The Impact of Social and Material Resources on Resilience Communication at the Intersection of Race and Gender

Abstract

In the communicative theory of resilience (CTR), communication constructs resilience processes that can promote change or continuity during hardship. The enactment of resilience is theorized to depend on available resources. The current study tests this theoretical link in the context of the COVID-19 pandemic, examining differences at the intersection of race and gender (N=588). Job security, education quality, and friend support positively predicted continuity and change resilience. Four differences based on race/gender emerged: necessities negatively predicted continuity resilience for Black men and White women, healthcare and government representation positively predicted continuity resilience for Black women only, and family support positively predicted change resilience for Black women and White men. Findings support CTR’s claim that resource access influences resilience enactment and indicate that theoretical associations differ based on race/gender. The current research emphasizes the importance of considering intersectionality in relation to CTR processes and structural barriers to enacting resilience.

The COVID-19 pandemic constituted a global health emergency (WHO, 2020), creating new hardships and exacerbating existing inequities faced by communities in the United States. Across the globe, countries and states engaged in social distancing protocols to stop the spread of COVID-19, including shutting down or restricting businesses, schools, and churches, prohibiting gatherings, and working from home (Block et al., 2020). While these measures were necessary to slow the virus, they resulted in hardships, including job loss, unstable housing, domestic abuse, and food insecurity (see CBPP, 2021, Keith-Jennings et al., 2021; Silva et al., 2020).

One beneficial framework for studying different communities’ responses towards hardship is the communication theory of resilience (CTR; Buzzanell, 2010). CTR conceptualizes resilience as a process that resides within communication, identifying five key resilience communication processes that promote adaptation and growth (Buzzanell, 2010). Rather than placing the onus of being resilient on a person or community, CTR posits that the ability to enact resilience is determined by access to social and material resources (Buzzanell, 2019; Houston & Buzzanell, 2018). Although findings from recent studies suggest this connection exists (Dorrance Hall & Scharp, 2021; Lillie et al., 2021), the association between access to resources and the enactment of CTR’s resilience communication processes has not been systematically tested. It is the primary goal of the current study to establish the theorized link between resource access and the enactment of resilience communication processes.

Further, it is important to establish whether these associations differ across people groups. Qualitative research has identified CTR’s processes in diverse samples (e.g., Kam et al., 2018; Long et al., 2015; Tian & Bush, 2020). Research quantitatively testing CTR’s postulates, however, is in its early stages (Buzzanell, 2019). Thus far, quantitative studies have utilized predominantly White samples (e.g., Lillie et al., 2021; Wilson et al., 2021), with the exception of Kuang et al.’s (2021) recent work examining anticipatory resilience in Chinese adults. The current study, therefore, assesses whether relationships between resource access and resilience communication differ based on race and gender identity.

Black residents were disproportionately burdened by the COVID-19 pandemic. Black Americans accounted for 12.4% of all cases and 13.2% of covid-related deaths as of October 5, 2022 (CDC, 2022), indicating a higher risk of dying from COVID-19. Black residents, broadly, are more likely to be exposed to COVID-19 due to their occupations (CDC, 2020). Additionally, Black residents are more likely to be diagnosed with chronic illnesses and less likely to have insurance covering hospitalizations and COVID-19 testing (Egede & Walker, 2020). Preliminary information has also hinted at the role of policing during the pandemic for Black residents who were more likely to be cited for violating stay-at-home and social distancing orders (Dunbar & Jones, 2021) and less likely to wear face coverings out of fear of being perceived as threatening (Alfonso III, 2020; Natividad, 2020).

In addition to the racial disparities exacerbated by the COVID-19 pandemic, gender inequities have also been observed. Women are disproportionately more likely to take on caregiver roles both for children and for sick family members (Connor et al., 2020), resulting in greater lost work hours and potential exposure to COVID-19 (Morgan et al., 2021). Further, partner violence against women increased during the pandemic, linked to financial struggles, alcohol abuse, and lack of support from women’s social networks (Silva et al., 2020). Combined with the racial disparities discussed above, this suggests that Black women were at greater risk for negative health and life impacts compared to other groups—a reflection of the intersection of multiple structural inequities (see Crenshaw, 1991). In her seminal article, Crenshaw (1991) wrote that “intersectional subordination…is frequently the consequence of the imposition of one burden that interacts with preexisting vulnerabilities to create yet another dimension of disempowerment” (pg. 1249). The COVID-19 pandemic, as we begin to argue above, created a new “dimension of disempowerment” by building upon already existing inequities. Therefore, understanding what resources enable Black women to be resilient during times of hardship such as the COVID-19 pandemic is vital for guarding against such inequities in the future.

It is crucial both for theory-building and practical application to determine if the predictors of CTR’s resilience communication processes vary across people groups. A key problem within communication research has been an emphasis on scholarship by and about White populations (Afifi & Cornejo, 2020; Chakravartty et al., 2018). Therefore, the current study seeks to identify if the associations between resource access and the CTR processes differ at the intersection of race and gender within the COVID-19 context. Buzzanell (2019) states that resilience research that takes “a critical approach would challenge people to consider the ways in which difference (e.g., gender, age, ability, class, nationality) constitutes how people enact resilience processes” (p. 78), and Kuang et al. (2021) called for cross-cultural research that assesses how predictors (and outcomes) of resilience communication both differ and agree across cultures. Therefore, this study has stratified sampling across four groups: Black women, Black men, White women, and White men. This application will serve as a step in understanding the generalizability of CTR.

Communication Theory of Resilience

The communication theory of resilience (CTR; Buzzanell, 2010) defines resilience as “a constitutive process through which people reintegrate and actively construct their new normal through language, interaction, networks, and attention to their identities and identifications, within their material environments and societal discourses” (Buzzanell, 2019, p. 68). In CTR, resilience resides within interaction and discourse, making resilience communicative in nature (Buzzanell, 2018a). Resilience can either be anticipatory, meaning in preparation for an expected hardship, or reactive, meaning in response to a disruption that has occurred (Betts et al., 2021; Buzzanell, 2019).

Resilience is constituted through five core resilience processes (Buzzanell, 2010). First, crafting normalcy includes maintaining routines and rituals from before the hardship as well as creating new routines to adapt to the hardship (Wilson et al., 2021). Second, foregrounding productive action while backgrounding negative feelings means finding ways to move forward and take beneficial action while acknowledging that negative emotions are valid (Buzzanell, 2019). Third, affirming identity anchors focuses on emphasizing and validating aspects of one’s identity that were important before the hardship (Buzzanell, 2018a). Fourth, utilizing communication networks involves connecting with and receiving support from one’s social network and/or building a hardship-specific network (e.g., online cancer support group). Finally, using alternative logics involves reframing the hardship in a positive or beneficial manner (e.g., an adventure; Villagran et al., 2013), with humor being a common strategy for reframing (Chernichky-Karcher et al., 2019).

A recent addition to CTR classifies the resilience communication processes into two larger umbrellas: continuity and change (Wilson et al., 2021). Continuity resilience processes enable individuals to maintain important facets of their lives from before the hardship. This includes maintaining routines, affirming identity anchors, relying on communication networks, and taking productive action. Change resilience processes involve adaptation in the face of hardship, including creating new routines, utilizing humor, and reframing the hardship. Some processes could arguably contribute to either continuity or change (e.g., communication networks), but each process primarily facilitates one or the other.

Although continuity and change appear to be opposite methods for managing hardship, they often occur in tandem (Kuang et al., 2021). For example, an individual may be able to maintain routines in one area of life while needing to adapt and create new routines in another. Flexibility between continuity and change is likely more adaptive because it allows individuals to tailor their hardship responses as their circumstances change. Rather than being opposites, continuity and change are distinct, adaptive strategies whose true opposites are maladaptive responses such as apathy, inaction, negative rumination, or violence. Further, studies have consistently found that continuity and change processes are significantly, positively correlated, suggesting individuals typically do perform both (Chernichky-Karcher et al., 2019; Kuang et al., 2021; Lillie et al., 2021; Wilson et al., 2021). The relationship between continuity and change resilience processes is explicated by Kuang et al., (2021), explaining that continuity-change can serve as a dialectical tension.

The ability to enact resilience is theorized to be dependent on access to social and material resources (Buzzanell, 2019; Wilson et al., 2021). Yet, research has not systematically tested the association with multiple types of resources. Dorrance Hall and Scharp (2021) identified that family and friend support are positively associated with social network resilience, part of communication networks. Importantly, recent qualitative CTR scholarship has explored the social and material barriers to resilience for marginalized groups, including migrants and first-generation college students (Scharp et al., 2021; 2022). Additionally, Hintz et al. (2021), focusing on the health care workers experiences during the COVID-19 pandemic, explore the impact of a shortage of personal protective equipment, a material resource vital in healthcare settings. Understanding how resources are related to resilience is of practical, as well as theoretical, importance because resilience communication has been linked to important outcomes such as improved mental health, coping, and health management (Lillie et al., 2021; Kuang et al., 2021; Venetis et al., 2020; Wilson et al., 2021).

Access to Social and Material Resources

CTR posits that enactment of resilience is predicated by resource access (Buzzanell, 2010). Additionally, differing access to resources are an important component of intersectional experiences (Crenshaw, 1991). Particular resources may be applicable across contexts, such as family support and housing. Others may be context specific, such as healthcare quality (Buzzanell, 2019). The current study examines the influence of eight resources on continuity and change resilience. These resources were selected based on three criteria: (a) past CTR literature and theorizing (Buzzanell, 2010), (b) indicators of structural inequality (Bailey et al., 2017), and (c) relevance to the COVID-19 pandemic. Several of the resources fit all three criteria, such as job security. Each resource, including family support, friend support, necessities, education quality, job security, healthcare quality, police support, and government representation, and the rationale for their inclusion are discussed below.

The first two resources are family support and friend support. According to CTR, resilience is constituted via human interaction (Buzzanell, 2010). Having access to quality social support from close others is, therefore, vital to the enactment of resilience. Family members can be particularly helpful at continuity resilience because of their long-term connections, better enabling maintenance of routines and identity affirmation (Buzzanell, 2018b; Kam et al., 2018). Conversely, families may inhibit resilience by enforcing unhelpful gender norms or constraining communication (Buzzanell, 2018b; 2019; Donovan-Kicken & Caughlin, 2010; Scharp et al., 2021). When families fail to help, individuals rely on other relationships, particularly friendships (Dorrance Hall, 2018). Dorrance Hall and Scharp (2021) found that friend support was vital for promoting social network resilience, part of the communication network process.

Insecurity around necessities like food and shelter is a key indicator of structural inequality (Bailey et al., 2017). During the COVID-19 pandemic, many families faced food and housing insecurity, especially families of color (CBPP, 2021; Keith-Jennings et al., 2021)​. Further, Black residents had more limited access to food and other necessities compared to White residents during the COVID-19 pandemic (Ruprecht et al., 2020), and material hardship (primarily assessed via food and housing insecurity) has been related to diminished support within Black families (Taylor et al., 2021). Limited access to necessities can diminish individuals’ ability to maintain normalcy (Sánchez Sánchez & Lillie, 2019) and would complicate taking productive action. Further, food and housing insecurity could limit the flexibility required of change resilience.

Similarly, education quality serves as a litmus test for structural inequality (Bailey et al., 2017). Here, we do not refer to education level but to the quality of resources and the respect provided to students. Limited funding and lack of quality resources are common problems in K-12 schools that predominantly serve students of color (Noguera, 2017). In the United States, educational inequalities persist (Aylward, 2017). For example, discrimination and racial injustices have resulted in lower admittance of Black students to gifted programs and an increased likelihood of expulsion, suspension, arrest, and transfer to alternative schools (Grissom & Redding, 2016; U.S. Department of Education, 2021). Education quality can influence how well people adjust to hardships (Frankenburg et al., 2013; Lightfoot et al., 2020). For example, higher quality education develops critical thinking skills beneficial for productive action and the development of adaptive routines.

Job security was a key resource for navigating the COVID-19 pandemic. The rate of unemployment rose to a height that had not been observed for nearly a century (CBPP, 2021). The hardest hit workers tended to be those with the least formal education, women, and other minoritized groups (Center on Education and the Workforce, 2021). Building resilience in relation to job insecurity is a highly gendered process, favoring the identity and normalcy needs of men (Buzzanell & Turner, 2003; Lucas & Buzzanell, 2012). Job insecurity adds significant challenge to normalcy maintenance, and beneficial communication and respect in the workplace can spillover to increase resilience in other areas of life (LaGree et al., 2021; Lucas & Buzzanell, 2012; Roeder et al., 2021). Job security could enable positive alternative logics, like the sense that one is lucky despite the hardship (Lillie et al., 2018).

Healthcare quality was a crucial resource during the COVID-19 pandemic, including not only insurance concerns, but also respect and overall quality of care. Pre-pandemic studies had observed the negative effects of racism on individual health, particularly for Black residents (Egede & Walker, 2020; Paradies et al., 2015). Several studies conducted at the state level have found an overrepresentation of Black patients needing hospitalization due to COVID-19 (Gold et al., 2020; Price-Haywood et al., 2020). Further, Black COVID-19 patients, compared to White patients, were more likely to live in low-income neighborhoods and have Medicaid insurance (Price-Haywood et al., 2020). Low healthcare quality could be dehumanizing, disaffirming identity anchors, and could make adaptation/change challenging when carrying a risk of COVID-19 exposure. During a time when many areas faced triaging in medical care (Supady et al., 2021), it is likely that unequal access to quality healthcare would limit resilience.

Alongside the pandemic, the summer of 2020 was punctuated with protests about policing and government policy (Njoku et al., 2021). These protests, combined with fear-based hesitancy related to mask-wearing among Black community members (Alonso III 2020; Dunbar & Jones, 2021; Natividad 2020), add to the distrust of government and police structures which hold resources that could aid during a pandemic. Government representation is a key resource during times of hardship and an indicator of structural inequality (Lukachko et al., 2014). Importantly, government officials determine institutional-level responses to crises. For groups that do not have adequate representation, those responses will likely not be reflective of their needs or may even run counter to them. Voter suppression and inadequate government representation have been linked to the disproportionate number of Black residents who have died from COVID-19 (Moore et al., 2020).

In sum, CTR proposes that greater resource access will facilitate resilience (Buzzanell, 2019). We hypothesize that the eight resources described above will predict both continuity and change resilience.

• H1: Continuity resilience will be positively related to (a) family support, (b) friend support, (c) necessities, (d) education quality, (e) job security, (f) healthcare quality, (g) police support, and (h) government representation.

• H2: Change resilience will be positively related to (a) family support, (b) friend support, (c) necessities, (d) education quality, (e) job security, (f) healthcare quality, (g) police support, and (h) government representation.

Further, we anticipate that these associations differ based on race and gender. Specifically, due to the increased burden of discrimination placed upon them, we hypothesize the following:

• H3: The associations proposed in H1 and H2 will be moderated by race, such that the associations will be stronger for Black individuals compared to White individuals.

• H4: The associations proposed in H1 and H2 will be moderated gender, such that the associations will be stronger for women compared to men.

• H5: The associations proposed in H1 and H2 will be moderated by race and gender, such that the associations will be stronger for Black women compared to Black men, White women, and White men.

Methods

Participants and Procedures

Participants were recruited via Qualtrics Panels, a company that maintains a nationally representative (United States) participant pool and hosts an online survey platform. Qualtrics contacted potential participants from their pool. Interested participants followed a link to the survey on the Qualtrics platform. After providing electronic consent, participants completed an anonymous survey, assessing their access to social and material resources and their use of resilience communication. Qualtrics ensured even stratification into the four racial/gender categories (Black women: n = 158, Black men: n = 130, White women: n = 155, and White men: n = 145). Biracial individuals who identified as both Black and White (n = 7) were included in the Black participant groups. Gender was coded as a binary to allow for comparative statistical analysis. Participants had an average age of 44.05 years (SD = 15.34, Range = 18–87). For additional demographic information, see Table 1. The research was approved by the university Institutional Review Board (IRB).

Table 1.

Participant Demographics

Variables
Education
High school or less	24.1% (142)
Some college, no degree	25.3% (309)
2-year degree	13.6% (80)
4-year degree	18.9% (111)
Master’s degree	13.1% (77)
PhD, MD, JD	4.9% (29)
Race/Ethnicity
American Indian or Alaska Native	0.5% (3)
Asian or Pacific Islander	0.6% (4)
Black or African American	49.0% (288)
Hispanic or Latina/o	0.9% (5)
White or Caucasian American	52.2% (307)
Marital Status
Single	42.5% (250)
In a relationship (not married)	13.8% (81)
Married	43.7% (257)
Essential Worker
Yes	41.8% (246)
No	58.2% (342)
Work from Home
Yes	40.0% (235)
No	60.0% (353)
Children in Home
Yes	36.1% (212)
No	63.9% (376)

Note. Participants could identify as more than one race/ethnicity.

Participants were given a description of what it meant to be an “essential worker”: “Essential workers hold jobs that are considered necessary for the community, state, country, etc. to continue functioning, such as those who work in healthcare, food production, delivery, water and wastewater management, and emergency services.”

Measures

Resilience Communication

Resilience communication was measured using the communication resilience processes scale (CRPS; Wilson et al., 2021). Participants responded to 32 items on a scale from 1 (strongly disagree) to 6 (strongly agree), regarding their response to the COVID-19 pandemic. The items capture seven resilience sub-processes which are grouped into two overarching categories: continuity and change. Continuity resilience (α = .91, M = 4.47, SD = 0.74) includes maintain routines (4 items, “I made an effort to keep up with my daily routines”), affirming identity anchors (6 items, “I kept in mind who I wanted to be throughout the situation”), networks (5 items, “I turned to other people in my network for what I needed”), and productive action (4 items, “Despite how I was feeling, I focused on taking constructive actions”). Change resilience (α = .92, M = 4.34, SD = 0.89) includes new routines (4 items, “I started to do new things that over time became ordinary”), reframing (5 items, “I tried to see the difficult situation in a new light”), and humor (4 items, “I relied on humor to get through the challenging times”). For correlations between study variables, see Table 2.

Table 2.

Correlation Matrix

Variable	1	2	3	4	5	6	7	8	9
1. Continuity Resilience	----
2. Change Resilience	.78**	----
3. Family Support	.47**	.44**	----
4. Friend Support	.49**	.48**	.65**	----
5. Necessities	.28**	.24**	.40**	.42**	----
6. Education	.46**	.40**	.46**	.48**	.42**	----
7. Job Security	.39**	.40**	.36**	.36**	.42**	.38**	----
8. Healthcare	.36**	.29**	.44**	.46**	.58**	.48**	.35**	----
9. Police Support	.24**	.13*	.27**	.32**	.43**	.31**	.30**	.45**	----
10. Government Representation	.40**	.39**	.42**	.45**	.40**	.42**	.39**	.46**	.44**

^*p < .01,

^**p < .001

Social and Material Resources

Access to social and material resources was measured using two different scales. First, family and friend support was assessed using the family (4 items; α = .92, M = 5.19, SD = 1.57; “I get the emotional help and support I need from my family”) and friends (4 items; α = .93, M = 5.18, SD = 1.50; “I can count on my friends when things go wrong”) subscales of the multidimensional scale of perceived social support (Zimet et al., 1988). Participants indicated agreement with each item on a scale from 1 (strongly disagree) to 7 (strongly agree).

Second, a self-report resource access measure was developed. Items reflect access to necessities (e.g., housing, food), education quality, job security, healthcare quality, police support, and government representation. A confirmatory factor analysis (CFA) was used to confirm the six-factor structure in Amos 27. The model achieved good fit CFI = .96, RMSEA = .05, χ2/df = 2.96 (Schermelleh-Engel et al., 2003). All items loaded above .6 on their factor (.66–.88). See Table 3 for items, factor loadings, alphas, means, and standard deviations.

Table 3.

Confirmatory Factor Analysis for Access to Resources Measure

Item		2	3	4	5	6
I can easily pay for medical care (through insurance and/or out-of-pocket).	.74
The healthcare I receive is high quality.	.85
I know I will be treated with respect at a hospital/clinic/doctor’s office.	.76
I have good health insurance.	.80
I have had a quality education.		.75
My elementary/high school had quality resources.		.71
In elementary/high school, my teachers treated me with care and respect.		.66
I never worry about having housing.			.79
It’s been easy for me to have/afford electricity.			.79
I never worry about being able to afford food.			.83
It is easy for me to get healthy foods (like fresh fruits and vegetables).			.73
I never worry about being able to afford toiletries (soap, shampoo, etc.).			.83
My vote counts towards who will be elected to government positions.				.73
Government policies often reflect what is best for me and my community.				.81
I have a say in what policies the government creates.				.88
I can trust law enforcement/police to help me.					.86
I am not afraid of law enforcement/police.					.79
My experiences with law enforcement/police have been positive.					.79
I have a secure job.						.87
I am respected where I work.						.83
Even if I had to leave my job, it would be easy for me to find work elsewhere.						.71
M	3.71	3.98	3.73	3.50	3.56	3.32
SD	1.03	0.86	1.04	1.09	1.14	1.17
α	.86	.75	.89	.85	.85	.84

Note. Factors include healthcare quality, education quality, necessities, government representation, police support, and job security.

Results

MANOVA

A two-way MANOVA was used to test for differences in variable means at the intersection of race and gender. Dummy-coded variables for racial identity (0 = White; 1 = Black) and gender (0 = male, 1 = female) were used in the analyses. Analyses identified differences in two variables including change resilience F(1, 584) = 4.70, p = .031 and police support F(1, 584) = 5.18, p = .023. See Table 4 for variable means and standard deviations by group.

Table 4.

Study Variable Means and Standard Deviations by Racial Identity and Gender

					Black Participants		White Participants
Variable	Black	White	Women	Men	Women	Men	Women	Men
Continuity	4.50 (0.76)	4.45 (0.73)	4.37 (0.72)	4.59 (0.75)	4.44 (0.77)	4.57 (0.75)	4.29 (0.66)	4.61 (0.75)
Change	4.43 (0.83)	4.26 (0.93)	4.22 (0.89)	4.48 (0.86)	4.38 (0.88)	4.48 (0.78)	4.06 (0.88)	4.48 (0.94)
Family	5.13 (1.62)	5.24 (1.51)	4.98 (1.66)	5.42 (1.42)	4.91 (1.66)	5.39 (1.53)	5.05 (1.65)	5.44 (1.33)
Friend	5.08 (1.52)	5.27 (1.49)	4.98 (1.58)	5.41 (1.38)	4.81 (1.58)	5.41 (1.37)	5.14 (1.57)	5.41 (1.39)
Necessities	3.72 (0.99)	3.73 (1.10)	3.56 (1.07)	3.91 (0.99)	3.55 (1.03)	3.92 (0.90)	3.57 (1.11)	3.91 (1.07)
Education	3.97 (0.87)	3.99 (0.84)	3.94 (0.83)	4.03 (0.88)	3.93 (0.87)	4.03 (0.87)	3.96 (0.79)	4.02 (0.90)
Job Security	3.56 (1.12)	3.29 (1.21)	3.00 (1.13)	3.69 (1.10)	3.04 (1.09)	3.74 (1.04)	2.96 (1.18)	3.65 (1.15)
Healthcare	3.62 (1.02)	3.79 (1.03)	3.56 (1.02)	3.88 (1.01)	3.43 (1.02)	3.86 (0.98)	3.68 (1.01)	3.90 (1.03)
Police	3.22 (1.11)	3.88 (1.08)	3.47 (1.14)	3.66 (1.14)	3.05 (1.04)	3.43 (1.15)	3.90 (1.08)	3.87 (1.09)
Government	3.50 (1.05)	3.50 (1.12)	3.36 (1.07)	3.66 (1.09)	3.37 (1.03)	3.65 (1.06)	3.36 (1.11)	3.66 (1.11)

Note. Bolded pairs represent significant differences between groups (p < .05).

Linear Regression Models

Two linear regression models were used to test H1 and H2, with separate models for continuity and change resilience (see Table 5). H1 predicted that the eight resources would be positively related to continuity resilience, meaning resilience communication processes typically utilized in maintaining important aspects of pre-hardship life. Continuity resilience was positively predicted by family support, friend support, education quality, job security, and government representation, and negatively predicted by necessities. H2 predicted that the eight resources would be positively related to change resilience, meaning resilience communication processes typically utilized in adapting to hardship-specific features. Change resilience was positively predicted by family support, friend support, education, job security, and government representation and negatively predicted by police support.

Table 5.

Linear Regression Models

	Continuity		Change
	Beta	R2	Beta	R2
Model		.36***		.35***
Family Support	.17**		.14**
Friend Support	.19**		.24***
Necessities	−.09*		−07
Education	.19***		.14***
Job Security	.17***		.21***
Healthcare	.05		−.01
Police Support	−.02		−.14***
Government	.13**		.17***

^***p < .001,

^**p < .01,

^*p < .05 α

Moderation

To test H3–5, differences in associations between variables by racial identity and gender were assessed using Hayes PROCESS in SPSS 27. H3 and H4 were tested with model 1, and H5 was tested using model 3. Only significant interactions are reported (see Tables 6–7).Regarding H3, three differences emerged based on racial identity. Government representation positively predicted continuity resilience only for Black participants. Education quality positively predicted change resilience only for Black participants. Job security positively predicted change resilience for both racial identity groups, but the association was stronger for White participants.

Table 6.

Differences in Predictors by Race and Gender Separately

	R2	MSE	Coef.	Effect	SE	t
Government → Continuity	.37***	.35	.09*
Black Participants				.13***	.04	3.44
White Participants				.04	.04	1.01
Education → Change	.36***	.51	.15*
Black Participants				.22***	.06	4.00
White Participants				.07	.06	1.20
Job Security → Change	.36***	.51	−.14**
Black Participants				.08*	.04	2.08
White Participants				.22***	.04	5.82
Police Support → Change	.35***	.52	−.10*
Female Participants				−.15	.04	−3.87***
Male Participants				−.05	.04	−1.13

Note. Coef indicates the coefficient for the two-way interaction between the predictor and either race or gender. Only significant interactions are reported.

^***p < .001,

^**p < .01,

^*p < .05

Table 7.

Differences in Predictors at the Intersection of Race and Gender

	R2	MSE	Coef.	Effect	SE	t
Necessities → Continuity	.39***	.35	.30**
Black Women				.02	.05	0.31
Black Men				−.17**	.06	−2.61
White Women				−.15**	.05	−3.21
White Men				−.03	.05	−0.63
Healthcare → Continuity	.39***	.35	.27**
Black Women				.14**	.05	3.00
Black Men				.01	.06	0.10
White Women				−.07	.05	1.38
White Men				.05	.05	0.99
Government → Continuity	.39***	.35	.20*
Black Women				.17***	.05	3.65
Black Men				.08	.05	1.55
White Women				−.03	.05	−0.63
White Men				.08	.05	1.64
Family Support → Change	.37***	.51	.22**
Black Women				.12***	.04	3.32
Black Men				.04	.05	0.88
White Women				.02	.04	0.48
White Men				.15**	.05	2.93

Note. Coef indicates the coefficient for the three-way interaction between the predictor, race, and gender. Only significant interactions are reported.

^***p < .001,

^**p < .01,

^*p < .05

Regarding H4, only one gender difference emerged. Police support negatively predicted change resilience only for women.

Regarding H5, four differences emerged at the intersection of race and gender. Necessities negatively predicted continuity resilience for Black men and White women. Healthcare and government representation positively predicted continuity resilience for Black women only. Family support positively predicted change resilience for Black women and White men.

Discussion

The current study furthers the communication theory of resilience (CTR) by both providing support for the association between resource access and resilience communication and examining associations at the intersection of race and gender. That resource access affects resilience communication is a key premise of CTR (Buzzanell, 2018a). This study provides empirical backing for this claim and illustrates how different types of resources are related to resilience communication. Importantly, findings demonstrate that these associations differ by race and gender. From a theoretical standpoint, this clarifies that race and gender serve as boundary conditions for associations between CTR variables.

Scharp et al. (2021) refer to privilege “as a silent partner to [communicative resilience] processes” (p. 18). Resources like education quality, healthcare quality, and government representation often signal and derive from positions of privilege (Bailey et al., 2017; Lukachko et al., 2014). In the current study, not only were such resources predictive of resilience enactment, but they were particularly predictive for individuals who typically hold less societal power. Social and material resources were likely necessary to counterbalance discriminatory societal and interpersonal barriers to resilience.

Key Resources Across Groups

Two resources positively predicted continuity and change resilience for all participant groups: friend support and job security. Additionally, education quality was positively predictive of continuity resilience for all participants and predictive of change resilience for Black participants. These findings further CTR by identifying key resources for enacting resilience across race, gender, or continuity/change. Friend support, job security, and education quality are important for both enabling resilience and accessing additional resources (Dorrance Hall & Scharp, 2021; Frankenberg et al., 2013; LaGree et al., 2021). It is important to note that the relationship between job security and change resilience was stronger for White participants, suggesting that they relied more heavily on the stability provided by work to adapt to the pandemic.

Negative Associations Between Resources and Resilience

Surprisingly, access to necessities and police support were negatively predictive of resilience. Recognizing that lacking a resource can catalyze resilience is important for theorizing. Resources are believed to facilitate resilience. Yet, police support was negatively related to change resilience for women, meaning that women with less police support were more likely to engage in change resilience. Black women reported significantly lower levels of police support. Black women’s experiences of sexual and physical violence from police could explain this difference (Crenshaw et al., 2015). Study data were collected in September 2020, following protests regarding policing. A lack of meaningful change may have prompted reframing or sensemaking. When tangible action or change seems unlikely, sensemaking is a key method for coping with hardship (Horstman, 2019). Additionally, policing issues related to COVID-19 mandates may have necessitated creation of new routines (Dunbar & Jones, 2021).

Access to necessities was negatively related to continuity resilience for Black men and White women, meaning that individuals from these groups engaged in more continuity resilience when they experienced housing and food insecurity. Additional analyses indicate that lack of necessities was specifically related to maintaining routines. The maintain routines subscale asks about trying to keep life normal and making an effort to keep routines. Those who are housing and food insecure may have to work harder to maintain normalcy. In an analysis of Syrian refugee narratives, refugees depicted maintaining normalcy as a substantial challenge, often because of housing insecurity (Sánchez Sánchez & Lillie, 2019). Those in the current study may be experiencing a similar struggle. An additional question is why the association holds only for Black men and White women. Further research is needed using qualitative methods to probe this finding.

Differences at the Intersection of Race and Gender

Findings illustrate the importance of examining intersectionality in resilience scholarship. Four differences were found at the intersection of race and gender, only one of which was reflected in analyses examining race and gender separately. In addition to the negative association between change resilience and access to resources discussed above, associations with healthcare quality, government representation, and family support varied at the intersection.

Black women currently and historically have faced greater oppression in healthcare than other groups examined in this study (Few-Demo et al., 2018; Sims, 2010). For example, Black women have reported being stereotyped in medical settings and experiencing differences in medical treatment (Okoro et al., 2020; Sims, 2010). These realities make it unsurprising that healthcare quality was specifically predictive of continuity resilience for Black women. Discrimination in healthcare limits Black women’s ability rely on the healthcare system if they contract COVID-19, likely restricting the routines Black women establish during a health crisis.

Findings initially indicated that government representation predicted continuity resilience for Black participants. Analyses at the intersection identified that this relationship was specific to Black women, again highlighting the importance of examining multiple identities in resilience research. Black women have been and continue to be underrepresented in government. For example, out of the 535 current members of the U.S. Congress, only 23 (4.3%) are Black women (CAWP, 2021). Due to underrepresentation, government policies are less likely to reflect the needs of Black women, potentially inhibiting their ability to maintain normalcy.

Family support was positively related to continuity resilience for all groups and to change resilience for Black women and White men. Although the differences between these two populations are abundant, both men and Black women are often cast as heads of the family and expected to be “strong,” potentially at the detriment of their own mental health (Acker, 1990; Collins, 2002; Davis, 2015). White men have one of the highest suicide rates reported (Curtin & Hedegaard, 2019). Black women are often burdened by stereotypes that paint them “as super-strong, selfless, and super-human” leaving no room for vulnerability (Jones, 2021, p. 4). Having a family that provides emotional support could enable Black women and White men to feel comfortable accepting change and adapting in positive ways rather than being “strong.” It is important to note that family support may influence change resilience differently for Black and White families. The cultural variant perspective articulates how Black family life is culturally unique from White family life (Allen, 1978; Taylor et al., 2021) – different organizational structures characterize each family (Sarkisian & Gerstel, 2004).

Practical Implications

Practically, interventions targeted to improve resilience should be aware of three key factors. First, intervention recipients’ ability to enact resilience processes is dependent upon their access to social and material resources. Finding avenues for increasing relevant resources should be done in tandem with resilience communication skill-building. Second, which resources matter can differ at the intersection of race and gender. In the current study, Black women’s continuity resilience was facilitated by healthcare quality and government representation, among other resources. Programs and social movements designed to increase these resources are, therefore, vital moving forward— including providing support for Black women entering politics and/or the medical field. Third, finding that Black women were more resilient in the absence of police support should prompt us to shift societal understanding about the role of policing during pandemics and crises more broadly.

Limitations

Although the current study advances CTR, a few limitations need to be addressed. First, the study would not have captured those with extremely low levels of access because data were collected online. Examining communities without internet access, a material resource, has the potential to provide theoretical advances. Second, this study examines two racial groups as distinct and does not account for nuance in biracial participants. Additionally, the presentation of the Black community as a monolith erases possible distinctions between various subgroups (e.g., Black immigrants, socio-economic variances; Taylor et al., 2021). Thirdly, it is possible that processes like communication networks could contribute to both continuity and change. Future research should utilize qualitative methods to further establish how the processes facilitate continuity and change. Finally, gender is examined as a binary. Although this decision was made because the statistical tests performed required equivalent participant numbers in each gender group, it does not account for non-binary individuals who may have faced additional healthcare-specific hardships, among others, during the pandemic (see HRC Staff, 2020).

Conclusion

This study identifies the influence that access to material and social resources has on the enactment of change and continuity resilience. The COVID-19 pandemic, compounded with already existing inequalities (i.e., in the healthcare system, policing, etc.), created unique experiences at the intersections of race and gender. Our findings provide support for the link theorized by Buzzanell (2010; 2019) and Wilson et al. (2021) that proposes that resilience may be dependent on access to material and social resources—furthering our understanding about how resilience processes may be more easily enacted by individuals with the most privilege.