Racial Discrimination, Mental Health, and Parenting Among African American Mothers of Preschool-Aged Children

Abstract

Objective:

Indirect exposure to racism experienced by a caregiver (ie, vicarious racism) is associated with poor outcomes for children, but mechanisms of vicarious racism transmission are poorly understood. The purpose of this study is to examine the relationship between experiences of racial discrimination and parenting among African American mothers, and to identify psychological mediators and moderators of this relationship.

Method:

African American mothers (N=250) with young children ($\overline{\mathrm{x}}=3.7$ years-old) reported on perceived racial discrimination (Race-Related Events Scale), parenting (Parenting Stress Index, Parenting Styles and Dimensions Questionnaire), coping (Coping Strategies Index), and mental health (Stress Overload Scale, Beck Depression Inventory). We used multivariable linear regression to examine associations between perceived racial discrimination and parenting, and to test coping as a moderator of these relationships. We used ordinary least squares regression- based path analysis with bootstrapping to examine mediation by stress overload and depressive symptoms.

Results:

Fifty-seven percent of women reported at least one experience of racial discrimination. Experiences of racial discrimination were associated with increased parenting stress (β=0.69, p=.02), and this relationship was mediated by stress overload (95% CI 0.35, 1.09) and depressive symptoms (95% CI 0.27, 1.18). Racial discrimination was not associated with parenting styles, and coping strategies largely did not moderate the relationships examined.

Conclusion:

Racial discrimination has harmful intergenerational effects on African American children and families. Systemic-level interventions are needed, including adoption of policies to promote racial justice, and eliminate structural racism in the United States. Future research on coping strategies specific to racism-related stress are needed to inform approaches to intervention.

Introduction

Racism is a significant stressor for African American families and a root cause of racial health inequities.^1,2 Structural racism describes the macrosystem-level social forces and institutions that generate and reinforce inequities through racist laws, policies, and social norms,^2,3 and also perpetuate interpersonal racism, or discriminatory actions between individuals.^3,4 Together, both structural and interpersonal racism contribute to racism-related stress among African American families, and increase risk for poor physical, psychological, and social life-course outcomes.^1,5 Further, emerging evidence suggests that racism-related stress has harmful intergenerational effects.⁶ Vicarious racism, or indirect exposure to racism experienced by others, such as a caregiver, is associated with depressive symptoms, behavior problems, and other poor socioemotional and mental health outcomes in children.⁶ To inform interventions and improve outcomes for families who bear a higher burden of stress due to structural racism and social inequality, it is critical to understand the mechanisms underlying intergenerational transmission of racism-related stress.

One mechanism through which vicarious racism may affect children is through influences on caregivers’ parenting.^6,7 In a recent study of African American families with school-age children, the relationship between parental experiences of racial discrimination and child emotional problems was mediated by undesirable parenting practices (e.g. over-reactivity, laxness).⁸ Parental experiences of racial discrimination have also been associated with less competence-promoting parenting among African American families.⁹ Other studies demonstrate that parenting may protect against the intergenerational transmission of racism-related stress. For example, Varner and colleagues found parents’ experiences of racial discrimination were associated with involved-vigilant parenting (i.e. parenting characterized by responsiveness and monitoring),¹⁰ and that this parenting moderated the relationship between adolescents’ experiences of racial discrimination and their psychological wellbeing.¹¹ Supportive parenting has also been shown to buffer the effects of experiencing racial discrimination on violent behaviors¹² and depressive symptoms¹³ among African American adolescents.

This small but compelling body of research suggests parenting may contribute to or protect against the intergenerational transmission of racism-related stress among African American families. However, very little is known about the influence of racial discrimination on parenting during the early childhood period (<5 years). This is a significant gap in knowledge, as young children are highly sensitive to environmental stressors due to rapid brain growth and development during this period.¹⁴ Young children may also be especially vulnerable to the effects of vicarious racism, as they are highly dependent on caregivers who may be experiencing racism-related stress.¹⁵ Given this critical gap in knowledge and the importance of early childhood for healthy brain development, empirical research is needed to understand how caregiver experiences of racial discrimination influence parenting during this sensitive period.

There is also a pressing need to understand the psychological mechanisms that mediate or moderate associations between caregiver experiences of racial discrimination and parenting. Improved understanding of these mechanisms has important implications for informing targeted, strengths-based approaches to intervention, including at the individual, family, and systemic levels. Previous research with African American families suggests that depressive symptoms may mediate relationships between experiences of racial discrimination and parenting.^8,9 Other studies of African American families also implicate maternal mental health; in a study of mothers with school-aged children, maternal depressive symptoms mediated the relationship between mothers’ perceived racial discrimination and child externalizing behaviors,¹⁶ and among mothers with 10 and 11 year-old children, an association between maternal psychological distress and poor parent-child relationship quality was stronger for mothers who experienced higher levels racial discrimination.¹⁷ Another mechanism that may mediate the relationship between caregiver experiences of racial discrimination and parenting is stress overload, or excessive stress that exceeds available personal and environmental resources.¹⁸ Stress is known to place strain on the parent-child relationship and increase risk for harsh parenting,¹⁹ but the role of stress overload in the intergenerational transmission of racism has not been well studied.

Finally, much research has been conducted on racial socialization and other coping strategies used by African American parents to prepare their children for experiencing racial bias.²⁰ However, less is known about how parents’ own coping mechanisms may help buffer the effects of racism-related stress on parenting practices. In a recent study of African American fathers with sons, Cooper and colleagues found that fathers with high levels of coping across multiple dimensions (e.g. seeking emotional and instrumental support, using direct problemsolving strategies) reported both higher experiences of racial discrimination and higher parenting involvement with their sons.²¹ Fathers with avoidant coping strategies (e.g. avoiding thinking about a problem), however, reported less engaged and involved parenting.²¹ Individual coping strategies have been associated with lower parenting stress and more supportive parent-child relationships in other studies of African American families.^22,23 However, to our knowledge, no researchers have explored whether coping moderates relationships between racial discrimination and parenting among African American mothers with young children.

Conceptual Framework and Study Aims

In the current study, we address the critical scientific gaps outlined above by examining data from a unique cohort of African American mothers with preschool-aged children. Our approach is informed by Murry’s “Integrative Model for the Study of Stress in Black American Families,” a comprehensive framework that describes how structural and interpersonal environmental stressors influence relationships, including mental health and parenting, but also may be moderated by strengths-based coping.⁷

Our first aim is to examine relationships between mothers’ perceived racial discrimination and two parenting constructs: parenting stress and parenting styles. Parenting stress is associated with child behavior problems, particularly externalizing behaviors (e.g. hyperactivity, aggression).²⁴ Parenting styles include authoritative parenting, which is characterized by emotional support, responsiveness, and firm limit-setting, authoritarian parenting, characterized by punitive discipline strategies, and permissive parenting, in which the parent places few demands on the child.^25,26 Authoritative parenting is generally viewed as a more favorable parenting style compared to the others, and is associated with adaptive social behaviors and fewer externalizing behaviors in young children.^25,27 We hypothesized that mothers’ lifetime perceived racial discrimination would be associated with increased parenting stress, authoritarian parenting, and permissive parenting, and decreased authoritative parenting.

Our second aim is to examine mediators and moderators of the relationship between mothers’ perceived racial discrimination and current parenting. We hypothesized that relationships between perceived racial discrimination and parenting (stress and styles) would be mitigated by positive coping strategies (problem solving, seeking social support), and exacerbated by avoidant coping strategies. We also hypothesized that relationships between mothers’ lifetime perceived racial discrimination and current parenting stress would be mediated by past-week maternal stress overload and depressive symptoms. As parenting styles are unlikely to change based on short-term fluctuations in stress or depressive symptoms, we did not develop or test a mediation hypothesis for the parenting styles outcomes.

Method

We conducted a secondary data analysis of mothers and children enrolled in InterGEN, a longitudinal cohort study examining the interaction of genes and psychological environmental variables on blood pressure in Black/African American mother-child dyads. A community-based sample of mother-child dyads (N=250) were recruited though partnerships with early care and education centers, which provide preschool education for socioeconomically disadvantaged children in ethnically and racially diverse urban communities.²⁸ Dyads were recruited from 2014 to 2019. Eligibility criteria included mothers who: (1) were at least 21 years of age, (2) self-reported identification as African American or Black, (3) spoke English, (4) did not have a cognitive/psychiatric disorder that could limit accuracy of reporting, and (5) enrolled with a biological child who was 3–5 years old. The study was approved by the appropriate Institutional Review Boards at participating institutions.

Women and children enrolled in InterGEN completed four study visits, each approximately six months apart, over the span of 18–24 months. Data used in the present cross-sectional analysis were collected during the baseline (Time 1) visit. We also conducted sensitivity analyses using data collected at 6 months post-baseline (Time 2). Demographic and psychological data were collected from mothers using Audio Computer-Assisted Self-Interview software (ACASI, version 16). Full study methods and procedures, including recruitment strategies and instrument psychometrics, have been previously described.^28,29

Variables and Measures

Exposure Variable

Perceived Racial Discrimination.

Our primary exposure was mothers’ lifetime experiences of perceived racial discrimination, which was measured using the Race- Related Events Scale (RES).³⁰ The RES is a 23-item instrument in which participants report whether they have experienced stressful and potentially traumatizing events because of their race or ethnicity. Examples include being told to leave a place and not come back, being harassed by police, and seeing someone of the same race being injured or killed because of their race or ethnicity. Participants respond whether they have ever experienced each event with a yes (1) or no (0). Scores are summed (0–22) and higher scores indicate greater experiences with race-based discrimination. The RES has adequate test-retest reliability (r = .66) and reliability was high (α = .93) in our sample.

Outcome Variables

Parenting Stress.

Parenting stress was measured using the Parenting Stress Index— Short Form (PSI-SF).³¹ The PSI evaluates current stress in the parenting role and parent-child relationship, and is appropriate for parents of children from birth to 12 years. The PSI-SF is a 36- item Likert scale with high reliability in our sample (α = .94). We used the PSI total score, which ranges from 36 to 180, with higher scores indicating more parenting stress.

Parenting Styles.

Parenting styles were measured using the Parenting Styles and Dimension Questionnaire (PSDQ).²⁶ The PSDQ is a 59-item Likert scale measuring self-reported current parenting style. The PSDQ has three subscales with good reliability in our sample: authoritative parenting (26 items; α = .91), authoritarian parenting (18 items; α = .83), and permissive parenting (15 items; α = .72). Higher scores on each subscale indicate a greater affinity for that parenting style.

Mediating Variables

Stress Overload.

Maternal stress was measured using the Stress Overload Scale (SOS), a 24-item instrument measuring subjective stress overload.¹⁸ The SOS includes two subscales: personal vulnerability and event load. The personal vulnerability subscale covers feelings of powerlessness, inadequacy, frailty, and debility, and event load refers to the burden of outside demands, responsibilities and pressures. Participants report how they felt during the past week using a 5-point Likert Scale ranging from “not at all” to “a lot.” In the current analysis, we used the SOS total score, which includes both subscales. Reliability was high in our sample (α =.96).

Depressive Symptoms.

Maternal depressive symptoms over the past week were measured using the Beck Depression Inventory (BDI).³² The BDI consists of 21 items scored based on severity of each symptom (low of 0, high of 3). A total depression score is calculated by summing the items. Any score greater than 16 indicates some clinical depression, with scores over 30 indicating severe depression. Reliability was high in our sample (α =.91).

Moderating Variables

Coping Strategies.

Coping was measured using the 33-item Coping Strategy Indicator (CSI).³³ For this instrument, respondents are first instructed to think about a problem that they have encountered during the last six months. For each item on the instrument, respondents then indicate how much they have used that coping strategy on a Likert scale ranging from “a little” to “a lot.” Scores are summed and used to create three subscales, which had high reliability in our sample: problem solving (α = .93), seeking social support (α = .91), and avoidance (α = .84). Subscale scores are used to create categorical variables to indicate the extent to which respondents engage in each coping strategy, ranging from very low to very high.³³

Covariates

Covariates were selected via a priori knowledge and included maternal age, child age, child gender, maternal education level, and maternal income level. We also controlled for use of services that support parenting, as engagement in such services may affect our outcomes of interest. We created a parenting services variable, coded as “1” for families who responded “yes” to receiving any of the following services: early intervention program (e.g. Birth to Three), family resource support center, home visiting program, prenatal/newborn program, or parenting program. Families responded “no” to these services were coded as “0” for this variable.

Statistical Analysis

Statistical analyses were completed using Version 9.4 of the SAS System for Windows (Cary, NC), except for path analyses with bootstrapping, which were conducted using the Preacher and Hayes PROCESS macro in SPSS ³⁴ Descriptive analyses were carried out to assess normality of variables and missing data. Predictor variables were centered at the mean to aid interpretation of parameter estimates. Missing data across all variables ranged from 0.8% (maternal education, use of parenting services) to 6.4% (SOS score). Perceived racial discrimination (RES) scores were missing for 3.2% (n=8) of participants. We used listwise deletion to handle missing data.

Because most women in the sample reported very few experiences of discrimination, we first conducted bivariate analyses by dichotomizing the RES scale responses into two groups: mothers who reported racial discrimination (RES score >0) and those who did not (RES score = 0). We also examined RES as a continuous variable in regression models. Chi-square tests and Mann-Whitney-U tests were used to compare demographic characteristics and all study variables (parenting stress, parenting styles, stress overload, depressive symptoms, and coping styles) between these two groups.

We used multivariable linear regression to examine associations between perceived racial discrimination (RES) and parenting (PSI, PSDQ). The RES scale score was treated as a continuous variable for these analyses and covariates were included as described above. All linear regression model assumptions were met. We then used ordinary least squares regression-based path analysis with bootstrapping to examine stress overload and depressive symptoms as mediators of the relationship between perceived racial discrimination and parenting stress. Bootstrapping is a nonparametric method that involves repeated sampling from the same dataset to estimate the size of the indirect effects, and allows for greater confidence in results for studies with relatively small sample sizes. ³⁵ We generated 5,000 bootstrap samples in this study, and thus 95% confidence intervals are based on the distribution of the 5,000 samples. The indirect effect is considered significant if the confidence interval does not include zero.

To test coping as a moderator of the relationship between perceived racial discrimination and parenting, we conducted multivariable linear regression as described above, and added coping (CSI) and an interaction term between coping and perceived racial discrimination (RES) to each model. We modeled each coping subscale (problem solving, seeking social support, avoidance) separately to examine independent effects.

Finally, to explore whether relationships changed over time, we conducted sensitivity analyses using data from Time 2 (6 months after baseline). We repeated the descriptive, bivariate, and multivariable regression analyses described above using cross-sectional data from Time 2. We also tested whether experiences of racial discrimination at Time 1 predicted parenting outcomes at Time 2.

Results

Study Sample

Children (n=250) in this sample were an average age of 3.7 years and 58% were female. Mothers ranged in age from 21 to 46 years, with a mean of 31.22 years. Though more than half of the participants had some college or more education (58.1%), nearly half (46.6%) reported an annual income of less than $15,000 and 77% reported receiving Medicaid or government/Affordable Care Act health insurance. Most families (70.6%) were living in a single parent household (i.e. not married or cohabitating). RES scores ranged from 0 to 21 with a mean of 3.44 (SD=4.54), and 138 (57%) women reported at least one experience of perceived racial discrimination. Women who reported experiencing racial discrimination were more likely to be older and report higher income and education levels than women who did not report experiencing discrimination, but no other differences in demographic characteristics were noted. See Table 1 for a full description of the sample.

Table 1.

Characteristics of Women and Children Enrolled in InterGEN Study

	Total Sample (N=250 dyads)	Maternal Experiences of Discrimination (n=138 dyads)	No Maternal Experiences of Discrimination (n=104 dyads)
	N (%)	n (%)	n (%)	p
Child characteristics
Gender
Female	146 (58.4)	88 (63.8)	53 (51.0)	.05
Male	104 (41.6)	50 (36.2)	51 (49.0)
Age
3 years	114 (45.6)	59 (42.8)	51 (49.5)	.38
4 years	97 (38.8)	59 (42.8)	35 (34.0)
5 years	39 (15.6)	20 (14.5)	17 (16.5)
Maternal characteristics
Age
20–29 years	105 (42.0)	45 (32.6)	56 (53.9)	.003
30–39 years	124 (49.6)	82 (59.4)	40 (38.5)
40–49 years	21 (8.4)	11 (8.0)	8 (3.3)
Education
Less than high school	13 (5.2)	4 (2.9)	9 (8.7)	.001
High school graduate	91 (36.7)	39 (28.3)	50 (48.1)
Some college	82 (33.1)	52 (37.7)	28 (26.9)
Associate’s degree or higher	62 (25.0)	43 (31.2)	17 (16.4)
Annual household income
< US$15,000	111 (46.4)	48 (35.6)	63 (63.6)	<.0001
US $15,000–50,000	102 (42.9)	68 (50.4)	32 (32.3)
> US$50,000	25 (10.5)	19 (14.1)	4 (4.0)
Enrolled in parenting services
Yes	35 (14.1)	22 (15.9)	13 (12.5)	.45
No	213 (85.9)	116 (84.1)	91 (87.5)
Current smoker
Yes	54 (21.9)	27 (19.6)	27 (26.2)	.22
No	193 (78.1)	111 (80.4)	76 (73.8)
Marital status
Single/ Divorced/ Separated	175 (70.6)	98 (71.0)	75 (72.1)	.85
Married/ cohabitating	73 (29.4)	40 (29.0)	29 (27.9)

Note: Chi-square tests used to compare mothers who reported experiencing discrimination versus mothers who reported no experiences of discrimination.

We describe bivariate associations between study variables and perceived racial discrimination in Table 2. There were no differences in parenting stress or parenting styles between women who reported experiencing racial discrimination and those who did not. However, women who experienced racial discrimination reported significantly higher levels of stress overload (p=.0001) and higher depressive symptoms (p=.006). Use of problem solving coping differed between women who reported experiencing racial discrimination and those who did not (p = .003), but there were no differences in seeking social support (p=.25) or avoidance (p=20).

Table 2.

Bivariate Associations Between Study Variables and Maternal Experiences of Discrimination

	Total Sample	≥1 Experiences of Discrimination	No reported Experiences of Discrimination
	Mean (SD)	Mean (SD)	Mean (SD)	p
Outcome Variablesa
Parenting Stress	69.4 (20.2)	70.1 (19.9)	68.0 (20.6)	.51
Authoritative Parenting	4.04 (0,64)	4.14 (0.45)	3.96 (0.73)	.12
Authoritarian Parenting	1.85 (0.50)	1.86 (0.44)	1.87 (0.55)	.74
Permissive Parenting	1.95 (0.49)	1.95 (0.43)	1.99 (0.50)	.54
Mediating Variablesa
Stress Overload	60.1 (24.0)	65.1 (23.4)	53.1 (23.3)	.0001
Depressive Symptoms	6.41 (7.32)	7.48 (8.13)	4.79 (5.73)	.006
Moderating Variablesb	N (%)	n (%)	n (%)
Problem Solving
Very Low or Low	41 (17.5)	14 (10.4)	26 (27.1)	.003
Average	111 (47.2)	72 (53.3)	37 (38.5)
High	83 (35.3)	49 (36.3)	33 (34.4)
Seeking Social Support
Very Low or Low	71 (28.4)	32 (23.2)	34 (32.7)	.25
Average	125 (50.0)	75 (54.4)	48 (46.2)
High	54 (21.6)	31 (22.5)	22 (21.2)
Avoidance
Low	47 (19.9)	22 (16.2)	24 (25.0)	.20
Average	129 (54.7)	80 (58.8)	47 (49.0)
High or Very High	60 (25.4)	34 (25.0)	25 (26.0)

Note:

^aComparisons conducted using Mann-Whitney-U tests.

^bComparisons conducted using Chi-square tests.

Associations between Perceived Racial Discrimination and Parenting

In Table 3, we report adjusted associations between mothers’ perceived racial discrimination and parenting outcomes. Perceived racial discrimination was associated with an increase in parenting stress (β=0.69, p=.02), but was not associated with authoritative, authoritarian, or permissive parenting styles.

Table 3.

Adjusted Associations Between Maternal Experiences of Racial Discrimination and Parenting

	β (SE)	95% CI	p
Parenting Stress	0.69 (0.29)	0.11, 1.27	.02
Authoritative Parenting	0.01 (0.01)	−0.01, 0.02	.30
Authoritarian Parenting	0.01 (0.01)	−0.01, 0.02	.36
Permissive Parenting	0.00 (0.01)	−0.01, 0.02	0.57

Note: Models conducted using linear regression (n=231). Adjusted for maternal age, child age, child sex, maternal education level, income level, and use of parenting services. SE = Standard Error.

Mediation by Stress Overload and Depressive Symptoms

Based on our linear regression findings, we tested stress overload and depressive symptoms as mediators of the relationship between perceived racial discrimination and parenting stress (Table 4). In the model testing mediation by stress overload, perceived racial discrimination was directly associated with increased stress overload (β=1.65, p<.0001), and stress overload was directly associated with increased parenting stress (β=0.43, p<.0001). The direct association between perceived discrimination and parenting stress was no longer statistically significant in the mediation model (p=.91), but the indirect pathway was significant (β=0.71, 95% CI 0.35, 1.09), indicating that stress overload fully mediates the relationship between perceived racial discrimination and parenting stress. We found similar results in our model testing mediation by depressive symptoms. Perceived racial discrimination was directly associated with increased depressive symptoms (β=0.52, p<.0001), and depressive symptoms were directly associated with increased parenting stress (β=1.30, p<.0001). The direct pathway between perceived racial discrimination and parenting stress was not statistically significant in our mediation model (p=.71), but the indirect pathway was significant (β=0.68, 95% CI 0.27, 1.18), indicating complete mediation by depressive symptoms.

Table 4.

Mediating Effects of Stress Overload and Depression on the Association Between Maternal Experiences of Racial Discrimination and Parenting Stress

	Stress overload		Depressive symptoms
	β (SE)	95% CI	β (SE)	95% CI
Direct Effects
Discrimination → mediator	1.65 (0.34)	0.98, 2.31*	0.52 (0.11)	0.31, 0.73*
Mediator → parenting stress	0.43 (0.05)	0.33, 0.53*	1.30 (0.17)	0.95, 1.63*
Discrimination → parenting stress	0.09 (0.27)	−0.51, 0.57	0.11 (0.27)	−0.44, 0.65
Indirect effect
Discrimination → mediator → parenting stress	0.71 (0.19)	0.35, 1.09*	0.68 (0.23)	0.27, 1.18*

Note: Both models adjusted for maternal age, child age, child sex, maternal education level, income level, and use of parenting services; SE = Standard Error.

^*indicates statistically significant effect because the confidence interval does not cross zero; mediation tested using 5,000 bootstrap samples; n=225 for the Stress Overload model; n=214 for the depressive symptoms model.

Moderation by Coping Strategies

In Table 5, we present the adjusted associations between perceived racial discrimination and parenting outcomes with moderation by coping strategies. Problem solving moderated the association between experiences of racial discrimination and authoritarian parenting (p=.03), but not in the hypothesized direction; experiences of racial discrimination were associated with increased authoritarian parenting among mothers reporting high levels of problem solving (β =.03, p=.02), and associated with decreased authoritarian parenting among mothers reporting very low to low (β = −.01,p=.56) and average (β = −.01,p=.26) problem solving. Problem solving did not moderate the association between experiences of racial discrimination and the other parenting outcomes.

Table 5.

Moderating Effect of Coping Strategies on the Association between Maternal Experiences of Racial Discrimination and Parenting

	Parenting stress		Authoritative parenting		Authoritarian parenting		Permissive parenting
	β (SE)	p	β (SE)	p	β (SE)	p	β (SE)	p
Coping model 1: problem solving
Racial discrimination	0.01(0.04)	.95	−0.01 (0.02)	.96	−0.35 (0.02)	.07	0.19 (0.09)	.33
Problem solving coping	−.22 (1.85)	.004	0.23 (0.05)	.002	0.01 (0.04)	.93	−0.07 (0.04)	.36
Discrimination*coping interaction	0.37 (0.34)	.21	0.02 (0.01)	.93	0.43 (0.01)	.03	−0.17 (0.01)	.41
Coping model 2: social support
Racial discrimination	0.07 (0.68)	.64	−0.05 (0.02)	.75	0.11 (0.02)	.48	0.15 (0.02)	.34
Social support coping	−0.22 (1.74)	.004	0.21 (0.05)	.01	0.08 (0.04)	.32	0.00 (0.04)	.96
Discrimination*coping interaction	0.13 (0.31)	.43	0.09 (0.01)	.58	−0.07 (0.01)	.67	−0.14 (0.01)	.41
Coping model 3: avoidance
Racial discrimination	−0.05 (0.52)	.69	−0.14 (0.01)	.24	−0.16 (0.01)	.19	0.11 (0.01)	.38
Avoidance coping	0.10 (2.13)	.23	−0.07 (0.06)	.40	0.22 (0.05)	.01	0.21 (0.05)	.02
Discrimination*coping interaction	0.24 (0.31)	.08	0.24 (0.01)	.08	0.08 (0.01)	.19	−0.17 (0.01)	.22

Note: Models conducted using linear regression and adjusted for maternal age, child age, child sex, maternal education level, income level, and use of parenting services; n=225 dyads for the problem solving model; n=226 dyads for the social support and avoidance models

Social support was associated with significantly lower parenting stress (β= −0.22, p=.004) and significantly higher authoritative parenting (β=0.21, p=.01), but did not moderate the relationship between experiences of racial discrimination and any of the parenting outcomes. Avoidance was associated with significantly increased authoritarian (β=0.22,p=.01) and permissive (β=0.21, p=.02) parenting, but did not moderate the relationship between experiences of racial discrimination and any parenting outcomes.

Sensitivity Analyses

We found that results at Time 2 were largely unchanged from Time 1. Mothers’ lifetime experiences of racial discrimination reported at Time 2 were associated with increased parenting stress (β=0.19, p=.01) at Time 2, but not with parenting styles. The relationship between racial discrimination and parenting stress was mediated by stress overload and depressive symptoms, but not moderated by coping strategies (results not shown). In our longitudinal sensitivity analyses, we found that reported racial discrimination at Time 1 was associated with increased parenting stress at Time 2 (β=0.95, p=.003), but not associated with parenting styles. Mediation and moderation results were consistent with our cross-sectional findings.

Discussion

The results of this study contribute to a critical and growing body of literature highlighting the harmful intergenerational effects of racial discrimination, particularly for Black and African American families living in the United States. We found that lifetime experiences of perceived racial discrimination were associated with a small increase in parenting stress among African American women with preschool-aged children, and that this relationship was mediated by increased stress overload and depressive symptoms during the past week. We did not support our hypothesis that maternal experiences of racial discrimination were associated with parenting styles, and results of this study suggest that individual coping strategies may not mitigate the harmful effects of racial discrimination on parenting. Policies to promote racial justice and tailored interventions to support mental health and reduce racism-related stress are necessary to combat the harmful intergenerational effects of structural and interpersonal racism.

Our findings are consistent with previous studies conducted with parents of older children that demonstrate maternal mental health mediates associations between racial discrimination and poor parenting outcomes.^8,9 We also report the novel finding that perceived racial discrimination is associated with increased parenting stress among African American mothers with young children. Parenting stress describes a negative psychological response to the obligations of being a parent, and young children may be particularly vulnerable to parental distress and dysfunctional parent-child interactions.^24,31 Parenting stress is also associated with child behavior and mood disorders across the childhood period.²⁴ Thus, by placing stress on the parent child system, our results suggest that stress overload, depressive symptoms, and parenting stress may be important mechanisms underlying transmission of vicarious racism.

Effect sizes in this study were small, including the effect of perceived racial discrimination on parenting stress, suggesting these results may be of limited clinical significance. However, the small effect sizes detected may be in part due to low rates of reported racial discrimination among women in the InterGEN sample. High levels of self-reported racial discrimination are often not observed in quantitative studies of clinical samples,^8–10 potentially due to social desirability bias, limitations of self-report measures, or other factors. Future research using qualitative or mixed-methods approaches may provide more in-depth and nuanced insight into the effects of racism-related stress on parenting outcomes.

Healthy childhood development requires an environment that is safe, emotionally supportive, and developmentally stimulating, and as such the early childhood period places unique and high demands on parenting.³⁶ While additional research on families with young children is needed, consistent findings related to the mediating effects of stress and depressive symptoms provide important insight into directions for family and individual level intervention. Stress and depression are known to have powerful effects on parenting and parent-child relationships, including harsh parenting practices and less sensitive parenting.^37,38 Maternal depression is also associated with poor child outcomes,³⁹ including child psychopathology.⁴⁰ Thus, existing interventions aimed at reducing maternal stress and preventing or treating maternal depression may be an important approach to mitigating the harmful effects of perceived racial discrimination on parenting, particularly those specifically developed for and tested in Black women and families.^41,42

In the United States, intervention at the systemic level is of utmost importance, and this includes eliminating the structural racism and historically racist policies that are the root causes of racial discrimination experienced by people of color.⁵ While we examined interpersonal experiences of racial discrimination in this study, other harmful effects of structural racism, such as mass incarceration and criminalization of youth of color, may also justifiably contribute to increased parenting stress.^2,3,43 Thus, while family and individual level interventions may help to buffer the harmful effects of racial discrimination on parenting and maternal mental health, upstream level approaches to reducing racism and racial discrimination can and should be prioritized. This includes adoption of policies that may reduce maternal stress or improve access to mental health services, such as Medicaid expansion or universal healthcare, universal childcare, paid family leave, and a living wage.⁴⁴

Among women in our study, we did not support our hypothesis that individual coping strategies would mitigate the harmful effects of racial discrimination on parenting. While effects were small, our findings suggest that individual coping strategies may be beneficial (in the case of seeking social support) or detrimental (in the case of avoidance) to parenting stress and styles, but they are inadequate for addressing the harmful effects of racial discrimination on parenting. Future research should examine coping strategies that may specifically buffer racism-related stress, including racial pride and identity, religious practices, or specific sources of social support, such as co-parenting support.^15,45 Racial socialization, a process through which caregivers convey information about race and prepare their children to cope with racial barriers, has been proposed as a strategy for healing past and current race-based trauma for both children and parents.⁴⁶ Utilization of collective coping strategies, including civic engagement, participation in protests or boycotts, or healing groups, may also help reduce the effects of racial discrimination on parental mental health.¹⁵ As research in this area continues to emerge, these strategies can be incorporated into existing parenting interventions, including parenting groups or early home visiting programs.

In our sample, experiences of racial discrimination were more commonly reported among women with higher education and income levels, consistent with previous studies.⁴⁷ These women may be more likely to live or work in an environment where individuals are predominantly white, and thus are more likely to be exposed to and harmed by both institutional and interpersonal racism. Numerous studies demonstrate racial inequities in physical and mental health, even among the most highly educated and highest earning Black women, demonstrating that the insidious effects of racism persist regardless of social class.^48,49 Thus, although 89% of participants in the InterGEN study reported an annual income of less than $50,000 per year, results of our study may be applicable to all Black and African American families who experience racial discrimination, as the effects of racism persist across the socioeconomic spectrum.

This study is strengthened by examination of a sample of Black/African American mothers with preschool-age children, which provides a unique opportunity to examine parenting among women with young children who are at risk for experiencing racial discrimination. Forty-three percent of women in this study reported no lifetime experiences racial discrimination. While this is similar to rates reported in other studies with similar samples,^8–10 it is possible that this reflects a social desirability response bias, and that the effect of racial discrimination on parenting is underestimated in our sample. However, the InterGEN study included a number of strategies to promote culturally acceptable data collection, including employing Black women from the community to recruit participants, and use of ACASI software to allow private collection of sensitive data. As women in our sample were relatively young and reported low income levels, our findings also may not be generalizable to the general population of African American families.

This study was a secondary analysis of previously collected data, and thus we were limited to the available study measures. While we controlled for many characteristics associated with parenting (e.g. child age) and racism (e.g. income level), we did not have information on other factors that may influence parenting and racism-related stress, such as racial socialization strategies or the availability of other caregivers (e.g. grandparents). We were also unable to control for the impact of systemic factors that may influence our relationships of interest, such as racial segregation or community resources. Other directions for future research include consideration of other mental health mediators, such as anxiety or post-traumatic stress symptoms, which have been linked with experiences of racial discrimination and may have different effects on parenting than stress and depression.³⁰ Examination of parenting outcomes that reflect the quality of the parent-child dyadic relationship, such as parental sensitivity or attachment, as well as child social, emotional, and behavioral outcomes, are also important directions for future research.

Finally, we modeled mental health and coping strategies separately in our regression models in order to understand independent effects and identify mediating and moderating mechanisms that could be addressed with intervention. We did not have adequate statistical power to conduct structural equation modeling, but this approach could be considered for future studies. Our relatively small sample size also did not allow us to conduct subgroup analyses, such as an examination of women with higher income levels who experienced more racial discrimination, and we also may not have had adequate power to detect small effects. However, our statistical approaches, including the use of bootstrapping, are considered powerful and valid approaches for examination of relatively small sample sizes.^35,50

Experiencing the stress of racism and racial discrimination is not only harmful for mothers’ mental health, but may affect children’s social-emotional and behavioral development through effects on parenting stress. Results of this study add to an important body of literature highlighting the pathways through which racism creates and perpetuates racial inequities, beginning in early childhood. While tailored and culturally specific interventions may promote coping and help reduce depression and stress among African American mothers, addressing systemic racism and adopting policies to promote racial and social justice are of the utmost importance.