Abstract
Leading up to the 2020 U.S. presidential election, Native American organizations and tribes launched get-out-the-vote campaigns that motivated Native peoples to vote in record numbers and helped flip battleground states. We conducted four studies (total N = 11,661 Native American adults) to examine the social and cultural factors explaining this historic Native civic engagement (e.g., campaigning). Results revealed that the more participants identified as being Native, the more they reported (a) engaging in civic activities, including get-out-the-vote behaviors during the 2020 election (Study 1); (b) civic engagement more broadly across a 5-year period (pilot study, Study 2); and (c) intentions to engage in civic activities in the future (Study 3). Moreover, participants who more strongly identified as Native were more likely to recognize the omission of their group from society and perceive greater group discrimination, which both independently and serially predicted greater civic engagement. These results suggest that leveraging the link between Native identification and group injustices can motivate action.
Keywords
Indigenous Peoples remain severely under-resourced, under-counted, and underestimated. . . . If this is important to you, . . . [and] your people, Rock the Native Vote!
Leading up to the 2020 U.S. presidential election, Native American organizers and communities launched a series of campaigns to motivate Native peoples to vote. Election data revealed that a record number of Native peoples voted (Smith, 2020), which contributed to flipping battleground states (i.e., Arizona, Minnesota, and Wisconsin; NoiseCat, 2020) and electing a record-setting six Native Americans to Congress (O’Kane, 2020). The effectiveness of these voting campaigns has meaningful and novel implications for social psychological theories. We posit that these campaigns successfully galvanized Native voters because they motivated individuals who strongly identified as Native to address bias and discrimination toward their group.
Take the Rock the Rez quote as an example. The campaign drew attention to the undercounting and underestimation of Native peoples, which hints at Native omission, a form of bias constituted by the scarcity and absence of Native representations that leaves Native peoples’ experiences, perspectives, and even existence out of public consciousness (Fryberg & Eason, 2017). “Under-resourced” alludes to the discrimination Native peoples face, which arguably stems from omission. When Native peoples are omitted from public discourse, resources and policies that promote equity are not directed toward their communities. Finally, by linking the importance of these issues to individuals and communities, the campaign tapped into social-identity theorizing suggesting that individuals who strongly identify as Native view their ingroup’s experiences as central to their self-concept (Leach et al., 2008). Together, the campaign capitalized on the idea that people who strongly identify as Native recognized more omission, which in turn shaped their perceptions of discrimination, thereby promoting greater civic engagement (defined as participation in actions, such as voting, that strengthen one’s community; Diller, 2001; see Fig. 1 for proposed model).
Proposed model.
Social psychological research demonstrates that people who strongly identify as members of a marginalized group engage in more civic activities that can enhance their group’s societal status (Ellemers & Haslam, 2012; Tajfel, 1978). This tendency is partially because higher group identification is related to greater perceptions of discrimination facing one’s group (Branscombe et al., 1999; Inzlicht et al., 2009; Kaiser & Wilkins, 2010; Major et al., 2003; Schmader et al., 2015; Schweigman et al., 2011; Sellers & Shelton, 2003; Simon et al., 1998; Tropp et al., 2012; van Zomeren et al., 2008; Wright & Littleford, 2002; Yasui et al., 2015). For example, students who more strongly identified as Latine perceived greater discrimination and subsequently reported greater intentions to vote, attend demonstrations, and sign petitions (Cronin et al., 2012).
Statement of Relevance
Native peoples and their experiences are constantly omitted from society (e.g., from news, congressional hearings, and policymaking). Consequently, U.S. social and legal systems fail to allocate resources and make policies to rectify community-based injustices, thereby perpetuating discrimination and injustices facing Native peoples. Like other domains, social science has long failed to give voices to Native peoples, and thus psychological literature has not empirically examined how Native peoples psychologically and behaviorally engage with this omission. Our research is the first to demonstrate that to the extent that Native peoples identify with their group, recognition of being omitted from society promotes a greater sense of being discriminated against and further motivates engagement in civic activities to help change the status quo. These findings suggest that Native omission is a critical component contributing to Native peoples’ understanding of their group’s position in society and that civic engagement may be one way they can contend with this omission.
Indigenous scholarship highlights that discrimination against Native peoples partly stems from omission, a unique form of bias facing Native peoples across numerous domains (Deloria et al., 2018; Eason et al., 2018; Fryberg & Eason, 2017; Lomawaima & McCarty, 2014; Rifkin, 2011; Robertson, 2015; Veracini, 2011), including media (Cooperative Children’s Book Center, 2020; Nielsen, 2020), education (Shear et al., 2015), and academic research (Brady et al., 2018; Fryberg & Eason, 2017; Lopez et al., 2022). For example, less than 1% of prime-time television, film, or video-game characters are Native (Tukachinsky et al., 2015). Given prior findings that individuals who strongly identify with their group are more likely to notice and search for information pertaining to their group (Operario & Fiske, 2001; Turner et al., 1987), we theorized that people who more strongly identified as Native would be more likely to recognize the absence or omission of Native representations.
We also anticipated that recognizing omission would shape individuals’ perceptions of the discrimination faced by their group. First, biased social representations lay the foundation for discrimination because they provide individuals with a script regarding the appropriate ways to think, feel, and behave (Moscovici, 1981). For example, exposure to Native mascots—a social representation that locates Native peoples as historical figures and thereby omits their contemporary existence—leads non-Native individuals to associate Native peoples with negative stereotypes (Angle et al., 2017; Burkley et al., 2017; Freng & Willis-Esqueda, 2011), which promotes discriminatory behaviors, such as devaluing Native students’ academic potential (Chaney et al., 2011; Dai et al., 2021).
Second, from the target’s perspective, social representations denote the group’s social standing and worth (Moscovici, 1988). Viewing derogatory or stereotypical representations of one’s group contributes to one’s perceptions of group discrimination (Ortiz & Behm-Morawitz, 2015; Rivadeneyra et al., 2007; Schmader et al., 2015). For example, the more Black individuals evaluate Black media representations as negative, the more they think outgroup members perceive Black people as having low status (Fujioka, 2005). Although empirical literature focuses on the consequences of misrepresentations, we reasoned that omission—an absence of representations—would send a similarly strong cue (i.e., the absence of a representation is a representation) that the group is devalued in society, thereby shaping perceptions of group discrimination (Fryberg & Townsend, 2008).
Overall, we hypothesized that (a) stronger Native identification would predict greater civic engagement, (b) recognition of Native omission and perceived group discrimination would each independently explain the link between Native identification and civic engagement, and (c) because recognition of Native omission is theorized to contribute to perceptions of group discrimination, these two social and cultural mechanisms would serially explain the link between Native identification and civic engagement.
Open Practices Statement
The pilot study was not preregistered; the preregistration for Studies 1 and 2 can be accessed at https://osf.io/5udrf/; the preregistration for Study 3 can be accessed at https://osf.io/nvd3g/. All supplements, complete materials, deidentified data for the four studies, and data-analysis scripts can be accessed at https://osf.io/76rt3/. All studies in this article were approved by the University of California, Berkeley Institutional Review Board (No. 2019-08-12485). We obtained informed consent from all participants.
Pilot Study
Because of a lack of empirical evidence from previous research on Native omission and its consequences, we conducted a pilot study to examine the plausibility of the proposed model.
Method
The data used in the pilot study were a segment of the data set from a large-scale survey named the Indigenous Futures Survey, which was designed to investigate Native peoples’ priorities, needs, and perspectives related to current political issues and systemic racism. We randomly selected 500 responses for preliminary analysis in the pilot study. The rest of the data set was used for Study 2, which was preregistered on the basis of the pilot study findings.
Participants
A 15-min online survey was distributed by 46 Native organizations, 75 tribes, 60 tribal colleges and/or Native college/university student organizations, and five Native media outlets. Given that Native peoples are hard-to-reach populations and consist of numerous tribes with rich languages and cultures, to represent the diverse voices within the Native communities, we tried to reach and recruit as many Native participants as possible. The data collection ended on a date that the research team and the organizations agreed on beforehand. We used this rule to determine the sample size for all the studies reported in this article. In all, 6,461 Native participants completed the survey in exchange for the opportunity to enter a raffle. The pilot study used responses from 500 randomly selected participants (see Table 1 for demographics).
Demographics of Samples in the Pilot Study and Studies 1 to 3
Note: Responses were missing for some demographic variables, resulting in percentages not summing to 100%.
Materials and procedure
For full materials for all the studies and for bivariate relationships between all variables in all the studies, see “Measures & Materials” and “Bivariate Correlation Tables,” respectively, at https://osf.io/76rt3/. In addition, supplements for all the studies can be accessed at this link. In this article, we focus on a subset of the measures from the larger study, which are presented below.
Native identification
To assess the extent to which participants identified with being Native, we asked participants to complete three items adapted from the Attachment—Importance subscale of Roccas and colleagues’ (2008) identification measure in regard to Native identity (e.g., “Being Native American is an important part of my identity”; 1 = Strongly Disagree, 5 = Strongly Agree; M = 4.62, SD = 0.59, α = .74).
Recognition of Native omission
A single item assessed the extent to which participants generally recognize the omission of Native peoples from public consciousness (“I notice when Native peoples are not represented”; 1 = Strongly Disagree, 5 = Strongly Agree; M = 4.65, SD = 0.55).
Perceived group discrimination
A single item assessed the extent to which participants perceived discrimination against their group (“Native Americans experience discrimination”; 1 = Strongly Disagree, 5 = Strongly Agree; M = 4.69, SD = 0.64). We counterbalanced the order in which participants received the measures of recognition of Native omission and perceived group discrimination.
Civic engagement
To assess civic engagement, we asked participants to indicate whether they engaged in a list of nine civic activities over the last 5 years (e.g., “participated in a community action group/grassroots organization,” “signed a petition”; M = 5.09, SD = 2.68, range = 0–9).
Primary model analytic plan
On the basis of prior social psychological and Indigenous scholarship, we hypothesized that higher Native identification would be related to greater reported engagement in civic activities. Further, we predicted that the link between Native identification and civic engagement would be at least partially explained by (a) participants who more strongly identified as Native being more likely to recognize the omission of their group, (b) participants who more strongly identified as Native being more likely to perceive group discrimination, and (c) the serial relationship between recognition of omission and perceptions of discrimination (see Fig. 1).
To test these predictions, we conducted serial mediation analyses. We also checked for the robustness of our results by conducting analyses with and without relevant demographic covariates (i.e., residence, political orientation, and political party affiliation; for details, see “Measures & Materials” at https://osf.io/76rt3/). All analyses were conducted with the lavaan package (Rosseel, 2012) in R (R Core Team, 2018) using a bias-corrected and accelerated bootstrapping procedure with 5,000 resamples. Because this study served as an initial test of the proposed model, we did not preregister the hypotheses or the analytic plan.
Results
Preliminary analysis: order effects
We first examined whether the order in which participants received the Recognition of Native Omission and Perceived Group Discrimination measures affected responses to these items and the proposed model. Because we counterbalanced the order of these two measures across all four studies and found consistent patterns throughout, we report the order effects for all studies here for ease of presentation. On average, participants’ responses to the Recognition of Native Omission and Perceived Group Discrimination measures did not differ on the basis of the order in which they received the measures (ps > .201). Moreover, we tested the proposed model separately among participants who received the Recognition of Native Omission measure first and those who received the Perceived Group Discrimination measure first. In this and all subsequent studies, the results and interpretations for each presentation order were consistent with the models using data collapsed across all participants (for results, see “Order Effects of Recognition of Native Omission and Perceived Group Discrimination Measures” at https://osf.io/76rt3/). Therefore, analyses reported in this article were collapsed across all participants.
Total effect: relation between Native identification and civic engagement
See Table 2 for model results. Although results were in the predicted direction, higher Native identification was not significantly associated with greater civic engagement over the past 5 years (p = .157).
Results of Serial Mediation Models Predicting Civic Engagement in the Pilot Study and Studies 1 to 3
Note: In the pilot study and Studies 2 and 3, we used only general recognition of Native omission as the operationalization of recognition of omission. CI = confidence interval.
p < .001.
Indirect effect through recognition of omission
Higher Native identification was related to greater recognition of omission (p < .001), and in turn, the more participants reported recognizing omission, the more they reported engaging in civic activities over the past 5 years (p < .001). The indirect effect was significant (see Table 2), indicating that participants who more strongly identified as Native reported greater engagement in civic activities, in part because they were more likely to recognize that their group is omitted from society.
Indirect effect through perceived group discrimination
Higher Native identification did not significantly predict perceived group discrimination (p = .079), and yet greater perceived group discrimination was related to greater engagement in civic activities over the past 5 years (p < .001). The indirect effect was significant (see Table 2). These results suggest that participants who more strongly identified as Native reported greater engagement in civic activities, in part because they were more likely to perceive discrimination against their group.
Serial indirect effect through recognition of omission predicting perceived group discrimination
Greater recognition of omission was related to greater perceived group discrimination (p < .001). Moreover, supporting the proposed model, the serial indirect effect was significant (see Table 2); specifically, the more Native participants identified with being Native, the more they recognized Native omission and subsequently perceived greater group discrimination, ultimately predicting greater engagement in civic activities across the past 5 years.
Covariates
Finally, including the covariates changed the model results of the pilot study such that the direct effect of perceived group discrimination on civic engagement, the indirect effect of Native identification on civic engagement through perceived group discrimination, and the serial indirect effect were no longer significant (see “Model Results With Covariates” at https://osf.io/76rt3/). Given the relatively small sample in relation to the large number of variables in this model, we will examine whether these inconsistencies remain in Studies 1 to 3 with larger samples.
Discussion
The pilot study provided preliminary support for the proposition that recognition of omission and perception of group discrimination are important factors explaining civic engagement of people who strongly identify as Native. These findings also constitute the first empirical evidence that being omitted from society may be a critical component shaping Native peoples’ understanding of the discrimination their group faces.
Study 1
Study 1 was a preregistered, confirmatory conceptual replication of the proposed model with increased statistical power. In the context of live coverage of the 2020 U.S. presidential election, CNN provided a graphic that identified White, Latino, Black, and Asian voters but lumped Native voters into a “something else” category (Forester, 2020). Capitalizing on the 2020 election and this incident of Native omission, we examined whether recognition of this specific incident of omission, along with recognition of omission in general, predicted one particular form of civic engagement—get-out-the-vote activities.
Method
Participants
A short 5-min online survey was distributed by three Native organizations (IllumiNative, Native Organizers Alliance, and the Center for Native American Youth at The Aspen Institute) via social media promotions and word of mouth. In all, 3,586 Native participants completed the survey in exchange for the opportunity to enter a raffle (see Table 1 for demographics).
Materials and procedure
The data presented in Study 1 came from a larger survey named the “Something Else” Survey, which was designed to investigate the relationship between Native peoples’ civic engagement leading up to the 2020 presidential election and their responses to a specific omission (i.e., CNN lumping Native voters into the “something else” category) during the election news coverage. The “Something Else” Survey was distributed after the Indigenous Futures Survey used in the pilot study and Study 2. We switched the order of the first two studies in this article for ease of presentation. For the purposes of this study, we focused on the relationship between a subset of measures relevant to the proposed model, as presented below.
Native identification
To assess the extent to which participants identified with being Native, we asked participants to complete Luhtanen and Crocker’s (1992) four-item racial-identity-centrality measure in regard to Native identity (e.g., “Being Native is an important reflection of who I am”; 1 = Strongly disagree, 7 = Strongly agree; M = 6.18, SD = 1.00, α = .61).
Recognition of omission
Because we defined Native omission as the absence of Native representations, thereby leaving Native peoples invisible, the extent to which Native individuals are able to notice daily incidents in which the representation of their group is left out should serve as an index of their ability to recognize Native omissions in general. Hence, we operationalized recognition of omission in two ways: (a) recognition of Native omission and (b) recognition of a specific incident of Native omission.
We used the same single item as in the pilot study to assess recognition of omission, except on a 7-point Likert scale (1 = Strongly disagree, 7 = Strongly agree; M = 6.56, SD = 0.91). Next, participants indicated the extent to which they felt invisible (1 = Not at all, 4 = A great deal; M = 3.51, SD = 0.84) after viewing a graphic from CNN’s live election coverage of the 2020 presidential election (Forester, 2020). In the graphic, White, Latine, Black, and Asian voters were explicitly labeled, but Native voters were lumped into a “something else” category.
Perceived group discrimination
We used the same single item as in the pilot study to measure perceived group discrimination, except on a 7-point Likert scale (1 = Strongly disagree, 7 = Strongly agree; M = 6.60, SD = 1.03).
Engagement in get-out-the-vote activities
To assess civic engagement, we asked participants to indicate whether they engaged in a list of seven get-out-the-vote activities during the 2020 election (e.g., “drove people to the polls,” “texted people to vote”; M = 1.71, SD = 1.17, range = 0–7).
Analytic plan
Regarding the primary model (Fig. 1), we preregistered and followed the same hypotheses and analytic plan as outlined in the pilot study Primary Model Analytic Plan section. As with the pilot study and as preregistered, we checked the robustness of our results by conducting analyses with and without relevant demographic covariates (i.e., residence, political orientation, and political party affiliation; for measures of covariates, see “Measures & Materials” at https://osf.io/76rt3/).
We also recognized that different civic activities involve different levels of commitment and risk, which may impact engagement. Hence, we preregistered our intention to examine whether the model would hold for civic activities high in commitment and risk. To determine which civic engagement items to include, we recruited participants from different racial/ethnic backgrounds (Study 1: n = 17, Study 2: n = 22, Study 3: n = 26) to rate the extent to which each of the civic activities required commitment and involved risk on 5-point scales (1 = Not at all, 5 = Extremely). As preregistered, high-commitment, high-risk activities were items with scores 1 standard deviation above the mean in both the commitment and the risk ratings (for items identified as high-commitment, high-risk in each study, see “High-Commitment High-Risk Civic Engagement” at https://osf.io/76rt3/). We then tested the primary model with high-commitment, high-risk civic engagement as the outcome.
In addition, given that the study was cross-sectional, we further clarified the relationships among the variables by testing two alternative models: one with mediators reversed such that perceived group discrimination predicted recognition of Native omission and the other with the two mechanisms acting as parallel mediators that independently explained the relation between Native identification and civic engagement. We did not preregister the alternative model analyses for Studies 1 to 3.
In Study 1, we conducted all the aforementioned analyses with recognition of Native omission in general as well as recognition of a specific incident of omission (i.e., CNN putting Native voters in the “something else” category).
Results
Primary analyses
Total effects: relation between Native identification and engagement in get-out-the-vote activities
See Table 2 for model results. As expected, across both the primary models, higher Native identification was associated with greater engagement in get-out-the-vote activities (all ps < .001). That is, the more participants identified as Native, the more get-out-the-vote activities they reported engaging in during the 2020 presidential election.
Indirect effects through recognition of omission
As hypothesized, higher Native identification was related to both greater recognition of omission in general (p < .001) and greater recognition of a specific incident of omission (p < .001), and in turn, the more participants reported recognizing omission (both generally and specifically), the more they reported engaging in get-out-the-vote activities (all ps < .001). The indirect effects were significant (see Table 2). Supporting the hypothesis, results showed that participants who identified more strongly as Native reported greater engagement in get-out-the-vote activities, partly because they were more likely to recognize that their group is omitted from society.
Indirect effects through perceived group discrimination
As predicted, higher Native identification was related to greater perceived group discrimination (all ps < .007). Greater perceived group discrimination, however, was not related to greater engagement in get-out-the-vote activities (all ps > .175). The indirect effects were not significant (see Table 2), indicating that perceptions of group discrimination could not explain why participants who identified more strongly as Native reported engaging in more get-out-the-vote activities.
Serial indirect effects through recognition of omission predicting perceived group discrimination
As anticipated, greater recognition of omission (both generally and specifically) was related to greater perceived group discrimination (all ps < .001). Diverging from the proposed model, however, the serial indirect effects were not significant (because perceived discrimination did not predict get-out-the-vote activities).
Covariates
Finally, unlike in the pilot study, both primary models remained consistent when we included the demographic covariates (see “Model Results With Covariates” at https://osf.io/76rt3/).
Models predicting high-commitment, high-risk civic engagement
The results remained consistent with the primary model results; specifically, participants who identified more strongly as Native recognized more Native omission (both generally and specifically) and subsequently engaged in more high-commitment, high-risk get-out-the-vote activities. Yet, the effects of perceived group discrimination on engagement were not significant. Neither were the serial indirect effects (for model results, see “High-Commitment High-Risk Civic Engagement” at https://osf.io/76rt3/).
Alternative model testing
For results, see “Alternative Models Testing” at https://osf.io/76rt3/. Different from the primary models, the alternative models—one using general recognition of omission and one using recognition of a specific incident of omission—with reversed mediators yielded significant serial indirect effects. Nonetheless, we do not necessarily take this as strong evidence that the alternative specification is more valid than the primary model. First, only one relationship accounts for the difference between models. Specifically, we found that group discrimination was not a significant predictor of civic engagement. Therefore, when group discrimination was the second mediator (as in the primary model), the serial indirect effects were not significant. In contrast, recognition of omission (both generally and specifically) significantly predicted civic engagement. Therefore, when recognition of omission was the second mediator (as in the alternative model), the serial indirect effects were significant. Second, the fact that perceived group discrimination was not a significant predictor of civic engagement is inconsistent with a large body of research (e.g., van Zomeren et al., 2008). Therefore, we would need to replicate these findings before reading too much into differential patterns of significance in the primary and alternative models.
Regarding the second set of alternative models with parallel mediators, both recognition of omission (both generally and specifically) and perceived discrimination were significant mediators. Nonetheless, the indirect effects of Native identification on civic engagement through recognition of omission were numerically stronger than the indirect effects through perceived discrimination. We further discuss the implications of these alternative model results in the General Discussion.
Discussion
Study 1 provided partial support for the proposed model. First, participants who identified more strongly as Native reported engaging in more get-out-the-vote activities during the 2020 election season. Second, participants who identified more strongly as Native were more likely to recognize general and specific omissions of Native peoples. This recognition predicted greater perceived discrimination and civic engagement.
Contrary to hypotheses and prior literature (van Zomeren et al., 2008), Study 1’s results showed that perceived discrimination did not predict engagement in get-out-the-vote activities. One potential explanation is that get-out-the-vote activities, especially during the COVID-19 pandemic, may be an overly narrow take on civic engagement. Specifically, Native peoples face disproportionately high disenfranchisement rates, including during the 2020 presidential election (Mansoor, 2020). Moreover, Native peoples have been among the groups most adversely impacted by COVID-19 (Centers for Disease Control and Prevention, 2020). Thus, expanding the types of civic engagement and the time frame to include prepandemic engagement may reveal an association between perceived discrimination and civic engagement.
Study 2
Study 2 was a preregistered test of the model that assessed engagement in a wider range of civic activities across a 5-year time span, including prepandemic engagement.
Method
The data in Study 2 came from the same “Indigenous Futures Survey” as used in the pilot study. Responses from the 500 randomly selected participants used in the pilot study were excluded from Study 2 analyses.
Participants
See the Participants section in the pilot study for the recruitment procedure. After excluding the 500 responses used in the pilot study, we had responses from 5,961 Native respondents for use in Study 2 (see Table 1 for demographics).
Materials and procedure
Because Study 2 data were from the same survey as used in the pilot study, the measures were the same as in the pilot study: Native identification (M = 4.60, SD = 0.59, α = .73), recognition of Native omission (M = 4.52, SD = 0.66), perceived group discrimination (M = 4.60, SD = 0.65), and civic engagement during the past 5 years (M = 3.73, SD = 2.41, range = 0–9).
Results
We followed the same analytic plan as in Study 1, including testing the primary model; checking the robustness of the model by accounting for demographic covariates; testing the primary model predicting high-commitment, high-risk civic engagement; and testing two alternative models. As with Study 1, the first three sets of analyses were preregistered, but the alternative model testing was not.
Primary analyses
Total effect: relation between Native identification and civic engagement
See Table 2 for model results. Consistent with Study 1’s findings, results showed that higher Native identification was associated with greater civic engagement (p < .001); specifically, the more participants identified with being Native, the more civic activities they reported engaging in over the past 5 years.
Indirect effect through recognition of omission
As hypothesized, higher Native identification was related to greater recognition of omission (p < .001), and in turn, the more participants reported recognizing omission, the more they reported engaging in civic activities over the past 5 years (p < .001). The indirect effect was significant (see Table 2). Consistent with the findings in the pilot study and Study 1, these results support the hypothesis that participants who more strongly identified as Native reported greater civic engagement, in part because they were more likely to recognize that their group is omitted from society.
Indirect effect through perceived group discrimination
As predicted, higher Native identification was related to greater perceived group discrimination (p < .001), and greater perceived discrimination was related to greater engagement in civic activities in the past 5 years (p < .001). In line with the proposed model and the results of the pilot study, but in contrast to those of Study 1, the indirect effect was significant (see Table 2), suggesting that participants who more strongly identified as Native reported greater engagement in civic activities, partly because they were more likely to perceive discrimination against their group.
Serial indirect effect through recognition of omission predicting perceived group discrimination
As anticipated, greater recognition of omission was related to greater perceived group discrimination (p < .001). Moreover, supporting the proposed model and pilot study results but diverging from those of Study 1, the serial indirect effect was significant (see Table 2); specifically, participants who more strongly identified as Native recognized more Native omission and subsequently perceived greater group discrimination, which ultimately predicted greater civic engagement over the past 5 years.
Covariates
Finally, the model results and interpretations remained consistent when we included the demographic covariates (see “Model Results With Covariates” at https://osf.io/76rt3/).
Model predicting high-commitment, high-risk civic engagement
Consistent with the findings in Study 1 and supporting our hypotheses, analyses showed that the model using high-commitment, high-risk civic engagement yielded similar results as in the primary analysis, suggesting that participants who identified more strongly as Native recognized more Native omission and subsequently perceived greater group discrimination, which ultimately predicted greater engagement in high-commitment, high-risk activities over the past 5 years (see “High-Commitment High-Risk Civic Engagement” at https://osf.io/76rt3/).
Alternative model testing
The alternative model with reversed mediators yielded results consistent with those of the primary model. Regarding the alternative model with parallel mediators, both recognition of omission and perceived discrimination were significant mediators. Still, the indirect effect of Native identification on civic engagement through recognition of omission was numerically stronger than the indirect effect through perceived discrimination (for results, see “Alternative Models Testing” at https://osf.io/76rt3/).
Discussion
In support of the proposed model, Study 2’s results showed that participants who identified more strongly as Native reported greater engagement in general civic activities over the past 5 years. This relationship was explained by recognition of Native omission and perceived group discrimination independently and serially. Nonetheless, measuring past civic engagement made it hard to ascertain whether civic engagement was indeed an outcome of the proposed psychological processes.
Study 3
Study 3 had two aims. First, to establish temporal ordering, we measured intentions to engage in future civic activities. Second, to further establish omission and perceived group discrimination as unique constructs, we added two discrimination measures.
Method
Participants
A 15-min online survey was distributed by Native organizations via social media promotions and word of mouth. In all, 1,609 Native participants completed the survey in exchange for the opportunity to enter a raffle (see Table 1 for demographics).
Materials and procedure
The data presented in Study 3 came from a larger survey named the “Indigenous Futures Survey 2.0,” which was conducted in the lead-up to the 2022 midterm elections and was part of an annual, cross-sectional project designed to investigate Native peoples’ priorities, needs, and perspectives related to current political issues and systemic racism. For the purposes of this study, we focused on the relationship between a subset of measures relevant to the proposed model. We used the same measures in the pilot study and Study 2: Native identification (M = 4.46, SD = 0.70, α = .75), recognition of Native omission (M = 4.43, SD = 0.72), and perceived group discrimination (M = 4.51, SD = 0.73). The only differences were two additional discrimination measures and the civic-engagement-intentions measure.
Alternative measures of group discrimination
In prior studies, one could argue that recognition of omission and perceived discrimination emerged as unique predictors because of superficial differences in wording. To alleviate this concern, we added two additional items measuring group discrimination using wording similar to the omission measure: “I experience discrimination because I am Native” (1 = Strongly disagree, 5 = Strongly agree; M = 3.74, SD = 1.05) and “I notice when Natives are discriminated against” (1 = Strongly disagree, 5 = Strongly agree; M = 4.40, SD = 0.70).
Civic engagement intentions
To assess civic engagement intentions, we asked participants to indicate how likely they would be to engage in a list of nine civic activities over the next 12 months (e.g., “Protest against the poor treatment of Natives”; 1 = I will never do this, 5 = I will do this for sure; M = 3.50, SD = 0.80, α = .84).
Results
We followed the same analytic plan outlined in Study 1, including testing the primary model; checking the robustness of the model by accounting for demographic covariates (the results were consistent with or without covariates; see “Model Results With Covariates” at https://osf.io/76rt3/); testing the primary model predicting high-commitment, high-risk civic engagement; and testing two alternative models. Finally, expanding on Study 1’s plan, we tested additional models using the alternative wordings of the perceived-group-discrimination measure. The first three sets of analyses were preregistered, but the alternative model testing and the models using additional discrimination measures were not.
Primary analyses
Total effect: relation between Native identification and civic engagement intentions
See Table 2 for model results. Conceptually replicating Studies 1 and 2, results showed that higher Native identification was associated with greater civic engagement intentions (p < .001); specifically, the more strongly participants identified as Native, the more they reported planning to engage in civic activities over the next 12 months.
Indirect effect through recognition of omission
As hypothesized, higher Native identification was related to greater recognition of omission (p < .001), and in turn, the more participants reported recognizing omission, the more they reported planning to engage in civic activities over the next 12 months (p < .001). The indirect effect was significant (see Table 2), suggesting that participants who identified more strongly as Native reported greater intentions to engage in civic activities, in part because they were more likely to recognize that their group is omitted from society.
Indirect effect through perceived group discrimination
As predicted, higher Native identification was related to greater perceived group discrimination (p < .001), and greater perceived discrimination was related to greater civic engagement intentions (p < .001). In line with Study 2’s results, the indirect effect was significant (see Table 2), suggesting that participants who identified more strongly as Native reported greater intentions to engage in civic activities, partly because they were more likely to perceive discrimination against their group.
Serial indirect effect through recognition of omission predicting perceived group discrimination
As anticipated, greater recognition of omission was related to greater perceived group discrimination (p < .001). Moreover, supporting the proposed model and replicating the results of Study 2, the serial indirect effect was significant (see Table 2); specifically, participants who identified more strongly as Native recognized more Native omission and subsequently perceived greater group discrimination, which ultimately predicted greater civic engagement intentions.
Covariates
Finally, differing from the results of the pilot study but in line with those of Studies 1 and 2, the model remained consistent when we included the demographic covariates (see “Model Results With Covariates” at https://osf.io/76rt3/).
Model predicting high-commitment, high-risk civic engagement intentions
Consistent with findings in Studies 1 and 2 and our hypotheses, results showed that the model using high-commitment, high-risk civic engagement were in line with those of the primary analysis, suggesting that participants who identified more strongly as Native recognized more Native omission and subsequently perceived greater group discrimination, which ultimately predicted greater intentions to engage in high-commitment, high-risk activities over the next 12 months (see “High-Commitment High-Risk Civic Engagement” at https://osf.io/76rt3/).
Alternative model testing
The alternative model with reversed mediators yielded results consistent with those of the primary model. Regarding the alternative model with parallel mediators, both recognition of omission and perceived discrimination were significant mediators. Still, the indirect effect of Native identification on civic engagement through recognition of omission was numerically stronger than the indirect effect through perceived discrimination (for results, see “Alternative Models Testing” at https://osf.io/76rt3/).
Finally, the models using the two alternative wordings of group discrimination items also yielded results and interpretations consistent with those of the model using the original discrimination measure. This provides additional evidence that the omission and perceived-discrimination measures had unique, independent effects that did not emerge in prior studies simply because of wording (for bivariate correlations and model results, see “Models Using Alternative Discrimination Measures” at https://osf.io/76rt3/).
Discussion
Study 3 demonstrated that participants who identified more strongly as Native reported stronger intentions to engage in civic activities over the next 12 months. Conceptually replicating the findings of the pilot study and Study 2, Study 3’s results showed that recognition of Native omission and perception of group discrimination both independently and serially explained the relation between Native identification and civic engagement intentions.
General Discussion
Across four studies, we found that individuals who identified more strongly as Native reported greater past civic engagement (pilot study, Studies 1 and 2) and greater future civic engagement intentions (Study 3). Two factors—recognition of omission and perceptions of group discrimination—independently (all four studies) and serially (pilot study, Studies 2 and 3) explained this link. Specifically, for individuals who identified strongly as Native, omission fueled perceptions of discrimination and galvanized civic engagement.
This research extends the civic engagement literature in two ways. First, the four studies constitute the first empirical evidence of factors that galvanize Native peoples’ civic engagement. Native peoples have long been omitted from mainstream society, which can trigger withdrawal responses such as unwillingness to participate in societal-level activities, including civic engagement (Mazzoni et al., 2020; Wiley et al., 2013). Opting out from mainstream society further reifies the cycle of silencing marginalized voices and overlooking and not addressing their needs. By investigating what keeps Native peoples engaged, we shed light on how to break this cycle and work toward a democracy representing all groups’ experiences and perspectives.
Second, prior research has primarily focused on how discrimination stemming from tangible biases motivates civic engagement. Although Native peoples experience these types of biases (e.g., the stereotype of Native peoples as warlike), we explored biases of omission that have never been empirically tested in psychological research. Across all studies, recognition of omission was a stronger and more unique predictor of civic engagement than perceptions of discrimination. The difference between tangible and invisible played out in Study 1 when CNN labeled Native voters as “something else.” Native individuals recognized this act as an omission of their contributions to the 2020 election. They responded with immense online and public outrage, called for a public retraction, and demanded an apology. Although omissions do not fit within traditional frameworks of racism and discrimination, the offensiveness and negative implications of CNN’s action are evident. Native individuals recognized CNN’s omission and viewed the omission as a social cue that their group faces discrimination, thereby prompting action. Future research can broaden conceptualizations of racism and discrimination to include omission and use this updated framework to understand the agency and resiliency of other systematically omitted groups (e.g., Black women and transgender individuals).
Although the relationships among omission, discrimination, and civic engagement are meaningful, the current correlational studies cannot answer whether recognizing omission leads to perceived group discrimination or vice versa. We tested alternative models by reversing relationships, but when one compares cross-sectional mediation models involving the same set of variables, reversing relationships does not reliably distinguish between viable models. Instead, researchers rely on theory to establish variable ordering (Fiedler et al., 2011; Thoemmes, 2015). Both Indigenous scholarship and social-representation theories align with our proposal that omission is a form of bias that shapes broader perceptions of group discrimination; future research is needed to establish causal direction.
Finally, although this work used the largest Native sample in the social science literature, compared with the Native population, our sample consisted of more female, highly educated, urban, and liberal people. Furthermore, because this work used online surveys, we might have missed the opportunity to collect data from many Native peoples who lack Internet access (U.S. Government Accountability Office, 2022). Further research should carefully examine whether the model generalizes to a sample that reflects the Native population.
Conclusion
By collaborating with Native organizations to build an expansive recruiting network, we collected an unprecedented number of responses (> 10,000) from Native individuals across the United States. Generally speaking, research routinely omits Native peoples and justifies such omission by suggesting that they are too small a population to recruit and reliably draw inferences about (Shotton et al., 2013). The present research pushes back by highlighting the feasibility of recruiting “hard-to-reach” populations and building a more complete understanding of the mechanisms (i.e., both omission and perceptions of group discrimination) underlying civic engagement. We show that in the face of omission from society, Native peoples continue to “turn the light of truth upon” the wrongs of the world by acting in support of their communities and people (Wells, 1892). Nonetheless, the onus of these efforts does not lie solely on Native peoples. An equitable future requires not only recognition and acknowledgment of the unjust systems that perpetuate Native omission and discrimination in society but also efforts from all people, especially non-Native people, to dismantle those systems.
Footnotes
Acknowledgements
C. E. Hawk, J. LeBlanc, and N. Santos contributed equally to this work and are listed in alphabetical order in the byline. A. E. Eason and S. A. Fryberg contributed equally as senior authors and are listed in alphabetical order in the byline. We thank IllumiNative, Native Organizers Alliance, and the Center for Native American Youth at The Aspen Institute for participant recruitment and feedback at all stages of the research process. We especially thank Leah Salgado, Cheyenne Brady, Shelly Means, Savannah Romero, and Kendra Becenti for their contributions to the research.
Transparency
Action Editor: Lasana Harris
Editor: Patricia J. Bauer
Author Contributions
