Social Disorganization Theory
One theory that has benefited from macro-micro theoretical integration is Social
Disorganization Theory. Shaw & McKay (1942; 1969) originally proposed that crime resulted
from the intersection of macro-social factors (poverty, racial heterogeneity, and social
mobility) and micro-social factors (informal social control). Previous studies that have
examined the impact of Social Disorganization variables on crime have found mixed support
for the theory; however, these tests were conducted only at the macro-level (Bursik &
Webb, 1982; Heitgerd & Bursik, 1987). While Social Disorganization Theory held a prominent
spot in Criminology for nearly four decades, the theory fell out of favor in the 1960s largely
as the result of the shift in theoretical (and research) attention to micro-level (individual)
explanations of crime (Bohm, 2001). The work of Robert Sampson (Sampson & Groves,
1989; Sampson, 1991; Sampson et al., 1997) and Robert Bursik (1988, 2000) has revitalized
the theory through the inclusion of micro-level concepts. In what could be considered a
macro-micro theoretical integration of Social Disorganization, Sampson (1991; Sampson et
al., 1997) introduces the concept of collective efficacy as a micro-level variable that
mediates the relationship between the structural context of a community and crime.
Subsequent tests of macro-micro Social Disorganization Theory have found support for
Sampson and colleagues’ arguments that social disorganization erodes levels of collective
efficacy, which in turn increases the probability of crime and delinquency among residents
within the neighborhood (Browning, 2002; Rountree, Land & Miethe, 1994; Sampson et al.,
1997; Sun, Triplett & Gainey, 2004).
For example, Sampson et al. (1997) set out to examine whether collective efficacy
(measured at the micro-level) mediates the relationship between social disorganization
(measured at the macro-level) and violence utilizing a hierarchical dataset (where 8,782
residents were nested within 343 Chicago neighborhoods). Using HLM, Sampson et al.
included macro-level neighborhood characteristics (concentrated disadvantage,
immigrant concentration and residential stability) and micro-level measures of collective
efficacy (informal social control, social cohesion and trust) in their multi-level model in
order to examine the effects these variables have on self-reported violence. Sampson
and colleagues (1997) found support for Sampson’s (1991) original arguments that
collective efficacy mediates the relationship between social disorganization and violence.
Subsequent studies that have utilized hierarchical datasets and hierarchical statistical
procedures (i.e., HLM) have also found support for macro-micro Social Disorganization
(Browning, 2002; Sun et al., 2004; Wooldredge & Thistlethwaite, 2002).
General Strain Theory
In 1992, Agnew expanded on the work of Merton (1938), introducing the possibility
that individuals experience additional sources of strain (beyond economic strain). This work
not only presented the addition of strains beyond those resulting from economics, but
created a micro-level strain theory. General Strain Theory speculates that crime or
delinquency is largely the result of feeling angry, which comes from experiencing strain. The
likelihood that an angry individual will turn to crime to alleviate the strain they are
experiencing depends largely upon the coping mechanisms available to the individual.
Micro-level tests have provided general support for the theory (Agnew & White, 1992;
Baron, 2004; Brezina, 1996; Broidy, 2000; Mazerolle, Burton, Cullen, Evans & Payne, 2000;
Paternoster & Mazerolle, 1994).
In 1999, Agnew proposed a macro-level version of General Strain Theory. Macro General
Strain Theory (MGST) is positioned to explain community differences in crime rates (Agnew,
2006, 1999). Drawing from other structural theories of crime (specifically Social Disorganization
and Social Learning and Social Structure), Agnew (2006) argues that MGST can explain
differences in crime rates across communities because individuals residing in deprived
communities “are more likely to experience strains conducive to crime and cope with strains
through crime” (p. 155). Thus, strain is thought to mediate the relationship between
community disorder and crime. In addition, Agnew (2006) contends that “deprived
communities are more likely to attract and retain strained individuals” (p. 155). Warner and
Fowler (2003) tested this theory, finding some support for Agnew’s macro-level
propositions.
Most recently, work by Wareham, Cochran, Dembo and Sellers (1999) has proposed
a macro-micro version of General Strain Theory. Distinctively, Wareham and associates
(1999) argue that:
While the structural/macro version of GST was not explicitly advanced as
a multi-level explanation of effect of strain on crime, this statement raises
the tantalizing possibility that GST may also be conceptualized and
empirically tested as a multi-level integrated theory. (p. 118)
Setting out to test the value of a macro-micro version of Agnew’s General Strain Theory,
Wareham et al. (1999) utilized a hierarchical dataset that consisted of 430 students nested
within 108 community blocks. Micro-level data were collected through the administration of
self-report surveys. Individual-level variables included in the analyses represented
individual strain, negative affects (i.e., anger), and self-reported delinquency. Macro-level
data were collected from the Census Bureau. Structural-level variables included poverty,
residential mobility, racial heterogeneity, and female-headed households. Using HLM,
Wareham et al. (1999) did not find initial support for a multi-level version of GST. The
researchers, however, correctly point out that their study is plagued by a relatively small
sample size (on average each community block contained only four students). Because of
the small sample size, the authors caution that potentially significant effects may have
been overlooked. As such, a more accurate test of macro-micro GST should be conducted
utilizing a larger sample.
Subsequent research has examined the robustness of multi-level GST (Boardman,
Finch, Ellison, William & Jackson, 2001; Hoffman, 2002; Hoffman & Ireland, 2004). For
instance, Boardman and colleagues (2001) examined the impact of neighborhood
disadvantage on individual levels of stress and subsequent drug use. Data were collected
from the Census Bureau for 139 census tracts while micro-level data were collected from a
self-report study conducted among 1,101 adults residing in Detroit, Michigan. While
acknowledging the problems associated with utilizing a standard logistic regression model
when data is hierarchical in nature (see Chapter 4 for an overview of these issues),
Boardman et al. (2001) nonetheless use OLS regression to conduct their analyses. Overall,
they found support for the argument that the relationship between neighborhood
disadvantage and drug use is mediated by variables representative of General Strain
Theory.
In another study, Hoffman (2002) examined the relationship between community
characteristics, delinquent peer associations, informal social control, general strain and
juvenile delinquency. A multi-level model was constructed using self-reported data
collected from the initial wave of the National Educational Longitudinal Study (10,868 10th
graders) and macro-level data from the Census Bureau (1,617 communities identified by
Zip Code). Because of the hierarchical nature of the data, Hoffman (2002) used HLM to
nest the students within their respective communities (averaging about 6.7 students per
community). Hoffman (2002) found that communities plagued by high rates of
unemployment were significantly more likely to have strained and poorly supervised
juvenile delinquents than communities with low rates of unemployment.
Finally, Hoffman and Ireland (2004) utilized longitudinal data to examine the impact
of strain (measured at the macro- and micro-level) on delinquency among 12,420 students
from 883 schools. Specially, they were interested in examining whether “reported strain or
stress in 1990 result[s] in subsequent increased involvement in delinquency reported in
1992” controlling for structural and individual effects (p. 273). In their multi-level study,
Hoffman and Ireland (2004) operationalize strain in two manners. First, relying on
traditional measures of strain, they include a variable representing the “disjunction among
economic goals and educational expectations” (p. 274). Second, a composite measure of
stressful life experiences from the past year is included. Hoffman and Ireland (2004) found
independent effects for contextual variables representing opportunity structures (macro
strain) and general strain (micro strain) on delinquency. However, they did not find that
individual-levels of strain vary across opportunity structures.
Institutional Anomie Theory
Institutional Anomie Theory (IAT) is a structural- (macro) level theory that has been
proposed to explain differences in criminal offending across nation states. Specifically, IAT
attempts to explain disparity in offending rates by examining differences in adherence to
cultural values and involvement in macro-social institutional domains (Messner & Rosenfeld,
2004). Institutions are an important component of the theory because they are viewed as
social structures that “regulate human conduct to meet the basic needs of a society”
(Messner & Rosenfeld, 2001, p. 65). The four institutions IAT focuses on are the economy,
polity, family, and education.
A second important component of the theory is culture. In societies where the
economy is dominant, IAT proposes that cultural values (i.e., the “American dream”)
encourage the achievement of success “by any means possible,” and as a result, crime
flourishes. Messner and Rosenfeld (2001) define the “American dream” as consisting of four
cultural values: achievement, individualism, universalism, and the fetishism of money.
Thus, the crux of Institutional Anomie Theory is that crime thrives in societies where the
institutional balance is skewed towards the economy, which is supported and reinforced by the
ideals of the “American dream.” In contrast, when there is equality among institutions, non-
economic institutions (i.e., family, education, and the polity) are capable of offsetting the
criminogenic effects of both a dominating, capitalist economy and the cultural ethos of the
“American dream.” While a relatively new theory, a growing body of research has evaluated the
explanatory power of IAT (Batton & Jensen, 2002; Chamlin & Cochran, 1995; Maume & Lee,
2003; Messner & Rosenfeld, 1997; Kim & Pridemore, 2005; Piquero & Piquero, 1998; Savolainen,
2000). Consistent with the theory’s macro social perspective, the majority of these tests have
examined IAT variables at the aggregate level only. In addition, each of these studies has
failed to include an important component of IAT: culture (for an exception see J. B. Cullen,
Parboteeah, & Hoegl, 2004; Muftić, 2006).
Single-level theories, such as Institutional Anomie Theory, may benefit from multi-
level theoretical integration. As previously defined, multi-level theoretical integration, or
macro-micro integration, differentiates the causal properties of structural and individual
factors, identifying mediating and moderating linkages between cross-level variables and
their relationship with crime and delinquency. In subsequent writings on IAT, Messner and
Rosenfeld (2004) hint at the necessity of multi-level analyses of crime and criminality. They
state that “given that institutions constitute a salient feature of the situation or social
environment in all societies, explaining individual behavior requires an understanding of the
institutional context” (Messner & Rosenfeld, 2004, p. 97). They also go on to say that:
Studies of individual criminal behavior from an institutional perspective,
therefore, will nearly always require multi-level methods. Such methods, in
principle, allow for the portioning of individual behavior into a component
associated with differences in social context and a component associated
with variation across individuals within a given context. (Messner &
Rosenfeld, 2004, p. 99)
These statements provide support for a multi-level interpretation (and test) of their theory.
CONCLUSION
Theoretical integration is not new3. In fact, work as early as Lombroso’s suggested
the need for integration of theoretical ideas4 (Bohm, 2001). We can also see integrative
practices in many of the leading criminological theories. For instance, in their development
of Social Disorganization Theory, Shaw and McKay (1942, 1969) integrated concepts from
ecology, subcultural and control theories. Sutherland’s (1947) Differential Association
Theory has its roots in the Chicago school as well as conflict sociological approaches.
Merton (1938) draws from Durkheim’s theory of anomie, as well as cultural deviance
theories, in his development of classical Strain Theory. It may be argued that virtually all
criminological theories are in some form or another integrated theories, having borrowed
concepts, propositions, and ideas from within and without the discipline (Osgood, 1998).
Criminology has been dominated by theories that have relied on either strictly
macro- or micro-level theoretical propositions. These theories, however, have generally
fallen short in their ability to explain crime and criminality. In response, some criminologists
have begun to advocate for the integration of theoretical arguments. Recent work by
Agnew (2006, 2005, 1999), Akers (1998), and Sampson and Laub (1993) have all included
propositions in their theories that implicate the need for cross-level theoretical models. For
instance, in the creation of his general theory of offending and delinquency, Agnew (2005)
proposes that criminal motivation (why people do or do not commit crime) is best explained
by an integrated analysis that includes variables from the community in which the
individual resides along with variables representing individual characteristics. Similarly,
Akers (1998) proposes a cross-level version of Social Learning Theory where social learning
variables mediate the relationship between social structure and individual behaviors.
Finally, Sampson and Laub (1993) have expanded upon Social Bonding Theory to include
an analysis of structural characteristics (i.e., residential mobility, socio-economic status,
and family disruption) and their impact on informal social control.
Despite research calling for the integration of macro- and micro-level theoretical
explanations, there remains a paucity of research (and theoretical) attention given to
macro-micro theoretical explanations. One possible explanation as to why there have been
so few attempts at macro-micro theoretical integration may be that until recently, it was
methodologically impossible to statistically test the propositions of a cross-level integrated
theory (Garner & Raudenbush, 1991). The ability to test the propositions of integrated
multi-level theoretical explanations has largely been made possible through advancement
in statistical techniques over the past two decades. Techniques like hierarchical linear
modeling (HLM) allow researchers to nest individual-level variables into community
structural variables. Multi-level analysis is possible because such techniques permit the
researcher to control for the effect of both proximal (micro) and distal (macro) level
variables on crime and delinquency. In addition, HLM provides the researcher a way in
which to model the implicit hierarchy involved between characteristics of individuals and
the communities in which they live (Rountree, Land, & Miethe, 1994).
The use of HLM and other similar statistical techniques has not only created
renewed interest, but has also produced more empirical support for traditionally macro-
level theories such as Social Disorganization Theory. For example, recent studies that have
included both micro-level (social capital and collective efficacy) and macro-level (poverty,
family disruption, racial heterogeneity, and social mobility) variables in their multi-level
analyses find more support for Social Disorganization Theory compared to previous
research that included only structural variables (Browning, 2002; Rountree, Land & Miethe,
1994; Sampson, Raudenbush & Earls, 1997; Sun, Triplett & Gainey, 2004; Wooldredge &
Thistlethwaite, 2002).
In addition, the use of multi-level regression techniques allows for the exploration of
causal heterogeneity (Steenbergen & Jones, 2002). In other words, hierarchical statistical
procedures allow for the examination of any direct effects of individual and contextual variables
on the dependent variable of interest. Additionally, such procedures permit the assessment of
whether macro-level variables are conditioned by micro-level variables (Guo
& Zhao, 2000). This is all done while taking into account the unique hierarchy of multi-level
data, including the proper causal order of multi-level variables (i.e., that a macro-level
variable may affect another macro-level variable or a micro-level variable, but that a micro-
level variable may only affect another micro-level variable, but not a macro-level variable;
Krull & MacKinnon, 2001).
Opponents of integration have long argued that the complexity of integrated theories
impedes their testability. However, with the increase in methodological sophistification in the last
decade, theorists should no longer shy away from “complex” theories. Rather, future theoretical
development (and subsequent theory testing) needs to consider such
complexity, while at the same time concentrating on the relationship between macro- and
micro-level variables and crime and deviance. The relevance of the current article is the
presentation of evidence that accentuates the importance of theoretical models including both
individual and structural variables, supporting the notion that multi-level theoretical models
provide a richer, more complete picture of the phenomenon of interest.
REFERENCES
Agnew, R. (1991). The interactive effects of peer variables on delinquency.
Criminology, 28, 47-72.
Agnew, R. (1999). A General Strain Theory of community differences in crime rates.
Journal of Research in Crime and Delinquency, 36(2), 123-155.
Agnew, R. (2005). Why do criminals offend? A general theory of crime and
delinquency. Los Angeles: Roxbury Publishing.
Agnew, R. (2006). Pressured into crime: An overview of General Strain Theory. Los
Angeles: Roxbury Publishing.
Agnew, R., & White, H. R. (1992). An empirical test of General Strain Theory.
Criminology, 30(4), 475-495.
Akers, R. L. (1989). A social behaviorist’s perspective on integration of theories of
crime and deviance. In S. F. Messner, M. D. Krohn, & A. E. Liska (Eds.),
Theoretical integration in the study of deviance and crime: problems and
prospects (pp. 23-36). Buffalo: State University of New York Press.
Akers, R. L. (1998). Social learning and social structure: A general theory of crime
and deviance. Boston: Northeastern University Press.
Akers, R. L., & Sellers, C. S. (2004). Criminological theories: Introduction, evaluation
and application (4th ed.). Los Angeles: Roxbury Publishing Company.
Alaird, L. F., Burton, V. S., & Cullen, F. T. (2000). Gender and crime among felony
offenders: assessing the generality of Social Bond and Differential Association
theories. Journal of Research in Crime and Delinquency, 37(2), 171-199.
Barak, G. (2008). Toward an Integrative Study of International and State-Corporate
Criminality:
A Reciprocal Approach to Gross Human Rights Violations. In R. Haveman and A.
Smeulers (eds.), Supranational Criminology: Towards Criminology of International
Crime. Antwerp, The Netherlands: Intersentia Press.
Barak, G. (1998). Integrating criminologies. Boston: Allyn and Bacon.
Baron, S. W. (2004). General Strain Theory, street youth and crime: A test of
Agnew’s revised theory. Criminology, 24, 457-483.
Batton, C., & Jensen, G. (2002). Decommodification and homicide rates in the 20th-
century United States. Homicide Studies, 6(1), 6-38.
Bellair, P. E., Roscigno, V. J., & McNulty, T. L. (2003). Linking local labor market
opportunity to violent adolescent delinquency. Journal of Research in Crime and
Delinquency, 40(1), 6-33.
Bernard, T. J., & Snipes, J. B. (1996). Theoretical integration in criminology. Crime
and Justice: An Annual Review of Research, 20, 301-348.
Boardman, J. D., Finch, B. K., Ellison, C. G., Williams, D. R., & Jackson, J. S. (2001).
Neighborhood disadvantage, stress, and drug use among adults. Journal of Health
and Social Behavior, 42, 151-165.
Bohm, R. M. (2001). A primer on crime and delinquency theory (2nd ed.). Belmont:
Wadsworth/Thomson Learning.
Bouffard, L. A. (2003). Examining the relationship between military service and
criminal behavior during the Vietnam era. Criminology, 41, 491-510.
Braithwaite, J. (1989). Crime, shame, and reintegration. Cambridge: Cambridg
University Press.
Brezina, T. (1996). Adapting to strain: An examination of delinquent coping
responses. Criminology, 34(1), 39-60.
Broidy, L. M. (2000). A test of General Strain Theory. Criminology, 39(1), 9-32.
Browning, C. R. (2002). The span of collective efficacy: Extending Social
Disorganization Theory to partner violence. Journal of Marriage and Family, 64, 833-
850.
Bursik, R. J. (1988). Social Disorganization and theories of crime and delinquency:
Problems and prospects. Criminology, 26(4), 519-551.
Bursik, R. J. (2000). The systemic theory of neighborhood crime rates. In S. S. Simpson
(Ed.), Of crime & criminality: The use of theory in everyday life (pp. 87-108).
Boston: Pine Forge Press.
Bursik, R. J., & Webb, J. (1982). Community change and patterns of delinquency. The
American Journal of Sociology, 88(1), 24-42.
Cattarello, A. M. (2000). Community-level influences on individuals’ social bonds,
peer associations, and delinquency: A multi-level analysis. Justice Quarterly, 17(1),
33-59.
Chamlin, M. B., & Cochran, J. K. (1995). Assessing Messner and Rosenfeld’s
Institutional Anomie Theory: A partial test. Criminology, 33(3), 411-429.
Colvin, M., Cullen, F. T., & Vander Ven, T. (2002). Coercion, social support, and
crime: An emerging theoretical consensus. Criminology, 40(1), 19-42.
