Satisfaction with the Police: Perceptions and Related Variables from an Urban Community Sample

Abstract

This quantitative, correlational, and transversal study was performed with a sample of 482 participants from an urban community at the Historic Centre of Porto (HCP). Participants answered to an enquiry designed to collect information about (dis)satisfaction with the police and its performance, with sociodemographic, victimisation, criminal, environmental, social control, and community variables as potential predictors. Findings revealed that the community was mainly satisfied with the police in its efforts to guarantee security, and there was no relationship between those variables, and sociodemographic and some community variables (e.g., years at the HCP, willingness to collaborate in security measures, and strength of attachment to HCP). On the other hand, there were relationships of (dis)satisfaction with the police and being the victim of crime, and some criminal and environmental variables (e.g., perception of increased criminality, conditions promoting crime, and incivilities). Regression analyses found that the perception of increased criminality and the need to adopt improvement measures were significant predictors of dissatisfaction with the police. This study promotes further discussion on factors that can be improved to increase satisfaction with the police and the connection of community–institutions to promote community security.

1. Introduction

The United Nations (2019) indicates a global perspective of the growth of urban spaces while rural spaces decrease. The transformation of physical and social environments due to urbanisation brings forth mainly positive changes, translated usually into economic growth, technological tools, and the speed of access to goods and services, increasingly extensible to the entire population (Sani et al. 2019). Nevertheless, it can also be associated with some concerns related to evident transformations on lifestyles, culture, and people behaviour (National Research Council 2003). Urbanisation may be a sign of improvements in quality of life of community (Nunes et al. 2017), despite not always being available to an entire population of a certain limited area. It also potentiates social inequalities through exclusion processes (Hope and Karstedt 2003) and many urban spaces become unsafe, as physical structures and the new social dynamics (Sani and Nunes 2013a) interfere with behavioural patterns (e.g., through the limitation of mobility, a decrease in social interaction with the community, and the adoption of strategies of self-protection) (Caliso et al. 2019). Such changes may deeply impact people’s lives, well-being, and overall quality of life (Makita et al. 2020).

Some concerns related to public peace and security in spaces of social interaction emerge from the occurrence of deviant and criminal phenomena in urban spaces (Fernandes and Rêgo 2017; Skogan 1987; Tilley 2013). Therefore, if phenomena such as violence and crime become part of the everyday life of populations and transform such areas into “disqualified peripheries” (Fernandes and Mata 2015), the search for an equilibrium regarding security is a pressing need that should be guaranteed by the state. So, policing arises as an urgent social response by the state that can (or cannot) be considered legitimate by populations (Oliveira et al. 2019), but that is established to prevent crime and victimisation, to promote community security, and to contribute to the sustainable development of countries (United Nations Office on Drugs and Crime 2010). Therefore, it is relevant to understand which the perception of the urban community is about social control mechanisms, and which variables support this perception. These aspects may be identified through the collection of (objective and subjective) indicators of the population on the basis of their perceptions (Austin et al. 2002; Cardoso et al. 2019), which can be assessed with tools such as the internationally tested and valued Diagnosis of Local Security (DLS; General Directorate of Internal Administration 2009; Sani and Nunes 2013b; Sani and Nunes 2016; Sani et al. 2020).

This work is part of Project LookCrim (supported by the Fundação para a Ciência e a Tecnologia—PTDC/DIR-DCP/28120/2017), that was designed to collect data from different sources (e.g., official statistics, self-reports, and space observations) concerning the specific geographic area of the Historic Centre of Porto (HCP), Portugal. This article addresses the perceptions of this specific community about formal social control, namely, (dis)satisfaction with the police and its related variables and predictors. Briefly, this paper is organised as follows: we first review the literature regarding the topic, introducing the main variables and perspectives; second, we present the methodological aspects followed by a detailed description of the findings. The discussion section ends this work with a comparison of our results with previous studies along with potential explanatory reasons beyond our findings. Implications and limitations of our research are also identified in the discussion.

1.1. Community Perceptions about Formal Social Control—Policing

Public satisfaction with police has been a concern for a long time in the debate regarding police effectiveness (Bolger et al. 2021; National Institute of Justice 2002). Policing characterises the professional performance of police agents that, in many countries and through different approaches, intervene in order to watch, fight against, and prevent crime (Mawby 2011). It is performed in accordance with emerging changes in physical and social contexts, considering the major concerns of focused populations (Bengochea et al. 2004). The greater the proximity between social control forces and the community is, the deeper the knowledge about geographical areas, and the better the acknowledgment of population needs can be (Cusson 2000). A community’s satisfaction with the police also contributes to increasing the involvement of the community itself in the identification and resolution of criminal and noncriminal problems (Merenda et al. 2021). This is a policing model of proximity or community (Skolnick and Bayley 2006) that may benefit from better confidence in police, promoting a higher civic participation of population in the maintenance of security (Bayley [1985] 2006; Community Oriented Policing Services 2009; Gill et al. 2014; Merenda et al. 2020). This model characterises the organisational structures of European countries such as Portugal, which from the 1990s onwards implemented “neighbourhood police stations” closer to the populations (Cruz and Fernandes 2021).

At urban contexts, aiming to promote civility and confidence of population, police may contribute to cooperative efforts in the construction of security at physical spaces and social interaction (Sani et al. 2019). The acceptance and legitimacy of the police are directly associated with the population trusting this formal control entity (Tyler 2006). This constitutes, among other things, an essential element in the definition of security and well-being of populations (Oliveira et al. 2019). In this sense, the perception of the community about the legitimacy of police may impact the confidence of the populations through their proximity to formal social control (Farren et al. 2018; Gray et al. 2019). “Procedural justice”, as mentioned by Tyler (2006), results from the respectful, dignified, and transparent way in which the population perceives police performance, legitimising their authority (Oliveira et al. 2019). If there is legitimacy, there may be trust, and the community may be more willing to cooperate in reporting situations and complying with police rules and the law in general, which also has implications for the effectiveness of the police in criminal prevention (Braga 2015; Gill et al. 2014). Merenda et al. (2020) found that procedural justice was the key predictor of satisfaction with police, particularly through indirect community contact.

To the best of our knowledge, available information about trust in police and justice institutions in the Portuguese context is based on the fifth round of the European Social Survey, previously analysed by Jackson et al. (2011), which is probably outdated. Concerning these data, most individuals did not have any contact with the police in the two previous years. Among those that had answered positively, most were satisfied, and similar percentages were neither satisfied nor dissatisfied. The authors also concluded that “there is no necessary correspondence or match between (a) how much contact people have with the police and (b) levels of satisfaction with that contact.” (p. 4). Moreover, Portuguese participants rated police effectiveness lower than the remaining nineteen countries did despite authors arguing that this perception seemed to be affected by being (or not) a victim of crime. Overall, authors concluded that, in comparison with Nordic countries, Southern European countries seemed to be less trusting in police.

1.2. Perception about Policing: Related Variables

Public satisfaction with police is linked to various variables, organised in three models asserted by Brown and Benedict (2002) after reviewing more than 100 empirical studies about this topic, which comprises: (i) demographic factors, (ii) prior contact with the police, and (iii) neighbourhood conditions.

1.2.1. Sociodemographic Characteristics Model

The first model is focused on sociodemographic characteristics such as age, gender, education, and socioeconomic status (Boggs and Galliher 1975; Peek et al. 1981). Regarding age, it is suggested overall that younger individuals demonstrate lower levels of satisfaction than those of older communities (De Angelis and Wolf 2016; Reisig and Parks 2000; Sullivan et al. 1987). Reisig and Correia (1997) clarified this from the perspective of younger individuals who look at police as a severe threat towards their independency, while older individuals appreciate the order proportioned by the police. Moreover, Dai et al. (2019) noted that this finding could also be explained by the fact that elderly people’s contact with the police is mostly initiated by them, and contact initiated by police usually happens to younger people. Both gender and education seem to have mixed outcomes throughout the literature. Some studies concluded that females report higher levels of satisfaction with police (Lai and Zhao 2010; Reisig and Correia 1997), which could be explained by the higher probability of men to be involved in situations that result in them being arrested or the use of force. Nonetheless, another study showed the opposite (Ratcliffe et al. 2015), and others suggested that it has no impact on this issue (Dunham and Alpert 1988; Reisig and Parks 2000). Studies about education showed that this is not a consistent variable when related to satisfaction with police. Indeed, some studies report that higher education positively influences levels of satisfaction with police (Dai and Jiang 2016; Reisig and Parks 2000), while others showed the opposite (Wu 2014) or even that education level was not a statistically significant predictor of satisfaction (Lai and Zhao 2010). Lastly, socioeconomic status is generally assessed as income and has not been considered a significant variable of satisfaction with police overall; however, those with higher incomes seem to usually be more satisfied with police services (De Angelis and Wolf 2016; Reisig and Parks 2000). In a recent meta-analytic work about global community perceptions of satisfaction with the police (Bolger et al. 2021), findings showed that females and older participants had greater levels of satisfaction in the police. It was claimed that, especially among dissatisfied groups, improving perceptions regarding police would benefit compliance and community willingness to report crimes. Despite this study including some data about European countries, such as Belgium and Netherlands, it dismissed Portugal.

1.2.2. Prior Contact with Police Model

Ongoing studies about satisfaction with police also concentrate on the contact with police and victimisation experiences (Russell and Light 2006), which are in the second model. For instance, Sani and Morais (2015) found that, in the case of domestic violence, the way that police respond to the victims significantly impacts victim satisfaction. Regarding the relationship between personal experience and level of satisfaction with the police, the topic has been addressed for a long time as a significant factor of community satisfaction (Boggs and Galliher 1975; Dean 1980) and as the most significant one by Scaglion and Condon (1980). Most findings suggested that those who were satisfied with police contact demonstrated higher levels of satisfaction regarding its intervention than those who had had no contact with police (Reisig and Parks 2000). Individuals who were not satisfied with police’s contact were even less satisfied with police intervention than those who had had no contact were (Dai and Jiang 2016; Li et al. 2016; Reisig and Parks 2000). Dai et al. (2019) noted that individuals without higher education are more influenced by their prior contact with police (that represented a negative experience) than those with higher education were. Research studies found significant differences between contacts that were initiated by the community (e.g., service calls) and those initiated by police (e.g., traffic stops). A community who initiated contact showed higher levels of satisfaction towards the police than those of those who had experienced police-initiated contact (Dean 1980; Rosenbaum et al. 2005; Reisig and Correia 1997). Furthermore, the knowledge that somebody has had contact with the police, either positive or negative, which comprises a vicarious experience, seems to influence community opinions about police (Rosenbaum et al. 2005). Regardless of the police contact’s nature, its quality and treatment are also relevant because the community think more greatly of the police when officers seem to be reasonable, respectful, fair, empathic, and concerned (Birzer 2008; Reisig and Correia 1997; Wells 2007).

Another significant variable seems to be police work ratings concerning their collaboration with the local community. A study conducted by Yuksel and Tepe (2013) confirmed this as the most important predictor of community satisfaction with the police, as a community that has a positive opinion on the police’s job in working with them to resolve their own problems showed being considerably more satisfied with the police. Other researchers, such as Orr and West (2007), have considered that the role of political and social beliefs on community impressions would be more impactful than direct previous experiences with crime (and thereby with the police). By examining attitudes about local police, these authors found that both were pertinent concerning satisfaction with the police, with personal experience with crime and the police influencing their impressions more. The authors also stated that prior contact was more important in regard to police courtesy and fairness.

1.2.3. Neighbourhood Conditions Model

Although studies about neighbourhood conditions’ effect on public satisfaction with police are growing, these are still restricted. According to Reisig and Parks (2000), factors such as racial or ethnic composition, poverty, unemployment, rates of violent crime, physical and social disorder, and the perception of security or quality of life could affect attitudes toward the police. For instance, the community perception of their neighbourhood and levels of satisfaction with the police seem to be related, as stated by Brown and Benedict (2002). Additionally, individuals who lived in neighbourhoods deprived of economic stability and resources are associated with lower satisfaction with the police (McNeeley and Grothoff 2016). Findings also suggest that perceptions of crime and incivilities are consistent with lower levels of satisfaction with the police (Ratcliffe et al. 2015; Reisig and Parks 2000). Incivilities, as Roché (1991) introduced, involve minor violations of norms adopted by the community, which can be disorderly physical surroundings, such as litter, graffiti, abandoned cars, or disruptive social behaviours, such as panhandlers, public drinking, and thoughtless neighbours (LaGrange et al. 1992). These, along with criminal occurrences, contribute to the loss of trust in the police’s job by reflecting the inability of authorities to control these problems and guarantee an environment of safety and social order (Lourenço 2010).

The feeling of security is also a relevant factor on the level of satisfaction with the police. The majority of research studies have shown that those who have a great sense of security are more satisfied with police (Reisig and Parks 2000), and those who feel unsafe or had been victimised demonstrated less satisfaction (Yuksel and Tepe 2013). It is pertinent to consider several other factors that influence the perception of (in)security, for instance, sociodemographic factors (e.g., age, gender, education, employment situation, and household status), previous victimisation experiences, and neighbourhood conditions (e.g., incivilities) (Carro et al. 2010; Curiel and Bishop 2016; Valera-Pertegas and Guàrdia-Olmos 2017). Additionally, victimisation experiences influence the individual’s perception on feelings of security (Azevedo et al. 2021, 2022). The perception of (in)security is influenced by both direct (Kullberg et al. 2009; Visser et al. 2013) and indirect victimisation experiences. As Covington and Taylor (1991) noted, residents who learnt about recent crime in their neighbourhood presented a high perception of insecurity. This in turn affects their level of satisfaction towards the police (Bolger et al. 2021; Lai and Zhao 2010).

1.3. Current Study

To the best of our knowledge, no previous study about this topic has been conducted in Portugal; even in the international context, most studies focused solely on one kind of variables. The current article explores the perception of (dis)satisfaction with the police and related variables. More specifically, on the basis of an urban community sample from the Historic Centre of Porto (HCP), the present study: (i) describes perceptions of (dis)satisfaction about the police and adequacy of policing; (ii) tests associations between (dis)satisfaction with the police and sociodemographic, criminal, environmental, social control, and community variables; and (iii) develops a predictive model of (dis)satisfaction with the police.

2. Materials and Methods

2.1. Participants

Residents, workers, and students at the HCP were randomly recruited from the streets of this geographic area of the city to participate in this study. The inclusion criteria were an age equal to or above 18 years and being able to speak Portuguese fluently. Thus, data were obtained from 554 participants, although only 87% answered the question about satisfaction with the police. Therefore, the community sample was constituted by 482 individuals aged between 18 and 96 years old (M = 44.53, SD = 18.28). As shown in

Table 1. Participant description (N = 482).

Variables	Frequency
	n	%
Gender
Men	204	42.3
Women	278	57.7
Nationality
Portuguese	445	92.3
Other	37	7.7
Brazilian	24	64.9
Spanish	2	5.4
French	3	8.1
Angolan	2	5.4
German	1	2.7
Dutch	1	2.7
Other	4	10.8
Marital status
Single	204	42.3
Married/cohabiting	199	41.3
Divorced/separated	57	11.8
Widowed	22	4.6
Age group
18 to 24 years	97	20.1
25 to 34 years	69	14.3
35 years to 44 years	81	16.8
45 to 54 years	93	19.3
55 to 64 years	63	13.1
65 or above	79	16.4
Education
1st to 4th years	53	11.0
4th to 6th years	46	9.5
7th to 9th years	57	11.8
10th to 12th years	168	34.9
University education	157	32.6
Other	1	0.2
Professional status
Student	74	15.4
Active worker	318	66.1
Unemployed	21	4.4
Retired/pensioner	65	13.5
Other/did not answer	3	0.6

, almost 58% of the participants were females, and 90% were Portuguese. The modal category for marital status was “single” followed by “married/cohabiting”. Additionally, nearly 35% had secondary education, while one-third had university education. Lastly, the majority of participants were active workers.

2.2. Measure

“Diagnosis of local security” (omitted reference) was the measure applied to gathered data. It includes both closed- and open-ended questions, allowing for both quantitative and qualitative coding. The questionnaire included 61 items organised into five sections: (i) sociodemographic information; (ii) perception of (in)security; (iii) victimisation; (iv) social control; and (v) community participation.

The central topic addressed in this study of satisfaction with the police was assessed in section (iv), social control, asking participants to “rate your level of satisfaction with the police at HCP”. Six answer options were provided: very satisfied, satisfied, a little satisfied, not satisfied, do not know, do not want to answer. Answers of “do not know/want to answer” were excluded from analysis. Besides the question about satisfaction with the police, section (iv) gathered information about the adequacy of policing, and seeking formal and informal support.

Moreover, the first section was devoted to sociodemographic information. Information was gattered regarding gender, age, nationality, marital status, and professional status. To assess victimisation experiences, we analysed two questions presented in the respective section that asked about direct and indirect experiences of victimisation in the last five years at HCP. For the participants that answered positively, other information related to the experience was collected. In this study, we considered as victims those participants that answered “yes” to at least one of the two questions “In the last five years, were you victim of crime?” and “In the last five years, was someone you know victim of crime?”. The section focused on the provided information about the perception of (in)security about criminal and environmental variables. We analysed the perception of increased criminality, asking participants if they thought that criminality had been increasing (yes vs. no vs. do not know/want to answer). Additionally, from a list of fourteen crimes (e.g., fraud, robbery, sexual offense, domestic violence), participants were asked to select those crimes that they described as frequent (the sum provided the total number of criminal occurrences) and those that they most feared (the sum provided the total number of most feared crimes). Then, from a list of twelve conditions (e.g., poverty/unemployment, poor lighting), participants were asked to choose those which promoted crime occurrences (the sum provided the total number of conditions promoting crime), and from a list of seven incivilities (e.g., urinating on public roads, produce noise), to identify the most common ones (the sum provided the total number of incivilities). We also asked through blank spaces about suggestions to improve security at the HCP (the sum provided the total number of suggested improvement measures).

2.3. Procedures

After the approval of the research project, authorisation for data collection was obtained from the Parish Council of HCP. Thereafter, participants were invited to collaborate in a study about perceptions of (in)security and crime at HCP, recruited in streets, squares, shops, offices, schools, green parks, and other public and private spaces at HCP. Study procedures and the conditions of participation were presented, and written informed consent was obtained from all individuals who had agreed to participate. A survey directed to HCP residents was performed by students with a psychology or criminology graduate or postgraduate degree (N = 10) who had received a formal training session with roleplay exercises before moving to the streets. Then, they observed a senior researcher performing a couple of surveys, and their initial surveys were also accompanied by a senior researcher. Participants did not receive any incentive to enrol in the study and were assured of the voluntary nature of their participation, their right to refuse to answer, and the privacy of their responses.

2.4. Statistical Analyses

Data were analysed through IBM Statistical Package for Social Sciences software (IBM SPSS for Windows, version 25.0, IBM Corp., Armonk, NY, USA). Dataset and syntaxes are available upon request to the authors.

According to our aims, univariate descriptive statistics (i.e., relative and absolute frequencies) were computed for satisfaction with policing performance and the adequacy of policing. Previous to further statistical analyses, four-point Likert scales were recoded into dichotomised variables, namely, satisfaction with the police was dichotomised into satisfied (including very or satisfied at all) and dissatisfied (including little or not satisfied at all). Additionally, the perception of adequacy of policing, seeking formal and informal support, the willingness to collaborate to improve security, and strength of attachment to HCP were dichotomised.

Moreover, to test associations between variables, inferential statistics were computed, namely, chi-squared. The involved variables were perception of satisfaction with the police and sociodemographic, victimisation, criminal, environmental, social control, and community variables. Considering those variables that were significantly associated at the previous step, logistic regression was performed to identify predictors of (dis)satisfaction with the police (third aim).

3. Results

3.1. How Satisfied Were Participants with the Police?

The modal category was “satisfied” (44.0%, n = 244), followed by “little satisfied” (26.9%, n = 149), “very satisfied” (10.3%, n = 57), and “not satisfied” (5.8%, n = 32); 13% (n = 72) of the participants did not know or not answer. When asked if the police performed in the best way to guarantee safety, 34.7% (n = 192) answered “rarely”, 33.8% (n = 187) answered “frequently”, 17.0% (n = 94) answered “always”, and 5.6% (n = 31) answered “never”; 9% (n = 50) of the participants did not know or did not want to answer the question.

3.2. Which Variables Were Related to (Dis)Satisfaction with the Police?

As shown in

Table 2. Frequencies and inferential tests for police performance satisfaction by sociodemographic, victimisation, criminal, environmental, social control, and community variables.

Sociodemographic Variables	Satisfied with Police Performance		Χ2	t(480)	p
	Yes % (n)/M (SD)	No % (n)/M (SD)
Gender (N = 482)
Women	63.3 (176)	36.7 (102)	0.21 a	na	0.649
Men	61.3 (125)	38.7 (79)
Education (N = 481)
Basic education	59.0 (92)	41.0 (64)	3.89 b	na	0.143
Secondary education	60.1 (101)	39.9 (67)
Higher education	68.8 (108)	31.2 (49)
Age (N = 482)
18 to 34 years	65.1 (108)	34.9 (58)	0.96 b	na	0.620
35 to 64 years	60.3 (143)	39.7 (94)
65 years or above	63.3 (50)	36.7 (29)
Marital status (N = 482)
Married/in cohabitation	61.8 (126)	38.2 (78)	1.45 b	na	0.485
Single	60.8 (121)	39.2 (78)
Other	68.4 (54)	31.6 (25)
Nationality (N = 482)
Portuguese	61.6 (274)	38.4 (171)	1.89 a	na	0.169
Other	73.0 (27)	27.0 (10)
Professional status (N = 481)
Student	64.9 (48)	35.1 (26)	2.36 b	na	0.307
Employed	60.1 (191)	39.9 (127)
Other	68.5 (61)	31.5 (28)
Victimisation
Yes	57.2 (111)	42.8 (83)	3.88 a	na	0.049
No	66.1 (189)	33.9 (97)
Criminal and environmental variables
Perception of increased criminality
Yes	50.0 (96)	50.0 (96)
No	70.0 (159)	30.0 (68)	17.54 a	na	<0.001
Total number of criminal occurrences	3.73 (2.53)	4.64 (2.58)	na	−3.80	<0.001
Total number of most feared crimes	3.51 (3.07)	3.76 (2.94)	na	−0.863	0.388
Total number of conditions promoting crime	4.94 (3.00)	5.90 (2.82)	na	−3.47	<0.001
Total number of incivilities	4.20 (1.90)	4.80 (1.74)	na	−3.45	<0.001
Total number of suggested improvement measures	0.91 (0.58)	1.15 (0.48)	na	−4.77	<0.001
Social control variables
Adequacy of policing
Most of the time/always	88.6 (234)	11.4 (30)	177.26 b	na	<0.001
Few times/Never	28.1 (56)	71.9 (143)
Did not know/answer	57.9 (11)	42.1 (8)
Seeking formal support
Most of the time/always	61.6 (90)	38.4 (56)	0.52 b	na	0.771
Few times/never	62.4 (199)	37.6 (120)
Did not know/answer	70.6 (12)	29.4 (5)
Seeking informal support
Most of the time/always	58.5 (69)	41.5 (49)	1.60 b	na	0.450
Few times/never	64.2 (219)	35.8 (122)
Did not know/answer	56.5 (13)	43.5 (10)
Community variables
Years at HCP
Three or less years	64.7 (101)	35.3 (55)
Between four and 9 years	61.6 (53)	38.4 (33)	0.56 b	na	0.755
Ten years or more	61.1 (146)	38.9 (93)
Willingness to collaborate in security measures
Most of the time/always	62.1 (261)	37.9 (159)	0.13 b	na	0.936
Few times/never	64.3 (27)	35.7 (15)
Did not know/answer	65.0 (13)	35.0 (7)
Strength of attachment to HCP
Strong/very strong	63.6 (248)	36.4 (142)	1.14 a	na	0.287
Little strong/weak	57.6 (53)	42.4 (39)

Note. ^a df = 1; ^b df = 2; na = not applicable.

, there were no associations between (dis)satisfaction with the police and sociodemographic variables. There was significant positive association between (dis)satisfaction with the police and victimisation, suggesting that victims seemed to be more dissatisfied than nonvictims were, with 42.8% vs. 33.9%, respectively. Concerning criminal and environmental variables, there were significant positive associations between (dis)satisfaction with the police and perception of security, perception of increased criminality, number of criminal occurrences, number of conditions promoting crime, total number of incivilities and number of suggested improvement measures. Lastly, on social control and community variables, only policing adequacy achieved statistical significance. Those dissatisfied with the police seemed to have poorer evaluations of adequacy of policing than those satisfied, with 71.9% vs. 28.1%, respectively.

3.3. Which Variables Predicted (Dis)Satisfaction with the Police?

The logistic regression revealed a significant model, Χ²(6) = 35.79, p < 0.001, which correctly predicted 63.5% of the cases. According to

Table 3. Logistic regression for (dis)satisfaction with police performance: results of individual predictors.

Variables	B	Wald (1)	p
Victimisation	−0.18	0.69	0.406
Perception of increased criminality	0.63	8.38	0.004
Total number of criminal occurrences	0.04	0.78	0.377
Total number of conditions promoting crime	−0.00	0.00	0.954
Total number of incivilities	0.06	0.77	0.379
Total number of suggested improvement measures	0.68	9.87	0.002

, only the perception of increased criminality and the total number of suggested measures to improve security were significant predictors of (dis)satisfaction with the police. More specifically, those that perceived that criminality was increasing and suggested more measures to improve security were more dissatisfied with the police.

4. Discussion

Policing represents a main social response by the state in order to prevent crime and victimisation, and to promote community security, contributing to sustainable development (United Nations Office on Drugs and Crime 2010). Despite other objective indicators to assess police effectiveness (e.g., arrests rates or decreasing in social disorders), subjective factors (such as community perceptions) are also quite relevant. Although policing is relevant at both rural and urban contexts, urbanisation represents an increasing global trend (United Nations 2019) that poses both gains and pitfalls for individuals and society as a whole. Research about policing perceptions and related variables focusing on urban samples represents a major current issue that is the main topic covered in this study. Briefly, community policing seeks to improve community satisfaction towards the police, enhance a sense that police worry about their problems, develop mutual trust, and and urge individuals to share information regarding criminality and public safety issues (Merenda et al. 2021; Yuksel and Tepe 2013). For these reasons, improving the understanding on this matter is extremely important.

In our study, the majority of participants reported being satisfied or very satisfied with the police. Most participants also described the police as performing in the best way to guarantee safety either frequently or always. These are important and encouraging results if positive perceptions regarding the police and its performance are considered to benefit compliance and community willingness to report crimes (Bolger et al. 2021). These findings seem to be in line of those by Jackson et al. (2011) when analysing Portugal in the context of the European Social Survey. Indeed, Portuguese police has been attempting to adopt a community policing model, implementing cooperative efforts in the construction of security at physical spaces and social interaction (Cruz and Fernandes 2021; Sani et al. 2019).

No sociodemographic variable was associated with (dis)satisfaction with the police. Although being contradictory to the findings of the meta-analysis performed by Bolger et al. (2021), this result is not entirely unexpected, as previous studies had reported the absence of association between satisfaction with police and variables as gender (Dunham and Alpert 1988; Reisig and Parks 2000) and education (Lai and Zhao 2010). Concerning age, the high levels of satisfaction among the total sample might explain the absence of significant association between (dis)satisfaction with the police and age. Additionally, at least in Portugal, the police has developed a broad set of programmes devoted to differently aged target groups, even at school settings, which may contribute to positive perceptions across all age groups.

Focusing on model variables regarding prior contact police, we concluded that (dis)satisfaction with the police presented a significant association with victimisation, suggesting that having a direct or indirect experience of victimisation is normally associated with a lower satisfaction level regarding the police. These findings are congruent with the fifth round of the European Social Survey (Jackson et al. 2011), which reported that victims consider the police less effective than nonvictims do. Indeed, in a phenomenon labelled as secondary victimisation, victims of crime may also be revictimised through contact with the justice system, including the police (Sani and Lopes 2018). In fact, some studies showed that the role of the police was highly dependent on police officers’ beliefs regarding certain criminal phenomena (Sani et al. 2018) and that these perceptions may affect their reactive role to the victimisation. Other studies (e.g., Russell and Light 2006; Sani and Morais 2015) showed that, if the police displays attitudes and responses that meet the victim’s needs and contribute to being empowered, this may promote a greater satisfaction with the institution, contributing to the greater likelihood of asking for police help in the future.

(Dis)satisfaction with the police was equally associated with environmental variables. Perception of security, perception of increased criminality, number of criminal occurrences, number of conditions promoting crime, total number of incivilities, and number of suggested improvement measures were the variables significantly associated. More specifically, those that were dissatisfied tended to feel less safe and that criminality was increasing. Additionally, they reported a higher number of criminal occurrences, conditions promoting time, and incivilities, and suggested more measures to improve security. Some studies reported that the disorder of the physical (LaGrange et al. 1992) and social (Fernandes and Mata 2015) environment produces a feeling of insecurity (Valera-Pertegas and Guàrdia-Olmos 2017), giving the idea that there is no control that can be exercised by the police.

When analysing predictive relationships, only the perception of increased criminality and the total number of suggested measures to improve security were identified as predictive factors of (dis)satisfaction with the police. More specifically, those who perceived increased criminality and those who suggested a high number of measures seemed to be less satisfied with the police. These and criminal occurrences contribute to the loss of trust in the police’s job by reflecting the inability of authorities to control these problems, and guarantee an environment of safety and social order (Lourenço 2010). In Tyler’s (2006) opinion, this can hurt the community perception of legitimacy regarding the intervention of the police and the trust placed in the formal control by this entity.

Lastly, among social control and community variables, policing adequacy solely presented a significant association with (dis)satisfaction regarding police, with people reporting a higher level of dissatisfaction, while also presenting poorer evaluations of police adequacy. This result seems obvious as, if someone is not satisfied with the police, they do not evaluate it as adequate. This perception is largely built on the basis of negative or positive experiences with this entity (Rosenbaum et al. 2005).

Despite the relevance of the obtained and presented results, the present study has limitations that must be mentioned. First, the sample being restricted to people who live, work, or study in the Historic Centre of Porto (this being a convenience sample) may have biased the results, since this is an extremely touristic area, and the experience of the foreign population was not considered. In this sense, the generalisation of the obtained results is extremely difficult because the present study is based solely and exclusively on a specific geographic area, and may have been influenced by selection bias. Additionally, this is an explorative study based on a cross-sectional design; as a result, causal effects cannot be established, and we could detect cohort effects but not developmental trajectories. Lastly, data were collected through self-reports, which can be biased by social desirability; consequently, caution should be taken when generalising our results.

The acceptance and legitimacy of police seem to be directly associated with the trust with which the population faces this entity of formal control (Merenda et al. 2020; Tyler 2006). This constitutes, among other things, an essential element in the definition of security and well-being of populations (Oliveira et al. 2019). In this sense, to promote the perception of legitimacy of police, it is important to improve confidence through the proximity of populations and some groups in particular, such as prior victims and communities that felt that criminality is rising (Farren et al. 2018; Gray et al. 2019).