The Formal-Informal Control Nexus During COVID-19: What Drives Informal Social Control of Social Distancing Restrictions During Lockdown?

Abstract

The public rely on the police to enforce the law, and the police rely on the public to report crime and assist them with their enquiries. Police action or inaction can also impact on public willingness to informally intervene in community problems. In this paper we examine the formal-informal control nexus in the context of the COVID-19 pandemic. Drawing on a survey sample of 1,595 Australians during COVID-19 lockdown restrictions, we examine the relationship between police effectiveness, collective efficacy, and public willingness to intervene when others violate lockdown restrictions. We find that perceptions of police effectiveness in handling the COVID-19 crisis has a positive impact on the public’s willingness to intervene when others violate lockdown restrictions.

Introduction

It is an understatement to say that COVID-19 has brought immense challenges to all aspects of modern society. At the time of writing, the pandemic has resulted in over 40 million infections and over 1 million deaths worldwide. In Australia, suppression and containment strategies have been adopted with the hope of minimizing outbreaks. These containment strategies have included stay-at-home orders, contact tracing and border closures. Members of the public have also been instructed to maintain “social distance” when leaving their homes and have been encouraged to practice vigilant personal hygiene to limit the spread of the virus (e.g., regular handwashing, coughing/sneezing into elbows, staying home, and getting tested when unwell, etc.). These strategies require wide-scale public cooperation backed by formal control mechanisms such as the threat of monetary fines and arrests for those violating restrictions.

Police in Australia have become the primary formal enforcers of COVID-19 restrictions. However, as with other types of laws and regulations, police are not alone in their “policing” role. Members of the public may also play a role in policing compliance with COVID-19 laws and regulations. Community members generally have the capacity to intervene when they observe others flouting social distancing rules, whether that be by calling the police, or intervening directly (Warner, 2007). A large body of research has emphasized the importance of informal social control in crime control and prevention (e.g., Bursik & Grasmick, 1993; Kornhauser, 1978; Sampson & Groves, 1989; Sampson et al., 1997; Shaw & McKay, 1942). During the COVID-19 pandemic, when rates of non-compliance are high (Murphy et al., 2020a), and when official resources are spread increasingly thin, informal social control may play a critical role in encouraging community compliance with new regulations.

In this paper we examine the willingness of ordinary members of the public to intervene when they observe others flouting the rules. We do so in the unique context of the COVID-19 pandemic. Specifically, we focus on the relationship between formal control (or the perceived ability of police to manage COVID-19 restrictions) and informal social control (or the willingness of community members to intervene when others break the rules). Theory and research in the traditional crime prevention context points to alternative hypotheses about the relationship between formal and informal social control. On the one hand, it is argued that when police are seen to be doing a good job to prevent and control crime and disorder, community members will feel more confident to intervene informally in response to community problems (Kubrin & Weitzer, 2003). On the other hand, when police are seen to be doing a good job at preventing and controlling crime, people may not see the need for informal intervention, instead believing that police can handle and deal with crime appropriately (Silver & Miller, 2004). To examine this relationship as it relates to the COVID-19 pandemic we employ a national survey of 1,595 Australians. Before presenting our methodology and reporting our results, we first overview the literature about the “formal-informal control nexus” (Kubrin & Weitzer, 2003, p. 384). We conclude by considering the implications of our results for both theory and police practice.

Literature Review

Black (1998) defines social control as a “mechanism by which a person or group expresses a grievance” and “a mode of conducting normative business” (p. 5). Social control can be roughly divided into formal and informal mechanisms of control. In the criminological context, informal mechanisms of social control are actions taken by members of the public to uphold community norms, laws, and regulations (Bursik & Grasmick, 1993). These may include “gossip,” “scolding,” “disapproval,” and “face-to-face discussion” of and with offenders (Black 1984, pp. 5–7). On the other hand, formal control refers to the “practices of the authorities to maintain order and enforce legal and regulatory codes” (Kubrin & Weitzer, 2003, p. 381). Informal activation of formal control mechanisms (i.e., a community resident calling the police to report a crime) sits somewhere between formal and informal social control. Here the resident intervenes, but they do so by asking police to take carriage of the problem. These various types of formal and informal social controls are believed to play a key role in the prevention and control of crime.

Proponents of social disorganization and collective efficacy theories have focused their research on the positive role of informal social control in crime control and prevention (e.g., Bursik & Grasmick, 1993; Kornhauser, 1978; Sampson & Groves, 1989; Shaw & McKay, 1942; Silver & Miller, 2004). For example in their now classic study of 8,872 residents living in 343 Chicago neighborhoods (i.e., the Project on Human Development in Chicago Neighborhoods or PHDCN), Sampson et al. (1997) found that neighborhoods with higher levels of collective efficacy (a composite measure of neighborhood informal social control and social cohesion and trust) had lower rates of violent crime. Here, informal social control was defined as neighbors’ “willingness to intervene on behalf of the common good” in the face of community problems (Sampson et al., 1997, p. 918).

The replication of these results across multiple contexts, and the popular uptake of collective efficacy theory (e.g., Browning & Cagney, 2002; Franzini et al., 2005; Mazerolle et al., 2010; Odgers, et al., 2009; Sampson & Wikstrom, 2008), has led scholars to explore the factors that underpin informal social control beliefs and behaviors (e.g., Silver & Miller, 2004; Warner, 2007; Wickes et al., 2017). One factor that has the potential to impact on the ability and willingness of community members to enact informal social control is formal control. In their theoretical paper, Kubrin and Weitzer (2003) hypothesize that formal social control can impact on crime and disorder either directly (by “influencing crime and disorder” through law enforcement practices) or indirectly (“by influencing residents’ informal control practices”) (p. 382). They argue that police behavior, or perceptions of police behavior, can impact on informal social control in a number of ways: (1) when police are seen to be doing a good job at preventing and controlling crime they may “enhance residents’ capacity to fight crime and disorder”; (2) when residents view police as “unresponsive or ineffective” they “may feel vulnerable when considering whether to try to stop street deviance”; and (3) when formal control is viewed as inadequate the “vacuum” left behind may encourage community members to intervene—perhaps in nefarious or retaliatory ways (Kubrin & Weitzer, 2003, p. 384). At the time of writing, Kubrin and Weitzer (2003) highlighted that limited research existed to unpack the relationship between policing and informal social control.

Since Kubrin and Weitzer’s (2003) paper, several scholars have begun to examine the relationship between formal and informal social control or the “formal-informal control nexus” (p. 384). For example, drawing on the same study of Chicago neighborhoods as Sampson et al. (1997), Silver and Miller (2004) investigated the relationship between satisfaction with police and neighborhood informal social control. They found that satisfaction with the police was positively associated with informal social control. Specifically, the higher the level of satisfaction with the police in their neighborhood, the more likely residents were to believe that their neighbors would be willing to intervene in community problems. Using data from 5,812 residents in 123 Seattle neighborhoods, Drakulich and Crutchfield (2013) found similar results. They found that individuals’ perceptions of police efficacy were positively related to their belief that neighbors would intervene in community crime and disorder problems (see also Warner & Birchfield, 2011). Most recently, Kochel and Gau (2019) examined the effect of police engagement, police visibility, and satisfaction with police on informal social control among residents of 71 crime hot spots in St Louis County. They found that police engagement, satisfaction with police visibility, and satisfaction with police tactics were positively related to perceived social cohesion in the neighborhood, which in turn predicted self-reported informal social control in the neighborhood. Comparable results have also been found in studies of neighborhood collective efficacy (e.g., Ferguson & Mindel, 2007; Kochel, 2012, 2018; Sargeant, 2017; Sun et al., 2004). These studies find support for the idea that “increasing the responsiveness of police to the concerns of local residents will stimulate residents to more effectively resist crime on their own” (Silver & Miller, 2004, p. 558).

Neighborhood research finds support for a positive relationship between police effectiveness and informal social control or collective efficacy. However, these studies tend to focus on the anticipated informal control actions of neighbors—rather than on one’s own willingness to intervene. Warner (2007) drew attention to how informal social control might be conceptualized at the individual-level in her paper. Here, she examined individuals’ own willingness to intervene in hypothetical neighborhood disputes and distinguished between direct and indirect informal social control. Warner (2007, p. 99) defined “direct informal social control” as “direct intervention” and “indirect informal social control” as the mobilization of formal authorities (e.g., calling the police to intervene). Warner (2007) collected survey data from 2,309 residents living in 66 neighborhoods in a southern US state and asked survey participants about their own willingness to engage in intervening behaviors to resolve a disagreement with someone in their neighborhood (either directly, by either insisting “they stop the behavior” or by talking “it out until a mutually satisfactory solution is found,” or indirectly, by letting “someone else, like a landlord or police officer, resolve the issue?”) (p. 111). Warner (2007) also measured “faith in the police” (e.g., “The police play an important role in preventing crime in this neighborhood”; “the police do a good job in responding to people in this neighborhood after they have been victims of crime”) (p. 113). Warner (2007) found that average neighborhood levels of faith in the police had no effect on the willingness to intervene in neighborhood disputes either directly or indirectly. However, in an ancillary analysis, Warner (2007) found that individual perceptions of faith in the police increased the willingness to intervene in both direct and indirect ways.

In sum, prior research provides support for the hypothesis that police effectiveness may encourage informal social control and collective efficacy. However more research is needed in order to replicate these findings in different contexts. To date, only one empirical study has examined the relationship between police effectiveness (formal control) and one’s own willingness to engage in informal social control behavior (i.e., Warner’s, 2007 study of neighborhood disputes). The COVID-19 pandemic presents a novel context in which to study the relationship between formal and informal social control mechanisms. COVID-19 has brought rapidly changing laws, regulations, and norms for behavior. For example, depending on the context, members of the public are required to stand at least 1.5 m away from each other, to wear masks, to limit the number of people in their homes or businesses, to abide by strict quarantine restrictions, to wash and sanitize their hands more frequently, and to stay at home when sick. The police rely on members of the public to voluntarily comply with these norms and regulations, but also to place social pressure on others to comply. Police also play a role in setting these new norms, rules, and regulations through police enforcement by issuing fines and arresting people who do not comply. Requiring all members of the public to abide by these changing norms for behavior is an expansive task, and not one that police can facilitate alone. As police resources are stretched, and, as recent research indicates, non-compliance is high (Murphy et al., 2020a), informal social control behaviors are arguably increasingly important in order to curtail the spread of COVID-19.

The Current Study

While a handful of studies have examined the relationship between victimization and crime reporting (Boateng, 2018; Goudriaan et al., 2006; Watkins, 2005), and a wealth of studies have examined the link between procedural justice, police legitimacy, and the willingness to cooperate with police (e.g., Sunshine & Tyler, 2003; Tyler & Fagan, 2008; Murphy et al., 2008), more research is needed to unpack the relationship between formal and informal social control. In particular, there is need for a clearer understanding of the drivers of individuals’ willingness to engage in informal social control behaviors. While several studies examine the relationship between perceptions of police and informal social control or collective efficacy (Drakulich & Crutchfield, 2013; Ferguson & Mindel, 2007; Kochel, 2012, 2018; Kochel & Gau, 2019; Sargeant, 2017; Sargeant et al., 2013; Silver & Miller, 2004; see also Sargeant et al., 2018), informal social control research tends to examine perceptions of what neighbors would do, rather than to examine the predictors of an individual’s own willingness to intervene (for exceptions see Warner, 2007; Wickes et al., 2017). Further, previous research has focused on traditional crime prevention contexts, where norms for behavior and the distinction between criminal and non-criminal behavior are usually clear and well-embedded in society. For COVID-19 social distancing restrictions, previously normal and acceptable behaviors (i.e., traveling; socializing with others; leaving the house to shop) have been criminalized in order to stem community transmission of COVID-19 virus. We argue that in this context, police play an important role in setting the norms for behavior (via their enforcements of these new restrictions) and, also, that police will rely more heavily on citizens to engage in informal social control behaviors. Examining how the formal-informal control nexus holds in this context remains untested.

In this paper we extend Warner’s (2007) work to examine the predictors of individual’s willingness to engage in informal social control behaviors in response to breaches of COVID-19 regulations. We specifically focus on the role of police effectiveness in encouraging or discouraging individuals to intervene informally when they observe others flouting the rules. Importantly, we also explore the role of the perceived collective efficacy of one’s neighborhood in encouraging people to informally intervene. Theory suggests that the reason collective efficacy is related to crime in neighborhoods is due to the informal social control actions taken by residents. As Wickes et al. (2017, p. 102; see also Sampson, 2012) explain “the presence of shared expectations for informal social control, one of the key drivers of collective efficacy, convey a shared norm that residents in the neighborhood will do something when a problem arises. These shared expectations are necessary for triggering action in response to a given problem.” Prior research also suggests that police effectiveness will encourage neighborhood collective efficacy (see, e.g., Sargeant et al., 2013). In turn we might expect that neighborhood collective efficacy, or shared expectations for action, will impact upon individuals’ own willingness to intervene in community problems. Based on the review of the literature we expect to find the following relationships between our key variables:

• H1: Police effectiveness will encourage informal social control actions when others flout COVID-19 social distancing regulations;

• H2: Collective efficacy will encourage informal social control actions when others flout COVID-19 social distancing regulations and, if so;

• H3: The relationship between police effectiveness and informal social control actions will be partially mediated by perceived neighborhood collective efficacy.

Methods

COVID-19 Timeline

Following the outbreak of COVID-19 in December 2019 in Hubei Province, China, it took less than a month for Australia to record its first case of COVID-19 on 25 January 2020. Community transmission began in early March and by mid-late March case numbers were doubling every 3 to 4 days (although many cases recorded were returning travelers). Aiming to prevent large scale community transmission, the Australian federal and state governments introduced lockdown restrictions. Preliminary measures included closing the Australian border (to non-citizens and non-permanent residents), non-essential businesses, schools, and universities. Australians were strongly encouraged to stay at home, work and study from home if possible, and to limit indoor social gatherings (Duckett & Stobart, 2020). March 15 to May 1 represented the most restrictive time period. During this time only certain activities were permitted outside of the home. These included traveling to and from medical appointments or work (if work could not be done at home), shopping for “essentials items,” and exercising in one’s immediate local area. Socializing in homes was also restricted. Incoming traveler numbers were limited, and those who were permitted to enter the county were ordered to “hotel quarantine.” Quarantined travelers were not permitted to leave their hotel room for 14 days. By May 2020 the Australian public were growing increasingly frustrated with these lockdown restrictions, resulting in protests and police action. Thousands of infringement notices were issued to those flouting lockdown restrictions. For example, by May 3, the Queensland Police Service had issued 1,664 fines totaling more than AUD$2 million (Cartwright, 2020), and by May 21, the Victorian Police had issued more than 5,719 fines totally more than AUD$9.4 million (Zagon, 2020). As the curve flattened (due to strict hotel quarantine measures for return travelers and other measures) restrictions were eased in June/July 2020 across the country. Unfortunately, due to an outbreak (that appears to have originated from flaws in the hotel quarantine program) Victorian residents were forced back into lockdown on August 2 2020 (with restrictions easing in October, 2020). It appears that Australians may be in and out of lockdown as authorities continue to suppress wide-spread community transmission.

Data Collection

To examine Australians’ attitudes toward the COVID-19 lockdown, and more specifically their attitudes toward authority during this time, researchers at Griffith University launched a nation-wide survey (Murphy et al., 2020a; Murphy et al., 2020b; see also McCarthy et al., In press). The Attitudes to Authority During COVID-19 Survey began on the April 24, prior to the easing of nation-wide restrictions, but at a period of time where many Australians were caught flouting social distancing rules. The survey was conducted in Limesurvey and advertised through Facebook (for more information about the survey see Murphy et al., 2020a). Facebook surveys, while limited (e.g., sample representativeness; difficult to calculate response rates), offer a quick turnaround to researchers wishing to collect data on time-dependent or topical subjects such as the COVID-19 pandemic. A convenience sample¹ of 1,595 participants were recruited. Upon completing the survey, participants could enter a draw to win a AUD$100 voucher. A response rate of 44% was achieved (based on the number of people who completed the survey/the number of people who clicked on the Facebook advertisement).

Sample Characteristics

As the sample was a convenience sample, we compared key sample characteristics to Australian Census data (for more details see Murphy et al., 2020a). We found that the sample was broadly representative of the Australian population but was slightly over-representative of females (+6.3%) and was substantially over-representative of those with a university education (Bachelor’s degree and above, +34.1%), and those living in the State of Queensland (+12.4%). These latter two differences may be attributed to the fact that the survey was run by a Queensland university and originally advertised on the Griffith university Facebook page. We control for key sample characteristics in our analyses.

Measures

Four dependent variables were included in the analyses. These dependent variables captured participants’ willingness to engage in informal social control actions specific to COVID-19 social distancing restrictions. Participants were asked to respond to the question “If you saw people breaking the rules on social distancing how likely would you be to do the following….”: (1) “shout at them to go home,” (2) “use social media to publicly shame them,” (3) “call the police to report them,” or (4) “do nothing.” Participants indicated how likely they would be to respond in each of these four ways on a 5-point Likert scale of 1 = Not at all likely to 5 = Very likely. According to Warner’s (2007) framework, these measures represent both direct (shout at them to go home) and indirect (call the police to report them) methods of intervention. Using social media to publicly shame perpetrators could be considered either/both a direct and/or indirect method of intervention, depending on how the public shaming is executed. Obviously, doing nothing, represents an unwillingness to intervene directly or indirectly.

The key independent variable in the analysis was police effectiveness. Police effectiveness was measured specifically regarding the police response to the COVID-19 pandemic. Participants were asked to indicate how much confidence they had in the ability of the police to handle the COVID-19 crisis (“How much confidence do you have in the ability of the following institutions to handle the COVID-19 crisis?. . .Police”; measured on a 5-point Likert scale ranging from 1 = No confidence at all to 5 = A lot of confidence).

In line with prior research on the formal-informal social control nexus we also included a measure of collective efficacy. As per the definition provided by Sampson et al. (1997), collective efficacy encapsulates social cohesion and trust as well as the perceived willingness of neighbors to intervene in community problems. Participants were asked to indicate their level of agreement with the following six statements: “People in my area can be trusted”; “People act with courtesy to each other in public space in my area”; “You can see from the public space in my local area that people take pride in their environment”; “My local area is a place where people from different backgrounds get on well together”; “If I sensed trouble whilst in my local area I could get help from people who live here”; and “If any of the children or young people in my area are causing trouble, local people will tell them off.” Responses were measured on a 5-point Likert scale ranging from 1 = Strongly disagree to 5 = Strongly agree. Items were combined as a mean scale of collective efficacy (Cronbach’s Alpha = .866).

Several demographic control variables were also included in the analysis. Specifically, we controlled for age (in years), gender (1 = male; 0 = female/other), country of birth (1 = born in Australia; 0 = other), and socio-economic status. Socio-economic status was measured with two variables capturing unemployment (1 = unemployed; 0 = other) and homeownership (1 = homeowner; 0 = other).

Results

Descriptive statistics are presented in Table 1. Descriptive statistics show that, overall, participants were more likely to indicate that they would not intervene, and less likely to indicate that they would intervene by reporting to the police, shaming the perpetrator on social media, and shouting at the perpetrator. Overall, participants’ perceptions of police effectiveness in handing COVID-19 were neutral (Mean = 3.025 on a scale from 1 to 5).

Table 1.

Descriptive Statistics (N = 1,595).

	Min	Max	Mean/%*	SD
Intervene—do nothing	1	5	3.377	1.390
Intervene—report to police	1	5	2.492	1.402
Intervene—shame on social media	1	5	1.712	1.044
Intervene—shout at them	1	5	1.836	1.080
Police effectiveness—COVID-19	1	5	3.025	1.217
Collective efficacy	1	5	3.639	0.758
Age	18	89	49.818	14.467
Malea	0	1	43.0%	−
Born in Australiaa	0	1	77.2%	−
Unemployeda	0	1	7.5%	−
Owns homea	0	1	65.1%	−
Tertiary educateda	0	1	56.1%	−

^*% Provided for dichotomous variables; a. reference category = other.

Regression results are presented in Table 2. Five multivariate outliers were detected using Mahalanobis distance and removed prior to the final analyses. To address concerns about normality (i.e., the distribution of each dependent variable was skewed toward not intervening) we employed robust standard errors. Separate regression analyses were run for each of the four dependent variables: “no intervention—do nothing”; “indirect intervention—report to police,” “indirect/direct intervention—shame on social media,” “direct intervention—shout at them.”

Table 2.

Regression Analyses of Independent Variables on Dependent Variables with Robust Standard Errors (N = 1,572).

	Intervene—do nothing			Intervene—report to police			Intervene—shame on social media			Intervene—shout at them
	b (SE)	t	β	b (SE)	t	β	b (SE)	t	β	b (SE)	t	β
Age	−.011 (.002)	−4.37***	−.113	.009 (.002)	3.81***	.095	.004 (.002)	1.92^	.054	.005 (.002)	2.39*	.065
Malea	.166 (.068)	2.45*	.059	−.206 (.063)	−3.25 **	−.073	−.009 (.055)	−0.17	−.004	.025 (.057)	0.43	.011
Born in Australiaa	.002 (.078)	0.02	.000	−.053 (.074)	−0.72	−.016	−.016 (.064)	−0.25	−.006	.194 (.061)	3.16**	.076
Unemployeda	.162 (.118)	1.37	.030	−.038 (.116)	−0.33	−.007	.032 (.102)	0.32	.008	−.069 (.098)	−0.70	−.016
Owns homea	.181 (.074)	2.44*	.062	−.127 (.071)	−1.79 ^	−.043	−.042 (.061)	−0.68	−.019	−.043 (.061)	−0.70	−.019
Tertiary educateda	.031 (.066)	0.47	.011	.030 (.062)	0.48	.010	−.047 (.053)	−0.90	−.022	−.021 (.055)	−0.38	−.010
Police effectiveness—COVID-19	−.416 (.028)	−15.09***	−.362	.548 (.024)	22.65***	.474	.111 (.022)	5.05***	.128	.116 (.023)	4.98***	.130
Collective efficacy	.003 (.046)	0.07	.002	−.077 (.043)	−1.80^	−.041	−.055 (.038)	−1.45	−.039	.004 (.040)	0.10	.003
Intercept	4.952 (.211)	23.49***		.849 (.200)	4.23***		1.448 (.178)	8.13***		1.111 (.183)	6.07***
F	43.07 (8, 1,563)***			83.37 (8, 1,563)***			4.34 (8, 1,563)***			6.06 (8, 1,563)***
R 2	0.1597			0.2511			0.0205			0.0272

^***p ≤ .001. **p ≤ .01. *p ≤ .05. ^p ≤ .10; ^areference category = other.

Results for the key independent variables are fairly consistent across each of the four models. We find that, in support of Hypothesis 1, confidence in the police ability to handle COVID-19, or police effectiveness, is the strongest predictor of each dependent variable. That is, we find that those participants in our sample who were confident in the ability of police to handle COVID-19 were more likely to indicate they would be willing to intervene both directly (shouting at them; t = 4.98, p ≤ .001) and indirectly (reporting to the police; t = 22.65, p ≤ .001) in response to seeing someone flouting social distancing rules. Participants who had more confidence in the police ability to handle COVID-19 were also more likely to indicate that they would intervene by shaming the perpetrators on social media (t = 5.05, p ≤ .001). Similarly, those participants who were confident in the ability of police to handle COVID-19 were less likely to report that they would “do nothing” in response to seeing someone flouting social distancing rules (t = −15.09, p ≤ .001). These results suggest that when community members are confident that the police can handle COVID-19 they will feel supported to intervene when people are flouting social distancing rules, whether that be by calling on the police to manage the problem or intervening on their own.

Collective efficacy on the other hand is not a predictor of the willingness to intervene (or of doing nothing) across the four models (although the p-value approached significance when predicting the willingness to intervene by reporting to the police: t = −1.80, p ≤ .10). These results suggest that collective efficacy is not particularly relevant to residents’ willingness to intervene either directly or indirectly in this context. Thus, we did not find support for Hypotheses 2 or 3.

When looking at the effects of the demographic variables in the models we again see consistencies across the four models. We find that age appears to have a positive impact on the willingness to intervene. Specifically, older people are more willing to intervene, whether that be directly (by shouting at them, t = 2.39, p ≤ .05) or indirectly (by reporting to the police, t = 3.81, p ≤ .001), and less likely to do nothing (t = −4.37; p ≤ .001). Interestingly, we also find that older people are also more likely to indicate that they would shame people on social media compared to their younger counterparts (t = 1.92, p ≤ .10).

While there are consistencies, we also find some differences when looking at the effects of the demographic variables across the four models. While gender and home ownership are not related to the willingness to intervene either directly (by shouting at them) or by shaming perpetrators on social media, these variables are related to the willingness to report to the police and the willingness to do nothing. That is, male participants were more likely to say they would do nothing (t = 2.45, p ≤ .05), and less likely to say they would report to the police (t = −3.25, p ≤ .01), when confronted with people flouting social distancing rules. Similarly, homeowners were less likely to indicate that they would intervene by reporting violations to the police (t = −1.79, p ≤ .10), and were also more likely to indicate that they would do nothing (t = 2.44, p ≤ .05), when confronted with people flouting social distancing rules. While this second finding seems counter-intuitive (i.e., homeowners are typically perceived to care more about what goes on in their local space/community), it could also be that homeowners are more likely to know their neighbors and that the resulting social bonds prevent them from engaging in conflict. This may be especially true during the COVID-19 pandemic because breaching social distancing regulations would typically not be viewed as “criminal” or “deviant” outside of the pandemic context. As such, homeowners may be less willing to engage in conflict over an issue that will likely be resolved when the pandemic is over.

The final three demographic variables included in the models are: born in Australia, unemployed and tertiary educated. Being unemployed or tertiary educated does not have a significant impact on the willingness to intervene (or not) across the four models. On the other hand, being born in Australia is positively and significantly associated with the willingness to intervene directly (by shouting at them, t = 3.16, p ≤ .01) but not to the two other forms of intervention or to doing nothing. This finding may indicate that those born in Australia feel a greater sense of ownership over shared spaces, and perhaps a greater familiarity with rules and regulations. They may consequently feel more confident to intervene directly when people are flouting social distancing rules. On the other hand, it may be that those born overseas are more likely to experience a language barrier, which may reduce one’s knowledge of the current rules and regulations and impact on one’s confidence to confront someone who is breaking the rules.

Discussion

Few studies have examined the “formal-informal control nexus” (Kubrin & Weitzer, 2003, p. 384). Those that do, tend to focus on neighborhood informal social control or collective efficacy, rather than on an individual’s own willingness to intervene. In this paper we examined the predictors of one’s willingness to enact both direct and indirect informal social control in the unique context of the COVID-19 pandemic. Specifically, we focused on the relationship between confidence in the ability of police to handle the COVID-19 crisis (i.e., police effectiveness), collective efficacy, and participants’ willingness to intervene in response to others flouting social distancing rules during the COVID-19 lockdown. Our results shed light on the factors that may contribute to informal social control actions in this unique context. However, we also suggest that these results make a broader contribution to theory and research about the formal-informal control nexus.

Before we discuss our key findings, it is important to note the limitations of the current study. First, the data utilized herein were collected via a cross-sectional Facebook survey with convenience sampling. The use of Facebook and other online platforms are becoming increasingly popular in policing and criminological research (see, e.g., Pickett et al., 2018). However, there are limitations associated with using Facebook as a sampling mechanism including that the sampling method is non-random, and that the sample includes only Facebook users who accessed the survey. It goes without saying that even as this method of sampling allowed us to act quickly to collect data during the COVID-19 lockdown, our results should be interpreted with caution.

A second limitation concerns our measures of intervention behaviors. When we developed our survey instrument at the onset of the COVID-19 pandemic, it was difficult to imagine what intervention regarding social distancing violations might look like. On reflection, our measures (which included the informal social control actions of “shouting,” “shaming,” and “calling the police”) are not comprehensive, and do not allow for an examination of more subtle forms of intervention. Future research in this vein could examine other informal social control actions such as actively modeling positive behaviors (moving to increase interpersonal space to 1.5 m), displaying non-verbal disapproval, or calmly advising someone of a restriction they may be contravening.

Despite these limitations, our findings do support our key hypothesis of interest. We found a positive and significant relationship between police effectiveness and all three modes of informal social control measured, and a negative relationship between police effectiveness and “doing nothing.” While the pattern of results are similar for each of our four dependent variables, it is likely that the mechanisms behind these relationships may be slightly different.

For example, when reporting a COVID-19 social distancing violation to the police, the individual will likely be expecting the police to take action in response to the report. As such, it makes sense that they would be less likely to contact the police to report such an incident if they did not expect that the police would have the capability to respond. The same could be said for other types of criminal behavior; people may be unlikely to call the police to report a crime if they anticipate that the police will not have the ability or capacity to respond to the complaint (Anderson, 1999; Warner, 2007).²

When examining more direct forms of intervention (i.e., shouting at the perpetrator, shaming them on social media), police effectiveness may be important because people rely on the indirect support of the police. For example, people may be more likely to intervene directly when they feel that they can rely on the police to “back them up” if the need arose (Kubrin & Weitzer, 2003; Silver & Miller, 2004).

There may also be a “normative” element to explaining these results. When police are seen to be doing a good job at responding to COVID-19 (by, e.g., issuing fines to people who flout social distancing rules), police are setting the norm that these types of violations will not be tolerated. Norm setting by police may instil confidence in community residents to intervene in community problems with the view that they have “right on their side.” Kochel (2012) explains how this process works in the neighborhood setting: “When significant proportions of residents doubt the legitimacy of institutions such as the police, the conventional values that police represent will become less respected, eroding a foundation on which neighborhoods can build consensus about the appropriate behaviors to expect in the neighborhood” (p. 389). These same mechanisms may explain why we found a negative relationship between police effectiveness and “doing nothing” in our survey. If people do not have confidence in the ability of police to handle the COVID-19 pandemic, they may be less likely to intervene either directly (because they cannot rely on the police to resolve the problem) or indirectly (because they cannot rely on the police to establish norms for behavior).

Given the positive relationship between police effectiveness and the willingness to enact informal social control, it is surprising that we did not find a similar relationship between collective efficacy and our dependent variables. We expected that local collective efficacy would support the willingness to intervene in the same way as police effectiveness. That is, if survey participants believed that their neighbors would enact informal social control in response to local problems then they may be more likely to do the same due to the establishment of norms for behavior. Indeed, one of the “longstanding theoretical assumptions” in the collective efficacy literature is that these “shared expectations affect crime rates by activating informal social control actions” (Wickes et al., 2017, p. 103). However, we found that collective efficacy had no effect on the willingness of survey participants to engage in informal social control actions in response to COVID-19 social distancing violations.

There are several reasons why this hypothesized relationship might not be evident in our data. First, our collective efficacy measure was not geared toward COVID-19 specifically (unlike the police effectiveness variable). Our measure of collective efficacy asked participants to report on their level of agreement with, for example,: “If I sensed trouble whilst in my local area I could get help from people who live here”; and “If any of the children or young people in my area are causing trouble, local people will tell them off.” None of the items employed to measure collective efficacy specifically referenced the COVID-19 pandemic. Second, our measure of collective efficacy referred to the neighborhood context, however our measure of the willingness to intervene did not refer specifically to neighbors or one’s neighborhood. That being said, during the period of COVID-19 lockdown, it would have been unlikely that survey participants were traveling far beyond their immediate neighborhoods in their day-to-day lives. Third, and relatedly, in prior research collective efficacy is primarily measured as a neighborhood construct. However, our data is not neighborhood data—we did not collect our sample using a hierarchical sampling method to capture residents living in particular neighborhoods, so we could not aggregate our measure of collective efficacy to the neighborhood level. Alternatively, it may be the case that collective efficacy does not work to encourage informal social control action in this context. This could be a result of the unique global context of COVID-19. While COVID-19 does affect people differently, and the risk is more prevalent in some areas compared to others, COVID-19 is not a neighborhood problem. It may therefore be that shared expectations for informal social control in one’s local area is irrelevant. Having said that, the absence of a relationship between expectations for action in one’s neighborhood and actual behavior is not new. These results are consistent with those of Wickes et al. (2017, p.101) who found that “shared expectations for action” did not predict individuals informal social control responses to community problems.

Conclusion

Since Kubrin and Weitzer (2003) noted the formal-informal nexus was an important direction for criminological research, few studies have examined this relationship and research rarely considers the link between perceived formal control, and one’s own willingness to engage in informal social control behavior. Our paper adds to the literature on the formal-informal control nexus by exploring the predictors of the willingness to intervene in the face of COVID-19 lockdown. Our results demonstrate that police effectiveness is an important predictor of one’s willingness to intervene informally when others flout COVID-19 lockdown restrictions. These results suggest that, in the face of challenging times that disrupt norms for public behavior, police play an important role in norm-setting and facilitating informal social control behaviors in the community. The role of police and the community in setting and enforcing norms for behavior are particularly salient in the context of COVID-19 when rules for behavior are rapidly changing, and non-compliance with the rules may be subsequently high (Murphy et al., 2020a).

Author Biographies

Elise Sargeant is a senior lecturer in the School of Criminology and Criminal Justice and the Griffith Criminology Institute at Griffith University. Elise graduated with her PhD in Criminology from the University of Queensland in 2012. Her research focuses on public perceptions of police, police legitimacy, procedural justice, ethnicity and neighborhood effects.

Kristina Murphy is a professor and Australian Research Council Future Fellow at the Griffith Criminology Institute at Griffith University. She received her PhD in psychology in 2002 from the Australian National University. Her research focuses on the association between procedural justice and trust in authorities.

Molly McCarthy is a postdoctoral research fellow at the Griffith Criminology Institute at Griffith University. Her research interests include police use of force, the impacts of policing responses with different communities, youth offending and life-course criminology.

Harley Williamson is a research fellow at the Griffith Criminology Institute at Griffith University. Harley’s research focuses on policing, countering violent extremism, and understanding the impacts of threat perceptions.