Young Adults’ Distress About Police Brutality Following the Death of George Floyd

Abstract

Following the homicide of George Floyd on May 25, 2020, young adults participated widely in national protests against racism and police brutality. This study examined temporal trends and correlates of self-reported distress about police brutality from May 18 to August 3, 2020 among a cohort of young adults (ages 19–22; N = 2,080) from Los Angeles, CA. Bivariate and adjusted linear regression models estimated the association between survey completion date and distress about police brutality. Distress about police brutality increased substantially in the 2-weeks after Floyd’s death (43% increase over baseline), but gradually declined over the ensuing 8 weeks to a 19% increase over baseline. Results suggest that police violence is contributing to stress and worry among young adults.

Keywords

police brutality young adults George Floyd

Introduction

On May 25, 2020, George Floyd, a 46-year old Black man from Minneapolis, MN, U.S. died by use of force by local police (The New York Times, 2021). Floyd’s homicide is not an aberration from the experiences of many Black Americans who are mistreated and subjected to excessive and unwarranted use of force by law enforcement—often resulting in murder—in the United States (Chaney & Robertson, 2013; Peeples, 2019; Weir, 2016); For instance, Black men are 2.5 times as likely as White men to be killed by police (Edwards et al., 2019). However, Floyd’s death was recorded on video and shared widely online in the news and social media, sparking a wave of over 10,000 demonstrations and protests across all 50 states and Washington, D.C. between May 26 and August 22, 2020 (ACLED, 2020). Protestors and activists have demanded changes to American policing—from increasing police oversight and accountability to reimagining how public safety is delivered altogether, among other reforms (Andrew, 2020; McKenna, 2020). The United States is now at a crossroads, as Americans reckon with issues of systemic racism, overt bigotry, and the role of police forces in reinforcing systems of oppression (Balko, 2020).

Dramatic societal events are a source of stress that affect wellbeing, especially among communities in which these events frequently occur (Aneshensel, 1992; Williams & Medlock, 2017). Witnessing an event of excessive police force, commonly referred to as “police brutality,” can evoke historical trauma and induce a stress response that contributes to poorer downstream physical and mental health outcomes (Alang et al., 2017). For example, in a recent quasi-experimental study (using population-based data from 2013 to 2015, around the time of the high-profile police killings of Michael Brown, Eric Garner, and Freddie Gray, among others), increases in the number of recent police killings of unarmed Black Americans were associated with higher rates of poor mental health among Black Americans in the states where those killings occurred (Bor et al., 2018).

Young adults participated widely in national protests against racism and police brutality spurned by Floyd’s murder (Bryant, 2020; Gajanan, 2020; Janfaza, 2020; Jurkowitz, 2020), yet empirical evidence quantifying young adults’ stress or concerns about police brutality are lacking. Obtaining rapid information on whether and which groups of young adults perceive police violence as a source of distress is vital to understanding the capacity for young adults to engage in activism and social policy reform (Diemer et al., 2006, 2021). In addition, recent information is needed about how societal unrest related to police violence affects the wellbeing of young adults (American Psychological Association, 2020).

This observational study capitalized on an ongoing cohort survey of young adults from the Los Angeles, CA metropolitan area (home to one of the largest police forces in the U.S., and where police shoot an average of three to four people each month (Levin, 2020)). The survey was already collecting data on police brutality distress before George Floyd’s murder, and continued to do so afterwards. In addition to estimating temporal trends in distress about police brutality as surveys were completed, this study examined sociodemographic and behavioral correlates (e.g., race/ethnicity, political affiliation, and social media use) of that distress. See Figure 1 for a brief timeline of several relevant events and concurrent levels of distress about police brutality in the sample (Ortiz, 2021; Taylor, 2021; University of California, San Diego, 2021; Wright, 2020).

Figure 1.

Brief timeline of events and concurrent distress about police brutality by survey completion date. Bar chart displays the mean value for police brutality distress by week of survey completion. The timeline above displays several relevant events occurring during the period of data collection.

Methods

Study Design

In 2013, 10 high schools from Los Angeles County, USA agreed to participate in a longitudinal cohort survey study. These schools comprised a convenience sample of schools from the county, though participating schools were situated across the county and included students from a wide diversity of socioeconomic and racial/ethnic backgrounds. Within each high school, all 9^th graders (freshmen) were then invited to participate and have since been followed prospectively. The initial surveys were administered in Fall, 2013 (Wave 1; N = 3,396; M_age = 14.1). Students were re-surveyed bi-annually through the 12th grade (Wave 1–Wave 8). After completion of high school, participants were contacted and surveyed in two additional annual waves (Wave 9 and Wave 10). Data were drawn from the most recent wave of data collection (Wave 10), when questions about police brutality distress were assessed. Surveys were completed between May 18th, 2020 and August 3rd, 2020 (analytic N = 2,080; M_age = 21.2). Participants were randomized into one of seven “batches.” Surveys were sent to participants sequentially by batch to ensure that surveys were completed over the entire course of follow-up. Participants completed the survey over a period of 10 weeks. The demographic composition of respondents completing surveys in each of the 10 weeks of follow-up were compared; respondents’ demographics did not vary aside from education, which differed slightly across weeks (range in post- high school degree program enrollment across weeks of follow-up: 43%−68%; p < .01). This study was approved by the University of Southern California Institutional Review Board.

Study Measures

Distress about police brutality

Participants were asked three questions about their levels of concern, worry, and stress about various societal issues, including police brutality (i.e., “Please rate your degree of [concern, worry, stress] toward the following issues in terms of their effects on you personally: Police brutality or the unfair treatment of members in your community by law enforcement”). Each item was measured using a 5-point Likert scale ranging from 1 (Not at all) to 5 (Extremely) (Cronbach α = .95). Mirroring an approach used by Leventhal et al. (2018), a composite score (“distress about police brutality”) was calculated for each participant as the mean of all three items (i.e., concern, worry, and stress; Range: 1–5, M = 3.43, SD = 1.38).

Survey Completion Date

Surveys were collected on an ongoing basis between May 18th, 2020 and August 3rd, 2020. Surveys were grouped into 7-day (1-week) intervals for analysis. This resulted in an 11-category variable, ranging from 0 to 7 days prior to Floyd’s death (Week 0; May 18–24, 2020) and from 64 to 70 days following Floyd’s death (Week 10; July 27–August 2, 2020).

Covariates

Race/ethnicity was assessed categorically (White, Black, Asian, Other/Native American/Hawaiian/Pacific Islander, and Multiracial) with one question (“Which of the following races do you identify with? Check all that apply”). Hispanic/Latino ethnicity was assessed through a separate question (“Are you Hispanic or Latino?”). Non-Hispanic participants were coded into the following racial categories: White, Black, Asian, Multiracial, or Other (inclusive of Native American and Hawaiian/Pacific Islander). The race/ethnicities included in the “Other” category were combined due to small Native American (N = 3), Pacific Islander (N = 15), and other ethnicity (N = 24) sample sizes. Participants that chose multiple races and were non-Hispanic were coded as Multiracial. Participants endorsing Hispanic/Latino ethnicity were coded as a separate Hispanic racial category. Missing values (N = 18) were coded as such.

Gender was assessed using two measures: sex assigned at birth (“What is your assigned sex at birth?” Response options: male, female) and gender identity (“With which gender identity do you most identify?” Response options: Male/masculine, Female/feminine, Transgender male, Transgender female, Gender variant/non-binary, and Additional gender category/identity). Respondents were categorized according to their current gender identity as follows: man (male/masculine), woman (female/feminine), transgender/non-binary (if identified as transgender male, transgender female, gender variant/non-binary, additional gender category, or if sex assigned at birth did not align with current gender identity), or prefer not to disclose (recoded as missing). Missing values (N = 37) were coded as such.

Sexual identity was assessed categorically with the question, “do you consider yourself to be:” (Response options: Asexual, Bisexual, Gay, Straight, Lesbian, Pansexual, Queer, Questioning or unsure, and Another identity not listed). Collapsed categories used for analysis were Straight, Lesbian/Gay, Bisexual, Pansexual/Queer, and Asexual/Questioning/Other. Missing values (N = 58) were coded as such.

Education was assessed with the question, “Are you currently enrolled in a degree program?” (Response options: yes, no, or don’t know). Unknown or unsure degree program status was coded as “no” for analyses. Missing values (N = 14) were coded as such.

Financial status was assessed with a question assessing participants’ subjective personal financial state (“Considering your own income and your income from any other people who help you, how would you describe your overall personal financial situation?”). Responses ranged from greatest to least financial stability (“Live comfortably,” “Meet needs with a little left,” “Just meet basic expenses,” and “Don’t meet basic expenses”) and was treated as a categorical variable in analyses. Missing values (N = 27) were coded as such.

Age was calculated in years based on the difference between date of birth and survey completion date (range = 19.4–22.7, M_age = 21.2, SD = 0.39). There were no missing values for age.

Political affiliation was assessed with one question (“Generally speaking, do you usually think of yourself as a Democrat, Republican, and Independent, or something else?”) with categorical responses (Democrat, Republican, Independent, Something else, Don’t know). Collapsed categories used for analyses were Democrat, Republican, Independent, and Something else/Don’t know. Missing values (N = 19) were coded as such.

Social media exposure. Participants were asked a series of questions about how frequently they used various social media sites (Facebook, Instagram, Tik Tok, Pinterest, Linkedin, Twitter, Snapchat, and Other). Frequencies used were “Monthly or less,” “Weekly,” “Daily,” “Several times a day,” “Don’t know,” and “I do not use this form of social media.” These items were coded into a 1 (Monthly or less) to 4 (Several times a day) scale, with non-use and unknown use coded as 0. A sum score of overall social media use was calculated for each participant. Based on this sum, participants were recategorized as High social media users (Social media use sum score >1 SD from the mean), Average social media users (within ±1 SD from the mean), Low social media users (scores <1 SD of the mean), and non-SM users (Social media score = 0).

Data analysis

Descriptive statistics were first estimated for the sample, including frequencies, percentages, means, and standard deviations, as appropriate. Then, mean differences in distress about police brutality were calculated by survey completion date. Finally, bivariate and adjusted (for all covariates) linear regression models estimated the association between survey completion date and distress about police brutality. There were no missing data, as missing categories were included in the model (e.g., those reporting “prefer not to disclose/missing” for gender identity or “something else/don’t know/prefer not to answer” political affiliation). However, coefficients for missing categories were not presented for ease of interpretation. All analyses were completed in R, version 3.5.3

Results

Descriptive statistics of this cross-sectional sample (n = 2,080; M_age = 21.2) are reported in Table 1. Collapsed across the data collection period, participant’s mean distress rating across the three types of reactions (concern, worry, and stress) to police brutality in their communities was 3.4 (SD = 1.4; range = 1 [not at all distressed] to 5 [extremely distressed]). Respondents were predominantly female/feminine (59%), heterosexual (77%), and Latinx (55%). The majority of respondents were currently enrolled in a degree program (62%). A plurality also reported having a financial status that was either “comfortable” (44%) or having financial needs met “with a little bit left” (30%). The largest proportion of respondents were Democrats (43%), though a sizable proportion were either unsure or unwilling to disclose their political affiliation (37%). Fifteen percent of participants reported no social media use or infrequent social media use (≤1 SD below mean usage), and an additional 15% reported frequent use (≥1 SD above mean usage).

Table 1.

Sample Characteristics (N = 2,080).

	N (%)	M (SD)
Distress about police brutality (range = 1–5)	–	3.4 (1.4)
Survey completion date
Week 0 (5/18/20–5/24/20)	61 (3)	—
Week 1 (5/25/20–5/31/20)	165 (8)	—
Week 2 (6/1/20–6/7/20)	247 (12)	—
Week 3 (6/8/20–6/14/20)	298 (14)	—
Week 4 (6/15/20–6/21/20)	347 (17)	—
Week 5 (6/22/20–6/28/20)	368 (18)	—
Week 6 (6/29/20–7/5/20)	228 (11)	—
Week 7 (7/6/20–7/12/20)	132 (6)	—
Week 8 (7/13/20–7/19/20)	102 (5)	—
Week 9 (7/20/20–7/26/20)	102 (5)	—
Week 10 (7/27/20–8/2/10)	30 (1)	—
Gender identity
Male/masculine	778 (37)	—
Female/feminine	1,220 (59)	—
Transgender/non-binary/other	45 (2)	—
Prefer not to disclose/missing	37 (2)	—
Sexual orientation
Straight	1,594 (77)	—
Lesbian/Gay	61 (3)	—
Bisexual	199 (10)	—
Pansexual/Queer	51 (2)	—
Asexual/Questioning/Other	117 (6)	—
Prefer not to disclose/missing	58 (3)	—
Race/ethnicity
White	229 (11)	—
Black	71 (3)	—
Latinx	1,145 (55)	—
Asian	377 (18)	—
Multiracial	198 (10)	—
Other	42 (2)	—
Missing	18 (1)	—
Age (range = 19.4–22.7)	—	21.2 (0.4)
Degree program
Yes	1,299 (62)	—
No	767 (37)	—
Missing	14 (1)	—
Financial status
Live comfortably	917 (44)	—
Meet needs with a little bit left	629 (30)	—
Just meet basic expenses	440 (21)	—
Don’t meet basic expenses	67 (3)	—
Missing	27 (1)	—
Political affiliation
Democrat	890 (43)	—
Republican	150 (7)	—
Independent	248 (12)	—
Something else/don’t know/prefer not to answer	773 (37)	—
Missing	19 (1)	—
Social media use frequency		—
No social media use	57 (3)	—
Infrequent use (≤1 SD)	253 (12)	—
Moderate use	1,455 (70)	—
Frequent use (≥1 SD)	315 (15)	—

Note. Distress about police brutality was assessed as a composite (mean) of three items: Please rate your degree of (concern, worry, and stress) toward the following issues in terms of their effects on you personally: Police brutality or the unfair treatment of members in your community by law enforcement. Each of the three items was rated on a 5-point Likert scale ranging from “not at all” to “extremely.”

The average value of respondents’ distress about police brutality varied as a function of survey completion date (see Table 2, Figure 1). Distress about police brutality increased immediately following Floyd’s death; the mean distress value in the overall sample rose nearly one point in the week (27% increase over baseline) immediately following Floyd’s death (Week 1; May 25–31, 2020) (B = 0.70; 95% CI [0.30, 1.11], p < .001), compared to the week directly preceding it (Week 0; May 18–24, 2020). Distress about police brutality reached a peak of 3.7 during Week 2 (June 1–7, 2020; 43% increase over baseline, B = 1.12; 95% CI [0.73, 1.51], p < .001), and then decreased gradually week-by-week. By Week 10 (July 27–August 2, 2020), the mean value for distress about police brutality was 3.1 (19% increase over baseline, B = 0.50; 95% CI [−0.10, 1.11], p = .100) and statistically indistinguishable from baseline levels of distress. Sociodemographic differences in trajectories of distress about police brutality were tested using multiplicative interaction terms (e.g., survey completion date × race/ethnicity). However, no statistically significant interactions were found (results not shown).

Table 2.

Associations Between Survey Completion Date and Distress About Police Brutality.

	Bivariate models		Adjusted model
	Coef (95% CI)	p-Value	Coef (95% CI)	p-Value
Survey completion date
Week 0 (5/18/20–5/24/20)	Ref		Ref
Week 1 (5/25/20–5/31/20)	0.70 [0.30, 1.11]	<.001	0.61 [0.24, 0.97]	.001
Week 2 (6/1/20–6/7/20)	1.12 [0.73, 1.51]	<.001	1.00 [0.65, 1.35]	<.001
Week 3 (6/8/20–6/14/20)	0.99 [0.61, 1.38]	<.001	0.90 [0.56, 1.24]	<.001
Week 4 (6/15/20–6/21/20)	0.92 [0.54, 1.3]	<.001	0.82 [0.48, 1.15]	<.001
Week 5 (6/22/20–6/28/20)	0.76 [0.38, 1.14]	<.001	0.68 [0.34, 1.01]	<.001
Week 6 (6/29/20–7/5/20)	0.86 [0.47, 1.25]	<.001	0.87 [0.52, 1.22]	<.001
Week 7 (7/6/20–7/12/20)	0.72 (0.3–1.15)	<.001	0.75 [0.38, 1.13]	<.001
Week 8 (7/13/20–7/19/20)	0.77 [0.33, 1.21]	<.001	0.97 [0.58, 1.36]	<.001
Week 9 (7/20/20–7/26/20)	0.57 [0.13, 1.01]	.010	0.67 [0.28, 1.06]	<.001
Week 10 (7/27/20–8/2/10)	0.50 (−0.10–1.11)	.100	0.56 [0.01, 1.10]	.040
Gender identity
Male/masculine	Ref		Ref
Female/feminine	0.88 [0.76, 1.00]	<.001	0.63 [0.52, 0.75]	<.001
Transgender/non-binary/other	1.12 [0.71, 1.52]	<0.001	0.86 (0.47–1.26)	<.001
Sexual orientation
Straight	Ref		Ref
Lesbian/Gay	0.14 [−0.21, 0.50]	.431	0.07 [−0.25, 0.39]	.660
Bisexual	0.60 [0.40–0.80]	<.001	0.31 [0.13, 0.49]	<.001
Pansexual/Queer	0.93 [0.54, 1.33]	<.001	0.61 [0.25, 0.96]	<.001
Asexual/Questioning/Other	0.11 [−0.14, 0.37]	.386	0.00 [−0.24, 0.23]	.970
Race/ethnicity
White	Ref		Ref
Black	0.36 [−0.01, 0.73]	.055	0.23 [−0.10, 0.56]	.170
Latinx	0.27 [0.07, 0.47]	.008	0.11 [−0.07, 0.29]	.220
Asian	0.15 [−0.08, 0.38]	.203	0.10 [−0.11, 0.30]	0.360
Multiracial	0.29 [0.03, 0.55]	.031	0.16 [−0.07, 0.40]	.170
Other	0.29 [−0.17, 0.74]	.220	0.36 [−0.05, 0.77]	.080
Age	−0.3 [−0.46, −0.15]	<.001	−0.12 [−0.26, 0.02]	.090
Degree program
Yes	Ref		Ref
No	−0.11 [−0.23, 0.02]	.091	0.03 [−0.09, 0.15]	.620
Financial status
Live comfortably	Ref		Ref
Meet needs with a little bit left	0.36 [0.22, 0.50]	<.001	0.25 [0.12, 0.37]	<.001
Just meet basic expenses	0.35 [0.19, 0.51]	<.001	0.32 [0.18, 0.46]	<.001
Don’t meet basic expenses	0.12 [−0.22, 0.47]	.483	0.00 [−0.32, 0.31]	.980
Political affiliation
Democrat	Ref		Ref
Republican	−1.57 [−1.80, −1.34]	<.001	−1.25 [−1.47, −1.03]	<.001
Independent	−0.45 [−0.64 to −0.26]	<.001	−0.27 [−0.45, −0.10]	.002
Something else/Don’t Know/Prefer not to answer	−0.70 [−0.83, −0.57]	<.001	−0.57 [−0.69, −0.45]	<.001
Social media use frequency
No social media use	−1.49 [−1.89, −1.10]	<.001	−1.31 [−1.68, −0.94]	<.001
Infrequent use (≤1 SD)	−0.6 [−0.78, −0.42]	<.001	−0.37 [−0.54, −0.2]	<.001
Moderate use	Ref		Ref
Frequent use (≥1 SD)	0.08 [−0.09, 0.24]	.365	−0.02 [−0.17, 0.13]	.800

Several empirically unique correlates of distress about police brutality were identified in a multivariable regression model controlling for the covariance across different sociodemographic variables (Table 2). Collapsed across the data collection period, both female/feminine ((B = 0.63; 95% CI [0.52, 0.75], p < .001) and transgender/non-binary participants (B = 0.86; 95% CI [0.47, 1.26], p < .001) reported greater levels of distress than male/masculine participants. Bisexual (B = 0.31; 95% CI [0.13, 0.49], p < .001) and pansexual/queer (B = 0.61; 95% CI [0.25, 0.96], p < .001) respondents reported higher levels of distress compared to heterosexual respondents averaged across the study period. Compared to respondents reporting that they lived comfortably, those reporting that they “just met basic needs” (B = 0.32; 95% CI [0.18, 0.46], p < .001) or met needs with “a little bit [of money] left” (B = 0.25; 95% CI [0.12, 0.37], p < .001) reported higher levels of distress about police brutality. Self-identified Republicans (B = −1.25; 95% CI [−1.47, −1.03], p < .001), Independents (B = −0.27; 95% CI [−0.45, −0.10], p < .005), and Something else/don’t know/prefer not to answer (B = −0.57; 95% CI [−0.69, −0.45], p < .001) reported lower levels of distress about police brutality, compared to Democrats. Compared to respondents reporting moderate social media use during this time period, those reporting infrequent social media use (≤1 SD; B = −0.37; 95% CI [−0.54, −0.2], p < 0.001) or no use at all (B = −1.31; 95% CI [−1.68, −0.94], p < .001) reported lower levels of distress about police brutality. Neither education nor race/ethnicity was associated with distress about police brutality.

Discussion

This study found that distress about police brutality reported by young adults from the Los Angeles metropolitan area increased during an 11-week period with historically unprecedented levels of attention toward policing and racism in the United States. The widely-publicized murder of George Floyd, a Black man killed by a police officer while in police custody, appeared to be a critical event for the subsequent unrest and protests ensuing during summer 2020. In our sample, mean levels of distress about police brutality increased sharply immediately following Floyd’s murder, remained elevated for several weeks afterwards, and gradually decreased over time. Parallel calls for police reform intensified shortly after Floyd’s death (Sanchez, 2020), yet the changes that many have called for have stalled, despite the wave of demonstrations and activism on this topic (McCaskill, 2020). In their 2013 to 2015 population-based study of police murders of Black Americans, Bor et al. (2018) found that reductions in mental health among Black Americans were largest in the 1 to 2 months following close-to-home police murders of other Black Americans. Our results similarly suggest there might be a brief window of greatest emotional urgency and heightened concern about social and political affairs in the aftermath of major events among young people, and so mobilizing policy change efforts quickly following major stressful events is key.

In this study, most of the sociodemographic and behavioral factors that were significantly associated with distress about police brutality were groups who have historically experienced greater police violence (transgender/non-binary respondents, several sexual minority subgroups, including bisexual and queer/pansexual respondents, and lower-income respondents). In addition, self-reported Democrats were the most distressed about police brutality, with Independents reported slightly lower distress and Republicans reporting substantially lower distress. This finding aligns with prior research showing Republicans’ generally warmer views toward police officers, compared to their Democratic counterparts (Pew Research Center, 2017). Perhaps surprisingly, race/ethnicity was not associated with distress about police brutality in our sample. It is possible that small sample sizes across racial/ethnic minority groups limited our ability to detect group-level differences in distress about police brutality; only 71 participants in our sample were Black. Also, and importantly, there was only 1 week of data collection prior to Floyd’s death, and the majority of respondents completed the survey afterwards. It is possible race/ethnicity differences would have been apparent before and after the study period, but that during the 10 weeks directly following Floyd’s death, distress about police brutality were increased for all races/ethnicities. Future surveillance is warranted.

There were also strong graded associations between level of social media usage and higher distress about police brutality. A recent report suggests nearly half of adults under 30 receive their news primarily from social media sites (Mitchell et al., 2020). Floyd’s murder and the ensuing demonstrations garnered significant media attention (Heaney, 2020). Protests and activism following Floyd’s murder were also discussed broadly across social media platforms such as Twitter (Anderson et al., 2020; Giorgi et al., 2020). Social media sites such as Twitter often play a vital role in providing young adults access to news information. A longitudinal analysis of activist Twitter hashtag #BlackLivesMatter reported spikes in use of the hashtag following relevant issues covered in traditional news media (Anderson et al., 2018), suggesting amplification of events such as Floyd’s murder likely reaches social media users who do not actively follow racial justice issues in their daily lives. Our data suggest strong associations between exposure to social media and distress about police brutality.

While social media usage may contribute to young adult stress, it is also central to how young adults’ engage with critical sociopolitical issues (Burton, 2019; Kahne & Bowyer, 2018; Pruitt, 2017; Velasquez & LaRose, 2015). Many of the discussions witnessed on, and events organized through, social media exemplify key stages of critical consciousness development, as described by Diemer et al. (2006). One notable example, while not specific to police brutality per se, came from a group of young people who were opposed to former President Trump’s bid for re-election in 2020. They coordinated through social media to reserve a large number of tickets to one of Trump’s political rallies, which they then intentionally did not attend in order to inflate expected attendance numbers and sabotage the rally on national broadcasts (Lorenz et al., 2020). Our results suggest that activist organizations may increase their reach and potential impact by identifying and supporting young activists through social media and fostering collaborations between youth and organizations during, and shortly after highly salient national events. Further, recent research suggests that by relating controversial social issues (e.g., police reform) to personal experiences—rather than by sharing facts—young people may be best able to begin bridging political divides over these seemingly unresolvable disputes (Kubin et al., 2021; van Bavel et al., 2021). Additional work in this area is sorely needed.

Regardless of the changes made (or not) to American policing practices, it is clear from our data that police violence contributes to stress and worry among young adults across a range of sociodemographic characteristics, which practitioners need to be aware of. Distress about police brutality may contribute to mental health morbidity among young adults, who are already reporting disproportionately high levels of anxiety, substance use, and suicidality during the COVID-19 pandemic (Czeisler et al., 2020). Consistent with the recommendations of Bor et al. (2018), in addition to enacting policies aimed at preventing the occurrence of police brutality against minoritized communities, social and behavioral interventions are also needed to help young people cope with acute stress when such events occur. To do so in a culturally relevant manner, behavioral health practitioners and scientists must themselves engage in life-long learning about the socio-historical contexts that shape their clients and research participants’ lives.

Limitations

First, the observational design precludes causal inferences. We cannot attribute the time trend results specifically to the murder of George Floyd or other high-profile reports of excessive police force. Second, although the order of invitation to complete the survey was randomly assigned, some participants took longer than others to respond to the invitation and complete the survey. Thus, the date of survey completion variable may be influenced by systematic factors, although statistical adjustment for the sociodemographic characteristics and social media use empirically controls for the influence of these factors on possible confounding with survey completion date. Third, this was a population-based convenience sample and may not be representative of the larger population of younger adults from the Los Angeles metropolitan area or other regions of the U.S. Given long-standing mistrust of researchers and medical professionals among racial/ethnic minority populations (Scharff et al., 2010), this may be especially true with regard to race/ethnicity. Notably, African Americans were underrepresented in our sample, which may have contributed to non-significant differences between Black and White participants in distress about police brutality. Further, with regard to the categorization of race/ethnicity, we had to collapse several racial/ethnic groups into one “other” category for analysis, though this minimizes their distinct experiences with police brutality distress. Fourth, because the study was originally designed to investigate substance use and health, it did not comprehensively assess young adults’ feelings about, and experiences with police violence. Respondent’s views on police reform and other types of public opinion polling measures were not included, nor were the measures of reactions to policy brutality targeted to black victims specifically addressed.

Conclusions

Using a large diverse sample of young adults, we estimated population trajectories of distress about police brutality immediately before and following the high-profile murder of George Floyd at the hands of Minneapolis police. Results from this study provide compelling evidence of increased levels of distress about police brutality in the 2 weeks immediately following his death. We also found several sociodemographic and behavioral factors to be associated with distress about police brutality. Today’s young adults have unprecedented access to media, providing them with vital information and ideas, while also exposing them to potentially disturbing content (e.g., videos of police brutality events). Clinicians and other practitioners must therefore be attuned to these social issues themselves and be prepared to engage with youth as they grapple with associated stress. Our results, together with other work in this field, may suggest ways in which researchers and activists can help young people leverage social media to begin bridging political divides on controversial topics such as police brutality and reform.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the National Institutes of Health (NIH), grant numbers U54CA180905, R01CA229617, and K01DA042950. This study was also funded by the Tobacco-Related Disease Research Program (TRDRP), grant number 27-IR-0034. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or TRDRP. The corresponding author affirms that all who contributed significantly to this study are listed as study authors.

ORCID iD

Evan A. Krueger

Author Biographies

Lauren C. Howard is a Researcher at the Department of Population and Public Health Sciences at USC Keck School of Medicine. Her work focuses on uncovering sociocultural, behavioral, and systematic factors that influence health disparities in minoritized communities.

Evan A. Krueger is an Assistant Professor at the Tulane University School of Social Work. His research focuses on identifying the social environmental determinants of disparities in health and wellbeing among marginalized populations.

Joshua O. Barker is a Research Director at MMR Strategy Group. His research focuses on the intersections between consumer protection, brand speech, and consumer behavior.

Tess Boley Cruz is an Associate Professor Emerita of Population and Public Health Sciences at the University of Southern California. Her research focuses on the use of health promotion and public health communication to reduce health disparities.

Sam N. Cwalina earned her PhD in Preventive Medicine from the University of Southern California. Her research focuses on tobacco-related mental health disparities among adolescent and young adult populations.

Jennifer B. Unger, Ph.D. is a Professor of Population and Public Health Sciences at the University of Southern California. Her research interests include psychosocial and cultural influences on adolescent health behaviors, tobacco regulatory science, and social media.

Jessica L. Barrington-Trimis is an epidemiologist and Assistant Professor of Population and Public Health Sciences at the University of Southern California. Her work aims to identify intra-individual psychological, behavioral, and social processes associated with nicotine use in adolescence and early adulthood, and to elucidate the behavioral consequences (e.g., transition to more harmful patterns of substance use) and physiological consequences (e.g., adverse respiratory health effects of e-cigarette use) of varying patterns of nicotine product use in adolescence, with the goal of informing regulatory efforts to protect adolescents and young adults.

Adam M. Leventhal is a Professor in the Departments of Population and Public Health Sciences and Psychology at the University. He is the Director of the USC Institute for Addiction Science. He studies the causes, consequences, treatment, and prevention of addiction and mental illness across the lifespan.

References

ACLED. (2020) Demonstrations & political violence in America: New data for summer 2020. https://acleddata.com/2020/09/03/demonstrations-political-violence-in-america-new-data-for-summer-2020/

Alang

McAlpine

McCreedy

Hardeman

(2017). Police brutality and black health: Setting the agenda for public health scholars. American Journal of Public Health, 107(5), 662–665.

American Psychological Association (2020). Stress in America™ 2020: Stress in the time of COVID-19 (Vol. 2). https://www.apa.org/news/press/releases/stress/2020/report-june

Anderson

Barthel

Perrin

Vogels

E. A.

(2020). #BlackLivesMatter surges on Twitter after George Floyd’s death. Pew Research Center. https://www.pewresearch.org/fact-tank/2020/06/10/blacklivesmatter-surges-on-twitter-after-george-floyds-death/

Anderson

Toor

Rainie

Smith

(2018). Activism in the Social Media Age. Pew Research Center. https://www.pewresearch.org/internet/2018/07/11/activism-in-the-social-media-age/

Andrew

(2020, June 17). There’s a growing call to defund the police. Here’s what it means. CNN. https://www.cnn.com/2020/06/06/us/what-is-defund-police-trnd/index.html

Aneshensel

C. S.

(1992). Social stress: Theory and research. Annual Review of Sociology, 18, 15–38.

Balko

(2020, June 10). There’s overwhelming evidence that the criminal justice system is racist. Here’s the proof. Washington Post. https://www.washingtonpost.com/graphics/2020/opinions/systemic-racism-police-evidence-criminal-justice-system/

Bor

Venkataramani

A. S.

Williams

D. R.

Tsai

A. C.

(2018). Police killings and their spillover effects on the mental health of black Americans: A population-based, quasi-experimental study. The Lancet, 392(10144), 302–310.

10.

Bryant

(2020, June 15). “It was time to take charge”: The Black youth leading the George Floyd protests. The Guardian. https://www.theguardian.com/world/2020/jun/15/black-youth-activism-george-floyd-protests

11.

Burton

(2019). Look at us, we have anxiety: Youth, memes, and the power of online cultural politics. Journal of Childhood Studies, 44(3), 3–17.

12.

Chaney

Robertson

R. V.

(2013). Racism and police brutality in America. Journal of African American Studies, 17(4), 480–505.

13.

Czeisler

M. É.

Lane

R. I.

Petrosky

Wiley

J. F.

Christensen

Njai

Weaver

M. D.

Robbins

Facer-Childs

E. R.

Barger

L. K.

(2020). Mental health, substance use, and suicidal ideation during the COVID-19 pandemic—United States, June 24–30, 2020. Morbidity and Mortality Weekly Report, 69(32), 1049.

14.

Diemer

M. A.

Kauffman

Koenig

Trahan

Hsieh

C.-A.

(2006). Challenging racism, sexism, and social injustice: Support for urban adolescents’ critical consciousness development. Cultural Diversity and Ethnic Minority Psychology, 12(3), 444.

15.

Diemer

M. A.

Pinedo

Bañales

Mathews

C. J.

Frisby

M. B.

Harris

E. M.

McAlister

(2021). Recentering action in critical consciousness. Child Development Perspectives, 15(1), 12–17.

16.

Edwards

Lee

Esposito

(2019). Risk of being killed by police use of force in the United States by age, race–ethnicity, and sex. Proceedings of the National Academy of Sciences of the United States of America, 116(34), 16793–16798.

17.

Gajanan

(2020, June 4). Protests are being held in small cities and towns across the U.S.—And young people are leading the charge. Time. https://time.com/5847228/george-floyd-nationwide-protests/

18.

Giorgi

Guntuku

S. C.

Rahman

Himelein-Wachowiak

Kwarteng

Curtis

(2020). Twitter corpus of the# BlackLivesMatter movement and counter protests: 2013 to 2020. ArXiv Preprint ArXiv:2009.00596.

19.

Heaney

(2020, July 6). The George Floyd protests generated more media coverage than any protest in 50 years. The New York Times. https://www.washingtonpost.com/politics/2020/07/06/george-floyd-protests-generated-more-media-coverage-than-any-protest-50-years/

20.

Janfaza

(2020, June 3). Across the country, young activists take different approaches in the name of justice for George Floyd. CNN. https://www.cnn.com/2020/06/03/politics/young-activists-george-floyd/index.html

21.

Jurkowitz

(2020, July 9). Younger adults differ from older ones in perceptions of news about COVID-19, George Floyd protests. Pew Research Center. https://www.pewresearch.org/fact-tank/2020/07/09/younger-adults-differ-from-older-ones-in-perceptions-of-news-about-covid-19-george-floyd-protests/

22.

Kahne

Bowyer

(2018). The political significance of social media activity and social networks. Political Communication, 35(3), 470–493.

23.

Kubin

Puryear

Schein

Gray

(2021). Personal experiences bridge moral and political divides better than facts. Proceedings of the National Academy of Sciences of the United States of America, 118(6), e2008389118.

24.

Leventhal

A. M.

Cho

Andrabi

Barrington-Trimis

(2018). Association of reported concern about increasing societal discrimination with adverse behavioral health outcomes in late adolescence. JAMA Pediatrics, 172(10), 924–933.

25.

Levin

(2020, September 6). “Reign of terror”: A summer of police violence in Los Angeles. The Guardian. https://www.theguardian.com/us-news/2020/sep/06/los-angeles-police-killings-dijon-kizzee-andres-guardado

26.

Lorenz

Browning

Frenkel

(2020, November 6). TikTok Teens and K-Pop Stans Say They Sank Trump Rally. New York Times. https://www.nytimes.com/2020/06/21/style/tiktok-trump-rally-tulsa.html

27.

McCaskill

(2020, September 23). Police reforms stall around the country, despite new wave of activism. Politico. https://www.politico.com/news/2020/09/23/breonna-taylor-police-reforms-420799

28.

McKenna

(2020, June 17). Police violence calls for measures beyond de-escalation training. Scientific American. https://www.scientificamerican.com/article/police-violence-calls-for-measures-beyond-de-escalation-training1/

29.

Mitchell

Jurkowitz

Oliphant

J. B.

Shearer

(2020). Americans who mainly get their news on social media are less engaged, less knowledgeable. Pew Research Center. https://www.pewresearch.org/journalism/2020/07/30/americans-who-mainly-get-their-news-on-social-media-are-less-engaged-less-knowledgeable/

30.

The New York Times. (2021, May 25). How George Floyd Died, and what happened next. The New York Times. https://www.nytimes.com/article/george-floyd.html

31.

Ortiz

(2021, May 6). What to know about the death of Rayshard brooks. The New York Times. https://www.nytimes.com/article/rayshard-brooks-what-we-know.html

32.

Peeples

(2019). What the data say about police shootings. Nature, 573(7772), 24–27.

33.

Pew Research Center. (2017). Partisans differ widely in views of police officers, college professors. https://www.pewresearch.org/politics/2017/09/13/partisans-differ-widely-in-views-of-police-officers-college-professors/

34.

Pruitt

(2017). Youth, politics, and participation in a changing world. Journal of Sociology, 53(2), 507–513.

35.

Sanchez

(2020, June 14). Police reforms quickly take hold across America. It’s only just getting started. CNN. https://www.cnn.com/2020/06/14/us/george-floyd-national-police-reforms/index.html

36.

Scharff

D. P.

Mathews

K. J.

Jackson

Hoffsuemmer

Martin

Edwards

(2010). More than Tuskegee: Understanding mistrust about research participation. Journal of Health Care for the Poor and Underserved, 21(3), 879.

37.

Taylor

D. B.

(2021, March 28). George Floyd protests: A timeline. The New York Times. https://www.nytimes.com/article/george-floyd-protests-timeline.html

38.

University of California, San Diego. (2021, September 2). Protests against police brutality (2020). https://ucsd.libguides.com/protests2020

39.

van Bavel

J. J.

Reinero

D. A.

Spring

Harris

E. A.

Duke

. (2021). Speaking my truth: Why personal experiences can bridge divides but mislead. Proceedings of the National Academy of Sciences of the United States of America, 118(8), e2100280118.

40.

Velasquez

LaRose

(2015). Youth collective activism through social media: The role of collective efficacy. New Media & Society, 17(6), 899–918.

41.

Weir

(2016, December). Policing in black & white. American Psychological Association. https://www.apa.org/monitor/2016/12/cover-policing

42.

Williams

D. R.

Medlock

M. M.

(2017). Health effects of dramatic societal events—ramifications of the recent presidential election. Mass Medical Society.

43.

Wright

(2020, June 23). Louisville police fire brett Hankison, officer in Breonna Taylor Shooting. The New York Times. https://www.nytimes.com/2020/06/23/us/breonna-taylor-brett-hankison-fired.html