Exploring Whether and How Black and White Parents Talk with Their Children about Race: M(ai)cro Race Conversations about Black Lives Matter

Acknowledgement: The authors would like to thank the Institute for Learning and Brain Sciences at the University of Washington, Seattle, for funding the data collection for the On Parenting About Race (OnPAR) Study, and the Development of Identities in Cultural Environments (DICE) Lab at the Northwestern University for research support and feedback on the article. All data and survey questions used in this analysis are available on the Open Science Framework: https://tinyurl.com/onparosf.
Leoandra Onnie Rogers served as lead for conceptualization, data curation, investigation, methodology, project administration, resources, supervision, writing–original draft, and writing–review and editing, contributed equally to funding acquisition, and served in a supporting role for validation and visualization. Katharine E. Scott served as lead for visualization, contributed equally to writing–original draft and writing–review and editing, and served in a supporting role for supervision. Finn Wintz served as lead for formal analysis and served in a supporting role for data curation, visualization, writing–original draft, and writing–review and editing. Sarah R. Eisenman served in a supporting role for data curation, visualization, writing–original draft, and writing–review and editing. Chiara Dorsi served in a supporting role for data curation, formal analysis, validation, and writing–review and editing. David Chae contributed equally to conceptualization and served in a supporting role for data curation and writing–review and editing. Andrew N. Meltzoff served as lead for funding acquisition and served in a supporting role for conceptualization, resources, and writing–review and editing. Leoandra Onnie Rogers, Katharine E. Scott, and Sarah R. Eisenman contributed equally to formal analysis. Katharine E. Scott, Finn Wintz, and Sarah R. Eisenman contributed equally to validation. David Chae and Andrew N. Meltzoff contributed equally to investigation. Katharine E. Scott, David Chae, and Andrew N. Meltzoff contributed equally to methodology.

In February 2020, Ahmaud Arbery, a 25-year-old Black man, was jogging on a Sunday morning in Brunswick, Georgia, when three white men identified Arbery as a threat, then proceeded to chase, shoot, and kill him (Andone, 2021). In March 2020, Breonna Taylor, a 26-year-old Black woman in Louisville, Kentucky, was fatally shot when police raided her apartment while she slept (History.com Editors, 2020). In May 2020, terrifying video footage of Arbery’s murder was circulated on social media, thrusting the case into the public eye, and in the same month, a video of Derek Chauvin, a white police officer, kneeling on the neck of George Floyd, a Minnesota Black man, suffocating him to death, went viral (Selby, 2020). Arbery, Taylor, and Floyd were not the first (or last) victims of anti-Black racial violence in 2020, but their stories reignited a racial justice movement in the United States. “Black Lives Matter (BLM),” the slogan rooted in the grass-roots racial justice organization, became the global rallying cry, as citizens organized rallies, held vigils, led protests, and demanded change in cities around the nation (Buchanan et al., 2020). In June 2020, CNN and Sesame Street cohosted an hour-long town hall meeting, “Coming Together: Standing Up to Racism,” on primetime television (Coming Together: Standing Up to Racism, 2020). The event included race experts, like developmental scientist, Beverly D. Tatum as well as the beloved Elmo character, talking to children (and families) about race, racism, and racial protests. It was a profound moment in a nation where racial silence and colorblindness still dominate (Bonilla-Silva & Dietrich, 2011).

Although social scientists have long argued that racial socialization—talking and teaching children about race—is part of healthy child development (e.g., D. Hughes et al., 2006; Spencer, 2006; H. Stevenson, 2014), current events reignited debates about these efforts (Abaied et al., 2022; Sullivan et al., 2021). In the current mixed-method study, we explore whether and how Black and white parents talked with their children (8- to 11-year-old children) about BLM in the months following the summer of 2020. We focus on middle childhood (8–11 years) for several reasons. First, by middle childhood, many children display robust racial biases (Gollwitzer et al., 2020; Shutts et al., 2013) and, by this age, many children of color report experiencing race-based discrimination by peers (Cave et al., 2020; Marcelo & Yates, 2019). This highlights the need to address race and racism with children (before, if not) by the time they reach this development stage (e.g., Skinner et al., 2017; Waxman, 2021). Second, social norms intensify during middle childhood; both adults and peers expect young people to comply with social rules and young people are themselves more aware of what is and is not socially normative or appropriate, particularly when it comes to talking about race (e.g., Apfelbaum et al., 2008). Third, middle childhood is a period of rapid growth in cognitive flexibility, allowing children to represent abstract concepts and engage in social perspective-taking in new ways. They are also forming their own racial identities—learning what it means to be Black or white, how these racial groups fit into society and their developing sense of self (e.g., Sani & Bennett, 2004; Rogers & Meltzoff, 2017; Rogers et al., 2012; Ruble et al., 2004). Whereas younger children are not as deeply introspective about their racial identities and adolescents have a more crystallized view of who they are and want to be, middle childhood has fluidity; during this period, children are both aware of social norms and roles and flexible in their emerging belief systems, questioning established norms for themselves. In this way, middle childhood is a unique time to consider the messages children receive about race, racism, and racial justice.

To situate our study, we first review recent research on parents’ racial socialization during heightened racialized sociopolitical moments. We then introduce the m(ai)cro model of development (Rogers, Niwa, et al., 2021) and discuss how BLM provides a window into the context of structural racism as it shapes child development. We note here our decision to capitalize “Black” with regard to parent race and to use lowercase for “white” in this manuscript (except when beginning a sentence or used in section headers and tables). Although both are racial labels, Black people have a shared culture and history of systematic marginalization that does not have a parallel collective experience among people categorized as white in the United States. This differentiation in capitalization when referencing the racial backgrounds of parents is meant to acknowledge such positionality within the U.S. racial hierarchy (see Rogers, Niwa, et al., 2021).

There is a robust developmental literature on whether and how parents teach children about race through conversations, or racial socialization (e.g., Rogers, Niwa, et al., 2021; Sullivan et al., 2021). This literature has examined the frequency, forms, and content of race-related conversations, which shows that parents vary considerably in how often they talk about race and what they say, identifying four main types of racial socialization messages: (a) cultural or group pride, (b) egalitarian messages that focus on equality, (c) preparation for bias that name discrimination and injustice, and (d) promotion of mistrust messages that raise awareness of interracial tensions (e.g., D. Hughes et al., 2006; Paasch-Anderson et al., 2019; Ruck et al., 2021). The frequency and content of these socialization practices are racialized; Black families and Parents of Color in the United States are more likely to have race conversations, to do so earlier in childhood, and to communicate messages that acknowledge and prepare Black children for racism and inequality (e.g., Umaña-Taylor & Hill, 2020; Wang et al., 2020). White parents in the United States, in contrast, tend to avoid talking about race or do so in ways that emphasize egalitarianism (“treat everyone the same”) and dismiss the significance of race (“race doesn’t matter”; Priest et al., 2014; Zucker & Patterson, 2018).

In the wake of high-profile racial events, several recent studies have examined how race talk within families shifts in response to publicized and politicized racial events. For example, in 2014, Underhill (2018) was conducting a study about racial socialization in white families when Michael Brown, an 18-year-old Black boy, was killed by police. The story was top of national news as protests erupted in St. Louis, Missouri; Underhill pivoted to ask white parents whether they talked with their children about the shooting or subsequent uprisings. In an interview sample of 40 white parents, just 35% had talked to their children about the current racial protests, riots, and violence. Following the 2015 mass shooting in which a white man opened fire in an African American church killing nine people in Charleston, South Carolina, Abaied and Perry (2021) found similar results: only 37% of white parents had talked with their children (8–12 years old) about “current race-related events (including the Charleston shooting, Trayvon Martin, and Michael Brown).”

Amid the race events of 2020, Sullivan et al. (2021) sampled white and Black parents (with children under 18 years old) 6 weeks before the murder of George Floyd and 3 weeks after. Black parents, compared to white parents, reported more race conversations, and this difference was more pronounced after the murder of George Floyd. For Black parents, the frequency of race conversations was higher in the postsample than the presample, but white parents showed the same frequency of race conversations across the two time points. Rogers, Niwa, et al. (2021) examined data from later that summer (August 2020) on race conversations of Black and white parents with young children (4–11, Mage = 7 years old). They found the majority of Black (75%) and white (66%) parents talked with their children about “Black Lives Matter rallies or protests against police brutality”; this nonsignificant racial difference seems due to the unexpectedly high percentage of white parents reporting these discussions, which may be explained by the specificity of the question (asking about BLM) and timing of data collection. Whereas Sullivan et al. (2021) collected data three weeks after the murder of George Floyd, Rogers, Niwa, et al. (2021) collected data in July and August 2020, the peak of BLM protests and high-profile society-wide conversations (Buchanan et al., 2020). Like Rogers, Niwa, et al. (2021), Abaied et al. (2022) also reported that 80% of white parents sampled months after the public outcry about the murder of George Floyd, discussed “current race-related topics” (e.g., police brutality, the May/June 2020, and the recent deaths of Floyd, Taylor, Arbery) with their 14- to 17-year-old adolescents.

In addition to whether and how much parents discuss race topics, what they say—the content of their conversations—also matters. Here, the data suggest that when white parents discuss race-related events and topics, they do so with color evasive or egalitarian themes, talking about race in ways that downplay or deny the realities of inequality, racism, and injustice (Abaied & Perry, 2021; Underhill, 2018). There is variability, however; for instance, Abaied and Perry (2021) found that 69% of white parents engaged in some form of “color-conscious” socialization, such as endorsing equal treatment or acknowledging racism. Sullivan et al. (2021) found similar patterns and clear racial differences in the content of white and Black parents’ race conversations. Black parents were significantly more likely than white parents to acknowledge inequality and prepare children for racism, and less likely to use “colorblind” messages focused solely on egalitarianism without recognition of inequality; these differences were most pronounced after the murder of George Floyd (Sullivan et al., 2021).

Collectively, these findings suggest that current events can influence parents’ race-related discussions and underscore the need to analyze in more depth what parents say; particularly because when white parents talk about racist events, they often do so in ways that do not address racism. Although informative, the extant studies vary considerably in at least three factors which may be critical for theory and data alike: (a) the age of children (4- to 18-year-olds), (b) the type of race conversation assessed (e.g., a specific race event, or racism broadly), and (c) analysis of parents’ racial background. In the current analysis, we focus intentionally on middle childhood and both quantitatively and qualitatively analyze the content of conversations about BLM among a large number of Black and white parents.

In addition, we consider four contextual factors that shape how race-related events are experienced and thus potentially discussed among Black and white parents. Previous research on race socialization has been inconsistent in examining such factors but suggests that these factors may play an important role in racial socialization (e.g., D. L. Hughes et al., 2016). First, we consider parent education. Existing research suggests that parents with some college education or higher tend to report more racial socialization than parents with less formal educational experience, but that the influence of education may depend on race (e.g., Rogers, Niwa, et al., 2021). Specifically, because most white people do not have personal experience with racism, higher levels of education may provide greater knowledge of racial history that could translate to more discussion of race with children, whereas Black parents across educational levels may recognize the importance of race conversations. Second, we consider child gender. Previous research on racial socialization among Black families has shown some gender differences in content: Black boys tend to receive more messages about racism whereas Black girls tend to receive more cultural, racial pride messages (e.g., Caughy et al., 2011; D. Hughes & Chen, 1997; H. C. Stevenson et al., 1997). In a similar way, D. Hughes et al. (2009) found that reports of preparation for bias, such as racism and discrimination from others, were more strongly correlated among mother–son dyads (compared to mother–daughter) whereas cultural pride socialization was more strongly correlated for mother–daughter (compared to mother–son) dyads. Thus, parents’ talk about BLM may be more frequent and different among Black boys versus Black girls. We are unaware of any gender-specific racial socialization messages among white families.

Third, we consider the relevance of community diversity, because the diversity of one’s neighborhood has been correlated with racial socialization. For example, Black families in predominately white neighborhoods engage in more racial socialization (e.g., H. C. Stevenson et al., 2005; Tatum, 2017). Caughy et al. (2011) also found that cultural socialization and coping were more common in neighborhoods characterized by more community involvement, perhaps suggesting that parents engage in more racial socialization in neighborhoods where their Black children are more likely to navigate race within their communities. Moreover, white parents talk very little about race, in part because of the racial homogeneity of their communities and everyday networks (Hagerman, 2018).

Fourth, we explore the role of child age, because a common reason for not talking about race is the belief it is not age-appropriate to do so (e.g., Abaied & Perry, 2021; Rogers, Niwa, et al., 2021). Intersectionality matters here and the relevance of child age may interact with race; for example, white families, who are more hesitant and less likely overall to discuss race, may be especially likely to avoid the topic with their elementary-school-aged children. In the current analysis, we explore whether parent race interacts with other demographic variables in predicting the likelihood of talking about BLM as well as the content they report.

Theoretically, we frame this research with the m(ai)cro model, which centers “the macrosystem, and specifically racism (and its partnering ideologies of sexism, heteronormativity, classism, and capitalism)” as the starting point for developmental research (Rogers, Niwa, et al., 2021, p. 2). This theoretical lens brings the larger sociopolitical context, moments, and movements, like BLM, to the core rather than periphery of child development. For the topic of talking about race, this means paying attention to racism—the power, oppression, and privilege that is ascribed to race. We intentionally shift the focus from talking about race (generally) to acknowledging racism (specifically). Our approach builds on the idea that racism can be considered a system in which everyone participates (Tatum, 2017); the question is not if we participate in this system, but how we resist or reinforce it. As such, we examine how parents’ talk about BLM may reinforce or disrupt prevailing narratives of racism. This approach also recognizes that talk about racism is relevant for Black and white families (and beyond) and attends to the public and political aspects of racial socialization (Rogers, Niwa, et al., 2021). Thus, while these are micro conversations among parents and children, they are inextricably tied to the macro sociopolitical sphere that we collectively navigate.

To operationalize these principles in our research, we examine how parents talk to their children about BLM. There are four reasons why the topic of BLM is useful to capture talk about racism as public, political, and systemic. First, the BLM movement is explicitly about racial injustice; the stated mission is “to eradicate white supremacy and build local power to intervene in violence inflicted on Black communities” (BlackLivesMatter.com, Black Lives Matter, n.d.). Second, BLM is widely recognized as a social movement for racial justice in the United States. Inasmuch as BLM was seen as striving for racial justice, then its counterpleas, “All Lives Matter” and “Blue Lives Matter,” were often used to refute this position, creating a public and political script for race discussions in the United States. Third, BLM is developmentally relevant to children and their racial experiences and identities in middle childhood (e.g., Rogers, Rosario et al., 2021). Fourth, the topic of BLM is timely; our research was intentionally conducted in the months immediately following the peak for BLM tweets, protests, street art, news, and children’s programs (e.g., Buchanan et al., 2020).

The current study examined whether and how U.S. Black and white parents of children (8–11 years old) talk about BLM in the months following the heightened racial protests of summer 2020. Our overarching research questions were: Do Black and white parents talk to their 8- to 11-year-old children about BLM? If not: What explanations do parents provide for not talking about race? If yes: What is the content of their explanations?

We further analyzed the following four issues. First, does the likelihood of talking about BLM differ for Black and white parents? Given the specific focus on BLM and timing of data collection, we predicted that Black parents would be more likely to talk about BLM with their children than white parents (e.g., Rogers, Niwa, et al., 2021). We also explored whether the likelihood of talking about BLM was moderated by race and parent education, child age, child gender, and/or neighborhood diversity. Second, in what ways does the content of parents’ responses about BLM conversations with their children align with or challenge ideologies of racism? We predicted that parental talk about race with their children includes both content that reinforces racism (such as denying the existence of racism) and content that disrupts it (such as characterizing certain events and actions as examples of racial injustice) (e.g., Sullivan et al., 2021). Among parents who report not talking about BLM at all, we expected age-appropriateness (e.g., child is too young) to be a common reason (Abaied & Perry, 2021) as well as explicit denial of racism. Third, does the content of parents’ responses differ between Black and white parents? Based on prior research, we predicted more responses from Black parents to acknowledge racism and more responses from white parents to use uncritical race messages (Priest et al., 2014; Sullivan et al., 2021). Fourth, we also explored whether patterns in the content of BLM discussions were moderated by race and parent education, child age, child gender, and/or neighborhood diversity.

Our focus on Black and white parents is not meant to convey that other racial and ethnic groups are irrelevant in this issue. However, the longstanding racial binary in the U.S. positions whiteness as superior and Blackness as inferior, such that these racial groups are anchors that tether the racial hierarchy in place in the U.S. context (see Molina, 2014; Rogers, Niwa, et al., 2021). Moreover, BLM directly challenges anti-Blackness and is explicitly about Black people, so we focus on Black and white parents’ socialization within this context.

Data for the current analyses were collected for the On Parenting About Race (On PAR) study. On PAR was a larger, mixed-method study administered via Qualtrics that surveyed Black and white parents (N = 725) from across the United States, from November 2020 to January 2021, about a variety of topics including racial identity, the BLM movement, parenting, and mental health (Rogers et al., 2023). The current study focused on parents’ reported discussions about BLM with their children. The project was not preregistered. All data and survey questions used in this analysis are available on the Open Science Framework (OSF): https://tinyurl.com/onparosf. All research activities were reviewed and approved by the Institutional Review Board [IRB# STU00213512].

We recruited a sample of 725 Black and white parents of 8- to 11-year-old children in the United States (47.45% Black, Mage = 38.08 years old, SD = 7.00). Table 1 presents a full description of sample demographics by parent racial group. Of the 344 Black parents, 27.91% were men (n = 96), 71.80% were women (n = 247), and one identified their gender as “Other.” Of the 381 white parents, 55.12% were men (n = 210) and 44.88% were women (n = 171). The majority of the sample was married (68.28%; n = 495). The child sample (Mage = 9.45 years, SD = 1.08) was composed of 398 boys (54.90%) and 326 girls (44.97%); one parent identified their child’s gender as “other/nonbinary.” Families were economically diverse, with median reported annual income between $54,000 and $59,988. Parents’ reported levels of educational attainment ranged from “less than a high school diploma” (3.17%) to earned “doctorate” degree (5.24%), with roughly half (48.69%) earning a bachelor’s or master’s degree (see Table 1). Community racial diversity was calculated using census tract data; the mean percentage of Black residents in participants’ communities based on the designated market areas (DMA) location was 24.71% (SD = 29.42%). The mean percent of white residents in participants’ neighborhoods was 59.36% (SD = 29.68%).

Participants were recruited via Qualtrics Survey Panels using specified DMA which included the following cities for our research: Atlanta, Baltimore, Cleveland, Chicago, Detroit, Houston, Los Angeles, Memphis, New York City, Philadelphia, and St. Louis. DMAs were selected for diverse geographic representation as well as racial diversity for recruitment purposes. The survey was open for recruitment from October 2020 through January 2021. Eligible participants were parents located in the aforementioned DMAs who self-categorized themselves as: (a) monoracial Black or white, (b) having at least one 8- to 11-year-old child, and (c) able to answer questions in English. Parents gave the ages of their children and were then randomly prompted to focus on a child either between 8–9 years old or 10–11 years old. To balance recruitment for race and child age, Qualtrics personnel monitored the survey demographics continually, setting quotas within each DMA to ensure that it was balanced across Black and white parents in the 8- to 9-years-old or 10- to 11-years-old age groups. To further assure balance, Qualtrics opened recruitment in a series of smaller samples of 10 white and 10 Black participants within a DMA; once both racial numbers were filled, they then reopened the survey in that DMA. This process was repeated throughout the data collection process. Our final sample included no less than 10 and more than 40 participants of each racial group from each DMA. We used this approach because online research samples are skewed toward larger numbers of white participants; thus, it was likely that our sample of white parents would be recruited quicker than our sample of Black parents, and we wanted to ensure our data collection for the Black and white samples occurred concurrently and remained balanced throughout the study.

Interested participants received an online screener with questions to confirm eligibility before signing consent and moving to the full survey. In addition to demographic questions, participants answered a series of multiple-choice and short-answer questions about a variety of topics related to race, racism experiences, and conversations with their children. Relevant survey measures are discussed below. Once participants completed the survey, they received compensation in accordance with terms for Qualtrics Survey Panels. Participants were also given the option to provide their name and email address if they were comfortable with future contact from researchers, but this was collected separately from their survey responses which were linked to a randomly assigned participant ID number for confidentiality purposes.

The survey was self-administered online via Qualtrics. Parents responded to a range of written and multiple-choice questions about their experiences with civic engagement, conversations with their children about race and racism, and the racial socialization of both themselves and their children. All responses were typed into the Qualtrics platform. Focal questions for the current research are detailed below.

Talking About BLM

To assess BLM talk, parents answered two questions: a closed-ended prompt and an open-ended typed response. First, all parents saw the survey item: Have you talked to your (child age) year-old child about BLM?, a closed-ended prompt with response selections of “yes” or “no.” These dichotomous self-report data were analyzed to report on frequency of “yes” and “no” responses. If participants selected “yes,” they were routed to the open-ended prompt: What have you said to your (child age) year-old child about BLM? If participants selected “no,” they were routed to the open-ended prompt: Why have not you talked to your (child age) year-old child about BLM? The short answers to these prompts were captured through a qualitative coding system and submitted to statistical analyses (described below).

Demographic Variables

Participants self-reported their demographics (see Table 1), which were used to explore the broader sociopolitical and economic contexts of parents’ discussions of BLM.

All data were checked and cleaned for rote responses, nonsense, and incoherence across the survey responses, and suspicious cases were removed from the data file. First, the Qualtrics data team followed procedures using their multidimensional check for response quality and then the verified “quality responses” were delivered from Qualtrics to our research team. Second, we conducted additional data quality checks, including: (a) straight lining and response patterns, (b) response duration (too fast or short), (c) attention checks embedded in the survey (participants were asked to report their zip code at different points in the survey and removed if they failed the attention check), and (d) evidence of duplicates (identical IP address, identical demographics, similar open-ended responses). Open-ended responses for talking about BLM were then imported into NVivo QSR Qualitative software for coding. Our sequential, mixed-data analytic approach began with qualitative coding in NVivo, followed by data exportation to R for quantitative analysis.

Our coding process uses Wolcott’s (1994) three levels which include: description, analysis, and interpretation. In Level 1, we developed codes that describe the content of the data. In Level 2, we analyzed the prevalence and patterns of codes within our sample. In Level 3, we used the m(ai)cro framework (Rogers, Niwa, et al., 2021) to interpret the codes and patterns in our data within the broader sociopolitical context. This final level of interpretation is essential to the analysis, doing the interpretive work of making sense of the codes and drawing systematic meaning from them (Wolcott, 1994). Here, using m(ai)cro framework, we focus less on the specific content of each response and more on whether and how the codes function to reinforce or disrupt societal narratives and systems of racism (Rogers, Moffitt, Jones, & McLean, 2021).

The first step for describing the data to generate codes was to organize the dataset into “yes” and “no” responses; parents who reported talking about BLM (n = 581) and parents who did not (n = 144). We coded the data without participant demographics visible; however, parent race was referenced to make final coding decisions consistent with the codebook. For example, a white parent who wrote, “our lives matter” is interpreted differently in the context of BLM than a Black parent who responded, “our lives matter.” In this way, the codebook centers a m(ai)cro lens so that racial comparisons are interpretable within the racial structure. Three undergraduate research assistants read all of the responses and wrote analytic memos (Birks et al., 2008) to describe the themes, patterns, and questions they noticed in the data. We discussed and translated these memos into preliminary codes to organize the data by content or key topics related to BLM. At this stage, we noted nonsense responses to this open-ended prompt (e.g., “dkuelbulekgh didkkjkhe!”) as well as responses pasted directly from websites (e.g., the word-by-word copy–paste definition of BLM from Wikipedia). Thus, some of the participants who otherwise answered survey questions reasonably (and passed data quality checks) did not answer the specific question about BLM in a sensical manner; we detail this in the codebook. It was also apparent that the content of “yes” and “no” responses were noticeably different. Thus, we developed the codebooks for the “yes” and “no” responses separately.

We used open coding to generate keywords to describe for each response, which was used to create a preliminary coding scheme. We followed a general inductive approach (D. R. Thomas, 2003), paralleling prior studies examining the content and thematic coherence of open-ended race responses (e.g., Rogers & Meltzoff, 2017; Rogers et al., 2012). Our data-based conversations in developing the codebook situated responses within the m(ai)cro lens of reinforcing or disrupting racism; that is, we developed a coding system to categorize participants’ responses to address the following overarching question: Does this answer/comment suggest a message that disrupts racial silence and acknowledges racism or does it reinforce racial hierarchy and uphold racial injustice? Our final codebook reflects this orientation. Once established, we divided the data equally among the three research assistants to code all responses. See Table 2 for full descriptions of the coding scheme.

“Yes” Responses

For parents who said they had talked to their child about BLM, we identified seven codes. The first two themes indicate an antiracist stance: Acknowledge Inequality responses made explicit references to race, Blackness, and inequality; and Affirm and Support responses emphasized the value of Black lives and/or support for the organization’s mission to protect Black people. The remaining codes captured responses that passively accepted or actively enforced the racial status quo: Lacks Substance messages were vague, with no real content for analysis (e.g., “we talked about it”); Uncritical Equality messages emphasized that all people are equal but without acknowledging racial injustices; Delegitimizing responses questioned the legitimacy of the BLM organization and/or racism; and Other Content responses were relevant to the topic but came up too infrequently to substantiate a clear code. Finally, we noted responses that were Uncodable; these responses were nonsensical or copy and pasted from the internet (see Table 2). The subcodes of Uncodable—Nonsensical and Copy/Paste—responses were mutually exclusive. All other codes could co-occur, though most responses received a single code (91.39% of all “yes” responses). The exceptions were the Acknowledge Inequality and Affirm & Support codes which overlapped in 6.21% of all “yes” responses. There is a distinct message in each of these two codes that warrants specificity and relation, rather than a full collapse of the differences between the two codes. This coding conveys that affirming Black life (“affirm and support”) is separate from but equally integral to racial justice as naming inequality (“acknowledge inequality”).

“No” Responses

For parents who said they did not talk to their child about BLM, we identified seven codes: BLM Delegitimizing dismissed or actively undermined the legitimacy of the BLM organization; Racism Delegitimizing was the expressed denial of the existence of racism broadly; Indifference responses suggested a passive avoidance of the conversation, stating that the topic simply had not come up or child had not asked; Not Appropriate responses focused on parents’ desire to preserve children’s innocence and/or children’s naivete; Not Prepared responses included parents reporting they did not feel equipped or knowledgeable enough to discuss BLM; and response coded as Other Race Conversations included parents who noted they talked about race/racism with their child but not BLM specifically. Finally, Uncodable responses that were nonsensical or copy and pasted from the internet. Again, most responses (97.92% of all “no” responses) were assigned a single code.

Throughout the codebook development, the three coders double-coded responses (two coders per response) to reach consensus. Once finalized, we conducted reliability coding. A random 20% of the “yes” and “no” responses were assigned to the lead researchers (Rogers, Scott) who did not participate in consensus coding. These responses were cross-checked with the data final coding by the research assistants. We obtained 90% or higher coder agreement; all discrepancies were resolved with discussion. After coding was complete, we finalized code names to ensure they accurately captured the tone and tenor of the responses.

The qualitative coded results were exported as dichotomous variables (as 1 = present, 0 = absent) to R for analysis. For our core quantitative analyses that compare responses on the basis of parent race, parent education, child age, child gender, and community racial diversity, we fit generalized linear models with the binomial family for the logit link function because the responses were coded as dichotomous outcome variables (i.e., each response was coded for the presence or absence of each code). We then regressed the codes onto parent race (White = −0.5; Black = 0.5). All p values were adjusted using a Holm–Bonferroni adjustment to correct for multiple comparisons.

The coding team consisted of three undergraduate students trained in qualitative and mixed-method developmental research. Rather than an assumption of neutrality or distance from the research and participants (Nzinga et al., 2018), we engaged a critical epistemological stance with active awareness of the ways our own biases and lived experiences shape our data analysis and interpretations. The coders self-identify as white and male; white and genderqueer; and multiracial (Black and white) and female. The team was supervised by the principal investigator, a Black female faculty member and also by a postdoctoral scholar, who is a white woman. Having a racially diverse team was intentional, designed to encourage critical reflection and engagement with the data and our interpretations. This was especially useful in establishing the parameters and names for codes; we stayed close to the data (Maxwell, 2005), while also grounding its significance within a racialized system (Rogers, Niwa, et al., 2021). Our systematic, collaborative approach to coding within an “interpretive community” of scholars with diverse knowledge, perspectives, and experiences (Marecek et al., 2001) builds credibility and accuracy of the results, and the development of a detailed codebook provides access to validity, transparency, and transferability of the analysis process (Lincoln & Guba, 1986).

We first report descriptive and inferential findings on the prevalence and content of parents (not) talking about BLM, and then discuss variation by race and related demographics.

When asked to provide a dichotomous response to whether or not they had discussed BLM with their child, the majority of parents (80.14%) in our sample reported “yes.” This finding aligns with the few recent studies suggesting that parents are talking about race in response to the sociopolitical context (e.g., Abaied et al., 2022; Rogers, Niwa, et al., 2021).

Racial Differences in Prevalence of BLM Talk

While most parents reported talking with their child about BLM, we also found racial differences in prevalence of talking about BLM. Specifically, Black parents (84.30%) were significantly more likely than white parents (76.38%) to talk about BLM with their children (p = .008). See Table 3 for analysis details.

Next, we explored the relevance of other demographic variables, alongside race of the parent, on the prevalence of talking about BLM. Parents with higher levels of education, were significantly more likely to talk about BLM with their children (p < .0001; see Figure 1). No other demographic factors influenced the likelihood of parents discussing BLM with their children, and there were no significant interactions between race and any demographic variables. See Table S1 in the supplemental materials for details.

We coded the content of the responses from the 581 parents who reported “Yes” to having talked about BLM with their children (see Table 2). The most common type of response was Acknowledging Inequality (33.56%), for example, a 41-year-old white father said, “I talk with my son about how wrongful deaths of men and women of color at the hands of police.” The next most frequent response was Affirm and Support (23.06%), for example, a 28-year-old Black mother said, “I try to remind him that he is important and worthy despite what the media tells us.” These two types of responses co-occurred at times, particularly among Black parents who both emphasized the racial injustices and affirmed the value of Black identity. The remaining responses were distributed into the following coding categories: (a) Uncritical Equality (14.80%; 33-year-old white mother: “That all lives matter no matter your skin color”), (b) Lacks Substance (11.88%; 30-year-old Black mother: “I have explained to him what it is and what it means.”), and (c) Delegitimizing (5.68%; 42-year-old white mother: “I have mentioned that they are the cause of the riots that she sees on the news…. We feel differently than most and do not sympathize with the cause.”). A small percentage of “Yes” responses were coded as Other content (1.72%; 36-year-old white father: “Be careful around any protest movement. Make sure to study what the movements are all about and what their guiding beliefs are as well as what they hope to accomplish in the end.”).

Approximately 16% of responses were Uncodable because they were either: (a) Nonsensical or confusing (8.43%; 33-year-old Black father: “It’s nice to consider the black lives matter.”), or (b) directly copy and pasted from Internet sources (7.57%; multiple participants: “Black Lives Matter is a decentralized political and social movement advocating for nonviolent civil disobedience in protest against incidents of police brutality and all racially motivated violence against Black people.”). These nonsensical and copy–pasted responses raise questions about the content of “Yes” responses overall; we analyze the patterns the uncodable responses in our data and return to this issue in the Discussion section.

Racial Differences in the Content of BLM Talk

As shown in Tables 2 and 3, Black parents were significantly more likely than white parents to Acknowledge Inequality (44.48% vs. 22.68%) and to Affirm and Support (34.14% vs. 12.03%) the lives, humanity, and identities of Black people in their responses about BLM. At the same time, Black parents were significantly less likely than white parents to give response coded as Uncritical Equality (9.31% vs. 20.27%) or Uncodable (6.90% vs. 28.52%); the majority of Uncodable responses were copy and pasted from Internet sources (e.g., Wikipedia, news articles, etc.). The prevalence rates for responses coded as Delegitimizing and Lacks Substance did not differ significantly by parent race.

Exploratory Analyses

Next, to explore the relevance of other demographic variables (e.g., parent education) in our exploratory analyses, we selectively focused on three robust and substantive codes and fit one logistic regression model per code: (a) Acknowledge Inequality; (b) Affirm and Support; and (c) Uncritical Equality. We selected these categories because they were substantive (i.e., not uncodable) and at least 10% of responses were coded into each category, providing ample variability for analysis. Each of these analyses evaluated how race, as it interacts with different demographic variables, may impact the likelihood of discussing BLM as well as the content of and reasons for not having BLM conversations. For the three codes of interest, we regressed parents’ coded responses on each demographic variable, parent race, and the interaction of parent race and each demographic variable of interest, in turn (parent education, child age, child gender, and racial demographics of DMA). Parent education was operationalized as the highest grade or level of school parents completed (Coded 1 [lowest] to 9 [highest]; see Table 1). Although educational attainment is not necessarily linear, we treated it as such in the analyses for interpretive purposes.

Given prior research recognizing how racial positionality intersects with other social variables, we anticipated that demographic factors may function differently among Black and white parents. Thus, to evaluate each demographic variable, we regressed the codes onto the demographic variables, parent race, and the interactions between each demographic variable and parent race. For ease of presentation, we report significant main effects and interactions with race in text and details of nonsignificant analyses are included in Tables S1–S4 in the supplemental materials. As with the core analyses reported above, all p values were adjusted using a Holm–Bonferroni adjustment to correct for multiple comparisons.

Across all of these analyses, the only significant effect was an interaction between parent race and parent education on Acknowledging Inequality (B = 0.35, χ2 = 8.64, p = .03, OR = 1.43). Simple slopes analyses revealed that for Black parents, the likelihood of Acknowledging Inequality was unrelated to parents’ education (B = 0.10, χ2 = 1.53, p = 1.00, OR = 1.10); across levels of education, Black parents were more likely to acknowledge inequality in their BLM talk. Among white parents, however, there was a significant effect of parent education on Acknowledging Inequality such that the responses of white parents with higher levels of education were less likely to acknowledge inequality (B = −0.26, χ2 = 7.85, p = .04, OR = 0.77). See Figure 2 for a visual depiction of the interaction effect. This effect was in the opposite direction of what one might expect, which warrants further research. See Tables S2–S4 in the supplemental materials for details of analyses.

A total of 144 parents (19.8% of the full sample) provided “No” responses, indicating that they did not talk to their children about BLM. The most common reason for not talking about BLM was Delegitimization. This code had two subcodes: BLM Delegitimization was present in 20.83% of the “No” responses and included responses that discredited the organization, mission, and actions (e.g., 42-year-old white father: “Because Black lives matters is a racist group and I do not teach ignorance.”). The other subcode, Race Delegitimization, was present in 15.28% of the “No” responses, which denied race and racism broadly (e.g., 41-year-old white mother: “I don’t talk to my child about subjects like this. Color should not matter. Every life matters.”).

The next most common reasons for not talking about BLM were: (a) Indifference, in which parents indicated that they did not have a motive or reason to do so or that they had not thought about it (24.31%; 45-year-old Black mother: “I haven’t really thought about it.”); and (b) Not Appropriate, in which parents indicated that children were too young or innocent to talk about BLM, or the child simply would not understand (20.14%; 42-year-old white mother: “Because he is too young to talk about the dynamics of something like protestors and racial equality. He can understand the basics of being kind and loving of all people, but he doesn’t need to bear the burden of something so heavy at his age. Our kids have been bombarded with scary things this year. I just want to keep him young longer.”). A smaller percentage of parents were coded as reporting Other Race Conversations but not discussing BLM specifically (6.25%; 36-year-old white father: “I haven’t mentioned that organization specifically, but have talked about inequity and how things must change.”). Less than 5% of parents (4.17%) reported feeling Unprepared to have these conversations (34-year-old white mother: “I am not sure I fully understand it enough to explain it to him.”). Finally, 11.11% of the “No” responses were Uncodable (33-year-old white mother: “Little explaining that about things story”). See Table 2 for a summary with sample responses.

Racial Differences in Reasons for Not Talking About BLM

Amongst parents who reported not talking about BLM with their children, both Black and white parents were equally likely to provide responses that question the legitimacy of the BLM movement (BLM Delegitimization), 18.52% and 22.22%, respectively. However, Black parents were significantly less likely to give responses that were coded as denying systemic racism (Racism Delegitimization), 7.41% of Black parents compared to 20.00% of white parents. Black parents were also significantly more likely than white parents to report Other Race Conversations (12.96% vs. 2.22%) with their children and were marginally more likely than white parents to provide responses coded as Indifference (33.33% vs. 18.89%), indicating that they “just haven’t thought about” raising the topic or were waiting for the child to do so, for example.

Given the overall infrequency of “No” responses and therefore, the low numbers of codes in each category, we did not have sufficient power to evaluate the effect of the exploratory demographics variables on parents’ reasons for avoiding conversations about BLM.

Guided by a premise of the m(ai)cro model (Rogers, Niwa, et al., 2021) that racial socialization is responsive to sociopolitical and historical events, this research examined what parents said about race and racism during a sociopolitical moment of racial unrest and global focus on racial justice. We found that most parents (80%) in our sample reported talking with their 8- to 11-year-old child about BLM. This aligns with recent research, specifically for Black parents (Sullivan et al., 2021), but the pattern among white parents is particularly striking from a sociopolitical lens. For example, in studies conducted prior to 2020 (Abaied & Perry, 2021; Underhill, 2018) researchers investigated whether white parents talked with their children about publicized racial violence, and in both studies, only about one-third of white parents talked about the events. Yet, in studies of white parents following the murder of George Floyd, those numbers are notably higher; 76% of white parents in the present study and 80% in Abaied et al.’s (2022) recent work. This seems to reflect the national response to racism in Summer 2020, marking a shift in the landscape for race conversations, especially among white parents who were more willing to report talking about racism with their children. In the face of national racial injustice, such findings point to the possibility and potential to meaningfully disrupt the cultural norm of racial silence. Still, Black parents were more likely to talk about BLM than white parents, and we found predicted racial differences in the content of these reports that are informative to theory and practice. In this way, these findings parallel the robust literature on racial socialization (e.g., D. Hughes et al., 2006; Priest et al., 2014; Ruck et al., 2021), and suggest that even in response to explicit and public racialized violence, Black and white parents markedly different approaches to talking with their children about race that pattern the established social “regularities” of racial positionality in the U.S (D. L. Hughes & Watford, 2022).

Among the parents who reported talking about BLM (n = 581), we found that most frequently used code was Acknowledging Inequality, followed by the Affirm and Support code. As a base rate, this finding points toward more critical, antiracist socialization. At the same time, parents’ messages across the sample were quite varied. Acknowledging Inequality and Affirm and Support were represented in just 33% and 23%, respectively, of the “Yes” responses. The prevalence of these types of responses was also driven largely by Black parents who were more than twice as likely as white parents to talk about BLM in ways that acknowledge racial injustice and affirm the human dignity and value of Black lives. Collectively, 78% of responses from Black parents were coded for Acknowledging Inequality and/or Affirm and Support. These patterns align with previous research on racial socialization which suggests that Black families are more likely to discuss race, racism, and preparation for bias with their children than are white families, and underscores the value of supporting families in this component of their parenting (R. E. Anderson & Stevenson, 2019; D. Hughes et al., 2006; Priest et al., 2014; Ruck et al., 2021).

For white parents who talked about BLM, there was no dominant or cohesive message. The most common codes were Uncodable (27%), followed by Acknowledging Inequality (22%), and Uncritical Equality (21%). The fact that the Acknowledging Inequality code, including references to racism and inequality, was the second most common way that white parents talked about BLM, may reflect the increased attention to race and racism in media and children’s programming. Still, white parents were twice as likely as Black parents to use color- and power-evasive language, emphasizing equal treatment for all people without acknowledging or naming injustices, presenting a narrative of Uncritical Equality (Abaied & Perry, 2021; D. Hughes et al., 2009; Sullivan et al., 2021). More attention-getting, however, was that nearly 27% of white parents who reported talking with their child about BLM gave open-ended responses that were Uncodable. White parents were more than four times as likely as Black parents to give nonsensical responses (13% vs. 3%) and 14 times more likely to directly copy and paste an answer from the Internet (14% vs. 1%). Thus, while many white parents said they talked about BLM, they were unable (or unwilling) to provide a substantive description. These racialized patterns call us to question how to interpret white parents’ reported conversations.

One interpretation is that the foregoing pattern of responses reflects white parents’ performativity of antiracism—a response to social pressure to appear “antiracist” or at least not racist. This has been referred to as “performative allyship” in which people engage in surface-level behaviors, such as posting messages of solidarity with minoritized groups instead of genuine forms of challenging racism and supporting of social justice (e.g., Kalina, 2020). Prior to 2020, there may have been less of a cultural expectation for white people to speak out on race and racial issues, but this norm was disrupted in the wake of BLM protests in 2020. The urge for “performative allyship” may have been tapped among the white parents in our sample who reported talking with their child about BLM but did not offer substantive answers. The prevalence of this type of response among white parents is concerning and warrants further attention in research designed to investigate ways to reduce performativity and increase deeper engagement. Indeed, in a recent report by the Pew Research Center (2023) found that support for BLM has declined from its 67% at its peak in 2020 to 51% in 2023, suggesting that the sharp uptick in racial justice advocacy we observed was not sustained. With regard to developmental research and theory, these findings also support the relevance of the chronosystem on micro-procsses (Rogers, Niwa et al., 2021), and call us to use caution when relying on parents’ “yes/no,” or even frequency, responses on surveys or test items about racial socialization without probing the content and meaning of race-related messages within families.

While most parents reported talking about BLM, we also found interesting and important trends among the 144 parents who reported not talking about BLM with their children. The most common reason parents gave for not talking their child about BLM was some form of Delegitimization (35%, 52 of 144); these parents did not support the mission of BLM or believe that racism itself exists. Here our coding distinction between parents who gave responses that were directed toward the Delegitimization of BLM specifically versus a Delegitimization of Racism in general becomes very useful. Interestingly, Black and white parents were equally likely to reference BLM Delegitimization 22% and 20%, respectively. In their responses, some parents tied the BLM movement to fake news and conspiracy theories that made them distrustful of the organization, its intentions, and potential for addressing racism. In this way, some parents—both Black and White—were skeptical of the politicization and actions of the BLM movement. For Racism Delegitimization, however, there was a very different pattern. White parents were three times as likely as Black parents to deny racism and assert explicitly racist comments; 18% of white parents delegitimized racism compared to 7% of Black parents. This finding is consistent with prior research in which white adults are significantly less likely than Black adults to recognize racial injustice in the United States (Eibach & Ehrlinger, 2006), and parallels the literature on racial socialization wherein reports of “preparation for bias” among Black youth and families is a common message (e.g., D. Hughes et al., 2006; Paasch-Anderson et al., 2019).

For other “No” responses, both Black and white parents were equally likely to report Indifference, stating that they had not thought about it, or the child had not mentioned the topic. In a system of racial inequality, a passive response of not naming injustice can contribute to its maintenance (Bonilla-Silva & Dietrich, 2011; Tatum, 2017). Yet, for Black parents who did not talk about BLM, we observed a distinct pattern: Black parents reported having Other Race Conversations. Although white parents who did not talk about BLM also seem to not talk about racism at all, Black parents were more than six times as likely to report that they discussed “other relevant” race topics with their child. This finding also seems to align fit with our observed pattern whereby Black and white parents are equally likely to delegitimize the BLM movement but not racism in general.

Finally, we were surprised that very few parents stated that talking about BLM was Not Appropriate for their child, given the prevalence of the preservation of (white) childhood innocence (R. A. Anderson & Masicampo, 2017) and some evidence of this reasoning among white parents from prior research (e.g., Abaied & Perry, 2021; Rogers, Niwa, et al., 2021). In a similar vein, we found that very few parents (∼4%) reported being Unprepared to talk with their child about BLM, countering the common societal narrative that white parents, in particular, may recognize the need to talk about race but are uncertain about how to do so. These patterns may show the impact of the sociopolitical moment and mass media efforts with many sources (e.g., Sesame Street Workshop and PBS Kids) promoting resources for and the value of race conversations with young children. Our research findings, however, indicate that parents’ reasoning varies considerably, and very few state that it has to do with lack of preparation. That said, the findings suggest the need for more empirical focus on why parents do not talk about racialized events with their children. For example, R. E. Anderson et al. (2020) propose the need to evaluate how parents perceive their own competency and preparation to talk about race with their children as key to effective racial socialization practices. Our finding that few parents reported lack of preparation as to why they did not discuss BLM may call attention to the need for greater accuracy in parents’ understanding of their own capacity for racial socialization.

Our exploratory analyses investigating whether race interacted with relevant contextual variables were largely nonsignificant. These contextual variables and variations remain underexamined in the racial socialization literature more broadly (e.g., D. L. Hughes & Watford, 2022; Ruck et al., 2021; H. C. Stevenson et al., 2005). Still, we observed the two significant interactions in our exploratory analyses that highlight the intersection role of educational attainment and race. Overall, parents who reported higher levels of education were more likely to discuss BLM than parents with lower levels of education. This aligns with previous research on racial socialization (Rogers, Niwa, et al., 2021). The high media and news coverage on the topic during a heightened moment like summer 2020 may also contribute to the pattern, inasmuch as adults with higher levels of education consume more news. For example, 57% of U.S. adults with college education or higher are more likely to read “in-depth” news articles and stories on a daily basis compared to those with only (36%) or without (20%) a high school diploma (Media Insight Project, 2014).

That said, we found an unexpected pattern among parents who reported talking about BLM: among the white parents, the likelihood of Acknowledging Inequality was higher among white parents who reported lower levels of education. Although we might expect higher levels of education to support more critical engagement with racial topics, here we see the opposite pattern. Although Black parents seem equally likely to Acknowledge Inequality across education levels, at the highest levels of education, we see the largest gap between Black and white parents. One explanation is simply that few white parents overall (n = 62, 22%) gave responses coded as Acknowledging Inequality, which impacts the distribution and thus ability to detect effects. White parents who are highly educated are also highly likely to reside in white communities and social circles, with very limited interaction with Black families (Hagerman, 2018); such homogeneity is itself an example of the "racial regularities" that characterize social context (D. L. Hughes & Watford, 2022) and may also undermine parents’ motivation to discuss BLM with children. Another explanation is the link in the United States between socioeconomic status and meritocracy beliefs. For example, belief in meritocracy is positively associated with system justification (Darnon et al., 2018) and some previous work shows that individuals from higher class backgrounds hold stronger meritocratic beliefs that hard work overcomes inequality (Xian & Reynolds, 2017). Such beliefs would be inconsistent with acknowledging (racial) inequality. That is to say, higher education among white parents in our sample may be correlated with more general ideologies that undermine recognition of systemic inequality. Still, the interactions discussed here are based on a small subsample of responses and additional research is needed to further attend to the intersecting contexts that influence whether and how parents discuss racial (in)justice with their children.

We did not observe significant interaction effects for child age, gender, or neighborhood diversity. The age span of 8–11 years old is a relevant yet narrow range. Some prior research on children’s racial identities suggests that during middle childhood, the racial background of the child is more relevant to the content of children’s race narratives than age within this developmental period (Rogers & Meltzoff, 2017; Rogers et al., 2012). Our interpretation is not that there is an absence of “development” but rather that the racialized context is salient to children’s development. This pattern holds in our sample; racial background is more predictive of the race conversation than child age. With gender, prior research is limited—and we are unaware of research with white families examining gendered patterns in race socialization. Finally, the data based on general DMA information for diversity may not be sensitive enough to reliably detect differences in frequency of talk about BLM. While we attempted to analyze for contextual factors in relation to racial group membership, future research will benefit from larger projects that are designed to robustly examine intersectional and contextual factors within and across racial groups.

We acknowledge several limitations. First, the use of online data collection brings limitations. Although we used strict strategies to ensure data quality, the online Qualtrics sample may include expert survey takers and those who may not have devoted deep attention to the questions and topics. Stratifying and pacing our recruitment by DMAs were strengths of our online data collection design as it offsets the tendency to recruit from a single area of the country and from whoever is first to complete the survey. Still, unlike Census tract data, for example, DMAs do not provide guaranteed data on each participant’s neighborhood which limits our ability to analyze the specificity of racial diversity for each participant. We also have only first-person reports, without ways to verify the substance and accuracy of the reported information. Did parents actually talk about BLM? We have some evidence—the nonsensical, lack of substance, and copy–pasted responses—that suggests some did not actually do so in any meaningful way. Moreover, even if parents did talk about BLM, we are only able to interpret what is reported, not able to verify it against actual behavior. With brief open-response data such as used here, the codes are necessarily tied to the explicit words used. Thus, we are relying on parents’ reported moments rather than documentation of their child’s lived experiences.

An additional limitation arises when considering the specificity of the sociopolitical moment. These data represent patterns detected during the height of the 2020 BLM racial justice movement. Although this is a strength for evaluating conversations about racism when they are perhaps most likely to occur and contemporaneous with major societal upheavals, questions remain about how the frequency and content of these parent–child conversations will evolve or fade over time (e.g., Pew Research Center, 2023). Longitudinal designs that follow families and capture the lived and nuanced ways in which parents talk about sociopolitical events would provide invaluable insight. It would also provide an important missing piece: children’s responses, interpretations, and engagement with such conversations. We cannot know from these parent reports what the children said or did not say, how they responded, and how these BLM moments support children’s own developing critical awareness of racial injustices. Moreover, the duration and repetition of racial dialogue likely matters. Certainly, a single conversation or nod to the history of racism in the face of a current event of racial violence is not enough to change the course of the child’s racial socialization (and prevailing messages of white supremacy and anti-Blackness) that children will encounter from various sources within the family system and beyond it, including, school curricula, teachers, friends, and the media.

Our interpretations are also limited by our intentional design decision to recruit only Black and white parents. We restricted our focus in this way to ensure a sufficiently large sample of Black and white parents with children within a 3-year age span of interest. We also acknowledge the conceptual and practical challenges of accounting for the dynamics of multiple racial groups and positionalities, as well as the constraints imposed by the COVID-19 pandemic. That said, we encourage future research that engages a more racially diverse sample, as well as the relevant intersecting positionalities. For example, how did Asian American families navigate BLM as well as other racialized sociopolitical pressures such as anti-Asian rhetoric tied to COVID-19 and the movement to Stop Asian American Pacific Islander Hate? How were multiracial families moving through this period of racial salience and socializing their children to navigate their racial experiences and positionalities in this highly (bi)polarized context (Jones & Rogers, 2023)? Moreover, parents’ own racial identities and ideologies are likely important contributors to this process and more in-depth interviews with the parents about themselves could offer additional insight to whether, how, and the motivation for their discussing BLM with their children (e.g., Dull et al., 2023; A. J. Thomas & Speight, 1999). Each of the limitations provides a key directions for future research.

The pattern of empirical findings and our interpretations of them offer a window into how parents were navigating race at the height of the BLM movement, and what they said or thought they should say to their young children about the racism and racial violence happening in the United States. The findings suggest some progress, at least in terms of the prevalence of talking about race with children, especially among white parents, but there is still work to do. Our analysis focused on how conversation content may work to disrupt racial injustice or passively/actively allow it to persist. The parental responses within the Acknowledge Inequality and Affirm and Support codes suggested an antiracist stance by countering racial silence fueled by the dominant ideology of white supremacy and anti-Blackness, whereas the Uncritical Equality and Indifference codes are analogous to standing on a conveyor belt but not actively challenging, questioning, or disrupting its rhythm. Finally, a small but notable number of parent responses were Delegitimizing narratives about racism and BLM and can be seen as actively upholding racial injustice through harmful racial ideologies that perpetuate racism. Adding this interpretative frame broadens the racial socialization literature by situating parental race conversations as transformative levers that are both micro and macro, personal and political (Rogers, Niwa, et al., 2021, p. 280).

The pattern of findings obtained in this study has important implications for the design of interventions seeking to enrich parent–child conversations about racism. We do not think that promoting more race conversations is enough. As we further the work on racial socialization, attending to how parents address racism is important for understanding the role and potential of antiracist socialization (R. E. Anderson & Stevenson, 2019; Rogers, Niwa, et al., 2021). In particular, the current findings suggest that feeling unprepared is not a primary reason for not talking about race with children; this may mean that we want to do more to convey to parents why such conversations are necessary in the work of disrupting racial injustice (see Scott et al., 2022; Waxman, 2021). Nurturing critical discussion and awareness of racism in a society designed to make racial inequality and injustice invisible requires intentionality. Parents play a key role in socializing children, but we must also attend to the dynamic sociopolitical context in which parenting is situated and consider the ways we can leverage related systems, such as media and schooling, to support the practice of antiracist socialization (e.g., Rogers, Niwa, et al., 2021; H. Stevenson, 2014). Studying children’s developmental contexts as dynamic and interdependent within the broader chronosystem and sociopolitical systems of society will help disrupt notions of unidirectional impacts and allow us to see how parents are themselves responding social forces that may prompt them to shift what and how they converse with their children. The present work lays the foundation for evaluating such factors but is only a start for a m(ai)cro analysis and approach to studying racial development.

Abaied, J. L., & Perry, S. P. (2021). Socialization of racial ideology by White parents. Cultural Diversity and Ethnic Minority Psychology, 27(3), 431–440. 10.1037/cdp0000454

Abaied, J. L., Perry, S. P., Cheaito, A., & Ramirez, V. (2022). Racial socialization messages in white parents’ discussions of current events involving racism with their adolescents. Journal of Research on Adolescence, 32(3), 863–882. 10.1111/jora.12767

Anderson, R. A., & Masicampo, E. J. (2017). Protecting the innocence of youth: Moral sanctity values underlie censorship from young children. Personality and Social Psychology Bulletin, 43(11), 1503–1518. 10.1177/0146167217722557

Anderson, R. E., Jones, S. C. T., & Stevenson, H. C. (2020). The initial development and validation of the Racial Socialization Competency Scale: Quality and quantity. Cultural Diversity and Ethnic Minority Psychology, 26(4), 426–436. 10.1037/cdp0000316

Anderson, R. E., & Stevenson, H. C. (2019). RECASTing racial stress and trauma: Theorizing the healing potential of racial socialization in families. American Psychologist, 74(1), 63–75. 10.1037/amp0000392

Andone, D. (2021, November12). A timeline of the killing of Ahmaud Arbery and the case against 3 men accused of his murder. CNN. https://www.cnn.com/2021/10/18/us/ahmaud-arbery-case-timeline/index.html

Apfelbaum, E. P., Pauker, K., Ambady, N., Sommers, S. R., & Norton, M. I. (2008). Learning (not) to talk about race: When older children underperform in social categorization. Developmental Psychology, 44(5), 1513–1518. 10.1037/a0012835

Birks, M., Chapman, Y., & Francis, K. (2008). Memoing in qualitative research: Probing data and processes. Journal of Research in Nursing, 13(1), 68–75. 10.1177/1744987107081254

Black Lives Matter. (n.d.). Black Lives Matter. https://blacklivesmatter.com/

Bonilla-Silva, E., & Dietrich, D. (2011). The sweet enchantment of color-blind racism in Obamerica. The ANNALS of the American Academy of Political and Social Science, 634(1), 190–206. 10.1177/0002716210389702

Buchanan, L., Bui, Q., & Patel, J. K. (2020, July3). Black Lives Matter may be the largest movement in U.S. history. The New York Times. https://www.nytimes.com/interactive/2020/07/03/us/george-floyd-protests-crowd-size.html

Caughy, M. O., Nettles, S. M., & Lima, J. (2011). Profiles of racial socialization among African American parents: Correlates, context, and outcome. Journal of Child and Family Studies, 20(4), 491–502. 10.1007/s10826-010-9416-1

Cave, L., Cooper, M. N., Zubrick, S. R., & Shepherd, C. C. J. (2020). Racial discrimination and child and adolescent health in longitudinal studies: A systematic review. Social Science & Medicine, 250, Article 112864. 10.1016/j.socscimed.2020.112864

Coming Together: Standing Up to Racism. (2020, June6). In CNN. https://www.cnn.com/2020/06/06/app-news-section/cnn-sesame-street-race-town-hall-app-june-6-2020-app/index.html

Darnon, C., Wiederkehr, V., Dompnier, B., & Martinot, D. (2018). “Where there is a will, there is a way”: Belief in school meritocracy and the social-class achievement gap. British Journal of Social Psychology, 57(1), 250–262. 10.1111/bjso.12214

Dull, B. D., Rogers, L. O., Chae, D., & Meltzoff, A. (2023, March). The role of critical consciousness in white and Black parents’ racial socialization during summer 2020 [Poster presentation]. The Biennial Meeting for Society for Research on Child Development, Salt Lake City, UT, United States.

Eibach, R. P., & Ehrlinger, J. (2006). “Keep your eyes on the prize”: Reference points and racial differences in assessing progress toward equality. Personality and Social Psychology Bulletin, 32(1), 66–77. 10.1177/0146167205279585

Gollwitzer, A., Marshall, J., & Bargh, J. A. (2020). Pattern deviancy aversion predicts prejudice via a dislike of statistical minorities. Journal of Experimental Psychology: General, 149(5), 828–854. 10.1037/xge0000682

Hagerman, M. A. (2018). White kids: Growing up with privilege in a racially divided America. New York University Press. 10.18574/nyu/9781479880522.001.0001

History.com Editors. (2020). This day in history: Breonna Taylor is killed by police in botched raid.

Hughes, D., & Chen, L. (1997). When and what parents tell children about race: An examination of race-related socialization among African American families. Applied Developmental Science, 1(4), 200–214. 10.1207/s1532480xads0104_4

Hughes, D., Hagelskamp, C., Way, N., & Foust, M. D. (2009). The role of mothers’ and adolescents’ perceptions of ethnic-racial socialization in shaping ethnic-racial identity among early adolescent boys and girls. Journal of Youth and Adolescence, 38(5), 605–626. 10.1007/s10964-009-9399-7

Hughes, D., Rodriguez, J., Smith, E. P., Johnson, D. J., Stevenson, H. C., & Spicer, P. (2006). Parents’ ethnic-racial socialization practices: A review of research and directions for future study. Developmental Psychology, 42(5), 747–770. 10.1037/0012-1649.42.5.747

Hughes, D. L., & Watford, J. A. (2022). Racial regularities: Setting-level dynamics as a source of ethnic-racial socialization. American Journal of Community Psychology, 70(1–2), 3–17. 10.1002/ajcp.12565

Hughes, D. L., Watford, J. A., & Del Toro, J. (2016) A transactional/ecological perspective on ethnic–racial identity, socialization, and discrimination. In S. S.Horn, M. D.Ruck, & L. S.Liben (Eds.), Advances in child development and behavior: Volume 51. Equity and justice in developmental science: Implications for young people, families, and communities (pp. 1–41). Elsevier Academic Press.

Jones, C. M., & Rogers, L. O. (2023). Family racial/ethnic socialization through the lens of Multiracial Black Identity: A m(ai)cro analysis of meaning-making. Race and Social Problems, 15(1), 59–78. 10.1007/s12552-023-09387-6

Kalina, P. (2020). Performative allyship. Technium Social Sciences Journal, 11(1), 478–481. 10.47577/tssj.v11i1.1518

Lincoln, Y. S., & Guba, E. G. (1986). But is it rigorous? Trustworthiness and authenticity in naturalistic evaluation. New Directions for Program Evaluation, 1986(30), 73–84. 10.1002/ev.1427

Marcelo, A. K., & Yates, T. M. (2019). Young children’s ethnic–racial identity moderates the impact of early discrimination experiences on child behavior problems. Cultural Diversity and Ethnic Minority Psychology, 25(2), 253–265. 10.1037/cdp0000220

Marecek, J., Fine, M., & Kidder, L. (2001). Working between two worlds: Qualitative methods and psychology. In D. L.Tolman & M.Brydon-Miller (Eds.), From subjects to subjectivities: A handbook of interpretive and participatory methods (pp. 29–41). New York University Press.

Maxwell, J. A. (2005). Qualitative research design: An interactive approach. SAGE.

Media Insight Project. (2014, March17). The personal news cycle: How Americans choose to get news. American Press Institute. https://www.americanpressinstitute.org/publications/reports/survey-research/personal-news-cycle/

Molina, N. (2014). How race is made in America: Immigration, citizenship, and the historical power of racial scripts. University of California Press.

Nzinga, K., Rapp, D. N., Leatherwood, C., Easterday, M., Rogers, L. O., Gallagher, N., & Medin, D. L. (2018). Should social scientists be distanced from or engaged with the people they study?Proceedings of the National Academy of Sciences, 115(45), 11435–11441. 10.1073/pnas.1721167115

Paasch-Anderson, J., Lamborn, S. D., & Azen, R. (2019). Beyond what, to how: Different ways African American adolescents receive ethnic and racial socialization messages. Cultural Diversity and Ethnic Minority Psychology, 25(4), 566–578. 10.1037/cdp0000269

Pew Research Center. (2023, June14). Support for the Black Lives Matter movement has dropped considerably from its peak in 2020. https://www.pewresearch.org/social-trends/wp-content/uploads/sites/3/2023/06/ST_2023.06.14_BLM-Support_Report.pdf

Priest, N., Walton, J., White, F., Kowal, E., Baker, A., & Paradies, Y. (2014). Understanding the complexities of ethnic-racial socialization processes for both minority and majority groups: A 30-year systematic review. International Journal of Intercultural Relations, 43(2014), 139–155. 10.1016/j.ijintrel.2014.08.003

Rogers, L. O., Meltzoff, A. M., & Chae, D. (2023, May15). On parenting about race. https://osf.io/ne35y/?view_only=b79ca110b6d54faa9c8e4fd1d57d6db7

Rogers, L. O., & Meltzoff, A. N. (2017). Is gender more important and meaningful than race? An analysis of racial and gender identity among Black, White, and mixed-race children. Cultural Diversity and Ethnic Minority Psychology, 23(3), 323–334. 10.1037/cdp0000125

Rogers, L. O., Moffitt, U., & Jones, C. M. (2021). Listening for culture: Using interview analysis to understand identity in contexts. In K. C.McLean (Ed.), Cultural methodologies in psychology: Describing and transforming cultures (ch. 3, pp. 45–75). Oxford University Press. https://global.oup.com/academic/product/cultural-methods-inpsychology-9780190095949?cc=us&lang=en&#

Rogers, L. O., Niwa, E. Y., Chung, K., Yip, T., & Chae, D. (2021). M(ai)cro: Centering the macrosystem in human development. Human Development, 65(5–6), 270–292. 10.1159/000519630

Rogers, L. O., Rosario, R. J., Padilla, D., & Foo, C. (2021). “[I]t’s hard because it’s the cops that are killing us for stupid stuff”: Racial identity in the sociopolitical context of Black Lives Matter. Developmental Psychology, 57(1), 87–101. 10.1037/dev0001130

Rogers, L. O., Zosuls, K., Halim, M., Ruble, D., Hughes, D., & Fuligni, A. (2012). Meaning making in middle childhood: An exploration of the meaning of ethnic identity. Cultural Diversity & Ethnic Minority Psychology, 18(2), 99–108. 10.1037/a0027691

Ruble, D. N., Alvarez, J., Bachman, M., Cameron, J., Fuglini, A., Garcia Coll, C., & Rhee, E. (2004). The development of a sense of “we”: The emergence and implications of children’s collective identity. In M.Bennett & F.Sani (Eds.) The development of the social self (pp. 29–76). Psychology Press.

Ruck, M. D., Hughes, D. L., & Niwa, E. Y. (2021). Through the looking glass: Ethnic racial socialization among children and adolescents. Journal of Social Issues, 77(4), 943–963. 10.1111/josi.12495

Sani, F., & Bennett, M. (2004). Developmental aspects of social identity. In M.Bennett & F.Sani (Eds.), The development of the social self (ch. 3, pp. 91–114). Psychology Press.

Scott, K. E., Ash, T. L., Immel, B., Liebeck, M. A., Devine, P. G., & Shutts, K. (2022). Engaging White parents to address their White children’s racial biases in the Black-White context. Child Development, 94(1), 74–92. 10.1111/cdev.13840

Selby, D. (2020). They executed my brother in broad daylight: The unjust killing of George Floyd. The Innocence Project. Retrieved May, 2020, from https://innocenceproject.org/george-floyd-minneapolis-police-killing/

Shutts, K., Roben, C. K. P., & Spelke, E. S. (2013). Children’s use of social categories in thinking about people and social relationships. Journal of Cognition and Development, 14(1), 35–62. 10.1080/15248372.2011.638686

Skinner, A. L., Meltzoff, A. N., & Olson, K. R. (2017). “Catching” social bias: Exposure to biased nonverbal signals creates social biases in preschool children. Psychological Science, 28(2), 216–224. 10.1177/0956797616678930

Spencer, M. B. (2006). Phenomenology and ecological systems theory: Development of diverse groups. In R. M.Lerner & W.Damon (Eds.), Handbook of child psychology: Theoretical models of human development (pp. 829–893). John Wiley & Sons. https://psycnet.apa.org/record/2006-08774-015

Stevenson, H. (2014). Promoting racial literacy in schools: Differences that make a difference. Teachers College Press.

Stevenson, H. C., McNeil, J. D., Herrero-Taylor, T., & Davis, G. Y. (2005). Influence of perceived neighborhood diversity and racism experience on the racial socialization of Black youth. Journal of Black Psychology, 31(3), 273–290. 10.1177/0095798405278453

Stevenson, H. C., Reed, J., Bodison, P., & Bishop, A. (1997). Racism stress management: Racial socialization beliefs and the experience of depression and anger in African American youth. Youth & Society, 29(2), 197–222. 10.1177/0044118X97029002003

Sullivan, J. N., Eberhardt, J. L., & Roberts, S. O. (2021). Conversations about race in Black and White US families: Before and after George Floyd’s death. Proceedings of the National Academy of Sciences, 118(38), Article e2106366118. 10.1073/pnas.2106366118

Tatum, B. D. (2017). “Why are all the black kids sitting together in the cafeteria?”: And other conversations about race (3rd ed.). Basic Books.

Thomas, A. J., & Speight, S. L. (1999). Racial identity and racial socialization attitudes of African American parents. Journal of Black Psychology, 25(2), 152–170. 10.1177/0095798499025002002

Thomas, D. R. (2003). A general inductive approach for qualitative data analysis. https://www.fmhs.auckland.ac.nz/soph/centres/hrmas/_docs/Inductive2003.pdf

Umaña-Taylor, A. J., & Hill, N. E. (2020). Ethnic–racial socialization in the family: A decade’s advance on precursors and outcomes. Journal of Marriage and Family, 82(1), 244–271. 10.1111/jomf.12622

Underhill, M. R. (2018). Parenting during Ferguson: Making sense of white parents’ silence. Ethnic and Racial Studies, 41(11), 1934–1951. 10.1080/01419870.2017.1375132

Wang, M.-T., Henry, D. A., Smith, L. V., Huguley, J. P., & Guo, J. (2020). Parental ethnic-racial socialization practices and children of color’s psychosocial and behavioral adjustment: A systematic review and meta-analysis. American Psychologist, 75(1), 1–22. 10.1037/amp0000464

Waxman, S. R. (2021). Racial awareness and bias begin early: Developmental entry points, challenges, and a call to action. Perspectives on Psychological Science, 16(5), 893–902. 10.1177/17456916211026968

Wolcott, H. F. (1994). Transforming qualitative data: Description, analysis, and interpretation. SAGE Publications.

Xian, H., & Reynolds, J. (2017). Bootstraps, buddies, and bribes: Perceived meritocracy in the United States and China. The Sociological Quarterly, 58(4), 622–647. 10.1080/00380253.2017.1331719

Zucker, J. K., & Patterson, M. M. (2018). Racial socialization practices among white American parents: Relations to racial attitudes, racial identity, and school diversity. Journal of Family Issues, 39(16), 3903–3930. 10.1177/0192513X18800766

Submitted: January 25, 2023 Revised: November 8, 2023 Accepted: November 14, 2023