Abstract
Dominican immigrants are increasingly turning away from traditional metropolitan gateways to settle in relatively small and medium-size cities in the Northeast. This study examines their views about neighborhood social disorder and cohesiveness in Reading, Pennsylvania. The results indicate that residents are divided about the pervasiveness of disorder-related problems in their neighborhoods. Moreover, views about social disorder have implications for social cohesiveness, but neither of these dimensions of urban life can be understood apart from immigrant incorporation. Among those who live in areas without disorder, naturalized citizens are especially likely to feel that they live in a tight-knit neighborhood and to interact with neighbors. The study concludes with an examination of perceptions of neighborhood safety.
Immigrants and cities are closely linked in historical narratives of America’s ascendance as an industrial power. Although immigrant lives still largely unfold in urban areas, a shift in perspective is necessary. The current volume of immigration rivals previous historical highs, but source countries are increasingly diverse and immigrant destinations now extend beyond traditional metropolitan gateways (U.S. Department of Homeland Security 2010; Singer, 2004). These changes are potentially significant if immigrants fare better in new destinations (Alba et al. 2009; see Lichter et al. 2010).
Consistent with these trends, the number of immigrants from the Dominican Republic has grown from 12,000 to 763, 000 since 1960. Dominicans are the fourth largest Hispanic group, among the ten largest immigrant groups, and among the five largest groups of legal permanent residents (Grieco, 2010; Monger, 2010). They are also expanding beyond their traditional port-of-entry, with the percentage living in New York falling from 77% in 1980 to 54% in 2005.1
The literature on immigrants in new destinations is burgeoning. However, few studies of new destinations examine the neighborhood circumstances of Dominican immigrants. This is a non-trivial matter because they are an economically-disadvantaged (Reimers, 2006) and highly segregated group, regardless of time spent in the U.S. (Iceland, 2009). Indeed, Dominicans in New York are segregated in lower quality neighborhoods that have deteriorated physically. Such neighborhoods are often associated in the public’s mind with isolation, a fraying social fabric, and a tangled web of “pathologies.” Thus, it is important to determine whether Dominicans view their neighborhoods in new destinations similarly.
Using a 2004 survey, this study examines their experiences in Reading, Pennsylvania. The first objective is to describe Dominicans’ views about the prevalence of neighborhood social disorder. The second objective is to demonstrate how these views are associated with psychological and behavioral dimensions of neighborhood cohesiveness. The third objective is to investigate how experiences with disorder and cohesiveness are tied to immigrant incorporation. Last, the implications of these experiences for perceptions of personal safety are explored.
DISORDER AND SOCIAL COHESION
Social disorder refers to the absence of rules and mechanisms for the protection of public safety. Social disorder is reflected in criminal activities, as well as a lack of civility and comportment. Signs of neighborhood disorder include “people hanging out on the streets, drinking, taking drugs, panhandling, and creating a sense of danger” (Ross and Jang, 2000, p. 402). A related concept is social disorganization; that is, the lack of an infrastructure for social control. Disorganization is evident where there is little local organizational participation, few social ties and people feel powerless to address problems collectively (Sampson and Raudenbush, 1999; Taylor, 2001).
Social disorder has an illustrious provenance that dates from the Chicago School at the end of the Progressive Era when concerns about immigrants and urban problems were inextricably linked. Handlin (1959, p. 36) pointed out a half century ago that signs of social disorder were “inescapable” during the era and complaints about “pauperism, crime, juvenile delinquency, disease, and other evils of the slum… often associated with immigration” were rampant. Indeed, immigrant settlements were viewed as being rife with social ills. Zorbaugh’s vivid description of Sicilians in Chicago’s Little Hell (1965, pp. 155-157) claimed “the foreign slum, is gangland” with “its lack of group life, common social definitions” and contempt for law enforcement. Such neighborhoods were plagued by a general absence of trust, the isolation of neighbor from neighbor, and predatory behavior that compromised public safety.
Apparently, these conditions could co-exist alongside family and village-like social relations. Wirth (1998, p. 222) saw ghettos as “close knit” worlds. Whyte (1993) later emphasized the high level of social organization in Boston’s North End, a community formed by socially connected Italians with mutual obligations despite appearing to outsiders as a “mass of confusion, a social chaos” (see also Gans 1982). Apparently, social order may prevail in inner cities because ethnic-based territorial divisions enhance “circles of trust” among residents (Suttles 1968).
Contemporary studies which serve as benchmarks pay less attention to immigrants, despite their greater likelihood of living in areas of concentrated poverty (Jargowsky, 2009), their segregation from native Whites (Iceland, 2009), and their difficulty moving from high-poverty to low poverty neighborhoods (South et al. 2005a). However, various collaborative efforts in the social disorder literature provide the foundation for additional inquiry. Consistent with classic accounts, Ross et al. (2001) claim that the disadvantaged operate in a “survival mode” characterized by exposure to exploitation through force and deception. This undermines trust in others, but social ties may lessen the brunt of neighborhood problems (Ross and Jang, 2000). Interaction with neighbors strengthens bonds that neutralize the alienating effects of disorder.
Sampson and colleagues draw attention to disorder and collective efficacy, a term encapsulating expectations about the ability to mobilize residents for informal social control (Sampson et al., 1999, 2002). Social disadvantage reduces efficacy through the dissolution of social bonds (Sampson and Raudenbush, 2004; Franzini et al., 2008). Moreover, disadvantaged residents may not act collectively because of powerlessness that is rooted in cynical attitudes about the responsiveness of the criminal justice system to their plight (Sampson, 2002; Sampson and Bartusch, 1998).
A key issue is the extent to which perceptions of local conditions mirror reality. Sampson and Raudenbush (2004) show that perceptions of disorder are strongly associated with objective levels of disorder that are revealed by systematic observations (see also McCord et al., 2007). Views are also linked to interests in the exchange and use values of neighborhoods (e.g., Garcia et al., 2007; Guest et al., 2006; Hipp 2010; Ross et al., 1999, 2001; Sampson and Raudenbush 2004). Homeownership promotes neighborhood social ties, fosters vigilance and increases sensitivity to signs of disorder because home values reflect local conditions. Similarly, neighborhood use value is linked to lifecycle characteristics such as age, marriage, and parenthood. Interests associated with parenthood, for example, encourage neighborhood monitoring and integration into local social networks to protect children’s safety.
Few studies provide insights into the role of immigrant incorporation.2 This is surprising given that English proficiency and citizenship affect the types of areas that immigrants live in. South et al. (2005a, 2005b, 2005c) show that the spatial assimilation of Hispanic immigrants is generally consistent with classic assimilation models. English proficiency, citizenship, and years in the U.S. promote mobility to non-Hispanic White neighborhoods. In so doing, incorporation may expand opportunities to live in areas where problems with social disorder are minimal. At the same time, opportunities differ among Hispanic groups. The mobility of Afro-Caribbean Spanish speakers to “better” areas is relatively limited (South et al. 2005b; also see Alba et al., 2009). Such findings demonstrate the need for inquiries into neighborhood experiences in new destinations.
Although perceptions may reflect social differences in exposure to deleterious circumstances, shifts in expectations and interpretive frameworks must also be recognized. Specifically, immigrants may become more “street smart” and sensitive to signs of neighborhood danger with the accumulation of experiences over time. Heightened sensitivity may also result from stereotypes learned in adjusting to the U.S. (Portes and Bach, 1985). Sampson and Raudenbush (2004) offer this to explain why Latinos in neighborhoods with high concentrations of African Americans disproportionately mention disorder-related problems, irrespective of observed disorder.
Dominican Neighborhoods
Insights into the neighborhood experiences of Dominicans are primarily available from studies conducted in New York. On the one hand, Dominicans in Washington Heights are portrayed as having a high degree of formal organization with a business community, public celebrations, community clubs, sports clubs, and newspapers (Torres-Saillant and Hernández, 1998). Duany (1994) points to a high degree of informal organization on a “representative block” where “major networks of social interaction” and “self-enclosed little towns” form inside buildings. Dense webs of social ties exist in an institutionally complete community.
On the other hand, New York Dominicans are described as “ghettoized” into areas where social ills and complaints about unfair treatment by the police and government agencies are commonplace (Duany, 1994; Pessar, 1995). Suro (1998, p. 188) suggests “Washington Heights became synonymous with cocaine…and the NYPD became the chief representative of American society in the neighborhood, and conflict with the police came to signify the way the whole community related to the outside world.” Kasinetz et al. (2008, p. 119) concur in claiming that Dominican immigrants throughout the city feel “‘streets’ are to be feared and avoided. Parents talk about ‘losing’ children ‘to the streets’.” Criminal activities “shaped the friendships people would make, constrained the public facilities people would use, restricted whether residents could safely go out at night, and even determine what route young people could take to and from school” (p. 122).
Similar themes echo throughout Levitt‘s (2001) study of Boston (Jamaica Plain). Some Dominicans are socially isolated, but others recreate life on the island, leave doors unlocked, and trust neighbors. Still, migrants display “mistrustful solidarity;” social circles consist of family and fictive kin to minimize the risk of victimization by co-ethnics who are allegedly “against the law.” Conceptions of “right and wrong” are flexible with an “open to interpretation stance toward morality.” (Levitt 2001, p. 115). Suro (1998, p. 190) asserts that this “inherent tolerance for misbehavior” fostered social problems in New York.
Dominicans in a New Destination: Reading, PA
These descriptions allude to some of the reasons why places like Reading, Pennsylvania are emerging as destinations in a Dominican migration stream that is increasingly oriented away from New York (Figure 1). Reading had the second highest growth rate (127%) of Dominican immigrants during the past decade among cities with at least 2,500 persons born on the island. It was eclipsed only by neighboring Allentown-Bethlehem (183%). Sixty-five miles to the south, Philadelphia also had a high growth rate (58%, ranked 7th). Such figures are impressive given that Pennsylvania includes numerous municipalities with “anti-immigrant” ordinances and has only recently emerged as a “new destination” state (Massey and Capoferro, 2008).
Figure 1.
The Expansion of New Settlements for U.S. Residents Born in the Dominican Republic: 2000 and 1980
Reading is a city that faces numerous economic challenges, along with a growing economically-disadvantaged Hispanic population, diversity-related tensions and crime. It is the fifth largest city in Pennsylvania with a population of 81,000. Situated on the Schuykill River, its growth historically was tied to industrial production and its development as a transportation hub for moving coal by rail to eastern cities before demand for coal, railroads, and manufactured products declined. The economy can be described as lackluster (Brookings Institution, 2003) with remnants of the past evident in “big industrial carcasses…with faded white painted letters giving the names of things, a catalogue of the order of the industrial universe when this country hammered, welded, and created form out of the void” (Montgomery, 2008).
Similar to other places with a growing Hispanic presence, Reading is a “majority minority” city.3 Hispanics are half, Whites one-third and African Americans one-tenth of the population. Puerto Ricans form the majority among Hispanics, followed by Mexicans and Dominicans. Reading also has a high poverty rate (33% vs. 12% for Pennsylvania) and a large population without a high school degree (35% vs. 13% for Pennsylvania). Only moderately segregated in the city, Dominican immigrants fare poorly there.4 Half of those age 25-64 are not high school graduates (52%) and few are proficient in English (37%) or naturalized citizens (42%). These characteristics contribute to a high poverty rate (43%) that cannot be attributed to recency of arrival given that 92% have 5+ years in the U.S.
Such changes have created numerous challenges. The mayor acknowledges that “Reading has been going through some real assimilation problems. It is a threat to a whole lot of people” (Montgomery, 2008). Indeed, a Dominican Independence Day “riot” resulted in property damage, an injured police officer, and arrests (Kahl, 2008). Moreover, representation in city government is an issue. The mayor’s office is devoid of Hispanics, with one each on the city council and school board. Lastly, crime rose during the middle part of the past decade when Reading had the fourth highest violent crime rate (124 per 10,000 population) and second highest property crime rate (573 per 10,000) among Pennsylvania cities (United States Department of Justice, 2007). Both rates surpassed those for New York (64 and 188, respectively).5
The potential exposure of Dominicans to such activities can be seen in Figure 2, which shows the spatial distributions of this population and crimes reported to the police around the time of the survey. Dominicans are largely concentrated in neighborhoods that are adjacent to the commercial core, as well as areas where crime and poverty are concentrated (not shown).
Figure 2.
Spatial Distribution of Dominicans and Crime in Reading, PA
RESEARCH ISSUES
Much empirical research has advanced understanding of the linkages between social disorder and social cohesiveness. However, the scope conditions of benchmark studies reduce their value for shedding light on the circumstances of immigrants in new destinations. Prior studies tend to focus on ethno-racial differences in the largest American cities with relatively little attention to the circumstances of immigrants.6 Prior studies also tend to examine “Hispanics” as an ethno-racial aggregate and, in so doing, lose sight of heterogeneity in the circumstances of specific subgroups stratified by nativity. Thus, the generalizability of prior research remains to be demonstrated.
This investigation has four objectives. Consistent with efforts to understand immigrants’ quality-of-life in new destinations, the first objective is to describe how Dominican immigrants in Reading portray their neighborhoods in terms of social disorder and the strength of the social fabric (i.e. cohesiveness). The second objective is to demonstrate how these dimensions of urban life are associated; that is, to determine whether perceived social disorder and a frayed social fabric go hand-in-hand. This is followed by an investigation into the roles of immigrant incorporation and interests in exchange/use values. Lastly, the implications of social disorder, social cohesiveness, and perceived threats to personal safety are examined. Perceived disorder and weak bonds should heighten fears about safety.
DATA AND METHODS
This study uses a 2004 survey of 61 Dominican immigrants in Reading, Pennsylvania. Respondents were recruited using non-probability sampling methods because probability sampling of rare populations is cost prohibitive (see also Levitt 2001; Itzigsohn 2009). These methods included a mass mailing, home canvassing, referrals, and street-level recruitment in areas where Dominicans are concentrated. The survey was administered as a computer-assisted personal interview, available in Spanish and English, by bilingual interviewers.
Although representativeness is problematic for non-probability samples, this concern is assuaged here by the similar socio-economic profiles of survey respondents and their counterparts in the 2000 Census (identifying reference).7 In addition, multiple imputation was used for missing data. Thus, all observations are analyzed.
Dependent Variables
Social Disorder
Respondents indicated whether several signs of neighborhood social disorder were “not a problem” (coded 1), “somewhat of a problem” (2) or a “big problem” (3).8 The signs include “little respect for rules, laws, and authority,” “assaults or muggings,” “gangs,” and “drug use or dealing in the open.” With inter-item correlations between .64 and .93 (Cronbach’s Alpha=.92), a four-item index ranging in value from 4 to 12 was initially created. Because the distribution of the index is U-shaped, it is trichotomized into no or little (4,5), moderate (6-8), and high perceived levels of disorder (9-12). “No disorder” is the reference category.
Social Cohesiveness
Social cohesiveness has behavioral and psychological dimensions (Sampson et al., 1999). Behaviorally, it reflects informal interactions; that is, how frequently respondents and neighbors “chat or drop in for a social visit,” “talk about job openings,” and do “favors for each other.” Frequency was measured with a seven-point scale ranging from “never” (1) to “several times a week or more” (7). Responses were summed to create an Index of Neighborhood Interaction (Cronbach’s Alpha=.7).
The Index of Neighborhood Cohesiveness utilizes five-point Likert items that indicate strength of agreement with the statements: “this is a close-knit neighborhood,” “people around here are willing to help their neighbors,” “people in this neighborhood can be trusted,” and “people in the neighborhood generally do not get along with each other” (reverse coded). Summed responses form an index ranging from 4 to 20 (Cronbach’s Alpha=.78).
Personal Safety
Respondents were asked: “How safe is it to walk around in your neighborhood after dark?” Categories varied between “extremely dangerous” (1) and “completely safe” (4).
Independent Variables
Victimization is potentially important if experiences matter for perceptions. Information was provided on whether any member of the household had anything stolen or damaged in the neighborhood.
Incorporation is determined by English language proficiency and citizenship. Using categories ranging from “not at all” (1) to “very well” (4), language is assessed with self-reported ability to speak, understand, write, and read English. Responses were summed to form an index ranging in value from 4 to 16. Citizenship distinguishes naturalized from non-naturalized residents. Incorporation is also a function of time, measured here as the logged number of years in the United States and in Reading. No information is available on time in the neighborhood.
Interests in the exchange value of property are a function of homeownership. Life cycle characteristics that determine interests in the use value of neighborhoods are measured in terms of parenthood, age, and marital status. Sex is an additional covariate.
Methodological Issues
Omitted variable bias is an issue for all non-experimental research. This analysis does not include measures of socioeconomic status and “objective” measures of disorder. As for the former, education, income, and employment are excluded for parsimony since they were not significant in preliminary analyses. As for the latter, independently observed measures of disorder cannot be included because respondents’ specific addresses were not recorded due to privacy considerations. This information would be useful if perceptions are biased and influenced by how neighborhoods are defined (Hipp 2010). At the same time, previous studies show that the association between subjective and objective measures is strong (Ross et al., 2001; Sampson and Raudenbush, 2004). Thus, we recognize that social characteristics influence both the types of areas people live in and how they view them. Perceptions generally reflect exposure to objective conditions that are interpreted in light of social experiences.
Second, causal inferences from associations between attitudes and behaviors are tenuous. For example, some prior studies assume that fear of crime affects perceptions of neighborhood problems (Hipp 2010) or exchanges affect concerns about disorder (Sampson and Raudenbush 2004), rather than the reverse. These issues are intractable with cross-sectional data. Nevertheless, the approach used here is generally consistent with prevailing theoretical frameworks (e.g., see Woldoff 2002).
Third, the small sample size required adjustments in modeling and statistical procedures. Multivariate analyses are limited to significant covariates in bivariate models and reduced models that tested for interactions. As for statistical procedures, the p-value is set at .10 to compensate for the low power of statistical tests for small samples and to reduce the chance of a Type II error. Needless to say, an additional caveat is necessary because statistical tests assume probability-based sampling methods. Such tests are relied on here for their value in providing a principled way to identify results that merit attention, even for non-probability samples.
RESULTS
The descriptive statistics in Table 1 provide insights into whether the aforementioned overlap in the spatial distribution of Dominicans and crime is generally reflected in perceptions of problems with neighborhood disorder. At one extreme (Panel A), 36% score 9 or above with most of these at 11 or 12 (29% of the total). Such high scores require identification of three or four major problems. The most frequently mentioned major problems involve gangs. 43% of respondents indicate gangs are a big problem and 21% indicate they are a minor problem. In addition to lack of respect for “rules and laws,” substantial shares also point to muggings and drugs as serious. At the other extreme, about 40% indicate that there are no signs (32% score 4) or just one sign of disorder (8% score 5) that is “somewhat” of a problem.
Table 1.
Descriptive Statistics for Measures of Social Disorder and Social Cohesion
| Panel A. Perceived Social Disorder Index | |
| Disorder Index - Mean (Median) | 7.0(7.4) |
| % High disorder (9+ on additive index) | 36.1 |
| % Moderate disorder (6 – 8 on additive index) | 23.0 |
| % Low disorder (4 or 5 on additive index) | 41.0 |
| Individual Items | |
| Little respect for rules, laws, and authority in neighborhood | |
| % A big problem | 22.6 |
| % Somewhat of a problem | 24.6 |
| % Not a problem | 52.8 |
| Assaults or muggings in neighborhood | |
| % A big problem | 33.4 |
| % Somewhat of a problem | 21.0 |
| % Not a problem | 45.6 |
| Delinquent gangs or drug gangs in neighborhood | |
| % A big problem | 43.3 |
| % Somewhat of a problem | 21.3 |
| % Not a problem | 35.4 |
| Drug use or drug dealing in the open in neighborhood | |
| % A big problem | 36.7 |
| % Somewhat of a problem | 18.0 |
| % Not a problem | 45.4 |
| Panel B. Index of Neighborhood Cohesiveness | |
| Perceived Social Cohesiveness Index - Mean (Median) | 14.2 (14.0) |
| Individual Items | |
| % Strongly agree …“people around here are willing to help their neighbors” | 33.8 |
| % Strongly agree …“this is a close-knit neighborhood” | 32.5 |
| % Strongly agree …“people in neighborhood generally…get along”a | 25.3 |
| % Strongly agree …“people in this neighborhood can be trusted” | 24.6 |
| Panel C. Index of Neighbor Interaction | |
| Behavioral Social Cohesion Index – Mean (Median) | 10.4 (10.0) |
| Individual items | |
| % At Least Once a Week … “Chat with neighbors or drop in for a visit” | 47.4 |
| % At Least Once a Week … “Talk to neighbors about job openings” | 23.3 |
| % At Least Once a Week … “Give and receive favors such as watching children or lending or borrowing” | 25.3 |
More precisely, the percentage who “strongly disagree” that people in the neighborhood “don’t get along with each other.”
The means for the indexes of cohesiveness and interaction are 14 (Panel B) and 10 (Panel C). The individual items show one-third strongly agrees that they live in a close-knit neighborhood where people are willing to help each other. A quarter feels similarly about whether residents get along and are trustworthy. Overall, 20% strongly agrees with at least three items (not shown). As for interaction, half have informal conversations with neighbors at least once a week and one-quarter frequently exchange favors or information. At the other extreme, nearly half do not strongly agree with any attitudinal measure of cohesiveness and a fifth has no contact with neighbors (not shown).
Table 2 profiles those with different views about disorder. Those with high scores on the disorder index perceive relatively low levels of neighborhood cohesiveness, a result that is not replicated for neighboring behavior. Also, the typical respondent in a (perceived) high disorder neighborhood has not been victimized, but a substantial share (38%) claims to have had property damaged or stolen in the area. This is substantially higher than the figure for those in areas with low disorder (9.8%).
Table 2.
Descriptive Statistics (N=61)
| Index of Perceived Social Disorder |
|||
|---|---|---|---|
| Low (N=25) |
Moderate (N=14) |
High (N=22) |
|
| Index of Social Cohesiveness (Mean) | 15.0 | 14.0 | 12.8 |
| Index of Neighbor Interaction (Mean) | 10.3 | 9.7 | 10.5 |
| Victimization % Yes |
9.8 | 30.0 | 37.5 |
| Incorporation & Exposure | |||
| English Proficiency Index (Mean) | 8.4 | 8.1 | 10.4 |
| Citizenship Status | |||
| % U.S. Citizen | 26.8 | 45.7 | 40.2 |
| % Not U.S. Citizen | 73.2 | 54.3 | 59.8 |
| Years in Reading (Mean) | 4.1 | 5.1 | 5.8 |
| Years in U.S. (Mean) | 13.3 | 13.0 | 11.0 |
| Life Cycle & Tenure | |||
| Age (Mean) | 41.3 | 45.2 | 36.7 |
| Marital Status | |||
| % Married/Cohabiting | 64.2 | 71.4 | 56.3 |
| % Single | 35.8 | 28.6 | 43.7 |
| Parenthood | |||
| % Children | 39.8 | 51.4 | 53.6 |
| % No Children | 60.2 | 48.6 | 46.4 |
| Homeownership | |||
| % Own | 22.0 | 40.0 | 22.3 |
| % Rent | 78.0 | 60.0 | 77.7 |
| Sex | |||
| % Male | 53.7 | 52.9 | 46.4 |
| % Female | 46.3 | 47.1 | 53.6 |
The results for measures of incorporation are intriguing; especially given studies which suggest that residence in better neighborhoods is generally a concomitant of the assimilation process, but spatial assimilation is problematic for groups from the Spanish Caribbean. The typical respondent who feels disorder is pervasive is a non-citizen at the mid-point of the English proficiency scale who has been in Reading for about 6 years and about 11 years in the U.S. However, those who perceive high levels of disorder appear to be more likely than their counterparts to be citizens (40% vs. 27%), adept at English (10.4 vs. 8.4), and to have lived in Reading for a longer period of time (5.8 vs. 4.1). In addition, those who perceive high levels of disorder tend to be younger renters in their mid-30s who are as likely as not to be men, married and to have children. Their low disorder counterparts are similar, though perhaps a little older and more likely to be non-parents.
Table 3 presents odds ratios from multinomial logistic regressions for social disorder. Here we see that just three variables approach significance in bivariate models. The odds of perceiving a high level of disorder is 5.7 times greater and the odds of perceiving a moderate level of disorder is 4.0 times (albeit non-significant) greater for crime victims than for non-victims. As for the other covariates, incorporation matters more than interests in the exchange and use value of neighborhoods for perceived exposure to problems. Perceived disorder increases with language proficiency (1.21), along with time in Reading (2.24). Those who have a greater facility with English and who have spent more time in Reading are more likely to “see” problems in the streets. Life cycle measures and homeownership fail to approach significance.
Table 3.
Odds Ratios from Multinomial Logistic Regressions: Perceived Social Disorder Index
| Bivariate |
Best Predictors |
|||
|---|---|---|---|---|
| High vs. Low |
Moderate vs. Low |
High vs. Low |
Moderate vs. Low |
|
| Victimization | 5.66* | 3.99 | 4.79+ | 3.37 |
| Incorporation & Exposure | ||||
| English Proficiency Index | 1.21+ | .96 | 1.18 | .91 |
| Citizenship Status | ||||
| U.S. Citizen | 1.85 | 2.32 | ||
| Not U.S. Citizen (ref.) | ---- | ----- | ||
| Years in Reading | 2.24* | 1.54 | 1.14 | 1.51 |
| Years in U.S. | .72 | .85 | ||
| Life Cycle & Tenure | ||||
| Age | .97 | 1.03 | ||
| Marital Status | ||||
| Married/Cohabiting | .72 | 1.42 | ||
| Single (ref.) | ----- | ----- | ||
| Parenthood | ||||
| Children | 1.74 | 1.60 | ||
| No children (ref.) | ----- | ----- | ||
| Homeownership | ||||
| Own | 1.03 | 2.39 | ||
| Rent (ref.) | ----- | ----- | ||
| Sex | ||||
| Male | .75 | .97 | ||
| Female (ref.) | ----- | ----- | ||
p < .10
p < .05
p < .01
p < .001
In the multivariate model, victimization alone remains significant. The odds ratio for the contrast between victims and non-victims is 4.8 after English proficiency and exposure are controlled. The failure of the latter variables to approach significance results from their mutual association (r = .47). English proficiency and victimization are both significant when exposure is excluded.
Table 4 provides results for social cohesiveness. Column 1 shows that perceived social bonds between neighbors decline with disorder. The unstandardized coefficient of −2.72 indicates that the mean for those who feel disorder is pervasive is nearly 3 points lower than the mean for those who perceive little disorder. Put differently, those who see few signs of neighborhood disorder tend to feel closer to their neighbors. Similarly, those who interact with neighbors feel a sense of connection to them (.16), as do those who are Spanish dominant (i.e. less English proficient, −.45) and older.
Table 4.
Unstandardized Regression Coefficients from OLS Regressions: Indexes of Perceived Neighborhood Cohesiveness and Interaction with Neighbors
| Index of Neighborhood Cohesiveness |
Index of Neighborhood Interaction |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Bivariate - Level of Disorder |
Bivariate -Level of Disorder |
||||||||||
| Bivariate (1) |
(2) | (3) | Low (4) | Moderati (5) | s High (6) |
Bivariate (7) | (8) | Low (9) | Moderate (10) |
High (11) |
|
| Perceived Disorder | |||||||||||
| High | −2.72* | −2.79* | −1.51 | −.16 | −.39 | ||||||
| Moderate | −1.59 | −1.50 | −1.77 | −1.00 | −2.54 | ||||||
| Low | |||||||||||
| Victimization | .13 | .30 | 4.10* | 4.00* | |||||||
| Index of Neighbor Interaction |
.16+ | .15 | .10 | ||||||||
| Incorporation & Exposure | |||||||||||
| English Proficiency Index | −.45** | −.31+ | .10 | −.36 | −.67** | −.00 | |||||
| Citizenship | |||||||||||
| Citizen | .47 | 1.59 | 4.37*** | .39 | −2.18 | .91 | .49 | 6.27* | −2.49 | −1.46 | |
| Not Citizen (ref.) | |||||||||||
| Years in Reading (log) | −.86 | 1.41 | |||||||||
| Years in U.S. (log) | .41 | 1.78+ | .86 | ||||||||
| Life Cycle & Tenure | |||||||||||
| Age | 14*** | .09* | .12* | .11+ | |||||||
| Marital Status | |||||||||||
| Married/Cohabiting | .32 | 3.98** | 3.59* | ||||||||
| Single (ref.) | |||||||||||
| Parenthood | |||||||||||
| Children | −1.18 | 1.59 | |||||||||
| No Children (ref.) | |||||||||||
| Homeownership | |||||||||||
| Own | −.20 | 2.71 | |||||||||
| Rent (ref.) | |||||||||||
| Sex | |||||||||||
| Male | .58 | 1.39 | |||||||||
| Female (ref.) | |||||||||||
p<.10
p < .05
p < .01
p < .001.
The multivariate models indicate that the association of cohesiveness with disorder is unaffected by victimization and neighboring (column 2), but it weakens considerably with the inclusion of other controls (column 3). Additional analysis reveals, however, that citizenship and language interact with social disorder.9 Among those who see few signs of disorder (column 4), naturalized citizens are more likely than non-citizens to feel that the social fabric is strong. The mean for citizens is four points higher (4.4) than that non-citizens. The other estimates suggest that differences become more similar as perceived disorder increases.
In contrast to citizenship, the effect of English is increasingly negative as disorder increases. No association exists between language and cohesiveness among those who see few signs of disorder, but the perception of a tightly-woven social fabric is negatively associated with English among those who describe disorder as pervasive (−.67).10
Although no association is evident for perceived disorder in additive models, columns 7 and 8 suggest that neighborhood experiences matter for neighboring behavior. Crime victims interact more than non-victims with neighbors (4.1), suggesting that victimization may promote vigilance and efforts to establish protective circles of trust through interaction. In this vein, it is telling that respondents were also asked how many neighbors they knew by name: 60% of victims and only 24% of non-victims know all or nearly all of their neighbors (not shown).
The positive parameter estimate for years of exposure to the U.S. is the only measure of incorporation to approach significance in the bivariate or multivariate additive models, but it is not significant in the latter. These models suggest that attention should turn instead to how interests in the exchange and use value of neighborhoods encourage investments in social relationships. Neighboring is more common among older and married residents even after accounting for other covariates. The coefficient for homeowners (2.71, p = .11) is noteworthy as well.
At the same time, multiplicative models re-direct attention to the process of incorporation where an important finding for the subjective measure of cohesiveness is replicated (columns 9-11). Naturalized citizens are more likely than non-citizens to interact with neighbors among residents who perceive little neighborhood disorder (6.27). Citizens are not different from non-citizens among those who perceive disorder. Unlike citizenship, the test for an interaction between disorder and language was not significant.
As might be expected from the results for perceived disorder, considerable variation is evident perceptions of personal safety. Approximately half feel “completely” (31%) or “fairly” safe (21%) walking alone in their neighborhoods at night. The remainder says it would be “somewhat dangerous” (46%) or “extremely dangerous” (11%) to do so. Differences in perceived safety by disorder are striking as well. Two-thirds of those in areas characterized as having high or moderate levels of disorder feel they are “somewhat” or “extremely” dangerous places. Just 13% of those who see no signs of disorder responded similarly.
Table 5 formally replicates and extends these results. The regression coefficients show that safety concerns are associated with high (-.95) and moderate levels of disorder (-.68). In other words, those who see few signs of disorder tend to feel the safest. Safety is associated with perceptions of cohesiveness as well, especially among those who describe their areas as orderly. Among those who see few signs of disorder, viewing neighbors as trustworthy and helpful tends to promote a sense of safety. Social cohesiveness does not alleviate concerns about safety among those who feel neighborhood disorder is pervasive.
Table 5.
Unstandardized Regression Coefficients from OLS Regressions: Safety of Neighborhood at Night
| Bivariate -Perceived Disorder |
||||
|---|---|---|---|---|
| Bivariate (1) |
Low (2) |
Moderate (3) |
High (4) |
|
| Perceived Disorder | ||||
| High | −.95** | |||
| Moderate | −.68+ | |||
| Low (ref.) | ----- | |||
| Index of Neighborhood Cohesiveness | .07+ | .14* | .13 | −.00 |
| Index of Neighbor Interaction | .01 | .07 | −.04 | −.02 |
p<.10
p < .05
p < .01
p < .001.
CONCLUSIONS
This study has pursued several objectives in the hope of shedding light on the neighborhood experiences of Dominican immigrants in a new destination. The first objective was to describe their perceptions of neighborhood social disorder and cohesiveness. Dominican immigrants are bifurcated into those who claim that social disorder is rife and those who claim that social order prevails. Dominican immigrants also vary widely in their interactions with neighbors and perceptions of neighborhood cohesiveness. Such findings are important because interviews with our respondents revealed that quality-of-life considerations are major reasons for moving to Reading (not shown). Reading may promise a lower cost-of-living and an escape from the “mean streets” of New York, but the results suggest that escape from problems proves elusive for many.
The second and third objectives focused on linkages between perceived disorder, cohesiveness, and various social characteristics. The results suggest a negative association between perceived social disorder and the strength of the social fabric. Those who are cognizant of numerous disorder-related problems tend to feel that their neighborhood is not close-knit and residents are not trustworthy. Moreover, perceptions of social disorder and cohesiveness are intertwined with measures of incorporation. Among those who see few signs of social disorder, citizens are more likely than non-citizens to view their neighborhood as cohesive and to interact with neighbors. Citizenship matters little for those who view their areas as having pervasive problems.
Although an “air-tight” interpretation would be aided by additional information on how citizenship affects the residential locations of Dominicans in new destinations, this finding is consistent with accounts of naturalization as a transformative process that confers full membership in the larger community. Naturalization is facilitated by the development of roots in the U.S. through property ownership and family ties, as well as the acquisition of forms of capital that expand opportunities for participation (e.g. English proficiency). Naturalization also involves psychological attachments (Desipio and de la Garza 1998). Ambivalence about life in America and the appearance of forsaking allegiance to the island are barriers to the acquisition of citizenship among some Dominicans (Gilbertson and Singer 2000).11 Thus, citizenship and social networks may reinforce one another in areas characterized by the absence of disorder. Citizenship lowers a social barrier to interaction and fosters cohesion among neighbors.
The results for language are intriguing, especially given that English proficiency is associated with citizenship. English is only associated with views about cohesiveness among those who identify several major neighborhood problems. Cohesiveness and trust decline with English proficiency among those who perceive high levels of disorder. Given that the bivariate association between language and disorder is weakly positive, it is possible that those who are adept at English may seek to distance themselves from residents of areas with activities that reflect poorly on both themselves and the neighborhood population. English proficiency may increase the ability to maneuver among the majority and the desire to distance oneself from fellow residents who are “against the law.”
The last objective focused on personal safety. As with disorder, residents differ greatly in their perceptions. About half feels their neighborhoods are safe and half feels they are unsafe. Perceived safety is promoted by both order and a close-knit social fabric. Social cohesiveness fosters safety among those who feel their area is orderly. It is inconsequential among those who feel neighborhood disorder is severe. Such findings reinforce suspicions that disorder “equalizes” and diminishes social differences in perceptions (notwithstanding results for language).
Some of the main findings from additive models regarding associations between perceptions of disorder and cohesiveness are consistent with “conventional wisdom.” It is worth reiterating, however, that the analysis reveals blind spots in the literature. The emerging literature on new destinations has not drawn substantial attention to neighborhood conditions, perhaps because of evidence that migration to new places generally improves welfare (Alba et al., 2009). Still, new destinations may present the same problems as old destinations to a substantial share of immigrants. The findings also provide a corrective to previous studies that tend to view these issues through the lens of race. Studies that focus exclusively on race are likely to miss how immigration and incorporation shape urban experiences. No study has demonstrated the intricacies of how citizenship, language, and neighborhood disorder jointly affect cohesiveness. In doing so here, additional avenues for a more nuanced understanding of neighborhood life are revealed. Perceptions cannot be reduced to interests in use and exchange values.
In closing, the foregoing has provided a case study of a single group in a single place at a single point in time. Future studies should strive to expand these scope conditions to investigate how different immigrant groups fare in different destinations. This will require the generic “Hispanic” category to be disaggregated to generate detailed portraits of the lives of specific groups in their neighborhoods. Additionally, future studies of new destinations must continue to examine the assumption that perceptions reflect actual neighborhood conditions filtered through the lens of social experience. Information on “objective” neighborhood conditions and changes in residential locations with incorporation would provide insights into how immigrants subjectively experience their new neighborhoods. This obviously needs to be expanded as well to examine not only the immigrants own perceptions of their neighborhoods, but also the reputations of their neighborhoods among others. Thus, the challenges are formidable for future efforts to determine the role of neighborhoods in shaping “the” American experience of immigrants in new urban destinations.
Acknowledgments
Support for this research was provided by the Russell Sage Foundation (project # 88-04-01) and the Population Research Institute at The Pennsylvania State University, which has core funding from the National Institute of Child Health and Human Development (2 R24 HD041025-06). The author appreciates the helpful comments of Nancy Landale, Barry Lee and Pete Guest, along with the map making assistance of Luis Sanchez.
Footnotes
All unattributed figures throughout are from the 2000 Census and the 2006-2008 American Community Surveys.
Incorporation refers to the “ways in which immigrants become American” (Itzigsohn 2009, p. 5) or the “processes by which new groups establish relationships with host societies” (Bean and Stevens 2003, p. 95).
Nether Dominicans nor Reading are claimed to be a “representative” case. As a Spanish-speaking Afro-Caribbean population, Dominicans do not represent Hispanics and their growing presence in cities from Boston to Miami suggests that a prototypical new destination does not exist.
The Dominican-White Index of Dissimilarity is .46 for Reading and .81 for Berks County.
No information is available on the race/nativity of perpetrators and victims of crime. The “most wanted” list is dominated by Hispanics (www.readingpa.gov/police_most_wanted.asp).
These difficulties are amplified in studies limited to the English-speaking population (e.g., see Ross and colleagues).
This was undoubtedly aided by the recruitment of respondents from several areas (Figure 2).
In answering questions about perceived disorder and cohesiveness, respondents were instructed to think about the street they lived on and several streets in each direction.
Results for citizenship and English are robust when both are in the same model.
Citizenship and English proficiency also interact (t = −2.07, p < .05). English proficiency more strongly inhibits the sense that neighbors are connected among citizens (−.83, p < .001) than among non-citizens (b = −.38, p <.05).
Naturalized are less likely than noncitizens in Reading to expect to move back to the Dominican Republic.
REFERENCES
- Alba R, Denton N, Hernandez D, Disha I, McKenzie B, Napierala J. Nowhere near the same: The neighborhoods of Latino children. In: Landale NS, McHale S, Booth A, editors. Growing up Hispanic. The Urban Institute; Washington, D.C.: 2009. [Google Scholar]
- Brookings Institution . A Profile of the Reading Area. Center on Urban and Metropolitan Policy; Washington, D.C.: 2003. [Google Scholar]
- DeSipio L, de la Garza RO. Making Americans, Remaking America. Westview; Boulder, CO: 1998. [Google Scholar]
- Duany J. Quisqueya on the Hudson: The Transnational Identity of Dominicans in Washington Heights. The CUNY Dominican Studies Institute; New York: 1994. [Google Scholar]
- Franzini L, Caughy MO, Nettles SM, O’Campo P. Perceptions of disorder: contributions of neighborhood characteristics to subjective perceptions of disorder. Journal of Environmental Psychology. 2008;28:83–93. [Google Scholar]
- Garcia RM, Taylor RB, Lawton BA. Impacts of violent crime and neighborhood structure on trusting your neighbors. Justice Quarterly. 2007;24:679–704. [Google Scholar]
- Gilbertson G, Singer A. Naturalization under changing conditions of membership: Dominican immigrants in New York City. In: Foner N, Rumbaut RG, Gold SJ, editors. Immigration Research for a New Century. Russell Sage; New York: 2000. [Google Scholar]
- Grieco E. American Community Survey Reports. U.S. Census Bureau; Washington, D.C.: 2010. Race and Hispanic origin of the foreign-born population in the United States: 2007. [Google Scholar]
- Guest AM, Cover JK, Matsueda RL, Kubrin CE. Neighborhood context and neighboring ties. City & Community. 2006;5:363–385. [Google Scholar]
- Handlin O. The Newcomers: Negroes and Puerto Ricans in a Changing Metropolis. Harvard University Press; Cambridge, MA: 1959. [Google Scholar]
- Hipp JR. Residents perceptions of crime and disorder: How much is “bias,” and how much is social environment differences? Criminology. 2010;48(2):475–507. [Google Scholar]
- Iceland J. Where We Live Now. University of California; Berkeley: 2009. [Google Scholar]
- Itzigsohn J. Encountering American Faultlines. Russell Sage; New York: 2009. [Google Scholar]
- Jargowsky Paul A. Immigrants and neighbourhoods of concentrated poverty: Assimilation or stagnation? Journal of Ethnic and Migration Studies. 2009;35:1129–1151. [Google Scholar]
- Kahl J. Hundreds riot at Reading High School. The Reading Eagle. 2008 Feb 28; [Google Scholar]
- Kasinitz P, Mollenkopf JH, Waters MC, Holdaway J. Inheriting the City: The Children of Immigrants Come of Age. Russell Sage; New York: 2008. [Google Scholar]
- Levitt P. The Transnational Villagers. University of California; Berkeley: 2001. [Google Scholar]
- Lichter DT, Parisi D, Taquino MC, Grice SM. Residential segregation in new Hispanic destinations: Cities, suburbs, and rural communities compared. Social Science Research. 2010;39:215–230. [Google Scholar]
- McCord ES, Ratcliffe JH, Garcia RM, Taylor RB. Nonresidential crime attractors and generators elevate perceived neighborhood crime and incivilities. Journal of Research in Crime and Delinquency. 2007;44:295–320. [Google Scholar]
- Massey DS, Capoferro C. The geographic diversification of American immigration. In: Massey DS, editor. New Faces in New Places. Russell Sage; New York: 2008. [Google Scholar]
- Monger R. Annual Flow Report. Department of Homeland Security; Washington, D.C.: 2009. U.S. Legal Permanent Residents. [Google Scholar]
- Massey DS, Capoferro C. The Geographic Diversification of American Immigration. In: Massey DS, editor. New Faces in New Places. Russell Sage; New York: 2008. pp. 25–50. [Google Scholar]
- Montgomery D. The Engine of Change. The Washington Post. 2008 Apr 22; [Google Scholar]
- Pessar Patricia R. Visa for a Dream:Dominicans in the United States. Allyn & Bacon; Boston: 1995. [Google Scholar]
- Portes A, Bach RL. Latin Journey: Cuban and Mexican Immigrants in the United States. University of California; Berkeley, CA: 1985. [Google Scholar]
- Reimers C. Economic well-being. In: Tienda M, Mitchell F, editors. Hispanics and the Future of America. The National Academies; Washoington, D.C.: 2006. [PubMed] [Google Scholar]
- Ross CE, Mirowsky J. Disorder and decay: The concept and measurement of perceived neighborhood disorder. Urban Affairs Review. 1999;34:412–432. [Google Scholar]
- Ross CE, Mirowsky J, Pribesh S. Powerlessness and the amplification of threat: Neighborhood disadvantage, disorder, and mistrust. American Sociological Review. 2001;66:568–591. [Google Scholar]
- Ross CE, Mirowsky J, Pribesh S. Disadvantage, disorder, and urban mistrust. City & Community. 2002;1:59–82. [Google Scholar]
- Ross CE, Jang SJ. Neighborhood disorder, fear, and mistrust: The buffering role of social ties with neighbors. American Journal of Community Pyschology. 2000;28:401–420. doi: 10.1023/a:1005137713332. [DOI] [PubMed] [Google Scholar]
- Sampson RJ. Transcending tradition: New directions in community research, Chicago style. Criminology. 2002;40:213–230. [Google Scholar]
- Sampson RJ, Bartusch DJ. Legal cynicism and (subcultural?) tolerance of deviance: The neighborhood context of racial differences. Law & Society Review. 1998;32:777–804. [Google Scholar]
- Sampson RJ, Morenoff JD, Earls F. Beyond social capital: Spatial dynamics of collective efficacy for children. American Sociological Review. 1999;64:633–660. [Google Scholar]
- Sampson RJ, Morenoff JD, Gannon-Rowley T. Assessing ‘Neighborhood Effects’: Social Processes and New Directions in Research. Annual Review of Sociology. 2002;28:443–778. [Google Scholar]
- Sampson Robert J., Raudenbush Stephen W. Systematic social observation of public spaces: A new look at disorder in urban neighborhoods. American Journal of Sociology. 1999;105:603–51. [Google Scholar]
- Sampson RJ, Raudenbush SW. Seeing disorder: Neighborhood stigma and the social construction of ‘broken windows. Social Psychology Quarterly. 2004;67:319–342. [Google Scholar]
- Sampson RJ, Raudenbush SW, Earls F. Neighborhoods and violent crime: A multilevel study of collective efficacy. Science. 1997;277:918–924. doi: 10.1126/science.277.5328.918. [DOI] [PubMed] [Google Scholar]
- South SJ, Crowder K, Chavez E. Exiting and entering high-poverty neighborhoods: Latinos, Blacks, and Anglosd compared. Social Forces. 2005a;84 [Google Scholar]
- South SJ, Crowder K, Chavez E. Geographic mobility and spatial assimilation among US Latino Immigrants. International Migration Review. 2005b;39:577–607. [Google Scholar]
- South SJ, Crowder K, Chavez E. Migration and spatial Assimilation among U.S. Latinos: Classical versus segmented assimilation. Demography. 2005c;42:497–521. doi: 10.1353/dem.2005.0025. [DOI] [PubMed] [Google Scholar]
- Singer A. The Rise of New Immigrant Gateways. The Brookings Institution; Washington, D.C.: 2004. [Google Scholar]
- Suro R. Strangers among Us: How Latino Immigration is Transforming America. Alfred Knopf; New York: 1998. [Google Scholar]
- Taylor RB. Breaking Away from Broken Windows. Westview; 2001. 1996. [Google Scholar]
- Torres-Saillant S, Hernández R. The Dominican Americans. Greenwood; Westport, CT: 1998. [Google Scholar]
- U.S. Department of Homeland Security . Yearbook of Immigration Statistics: 2009. Office of Immigration Statistics; Washington, D.C.: 2010. [Google Scholar]
- U.S. Department of Justice Crime in the United States, 2006. 2007 http://www.fbi.gov/ucr/cius2006/
- Wirth L. The Ghetto. Transaction; New Brunswick, NJ: 1998[1928]. [Google Scholar]
- Whyte WF. Street Corner Society: The Social Structure of an Italian Slum. University of Chicago Press; Chicago: 1993[1943]. [Google Scholar]
- Woldoff RA. The effects of local stressors on neighborhood attachment. Social Forces. 2002;81:87–116. [Google Scholar]
- Zorbaugh H. The Gold Coast and the Slum: A Sociological Study of Chicago’s Near North Side. University of Chicago Press; Chicago: 1965[1929]. [Google Scholar]