Abstract
Although policing scholars have discussed possible relationships between discrimination and perceptions of the police, assessing the impact of perceived cultural discrimination on perceptions of the police among immigrants is virtually non-existent and deserving of empirical scrutiny. The current study addresses this gap in the literature by examining the effect of perceived cultural discrimination on perceptions of the police using a sample of 1,272 married immigrant women living in South Korea. Our results show that perceived cultural discrimination is negatively and significantly associated with perceptions of the police, supporting the expressive nature of perceptions of the police among immigrants.
Introduction
Research assessing attitudes toward the police have consistently found age (Choi et al., 2020; Hadar & Snortum, 1975; Murphy & Worrall, 1999; Smith & Hawkins, 1973; Worrall, 1999), gender (Bridenball & Jesilow, 2008; Correia et al., 1996; Miller & Davis, 2008), and race (Carter, 1985; Peck, 2015; Reisig & Parks, 2000; Tyler, 2005; Weitzer & Tuch, 2005) to be significantly related to perceptions of the police. Collectively, this work has shown that younger persons, men, and members of minority groups, hold more negative attitudes toward the police than members of comparison groups—although the relationship between these demographic characteristics seems to be mediated by situational components surrounding the nature and type of police contact (i.e., voluntary or involuntary, police effort, communication, and assistance) (Reisig & Parks, 2000; Skogan, 2005). Recently, policing researchers have extended this line of inquiry to immigration status (e.g., Choi, 2019; Chu & Song, 2015; Correia, 2010; Jung et al., 2019; Röder & Mühlau, 2012) This is an important, understudied area of research, particularly considering the increasing global immigrant population (Castles et al., 2013). The police serve as one of the most visible symbols of government in a country (Choi & Kruis, 2020), and as such, the actions of police can shape immigrant perceptions of a host country’s system of government. It is important for police to establish good relationships with immigrants from diverse backgrounds to obtain cooperation and compliance with this population (Skogan, 2009; Tyler et al., 2010; Weitzer, 2014).
Several recent studies have explored immigrant perceptions of the police (e.g., Bradford & Jackson, 2018; Bradford, Jackson, & Hough et al., 2017; Bradford, Sargeant, Murphy, & Jackson, et al., 2017; Choi, 2019) Although these studies are important in documenting similarities and differences in perceptions of police between immigrants and native-born individuals, it remains unclear what variables predict perceptions of the police within immigrants. Notably, it has been suggested that discrimination experiences at the cultural level, as well as social capital, could be important drivers of perceptions of authority figures and institutions (Van Craen, 2013; Van Craen & Skogan, 2015). Accordingly, using data from a sample of 1,272 married immigrant women in South Korea, the current study seeks to fill these gaps in the literature by examining the impact of social capital and perceived cultural discrimination on immigrant perceptions of the police
Immigration and Perceptions of the Police
Explaining Immigrant Perceptions of Police
Some researchers have argued that scholars should expect immigrants to exhibit more negative perceptions of police than non-immigrants (McCluskey et al., 2008; Skogan et al., 2002). As outlined by Choi (2019), there are three plausible theoretical explanations for why one would expect immigrants to exhibit more negative perceptions of the police than native-born citizens. First, immigrants may project their perceptions of the government in their home country onto governmental entities in their host country. Considering that many first-generation immigrants come from countries with oppressive governmental regimes (Brown et al., 2006), one would expect them to hold a negative attitude toward government officials, such as police officers, based on their past experiences with corrupt government personnel. As such, they may carry this “institutional lens” or framework for understanding government officials from their home country to their host country (Bradford & Jackson, 2018). Second, immigrants face many cultural and social barriers that can lead to miscommunication with police officers. These barriers, accompanied by an intense fear of deportation (Sayad, 2004), can make immigrants skeptical of police interactions, and subsequently, hesitant to engage in dialogue with police officers. Third, immigrants’ experiences relating to prior contact with police officers, such as immigration officials or border patrol officers, could negatively influence their impression of the system of law enforcement in their host country (Bradford & Jackson, 2018; Wu et al., 2017), but only if they perceive such interactions as unsatisfactory.
Others have suggested that it is logical to expect immigrants to hold more positive attitudes toward the police than non-immigrants (see Correia, 2010). Choi (2019) argued that there are also three plausible explanations as to why immigrants would hold more positive views of the police than native-born persons. First, immigrants’ negative perceptions of their oppressive home government or system of law enforcement can set a “low standard expectation” for the police (Choi, 2019, p. 1040). As such, when comparing police in their host country to police in their home country with such an institutional lens (Bradford & Jackson, 2018), immigrants will regard the police in their host country as being more trustworthy and legitimate. Second, immigrants can lack a “real” understanding of the realities of their host country’s system of government. Thus, they have a distorted, romanticized perception of their host countries’ government, and subsequently, its system of justice. These views incline immigrants to place more confidence in police than native-born citizens who have experienced or witnessed the shortcomings of their country’s government, including policing. Third, immigrants hold a host of protective factors, such as strong informal social control and positive ethnic identity (Desmond & Kubrin, 2009; Sampson, 2008), that regulate their behaviors and attitudes toward government institutions, of which, police are seen as one of the most visible symbols (Choi & Kruis, 2020). Relatedly, many immigrants come from collectivist cultures and are motivated to better their life and family name in their new society. Accordingly, they may recognize and hold more positive attitudes toward the police because they symbolize their host society, which is providing them with an opportunity for advancement (Piquero et al., 2016).
Following their review of the literature on immigrant perceptions of the police, Wu et al. (2017) offered a more parsimonious conceptual framework for understanding immigrant perceptions of the police. This framework posited that both (1) universal and (2) group-specific factors influence perceptions of the police, and therefore, can be used to explain differences in perceptions of police between immigrants and native-born persons. Universal factors are those that shape all residents’ attitudes toward the police, whereas group-specific factors are those that only apply to immigrants. Within this conceptual dichotomy exist four subcategories of variables for both factors—(1) demographic (i.e., race/ foreign-born status), (2) experiential (i.e., experiences with police in home and host country), (3) structural (i.e., neighborhood conditions, city and region context, and the structural regime of home country), and (4) attitudinal (i.e., perceptions of injustice, group consciousness, and authoritarian values). Variables within the “Group-specific” factors domain consist of both pre-arrival (i.e., prior to immigration) and post-arrival (i.e., post immigration) variables, suggesting, as outlined by Choi (2019), that experiences in host countries and home countries can interact in shaping immigrant perceptions of police. The following provides a brief overview of the empirical literature in this area.
Prior Assessments of Immigrant Perceptions of Police
The literature on immigrants’ perceptions of the police is limited (Bradford & Jackson, 2018; Jung et al., 2019; Weitzer, 2014) and relatively atheoretical in nature (Wu et al., 2017), and has produced mixed results. For instance, some studies have found that immigrants to report more positive perceptions of the police than non-immigrants (Correia, 2010; Davis & Hendricks, 2007; Rengifo & Fratello, 2015; Wortley & Owusu-Bempah, 2009). Other studies have found immigrants to exhibit more negative views of the police than native-born citizens (Davis & Mateu-Gelabert, 2000; McCluskey et al., 2008; Wu et al., 2011). Still, others have failed to find significant differences in perceptions of the police between immigrants and non-immigrants (Skogan et al., 2002), or shown that other factors condition the relationship between immigration status and perceptions of the police (Bradford & Jackson, 2018; Jung et al., 2019).
Notably, in one of the first tests in this area, Correia (2010) compared perceptions of the police between Latino immigrants (N = 172) and non-immigrants (N = 253) living in Reno, Nevada. Interestingly, he found that the Latino immigrants in his sample indicated statistically significantly more positive perceptions of the police in Reno than did non-immigrants. Specifically, immigrants were significantly more likely than non-immigrants to perceive the local police as treating them with respect, honesty, and equality. Similarly, Bradford, Sargeant, Murphy, & Jackson, et al. (2017) found that immigrants to the United Kingdom reported higher levels of trust in police than native-born citizens, even when controlling for direct experience with police. However, their work highlighted the importance of the length of residency in influencing perceptions of police, with results showing trust levels to be lowest among those who had lived in the United Kingdom the longest. Relatedly, Choi (2019) used data collected from the General Social Survey (N = 1,148) to determine if global and situational support for police use of force varied across first-generation immigrants, second-generation immigrants, and native-born Americans. His results indicated that, compared to native-born Americans, first-generation immigrants indicated less global support for police use of force and less support for police use of reasonable force. However, first-generation immigrants expressed more support for police use of excessive force than both second-generation immigrants and native-born Americans.
Although limited, some research has also shown locational differences in immigrant perceptions of police. For example, Chu and Song (2015) compared perceptions of police among Chinese immigrants living in New York City (N = 151) and Toronto (N = 267). They found that, compared to Chinese immigrants living in Toronto, Chinese immigrants living in New York indicated less respect for the police. The Chinese immigrants living in New York also felt that police “slow response” to calls and prejudice against Chinese persons were more serious problems than did the Chinese immigrants living in Toronto. No significant group differences were found between the two groups in terms of perceptions of police being able to effectively deal with crime or providing adequate protection against crime in the cities where respondents lived. Similarly, Bradford and Jackson (2018) analyzed data from 27 European countries in the European Social Survey (ESS) and found immigrants to report differences in perceptions of two measures of police legitimacy (i.e., normative alignment with police and perceived duty to obey police). Notably, they found a partial positive association between immigrant status and perceptions of police legitimacy—with those immigrants who had recently immigrated (i.e., less than 15 years ago) as adults indicating greater perceptions of police legitimacy than reference groups (i.e., native citizens and immigrants who immigrated as children). However, upon including measures of police contact, group position (i.e., social inclusion), and changes in institutional frames, the relationship between immigration status and perceptions of police legitimacy was insignificant, suggesting that the relationship between immigration status and attitudes toward police is complex. Recent work by Jung et al. (2019) extended this line of research by showing that type of government of the country of origin and recency of immigration interact in predicting perceptions of police officers. Collectively, this work provides partial support for Wu et al.’s (2017) theoretical framework by showing that a host of situational characteristics and contextual factors, pre-and post-arrival, influence the relationship between immigration status and perceptions of police. Accordingly, tests within this area need to control for such variables when exploring this relationship.
The Effect of Perceived Discrimination on Perceptions of the Police
One variable that has been relatively ignored in this area of research is the effect of perceived cultural discrimination on perceptions of police. Technically, using Wu et al. (2017) conceptual framework, perceived discrimination, broadly speaking, could be considered a sense of injustice falling under the umbrella of “attitudinal” universal Factors. We argue, however, that discrimination at the cultural level (i.e., cultural discrimination), is a sense of injustice experienced more commonly by non-native citizens, who come from diverse cultural/ethnic backgrounds with unique customs, than by native-born citizens. As such, we suggest that perceived cultural discrimination is a “group-specific factor” because it relates to specific groups of persons. Therefore, we treat it as such in Wu et al. (2017) conceptual framework for the remainder of this paper.
Research has found that members of ethnic minority groups, particularly African Americans, hold more negative perceptions of the police than members of other ethnic groups (Frank et al., 2005; Peck, 2015; Tyler, 2005; Weitzer & Tuch, 1999, 2005). Scholars have argued that these differences in attitudes are a product of systemic discrimination experienced at the hands of actors within the criminal justice system, such as disproportionate police violence (Tyler & Huo, 2002), excessive use of force (Fryer, 2019; Weitzer, 2002), and racial profiling (Reitzel & Piquero, 2006; Tyler & Wakslak, 2004; Weitzer & Tuch, 2002). However, surprisingly, very little empirical work has been done to assess the effects of perceived cultural discrimination on perceptions of the police, and almost no work has examined the impact of perceived cultural discrimination on immigrant perceptions of the police in a non-Anglo-American context (cf. Röder & Mühlau, 2012). In the lone exception known to the authors, Röder and Mühlau (2012) used secondary data from the European Social Survey to examine immigrant perceptions of police across geographic boundaries. Those authors found that immigrants who felt that they belonged to an ethnic minority group, or were a member of a “discriminated group,” tended to exhibit less trust in the police. Additionally, they employed a country-level proxy variable, the scope of the anti-discrimination policy in a host country, to consider a different feature of discrimination against immigrants. Röder and Mühlau’s study indicated that immigrants residing in countries with “good” anti-discrimination policies were more likely to report higher levels of trust in the police than countries with “poor” anti-discrimination policies—thus, providing evidence of an empirical relationship between perceived discrimination and immigrant perceptions of police.
Different models propose different directions for the effects of perceived cultural discrimination. From an expressive model of public confidence, citizens perceptions of police legitimacy are not driven merely by perceptions of crime control and order maintenance, but rather, by feelings of social cohesion, collective efficacy, and moral consensus produced and enforced by government institutions (Choi & Kruis, 2020; Jackson & Bradford, 2009; Jackson & Sunshine, 2007). Within this “neo-Durkheiman model,” the public are as concerned with officers treating citizens with fairness and dignity when enforcing social norms as they are with the actual enforcement of norm control (Jackson & Bradford, 2009, p. 496). Therefore, according to this perspective, perceived cultural discrimination will be associated with more negative perceptions of the police. Conversely, System Justification Theory assumes that perceptions of authorities are enhanced whenever a group of subordinates is dependent on them, independent of perceptions of fairness in procedures and favorable outcomes (Jost et al., 2004; van der Toorn et al., 2011). Because the police represent an official authority of a host country’s system of government, which many immigrants depend on to handle unfair treatments by other private citizens in a host country, this perspective assumes that cultural discrimination will promote positive perceptions of the police among immigrants. In short, immigrants may be motivated to perceive the police more positively to justify the official system that they need to rely on when they encounter discrimination by other private citizens. However, given the limited research in this area, the relationship between these variables remains unclear.
Current Focus
Although scholars have speculated a possible relationship between cultural discrimination and perceptions of the police, assessing the impact of perceived discrimination at the cultural-level on perceptions of the police among immigrants is limited, and, to the best of the authors’ knowledge, no research has examined this relationship in the context of Asia. This investigation is important not only because immigrants are often subject to discrimination (Berry, 1997; Ra et al., 2019) but also because very little is known about perceptions of the police among immigrants who reside in non-Anglo-American settings. As such, the current study seeks to help fill this gap in the literature by examining the effect of perceived cultural discrimination of immigrant women living in South Korea on their perceptions of the police. We anticipate that perceived discrimination will be inversely related to perceptions of the police.
Methods
Data
Data for the current study were obtained from a self-report survey of married immigrant women living in South Korea. Due to rising immigrant populations (Ra et al., 2019), South Korea provides a good context to explore immigrant perceptions. According to the Korea Immigration Service Statistics (2019), the number of registered foreign residents living in South Korea is 1,257,366. Many of these immigrants are married immigrant women. In fact, the number of married immigrants is 162,582, and more than 80% of these immigrants are women (Korea Immigration Service Statistics, 2019). Immigration patterns have disrupted the traditional Korean monoculture and challenged Korean nationalists’ push for Danil Minjok—or a one “one blood” ethnicity among South Koreans (Kim, 2010; Moon, 2010; Watson et al., 2011). Still, due to a rich monocultural history advocating intolerance of foreigners, married immigrant women living in South Korea face many difficulties in adjusting to Korean Culture (Kim, 2010; Lee & Park, 2018). Given the unique history of South Korea and the recent cultural changes, this country provides a unique opportunity to explore the relationship between perceived cultural discrimination and immigrant perceptions of the police.
Unfortunately, due to the difficulties related to obtaining a sample of immigrants, we had to rely on non-probability, purposive sampling methods. Specifically, in 2013, the Korean National Police Agency was operating nationwide free driving schools for all married immigrant women to help them obtain their driver’s license. Police officers at different stations served as the driving school instructors, and each driving course typically lasts 14 days. This community policing program was developed in part to help establish positive relationships between the police and married immigrant women. Accordingly, officers were instructed to try to ensure quality interactions with students of the driving course.1 All of the participants in this program were married immigrant women who had resided in South Korea for a relatively short period of time (≤7 years). One of the authors contacted representatives of each of the 16 provincial sub-divisions of the National Police Agency to distribute the survey to married immigrant women who registered in the driving schools. All police representatives agreed to assist in survey administration. One of the authors created a five-page questionnaire used in this study to tap several domains of adjustment of married immigrant women living in South Korea, such as fear of crime, discrimination, and victimization. In developing this survey, we partially drew on the original questionnaire used by several researchers who investigated the unique challenges of different types of immigrants in South Korea (Jang & Lee, 2006; Yun, 2009). Trained researchers were dispatched to driving schools and administered the survey. Paper-and-pencil surveys were distributed to married immigrant women in the classroom of each driving school during September of 2013. Participants were informed that their participation in this study was both voluntary and anonymous. Even though the survey was presented as voluntary, the participants of this study may have felt pressured to participate. That is, it is possible that many thought that their participation could serve as compensation for the free driving instruction provided by the officers. This situation could explain why there was such a high response rate. Professional translators were present in cases where participants needed a translation. The questionnaire was offered to a sample of 1,284 respondents, and a total of 1,272 married immigrant women completed the survey (response rate = 99.06%). Although our sample is not a random selection of all married immigrant women living in South Korea, our sample provides valuable data into this hard-to-reach population.
Measures
Dependent variable
The dependent measure in this study was perceptions of the police. Perceptions of the police were measured using a five-item, multidimensional scale, designed to capture not only general perceptions of the police, but also foreigner specific attitudes toward the police. Our scale was influenced by measures used in previous research on perceptions of the police (Bradford et al., 2009; Frank et al., 1996). Specifically, participants were asked to rate their agreement with the following statements: (1) “The local police in my neighborhood are quickly dispatched to the crime scene if a crime takes place,” (2) “The local police in my neighborhood are good at patrol activity,” (3) “The local police in my neighborhood are working hard to ensure foreigners’ safety,” (4) “The police are critical in preventing foreigners’ criminal victimization,” and (5) “The police are working hard to prevent criminal victimization (e.g., school bullying) among foreigners’ children.” All items were rated on a five-point Likert scale ranging from 1 (completely disagree) to 5 (completely agree). Exploratory Factor Analysis (EFA) indicated that the items loaded well onto one latent factor, and assessment of Cronbach’s alpha (α = .894) suggested good internal consistency.
Independent variable
The key independent variable in this study was perceived cultural discrimination. Perceived cultural discrimination was measured using a six-item scale derived from Yun’s (2009) study that centered on married immigrant women’s psychological adjustment to South Korean society. Specifically, respondents were asked to answer the degree to which they agree or disagree with the following statements: (1) “I have been discriminated against by my neighbors and Koreans for economic reasons,” (2) “I have been looked down upon by my neighbors and Koreans because of my level of Korean language,” (3) “I have been looked down upon by my neighbors and Koreans just because I am a foreigner,” (4) “Koreans do not understand foreigners’ cultures,” (5) “My neighbors and Koreans have prejudice towards foreigners,” and (6) “Koreans do not regard me as Korean.” All six items for this scale loaded onto one latent factor that was shown to have good reliability (α = .864).
Control variables
We included the following controls in our statistical model: (1) age, (2) educational attainment, (3) family monthly income, (4) city residence, (5) direct victimization, (6) indirect victimization, (7) fear of crime, (8) length of residence in South Korea, (9) natural citizenship, (10) social capital, and (11) crime-related media consumption. All of these controls have been linked to perceptions of the police in the literature (Brown & Benedict, 2002; Choi, 2019; Choi & Kruis, 2020; Chu & Song, 2015; Sun & Wu, 2018; Van Craen, 2013; Van Craen & Skogan, 2015) Demographic controls included age (i.e., between 15 and 20 years, between 21 and 25 years, between 26 and 30 years, between 31 and 35 years, between 36 and 40 years, between 41 and 45 years, between 46 and 50 years, between 51 and 55 years, and equal to or more than 56 years), educational attainment (i.e., less than or equal to college degree and college degree or more), and monthly family income (i.e., less than or equal to $2,000, $2,000 –$3,000, and $3,000 or more). City residence was also included as a dichotomous control variable (1 = city resident, 0 = non-city resident).
To measure direct victimization, we asked respondents if they had been the victim of any of the following eight categories of crime since they came to South Korea: (1) home invasion, (2) home burglary, (3) theft, (4) robbery, (5) assault by a stranger, (6) assault by someone they knew, (7) sexual assault by a stranger, and (8) sexual assault by someone they knew. Respondents who had experienced any victimization were coded as 1, whereas those who had not were coded as 0. Indirect victimization was measured by asking respondents if they knew a person close to them who had experienced any of the eight categories of crime in the previous year. Those who indicated that they knew someone who had experienced victimization were coded as 1, and those who did not were coded as 0.
Fear of crime was measured using an eight-item scale adapted from Ferraro (1995). Respondents were asked to indicate their levels of fear to the following statements: (1) “I am afraid someone will break into my house,” (2) “I am afraid of home burglary,” (3) “I am afraid of having my money stolen,” (4) “I am afraid of getting robbed,” (5) “I am afraid of being assaulted by a stranger,” (6) “I am afraid of being assaulted by someone I know,” (7) “I am afraid of being sexually assaulted by a stranger,” and (8) “I am afraid of being sexually assaulted by someone I know.” The responses followed a five-point Likert scale ranging from 1 (completely disagree) to 5 (completely agree). The fear of crime scale was constructed by averaging across the items so that higher scores were indicative of higher levels of fear. The computed alpha score exceeded an acceptable level of internal consistency (α = .729), and the items for this scale loaded onto one latent factor in EFA.
Several immigrant-specific variables were also controlled for in our multivariate analysis, including length of residence, citizenship, and social capital. Length of residence in South Korea was measured by the number of years that respondents had stayed in South Korea at the time of the survey. South Korean citizenship was measured using one question asking whether respondents had obtained South Korean citizenship. Social capital was measured using a six-item scale derived from the work of Yun (2009). Participants were asked to indicate the degree to which they agree or disagree with the following statements: (1) “I have good relationships with my neighbors,” (2) “I often talk about the country that I am from with my neighbors,” (3) “I often participate in the events held in my neighborhood,” (4) “My neighbors help me whenever I am going through a hard time,” (5) “I get along with Koreans other than my neighbors,” and (6) “I often participate in social programs helping foreigners.” The magnitude of the reliability for the social capital scale was above the acceptable range (α = .863) and all six items loaded onto one latent factor in EFA.
Table 1 presents the descriptive statistics for the variables used in the current study.
Descriptive Statistics (N = 1,326).
Analytic Strategy
Multiple Ordinary Least Squares (OLS) regression was used as the primary multivariate analysis in this study. All assumptions of OLS regression were checked and ensured prior to constructing the final model reported below. Notably, all variables inputted into the regression model had tolerances above .1 and Variance Inflation Scores (VIFs) below 2. Further, analysis of normal Quantile-Quantile and Probability-Probability plots suggested the presence of a normal distribution, and the assessment of skewness and kurtosis values for the dependent measure fell in the accepted conventional range.
Results
As a first step in the analysis, we assessed the bivariate correlations between perceptions of the police and variables of interest. Table 2 presents the zero-order correlation coefficients. Surprisingly, the key independent variable and perceptions of the police were not significantly correlated to each other. However, we speculate that the relationship between them may be obscured because other variables were not held constant. City residence was positively and significantly associated with perceptions of the police (r = .073, p < .05), while fear of crime was negatively and significantly related to perceptions of the police (r = −.109, p < .05). Further, both the length of residence (r = −.093, p < .05), and South Korean citizenship (r = −.113, p < .05), were negatively associated with perceptions of the police. Social capital emerged as the strongest correlate of perceptions of the police (r = .262, p < .05), showing that married immigrant women with more social capital tended to display more positive perceptions of the police compared to their counterparts with less social capital.
Zero-Order Correlation Matrix among Variables.
Notes. (1) = perceptions of the police, (2) perceived discrimination, (3) age, (4) education, (5) income, (6) residence (1 = city residence), (7) direct victimization (1 = yes), (8) vicarious victimization (1 = yes), (9) fear of crime, (10) length of residence, (11) South Korean citizenship, (12) social capital, and (13) media consumption. Household income was coded as 1 = less than or equal to $2,000, 2 = $2,000–$3,000, and 3 = $3,000 or more. Pairwise deletion of missing values was used.
p≤ .05 (two-tailed tests).
Table 3 presents the multivariate results from our OLS regression analysis. Overall, the model fit the data well and explained about 8 percent of the variance in perceptions of the police (Adjusted R2= .079). The results of the regression analysis indicated that perceived cultural discrimination was negatively and significantly associated with perceptions of the police (b = −.073, p < .01), net of other variables. Fear of crime (b = −.081, p < .001) was also statistically significant, with results indicating that those with higher levels of fear tended to hold more negative views of the police compared to their peers. Additionally, social capital was a statistically significant, positive predictor of perceptions of the police (b = .229, p < .001), and remained the strongest predictor even after other variables were entered in the multivariate model.
OLS Model Predicting Perceptions of the Police.
Note. N = 1,326; OLS = Ordinary Least Squares; SE = standard error. Pairwise deletion of missing variables was used.
p ≤ .05, ** p ≤ .01, *** p ≤ .001.
Discussion
Although still in a nascent stage of development, the topic of immigrants’ perceptions of the police has started to acquire the attention of policing scholars (Choi, 2019; Chu & Song, 2015; Sun & Wu, 2018). This line of inquiry has examined both universal and immigrant-specific factors to understand immigrants’ perceptions of the police (Wu et al., 2017). Collectively, these studies are important for documenting the covariates of perceptions of the police among immigrants. Still, though, empirical research is very limited regarding one key factor that is particularly relevant to immigrants, perceived cultural discrimination (cf. Röder & Mühlau, 2012). Additionally, previous studies examining immigrants’ perceptions of the police have been based primarily on data within Anglo-American contexts (Chu & Song, 2015; Correia, 2010; Menjívar & Bejarano, 2004; Wu et al., 2011). In an effort to help fill these gaps in the literature, this article examined whether perceived cultural discrimination influences immigrants’ perceptions of the police. Specifically, drawing on the literature suggesting that perceived discrimination is an important contributing factor to minority group members’ perceptions of the police (Van Craen, 2013; Van Craen & Skogan, 2015), we used data from a large sample of married immigrant women living in South Korea to explore the role perceived cultural discrimination plays in shaping perceptions of the police among immigrants. Three key findings from this study emerged that warrant further discussion.
The first key finding concerns the relationship between perceived cultural discrimination and perceptions of the police. To explore this relationship, we measured perceptions of cultural discrimination experienced by married immigrant women in different realms of their lives. It is important to note that immigrants can experience unfair and ostracizing treatment merely because of cultural differences, such as customs, dress, accents, and language barriers (Berry, 1997; Lippi-Green, 1997; Yoo et al., 2009). Accordingly, our measure of perceived discrimination attempted to capture discrimination at the cultural level. Our multivariate analysis showed that perceived cultural discrimination was negatively and significantly associated with perceptions of the police. This finding is consistent with findings from previous studies showing that discrimination can erode positive perceptions of the police among minority group members (Van Craen, 2013; Van Craen & Skogan, 2015). To date, only one other study has examined the impact of discrimination on perceptions of police using an immigrant sample (see Röder & Mühlau, 2012). Röder & Mühlau’s study indicated that immigrants residing in countries with good anti-discrimination policies were more likely to report higher levels of trust in the police. One important difference between the current study and that of Röder and Mühlau is that Röder & Mühlau’s measures of discrimination did not consider immigrants’ direct experience of prejudice or ignorance in host countries. As such, the current study expanded their work by using a measure of discrimination based on immigrants’ direct discriminating experience, as perceived by them. Despite the methodological differences, our finding aligns with Röder and Mühlau’s research revealing that feelings of discrimination can negatively influence perceptions of the police.
Another important finding of this study involves the relationship between social capital and perceptions of the police among married immigrant women. According to social capital theory, the quality of one’s social capital is associated with public confidence in governmental institutions in democratic societies, such as confidence and trust in the police (Putnam, 2000; Van Craen, 2013). Our measure of social capital captured various aspects of social organizations among married immigrants, including social networks and cohesion. OLS regression revealed that the relationship between social capital and perceptions of the police was positive and significant, supporting the notion that social capital can facilitate not only interpersonal relationships but also confidence in public institutions such as the police (Putnam, 1993). The positive relationship between social capital and perceptions of the police observed in this study is also in line with the previous research in this area (Sun et al., 2013; Van Craen, 2013). The simple, but difficult to implement, policy implication based on these findings is that to help improve perceptions of the police, policymakers should attempt to improve the social capital of marginalized groups.
Lastly, our study revealed that fear of crime is significantly and inversely related to perceptions of the police among married immigrant women. We suggest that this finding supports tenets of the instrumental model of policing. The instrumental model of policing suggests that individuals with higher levels of fear of crime would hold a more negative view of the police because they perceive the police as being responsible for failing to effectively control crime (Choi & Kruis, 2020; Jackson & Sunshine, 2007). In short, heightened levels of fear of crime can question the effectiveness of police performance, resulting in lower levels of confidence in the police. Our finding was consistent with this theoretical expectation and previous research showing that instrumental concerns, such as fear of crime, can significantly impact perceptions of the police (Jackson et al., 2009; Sun et al., 2014).
Taken together, the pattern of our findings suggests that although universal factors, such as fear of crime, can be important in understanding perceptions of the police, it is critical to consider immigrant-specific variables. Notably, our results highlight that perceived discrimination by immigrants can be reflected in perceptions of the police even when the police were not the subject(s) who mistreated immigrants. The expressive model of policing posits that the police represent the defenders of community morals in the eyes of the public (Jackson & Bradford, 2009; Sunshine & Tyler, 2003). Accordingly, using this model, immigrants may view the police as prototypical representatives of the society of settlement. Thus, when immigrants experience hostility and rejection by their host society, they may project this perception of discriminatory treatment onto the hands of police officers. That is, they may think that members of the police force are no different from the members of a broader society that the institution of policing is commissioned to defend and uphold. As such, the current results highlight the importance of considering the role of a host society in shaping perceptions of the police among immigrants.
The ability of the police to work effectively is dependent on public perceptions of the police (Tyler & Huo, 2002). This is true regarding all members of the public, including immigrants (Skogan, 2009). Therefore, it is critical for policymakers to incorporate attempts not only to improve the quality of policing but also to increase social support for cultural pluralism (Berry, 1997; Sam & Berry, 2010). We suggest that this finding is mirrored in our measure of social capital. As noted, our results show that immigrants hold the police in higher regard when they have more social ties, trust, and cohesion with other Korean members. By definition, this is what cultural pluralism is supposed to represent—smaller groups of persons with different cultures assimilating into a larger group that is accepting of individual, cultural differences. It is difficult for immigrants to have great social capital in a society that does not advocate cultural pluralism. Accordingly, this pattern of findings suggests that we need to take steps to expand social networks of immigrants, facilitating positive experiences with Koreans. This approach not only corresponds to the effective integration strategy but also represents a successful policing strategy to build “a reservoir of goodwill” from immigrants in a host society (Berry, 1997; Dennis, 1976). At the community level, an example of a policy would be the creation of social programs that help foreigners and native-born citizens interact with one another in prosocial settings. At the societal level, such a policy would involve anti-stigma campaigns directed at alleviating misperceptions of foreigners and encouraging cross-cultural friendships—although, admittedly, such a policy would be difficult to implement. We suggest that social media platforms would provide a good starting point for such campaigns. Increasing social capital while decreasing cultural discrimination could increase immigrants positive sentiment toward the police, which in turn, can promote coordination and cooperation with the police, producing a mutually beneficial relationship for all (Putnam, 1993, 2000; Tyler & Huo, 2002).
Limitations
There are several limitations to the current study that should be noted. First, although a strength of the study was its ability to explore the relationship between perceived cultural discrimination and perceptions of the police outside of the Anglo American context, our participants are an immigrant sample that reflects the unique setting and circumstances in South Korea. As such, results are not generalizable beyond this region. Due to the limited research conducted in this area, and lack of generalizability of our findings, we encourage replications of our study to other regions, including the US, where immigrants account for a large proportion of the population (Weitzer, 2014). Second, the definitional issues of some measures used in this article should be kept in mind. For instance, while we used the perceived discrimination scale derived from the work of Yun (2009) that concerns cultural level discrimination, there is no single definition or operational measure of discrimination, especially with respect to immigrants (Röder & Mühlau, 2012; Schmitt et al., 2014; Van Craen & Skogan, 2015). We encourage replication of the current work with other measures of discrimination. Third, this work did not examine many of the pertinent factors associated with perceptions of the police, which could confound the effect of perceived discrimination on perceptions of the police. For example, we did not have any questions to address prior contact with the police in our survey. Prior contact with the police can significantly shape one’s perceptions of the police, whether it is with immigration authorities (e.g., Wu et al., 2011), with police in one’s home country (e.g., Davis et al., 1998), or with the police in South Korea (e.g., Chu et al., 2005). Our failure to include these variables preclude firmer conclusions regarding the relationship between perceived cultural discrimination and perceptions of the police. More empirical investigations are needed to examine the interrelationships between prior contact with the police, perceived discrimination, and perceptions of the police. Relatedly, immigrants’ perceptions of government institutions discriminating against them may be an important source of their perceptions of the police (Tyler & Huo, 2002). Although our study is focused on immigrants’ discrimination experienced by society, we call for future research to examine other factors (e.g., distributive justice or procedural justice) that could influence immigrant perceptions of the police. We suggest that it may be best to approach this topic from a qualitative lens to allow these predictors to emerge from a more naturalistic approach.
Conclusion
The number of immigrants has been rapidly growing worldwide (Castles et al., 2013). However, policing research on this population has yet to identify factors specific to understanding immigrants’ perceptions of the police (Weitzer, 2014). The current study suggests that perceived cultural discrimination and social capital are critical to understanding immigrants’ perceptions of the police. Continued research in this area is of great importance to help inform policing strategies regarding immigrants that are supported with empirical data.
Footnotes
Authors’ Note
1. It is possible that the positive nature of this interaction shaped married immigrant women’s perceptions of the police more favorably. However, this possibility does not preclude us from investigating relationships between variables of interest in the study.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
