Community Disadvantage,Parental Network,and Commitment to Social Norms

Abstract

Objectives:

Social disorganization theory implies that neighborhood disadvantage influences delinquency in part through the weakening of neighborhood-level social ties and residents’ commitment to social norms. We test these associations by focusing on social ties among neighborhood parents and adolescent commitment to social norms.

Methods:

We use a population survey of adolescents and combine it with administrative (population) data on school neighborhood characteristics in Iceland. We use multilevel data on 83 school communities and 5,865 adolescents in Iceland to analyze our hypotheses.

Results:

We find partial support for our hypotheses. Thus, adolescents living in neighborhoods characterized by concentrated disadvantage are more delinquent, net of individual-level (household) characteristics. Moreover, neighborhood-level parental networks and adolescent commitment to social norms mediate a part of this contextual effect.

Conclusion:

By supporting community theories in a different societal context (i.e., in a small homogeneous society) than most prior work, our study strengthens the external validity of the existing research. As we use cross-sectional data, the study faces the limitation of not being able to separate the constructs temporally.

Keywords

social disorganization criminological theory communities and crime multilevel analysis adolescent delinquency parental networks

One of the major propositions of social disorganization theory is that widespread disadvantage in the neighborhood can weaken the types of neighborhood-based social ties that produce social control over children and adolescents (Kornhauser 1978; Sampson and Groves 1989). The theory thus argues that social ties among neighborhood residents entail supervision, surveillance, and scrutiny of adolescents (Coleman 1988; Krohn 1986), and increase normative consensus among residents, thus strengthening adolescent commitment to community norms (Bernburg and Thorlindsson 2007).

While prior work generally provides support for social disorganization theory, research findings on the intermediate role of neighborhood-based social ties have been inconsistent (see reviews by Bursik and Grasmick 1993; Sampson, Morenoff, and Gannon-Rowley 2002; Steenbeek and Hipp 2011). Moreover, ethnographic work has described strong neighborhood networks in impoverished, high-crime communities (Anderson 1999; Patillo 1998). Such findings have led scholars to downplay the intermediate role of neighborhood ties, arguing that such ties may not be sufficient to provide effective social control and may even inhibit it (Sampson, Morenoff, and Earls 1999:635).

However, while most research has focused on ties among general residents, the effects that ties among neighborhood residents have on social control and delinquency may depend on the type of actors involved in the networks (Kubrin and Weitzer 2003; Rountree and Warner 1999). In particular, ties among parents in the neighborhood may provide a particularly important source of social control of children and adolescents. Parents have a greater interest in the conformity of adolescents than do other residents, and they have more means of controlling their behavior. Active parental networks facilitate parental supervision, surveillance, and enforcement of sanctions, and they may strengthen consensus among neighborhood parents about the social norms (Bernburg and Thorlindsson 2007; Coleman 1988). The few articles that focus on community-based parental ties suggest that such networks may provide an important source of adolescent social control (Bernburg and Thorlindsson 2007; Bjarnason et al. 2005; Thorlindsson, Bjarnason, and Sigfúsdóttir 2007). Still, limited research exists on the role of neighborhood-based parental ties in mediating the effect of neighborhood disadvantage on delinquency. We extend the research by examining the contextual effects of neighborhood disadvantage and residential instability on adolescent delinquency, and testing whether neighborhood parental ties and adolescent commitment to social norms mediate the effects.

We use Icelandic data that are uniquely well suited for this purpose. We combine a population survey of adolescents with administrative (population) data on neighborhood characteristics, allowing us to construct reliable, measurement-rich multilevel data on both neighborhood and individual-level characteristics. We identify community boundaries by using officially defined geographical boundaries of public schools. As prior community research has underscored (Bernburg and Thorlindsson 2005, 2007; Bernburg, Thorlindsson, and Sigfúsdóttir 2009a, 2009b; Thorlindsson and Bernburg 2004), this method provides “natural” community boundaries for Icelandic adolescents, thus we avoid the common problem of using arbitrary criteria (e.g., postal codes) for drawing community boundaries (see Bursik 1988; Sampson et al. 2002). Furthermore, characterized by a small, homogenous population, a generous welfare system, and a relatively low level of income inequality,¹ the Icelandic setting differs markedly from the setting in which most prior research has been conducted (urban settings in the United States), offering an opportunity to address the external validity of the research (see Lucas 2003).

Neighborhood Disadvantage and Delinquency: The Intermediate Role of Neighborhood Ties and Commitment to Social Norms

We frame our study with social disorganization theory (Shaw and McKay 1942; Kornhauser 1978), and its later modifications to include social capital (Coleman 1988; Putnam 2000), systemic model (Bursik and Grasmick 1993; Kasarda and Janowitz 1974), and collective efficacy (Sampson et al. 1999; Sampson, Raudenbush, and Earls 1997). The basic theoretical premise in these works is that neighborhood disadvantage increases crime by undermining informal social control. The weakening of neighborhood-based social ties may play an important role in that respect. Neighborhood disadvantage, such as high levels of poverty, residential mobility, single-parent households, and ethnic/cultural heterogeneity, may weaken social ties among neighborhood residents. Thus, the creation and maintenance of ties to neighbors and community institutions, for example, through regular interaction and participation in community activities, requires time and energy, and poverty and family disruption may reduce both (Sampson and Groves 1989). High-income and educated residents are more likely than those in disadvantaged communities to have regular work hours and consequently residents’ free time in well-off neighborhoods may coincide, which should increase their chances of interacting (Bellair 1997). Likewise, residential instability, frequently generating new neighbors, makes the maintenance of social ties difficult. People are also less likely to be involved in a community that they wish to leave at the first opportunity (Kornhauser 1978). Finally, ethnic heterogeneity and immigrant concentration may weaken neighborhood-based ties and normative consensus. Different values and cultural roots, as well as linguistic barriers, may impede communication, and give rise to tensions that create obstacles to form ties (Sampson et al. 1999; Sampson and Groves 1989).

Neighborhood-based Social Ties Among Residents and Parents

Network theory implies that dense social ties produce supervision, surveillance, and scrutiny of individuals belonging to the networks (Coleman 1988; Krohn 1986). But, while research has underscored the role of informal surveillance (Bellair 2000; Bellair and Browning 2010; Browning, Feinberg, and Dietz 2004; Mazerolle, Wickes, and McBroom 2010; Morenoff, Sampson, and Raudenbush 2001; Sampson et al. 1997) and unsupervised peer activity (Bernburg and Thorlindsson 2007; Sampson and Groves 1989) in reducing neighborhood crime and delinquency, research on the role of neighborhood social ties in social control and crime is mixed. Some studies have confirmed that neighborhood-based ties (Bellair 1997; Sampson and Groves 1989; Sun, Triplett, and Gainey 2004; Veysey and Messner 1999) and participation of residents in community affairs (Hawdon and Ryan 2009; Lowenkamp, Cullen, and Pratt 2003; Sampson and Groves 1989; Simcha-Fagan and Schwartz 1986; Veysey and Messner 1999) negatively influence some types of crime. In contrast, other studies find a positive effect of neighborhood ties on crime rates (Bellair 2000; Bellair and Browning 2010; Browning et al. 2004; Kingston, Huizinga, and Elliott 2009), and some research find insignificant effects (Morenoff et al. 2001; Simcha-Fagan and Schwartz 1986; also see review in Cantillon, Davidson, and Schweitzer 2003; Sampson et al. 2002).

Moreover, in his work on predominantly Black neighborhoods in Chicago, Wilson (2008:344) argued that although many residents have strong ties to one another in poor neighborhoods, these ties often do not create informal social control of children and adolescents. The neighborhoods are isolated from mainstream society and lack ties to local institutions, such as schools and churches, that promote an atmosphere of conformity with conventional norms. Furthermore, due to the social problems in these neighborhoods, adolescents often have ties with antisocial role models, who promote delinquent behavior rather than discouraging it. Relatedly, Patillo’s (1998) ethnographic work has described how gang members and drug dealers are a part of dense networks in Black middle-class neighborhoods. In short, the effects that ties among neighborhood residents have on adolescent social control depend on the actors involved and the purpose that particular networks may have (Kubrin and Weitzer 2003).

Social ties among neighborhood residents whose social role is to control children and adolescents may be more important in controlling adolescents than other ties. Most parents want their children to stay away from delinquency, and thus neighborhood-based ties among parents may be more likely than ties among other residents to decrease youth problem behavior. When parents in a neighborhood interact with one another, it is easier for them to form a consensus on the rules and values they consider best for the well-being of their children (Bernburg and Thorlindsson 2007). If parents in a neighborhood are applying conflicting rules, it is likely that parents will be more reluctant to interfere in the misbehaviors of children other than their own (Sampson et al. 1997). Moreover, parental ties enable parents to indirectly monitor their children across places and situations (Coleman 1988). It is therefore possible to respond to deviant behavior in the appropriate manner which strengthens informal social control.

We focus on three different types of neighborhood-based parental ties. First, following Coleman (1988), we examine intergenerational closure, which refers to ties that link parents and their adolescent children residing in the same community—whereby “the parent’s friends are the parents of their children’s friends” (Coleman 1988:106; see also Sampson et al. 1999:635). Research has found that adolescents involved in such ties are less delinquent (Bernburg and Thorlindsson 2007; Thorlindsson et al. 2007). Second, we examine parental neighboring, that is, the degree to which parents interact with their neighbors for personal reasons. Parental neighboring may increase mutual trust and obligations among neighbors and thus increase informal social control (Browning et al. 2004; Elliott et al. 1996; Sampson and Groves 1989; Warner and Rountree 1997). Finally, scholars have emphasized that participation in local organizations increase social control capacities of residents (Kornhauser 1978; Simcha-Fagan and Schwartz 1986). We focus on parental participation in local institutions, particularly the school, assuming that ties between parents and the school may increase social control of adolescents (Kornhauser 1978; McNulty and Bellair 2003; see Krohn 1986).

Neighborhood Context and Adolescent Commitment to Social Norms

Social disorganization theory implies that neighborhood disadvantage influences delinquency in part by undermining the commitment of residents to social norms. As Shaw and McKay (1942:164) noted in their seminal work, low-crime areas tend to be characterized by “more or less uniformity, consistency, and universality of conventional values and attitudes with respect to child-care, conformity to law and related matters; whereas in the high-rate areas systems of competing and conflicting moral values have developed.”

Neighborhood disadvantage may undermine residents’ commitments to social norms through several mechanisms, both directly and indirectly through neighborhood-based ties. Due to widespread poverty and lack of opportunities, adolescents in disadvantaged neighborhoods may tend to believe that success in life without breaking the law, cheating, or being dishonest, is unrealistic, resulting in cynicism toward the law (Anderson 1999; Sampson and Bartusch 1998; Warner 2003). Also, disadvantaged neighborhoods may have more diverse role models (Warner 2003), including adults that rationalize the use of criminal activities. Thus, parents and teachers in most neighborhoods (both affluent and disadvantaged) are likely to emphasize conventional norms—that people should obey the law, be honest, get an education and a decent job—but these norms may be more visible in the everyday life of more affluent communities, reinforcing socialization in such neighborhoods (Kornhauser 1978:76; Wilson 2008:342). As Shaw and McKay (1942) noted, although most people in most neighborhoods have conventional norms, adolescents in poor neighborhoods are more likely to be exposed to normlessness (i.e., legal cynicism) and tolerance for deviance (Sampson and Bartusch 1998:781).

Neighborhood disadvantage may also undermine residents’ commitment to social norms indirectly through the weakening of neighborhood-based ties. As Warner (2003) has argued, social ties among neighborhood residents provide active channels to communicate and reinforce social values and norms (see Berg et al. 2012; Kornhauser 1978; Warner and Burchfield 2011). In this vein, dense parental networks may facilitate consensus among neighborhood parents about the setting of rules and enforcement of sanctions, which in turn can bring normative consistency to adolescents, and thus increase their commitment to social norms (Bernburg and Thorlindsson 2007; Coleman 1988). Moreover, when parents participate in school activities, parents are not only able to observe what values are being taught at school, but they also send the message to their children that school is important, reinforcing the values being taught there.

Community research rarely examines individual commitment to social norms directly, despite the theoretical significance of doing so. But, the few studies that have done so find the expected relationship. Felson et al. (1994) found that conduct norms (referred to as subculture of violence in their article) were associated with interpersonal violence, theft/vandalism, and school delinquency both on individual and school level (also see Bernburg and Thorlindsson 2005). Moreover, Berg et al.’s (2012) longitudinal, multilevel analysis showed that African American adolescents in disadvantaged neighborhoods were more likely to have positive attitudes toward the use of violence, which in turn predicted violent behavior (for similar findings, see Stewart and Simons 2010, 2006; Stewart, Simons, and Conger 2002). One Icelandic study found normlessness to mediate the contextual effect of community structural instability on delinquency (Bernburg and Thorlindsson 2007), but the study did not measure community (or individual) economic disadvantage. Finally, Oberwittler (2004) found that tolerance for violence mediated a positive effect of neighborhood disadvantage on serious offending among German teenagers.²

This study tests the role that adolescent commitment to social norms plays in mediating the effects of both neighborhood disadvantage and neighborhood-based parental ties on adolescent delinquency. Our measure comprises two dimensions of commitment to norms. First, we include normlessness, defined as weak attachment to social norms and perceived norm inconsistency (Bernburg and Thorlindsson 2007; see Sampson and Bartusch 1998, for discussion about legal cynicism). Coleman (1988) implies that social ties between parents in a community may help and encourage parents to set uniform standards about adolescent conduct. Second, we focus on conduct norms about proper and appropriate behavior, for example, whether adolescents think that certain situations justify aggression and violence (Bernburg and Thorlindsson 2005).

The Current Study

We use two sources of data. First, we use a population survey of Icelandic adolescents. Surveying almost the entire population allows us to construct measures of neighborhood parental ties that are based on survey answers of the majority of adolescents residing in each neighborhood. Second, we obtain administrative, population data on neighborhood disadvantage. Combining these data enables us to construct multilevel data on neighborhood disadvantage, neighborhood-based parental ties, and personal situations, characteristics, and delinquent behaviors of the majority of the adolescents residing in the neighborhoods.

Social disorganization theory implies that community characteristics influence adolescent delinquency net of the personal situations and characteristics of individual adolescents (Simcha-Fagan and Schwartz 1986). The data allow us to test such “contextual” effects of community factors on individual behavior (see Bursik 1988; Liska 1990). We thus use multilevel regression to test whether neighborhood disadvantage and neighborhood-based parental ties influence adolescent delinquent behavior, controlling for chosen individual-level characteristics (Bernburg and Thorlindsson 2007).

Specifically, Figure 1 presents our hypotheses. First, we expect the level of neighborhood disadvantage and residential instability to have a positive, contextual effect on adolescent delinquency (path A). Second, these neighborhood factors should influence adolescent delinquency in part through neighborhood-based parental ties (paths B and C), and in part through adolescent commitment to social norms (paths D and F). Again, neighborhood-based parental ties may influence adolescent delinquency directly (path C; insofar that ties entail supervision, surveillance, and other forms of external control) as well as indirectly through increased adolescent commitment to social norms (paths E and F).

Figure 1.

Hypotheses.

By focusing on the behavior of adolescents residing in the neighborhood, we aim to capture the “pure” effects of neighborhood social organization on adolescents (Bernburg and Thorlindsson 2007). We may contrast this approach with the common approach of studying the level of delinquency in the neighborhood area, which arguably is less likely to provide a pure test of social disorganization effects (including the effects of weak social ties) since hot spots, available targets, and other opportunity factors may strongly influence area delinquency rates (Cohen and Felson 1979; Weisburd, Morris, and Groff 2009). Moreover, studies that compare area crime rates tend not to distinguish between adult crime rates and crimes committed by adolescents. This study specifically focuses on the latter.

Public Schools as Neighborhood Communities

We define the boundaries of neighborhood communities by using official geographic boundaries of public schools. This method provides meaningful community boundaries in the Icelandic context (Bernburg and Thorlindsson 2005, 2007; Bernburg et al. 2009a, 2009b; Thorlindsson and Bernburg 2004). In Iceland about 85 percent of children and adolescents between ages of 6 and 16 attend small, neighborhood-based public schools. They are selected into the schools based on residence in officially defined geographical areas, regardless of their backgrounds (public education is in principle free of charge). Thus, in contrast to many other countries, most adolescents attending a given school comprise most of the adolescents residing within the geographical boundaries of the school district. Moreover, many social club activities, such as organized sport, are organized based on the school neighborhoods. In short, Icelandic public schools districts comprise small, neighborhood communities within which social ties among parents can serve the purpose of attending to and controlling children and adolescents (Bernburg and Thorlindsson 2007).

External Validity

The majority of community research has focused on urban neighborhoods in the United States and in the United Kingdom. After all social disorganization theory emerged in response to expansion of large cities like Chicago where the development of differentiation between city areas resulted in some areas having high concentration of social problems such as poverty, mental disorders, and youth crime (Shaw and McKay (1942:xi, 17-19).

The Icelandic social setting differs substantially from the setting in which most prior tests of the theory have been conducted. Apart from enormous differences in population size (Iceland has little over 300,000 inhabitants), Iceland is very homogenous relative to most western countries, both racially and culturally (Olafsson and Stefansson 2005). Moreover, Iceland has a relatively strong welfare system and low levels of poverty and economic inequality (Organization for Economic Cooperation and Development [OECD] 2013). Social disorganization theory generally expects these characteristics of the Icelandic context to create denser social ties and stronger normative controls, implying more overall informal social control in Iceland than in urban America or Britain. In particular, the welfare policies in Iceland may serve as a buffer on the harmful effects of the concentrated disadvantage in the community on reduced social control and delinquency. For example, all public schools are equally funded by the government, and thus their quality is not dependent on the socioeconomic status of the neighborhood residents (Bernburg and Thorlindsson 2007).

We contribute to the external validity of social disorganization theory by examining whether it applies in the Icelandic context. Notwithstanding the uniqueness of the Icelandic setting, the theory expects differences that exist across neighborhoods with regard to structural disadvantage and social organization (i.e., neighborhood social ties, commitment to social norms) to influence the neighborhood level of adolescent delinquency. Supporting this point, recent research from rural areas (Kaylen and Pridemore 2013) and countries that have less income inequality than the United States, such as Canada (Thompson and Gartner 2013), Sweden (Sampson and Wikström 2008), and the Netherlands (Steenbeek and Hipp 2011), has supported social disorganization theory.

In the years preceding our data collection, Iceland was becoming more market-oriented (Olafsson 2003), there was an increase in low-income immigrant groups and single-parent households (Statistics Iceland 2014). Although Iceland generally has less variation between individuals, property values differ between neighborhoods in Iceland as they do in other countries (skra.is 2014), resulting in some neighborhoods having a relatively high proportion of disadvantaged groups. Prior Icelandic research in fact indicate that neighborhood characteristics have contextual effects on youth problem behavior (Bernburg and Thorlindsson 2007, 2005; Thorlindsson, Valdimarsdottir, and Jonsson 2012), but it has not examined the effect of concentrated disadvantage on adolescent crime and delinquency.

Method

Data

First, we used a national population survey of students born in 1990 and 1991 (15 and 16 years old at the time of the survey). Anonymous questionnaires were administered to all students attending the compulsory ninth and tenth grade and that were present in class on the day of the survey, in March 2006. The students completed the survey within two class periods and received no instructions other than to answer the questions to the best of their ability. Questionnaires were administered with sealed envelopes by teachers and research assistants (Bernburg and Thorlindsson 2007). Valid questionnaires were obtained from 7,430 respondents, representing approximately 84 percent of the population of the two cohorts. A total of 129 schools (96 percent of all secondary schools in Iceland) participated in the survey. We excluded 46 schools with less than 20 respondents, all located in large and very sparsely populated rural areas and thus do not constitute a meaningful neighborhood. Neighborhood-level measures of the intervening variables were created by aggregating the survey items, and hence we wanted a sufficient number of respondents in each school neighborhood. We also excluded 1,002 respondents who did not attend their local neighborhood school. Our sample thus included 5,865 individuals in 83 neighborhoods. No variable had more than 5 percent of missing cases, so we used listwise deletion when running analysis, resulting in 5,598 to 5,842 cases on individual level. There were no missing values on neighborhood level. The number of respondents from each school ranged from 21 to 286 (M = 71, SD = 48).

Second, data from Statistic Iceland, a center for official statistics that collects registered data of all persons in Iceland, were used to obtain measures of the community characteristics: concentrated disadvantage and residential instability.

The average number of households with children in each neighborhood is 410 (median = 348). The urban neighborhoods have a radius between about .6 and 1.0 miles, with a few large urban neighborhoods having a radius between 1.0 and 2.0 miles (for additional information, see Bernburg et al. 2009a).

Measures

Dependent variable

Delinquency is measured with the average score for six survey items. The respondents were asked how often (if ever) they had done any of the following during the last 12 months: (1) stolen something worth less than 5,000 ISK (about 45 USD), (2) stolen something worth more than 5,000 ISK, (3) used physical violence in order to rob/steal, (4) broken into a building or a car to steal, (5) damaged or vandalized things that did not belong to you, and (6) committed other offences. Cronbach’s α = 0.73.

Independent variables (level 2)

Neighborhood economic status is the mean income of all households with children that reside within the geographical boundaries of the neighborhood, a measure extracted from actual tax reports. The measure is positively skewed and was transformed using the natural logarithm of the original values. A measure for single parents is the proportion of neighborhood households registered as single parent households. Immigrant concentration was defined as the neighborhood proportion of parents that are registered as foreign nationals. The three neighborhood measures of community social structure are moderately to strongly correlated (Pearson’s r from .40 to .58), as is common in community research. Following prior research (Bellair 2000; McNulty and Bellair 2003; Morenoff et al. 2001; Sampson et al. 1997), we used factor analysis to create a summary index, concentrated disadvantage. The contribution of each item to the index was weighted by its factor loading score.

Residential instability was defined as the neighborhood proportion of parents that lived in the neighborhood in the year 2005 but who did not live there in the year 2000. To adjust for skewness, we transformed the measure of residential instability using the natural logarithm of the original value.

Intervening variables—Neighborhood-based parental ties (level 2)

We created measures of neighborhood-based parental ties by using the neighborhood mean score on three survey measures.³ First, following Coleman (1988) and Bernburg and Thorlindsson (2007), we measured intergenerational closure with a four-item, four-point Likert scale. The respondents were asked how well or poorly following statements applied to them: (1) my parents know my friend’s parents, (2) my parents often talk to my friend’s parents, (3) my parents and my friend’s parents sometimes meet to talk, and (4) my parents know my friends. The answers ranged from 1 (applies very poorly to me) to 4 (applies very well to me). Cronbach’s α = 0.82.

Second, parental neighboring is measured with a six-item, five-point Likert-type scale. The respondents were asked whether they agreed or disagreed to the following statements: (1) my parents have friends in the neighborhood, close to our home; (2) my parents know many of our neighbors with name; (3) my parents sometimes visit some of our neighbors; (4) our neighbors sometimes visit my parents; (5) we sometimes borrow things from our neighbors; and (6) our neighbors sometimes borrow things from us. The answers ranged from 1 (disagree completely) to 5 (agree completely). Cronbach’s α = 0.90.

Third, we measure parental participation in the adolescent community with a three-item, five-point Likert-type scale. The respondents were asked how well the following statements applied to their parents: (1) they participate in the parental association at the school I go to, (2) they participate in the sport affiliation that I go to, and (3) they participate in the social club that I go to. The answers ranged from 1 (nothing) to 5 (very much). Some adolescents do not participate in sports or social clubs and therefore these questions do not apply to them. Instead of increasing missing values, we do include these individuals and they therefore get a lower score since their parents miss the opportunity to form bonds to other parents. Cronbach’s α = 0.73. All three scales for parental ties were standardized.

Independent variables (level 1)

Household poverty is measured using four survey items (see Bernburg et al. 2009a): (1) your parents’ financial status is bad; (2) your parents cannot afford to own and operate a car; (3) your parents hardly have enough money to pay for basic necessities, such as food, housing, and phone; and (4) your parents cannot afford the type of leisure activity that you would most prefer (e.g., music or sports). The responses ranged from 1 (almost never) to 5 (almost always). We combined the four scales into an index using the mean score. Cronbach’s α = 0.77.

Family disruption is measured using a dummy variable coded “1” if the respondent lived with both biological parents only and “0” otherwise. Residential instability on individual level was coded “1” if the respondent answered yes to the question “have you moved to a new neighborhood in the last 12 months?” Foreign born parent is measured using a dummy variable coded “1” if the respondent replied no to either of the questions “was your mother born in Iceland” and “was your father born in Iceland”, and “0” otherwise.

Intervening variables—Commitment to social norms (level 1)

Commitment to social norms is measured with two main constructs, normlessness and conduct norms. Normlessness is measured with eight Likert score items (Thorlindsson and Bernburg 2004). The respondents were asked if they agreed or disagreed to the following statements: (1) you can break most rules if they don’t seem to apply; (2) I follow whatever rules I want to follow; (3) in fact, there are very few absolute rules in life; (4) it’s hard to trust anything because everything is so variable; (5) in fact, nobody knows what is expected of her or him; (6) one can never be certain of anything in this life; (7) you have to break the rules to achieve what you want; and (8) following the rules does not guarantee success. The answers ranged from 1 (agree completely) to 5 (disagree completely).

We measured conduct norms with answers to five survey items. Respondents were asked if it was a good thing for people to (1) forgive those who treat you badly, (2) respond kindly when someone gets angry at you, and (3) being nice to others even when they do something that you do not approve of (taken from Felson et al. 1994). The answers ranged from 1 (very good) to 4 (very bad; these items were reversed, so that all 13 survey items would reflect strong commitment to social norms). Additionally, we used two questions from Bernburg and Thorlindsson (2005): (4) sometimes there are situations that justify people being beaten up or hit and (5) when someone treats me badly it is ok to beat him or her up or hit him or her. The response categories ranged from 1 (agree completely) to 4 (disagree completely).

After reversing the first three questions measuring conduct norms, the z scores of all 13 survey items were combined into an index for commitment to social norms, Cronbach’s α = 0.84.

Controls

We control for the respondents’ gender on individual level with a dummy variable coded “1” for females and “0” for males. A control variable for urban location is also included in the analyses. School neighborhoods that are located in the larger Reykjavik capital area have more dense population than other neighborhoods. The concept of population density in Iceland differs from larger cities in the United States and other countries. Reykjavik, the capital and its surrounding towns, are 1.03 percent of the area (1.062 km²) of Iceland but home to 64 percent (205,675) of the total population of the country. Accordingly, we designate school neighborhoods in the capital area as “urban” as a proxy for population density (coded “1” for urban and “0” for rural areas). Table 1 shows descriptive statistics for all variables.

Table 1.

Descriptive Statistics.

	Mean	Standard deviation	Min	Max
Dependent variable
Delinquency	1.20	.48	1	7
Independent variables (level 2), N level 2 = 83
Concentrated disadvantage	0	1	−1.86	4.26
Residential instability logged	−.90	.38	−1.91	0
Intervening variables (level 2), Neighborhood-based parental ties
Intergenerational closure	0	1	−1.73	3.01
Parental neighboring	0	1	−2.15	2.53
Parental participation	0	1	−2.01	2.99
Independent variables (level 1), N level 1 = 5,865
Household poverty	1.33	.59	1	5
Both biological parents (vs. family disruption)	.72	.45	0	1
Moved in last 12 months	.08	.27	0	1
Foreign-born parents	.02	.12	0	1
Intervening variable (level 1)
Commitment to social norms	.01	.55	−2.83	1.60
Controls
Urban area	.57	.50	0	1
Female	.51	.50	0	1

Statistical Analysis

We use ordinary least squares (OLS) regression to examine the aggregate effects of neighborhood disadvantage and residential instability on parental ties (path B in Figure 1). We examine all other effects using hierarchical regression models⁴ (HLM 7; Bryk and Raudenbush 1992), which enable us to examine individual and neighborhood-level effects simultaneously. In our multilevel models, only the intercept is allowed to vary randomly on level 2 but the slopes are assumed to be fixed.

We ran an intercept only model (without any predictor variables) to determine if variability of adolescent delinquent behavior and commitment to social norms can be attributed to neighborhoods. The fully reduced model showed a significant intercommunity variation in delinquency (χ² = 188,546 df = 82, p < .001, intraclass correlation [ICC] = 0.015) and commitment to social norms (χ² = 167,622 df = 82, p < .001, ICC = 0.016). We used grand mean centering for all continuous variables.⁵

Results

The findings in Table 2 show that neighborhoods characterized by high levels of concentrated disadvantage have significantly lower levels of parental intergenerational closure (b = −.200), parental neighboring (b = −.194), and parental participation (b = −336). The coefficients for these effects can be interpreted as if they were standardized, since the variables are standardized. The standardized effects indicate moderately strong associations. For example, a standard deviation increase in neighborhood concentrated disadvantage is associated with .34 standard deviation decrease in parental participation in the adolescent community. As hypothesized, the results show that residential instability in the school neighborhood is associated with weaker neighborhood-based parental ties, but the effect is significant only for parental neighboring (b = −.790).

Table 2.

OLS Regression Estimates of Models Predicting Neighborhood-based Parental Ties (Path B).

Dependent variables→	Intergenerational closure	Parental neighboring	Parental participation
Independent variables	Model 1	Model 2	Model 3
Independent variables	b (SE)	b (SE)	b (SE)
Intercept	.280 (.406)	−.277 (.369)	.278 (358)
Concentrated disadvantage	−.200 (.096)*	−.194 (.087)*	−.336 (.085)**
Residential instability (logged)	−.265 (.323)	−.790 (.293)**	−.327 (.285)
Urban area	−.895 (.244)**	−.749 (.244)**	−.989 (.215)**
R ²; F-value	.31; 12.068**	.43; 20.126**	.47; 17.42**

Note: N = 83.

b (SE) = unstandardized coefficient (standard error); OLS = ordinary least squares.

*p < .05. **p < .01.

Although not our main focus, the results in Table 2 show that all measures of parental ties are significantly weaker in urban neighborhoods than in rural neighborhoods, net of concentration of disadvantage and residential instability. The dummy variable for urban areas is essentially an index for population density and thus our findings are in line with prior studies showing that ties between neighbors tend to be weaker in more dense areas (e.g., Sampson and Groves 1989; see also discussion in Krohn 1986).

In Table 3, we focus on paths D and E in Figure 1. The findings in model 1 lend partial support for the hypothesis implied by path D, showing a significant negative relationship between concentrated disadvantage and adolescent commitment to social norms (b = −.018). Importantly, this effect is a contextual effect, that is, regardless of whether an adolescent has poor parents, lives with both his or her parents, has recently moved to a new neighborhood, or has foreign parents, living in a disadvantaged neighborhood is associated with lower commitment to social norms.

Table 3.

HLM Regression Estimates of Models Predicting Commitment to Social Norms (Paths D and E).

Level 2, N = 83	Model 1	Model 2	Model 3	Model 4
Level 2, N = 83	b (SE)	b (SE)	b (SE)	b (SE)
Intercept	−.144 (.024)**	−.156 (.025)**	−.148 (.023)**	−.166 (.024)**
Concentrated disadvantage	−.018 (.009)*	−.012 (.008)	−.015 (.008)	−.006 (.008)
Residential instability (logged)	.121 (.034)**	.130 (.034)**	.135 (.038)**	.124 (.033)**
Urban area	−.014 (.024)	.010 (.025)	.003 (.026)	.026 (.032)
Neighborhood-based parental ties
Intergenerational closure	—	.031 (.010)**	—	—
Parental neighboring	—	—	.017 (.014)	—
Parental participation	—	—	—	.038 (.012)**
Level 1, N = 5598
Household poverty	−.117 (.016)**	−.117 (.016)**	−.117 (.016)**	−.117 (.016)**
Both biological parents	.036 (.014)*	.035 (.014)*	.035 (.014)*	.035 (.015)*
Moved in last 12 months	−.042 (.023)	−.040 (.023)	−.041 (.023)	−.040 (.023)
Foreign-born parents	−.284 (.067)**	−.284 (.067)**	−.284 (.067)**	−.283 (.067)**
Female	.295 (.017)**	.294 (.017)**	.294 (.017)**	.295 (.017)**
χ²	146.476**	134.188**	144.607**	131.947**

Note: b (SE) = unstandardized coefficient (standard error). Population-average model with robust standard errors.

*p < .05. **p < .01.

The remaining models in Table 3 examine whether neighborhood-based parental ties mediate some part of the effect of concentrated disadvantage on commitment to social norms (see paths B and E in Figure 1). The results lend substantial support for this hypothesis. Thus, intergenerational closure mediates about 33 percent of the effect of neighborhood concentrated disadvantage on adolescent commitment to social norms (model 2), and parental participation mediates about 67 percent of the effect (model 4). Moreover, both these intermediate factors have significant, positive, contextual effects on individual commitment to social norms. But, parental neighboring (interaction between parents and neighbors) does not emerge as a significant predictor of adolescent commitment to social norms (model 3).

The contextual effect of neighborhood-level residential instability on commitment to social norms is in the opposite direction from what we have hypothesized (b = .121), across all the four models. Thus, adolescents living in neighborhoods that have high residential instability actually tend to have a stronger commitment to social norms than those living in more stable communities, net of controls.

Table 4 examines the effects of all independent variables on adolescent delinquency (paths A, C, and F in Figure 1). First, model 1 supports the key hypothesis concerning the contextual effect of neighborhood concentrated disadvantage on delinquency, showing a significant and positive effect (b = .027). Thus, independent of their personal household situations, adolescents living in neighborhoods characterized by many disadvantaged families report more delinquency. But, neither neighborhood residential instability nor urban location exhibits any significant effects on delinquency. The individual-level relationships are consistent with prior research (Bernburg et al. 2009b); both household poverty and not living with both parents are associated with increased delinquency. But, having moved recently and having foreign-born parents is not associated with delinquency.

Table 4.

HLM Regression Estimates of Models Predicting Self-reported Delinquency (Paths A, C, and F).

Level 2, N = 83	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7
Level 2, N = 83	b (SE)	b (SE)	b (SE)	b (SE)	b (SE)	b (SE)	b (SE)
Intercept	1.317 (.024)**	1.327 (.024)**	1.287 (.021)**	1.319 (.025)**	1.280 (.021)**	1.319 (.024)**	1.27 (.022)**
Concentrated disadvantage	.027 (.008)**	.023 (.008)**	.017 (.006)**	.024 (.009)**	.020 (.008)**	.026 (.010)*	.025 (.008)**
Residential instability (logged)	−.040 (.033)	−.057 (.036)	−.032 (.031)	−.047 (.034)	−.014 (.031)	−.040 (.033)	−.020 (.030)
Urban area	.010 (.023)	−.006 (.024)	−.000 (.021)	.005 (.026)	−.004 (.011)	.009 (.027)	.017 (.024)
Neighborhood-based parental ties
Intergenerational closure	—	−.027 (.013)*	−.020 (.012)	—	—	—	—
Parental neighboring	—	—	—	−.009 (.013)	−.008 (.011)	—	—
Parental participation	—	—	—	—	—	−.003 (.013)	−.006 (.012)
Level 1, N = 5,482
Household poverty	.087 (.008)**	.086 (.019)**	.059 (.018)**	.087 (.019)**	.060 (.018)**	.087 (.007)**	.028 (.007)**
Both biological parents	−.060 (.016)**	−.060 (.016)**	−.051 (.015)**	−.060 (.016)**	−.051 (.015)**	−.060 (.016)**	−.051 (.015)**
Moved in last 12 months	.060 (.032)	.059 (.032)	.050 (.032)	.060 (.032)	.050 (.032)	.061 (.032)	.051 (.032)
Foreign born parents	.038 (.112)	.037 (.111)	−.027 (.103)	.038 (.111)	−.027 (.103)	.038 (.111)	−.027 (.102)
Female	−.176 (.014)**	−.176 (.014)**	−.108 (.013)**	−.176 (.014)**	−.108 (.013)**	−.176 (.014)**	−.107 (.013)**
Commitment to social norms	—	—	−.232 (.020)**	—	−.232 (.020)**	—	−.232 (.020)**
χ²	155.929**	146.734**	128.719**	155.726**	135.834**	155.589**	134.117**

Note: b (SE) = unstandardized coefficient (standard error). Population-average model with robust standard errors. HLM = hierarchical regression models.

*p < .05. **p < .01.

Models 2 through 7 examine the role of neighborhood-based parental ties and adolescent commitment to social norms in mediating the contextual effect of neighborhood concentrated disadvantage on delinquency. These models add one parental ties measure at a time and then include commitment to social norms to examine its mediating role. The results indicate that intergenerational closure and commitment to norms mediate a substantial part (about 37 percent) of the contextual effect of neighborhood concentrated disadvantage on delinquency, while the other two measures for parental ties do not significantly influence delinquency (model 4 through model 7).

Thus, focusing on the supportive findings, a comparison of model 1 and model 2 shows that intergenerational closure mediates about 15 percent of the contextual effect of neighborhood concentrated disadvantage on delinquency (the effect decreases from .027 to .023). Further, comparison of model 2 and model 3 shows that commitment to social norms mediates 26 percent of the remaining effect (the effect decreases from .023 to .017). Finally, a comparison of model 2 and model 3 reveals that commitment to social norms mediates 25 percent of the effect of intergenerational closure on delinquency, implying that intergenerational closure deters delinquency not only through increased external control (supervision, surveillance, and scrutiny; see path C in Figure 1) but also through internal control (commitment to norms; see path E through F).

In sum, the findings lend partial support for the hypotheses presented in Figure 1. Thus, they indicate that adolescents in disadvantaged neighborhoods report more delinquency (net of their household situations), in part, because such neighborhoods have both weaker intergenerational closure and reduced adolescent commitment to social norms. However, parental neighboring and parental participation do not emerge as significant mediators, although parental participation mediates some of the effect of neighborhood concentrated disadvantage on adolescent commitment to norms.

Discussion

Social disorganization theory implies that community structure influences crime and delinquency through its effect on social ties among community residents and their commitments to social norms (Bursik 1988; Bursik and Grasmick 1993; Kornhauser 1978; Sampson and Groves 1989). We obtained multilevel data on self-reported delinquent behavior and household situations of individual adolescents, as well as on the structural characteristics of the neighborhoods to which they belong, enabling us to examine the contextual effect of neighborhood structure on adolescent delinquency. The richness of these data has allowed us to test whether neighborhood parental ties (rather than ties among general residents) and adolescent commitment to social norms mediate the contextual effect of neighborhood structure on adolescent delinquency. Moreover, using Icelandic school neighborhoods has enabled us to rely on meaningful neighborhood boundaries, especially since our analysis is focused on the social world of middle adolescence (as opposed to, say, young adults who’s social world may be further reduced from the neighborhood in which they reside in). Many community studies ignore age distinctions in community effects on delinquency, as they rarely disaggregate delinquency rates by age nor focus on particular age groups (exceptions include Bernburg and Thorlindsson 2007; Elliott et al. 1996; Oberwittler 2004). Unlike the majority of community studies, we used self-reported delinquency, avoiding police data bias (Baumer 2002; Kirk 2006; Maguire 2007; see also discussion in Bursik 1998; and review in Goudriaan, Witterbrood, and Nieuwbeerta 2006).

Underscoring the salience of the community approach in the Icelandic context, and supporting the external validity of prior research, concentrated neighborhood disadvantage has a substantial, contextual effect on delinquency in Iceland. Adolescents belonging to neighborhoods that are high in concentrated disadvantage (i.e., have a high proportion of low-income families, single-parent households, and immigrant parents) report more delinquency, net of their personal family and economic situations.

However, we find no significant effect of neighborhood residential instability on delinquency. The effect of residential instability on delinquency has been inconsistent in prior work (see Bellair 1997, 2000; this also is the case in Iceland; Bernburg and Thorlindsson 2007 found a positive effect of residential instability on delinquency). Also, we find an unexpected positive effect of residential instability on adolescent commitment to social norms, although residential instability has the expected effect on parental neighboring (but not on intergenerational closure or parental participation in the community). In this respect, our data from Iceland in 2006 do not show the positive relationship usually observed between concentrated disadvantage and residential instability (Pearson’s r = .19 n.s.). Perhaps the reason is that in the years leading to 2006, there was an economic uptrend in Iceland that may have changed this relationship. During the economic uptrend, many middle- and higher-income families bought new houses and moved to new neighborhoods. In fact, our measure of residential instability only measures immigration to the neighborhood as instability but does not tackle emigration. Neighborhoods that had relatively many new residents (immigration) at that time thus had high instability rates but neighborhoods that had many residents move out of the neighborhood (emigration) did not. More research is needed to study the role of residential instability in delinquency in Iceland.

We have argued that neighborhood-based parental ties create an important source of social control over adolescents (Bernburg and Thorlindsson 2007; Coleman 1988; Thorlindsson et al. 2007). Accordingly, we have predicted that disadvantaged neighborhoods have higher levels of adolescent delinquency because they have (1) weaker ties among parents in the community and (2) less adolescent commitment to social norms. Moreover, (3) weaker neighborhood parental ties should increase delinquency in part through weaker adolescent commitment to social norms. The findings lend some support for all these intermediate processes. Thus, they show that intergenerational closure, that is, social ties linking parents and adolescents in the same community, and adolescent commitment to social norms both mediate a part of the contextual effect of neighborhood concentrated disadvantage on adolescent delinquency. Moreover, commitment to social norms mediates a part of the effect of intergenerational closure on delinquency, thus implying that intergenerational closure deters delinquency through both external control (supervision, surveillance, and scrutiny) and internal control (commitment to norms).

The findings underscore that families that live in neighborhoods where parents are not only well connected to each other, but willing to work together and participate in the children’s school and leisure activities, are able to mutually support each other in their efforts to influence youth problem behavior (Bernburg and Thorlindsson 2007; Coleman 1988). Intergenerational closure appears to be particularly important in this respect. Also, while it does not directly influence delinquency, parental participation in community affairs involving adolescents (e.g., school activities or sport clubs) positively influences adolescent commitment to social norms. In contrast, parental neighboring or parental ties to neighbors that do not extend to other parents or the adolescent community does not generate adolescent control in our study. Thus, external (i.e., supervision and surveillance) and internal social control (commitments to social norms) become more effective when many parents in the neighborhood are involved in their children’s lives. This further supports the knowledge that has been cumulating in neighborhood studies for the last decade or so, that social ties between neighbors may not be sufficient to create informal social control. The distinctive factor in the capacity of ties to produce social control seems to be the purpose of the ties (see Kubrin and Weitzer 2003; Sampson et al. 1999).

Our study has limitations. Although our findings are mostly consistent with the assumed theoretical pathways, we have not been able to separate the constructs temporally with our cross-sectional data. Prior research has shown that community effects on crime and delinquency hold while controlling for prior community rates of delinquency (Bernburg and Thorlindsson 2007; Sampson et al. 1997). Still, youth delinquency could theoretically both increase and decrease neighborhood-based ties among parents. On one hand, crime and delinquency may increase community ties and cohesion (Durkheim 1984). For example, when discovering that their child has been involved in crime, parents might reach out to other parents to try to prevent further delinquency. On the other hand, crime can increase fear and thus decrease interactions and participation in the neighborhood (Liska and Warner 1991). Although there are some indications that serious crime increases fear and subsequently decreases neighborhood social ties (Liska and Warner 1991; Sampson and Raudenbush 1999), these relationships do need more empirical work to further improve our understanding of these processes (for further discussion on nonrecursive models, see Bellair 2000; Markowitz et al. 2001; Steenbeek and Hipp 2011).

An important contribution of our study has been to test social disorganization theory in a different societal context from most prior studies. Unlike what is seen in many other countries, almost all children in Iceland attend public schools that are equally funded by the state and use the same official curriculum. The schools in Iceland are not dependent on the socioeconomic status of the neighborhood residents (Bernburg and Thorlindsson 2007). We were thus better able to isolate the effects of community disadvantage and informal social control on adolescent delinquency than many prior studies.

Our study indicates that social disorganization research can be generalized to a small homogeneous country with a developed welfare system. Because of a relatively strong welfare system, Iceland is not only ethnically and culturally homogeneous but has relatively low levels of poverty and income inequality. Nevertheless, our findings support one of the major propositions of social disorganization theory, namely that neighborhood disadvantage increases crime and delinquency by weakening neighborhood-based social ties and normative commitments.

Cross-societal comparisons have important implications for theory advancement because they give us information if the proposed processes work differently under different cultural contexts. Although Iceland is in many important respects different from cities within the United States, these western cultures still share some common characteristics, such as individualism, independence, and a moderate support for authorities (Olafsson 2003). The field would possibly benefit from testing social disorganization theory in non-Western cultures. Some countries within Asia or Latin America have high levels of poverty but have cultural characteristics that may influence informal social control, such as collective orientations and strong extended family ties.

Footnotes

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.

Notes

References

Anderson

Elijah

. 1999. The Code of the Streets: Decency, Violence and the Moral Life of the Inner City. New York: W.W. Norton & Company.

Baumer

Eric P.

2002. “Neighborhood Disadvantage and Police Notification by Victims of Violence.” Criminology 40:579–616.

Bellair

Paul E.

1997. “Social Interaction and Community Crime: Examining the Importance of Neighbor Networks.” Criminology 35:677–704.

Bellair

Paul E.

2000. “Informal Surveillance and Street Crime: A Complex Relationship.” Criminology 38:137–70.

Bellair

Paul E.

Browning

Christopher R.

2010. “Contemporary Disorganization Research: An Assessment and Further Test of the Systemic Model of Neighborhood Crime.” Journal of Research in Crime and Delinquency 47:496–521.

Berg

Mark T.

Steward

Eric A.

Schreck

Christopher J.

Simons

Ronald L.

2012. “The Victim-offender Overlap in Context: Examining the Role of Neighborhood Street Culture.” Criminology 50:359–89.

Bernburg

Jon Gunnar

Thorlindsson

Thorolfur

. 2005. “Violent Values, Conduct Norms, and Youth Aggression: A Multi-level Study in Iceland.” The Sociological Quarterly 46:455–76.

Bernburg

Jon Gunnar

Thorlindsson

Thorolfur

. 2007. “Community Structure and Adolescent Delinquency: A Contextual Analysis.” Criminology 45:415–44.

Bernburg

Jon Gunnar

Thorlindsson

Thorolfur

Sigfúsdóttir

Inga Dóra

. 2009a. “Relative Deprivation and Adolescent Outcomes in Iceland: A Multilevel Test.” Social Forces 3:1223–50.

10.

Bernburg

Jon Gunnar

Thorlindsson

Thorolfur

Sigfúsdóttir

Inga Dora

. 2009b. “The Neighborhood Effects of Disrupted Family Processes on Adolescent Substance Use.” Social Sience and Medicine 69:129–37.

11.

Bjarnason

Thoroddur

Thorlindsson

Thorolfur

Sigfusdottir

Inga Dora

Welch

Michael R.

2005. “Familial and Religious Influences on Adolescent Alcohol Use: A Multi-level Study of Students and School Communities.” Social Forces 84:375–90.

12.

Browning

Christopher R.

Feinberg

Seith L.

Dietz

Robert D.

2004. “The Paradox of Social Organization: Network, Collective Efficacy, and Violent Crime in Urban Neighborhoods.” Social Forces 83:503–34.

13.

Bryk

Anthony S.

Raudenbush

Stephen W.

1992. Hierarchical Linear Models: Applications and Data Analysis Methods. Newbury Park, CA: Sage.

14.

Bursik

Robert J.

Jr . 1988. “Social Disorganization and Theories of Crime and Delinquency.” Criminology 26:519–52.

15.

Bursik

Robert J.

Jr Grasmick

Harold G.

1993. Neighborhood and Crime: The Dimensions of Effective Community Control. New York: Lexington Books.

16.

Cantillon

Dan

Davidson

William S.

Schweitzer

John H.

2003. “Measuring Community Social Organization: Sense of Community as a Mediator in Social Disorganization Theory.” Journal of Criminal Justice 31:321–39.

17.

Cohen

Lawrence E.

Felson

Marcus

. 1979. “Social Change and Crime Rate Trends: A Routine Activity Approach.” American Sociological Review 44:488–608.

18.

Coleman

James S.

1988. “Social Capital in the Creation of Human Capital.” American Journal of Sociology 94:95–120.

19.

Durkheim

Emile

. 1984. The Division of Labor in Society. Translated by George Simpson. New York: Free Press.

20.

Elliott

Delbert S.

Wilson

Wiliam J.

Huizinga

David

Sampson

Robert J.

Elliott

Amanda

Rankin

Bruce

. 1996. “The Effects of Neighborhood Disadvantage on Adolescent Development.” Journal of Research in Crime and Delinquency 33:389–426.

21.

Enders

Craig K.

Tofighi

Davood

. 2007. “Centering Predictor Variables in Cross-sectional Multilevel Models: A New Look at an Old Issue.” Psychological Methods 12:121–38.

22.

Felson

Richard B.

Liska

Allen F.

South

Scott J.

McNulty

Thomas L.

1994. “The Subculture of Violence and Delinquency: Individual vs. School Context Effect.” Social Forces 73:155–73.

23.

Garson

David G.

2013. Hierarchical Linear Modeling: Guide and Applications. Thousand Oaks, CA: Sage.

24.

Goudriaan

Heike

Witterbrood

Karin

Nieuwbeerta

Paul

. 2006. “Neighborhood Characteristics and Reporting Crime: Effects of Social Cohesion, Confidence in Police Effectiveness and Socio-economic Disadvantage.” British Journal of Criminology 46:719–42.

25.

Hawdon

James E.

Ryan

John

. 2009. “Social Capitla, Social Control and Changes in Victimization Rates.” Crime and Delinquency 55:526–49.

26.

Kasarda

John D.

Janowitz

Morris

. 1974. “Community Attachment in Mass Society.” American Sociological Review 39:328–39.

27.

Kaylen

Maria T.

Pridemore

William A.

2013. “Social Disorganization and Crime in Rural Communities.” British Journal of Criminology 53:905–23.

28.

Kingston

Beverly

Huizinga

David

Elliott

Delbert S.

2009. “A Test of Social Disorganization Theory in High Risk Neighborhoods.” Youth & Society 41:53–79.

29.

Kirk

David S.

2006. “Examining the Divergence across Self-report and Official Data Sources on Inferences about the Adolescent Life-course of Crime.” Journal of Quantitative Criminology 22:107–29.

30.

Kornhauser

Ruth R.

1978. Social Sources of Delinquency: An Appraisal of Analytic Models. Chicago, IL: University of Chicago Press.

31.

Krohn

Marvin D.

1986. “The Web of Conformity: A Network Approach to the Explanation of Delinquent Behavior.” Social Problems 33:81–93.

32.

Krull

Jennifer L.

MacKinnon

David P.

1999. “Multilevel Mediation Modeling in Group-based Intervention Studies.” Evaluation Review 23:418–44.

33.

Krull

Jennifer L.

MacKinnon

David P.

2001. “Multilevel Modeling of Individual and Group Level Mediated Effects.” Multivariate Behavioral Research 36:249–77.

34.

Kubrin

Charis E.

Weitzer

Ronald

. 2003. “New Directions in Social Disorganization Theory.” Journal of Research in Crime and Delinquency 40:374–402.

35.

Liska

Allen E.

1990. “The Significance of Aggregate Dependent Variables and Contextual Independent Variables for Linking Macro and Micro Theories.” Social Psychology Quarterly 53:292–301.

36.

Liska

Allen E.

Warner

Barbara D.

1991. “Functions of Crime: A Paradoxical Process.” American Journal of Sociology 96:1441–63.

37.

Lowenkamp

Christopher T.

Cullen

Francis T.

Pratt

Travis C.

2003. “Replicating Sampsons and Groves’s Test of Social Disorganization Theory: Revisiting a Criminological Classic.” Journal of Research in Crime and Delinquency 40:351–73.

38.

Lucas

Jeffrey W.

2003. “Theory-testing, Generalization, and the Problem of External Validity.” Sociological Theory 21:236–53.

39.

Maguire

Mike

. 2007. “Crime Data and Statistics.” Pp. 241–302 in The Oxford Handbook of Criminology, edited by Maguire

Mike

Morgan

Rod

Reiner

Robert

. Oxford, UK: Oxford University Press.

40.

Markowitz

Fred E.

Bellair

Paul E.

Liska

Allen E.

Liu

Jianhong

. 2001. “Extending Social Disorganization Theory: Modeling the Relationships between Cohesion, Disorder, and Fear.” Criminology 39:293–319.

41.

Mazerolle

Wickes

McBroom

“Community Variations in Violence: The Role of Social Ties and Collective Efficacy in Comparative Context.”

Journal of Research in Crime and Delinquency 47:3–30.

42.

McNulty

Thomas L.

Bellair

Paul

. 2003. “Explaining Racial and Ethnic Differences in Adolescent Violence: Structural Disadvantage, Family Well-being, and Social Capital.” Justice Quarterly 20:1–31.

43.

Morenoff

Jeffrey D.

Sampson

Robert J.

Raudenbush

Stephen W.

2001. “Neighborhood Inequality, Collective Efficacy, and the Spatial Dynamics of Urban Violence.” Criminology 39:517–58.

44.

Oberwittler

Dietrich

. 2004. “A Multilevel Analysis of Neighborhood Contextual Effects on Serious Juvenile Offending: The Role of Subcultural Values and Social Disorganization.” European Journal of Criminology 1:201–35.

45.

OECD (Organization for Economic Cooperation and Development). 2013. Income Distribution and Poverty. Accessed November 15 2013, http://stats.oecd.org/Index.aspx?DatasetCode=IDD.

46.

Olafsson

Stefan

. 2003. “The Identity of Icelanders—Scandinavian or American?” Scandinavian Review 91:6–14.

47.

Olafsson

Stefan

Stefansson

Kolbeinn

. 2005. Hnattvaeding og thekkingarthodfelag. Reykjavik, Iceland: Haskolautgafan.

48.

Patillo

Mary E.

1998. “Sweet Mothers and Gangbangers: Managing Crime in a Black Middle-class Neighborhood.” Social Forces 74:747–74.

49.

Preacher

Kristopher J.

Zhang

Zhen

Zyphur

Michal J.

2011. “Alternative Methods for Assessing Mediation in Multilevel Data: The Advantages of Multilevel SEM.” Structural Equation Modeling 18:161–82.

50.

Preacher

Kristopher J.

Zyphur

Michael J.

Zhang

Zhen

. 2010. “A General Multilevel SEM Framework for Assessing Multilevel Mediation.” Psychological Methods 15:209–33.

51.

Putnam

Robert D.

2000. Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster.

52.

Rountree

Pamela W.

Warner

Barbara D.

1999. “Social Ties and Crime: Is the Relationship Gendered?” Criminology 37:789–814.

53.

Sampson

Robert J.

Bartusch

Dawn Jeglum

. 1998. “Legal Cynicism and (Subcultural?) Tolerance of Deviance: The Neighborhood Context of Racial Differences.” Law and Society Review 32:777–804.

54.

Sampson

Robert J.

Groves

Byron W.

1989. “Community Structure and Crime: Testing Social Disorganization Theory.” American Journal of Sociology 94:774–802.

55.

Sampson

Robert J.

Morenoff

Jeffrey D.

Earls

Felton

. 1999. “Beyond Social Capital: Spatial Dynamics of Collective Efficacy for Children.” American Sociological Review 64:633–60.

56.

Sampson

Robert J.

Morenoff

Jeffrey D.

Gannon-Rowley

Thomas

. 2002. “Assessing ‘Neighborhood Effects’: Social Processes and New Direction in Research.” Annual Review of Sociology 28:443–78.

57.

Sampson

Robert J.

Raudenbush

Stephen W.

1999. “Systematic Social Observation of Public Spaces: A New Look at Disorder in Urban Neighborhoods.” American Journal of Sociology 105:603–51.

58.

Sampson

Robert J.

Raudenbush

Stephen W.

Earls

Felton

. 1997. “Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy.” Science 277:918–24.

59.

Sampson

Robert J.

Wikström

Per-Olof H.

2008. “The Social Order of Violence in Chicago and Stockholm Neighborhoods: A Comparative Inquire.” Pp. 97–119 in Order, Conflict, and Violence, edited by Kalyvas

S. N.

Shapiro

Masoud

New York: Cambridge University Press.

60.

Shaw

Clifford R.

McKay

Henry D.

1942. Juvenile Delinquency and Urban Areas: A Study of Rates of Delinquency Characteristics of Local Communities in American Cities. Chicago, IL: University of Chicago Press.

61.

Simcha-Fagan

Schwartz

Joseph E.

1986. “Neighborhood and Delinquency: An Assessment of Contextual Effects.” Criminology 24:667–99.

62.

Skra.is. 2014. “Registers Iceland.” Accessed April 16 2014, http://www.skra.is/markadurinn/talnaefni/.

63.

Statistics Iceland. 2014. “Population Statistics.” Accessed January 10 2014, http://www.statice.is/Statistics/Population.

64.

Steenbeek

Wouter

Hipp

John R.

2011. “A Longitudinal Test of Social Disorganization Theory: Feedback Effects among Cohesion, Social Control, and Disorder.” Criminology 49:833–71.

65.

Stewart

Eric A

Simons

Ronald L.

2006. “Structure and Culture in African-American Adolescent Violence: A Partial Test of the Code of the Street Thesis.” Justice Quarterly 23:1–33.

66.

Stewart

Eric A.

Simons

Ronald L.

2010. “Race, Code of the Street, and Violent Delinquency: A Multilevel Investigation of Neighborhood Street Culture and Individual Norms of Violence.” Criminology 48:569–605.

67.

Stewart

Eric A.

Simons

Ronald L.

Conger

Rand D.

2002. “Assessing Neighborhood and Social Psychological Influences on Childhood Violence in an African American Sample.” Criminology 40:801–30.

68.

Sun

Ivan

Triplett

Ruth

Gainey

Randy

. 2004. “Social Disorganization, Legitimacy of Local Institutions and Neighborhood Crime: An Exploratory Study of Perceptions of the Police and Local Government.” Journal of Crime and Justice 27:33–60.

69.

Thompson

Sara K.

Gartner

Rosemary

. 2013. “The Spatial Distribution and Social Context of Homicide in Toronto’s Neighborhoods.” Journal of Research in Crime and Delinquency 51:88–117.

70.

Thorlindsson

Thorolfur

Bernburg

Jón Gunnar

. 2004. “Durkheim’s Theory of Social Order and Deviance: A Multi-level Test.” European Sociological Review 20:271–85.

71.

Thorlindsson

Thorolfur

Bjarnason

Thoroddur

Sigfúsdóttir

Inga Dóra

. 2007. “Individual and Community Processes of Social Closure: A Study of Adolescent Academic Achievement and Alcohol Use.” Acta Sociologica 50:161–78.

72.

Thorlindsson

Thorolfur

Valdimarsdottir

Margret

Jonsson

Stefan H.

2012. “Community Social Structure, Social Capital, and Adolescent Smoking: A Multi-level Analysis.” Health & Place 18:796–804.

73.

Veysey

Bonita M.

Messner

Steven F.

1999. “Further Testing of Social Disorganization Theory: An Elaboration of Sampson and Groves’s “Community Structure and Crime.” Journal of Research in Crime and Delinquency 36:156–74.

74.

Warner

Barbara D.

2003. “The Role of Attenuated Culture in Social Disorganization Theory.” Criminology 41:73–98.

75.

Warner

Barbara D.

Burchfield

Keri

. 2011. “Misperceived Neighborhood Values and Informal Social Control.” Justice Quarterly 28:606–30.

76.

Warner

Barbara D.

Rountree

Pamela W.

1997. “Local Social Ties in a Community and Crime Model: Questioning the Systemic Nature of Informal Social Control.” Social Problems 44:520–36.

77.

Weisburd

David

Morris

Nancy A.

Groff

Elizabeth R.

2009. “Hot Spots of Juvenile Crime: A Longitudinal Study of Arrest Incidents at Street Segments in Seattle, Washington.” Journal of Quantitative Criminology 25:443–67.

78.

Wilson

William J.

2008. “Jobless Poverty: A New Form of Social Dislocation in the Inner-city Ghetto.” Pp. 340–49 in Social Stratification: Class, Race, and Gender in Sociological Perspective, edited by Grusky

David D.

Boulder, CO: Westview Press.

79.

Wolfgang

Marvin E.

Ferracuti

Franco

. 1967. The Subculture of Violence: Towards an Integrated Theory in Criminology. London, UK: Travistock Publications.