COVID-19, Diffuse Anxiety, and Public (Mis)Trust in Government: Empirical Insights and Implications for Crime and Justice

Abstract

Given that trust in government is a critical feature of a well-functioning democracy, research into its determinants has long been a priority among public opinion scholars. The consensus in the literature is that short-term factors drive the ebbs and flows of public trust, and a climate of mistrust has significant consequences on the government's ability to deliver on policies and enforce the law. Despite decades of extensive research on public trust, changing circumstances related to the COVID-19 pandemic support the need to investigate the factors shaping trust in this distinct period. This article, using data from the American National Election Study, explores how economic, social, and political anxieties pervasive throughout the pandemic influence trust in the United States government. Findings from ordered logistic regression analyses indicate that public trust in government is associated with views of the government's COVID-19 response, beliefs about the state of the country and government corruption, economic anxieties, and concerns about election fraud and the status of American democracy. Findings also reveal that sentiments toward institutions—including the police and the Center for Disease Control—contribute to variability in public trust. The implications of these findings for criminal justice research and policy are also considered.

Trust plays a very tangible role in the effectiveness of government. Few perceptions are more palpable than that of trust or its absence. Governments ignore this at their peril. ( OECD (2017) )

In times of national crisis, citizens turn to their government for leadership, guidance, and assistance. How governments respond, or how citizens perceive this response, can shape public attitudes toward political leaders and impact public trust in government (Chanley, 2002; Citrin & Stoker, 2018). For example, people dissatisfied with government performance may express their dissatisfaction in terms of cynicism toward the government (Citrin, 1974). An increase in cynicism, however, is not without consequence, as public mistrust can undermine government systems and diminish voluntary compliance with the law (Marien & Hooghe, 2011; Miller, 1974a, 1974b). Outcomes of this nature provide ample reason for scholars to investigate and comprehensively understand the factors influencing public trust in the United States.

While prior research has linked various antecedents to Americans’ trust in government (e.g., Avery, 2006; Chanley et al., 2000; Citrin, 1974; Citrin & Stoker, 2018; Filindra et al., 2022; Hetherington & Rudolph, 2008; Orren, 1997), far less is known about the correlates of trust in the era of the COVID-19 pandemic. This once-in-a-century health crisis swept across the nation in the spring of 2020, transforming the economic, social, and political landscape in short order. Compounding the instability caused by the pandemic was a series of overlapping crises ranging from police abuse and racial injustice to pandemic-related unrest and an upsurge in lethal violence. The convergence of these issues promoted the diffusion of panic and concern throughout the public sphere (Bethune, 2021; Lati, 2021; Nicomedes & Avila, 2020), polarizing public sentiment (Hegland et al., 2022) and potentially diminishing citizens’ trust in government.

This article aims to inform the scientific understanding of public trust in the era of the COVID-19 pandemic. We use data from the American National Election Study to explore the factors predicting trust in government circa the November 2020 presidential election. Focusing on this period of the pandemic enables investigation into a broad range of potential predictors relating to salient social conditions and public concerns, such as issues of political unrest, election integrity, racial injustice in law enforcement, economic insecurity, and leaders’ handling of the surging pandemic. Moreover, because the nation was at a heightened risk of political violence and instability going into the November elections (see Kishi & Jones, 2020), concentrating our analysis on this period allows for inquiry into public attitudes at perhaps the most crucial political moment of the pandemic in the United States. The following section provides an overview of the literature on political trust and then outlines relevant public concerns and social conditions that may influence attitudes about trust in government. Next, we detail our methodology and present findings deriving from our analyses. We conclude with a discussion of the implications of this work for criminological research and criminal justice policy.

Trust in Government

The concept of trust in government is one of a family of terms referring to citizens’ feelings about their government. Although the concept appears straightforward, it is not easy to conceptualize and measure. For Easton (1965), trust signifies support for governmental policies, but it could also represent confidence in the political system itself. Likewise, trust in government could be considered an aspect of legitimacy (Tyler & Jackson, 2013) or consequential in the conferment of legitimacy (Hough et al., 2010), which grants authorities the moral foundations for obedience (Citrin & Stoker, 2018). The most common measure of trust is the American National Election Studies (ANES) item, “How much of the time do you think you can trust the government in Washington to do what is right?” Despite the various shortfalls of this item (see Citrin & Stoker, 2018), scholarly investigations, U.S. media polls, and Gallup and Pew surveys have used it for decades. ¹

A significant body of research has examined the determinants of trust in government. Studies have explored, among other factors, public evaluations of incumbent officials, government institutions, satisfaction with public policy, and the actions of government officials as predictors of public trust (e.g., Chanley et al., 2000, 2001; Citrin, 1974; Citrin & Green, 1986; Citrin & Luks, 2001; Miller, 1974a, 1974b). Extant evidence indicates that citizens who are dissatisfied with the economy's direction (Hetherington & Rudolph, 2008; Weatherford, 1984) or those facing economic insecurity (Citrin & Luks, 2001; Chanley et al., 2000, 2001) are more likely to mistrust the government. Public concerns about crime (Chanley et al., 2000), political scandal (Chanley et al., 2000), and media focus on political corruption and scandal (Orren, 1997) also reduce trust in government. Similarly, increased public focus on domestic versus international concerns is likely to diminish trust (Alford, 2001; Chanley, 2002; Chanley et al., 2000, 2001).

Diffuse Anxieties and Public Trust

In 2020, the United States was fraught with many concurrent crises. Most notable among them was the surge of the COVID-19 pandemic that brought significant mortality (Shiels et al., 2022), social and economic disruption (Sharif et al., 2020), and widespread fear and panic to the public sphere (Lati, 2021; Nicomedes & Avila, 2020). Other emergent issues concerning crime and electoral politics further deepened the collective anxiety of the period. For example, gun violence sharply increased within many large cities (Rosenfeld et al., 2021). The national murder rate also rose by 30% between 2019 and 2020, the most significant single-year increase in over a century (Gramlich, 2021; Rosenfeld et al., 2021). Roughly 80% of Americans believed crime was rising, the highest level observed since the early 1990s (McCarthy, 2020).

Hate crimes surged, with attacks targeting a range of minority groups. Persons of Asian descent saw a 73% increase in hate crime victimization—fueled partly by racist stereotypes tied to the pandemic (Han et al., 2023; Sganga, 2021). Though racially motivated hate crimes saw the most severe spike, bias attacks targeting LGBTQ+ persons also significantly increased (Hernandez, 2021; Kishi & Jones, 2020). Misinformation and divisive political rhetoric exacerbated polarization (Chen et al., 2021) to such an extent that, according to the US Crisis Monitor, the nation was at an elevated risk of political violence and instability approaching the November 2020 presidential election (Kishi & Jones, 2020). This confluence of issues was further complicated by the police-involved murder of George Floyd Jr. in Minneapolis, Minnesota. Floyd's death, alongside a series of other deadly police–citizen encounters, sparked an historic protest and reform movement spearheaded by Black Lives Matter (BLM) that mobilized 15–26 million people protesting racial injustice in law enforcement (Buchanan et al., 2020; Putnam et al., 2020).

As these and other incidents diffused panic and concern throughout the public, it is essential to recognize that such events can also impact broader societal perceptions of government. For example, in the wake of the Floyd protests, public perception of law enforcement's response to these demonstrations varied. The public was far less approving of then-President Trump mobilizing federal law enforcement, yet they were more accepting of localized police efforts to control or limit demonstrations (Kolcak & McCabe, 2022). Although public outcry regarding Floyd's death was overwhelming, Pryce and Gainey (2022) found that Black Americans were significantly less likely than all other racial groups to approve of how police responded to the rallies (see also Reid & Craig, 2021), demonstrating the impact of race and ethnicity on public trust.

Similarly, a small but growing body of literature has examined the impact of COVID-19 decision-making on public trust. For example, Fazio et al. (2021) found that COVID-19 lockdown policies were adjusted to prevent diminished trust and negative public perception that the government's actions were too “extreme.” In their study of French and Italian participants, Lalot et al. (2022) found that either trust in the government or concern about the pandemic was necessary to sustain compliance with COVID-19–related policies. However, the absence of both factors significantly reduces overall compliance. Schraff (2021) warns that the standard determinants of political trust, such as economic and social factors, lose explanatory power during the pandemic. Actions by individual decision-makers also likely drive how the public views their handling of the crises. For example, some sheriffs refused to enforce state mandates related to COVID-19, which was met by ideologically driven extreme responses (Kincaid & Leckrone, 2021; Turret et al., 2020). This effect was exacerbated by subsequent revelations that some leading law enforcement officials were also registered members of the Oath Keepers (Dickinson, 2021), further undermining public trust in government.

Beyond the COVID-19 pandemic, research has also found that the government's response to catastrophic events impacts public perception and trust. For example, in examining the 2010 Pakistan floods, which affected approximately 20% of the country's geographic area, Akbar and Aldrich (2017) found that perceived fairness in how disaster aid was distributed was positively correlated with levels of trust in the government. Moreover, research examining other catastrophic events centers on the influence of social capital on political trust (e.g., Aldrich & Meyer, 2015; Eadie & Su, 2018; Jovita et al., 2019; Veszteg et al., 2015), suggesting the importance of sociodemographic factors as mechanisms of support during national crises.

The Current Study

Prior research on the predictors of public trust has been informative. However, the outbreak of the COVID-19 pandemic in the United States and ensuing societal changes support the need to explore and holistically understand the correlates of public trust in this period. The current study uses a large, nationally representative survey of American adults to explore the factors influencing trust in government. Given the backdrop of co-occurring crises that emerged during the pandemic, we examine a comprehensive set of potential predictors relating to salient social, economic, and political topics and concerns. This study's findings are expected to have implications for ongoing criminal justice reform efforts and public support for the government's policy agenda in general.

Data and Methods

This study uses data from the American National Election Study (ANES), a mixed-mode survey using online, video, and telephone interviews and self-administered surveys (American National Election Studies, 2021). Our data comes from the 2020 survey, but the study has been conducted each election cycle since 1984. The 2020 survey was conducted between August 18 and November 3 (election day), with a follow-up survey conducted between November 8, 2020, and January 4, 2021. ²

For this iteration of the ANES, researchers first surveyed respondents who had taken the 2016 version of the survey (n = 2,839). Then, through multiple modes of recruitment, they also included randomly sampled first-time respondents (n = 5,441). After dropping 830 respondents who only took the preelection version of the survey, we had a potential sample of 7,449 respondents. Complete case analysis left us with a total of 6,563 respondents, with just under 12% of our sample lost due to missing data. ³

Dependent Variable

Our dependent variable is trust in government, focusing primarily on the federal level, based on a single item in which respondents were asked, “How often can you trust the federal government in Washington to do what is right?” Responses ranged on a five-item scale from “Always” to “Never,” with higher scores indicating more trust in government. Decades-long research on political trust has consistently used this item to measure public trust in the government (see, generally, Chanley, 2002; Citrin & Stoker, 2018). Descriptive information about this and all other variables in this study can be found in Table 1.

Table 1.

Weighted Full Sample Descriptive Statistics.

	Full sample		White sample		Nonwhite sample
	Mean	SD	Mean	SD	Mean	SD	Min.	Max.
Dependent variable
Trust in government	1.53	.01	1.52	.01	1.54	.02	0	4
Government pandemic response
Federal pandemic response	0.68	.01	0.78	.01	0.41	.02	0	3
Local pandemic response	1.35	.01	1.38	.01	1.29	.02	0	3
Political anxieties
Government corruption	2.04	.01	2.01	.01	2.10	.02	0	4
Concern about democracy	1.37	.01	1.35	.01	1.40	.02	0	3
Concern about elections	3.48	.03	3.36	.03	3.84	.05	0	8
Concern about political violence	1.77	.01	1.79	.01	1.73	.01	0	2
Concern about the country	0.75	.01	0.72	.01	0.84	.01	0	1
Negative emotions about the country	2.75	.01	2.71	.01	2.87	.02	0	4
Economic anxieties
National economic insecurity	2.12	.01	2.04	.02	2.35	.03	0	4
Personal economic insecurity	1.24	.01	1.14	.02	1.50	.03	0	4
Institutional sentiments
Warm feelings toward police	70.66	.31	74.15	.33	60.60	.65	0	100
Warm feelings toward BLM	53.68	.44	48.99	.51	67.23	.79	0	100
Warm feelings toward the CDC	70.45	.29	70.02	.34	71.72	.58	0	100
Controls
White	0.74	.01	–	–	–	–	0	1
Male	0.46	.01	0.47	.01	0.45	.01	0	1
Age	51.68	.23	53.54	.26	46.34	.41	18	80
Married	0.53	.01	0.57	.01	0.43	.01	0	1
Children	0.31	.01	0.28	.01	0.40	.01	0	1
Liberal	0.32	.01	0.32	.01	0.31	.01	0	1
Moderate	0.22	.01	0.20	.01	0.28	.01	0	1
Conservative	0.33	.01	0.38	.01	0.19	.01	0	1
Not sure	0.13	.00	0.10	.01	0.21	.01	0	1
N	6,563		4,874		1,689

Independent Variables

We include several independent variables in our models that together provide insight into respondents’ views on issues related to the government, politics, and various public institutions during the COVID-19 pandemic.

We include two measures focused on respondents’ views of how political leaders within the government handled the COVID-19 pandemic. In four separate questions, respondents were asked whether they approved or disapproved of the way (a) the president, (b) the federal government, (c) their state governor, and (d) their local government handled the pandemic. Their views about the president's role and the federal government's role were combined into an additive index called federal pandemic response (alpha = .8271), and their views about their governor's role and their local government's role were combined into an additive index called local pandemic response (alpha = .6209). Each index ranged from 0 to 2, with higher scores indicating more approval or higher support for that entity's response to the pandemic. These constructs seek to understand the approval of crisis management by the various levels of government.

Also included were several variables on views about the government and politics. Perceptions of government corruption were obtained by asking respondents, “How many of the people running the government are corrupt?” This variable was coded such that higher scores indicate that the respondent feels there is more government corruption. Past research has found that beliefs of corruption by leaders directly impacts an individual's trust in the government as an organization (Augustine & Enyi, 2020; Quah, 2013; Solé-Ollé & Sorribas-Navarro, 2014; Zhang & Kim, 2018). While one could argue that corruption is implicitly included within the composition of trust, the extant literature provides ample evidence for the need to evaluate these characteristics as two separate constructs. Respondents were also asked how satisfied they are with the way democracy works in the United States. This variable was coded such that a higher score indicates more concern about democracy. We also include a measure of concern about elections, which was obtained from two questions: respondents were asked how accurately they think the 2020 general election votes would be counted and how much they trusted officials who oversee their local elections. A higher score indicated more suspicion or concern about the election process. We included a measure to assess respondents’ concerns about political unrest. Respondents were asked whether they perceive political violence to have increased, decreased, or stayed the same compared to 4 years ago. This variable was coded such that higher scores indicate more concern about political violence.

We also examine how one's feelings about the state of the country relate to trust in the government. We include one measure of concerns about the country, in which respondents were asked, “Do you feel things in this country are generally going in the right direction, or do you feel things have pretty seriously gotten off on the wrong track?” Higher scores indicate more concern about the state of the country. Another measure capturing negative emotions about the country was included. In a series of questions, respondents were asked how afraid, outraged, angry, worried, irritated, or nervous they felt about how things were going in the country. They were also asked about positive emotions—including hope, happiness, and pride—which were all reverse coded. Their responses were compiled into an average index (alpha = .8977), with higher scores indicating more negative feelings toward the state of the country. ⁴

Because the COVID-19 pandemic may have contributed to economic anxieties, we include two variables to examine how economic insecurity may impact trust in the government. First, we include a measure of national economic insecurity. Respondents indicated how worried they were about the national economy, with higher scores on this measure indicating more concern. Second, we include a measure of personal economic insecurity. Respondents indicated how worried they were about their individual and family's financial well-being, with higher scores indicating greater insecurity or concern.

Finally, we recognize that respondents’ views about particular institutions may influence their views of or trust in the government. In this survey, respondents were asked to fill out feeling thermometers for several institutions, rating their feelings about that institution on a scale from 0 (cold feelings) to 100 (warm feelings). We included three specific measures for their warm feelings toward police, warm feelings toward BLM, and warm feelings toward the Center for Disease Control (CDC).

We also control for several key demographic indicators that are standard in the public opinion scholarship on trust in government. We control for the respondent's race by dichotomizing the sample as either white (coded as 1) or nonwhite (coded as 0). Gender is included, with male respondents coded as 1. Age was created by subtracting the respondent's birth year from 2020. Marital status indicates that the individual was either married or in a domestic partnership at the time of the survey, and we account for whether they had children under 18 in their household. Finally, respondents’ political ideology was included and disaggregated into four categories: liberal, moderate, conservative, and not sure. In all analyses, liberal is the reference category.

Plan of Analysis

In the subsequent section, we present ordered logistic regression models predicting trust in the government. Our analyses include a model for the full sample of respondents and subsamples of white and nonwhite respondents. Analytical models use ordered logistic regression because of the discrete, ordered nature of the dependent variable, with scores ranging from 0 (low trust in government) to 4 (high trust in government). In examining the model fit, the parallel regression assumption was violated, suggesting that multinomial logistic regression models may be appropriate. However, this finding is not unexpected due to the large sample size. Parallel regression assumption tests tend to be sensitive and reject the null hypothesis even when the assumption is met (Harrell, 2001). We, therefore, compared the results of our ordered logistic regression models to multinomial logistic regression models, and the results did not substantively vary. We present the more parsimonious ordered logistic regression models in the findings section. Tests examining multicollinearity among covariates were conducted, and no problematic findings emerged.

All models presented here use weighted data to enhance generalizability. Although numerous weights are available in the ANES, we use the postelection sample weight (V200010b) because it allows for generalizability to all noninstitutionalized adult U.S. citizens based on respondents who took the pre- and postelection surveys.

Results

How did the government's response to COVID-19—or people's perceptions of the government's handling of COVID-19—influence trust in the federal government? To answer this question, we examine the results of three ordered logistic regression models, which are displayed in Table 2.

Table 2.

Ordered Logistic Regression Models Showing Predictors of Government Trust, by Racial Group.

	Full sample		White sample		Nonwhite sample
	b	S.E.	b	S.E.	b	S.E.
Government pandemic response
Federal pandemic response	.32***	(.05)	.34***	(.05)	.25**	(.09)
Local pandemic response	.11**	(.03)	.07	(.04)	.23***	(.06)
Political anxieties
Government corruption	−.66***	(.03)	−.1***	(.04)	−.5***	(.07)
Concern about democracy	−.2***	(.04)	−.3***	(.04)	−.0**	(.07)
Concern about elections	−.4***	(.02)	−.3***	(.02)	−.7***	(.03)
Concern about political violence	−.3*	(.06)	−.9**	(.07)	−.3	(.10)
Concern about the country	−.7***	(.08)	−.7***	(.09)	−.0**	(.18)
Negative emotions about the country	−.2**	(.04)	−.0*	(.05)	−.5*	(.07)
Economic anxieties
National economic insecurity	−.5	(.03)	−.8*	(.04)	.01	(.06)
Personal economic insecurity	.11***	(.02)	.09***	(.03)	.13**	(.05)
Institutional sentiments
Warm feelings toward police	.01***	(.00)	.01***	(.00)	.01***	(.00)
Warm feelings toward BLM	.00*	(.00)	.00	(.00)	.00*	(.00)
Warm feelings toward the CDC	.01***	(.00)	.01***	(.00)	.00*	(.00)
Controls
White	−.5***	(.06)	–	–	–	–
Male	−.4**	(.05)	−.8***	(.06)	−.2	(.09)
Age	.00	(.00)	.00	(.00)	.00	(.00)
Married	−.8***	(.05)	−.9***	(.06)	−.7	(.10)
Children	.03	(.06)	.03	(.07)	.03	(.10)
Political identification (ref. liberal)
Moderate	−.1	(.07)	−.9***	(.09)	.27*	(.12)
Conservative	−.3**	(.09)	−.6***	(.11)	.04	(.16)
Not sure	−.8	(.09)	−.8	(.11)	−.1	(.14)

Note. All models use weighted data.

*p ≤ .05. **p ≤ .01. ***p ≤ .001.

In the full sample model, we find that one's perception of the federal COVID-19 response (b = .32, p ≤ .001) is directly correlated with one's trust in the government. Respondents who believe the federal government handled the COVID-19 pandemic well were significantly more likely to trust the government. While one's perception of their local government's response to COVID-19 was also significant (b = .11, p ≤ .01), the effect was muted in comparison to one's perceptions of the federal government's response.

This pattern varied in the race-specific subsample models. Among white respondents, only their perception of the federal government's response to COVID-19 was significant (b = .34, p ≤ .01). Among the nonwhite subsample, perceptions of both the federal COVID-19 response (b = .25, p ≤ .001) and the local COVID-19 response (b = .23, p ≤ .001) significantly related to respondents’ trust in the government. Coefficient comparisons ⁵ between the white and nonwhite subsamples revealed very few differences, but the effect of one's perception of the local government's response to COVID-19 was one of the few statistically and substantively significant differences between the models. ⁶

In addition to the government's handling of COVID, several other factors related to views of the government were significant correlates with trust. Respondents who believe the government is more corrupt (b = −.66, p ≤ .001), are more concerned about the state of democracy (b = −.22, p ≤ .001), are more concerned about the accuracy of elections (b = −.14, p ≤ .001), and believe political violence has escalated in recent years (b = −.13, p ≤ .05) are less trusting of the government. These patterns are similar in both the white and nonwhite subsamples (except for perceptions of political violence, which is not significant in the nonwhite model).

Other views about the country also correlated with one's trust in the government. Although one's perceptions of the nation's economic security were not significant, one's perceptions of their own personal economic insecurity (b = .11, p ≤ .001) was associated with higher trust in the government. One's concern about the country (b = −.47, p ≤ .001) and more negative emotions about the country (b = −.12, p ≤ .001) were both associated with more mistrust in the government. These patterns were consistent in race group subsample analyses.

Respondents’ views of the police and the CDC also play a role in their levels of trust in the government. For each one-unit increase (on a scale of 0–100) in warmth for each institution, one's level of trust in the government is associated with a .01 increase. All else being equal, a respondent who selected “100” for either variable would score a full point higher on the government trust outcome measure compared to a respondent who selected “0” for that same variable. Although views of BLM were statistically significant, they were not substantively significant. These patterns were consistent in the white and nonwhite subsample analyses, except for feelings about the CDC—which was not significant in the nonwhite subsample.

Several of the demographic controls were also significant. Males (b = −.14, p ≤ .01), those who are married (b = −.18, p ≤ .001), and those who identify as conservative (b = −.23, p ≤ .01) as compared to liberal were all less trusting in the government. These patterns varied slightly among the white and nonwhite subsamples and can be viewed in Table 2.

Standardized coefficients were also examined to compare the effect sizes of each variable. Across the full sample model, the white model, and the nonwhite model, perceptions of government corruption had the largest effect on one's trust in the government. In fact, the effect size for this variable was nearly twice as large as the second most influential variable in each model. In the main model and the white subsample model, concerns about the election, the federal government's response to COVID, and one's views of the police were the next most salient predictors. This pattern was similar among the nonwhite subsample, except one's perceptions of the local government's handling of COVID-19 exceeded one's perceptions of the federal government's handling of the COVID-19 pandemic.

Taken together, these results suggest that the government's handling of the pandemic plays a critical role in levels of public trust. Notably, the pandemic response of the federal government was often associated with people's trust in government more so than the response of their local government. The only correlate that was more consistently salient than the government's handling of COVID-19 was one's perceptions of corruption within the government. These findings were consistent across white and nonwhite subsamples.

Discussion and Conclusion

The year 2020 in the United States was saddled with many overlapping crises. The COVID-19 pandemic shut down schools, businesses, and many routine aspects of Americans’ day-to-day lives. Crime—especially violent crimes and hate crimes—increased dramatically in only a 1-year span (Gramlich, 2021; Hernandez, 2021; Kishi & Jones, 2020; Rosenfeld et al., 2021; Sganga, 2021), contributing to historic increases in public fear of crime (McCarthy, 2020). The police-involved killing of George Floyd Jr. spurred nationwide demonstrations (Buchanan et al., 2020; Putnam et al., 2020) that highlighted racial injustice in American law enforcement and sparked calls to defund or even abolish the police (Craig & Reid, 2022; Movement for Black Lives, 2020; Muhammad, 2020). Political instability was high (Hartigan et al., 2021; Kishi & Jones, 2020), and concerns about the integrity of the impending presidential election were prevalent (see Enders et al., 2021).

Although researchers have extensively examined public trust in the government over time, Hitlin and Shutava (2022, p. 5) note that “the current challenges facing the country require more than simply knowing that trust is on the decline.” Research should, instead, focus on studies “reflecting the current cultural moment and a deeper understanding of how governments should be present in people's lives” (p. 9). Hitlin and Shutava (2022) call for investigations of the determinants of trust, which researchers suggest can shift, especially during times of discontent and turmoil. International research during the early months of the pandemic suggests that the factors typically associated with public trust were less salient (Gozgor, 2022; Schraff, 2021). However, in the American context of the pandemic, little is known about the factors that predict trust or mistrust in the U.S. government. Given the various crises of the period, research in this space is needed to contextualize the environment in which the criminal justice system and its apparatuses operate, as these entities must act within the parameters set by policymakers. The current study aimed to help fill this void in the extant literature by examining the relationship between various economic, social, and political anxieties that emerged during the pandemic and Americans’ trust in government circa the 2020 presidential election.

Several key findings emerged from our analyses. First, respondents’ perceptions of the government's response to the pandemic had significant and substantial correlative effects on public trust. Notably, perceptions of the federal government's response were more salient than the local government's pandemic response. Second, political anxieties—including perceptions about government corruption, the state of democracy, the legitimacy of the impending election, increasing political violence, the country's overall direction, and negative emotions associated with the country—all contributed to diminishing public trust. Third, respondents’ warm feelings about various institutions, including the CDC and the police, were associated with greater trust in the government.

Finally, our findings demonstrate that race and ethnicity play an important role in predicting trust in government. Historically, there has been a racial divide in public trust, with communities of color typically demonstrating less trust in the government than their white counterparts (Avery, 2006; Michelson, 2003). However, our findings show a converse relationship, with white respondents significantly more likely to report lower trust than nonwhite respondents. This result could be a function of how our measure of race was operationalized—with all racial and ethnic minority groups combined into a single category labeled nonwhite. ⁷ Future research may better inform our understanding of this potential racial divide by disaggregating racial and ethnic minority groups in full-sample regression analyses. Moreover, given the period's salient sociopolitical and racialized climate, it may be worthwhile to consider the factors underlying interracial differences in trust. Additional research into race/ethnicity and government trust is therefore warranted. To the extent that whites have lower trust in government in this current context, scholars should seek to explain this otherwise unexpected relationship.

This study has several implications that also warrant emphasis. For example, our findings provide substantive context for understanding criminal offending and recent extremist violence in the United States. Prior research finds that public mistrust is associated with diminished government legitimacy and reduced compliance with established law (Hough et al., 2010; Mugarura et al., 2021). In particular, anxieties surrounding government corruption, election integrity, and the state of democracy, shown in the current study to be correlated with public trust—may contribute to political extremism and the willingness to use violence to achieve political aims. These political anxieties may have a reciprocal relationship with mistrust and political violence. For example, Šrol et al. (2022) found that diminished trust in the government was correlated with antigovernment protests and violent attacks. Edelson et al. (2017) found that increased trust negatively correlates with election fraud beliefs. If political anxieties increase mistrust and lead to violence, that violence could further erode public trust in the government. How the federal and state governments address political violence can, therefore, have repercussions for future political anxieties, public trust, and other extremist violence.

Politicians, policymakers, and other government entities should also be attentive to the role that perceptions of election integrity play on public trust. We find here that election fraud discussions—much of which was publicly fueled by politicians despite no evidence of fraud (Cummings et al., 2021; Somodevilla, 2022; Wheeler, 2021)—only serve to increase mistrust of the government. This concern about election fraud was not unique to the 2020 elections. Research finds that conspiratorial beliefs and populist values contributed to perceptions of election unfairness even in the 2016 presidential election (Norris et al., 2020). Of course, that is not to say that all concerns about election integrity are unfounded. Interference from foreign countries has been documented (see, for example, Henschke et al., 2020) and should be actively countered by U.S. officials. To the extent that the government wishes to improve public trust, it should implement safeguards against domestic and international interference to keep elections fair and transparent. When unfounded allegations of fraud are put forth, public servants should directly confront the (mis-)(dis-)information.

Trust in government may also play a salient, potentially indirect, role in reducing fear of crime. If public trust reduces offending, this may lead to perceptions that crime is decreasing, which has been shown to reduce fear of crime (Barton et al., 2017; Wanner & Caputo, 1987) and improve public trust (Chanley, 2002). These three factors—public trust, criminal offending, and fear of crime—can interact with and reinforce one another in ways that lead to favorable societal outcomes. In addition, trust in government could indirectly affect support for punitive punishment policies like the death penalty and three-strikes laws, as research has found that fear of crime contributes to punitiveness (Armborst, 2017; Callanan, 2005; Dowler, 2003; Singer et al., 2019).

Trust in government may also directly contribute to public support—or a lack of support—for crime policies and initiatives proposed by government officials and institutions. Americans less trusting of their government officials may be less likely to support proposed reforms, especially those requiring increased government funding. For example, a public that feels political violence is increasing and that the government has not done anything or perhaps cannot do anything to combat it may be less likely to support increased funding for federal agencies like the Department of Homeland Security or the Federal Bureau of Investigation. Even proposals to increase funding for crime prevention or treatment programs may face backlash from a disgruntled public simply because the public may not trust the intentions of the politicians proposing these programs. Indeed, research has found that Americans’ trust in their government impacts their support for various political initiatives (Chanley, 2002; Chanley et al., 2000; Herian, 2014; Lee et al., 2016). Future research should investigate the consequences of public trust and mistrust as it pertains to recent proposals for criminal justice policy and reform. For example, to what extent does public trust shape federal policy on proposals to reform the police after the 2020 policing crisis? Does public trust impact citizen views of broader criminal justice policy and funding?

Social science research is increasingly moving toward an interdisciplinary framework, integrating applied sciences to model complex sociological and behavioral phenomena more holistically. Consequently, it is imperative that research move beyond siloed, discipline-restricted research. Findings from the current study highlight the need for further integration of political constructs into criminological research as primary explanatory variables of consideration. We found that political factors, such as corruption, ideological leanings, political anxieties, and institutional responses, are highly connected to the public's broader trust in government. Building on a large corpus of research primarily drawn from the political science literature, we recommend that future researchers include constructs—like political trust—that may predict outcomes like criminal engagement and decision-making, support for crime reform measures, and punitiveness, among other outcomes. Such predictors would innovate quantitative modeling within criminological and criminal justice research by focusing on a more extensive range of sociopolitical factors.

Lastly, this study also points to the importance of homeland security's impact on the broader criminal justice system. For example, recent leaks of the Oath Keepers’ membership and subsequent analysis by the Anti-Defamation League (2022) explicitly identified members of law enforcement among their ranks. These revelations connect the homeland security enterprise and the criminal justice system but also pose a larger problem. The Oath Keepers subscribe to extreme conspiracies that frame the government as in opposition to American liberties. These actions directly impact social justice and governmental initiatives in the contemporary United States. Oath Keepers have been identified as a group that encouraged defiance of COVID-19 pandemic orders (MacNab, 2020) in addition to government standoffs and commentary related to the murder of George Floyd (Ai et al., 2021). Thus, “[i]t is highly problematic from a criminal justice perspective when law enforcement … are involved with these forms of far right activity, [because v]igilantism and the concept of ‘taking justice into your own hands’ run counter to the concept of a rule of law” (Dickinson, 2021).

Suppose officers, as agents of the state, are unable or unwilling to believe in the laws or the institutions that create them. In that case, the public cannot reasonably trust that they will faithfully discharge their mandates of enforcement and protection. Thus, the political and criminal components are intimately connected; consequently, criminological research would benefit from enhanced inclusion of sociopolitical correlates. To address issues with declining trust in government, a trend not restricted to the United States, the homeland security establishment should prioritize eliminating (mis-)(dis-)information spread by extremist actors and dissolving the ability of illicit networks to unduly sway public perception and, by extension, governmental policy. This aligns with an overwhelming body of literature that has identified distrust in government as a recruitment mechanism for extremist campaigns (e.g., Botha, 2013; Braizat et al., 2017).

Like all research, this study is not without limitations. First, we cannot establish causal order due to the data's cross-sectional nature. For example, one's preexisting trust in the government might predict views of how the government handled the pandemic, which is contrary to the causal order that we proposed here. Various studies have suggested alternative causal ordering between public trust and some of the independent variables included in this study (see, for example, Hitlin & Shutava, 2022; Van de Walle & Bouckaert, 2003). Future research would do well to use longitudinal analyses to examine broader trends and better establish causal order. Relatedly, it is possible (and arguably, plausible) that some of the proposed predictors may have a reciprocal relationship with trust in government. For example, someone who mistrusts their government might view the government's handling of the pandemic less favorably. These views of how the pandemic was handled may further reinforce a lack of trust in the government. These limitations have been raised in the public opinion literature in political science and criminology (see Feldman, 1983; Pickett, 2019).

Another limitation is the potential overlap between government trust and some predictors. For example, it is reasonable to suspect that people's views of the CDC, a federal institution, may intertwine with their overall views of the federal government. However, in light of the politicization of the pandemic (Stroebe et al., 2021), the CDC was under a figurative microscope that polarized public sentiment in a manner not seen before. Therefore, isolating the relationship between American attitudes toward this public health entity and the federal government's broader political structure is essential, especially during the current period. This logic also applies to other variables—such as views of the police, concerns about government corruption, and election fraud—all of which may be more salient and thus have unique effects in recent years.

Finally, it should be reinforced that the 2020 election occurred, in many ways, during unprecedented times with multiple overlapping crises—including a pandemic, a policing crisis, and concerns about the integrity of the election. It is plausible that future inquiries will demonstrate that the predictors of trust identified during this period are not generalizable to other periods. Although the United States may not find itself in this exact sociohistorical context again, these findings have implications for democratic governments that may face not just immediate crises—like hurricanes and other natural disasters—but large-scale social disruptions affecting the social, economic, and political systems of the country. In these instances, however rare they may be, governments should take care in how they respond to the crises, as perceptions of their response would likely be a key determinant in whether those governments face continued support and reelection or discontent and upheaval.

Author Biographies

Jonathan C. Reid is an Assistant Professor in the Department of Criminal Justice and Criminology at Sam Houston State University. His research examines sociopolitical inequalities, crime, and the criminal justice system, focusing primarily on violence, homicide, public opinion and policy, and police–community engagement. His most recent work appears in outlets such as Homicide Studies, Crime & Delinquency, and Social Problems.

Samantha J. Brown is an Assistant Professor at the University of North Florida's Department of Criminology and Criminal Justice. Her research focuses on the onset and maintenance of youth delinquency, school violence, social media and crime, and policing. Her most recent work appears in outlets such as Crime & Delinquency, Policing: A Journal of Policy and Practice, and Journal of Research in Crime and Delinquency.

Jared R. Dmello is an Assistant Professor of Criminal Justice & Criminology at Sam Houston State University (SHSU) and a Non-Resident Research Associate at the Rutgers University Center on Public Security. Dr. Dmello, an expert in illicit networks, focusing primarily on terrorism and criminal street gangs, has extensive program evaluation experience in the academic and nonacademic sectors. He is currently the Principal Investigator on an NIJ-supported project studying the networked evolution of gang violence in New Jersey. His work has been published in leading criminological and security studies journals, including Criminal Justice & Behavior, Crime & Delinquency, Journal of Interpersonal Violence, and Terrorism & Political Violence.