Introduction
The insurrection and its aftermath remain salient to contemporary American Politics. Existing scholarship shows the insurrection was fueled by an effort to return Donald Trump to power while also protesting the decline of the non-Hispanic white population. In examinations of attitudes toward the insurrection, its participants, and President Trump’s role, scholars find that attitudes are divided along partisan and ideological lines. Craig and Albertson (Reference Craig and Albertson2024) find members of Congress remain divided over the severity of the insurrection. These attitudes held by partisan elites filtered down to and were expressed by the electorate (Druckman, Peterson, and Slothuus Reference Druckman, Peterson and Slothuus2013; Malloy and Schwartz Reference Malloy and Schwartz2022). Additionally, perceptions of President Trump’s responsibility for the January 6th insurrection influence the public’s attitudes toward the event. What remains understudied is whether these attitudes vary across racial and ethnic groups.
Although a majority of non-Hispanic white, Latino, and Asian respondents recognize the severity of the insurrection and believe the rioters should be punished, Blacks express this sentiment at higher rates and believe January 6th was intended to be an insurrection by participants (Nadeem Reference Nadeem2021; Updegrove et al. Reference Updegrove, Dmello, Cooper and Cho2023; Wolf Reference Wolf2021). This is not surprising given their racial, social, and economic experiences in the United States. This introduces a broader question about the influence of racial attitudes in contrast to partisanship and ideology on attitudes toward the insurrection, which we address in this paper. Does the impact of partisanship, ideology, and attitudes toward Trump’s responsibility affect the attitudes of respondents from various racial and ethnic groups? Examining attitudes across racial and ethnic groups enables us to demonstrate that the political attitudes of respondents are as salient as their racial and ethnic identities in the context of the insurrection, the rioters, and perceptions of Trump’s role in it.
We use the 2020 Collaborative Multiracial Post-Election Survey (CMPS) to complete our analyses. Our examination of attitudes across racial and ethnic groups reveals that partisanship and ideology consistently mattered in the models examining attitudes toward the event itself, the rioters, and those examining attitudes toward Trump’s responsibility for January 6th. In all our models, we find that negative perceptions of Trump’s insistence that he did not lose the election are associated with more negative perceptions of the insurrection overall, of Trump’s responsibility for the insurrection, and of the rioters themselves.
First, we contribute to the existing literature on attitudes toward the insurrection by using timely, nationally representative public opinion data to evaluate attitudes across four racial and ethnic groups regarding the attack on U.S. democracy, its participants, and Donald Trump. Second, our analyses explore reactions to an event that continues to affect American politics. Third, our analyses demonstrate respondents’ political identities outweighed the influence of their racial and ethnic identities upon attitudes toward the insurrection, the rioters, and Trump’s responsibility for January 6th.
Background of January 6th, 2021
On January 6, 2021, rioters entered the U.S. Capitol Building in Washington, D.C., to delay the certification of the November 2020 presidential election results and attempt to overturn the results on behalf of President Trump. As a result, several people died, hundreds were injured, and four police officers committed suicide (Wise Reference Wise2021; Zengerle Reference Zengerle2022). Convicted rioters were pardoned at the beginning of President Trump’s second term (Dreisbach et al., Reference Dreisbach, Anderson and Van Woerkom2022; Johnson Reference Johnson2025).
Rioters were predominantly non-Hispanic white males and many non-Hispanic white females (Dreisbach et al., Reference Dreisbach, Anderson and Van Woerkom2022). They had a wide age range, came from nearly all states, and represented a variety of occupations (Dreisbach et al., Reference Dreisbach, Anderson and Van Woerkom2022; Tong and McMahon Reference Tong and McMahon2022). Many rioters felt racial threat; they were from counties won by then-candidate Biden in 2020 and were represented by members of Congress who rejected the results of the presidential election (Pape, Larson, and Ruby Reference Pape, Larson and Ruby2024). In addition, they perceived religious threats to their Christian nationalist identity, felt victimized, lost trust in political institutions, strongly identified with their white identity, and embraced conspiracy theories (Armaly, Buckley, and Enders Reference Armaly, Buckley and Enders2022; Armaly and Enders Reference Armaly and Enders2022). Overall, their concerns became part of the mainstream (Tong and McMahon Reference Tong and McMahon2022) and led them to engage in political violence.
The January 6th insurrection remains a prominent example of political violence. Scholars are divided regarding how prevalent political violence is in the United States (Kalmoe and Mason Reference Kalmoe and Mason2022; Westwood et al. Reference Westwood, Grimmer, Tyler and Nall2022). A growing area of research is the attitudes of racial minority groups to political violence. Blacks who have strong in-group attachments and have experienced unfair treatment by police are more likely to support the use of force by protestors responding to violent acts by the police toward their community (Gibson Reference Gibson2025). Also, Blacks are significantly less supportive of political violence compared to non-Hispanic white Republicans.
Scholars also argue conservative radicalization that developed as a result of partisan polarization, existing polarization within the media, white supremacists’ fear of losing power and political control, racial resentment, and the influence of right-wing groups and social media, led rioters to choose to engage in political violence on January 6th (K. B. Anderson Reference Anderson2021; Dirks Reference Dirks2022; Enders and Thornton Reference Enders and Thornton2022; Kydd Reference Kydd2021). Additionally, Trump’s supporters were more likely to express positive attitudes toward the insurrection and support the use of political violence, although racist attitudes mediate this effect (Piazza and Van Doren Reference Piazza and Van Doren2023). One consequence of political violence and the insurrection itself is that Americans anticipate more forms of political violence in the future after elections (Agiesta Reference Agiesta2021).
Given the unique nature of the events of January 6th and the findings in previous literature regarding political violence, we examine three facets of January 6th to determine how party identification, ideology, and perceptions of President Trump are related to support for the insurrection, and whether these factors vary across groups. If the factors that are associated with support for the insurrection are similar across groups, it could indicate a growing cross-racial conservative faction in the United States.
The Impact of Partisanship upon Perceptions of January 6
Since 2021, support for the insurrection and its participants has increased among Republicans (J. Anderson and Coduto Reference Anderson and Coduto2024; Van Noort Reference Van Noort2023; Weiner, Clement, and Guskin Reference Weiner, Clement and Guskin2024). The January 6th hearings revealed that Republican elites initially denounced the insurrection but later began to downplay its severity, claiming the election results were not credible, and censured Republican colleagues who investigated the insurrection (Craig and Albertson Reference Craig and Albertson2024; Gramlich Reference Gramlich2022; Zengerle Reference Zengerle2022). What also became clear is that elected officials from the Republican Party were participants or collaborators in the effort to prevent the election results from being certified (Dreisbach et al., Reference Dreisbach, Anderson and Van Woerkom2022; Wise Reference Wise2021; Zengerle Reference Zengerle2022). As a result, many Republicans in the electorate believe voter fraud occurred during the 2020 presidential election (Agiesta Reference Agiesta2021; Balz, Clement, and Guskin Reference Balz, Clement and Guskin2022; Frasure et al. Reference Frasure, Wong, Baretto and Vargas2021; Wolf Reference Wolf2021).
Although a majority of Americans recognize the insurrection was dangerous, partisans remain divided over the impact of January 6 upon democracy, how violent the rioters were, how severe the punishments should be for the rioters, and the amount of attention that has been given to the insurrection (Agiesta Reference Agiesta2021; Balz, Clement, and Guskin Reference Balz, Clement and Guskin2022; Nadeem Reference Nadeem2021). Partisans are also divided over Trump’s responsibility for the insurrection (Balz, Clement, and Guskin Reference Balz, Clement and Guskin2022; Malloy and Schwartz Reference Malloy and Schwartz2022). Furthermore, the belief in anti-white discrimination among non-Hispanic white republicans led them to continue to support Trump despite his role in the insurrection (Hernandez, Lee, and Roman Reference Hernandez, Lee and Roman2025).Footnote 1 Partisan and ideological divides increased the number of Republicans who believed Trump should have won the election and expressed less trust in the electoral process (Byler Reference Byler2021; Montanaro Reference Montanaro2021).
While the partisanship and ideology of non-Hispanic whites have been widely studied, there is a growing literature about the partisanship of Black, Hispanic, and Asian Americans. A majority of Blacks, Latinos, and Asian Americans are Democrats, although this may be changing (Jones and Saad Reference Jones and Saad2024; Newport Reference Newport2013; Wong and Ramakrishnan Reference Wong and Ramakrishnan2023). Asian American and Latino partisanship remained stable despite anti-immigrant attacks and policies during the first Trump presidency (Hopkins, Kaiser, and Pérez Reference Hopkins, Kaiser and Pérez2023). However, the number of Hispanics who identify as Republicans and who have voted for the Republican presidential candidate has increased (Alamillo Reference Alamillo2019; Basler Reference Basler2008; Jones-Correa, Al-Faham, and Cortez Reference Jones-Correa, Al-Faham and Cortez2018; Pérez Reference Pérez2015). This is evident in the amount of votes President Trump has received; his vote share among African Americans and Hispanics in 2016 was 6% and 28%, respectively, compared to 8% and 38%, respectively, in 2020, while he gained 28% of support from Asian American voters (Igielnik, Keeter, and Hartig Reference Igielnik and Keeter2021). In 2024, the vote share he received from Blacks, Hispanics, and Asians was 13%, 46%, and 40%, respectively.
Many non-Hispanic white respondents and respondents from marginalized groups oppose the actions that occurred on January 6th (Nadeem Reference Nadeem2021; Updegrove et al. Reference Updegrove, Dmello, Cooper and Cho2023). Additionally, it is worth noting that a majority of Black and non-Hispanic white respondents view the rioters as extremists (Bucci, Kirk, and Sampson Reference Bucci, Kirk and Sampson2022). Furthermore, partisanship and ideology influenced attitudes toward this event (J. Anderson and Coduto Reference Anderson and Coduto2024; Van Noort Reference Van Noort2023; Weiner, Clement, and Guskin Reference Weiner, Clement and Guskin2024).
We build upon Bucci et al. (Reference Bucci, Kirk and Sampson2022) by expanding the number of racial and ethnic groups in our analyses and by including additional dependent variables that encompass attitudes toward the event itself, as well as Trump’s culpability. Although we recognize that many factors contributed to January 6th, existing research demonstrates that the subsequent reactions and attitudes regarding January 6th have been primarily driven by partisanship and ideology. Thus, we argue that, compared to their racial and ethnic identities, respondents’ political identities—specifically, partisanship and ideology—demonstrate a greater influence on the attitudes of respondents toward the insurrection, its participants, and President Trump’s culpability. Many people argue that increased polarization is detrimental to democracy, and we examine how these partisan and ideological identities affect attitudes toward the insurrection, the rioters, and Trump’s responsibility for the insurrection.
Perceptions of the President’s Intentions during the Insurrection
During the rally before the insurrection, then-president Trump claimed that the election was stolen from him and encouraged supporters to march to the Capitol (Nawaz, Buhre, and Baldwin Reference Nawaz, Buhre and Baldwin2022).Footnote 2 As the President of the United States, Trump was considered a high-status leader with considerable personal power, who inspired his followers to trust and accept his message (Sauer 2011). With the help of the QAnon movement, Trump portrayed himself as the religious messiah his supporters sought in his quest to overturn the election results (Bond and Neville-Shepard Reference Bond and Neville-Shepard2023). Therefore, having a leader who embraced and heightened his followers’ grievances caused Trump’s followers to make the journey to Washington, D.C., to attempt to prevent Congress from certifying the election results. In addition, social media and right-wing media made his messages easily accessible to his supporters (Clayton et al. Reference Clayton, Davis, Nyhan, Porter, Ryan and Wood2021; Rudden Reference Rudden2021; Van Noort Reference Van Noort2023).
Trump’s rhetoric about the 2020 presidential election eroded trust in elections and increased belief in the rhetoric itself (Clayton et al. Reference Clayton, Davis, Nyhan, Porter, Ryan and Wood2021). We argue believing his rhetoric would lead to more favorable evaluations of the insurrection, the rioters, and Trump’s role in it, regardless of the respondents’ own race and ethnic background.Footnote 3 In contrast, respondents who did not believe Trump’s rhetoric and believed he knew he lost the election would not express favorable evaluations, regardless of their race or ethnicity. Trump has had a lasting impact on American politics, and we must understand the influence of attitudes toward him and how they are related to the insurrection and other associated topics. Given that he is currently in office, his impact on American politics remains salient.
Data and Methods
We utilize the 2020 CMPS to test whether partisanship, ideology, and perceptions of Trump exhibit similar associations with attitudes toward the insurrection across ethnoracial groups. The CMPS is an internet-based survey. More than 15,000 respondents were in the survey, which was offered in English and Spanish, among other languages (Frasure et al. Reference Frasure, Wong, Baretto and Vargas2021; Wolf Reference Wolf2021). The sample includes registered voters, non-registered adults, and non-citizens (Frasure et al. Reference Frasure, Wong, Baretto and Vargas2021; Wolf Reference Wolf2021).
The CMPS extensively asks respondents not only about their political views but also their views on current affairs and salient domestic and foreign policy. Survey responses were collected from April 2, 2021, through August 25, 2021. While there is a large number of respondents included in the CMPS, it is important to note that not all potential respondents are included in our models. As the survey was fielded three months after the events of January 6th, 2021, respondents should remember the event clearly and have accurate responses to the questions dealing with the insurrection.
Table 1 displays the total number of respondents in the CMPS. There are 2,987 white respondents, 4,048 Latino respondents, 4,132 Black respondents, and 3,996 Asian American respondents. The significant oversampling of minority populations in the CMPS is one of the strengths of this data source. A limitation of our study is that our models include fewer respondents than were included in the full survey due to the timing of when our questions of interest were asked to respondents. The principal investigators were able to add questions about January 6th to the survey while it was in the field. For this reason, not every respondent was asked these questions, resulting in a smaller sample size. Additionally, our models only include citizens who voted in the 2020 election, which also reduces the number of observations in our models. There were between 972 and 1175 respondents in each of our models.
Table 1. Total Number of Respondents
Hypotheses
Based on our previous discussion and the findings of the existing literature, we test six hypotheses in our analysis. Generally, existing research has found no racial differences in attitudes among non-Hispanic whites and Blacks toward the insurrection and rioters. Therefore, we expect partisanship and ideology to affect attitudes toward the president, the rioters, and the insurrection. The partisan and ideological division in attitudes leads to the following hypotheses:
H1: The belief that President Trump incited the riot is driven by partisanship and ideology among all racial and ethnic groups.
H2: Attitudes toward the rioters are driven by partisanship and ideology among all racial and ethnic groups.
H3: Attitudes toward the insurrection are driven by partisanship and ideology among all racial and ethnic groups.
We test the following hypotheses regarding the impact of perceived perceptions of Trump’s responsibility during January 6th. Generally, existing research has found that Trump’s rhetoric regarding his loss in 2020 not only contributed to the insurrection but also affected the perceptions of his supporters. Therefore, we expect this perception to affect attitudes toward the president, the rioters, and the insurrection. We test the following hypotheses:
H4: The belief that President Trump incited the riot is driven by perceptions of President Trump’s responsibility for the insurrection among all racial and ethnic groups.
H5: Attitudes toward the rioters are driven by perceptions of President Trump’s responsibility for the insurrection among all racial and ethnic groups.
H6: Attitudes toward the insurrection are driven by perceptions of President Trump’s responsibility for the insurrection among all racial and ethnic groups.
Dependent Variables
Trump Incited Riot
We have three primary dependent variables. First, the “Trump incited riot” dependent variable was based on the following question: “Some people have said that former President Donald Trump encouraged or incited the rioters to attack the U.S. Capitol, while other people have said Trump had no connection and the rioters did this on their own. Which comes closest to your own beliefs?” Respondents received the following options: “Trump encouraged or incited the attack, he shares blame for what happened,” “Rioters were inspired by Trump, but he should not be blamed for encouraging them.” And “Trump had no connection to the rioters, he should not be blamed at all.” Respondents who thought that Trump incited the riot were coded to 1, while other responses were coded to 0.
Rioters are White Supremacists
Second, our “rioters are white supremacists” dependent variable is based on the following question: “Thinking about the people who attacked the U.S. Capitol, do you think…” Respondents received the following options: “Many were White supremacists and racism was an underlying factor in their actions,” “Some may have been White supremacists, but they were mostly upset about Biden winning,” and “They were not White supremacists, racism had nothing to do with their actions.” Respondents who thought that the rioters were white supremacists were recoded to 1 while other responses were recoded to 0.
Coordinated Insurrection
Our final dependent variable is “coordinated insurrection.” Respondents received the following question: “On January 6, 2021, the U.S. Congress was scheduled to meet and vote on the final certification of the state electoral college votes for the presidential election. A group of angry people who supported Donald Trump gathered in Washington, D.C. and attacked the U.S. Capitol to stop Congress from voting to certify the final election results. Based on what you saw or heard about this incident, which comes closest to your view?” Respondents received the following options: “Mostly a protest that went too far” and “A coordinated act of insurrection against the United States.” This response was recoded to (1), and the other option was recoded to (0).
Independent Variables
Partisanship and Ideology
In this paper, we focus on three sets of explanations for attitudes toward January 6th, namely the perception that Trump knew that he lost the 2020 election and the effects of partisanship and ideology. Respondents were asked separate questions about both their party identification and ideology. Respondents were asked, “When it comes to politics, do you think of yourself as liberal, moderate, or conservative?” The variable was coded as a Likert scale from 1-5, where 1 = very liberal and 5 = very conservative. Respondents were asked, “Generally speaking, do you think of yourself as a Republican, Democrat, an independent, or something else?” Responses were coded where 1 = Democrat, 2 = Independent, and 3 = Republican.
Trump Knew He Lost the 2020 Election
Respondents were asked, “After all 50 states and D.C. certified the final vote results Joe Biden defeated Donald Trump by over 7 million votes. There were over 60 lawsuits investigating fraud and courts found no evidence of fraud. Some states that Biden narrowly won had Republican Governors and Secretaries of State who declared the results accurate and valid. Still Donald Trump challenged the results and wanted them over-turned. Given this information, which is closest to your belief?” The variable was recoded as a dichotomous variable where 1 = the belief that Trump lost the 2020 election and knew it, and 0 = the belief that Trump did not know these things. This question invited respondents to look beyond their own perceptions of whether election fraud occurred to focus on whether they thought the president believed he had lost the election. This measure indicates Trump’s mindset regarding the outcome of the election: whether he knew he lost or it was simply a case of not understanding. We include this variable as an important indicator of respondent attitude toward Trump’s culpability in the January 6th insurrection.
Control Variables
Finally, we have several control variables included in each model. We also include traditional demographic variables, such as gender, income, employment status, age, education, and marital status. Additionally, we include variables measuring political trust, political interest, and political efficacy (J. Anderson and Coduto Reference Anderson and Coduto2024; Armaly, Buckley, and Enders Reference Armaly, Buckley and Enders2022; Armaly and Enders Reference Armaly and Enders2022). We include variables on voting, social media usage, and frequency of church attendance. The media’s portrayal of social protests and the public’s political ideology influence perceptions of protesters, the police involved, and levels of support for the protest (Baranauskas Reference Baranauskas2022; McLeod and Detenber Reference McLeod and Detenber1999; Metcalfe and Pickett Reference Metcalfe and Pickett2022). It is expected that there would be more support among the Republican electorate for the insurrection, the rioters, and Trump, as the rioters and their co-partisans and ideologues exchanged messages via social media that aligned with their political views, including favorable portrayals of the insurrection (Clayton et al. Reference Clayton, Davis, Nyhan, Porter, Ryan and Wood2021; Rudden Reference Rudden2021). Church attendance is a predictor of feeling one’s religious liberty is threatened (Herrington Reference Herrington2021). It is possible that people who attend church frequently, specifically Protestant evangelicals, may be exposed to frequent messages about perceived threats to their Christian faith, which would contribute to more favorable views of each of the outcome variables we examine (Armaly, Buckley, and Enders Reference Armaly, Buckley and Enders2022). Finally, we include variables related to place and location, such as living in the South, being born in the United States, and living in an urban area. Many of the rioters originally came from large metropolitan areas that experienced significant population shifts, contributing to their participation in the insurrection (Anderson and Coduto, Reference Anderson and Coduto2024; Tong and McMahon, Reference Tong and McMahon2022). Therefore, it is possible respondents living in those areas would express favorable attitudes toward each of our outcome variables.Footnote 4
We employ probit models in this analysis. We examine these attitudes across four racial groups: non-Hispanic whites, Blacks, Latinos, and Asian American respondents. We created a dummy variable for each group where respondents belonging to that group were coded to 1, while all other respondents were coded to 0. We then examine the effect of independent variables and control variables within each group. This allows us to explore common themes across groups as well as diversity within each group. We present the results of alternative models and goodness of fit tests in the appendix.
FindingsFootnote 5,Footnote 6
Perceptions of Insurrection
The Impact of Partisanship and Ideology upon Attitudes Toward the Insurrection
In Table 2, we examine the impact of partisanship, ideology, and perceptions of Trump’s intentions on attitudes toward January 6, 2021. In each model for each racial and ethnic group, Democrat respondents and more liberal respondents were more likely to view the events of January 6th as an insurrection rather than a protest. It is clear that ideology and partisanship are the primary drivers of attitudes toward January 6, 2021. As a result, we find strong support for Hypothesis 3. Examining each ethnic group, we find that party and ideology are all meaningfully significant in explaining attitudes toward the insurrection. Our findings complement those of Nadeem (Reference Nadeem2021) and Balz, Clement, and Gushkin (Reference Balz, Clement and Guskin2022) regarding the differences in views of the insurrection among partisans and ideologues.
Table 2. Perceptions of Insurrection
The Impact of Perceptions of Trump’s Culpability on Attitudes Toward the Insurrection
Our results confirm Hypothesis 6. Examining each ethnic group, we find that attitudes toward the insurrection are driven by perceptions of President Trump’s intention for the insurrection. These results complement the findings of Agiesta (Reference Agiesta2021), Frasure et al. (Reference Frasure, Wong, Baretto and Vargas2021), and Wolf (Reference Wolf2021). We also find that respondents who believe Trump knew he lost the 2020 election were significantly more likely to view the events of January 6th as an insurrection rather than a protest.
Control Variables
While other control variables reach statistical significance in individual models, none of them are as consistent across the models. This is where differences across the racial and ethnic groups are apparent. We find that interest in politics is related to perceptions of the events of January 6th in these models. Non-Hispanic white, Black, and Asian American respondents who have more political interest were less likely to view January 6th as an insurrection and more likely to view it as a protest.Footnote 7 Among Latino respondents, Latinas, older Latinos, more highly educated Latinos, and Latinos with higher levels of religiosity are more likely to think of January 6th as an insurrection rather than a protest.Footnote 8 This indicates that there is within-group variation in attitudes toward January 6th.
Having voted in the 2020 election was only significant for whites and Blacks in this set of models. In these groups, respondents who voted were more likely to think of January 6th as an insurrection rather than a protest. The effect of education also differs between groups. For white and Asian respondents, higher education is associated with increased perceptions of January 6th as a protest. For Black respondents, lower levels of political trust are associated with the perception that January 6th was an insurrection. This complements the findings by Updegrove et al (Reference Updegrove, Dmello, Cooper and Cho2023) about Blacks perceiving the insurrection as another threat from white supremacy. We would expect to see consistent effects of these two factors; however, in this set of models, partisanship, ideology, and perceptions of Trump have a more pronounced impact on attitudes toward January 6th.
Asian American respondents with lower levels of political efficacy are more likely to view January 6th as a protest rather than an insurrection. This result can be explained by Wong and Ramakrishnan (Reference Wong and Ramakrishnan2023), who note that although turnout among Asian Americans has increased over time, many choose to engage in other forms of political participation, including protests and working on political campaigns (Wong et al. Reference Wong, Karthick Ramakrishnan, Lee, Junn and Wong2011). Therefore, these respondents may view the insurrection primarily as a protest, rather than something more dangerous. Wong and Ramakrishnan also note that political parties do not always mobilize Asian Americans, nor do they tend to live in swing states, which could contribute to lower levels of efficacy.
White respondents who use social media were more likely to view January 6th as an insurrection rather than as a protest. This is interesting, as previous scholarship has shown that non-Hispanic white rioters who participated in the insurrection used social media to spread their grievances and share information about the insurrection (Clayton et al., Reference Clayton, Davis, Nyhan, Porter, Ryan and Wood2021; Rudden, Reference Rudden2021; Van Noort, Reference Van Noort2023).
In sum, we conclude that across each ethnic group, the most important factors in determining attitudes toward the insurrection are partisanship, ideology, and perceptions of Trump’s electoral success in the 2020 election. While many other factors traditionally influence attitudes in political science research, none of them are consistently significant in our models.
Perceptions of Trump’s Incitement of the Riot
The Impact of Partisanship and Ideology upon Respondents’ Belief Trump Incited the Riot
In Table 3, we analyze the impact of partisanship, ideology, and perceptions of Trump’s intention upon attitudes toward Trump’s responsibility for the insurrection. In all models and each ethnic group, we find support for our first hypothesis; Democrat respondents and more liberal respondents were more likely to think that Trump incited the riot on January 6th. Attitudes toward President Trump’s intentions are driven by partisanship and ideology in each ethnic group. Our findings align with the University of Maryland Poll in 2022, which showed that a majority of Republicans and Trump supporters believed the president bore little to no responsibility for the insurrection (Balz, Clement, and Guskin Reference Balz, Clement and Guskin2022; Gramlich Reference Gramlich2022).
Table 3. Perceptions of Trump’s Incitement of the Riot
The Impact of Perceptions of Trump’s Culpability upon Respondents’ Belief that Trump Incited the Riot
Additionally, respondents who believed that Trump knew he had lost the 2020 election were more likely to think that the primary catalyst for the events of January 6th was Trump himself. As a result, we find strong support for hypothesis 4. Attitudes toward Trump’s intention during the insurrection are driven by perceptions of President Trump’s responsibility for the insurrection in each ethnic group. This consistent finding demonstrates respondents who supported Trump’s belief that voter fraud occurred did not view him as responsible for the riot (Balz et al, Reference Balz, Clement and Guskin2022).
Control Variables
Additionally, as in the first set of models, various control variables reach statistical significance, demonstrating racial and ethnic group differences. Wealthier Asian Americans, employed Asian Americans, and those with higher levels of education were more likely to think that Trump incited the riot. Asian American respondents who used social media and were married were less likely to think that Trump incited the riot. Again, these results demonstrate the diversity within group attitudes among Asian Americans.
Among Latinos, Latinos who voted, those with higher levels of political efficacy, and those with lower levels of political trust are more likely to think that Trump incited the riot. As Updegrove et al (Reference Updegrove, Dmello, Cooper and Cho2023) found for Blacks, Latino respondents who experienced discrimination may view the insurrection as another attack by white supremacy. This complements Alamillo (Reference Alamillo2019), who found Latinos who experienced discrimination were less likely to vote for President Trump compared to co-ethnics who denied racism existed.
For Black respondents, Black women, older Black respondents, and those with lower levels of political trust are more likely to think that Trump incited the riot. Older white respondents and those with higher levels of religiosity were more likely to believe that Trump incited the riot, while white respondents who voted and who are married were less likely to think that Trump caused the riot on January 6th. Overall, our results continue to demonstrate that the impact of partisanship, ideology, and perceptions of Trump’s intentions remain the same across each ethnic group in our examination.
Perceptions of the Rioters
The Impact of Partisanship and Ideology upon Attitudes Toward the Rioters
In Table 4, we examine the impact of partisanship, ideology, and perceptions of Trump’s intentions on the evaluation of rioters. This is the only set of models where we find exceptions to the consistency of the effects of partisanship and ideology. In three out of the four models, we find that Democrat respondents were more likely to think of the rioters as white supremacists. The model for white respondents did not conform to these findings; it did not reach statistical significance and was in the opposite direction. In three out of four models, we see that more liberal respondents were more likely to think of the rioters on January 6th as white supremacists instead of as other types of protestors. However, it did not reach statistical significance in our model for Black respondents. This result can be best explained by the findings of White and Laird (Reference White and Laird2020), who found that Blacks remain a solid part of the Democratic Party membership due to social norms and expectations. Thus, our result may demonstrate the impact of some Black conservatives among the respondents; however, social pressure and costs may be too great for them to diverge from the expectations of their co-racial group members fully. The authors acknowledge that although Blacks may hold conservative positions on various issues, remaining part of the Democratic party may be best for the group as a whole (White and Laird Reference White and Laird2020). Despite the exceptions, we find support for hypothesis 2 in our preferred specification. Among Black, Latino, and Asian respondents, partisanship is meaningfully significant in explaining attitudes toward the rioters. Among white, Latino, and Asian American respondents, ideology is meaningfully significant in explaining attitudes toward the rioters. Our findings align with other results from public opinion polls, which show a clear partisan and ideological divide over attitudes toward the rioters (Agiesta Reference Agiesta2021; Balz et al. Reference Balz, Clement and Guskin2022; Malloy and Schwartz Reference Malloy and Schwartz2022).
Table 4. Perceptions of the Rioters
The Impact of Perceptions of Trump’s Culpability on Attitudes toward the Rioters
In all the models, we see that respondents who believed that Trump knew that he lost were more likely to think of the rioters on January 6th as white supremacists instead of as other types of protestors. Here, we find broad support for our fifth hypothesis; examining each ethnic group, we find that perceptions of President Trump’s intentions regarding the insurrection are meaningfully significant in explaining attitudes toward the rioters.
Control Variables
Another interesting factor in this set of models concerns the effect of voting. In all models in this section, respondents who voted were more likely to think of the rioters as white supremacists. This is the only set of models where we consistently observe this result, which may have impacted the importance of partisanship and ideology in the two models where we see disparate results. It is possible that some respondents who voted have high political interest and high levels of education, which contributed to their awareness of the rioters’ intentions and their evaluation of them accordingly. Another possibility is that some voters were Democratic Party activists and inclined to evaluate the rioters according to their party’s narrative regarding the rioters. In all models in this section, respondents who voted were more likely to think of the rioters as white supremacists.
Additionally, as shown in the previous tables, several control variables reach statistical significance, indicating racial and ethnic group differences. Among white, Latino, and Asian American respondents, those who were born in the United States were less likely to think of the rioters as white supremacists. These Latino respondents may be second- or third-generation members who have acculturated to broader American society and political attitudes (Pedraza Reference Pedraza2014; Pérez Reference Pérez2024). It is also possible these respondents viewed media coverage that portrayed the rioters more positively, which led to more favorable evaluations of them (Edwards and Arnon Reference Edwards and Arnon2021; McLeod and Detenber Reference McLeod and Detenber1999).
Asian American respondents with higher levels of political interest are more likely to think of the rioters as white supremacists, while those with lower levels of political efficacy are less likely to do so. Black respondents with higher levels of political interest and lower levels of efficacy are less likely to think of the rioters as white supremacists. This finding complements Bucci, Kirk, and Sampson (Reference Bucci, Kirk and Sampson2022), who find that Black respondents were slightly more likely than white respondents to view the rioters as patriots. White respondents with lower levels of political interest are less likely to think of the rioters as white supremacists. They may be more susceptible to the messaging from Republican party leaders who offer positive narratives about the rioters (Craig and Albertson Reference Craig and Albertson2024; Gramlich Reference Gramlich2022). Overall, our results continue to demonstrate that the impact of partisanship, ideology, and perceptions of Trump’s intentions remain the same in each ethnic group.
Conclusions
Our purpose for our analyses was to examine potential differences among attitudes across multiple racial and ethnic groups toward Trump’s responsibility for the insurrection, the rioters, and the insurrection itself. We build upon existing research by evaluating the impact of partisanship, ideology, and attitudes toward President Trump’s intention for the insurrection upon attitudes of non-Hispanic white, Black, Latino, and Asian American respondents. In existing assessments of attitudes toward the insurrection and related topics, the results not only revealed a partisan and ideological divide but also highlighted divisions in perceptions of the president’s intentions regarding the insurrection. What was missing was a comprehensive examination that showed these components worked consistently and overrode potential racial and ethnic group differences. This is our main contribution.
A consistent finding is the importance of partisanship and ideology in attitudes toward the events of January 6th overall, of Trump’s responsibility for January 6th, and of the rioters themselves. Partisanship and ideology mattered most consistently in the models examining attitudes toward the event itself and those models examining attitudes toward Trump’s responsibility for January 6th. In the models examining attitudes toward the rioters, there were two instances where ideology and partisanship were not significant. It may be that the effects of partisanship and ideology are more salient in evaluating more institutional factors, but other factors for certain groups become more salient when evaluating their fellow citizens.
In all our models, we find that negative perceptions of Trump’s insistence that he did not lose the election, despite his claims, are associated with more negative perceptions of the insurrection overall, of Trump’s responsibility for the insurrection, and of the rioters themselves. This confirms all hypotheses in this category and further underscores the importance of Trump’s role in the events of January 6th, 2021.
Another important conclusion is the glaring lack of consistent results in many of the assumed predictors of political behavior. Where we would expect an influence of geographic region, income, education, gender, age, and several other factors, we do not find such an influence in this research. There are individual instances where these factors are significant. Still, they do not seem to have the same consistent impact as the effects of partisan identification, ideology, and perceptions that Trump believed he had lost the 2020 presidential election.
Our results lead to several implications. First, because there is still widespread division in our country about the results of the 2020 presidential election, the ongoing effects of another Trump term in office, and the ability to unite a country that seems (at this time) more divided than ever along partisan and ideological lines. Trump’s insistence that he won the 2020 presidential election was given widespread legitimacy among his supporters, and many still believe it (Frasure et al, Reference Frasure, Wong, Baretto and Vargas2021; Wolf, Reference Wolf2021).
Not only did this message have implications for the 2020 presidential election and its aftermath, but it also had implications for future elections. If a significant segment of the population does not believe elections are legitimate, then continued unrest, division, and conflict are likely to occur as a result. When people perceive politics as a zero-sum game and feel threatened, they are more likely to take violent actions to “defend” themselves against whatever they perceive as a threat. As discussed earlier, public opinion polls have found that people expect more political violence to occur in the future (Agiesta Reference Agiesta2021). Since scholars have found that partisans and ideologues support the use of political violence (K. B. Anderson Reference Anderson2021; Kydd Reference Kydd2021), and given the contentious nature of the general and presidential elections in 2024, the likelihood of more political violence occurring is higher than zero. Another consequence of believing the 2020 election results were illegitimate is that these members of the population continue to support state and federal legislation to curtail voting rights for various groups believed to have engaged in these actions. These developments have raised serious concerns about the health and longevity of democracy in the United States.
Donald Trump secured victory in both the Electoral College and the popular vote in 2024. Instead of diminishing polarization and division within the population, however, the country is more divided than ever. Along with ideological polarization, affective polarization and negative partisanship continue to grow. Many people dislike members of the opposing party and no longer claim to simply disagree about politics-instead, they claim to have differences on values, societal norms, and claims about who deserves fundamental rights in American society (Abramowitz and Webster Reference Abramowitz and Webster2018; Iyengar et al. Reference Iyengar, Lelkes, Levendusky, Malhotra and Westwood2019). In the 2024 election, nearly all groups in society shifted significantly to the right and supported Trump, including large numbers of the racial and ethnic groups discussed in this paper (Igielnik, Keeter, and Hartig Reference Igielnik and Keeter2021; Pérez Reference Pérez2024). It appears, then, that our findings remain applicable in 2024; ideology and partisanship continue to matter more in determining support for Donald Trump, regardless of group membership.
Second, our results also add to the ongoing discussion of the impact of parties, ideology, and the lasting impact of the president on American politics. The ongoing partisan polarization continues to shape the direction of politics, policy, and how the electorate perceives the events unfolding around them. We are not claiming that race and ethnic differences do not exist, since these differences became clear among our control variables; they shape some of the attitudes held by non-Hispanic whites, Blacks, Latinos, and Asian Americans. However, we demonstrate that the attitudes toward the insurrection and related topics among respondents from these racial and ethnic groups align with the attitudes of their party leadership.
Existing scholarship demonstrates that as a political figure, the president remains a polarizing figure, eliciting strong support as well as strong opposition within the electorate. Our analyses demonstrate that these factors override the impact of race and ethnicity upon evaluations of the insurrection, the rioters, and Trump’s responsibility for the insurrection. The reasons that led people to participate in the insurrection remained salient and unresolved going into and after the 2024 presidential elections. In addition, since returning to the office, the president has maintained steadfast support from the co-party leaders and supporters within the mass public. Therefore, the questions raised in our manuscript and other questions about the insurrection and related topics will remain salient as this political saga continues to unfold.
Acknowledgments
The authors would like to thank Dr. Andrea Silva for her valuable insights, which enhanced the quality of this manuscript and Dr. Parwinder Grewal for his support for this project.
Funding statement
This research did not receive any specific financial support.
Competing interests
The authors declare no competing interests.
Appendix 1. Robustness checks for Models
Table A1. Models with White and Non-White Samples
Table A2. Models with Full Sample and Interaction Terms
Table A3. Interaction models. Each interaction in the following models run as individual models to estimate effects
Table A4. Total number of conservatives in sample by group
Table A5. Total number of Republicans in sample by group
Table A6. Mean and Standard Deviation Per Group of Dependent and Independent Variables
Table A7. Robustness Models including Assimilation, Threat, and Linked Fate
Table A8. Goodness of fit statistics
