Effects of Vicarious Racism Exposure via the Media on College Students of Color

Acts of racism and racial injustice continue to plague today's society. According to the Pew Research Center (2016), 71% of Black and 52% of Hispanic adults have experienced discrimination. Of those surveyed, 11% of Black adults and 8% of Hispanic adults reported experiencing racial discrimination on a regular basis (Pew Research Center, 2016). Furthermore, in a national sample of Asian American immigrants (N = 1,639), 40% reported experiences of racial discrimination (Li, 2014). Moreover, according to hate crime reports disseminated by the Federal Bureau of Investigation (2017), of the 7,227 single‐bias hate crime offenses in 2016, 58.5% were racially motivated. Specifically, of the racially motivated hate crimes, 50.2% were anti‐Black, 20.7% were anti‐White, 10.6% were anti‐Hispanic, 4.2% were antimultiracial, 3.8% were anti‐Native American, 3.1% were anti‐Asian, and 1.3% were anti‐Arab (Federal Bureau of Investigation, 2017).

College students of color are not immune to these experiences of racism. Among 332 university students of color, 96% reported experiencing one or more microaggressions in the past 6 months (Pro et al., 2018). Additionally, among Asian and Pacific Islander university students, researchers found that 76.9% experienced at least one occurrence of discrimination (Chen et al., 2014). Experiences of racism can have substantial detrimental effects on college students of color. For example, racial discrimination can lead to internalized racism, which researchers have found to negatively associate with ethnic identity among college students (Hipolito‐Delgado, 2016). Internalized racism is a phenomenon in which individuals endorse biased and stereotypical beliefs about their own racial or ethnic group (Pyke, 2010). Additionally, the experience of microaggressions can lead to increased levels of psychological distress among university students of color (Robinson‐Perez et al., 2017). Finally, researchers have linked race‐based stress to risky drinking behaviors among Black female college students (Pittman & Kaur, 2018).

Personal experiences of racism can have detrimental consequences among college students, yet Harrell (2000) also noted that “racism exerts its influence not only through direct personal experience, but also vicariously, through observation and report” (p. 45). Indeed, viewing acts of racism against others can lead to negative psychological consequences (Li, 2014; Mason et al., 2017). Individuals can observe racial injustice in their immediate environment or via media coverage of discriminatory acts occurring across the globe. In light of the ubiquitous nature of the internet, modern college students may be exposed to more news coverage of acts of racial injustice than generations past. Consequently, it is important to investigate how vicarious racism affects college students of color, specifically in relation to changes in affect and substance use.

Direct Experiences of Racism, Negative Affect, and Substance Use

The fact that direct experiences of racism can create negative mood states among people of color is not surprising given the traumatic nature of racist incidents (Bryant‐Davis, 2007). In describing race‐based traumatic stress, Carter (2007) noted that “the concept of traumatic stress is valuable in assessing and recognizing race‐based experiences as stress and as traumatic” (p. 37). Furthermore, Ponds (2013) defined racial trauma as “the physiological, psychological, and emotional damage resulting from the stressors of racial harassment or discrimination” (p. 23). Thus, experiences of racism are often distressing and traumatic, inducing negative mood states among people of color.

One potential means of coping with dysphoric affect is substance use. Several researchers have found associations between direct, personal experiences of racism and substance use among people of color. In a study of 1,421 Native American, White, and biracial (Native American and White) youth from the Cherokee Nation, researchers found that those who experienced moderate frequencies of discrimination had greater odds of using marijuana and other drugs (Garrett et al., 2017). Additionally, researchers surveyed university students of color and determined that for every one standard deviation increase in microaggression scores, regular marijuana use increased by 31% (Pro et al., 2018). Finally, Pittman and Kaur (2018) found that increases in race‐based stress among Black college women were associated with increased high‐risk drinking behaviors. Therefore, evidence exists to support relationships among personal experiences of racism, negative affect, and substance use. A less studied form of racism that may also link to negative affect and substance use is vicarious racism.

Vicarious Racism, Negative Affect, and Substance Use

Harrell (2000) described six types of racism‐related stress, including experiences of vicarious racism. Specifically, vicarious racism refers to viewing racial injustice and oppression of others (e.g., friends, family, strangers) through direct observation or report (Harrell, 2000). Heard‐Garris et al. (2018) recommended the following definition for vicarious racism: “the secondhand exposure to the racial discrimination and/or prejudice directed at another individual” (p. 235). This exposure to racial discrimination can lead to myriad effects, including vicarious trauma and race‐based traumatic stress (Carter, 2007).

As with direct experiences of racism, vicarious racism can affect the emotional states of people of color. In a study of Black university students' experiences of the George Zimmerman trial (which investigated the shooting of Trayvon Martin, an unarmed Black adolescent), researchers found that students' level of race‐rejection sensitivity (i.e., the expectation of rejection based on race) predicted more thought intrusions about the Zimmerman trial, which statistically significantly predicted more negative affect and a decreased willingness to forgive (Mason et al., 2017). Researchers determined that 97% of the participants believed that Zimmerman was guilty; thus, his acquittal constituted vicarious racism, which led to intrusive thoughts and negative affect (Mason et al., 2017). Furthermore, in a qualitative investigation of 26 doctoral students of color and recent doctoral graduates, researchers found themes of vicarious racism (e.g., observed racism, trickle down by way of structural racism) and themes of responding to vicarious racism by realizing it is a pervasive feature of education (i.e., normalization of racism) and engaging in racial resistance (i.e., working against racism in education; Truong et al., 2016). The authors noted that the vicarious racism led to negative affective responses from participants (Truong et al., 2016).

Given that vicarious racism can lead to emotional pain and distress, the use of substances as a coping mechanism is possible among college students of color experiencing vicarious racism. Indeed, researchers surveyed 609 university students of color and found that awareness of stereotypes held by out‐group members about the students' own racial group (i.e., metastereotype awareness) statistically significantly correlated with mental health symptoms and drug and alcohol use for coping (Jerald et al., 2017). Using structural equation modeling, the researchers determined that more metastereotype awareness was associated with increased mental health symptomatology, which in turn correlated with decreased self‐care and increased substance use for coping (Jerald et al., 2017). Therefore, as with direct experiences of racism, vicarious racism has the potential to elicit negative affect and increase the risk of substance‐using behaviors among students of color.

Purpose of the Study

Although some research exists examining negative affect, substance use, and vicarious racism among college students of color, to our knowledge, no studies to date have investigated vicarious racism specifically via media coverage of race‐based incidents. Indeed, Heard‐Garris et al. (2018) noted that no study in their systematic review examined vicarious trauma from the media, citing examples of police violence or derogatory racial comments made by officials. Thus, given previously established findings associating vicarious racism with changes in affect and substance use, it is important to investigate whether vicarious racism through the media has similar effects on drug, alcohol, and emotional patterns among students of color. We sought to add to the literature by examining the following research question: Do differences in affect and substance use exist across time between students of color who view racial injustice in the news (experimental group) and students of color who view an emotionally neutral educational video (control group)? Specifically, we hypothesized that there would be statistically significant and meaningful interaction effects between time and group for all four variables (alcohol craving, marijuana craving, positive affect, and negative affect) and that those in the experimental group would demonstrate more alcohol and marijuana craving, more negative affect, and less positive affect at posttest compared with the control group.

Method

We used a quasi‐experimental control group design to examine the effects of vicarious racism via the media among undergraduate students of color. We conducted an a priori power analysis using G*Power software (Version 3.1; Faul et al., 2009). The analysis reported a minimum required total sample size of 34 participants to find statistical interaction effects between two groups using two times of measurement with an alpha level of .05, a moderate treatment effect size (ƒ = .25), and a minimum power of .80.

Participants

Participants were undergraduate students of color at a large southwestern university in the United States. Eligibility requirements included being at least 18 years old, being enrolled in an undergraduate course, and identifying with any race/ethnicity other than White. A sample of 217 participants met the eligibility requirements. We assigned each of the 13 classes to either a control or experimental group. The mean age of the control group (n = 112) was 20.98 years (SD = 2.11). With respect to race/ethnicity, most control group participants identified as Hispanic (n = 39, 34.8%), followed by Black (n = 31, 27.7%), multiple heritage (n = 13, 11.6%), Asian/Pacific Islander (n = 12, 10.7%), biracial (n = 12, 10.7%), Middle Eastern/North African (n = 3, 2.7%), Native American (n = 1, 0.9%), and other (n = 1, 0.9%). The majority of control group participants (n = 89, 79.5%) identified as female, 22 (19.6%) as male, and one (0.9%) as other. The mean age of the experimental group (n = 105) was 20.88 years (SD = 4.29). Experimental group participants reported the following racial/ethnic backgrounds: Hispanic (n = 43, 41.0%), Black (n = 32, 30.5%), Asian/Pacific Islander (n = 16, 15.2%), biracial (n = 6, 5.7%), multiple heritage (n = 4, 3.8%), and other (n = 4, 3.8%). The majority of experimental group participants (n = 67, 63.8%) identified as female, with the remaining participants identifying as male (n = 37, 35.2%) or other (n = 1, 1.0%).

Procedure

After obtaining institutional review board approval, research team members (i.e., the authors of the current study) used convenience sampling to contact undergraduate instructors for permission to conduct the research project in their classes. We gained access to 13 classes across a variety of disciplines within the university. The first author assigned classes either to the control or experimental group based on estimates of class sizes and attempts to create equal‐sized groups. We required 45 minutes of class time to conduct the study, which included completing a pretest survey packet, viewing a control or experimental video, and completing a posttest survey packet. Research team members read aloud the recruitment script describing the intent and voluntary nature of the study. Students who were under 18 years of age, had taken the survey in another class, or did not wish to participate were excused from class. To protect confidentiality, we had participants create a unique, individual code that we used to match pretests and posttests. At the end of data collection in each class, we conducted a drawing in which one to three students received a $5 gift card to a local restaurant.

The study was part of a larger data collection effort, so all students in each of the 13 classes were invited to participate. In all, 493 students had the option to participate in the study. Four‐hundred forty‐nine students (91.1% response rate) completed the pretest and posttest survey packets. Of the 449 participants, we removed 22 because of missing age or incomplete data, resulting in 427 participants (86.6% adjusted response rate). For the current study, we used data from only students of color (n = 217).

Measures

Demographic form. Participants completed a demographic form at pretest only. Demographic questions assessed age, gender, race/ethnicity, and sexual orientation. Additionally, we asked questions related to the research topic, including experiences of racism (i.e., “To what extent have you experienced racism, racial discrimination, and/or racial oppression in the last 30 days?”), news media exposure (i.e., “In a typical day, how much time are you exposed to the news [via social media, internet, or TV news outlets]”?), and alcohol and marijuana use (i.e., “Which option below best describes your alcohol use?” with answer choices ranging from “never drank in lifetime” to “drink daily”). Among members of the control group, 42.9% reported prevalence of racism as not at all, 28.6% as very little, 22.3% as somewhat, 5.4% as much, and 0.9% as a great deal. (Percentages may not total 100 because of rounding.) Among members of the experimental group, 44.8% reported prevalence of racism as not at all, 25.7% as very little, 21.0% as somewhat, 7.6% as much, and 1.0% as a great deal. In the control group, 95.5% of the participants identified as U.S. citizens, and in the experimental group, 83.8% of the participants identified as U.S. citizens.

We also inquired as to the amount of time participants were exposed to news media in a typical day. The most commonly endorsed response for both control and experimental group members was 1 to 2 hours (27.7% and 31.7%, respectively). With regard to substance use, control group members reported frequency of alcohol use as follows: never drank (9.8%), drank but not in last year (12.5%), drink a few times per year (41.1%), drink a few times per month (29.5%), drink a few times per week (6.3%), and drink daily (0.9%). Control group members reported frequency of marijuana use as never used in lifetime (42.9%), used but not in past year (22.3%), use a few times per year (9.8%), use a few times per month (11.6%), use a few times per week (5.4%), and use daily (8.0%). In contrast, experimental group members reported frequency of alcohol use as follows: never drank (22.9%), drank but not in last year (10.5%), drink a few times per year (37.1%), drink a few times per month (23.8%), and drink a few times per week (5.7%). Experimental group members reported frequency of marijuana use as never used in lifetime (53.3%), used but not in past year (18.1%), use a few times per year (10.5%), use a few times per month (7.6%), use a few times per week (7.6%), and use daily (2.9%).

Alcohol Craving Questionnaire–Short Form–Revised. The Alcohol Craving Questionnaire–Short Form–Revised (ACQ‐SF‐R; Singleton, 1997) is a 12‐item measure that assesses current alcohol craving. Items are rated on a 7‐point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Sample items include “If I had some alcohol, I would probably drink it,” “I would feel less irritable if I used alcohol right now,” and “Drinking would put me in a better mood.” Higher scores suggest stronger alcohol craving in that moment. Singleton (1997) reported acceptable reliability for obtained scores, with coefficient alphas ranging from .77 to .86 (cf. Henson, 2001). Items for the ACQ‐SF‐R were derived from a longer version, for which the developers confirmed evidence for score validity (Singleton et al., 1995). In the current study, we calculated acceptable reliability for ACQ‐SF‐R scores, with a coefficient alpha of .81 at pretest and .79 at posttest.

Marijuana Craving Questionnaire–Short Form. The Marijuana Craving Questionnaire–Short Form (MCQ‐SF; Heishman et al., 2009) is a 12‐item measure that assesses current marijuana craving. Respondents rate each item on a 7‐point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Sample items include “If I smoked marijuana right now, I would feel less tense,” “It would be great to smoke marijuana right now,” and “I could not easily limit how much marijuana I smoked right now.” Higher scores on the MCQ‐SF indicate stronger cravings for marijuana. Heishman et al. (2009) reported evidence for score validity and reported reliability with coefficient alphas ranging from .61 to .84. In the current study, we calculated strong reliability for MCQ‐SF scores, with a coefficient alpha of .93 at pretest and .94 at posttest.

Positive and Negative Affect Schedule. The Positive and Negative Affect Schedule (PANAS; Watson et al., 1988) is a 20‐item measure that assesses current mood. Respondents indicate the extent to which they feel each of the 10 positive and 10 negative affect words at that moment. Items are rated on a 5‐point Likert‐type scale ranging from 1 (very slightly or not at all) to 5 (extremely). Sample items from the Positive Affect (PA) subscale include “excited,” “proud,” and “enthusiastic.” Examples of items from the Negative Affect (NA) subscale include “distressed,” “guilty,” and “irritable.” Watson et al. (1988) reported strong reliability for PA subscale scores (α = .90) and for NA subscale scores (α = .87). In the current study, scores for each subscale were reliable, with coefficient alphas of .89 (PA) and .79 (NA) at pretest and .89 (PA) and .86 (NA) at posttest.

Control group video. Research team members used an emotionally neutral video for the control group. The video was a public YouTube clip describing how to best answer job interview questions. The video was approximately 7 minutes long and served as the neutral media stimulus for the control group.

Experimental group video. Research team members developed the experimental group video by editing together clips of media coverage depicting racial/ethnic injustice and discrimination. Team members intentionally chose clips portraying discrimination related to four racial/ethnic groups (Arab/Middle Eastern, Asian, Black, and Latinx) and included the same number of news clips for each racial/ethnic group. Clips included news coverage of racial slurs and gestures, verbal and physical assaults, police shootings of unarmed individuals, reactions to travel bans, and refusal of services because of race. We intentionally did not include news clips that focused on politicians' names or viewpoints but instead included coverage describing the racially oppressive acts. The video was approximately 6 minutes long.

Data Analysis

To address the research question, we used multiple split‐plot analysis of variance (ANOVA) models, which account for both repeated measures and between‐groups effects. Using the general linear model, we analyzed differences between the control and experimental groups on four variables at pretest and posttest. For clarity of interpretation, we used four separate split‐plot ANOVAs to examine each variable independently rather than a multivariate composite. The four variables were alcohol craving, marijuana craving, positive affect, and negative affect. Thus, we had two groups (experimental and control) and four variables assessed at two time points: before the video (pretest) and after the video (posttest). Results indicated whether statistically significant interaction effects between time and group existed for each variable.

Results

Prior to addressing the primary analysis, we examined the correlations between all study variables. Among pretest scores, marijuana craving and alcohol craving were significantly positively correlated (r = .28), as were negative affect and alcohol craving (r = .14). Among posttest scores, marijuana craving and alcohol craving were significantly positively correlated (r = .35), as were positive affect and negative affect (r = .27) and negative affect and alcohol craving (r = .14). All significant correlations were small to medium. To assess normality, we examined the skewness and kurtosis of the eight variables (four pretest and four posttest). One variable, negative affect, was skewed with notable kurtosis. A visual inspection of negative affect scores indicated that the majority of participants had low negative affect scores at pretest, and these scores became more normal at posttest. No other variable had strong distributional shifts. Finally, given the use of only two time points, the assumption of sphericity was not relevant.

We examined the means and standard deviations for all study variables at pretest and posttest (see Table 1). Next, we conducted four split‐plot ANOVAs and found that only one variable, negative affect, had a statistically significant interaction effect between time and group, F(1, 215) = 91.83, p < .001, with an observed power of 1.00. Specifically, those in the experimental group experienced a statistically significant increase in negative affect across time compared with the control group. The interaction effect explained 30% (η2 = .30; large effect size; Cohen, 1988) of the remaining variance in negative affect after controlling for the main effect for group and main effect for time. There were no statistically significant interaction effects for alcohol craving, F(1, 215) = 0.001, p = .949, η2 = .00, power = .05; marijuana craving, F(1, 215) = 0.31, p = .580, η2 = .001, power = .09; or positive affect, F(1, 215) = 1.19, p = .277, η2 = .005, power = .19.

Table 1

Means and Standard Deviations of Study Variables Between Groups

Discussion

In previous studies, researchers found that direct and vicarious racism were associated with negative affect (Carter, 2007; Mason et al., 2017; Ponds, 2013; Truong et al., 2016) and increased substance use (Jerald et al., 2017; Pittman & Kaur, 2018; Pro et al., 2018). To assess the effects of vicarious racism specifically via the media, we investigated whether exposure to racial injustice in the news would link to changes in affect and substance use among college students of color. We hypothesized that we would find statistically significant time and group interaction effects for four variables: positive affect, negative affect, alcohol craving, and marijuana craving. Specifically, we hypothesized that the experimental group would report lower positive affect, higher negative affect, increased alcohol craving, and increased marijuana craving after viewing news coverage of racially oppressive acts. Our hypothesis was only partially supported in that we found a statistically significant interaction effect between time and group for negative affect with a large effect size (η2 = .30; Cohen, 1988). Thus, after watching a 6‐minute media clip of racial injustice in the news, the experimental group experienced higher negative affect than did members of the control group who watched an emotionally neutral video related to job interviewing skills.

An important finding of our study is that vicarious racism via the media, even a brief clip, had a statistically significant effect on the emotional state of students of color. Negative affect increased among experimental group members, indicating that viewing racially oppressive acts in the media was distressing. This finding is consistent with previous research investigating vicarious racism (Mason et al., 2017). Given the prevalence of media coverage related to racial injustice, it is likely that university students are often subjected to increases in negative affect. Additionally, the change in affect occurred after viewing a brief (6‐minute) video. Therefore, experiences of negative affect can occur quickly and intensely.

In addition, it is interesting to note that although negative affect significantly increased at posttest for the experimental group, positive affect did not significantly change. This finding indicates that positive and negative emotions are not opposite ends on the same continuum or inverses of one another. Indeed, correlations between the two subscales of the PANAS for the experimental group were statistically significant but modest (r = .30). Furthermore, previous researchers have found changes in positive affect and negative affect independent of one another (Decker et al., 2018; Rzeszutek & Gruszczyn´ska, 2019). The fact that students in our sample could maintain positive affect from pretest (M = 27.77, SD = 8.99) to posttest (M = 27.11, SD = 9.02) despite experiencing vicarious racism may be an important aspect of coping.

There are several potential explanations as to why the experience of vicarious racism via the media did not influence substance‐using behaviors among our sample. First, it is possible that the dose of the intervention was too short. The video contained multiple short media clips covering racially oppressive acts totaling 6 minutes. It could be that exposure to vicarious racism was not long or powerful enough to affect behavioral outcomes (e.g., substance use). In addition, the students completed the posttest immediately after viewing the video, approximately 10 minutes after completing the pretest. It is possible that cravings for substances emerged later as a maladaptive coping strategy in response to a lasting negative emotional reaction stemming from the video. Finally, a portion of the control group (9.8%) and experimental group (22.9%) reported never drinking alcohol in their lifetime. Similarly, 42.9% of the control group and 53.3% of the experimental group reported never using marijuana in their lifetime. It is possible that for many undergraduate students of color in this sample, substance use is not a common form of coping with distress. Alternatively, participants may have been hesitant to report alcohol use behavior if they were under the age of 21 or marijuana use because of its illegal status.

Implications for College Counselors

The findings of this study have important implications for college counselors. It is noteworthy that almost 30% of the participants in both the control and experimental groups reported experiencing racism somewhat to a great deal. College counselors should be aware of the prevalence of racist incidents experienced by students of color and ensure that they are adequately equipped to address race‐based traumatic stress in their clinical work (Bryant‐Davis, 2007; Carter, 2007). To be equipped, college counselors should consistently assess their own cultural identities (including race and ethnicity) as well as accompanying beliefs, values, attitudes, systemic privilege, power, and biases (Ratts et al., 2016). This self‐awareness is essential for developing an antiracist identity and providing safe therapeutic spaces for clients to disclose experiences of racism. Additionally, it is imperative that counselors respond to client disclosures of racism with compassion, acceptance, and validation rather than minimization or intellectualization (Bryant‐Davis, 2007; Bryant‐Davis & Ocampo, 2006). As with other disclosures of trauma, the counselors' response to the client's report of racism is an important component of the treatment and healing process (Bryant‐Davis & Ocampo, 2006).

Our findings indicate that along with assessing clients' direct experiences of racism, it is important to assess experiences of vicarious racism. This assessment entails asking clients whether they have ever witnessed another person being targeted, harassed, victimized, or treated unfairly or cruelly because of their racial identity. It is important to clarify that this observation of racism could have occurred in person or in the media. As with reports of direct experiences of racism, college counselors should respond to reports of vicarious racism with validation, acceptance, and support.

Prior to assessing clients' experiences with direct or vicarious racism, counselors should broach cultural similarities and differences between themselves and their clients (Day‐Vines et al., 2007). Broaching behaviors should not be obligatory but instead should stem from a genuine curiosity to understand the unique experiences of clients and the recognition of the importance of culture. An example of a college counselor's broaching statement might be the following:

I am aware that you and I have some shared cultural identities—for example, we both identify as women and able‐bodied—yet we also have some differing identities with regard to our race and age. I am curious as to your thoughts about how these similarities and differences might affect our work together.

Counselor‐initiated broaching behaviors invite clients to address topics related to their cultural identities in counseling. Moreover, broaching acknowledges that the privileged or oppressed statuses of both the counselor and client can affect the counseling relationship and provides an opportunity to discuss these implications (Ratts et al., 2016). Indeed, the lack of broaching and exploring culture in counseling may be a contributing factor in research findings that reveal that Black college students are less likely than their White counterparts to return to counseling after intake (Levy et al., 2005).

In addition to broaching, college counselors can support clients of color by processing direct and vicarious experience of racism. Bryant‐Davis and Ocampo (2006) noted several themes relevant for addressing racist incident–based trauma in clinical work. One theme relates to coping strategies, and another relates to resistance strategies. The former theme includes addressing ways in which clients manage distress, and the latter theme pertains to fostering activism and advocacy efforts. Given our finding that students of color who observed racist incidents in the media had significant increases in negative affect, it is important for counselors to explore coping and emotion regulation strategies among clients of color. Importantly, negative affect among our participants changed after brief exposure (i.e., a 6‐minute clip); thus, effective coping strategies should be conducive to quick implementation. These coping strategies may include utilizing relaxation strategies, practicing breathwork, engaging in spiritual practices, or initiating contact with social support.

Along with coping strategies, Bryant‐Davis and Ocampo (2006) described resistance strategies, which include acts of activism and advocacy. Thus, college counselors can help clients of color explore opportunities for social activism in response to vicarious racism. Indeed, in a study of 147 Black undergraduate students, researchers found that when faced with racism‐related stress, participants used the following strategies most frequently: active coping (taking action to address the situation), emotional support (comfort and understanding from others), and instrumental support (seeking help or advice from others; Brown et al., 2011). These findings may explain the lack of statistically significant changes in alcohol and marijuana cravings among our sample. It is possible that students who experienced vicarious racism via the media used active coping strategies to address the issue and thus did not want to be inhibited by psychoactive substances. Therefore, college counselors may best support clients of color by creating safe therapeutic environments to discuss systemic‐level change efforts, acts of activism, and positive community responses to racist incidents.

Limitations

Readers should consider the results of this study in light of several limitations. First, we collected data from a university in one geographic location; therefore, results may not be generalizable to other regions of the country. Second, we used convenience sampling methods to obtain undergraduate participants, and thus, the benefits of random sampling do not apply. Third, our composite video used in the experimental condition included only four racial/ethnic groups, meaning that many racial/ethnic groups were not represented. Fourth, although we informed students in each class that they could participate in the study only one time, we cannot guarantee that all students abided by this guideline because of anonymous coding procedures. Finally, the use of multiple ANOVAs increases the risk of Type I error.

Suggestions for Future Research

Much research is needed to fully understand the experience of vicarious racism via the media. Although we found changes in negative affect immediately after viewing the clip, we did not investigate the intensity and persistence of this negative affect over time. Future research is needed to explore whether the negative affect persisted and intensified or whether the students used a positive coping strategy to address the negative affect quickly. Future researchers may choose to replicate this study among university students of color using a longer media clip or studying the effects of viewing one act of racial injustice in depth among participants belonging to the same racial/ethnic group. Additionally, we gave the posttest to students directly after viewing the 6‐minute clip of news coverage of racial injustice. Future researchers may choose to assess participants at several follow‐up time points in order to investigate the duration and intensity of negative affect among students of color. Furthermore, it is possible that college students have already developed effective coping strategies in response to vicarious racism in the media. Replicating the study with a younger sample would illuminate the effects of vicarious racism at different points of development.

Conclusion

Both direct and vicarious racism can be detrimental to people of color. Vicarious racism can occur from observing racist acts against another person in one's immediate environment as well as media coverage of racist incidents. The results of this study indicate that even brief exposure to racist acts in the media can alter the emotional state of college students of color. In light of the number of racist acts covered in the media, it is likely that college students of color must continually navigate swift and intense increases in negative affect. College counselors can best support clients of color by assessing experiences of direct and vicarious racism and exploring both coping and resistance strategies in counseling.