of media use with three different indicators of psychological well-being: the presence of positive affect and absence of negative affect as an indicator of subjective well-being (
We clearly are not suggesting that media use fully mediates the relationship connecting stress and anxiety with well-being. On the contrary, stress and anxiety are important factors in psychological well-being more generally. However, we do suggest that media use (and particularly coping-based, emotionally motivated, and need-satisfying media consumption) will influence this relationship, such that media use which serves to support coping and need satisfaction will reduce the effect of stressors on well-being, as follows:
Stress and anxiety will have negative total and direct effects on affect, mental health, and flourishing, but (H8) positive mediation effects via (a) quantity of media exposure, (b) using media to cope, (c) hedonic media use, (d) eudaimonic media use, and (e) intrinsically satisfying media use will partially suppress the negative influences of stress and anxiety on (i) affect, (ii) mental health, and (iii) flourishing.
Numerous protective factors, however, may alter both the initial stress reaction as well as the ways in which entertaining media are used as coping tools. In the psychological literature, such factors are frequently discussed in the context of resilience. The theoretical concept of resilience refers to positive adaptation after adversity (
Two theoretical perspectives differentiate resilience as either a relatively stable trait or a dynamic process (
One key resiliency factor is the presence and cultivation of positive emotions and affect (
In the context of the present study, both general trait resilience as well as more specific resilience factors such as optimism and hope appear relevant for multiple reasons. First, previous research suggests a direct positive influence of trait resilience and protective and promotive resiliency factors on adaptation to stress and psychological well-being (
While the empirical evidence on the interplay of media use and resilience factors is very limited, a number of theoretical mechanisms connect both concepts (
To explore the role of trait resiliency factors in the interplay of stress, media use, and well-being, we pose the following research questions: Do (RQ1) trait optimism, (RQ2) trait hope, and (RQ3) trait resilience have main effects on stress, anxiety, media use, and affect, mental health, and flourishing, and do they moderate hypothesized effects? Our conceptual model appears in
Conceptual model. Stress and anxiety are tested as two distinct independent variables. H3-7 predict different forms of media use (media exposure, using media to cope, hedonic media use, eudaimonic media use, and intrinsically satisfying media use) will be associated with three well-being measures (affect, mental health, and flourishing). H8 predicts media use will mediate the relationship between stress/anxiety and well-being. RQ1-3 examine the main effects and potential moderation role of resiliency trait factors hope, optimism, and resilience.
To test the hypotheses and research questions, students at two American universities that canceled face-to-face instruction due to COVID-19 were surveyed. Both universities canceled face-to-face instruction the week of March 9, 2020, and students completed a cross-sectional survey between March 23, 2020 and April 17, 2020. The study preregistration, data, and materials are available at
An initial 459 students accessed the questionnaire. Screening criteria removed 29 incomplete cases as well as 5 cases that reported more than 24 h per day on a single media activity. This left
All non-trait items were framed with instructions referring to “your feelings and thoughts since social distancing began.” Descriptives for all measures are reported in
Descriptive statistics for study variables.
Variable | M | SD | Min | Max | α | Skew | Kurtosis |
Stress | 3.099 | 0.480 | 1.43 | 4.86 | .797 | 0.110 | 0.352 |
Anxiety | 2.339 | 0.812 | 1.00 | 4.00 | .905 | 0.266 | −0.723 |
Media Exposure | 21.416 | 11.562 | 0.00 | 120.00* | − | 2.563 | 14.618 |
Problem-Focus Media Coping | 2.087 | 0.792 | 1.00 | 4.00 | .822 | 0.418 | −0.648 |
Avoidant Media Coping | 1.681 | 0.738 | 1.00 | 4.00 | .823 | 1.069 | 0.342 |
Escapist Media Coping | 2.649 | 0.772 | 1.00 | 4.00 | .827 | −0.143 | −0.829 |
Reframing Media Coping | 2.547 | 0.853 | 1.00 | 4.00 | .765 | −0.053 | −0.731 |
Humor Media Coping | 2.228 | 0.885 | 1.00 | 4.00 | .670 | 0.281 | −0.857 |
Hedonic Media | 5.411 | 1.080 | 1.00 | 7.00 | .910 | −0.817 | 0.773 |
Eudaimonic Media | 4.333 | 1.359 | 1.00 | 7.00 | .881 | −0.225 | −0.439 |
Media Need Satisfaction | 4.413 | 1.249 | 1.00 | 7.00 | .930 | −0.568 | 0.246 |
Optimism | 3.322 | 0.709 | 1.17 | 5.00 | .758 | −0.025 | −0.189 |
Hope | 3.058 | 0.453 | 1.38 | 4.00 | .842 | −0.223 | 0.421 |
Resilience | 3.226 | 0.754 | 1.00 | 5.00 | .805 | −0.186 | −0.008 |
Affect | 3.328 | 0.703 | 1.00 | 5.00 | .897 | −0.331 | −0.149 |
Mental Health | 3.155 | 0.724 | 1.00 | 5.00 | .810 | −0.234 | −0.136 |
Flourishing | 5.314 | 0.992 | 2.00 | 7.00 | .913 | −0.361 | −0.292 |
The two independent variables were stress and anxiety in the context of social distancing. Stress was measured with the 14-item Perceived Stress Scale (
Media variables included media exposure
Coping via media was measured with the 28-item Brief COPE (
Exploratory factor analysis for media coping.
Item | Original dimension | Media dimension | Factor loading |
5. I’ve been using media to get emotional support from others. | Emotional Support | Problem-Focus | .471 |
27. I’ve been using media as a kind of mediation or prayer. | Religion | Problem-Focus | .359 |
20. I use media to help accept the reality of the fact that this has happened. | Acceptance | Problem-Focus | .357 |
22. I try to find comfort, meaning, or spirituality through media. | Religion | Problem-Focus | .345 |
2. I’ve been using media to do something about the situation I’m in. | Active Coping | Problem-Focus | .315 |
26. I use media to blame myself for things that happened. | Self-Blame | Avoidant | .444 |
19. I’ve been using media to think about the situation less. | Self-Distraction | Escapist | .488 |
7. I’ve been using media to try to make the situation better. | Active Coping | Escapist | .409 |
24. I’ve used media as I’m learning to live with the situation. | Acceptance | Escapist | .395 |
15. I’ve been getting comfort and understanding from media. | Emotional Support | Reframing | .295 |
21. I express my negative feelings through media use. | Venting | Humor | .390 |
13. I use media to criticize myself. | Self-Blame | Humor | .361 |
Frequencies of consuming media perceived to meet hedonic motivations (6 items; e.g., “Lets me have fun”) and eudaimonic motivations (6 items; e.g., “Makes me more reflective”) were measured on a 7-point scale,
With regard to moderating traits, three established scales were administered. Trait optimism was measured with the 6-item Life Orientation Scale Revised (
Finally, affect, mental health, and flourishing outcomes were assessed with a set of established measures. Affect was measured with the 12-item Scale of Positive and Negative Experience (SPANE;
Descriptive statistics and correlations are presented as a preliminary analysis. To test initial hypotheses, regression analyses tested the effects of the following variables in three blocks: (a) demographics, (b) trait moderators, and (c) state stress and anxiety, on five dependent variables of media use (media exposure, coping, hedonic, eudaimonic, and intrinsically satisfying). Given the multidimensional nature of media-based coping from our EFA, effects were examined for each of the five dimensions of media coping separately. A fourth block was used to enter interaction terms between trait moderators (one trait at a time) and stress and anxiety (labeled as block 4a/b/c in
Correlations among study variables.
Variable | (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) | (12) | (13) | (14) | (15) | (16) |
(1) Stress | ||||||||||||||||
(2) Anxiety | .619*** | |||||||||||||||
(3) Media Exp. | .090 | .185*** | ||||||||||||||
(4) Prob. Cope | .145** | .254*** | .073 | |||||||||||||
(5) Avoid. Cope | .328*** | .402*** | .132** | .464*** | ||||||||||||
(6) Escap. Cope | .398*** | .536*** | .200*** | .440*** | .474*** | |||||||||||
(7) Refram. Cope | .087 | .221*** | .114* | .520*** | .273*** | .480*** | ||||||||||
(8) Humor Cope | .144** | .240*** | .121* | .379*** | .370*** | .444*** | .395*** | |||||||||
(9) Hedonic | .114* | .193*** | .043 | .057 | .006 | .360*** | .265*** | .245*** | ||||||||
(10) Eudaimonic | −.140** | .033 | .094 | .388*** | .142** | .166*** | .303*** | .176*** | .235*** | |||||||
(11) Need Satisfaction | −.021 | .068 | .107* | .393*** | .191*** | .307*** | .319*** | .271*** | .312*** | .507*** | ||||||
(12) Optimism | −.389*** | −.273*** | −.045 | −.075 | −.211*** | −.105* | .089 | −.051 | .107* | .101* | .060 | |||||
(13) Hope | −.220*** | .039 | .149** | .052 | −.091 | .099* | .195*** | .027 | .324*** | .220*** | .245*** | .456*** | ||||
(14) Resilience | −.429*** | −.331*** | −.043 | −.057 | −.142** | −.226*** | .045 | −.102* | −.016 | .151** | .007 | .490*** | .348*** | |||
(15) Affect | −.639*** | −.633*** | −.111* | −.094 | −.334*** | −.389*** | .021 | −.128** | −.009 | .201*** | .121* | .362*** | .235*** | .378*** | ||
(16) MentalHealth | −.722*** | −.695*** | −.139** | −.165*** | −.427*** | −.426*** | −.065 | −.104* | −.058 | .118* | .027 | .423*** | .210*** | .414*** | .758*** | |
(17) Flourishing | −.326*** | −.186*** | .073 | .017 | −.121* | .003 | .184*** | .013 | .318*** | .308*** | .344*** | .443*** | .548*** | .322*** | .394*** | .355*** |
Associations of stress and anxiety with media use variables.
Predictors | Media exposure | Hedonic media | Eudaimonic media | Media need satisfaction |
β | β | β | β | |
Block 1: Demographics | Δ |
Δ |
Δ |
Δ |
Woman | .078 | .036 | −.069 | −.041 |
Latinx | −.010 | .003 | −.029 | −.026 |
White | −.192*** | .018 | −.087 | −.043 |
Age | −.012 | −.027 | .049 | −.066 |
Education | −.028 | .048 | −.049 | −.016 |
Block 2: Traits | Δ |
Δ |
Δ |
Δ |
Optimism | −.109 | .023 | −.031 | −.008 |
Hope | .229*** | .378*** | .215*** | .300*** |
Resilience | −.059 | −.150** | .081 | −.100 |
Block 3: IVs | Δ |
Δ |
Δ |
Δ |
Stress | −.016 | .129* | −.167* | −.012 |
Anxiety | .160* | .093 | .159* | .046 |
Main Effects Model |
.103 | .160 | .090 | .088 |
Block 4a: Moderation | Δ |
Δ |
Δ |
Δ |
Optimism × Stress | .127* | .025 | .016 | −.079 |
Optimism × Anxiety | −.188** | −.074 | .030 | .103 |
Block 4b: Moderation | Δ |
Δ |
Δ |
Δ |
Hope × Stress | −.018 | .002 | −.008 | −.008 |
Hope × Anxiety | .074 | .002 | .055 | .012 |
Block 4c: Moderation | Δ |
Δ |
Δ |
Δ |
Resilience × Stress | .027 | .043 | .004 | −.021 |
Resilience × Anxiety | −.051 | −.056 | .073 | .051 |
Associations of stress and anxiety with media coping.
Predictors | Problem-focus coping | Avoidant coping | Escapist coping | Reframing coping | Humor coping |
β | β | β | β | β | |
Block 1: Demographics | Δ |
Δ |
Δ |
Δ |
Δ |
Woman | −.080 | .029 | .173*** | .135** | −.030 |
Latinx | .016 | −.026 | −.084 | −.026 | .005 |
White | −.058 | −.104* | −.024 | .020 | −.017 |
Age | −.064 | .000 | −.055 | .059 | .004 |
Education | .027 | −.015 | .035 | −.091 | .032 |
Block 2: Traits | Δ |
Δ |
Δ |
Δ |
Δ |
Optimism | −.085 | −.196*** | −.070 | .009 | −.026 |
Hope | .134* | .020 | .231*** | .192*** | .092 |
Resilience | −.070 | −.052 | −.255*** | −.014 | −.128* |
Block 3: IVs | Δ |
Δ |
Δ |
Δ |
Δ |
Stress | .040 | .128* | .133* | −.003 | .023 |
Anxiety | .263*** | .364*** | .424*** | .233*** | .238*** |
Main Effects Model |
.095 | .211 | .320 | .102 | .069 |
Block 4a: Moderation | Δ |
Δ |
Δ |
Δ |
Δ |
Optimism × Stress | .152** | .119* | .051 | .120* | .128* |
Optimism × Anxiety | −.039 | −.043 | .010 | −.033 | −.036 |
Block 4b: Moderation | Δ |
Δ |
Δ |
Δ |
Δ |
Hope × Stress | .067 | −.004 | .026 | .026 | .018 |
Hope × Anxiety | −.020 | .067 | −.013 | −.004 | .062 |
Block 4c: Moderation | Δ |
Δ |
Δ |
Δ |
Δ |
Resilience × Stress | .136* | .157** | .051 | .115* | .124* |
Resilience × Anxiety | −.068 | −.051 | .045 | −.067 | −.001 |
Next, regression analyses tested the effects of the following variables in four blocks: (a) demographics, (b) trait moderators, (c) state stress and anxiety, and (d) media use on the dependent variables of affect, mental health, and flourishing. An additional fifth block was used to enter interaction terms between trait moderators (one trait at a time) and, stress, anxiety, and media use (labeled as block 5a/b/c in
Associations of media use with affect, mental health, and flourishing.
Predictors | Affect | Mental health | Flourishing | |
β | β | β | ||
Block 1: Demographics | Δ |
Δ |
Δ |
|
Woman | −.170*** | −.235*** | .014 | |
Latinx | .090 | .053 | .037 | |
White | −.029 | −.003 | −.029 | |
Age | .107 | .129* | .031 | |
Education | −.105 | −.138* | −.103 | |
Block 2: Traits | Δ |
Δ |
Δ |
|
Optimism | .216*** | .307*** | .230*** | |
Hope | .050 | −.017 | .429*** | |
Resilience | .237*** | .238*** | .068 | |
Block 3: Stress/Anxiety | Δ |
Δ |
Δ |
|
Stress | −.320*** | −.385*** | −.120* | |
Anxiety | −.422*** | −.418*** | −.092 | |
Block 4: Media Use | Δ |
Δ |
Δ |
|
Media Exposure | −.033 | −.035 | .021 | |
Problem−Focus Coping | .005 | .007 | −.072 | |
Avoidant Coping | −.057 | −.157*** | .020 | |
Escapist Coping | −.162*** | −.078 | −.041 | |
Reframing Coping | .136** | .046 | .064 | |
Humor Coping | .002 | .106** | −.031 | |
Hedonic Media | .037 | .003 | .157*** | |
Eudaimonic Media | .094* | .048 | .094 | |
Need Satisfaction | .076 | .007 | .180*** | |
Main Effects Model |
.580 | .678 | .476 | |
Block 5a: Moderation | Δ |
Δ |
Δ |
|
Optimism × Reframing | −.100* | −.062 | −.061 | |
Block 5b: Moderation | Δ |
Δ |
Δ |
|
Hope × Anxiety | −.085 | −.050 | −.145** | |
Hope × Humor | −.005 | −.018 | .124** | |
Block 5c: Moderation | Δ |
Δ |
Δ |
|
Resilience | − | − | − |
The mediation hypotheses were tested with the PROCESS macro (
Media use partially mediates influence of stress and anxiety on affect. Note. Parallel mediation of media motives on affect. Gender, age, ethnicity, race, level of education, and traits (hope, optimism, and resilience) are used as covariates in PROCESS Model 4 using 10,000 bootstrap samples. Path coefficients reported are unstandardized,
Media use partially mediates influence of stress and anxiety on mental health.
Media use partially mediates influence of stress and anxiety on flourishing. Note. Parallel mediation of media motives on flourishing. Gender, age, ethnicity, race, level of education, and traits (hope, optimism, and resilience) are used as covariates in PROCESS Model 4 using 10,000 bootstrap samples. Path coefficients reported are unstandardized,
Finally, trait moderators from the earlier regression analyses were tested as moderators of the mediation effects using PROCESS.
Descriptive statistics appear in
In
Media exposure was not linked to differences in affect, mental health, or flourishing. Examining the effects of media coping dimensions, escapist coping was associated with less positive affect, and reframing coping with more positive affect. Avoidant coping was associated with lower mental health scores, but humor coping with higher mental health scores. This suggests that different media-related coping strategies were associated with different indicators of well-being, potentially suggesting adaptive or maladaptive functions.
Eudaimonic media use was connected to more positive affect, and both hedonic media and media need-satisfaction were associated with higher levels of flourishing.
Mediation tests (H8) found a mix of positive indirect effects, which were hypothesized to suppress the negative effects of stress and anxiety on well-being outcomes. Specifically, reframing coping suppressed the effect of anxiety on affect, β = .032,
We also found some negative indirect effects, suggesting that some (maladaptive) forms of media use may be associated with negative effects on well-being. Specifically, escapist coping mediated the effects of stress on affect, β = −.022,
After accounting for demographics, traits demonstrated some influence on both media use and well-being, supporting our research questions, as shown in Block 2 of
To examine the research questions’ interaction effects, the regression models reported in
Optimism positively moderated the effect of stress on media exposure, but negatively moderated the effect of anxiety on media exposure (
Hope negatively moderated the effect of anxiety on flourishing, and hope positively moderated the effect of humor coping on flourishing (
Resilience positively moderated the effect of stress on problem-focused, avoidant, reframing, and humor coping (
Finally, we considered how traits might moderate the observed mediation effects. The significant instances of moderated mediation are probed and presented in
Summary of hypothesis testing.
Prediction | Description | Supported | Details | |
H1a | Stress → Media Exposure | No | ||
H1b | Stress → Media Coping | Partial | Yes for avoidant and escapist dimensions | |
H1c | Stress → Hedonic | Yes | ||
H1d | Stress → Eudaimonic | No | Effect in opposite direction | |
H1e | Stress → Need Satisfaction | No | ||
H2a | Anxiety → Media Exposure | Yes | ||
H2b | Anxiety → Media Coping | Yes | ||
H2c | Anxiety → Hedonic | No | ||
H2d | Anxiety → Eudaimonic | Yes | ||
H2e | Anxiety → Need Satisfaction | No | ||
H3a | Media Exposure → Affect | No | ||
H3b | Media Exposure → Mental Health | No | ||
H3c | Media Exposure → Flourishing | No | ||
H4a | Media Coping → Affect | Partial | Yes for reframing; opposite effect for escapist | |
H4b | Media Coping → Mental Health | Partial | Yes for humor; opposite effect for avoidant | |
H4c | Media Coping → Flourishing | No | ||
H5a | Hedonic → Affect | No | ||
H5b | Hedonic → Mental Health | No | ||
H5c | Hedonic → Flourishing | Yes | ||
H6a | Eudaimonic → Affect | Yes | ||
H6b | Eudaimonic → Mental Health | No | ||
H6c | Eudaimonic → Flourishing | No | ||
H7a | Need Satisfaction → Affect | No | ||
H7b | Need Satisfaction → Mental Health | No | ||
H7c | Need Satisfaction → Flourishing | Yes | ||
H8a-i | Stress/Anxiety → Media Exp. → Affect | No | ||
H8b-i | Stress/Anxiety → Media Coping → Affect | Partial | Yes for anxiety via reframing. Opposite effect for stress and anxiety via escapist coping. | |
H8c-i | Stress/Anxiety → Hedonic → Affect | No | ||
H8d-i | Stress/Anxiety → Eudaimonic → Affect | Partial | Yes for anxiety. Opposite effect for stress. | |
H8e-i | Stress/Anxiety → Need Satisf. → Affect | No | ||
H8a-ii | Stress/Anxiety → Media Exp. → Mental Health | No | ||
H8b-ii | Stress/Anxiety → Media Coping → Mental Health | Partial | Yes for anxiety via humor. Opposite effects for stress and anxiety via avoidant coping. | |
H8c-ii | Stress/Anxiety → Hedonic → Mental Health | No | ||
H8d-ii | Stress/Anxiety → Eudaimonic → Mental Health | No | ||
H8e-ii | Stress/Anxiety → Need Satisf. → Mental Health | No | ||
H8a-iii | Stress/Anxiety → Media Exp. → Flourishing | No | ||
H8b-iii | Stress/Anxiety → Media Coping → Flourishing | No | ||
H8c-iii | Stress/Anxiety → Hedonic → Flourishing | No | ||
H8d-iii | Stress/Anxiety → Eudaimonic → Flourishing | Partial | Yes for anxiety. | |
H8e-iii | Stress/Anxiety → Need Satisf. → Flourishing | No | ||
RQ1 | Optimism → or X | |||
RQ2 | Hope → or X | |||
RQ3 | Resilience → or X |
Significant moderated mediation models.
Moderator level | Mediation effect | Indirect effect | SE | 95% CI |
Optimism | ||||
+1 SD | Stress→Reframing→Affect | 0.016 | 0.018 | [−0.018, 0.054] |
Mean | Stress→Reframing→Affect | −0.008 | 0.016 | [−0.045, 0.212] |
−1 SD | Stress→Reframing→Affect | −0.022 | 0.020 | [−0.068, 0.011] |
Optimism | ||||
+1 SD | Stress→Avoidant→Mental Health | |||
Mean | Stress→Avoidant→Mental Health | −0.023 | 0.018 | [−0.060, 0.010] |
−1 SD | Stress→Avoidant→Mental Health | −0.007 | 0.020 | [−0.047, 0.035] |
Optimism | ||||
+1 SD | Stress→Humor→Mental Health | 0.018 | 0.015 | [−0.007, 0.051] |
Mean | Stress→Humor→Mental Health | −0.003 | 0.013 | [−0.032, 0.023] |
−1 SD | Stress→Humor→Mental Health | −0.015 | 0.016 | [−0.054, 0.013] |
Resilience | ||||
+1 SD | Stress→Escapist→Affect | |||
Mean | Stress→Escapist→Affect | −0.027 | 0.018 | [−0.068, 0.001] |
−1 SD | Stress→Escapist→Affect | −0.011 | 0.017 | [−0.048, 0.022] |
Resilience | ||||
+1 SD | Anxiety→Escapist→Affect | |||
Mean | Anxiety→Escapist→Affect | |||
−1 SD | Anxiety→Escapist→Affect | |||
Resilience | ||||
+1 SD | Stress→Avoidant→Mental Health | |||
Mean | Stress→Avoidant→Mental Health | −0.022 | 0.016 | [−0.055, 0.006] |
−1 SD | Stress→Avoidant→Mental Health | 0.004 | 0.018 | [−0.031, 0.040] |
Resilience | ||||
+1 SD | Stress→Humor→Mental Health | 0.020 | 0.015 | [−0.005, 0.055] |
Mean | Stress→Humor→Mental Health | −0.002 | 0.012 | [−0.027, 0.023] |
−1 SD | Stress→Humor→Mental Health | −0.019 | 0.015 | [−0.055, 0.006] |
Resilience | ||||
+1 SD | Anxiety→Humor→Mental Health | |||
Mean | Anxiety→Humor→Mental Health | |||
−1 SD | Anxiety→Humor→Mental Health | 0.013 | 0.009 | [−0.002, 0.034] |
Hope | ||||
+1 SD | Anxiety→Media Exp.→Flourishing | −0.011 | 0.015 | [−0.049, 0.012] |
Mean | Anxiety→Media Exp.→Flourishing | 0.007 | 0.012 | [−0.014, 0.033] |
−1 SD | Anxiety→Media Exp.→Flourishing | 0.020 | 0.016 | [−0.005, 0.058] |
Resilience moderated the effect of stress on affect via escapist coping,
Resilience also moderated the indirect effect of stress on mental health via avoidant coping,
There was less evidence that traits interacted with media use to influence psychological well-being outcomes in the back half of the model. Neither trait optimism nor resilience moderated effects of media use on affect, mental health, or flourishing. Trait hope did moderate the influence of media exposure on flourishing,
In this study, we examined how stress and anxiety during a global pandemic—involving shutdowns and social distancing—related to different patterns of media use among university students, and how that media use was linked to affect, mental health, and flourishing. A survey of students at two American universities, conducted in the immediate weeks after face-to-face study and work were suspended, revealed that stress and anxiety were related to various patterns of media use and in particular a variety of coping strategies using media. In general, we find that students reporting heightened stress and anxiety reported different media-based coping styles, and these were associated with differential relationships with our measures of well-being. Prior literature on media use as a tool for coping tends to paint media use as a monolithic, and often problematic, coping behavior (e.g.,
Generally, results suggest that acute stress and anxiety resulting from the COVID-19 situation were associated with an increased tendency to use media as a coping tool, and some (but not all) media coping strategies were associated with positive affect, positive mental health, and flourishing. These results underscore the relevance of media use for coping during the pandemic, and the potential importance of media use as a psychological resource in times of crisis. Further, findings suggest trait resilience, hope, and optimism interact to influence these effects, and that stress and anxiety were both associated with adaptive
First, we would note that reports of stress and anxiety were very present in our sample, and they were, as predicted, negatively associated with psychological well-being indicators of positive affect, mental health, and (in the case of stress) flourishing. These results underscore the need to understand how students coped with these negative psychological states given the limited physical and social resources available to them during social distancing. The particularities of stress and anxiety provoked by COVID-19 and the associated stay-at-home orders resulted in clear patterns of media use for coping with negative emotions.
Yet, stress and anxiety were differentially associated with unique patterns of media use, including both the media-based coping strategies employed and the entertainment outcomes experienced. Stress was associated with more hedonic media use and less eudaimonic media use than anxiety. Stress was also associated with avoidant and escapist coping via media (but less than anxiety). These results are in line with escapist theories of media use (e.g.,
Students experiencing high anxiety, on the other hand, were more likely to report higher overall media exposure, as well as more eudaimonic media use. This appears to be a more adaptive form of media coping, as eudaimonic media was associated overall with more positive affect. Additionally, anxiety provoked multiple types of coping strategies, showing medium-sized positive associations with all five forms of media coping which emerged in our analysis. Although, like stress, anxiety was associated with escapist and avoidant coping, anxious individuals also used media for problem-focused coping as well as to reframe the current situation, and to provide humor and insight. These latter forms of coping via media are of particular interest as they were positively related to our psychological well-being outcomes.
These different patterns of media use seem to suggest that media exposure is used differently in response to the psychological states of stress and anxiety. While students reporting stress and students reporting anxiety both reported using media to cope in short-term ways, such as escapism, anxious individuals were far more likely to report adaptive forms of media coping, such as problem-focused media use. These differences may be due to the ways in which stress and anxiety differ, particularly in terms of duration of the experience. Whereas stress refers to more ephemeral perceptions of situational threat (
The fact that anxious individuals reported problem-focused coping played a role in their media use corresponds with their preference for eudaimonic entertainment. Eudaimonic content, in contrast to hedonic content, frequently provides role models for positive adaptation to critical life events, rather than short-term mood enhancement (e.g.,
The mediation findings emphasize the role of diverse media-based coping strategies in the relationships between stress, anxiety, and psychological well-being. Both reframing and humor coping suppressed the effect of anxiety on negative well-being outcomes, specifically affect and mental health. On the other hand, escapist and avoidant coping styles had negative indirect effects of stress and anxiety on affect and mental health. These findings suggest that differentiating media-based coping styles has the potential to explicate the diverse outcomes associated with media use in times of distress – and potentially address the underlying complexity which drives the conflicting findings associating media use and well-being in other literature. Previous contradictory findings on the role of media use as a coping mechanism may be due to different coping strategies used by the individuals experiencing negative mood states. These findings emphasize the need for future work to further explore the boundary conditions and individual predictors of functional versus detrimental forms of media use for stress coping.
The present study further reveals the important role of trait resiliency factors in individual responses to stress, and the role of media use in the stress-coping context. First, our results replicate the findings of previous research on the beneficial effects of psychological resilience: all three resiliency factors showed negative zero-order correlations with stress, and optimism and resilience also showed negative zero-order correlations with anxiety. Furthermore, all three trait resiliency variables positively predicted all three psychological well-being variables assessed in the present study. In sum, this suggests that individuals high in trait resilience, hope, and optimism were less negatively affected by the COVID-19 related social distancing measures, and more successfully upheld psychological well-being in the face of adversity.
In addition to this general buffer effect, the three trait variables also significantly shaped the way that individuals used media in the coping process. Interestingly, while both optimism and trait resilience were negative predictors of media use for coping, hope showed positive associations with three out of the five media-related coping strategies. The negative associations found between optimism and avoidant coping, and also resilience with escapist and humor coping (
Trait resiliency factors also moderated the relationship between stress, anxiety, and many of the media use variables addressed in the present study. Overall, optimism and trait resiliency intensified the relationship between stress and media use for coping. Optimism also moderated the effect of stress on media exposure. This reveals an interesting pattern: As discussed above, optimism and trait resilience showed negative main effects on media use for coping, presumably because individuals high in these traits experienced less stress and anxiety and thus had a lower need for coping. However, when individuals did experience high levels of stress
Finally, the three trait variables also moderated some of the relationships of media use with psychological well-being as well as some of the indirect effect of stress and anxiety on psychological well-being via media use. Resilience factors were generally less likely to moderate effects of media on well-being than they were to moderate effects of stress and anxiety on media use. Pessimists saw helpful effects of their reframing coping on their affective states. Those with high hope experienced less flourishing when anxious, and those with low amounts of hope experienced less flourishing in response to humor.
The moderated mediation effects found for trait optimism and resilience showed mixed patterns, mostly driven by those with lower levels of the resiliency factors. Under high levels of optimism or resilience, stress and anxiety were more likely to lead to avoidant and escapist media use which was harmful for well-being. However, in contrast to that maladaptive coping, those same optimistic or resilient individuals were also more likely to find adaptive coping through humor. Trait variables increased the likelihood for both adaptive and maladaptive media-related coping attempts as a reaction to stress and anxiety, and thus increased both the positive and negative indirect effects of stress on psychological well-being via media coping.
Overall, these results demonstrate that media use and other coping resources, such as the protective and promotive traits addressed in this study, show complex interactions in the context of stress and anxiety, emphasizing the need for future research to explore the boundary conditions of beneficial media effects in response to negative psychological states more systematically. Furthermore, the direction of the relationship between media use and resiliency factors remains an open question. In the present study, resiliency traits were treated as predictors of media use and the resulting relationships with psychological well-being outcomes. However, other research suggests that media use may also have long-term effects on resilience and facilitate or impair the development of psychological traits that facilitate positive adaptation to adversity (
First, we note that the findings presented here are limited by the use of a cross-sectional survey design. Although our theoretically grounded model conceptualizes psychological well-being variables (affect, mental health, and flourishing) as outcomes of media use, it is likely that pre-existing levels of psychological well-being impact media use (cf.
In regard to our measures, a recent study (
Finally, we would note that some effect sizes in the study were small. We would hesitate to describe small effect sizes as a limitation, as the effect sizes may reflect the true parameter in the population, particularly when dealing with distal effects such as those of trait variables on state appraisals. That said, we would caution overinterpretation of our results where the dataset values are close to zero, without subsequent replication of these findings with a larger sample. Similarly, we would caution that including multiple testing of mediators and moderators in one study may have led to alpha error inflation. Again, future work to replicate these findings is needed, particularly to lend robust estimation to our model parameters. A separate point with regard to effect sizes is the extent to which these effects are practically consequential. Small to medium effect sizes suggest that media played a modest role in university students’ well-being during the initial stage of COVID-19. Media are one piece in the puzzle of coping and well-being, especially during a complex and dynamic situation such as a global pandemic.
Media may be a productive tool for strategic coping, however it is not a panacea. The findings reported here demonstrate users’ traits and motivations interact with media use behavior to influence functional and dysfunctional outcomes of media-based coping, and results clearly demonstrate a range of coping styles may be associated with media use. Continued exploration of different media-based coping strategies employed by individuals—and their unique contributions to stress and anxiety reduction and increased well-being in times of crisis—may elucidate long-standing conflicting findings relating media use with both detrimental and positive psychological outcomes, and better explicate the ways in which media use may be adaptive or maladaptive, based on users’ individual traits, needs, selections and motivations.
The dataset presented in this study can be found in an online repository at:
The studies involving human participants were reviewed and approved by the Human Research Protection Program, Michigan State University, and the Behavioral/NonMedical Institutional Review Board, University of Florida. The participants provided their informed consent to participate in this study.
AE contributed ideas, theorizing, data collection, and writing. BJ contributed theorizing, data collection and analysis, and writing. LR contributed to theorizing and writing. SG contributed to data collection and writing. All authors contributed to the article and approved the submitted version.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Additional open-ended items asked participants to list or describe media content that they had used more, used less, and actively avoided during social distancing, as well as media content they used that was especially hopeful, stressful, connective, depressing, joyful, or guilt-inducing. Analysis of these items will be reported elsewhere.
This variable was named “time spent with media” in the pre-registration, but was changed to more accurately reflect the measurement of media exposure.