Download Theory, Method, and Data in Comparative Criminology and more Study notes Criminology in PDF only on Docsity!
VOLUME 4
A B S T R A C T
Theory, Method, and
Data in Comparative
Criminology
by Gregory J. Howard, Graeme Newman, and
William Alex Pridemore
The goal of this chapter is to provide an overview of the recently reju-
venated field of comparative criminology. It begins by considering the
context and history of comparative criminology and continues by out-
lining the contemporary comparative perspective. After identifying
several goals for comparative criminology that are often advanced,
including theory elaboration and testing as well as policy evaluation
and critique, the chapter describes the common approaches to com-
parative criminological research. The main theoretical traditions of
comparative criminology are examined first, with particular attention
directed to metanarratives such as modernization, civilization, oppor-
tunity, and world system theories and to structural theories based on
culture, social bonds, and the distribution of economic resources.
Taking up methodological concerns next, the chapter summarizes
some of the more common dependent variables studied by compara-
tive criminologists, noting how these variables have been operational-
ized in the literature, then explores the three methodological
approaches most typically deployed in the field, specifically metalev-
el, parallel, and case studies.
Gregory J. Howard is Assistant Professor of Sociology at Western Michigan University; Graeme Newman is Distinguished Teaching Professor of Criminal Justice at the University of Albany, State University of New York; and William Alex Pridemore is Assistant Professor of Sociology at the University of Oklahoma in Norman.
MEASUREMENT AND ANALYSIS OF CRIME AND JUSTICE
CRIMINAL JUSTICE 2000
THEORY, METHOD , AND DATA IN COMPARATIVE CRIMINOLOGY
With the growth of international “transparency” and the capacity of the World Wide Web to disseminate information, data about crime and justice around the world are more accessible then ever. The chapter discusses the three most common types of data on international crime and justice (i.e., official, victimization, and self-report data), describing the threats to the reliability and validity of each type and directing interested readers to exist- ing sources of data relevant to frequently employed explanatory concepts.
In conclusion, the chapter observes that, while comparative criminology is a growing area of study owing to the influence of globalization and concerns about transnational crime, the relative neglect of systematic comparative work in criminology throughout the 20th century means that the field is still in its infancy. Growth in this promising area of inquiry should be nurtured with a renaissance in theory so that research is driven by theory and not by the mere existence of more data.
A B S T R A C T
THEORY, METHOD , AND DATA IN COMPARATIVE CRIMINOLOGY
CRIMINAL JUSTICE 2000
Although many theoretical, methodological, and philosophical problems cer- tainly have dogged comparative criminology since its inception, there is little doubt that this field of investigation is currently in a state of rapid expansion. While this chapter outlines some of the main problems that confront compara- tive criminology, the discussion also focuses on what cross-national research has accomplished and what it can do for the field of criminology in the future. We begin with two questions often asked of comparative criminology: What is the comparative perspective, and why employ it? Following this discussion, we move to a consideration of the substantive and theoretical issues that lie at the root of comparative criminological inquiry. We must begin with theory, because the plethora of databases and other information now available from many coun- tries provides an environment that tempts rash comparisons and sometimes unsubstantiated conclusions based on what may be incomparable data. Faced with such a challenge, theoretically informed research supported by sound methodology is the wisest defense. Consequently, we look at the theoretical perspectives that have been brought to bear in understanding crime from a comparative perspective. Following this, we consider crime as a dependent vari- able in comparative work, then stake out the methodological approaches that are often used in this type of investigation. We then consider the data available to researchers interested in pursuing comparative studies and conclude with some observations about the future of comparative research in criminology.
The comparative perspective
Globalization
In its broadest sense, all social science research is comparative. As Durkheim noted: “Comparative sociology is not a special branch of sociology; it is sociology itself” ([1895] 1938, 157). To the extent that the scientific method depends on comparison, Durkheim is no doubt correct. But comparative crimi- nology demands more than comparison. Comparative investigations do not involve so much a method as a perspective, one that demands a gestalt that in today’s terminology might be called a global view. It may be argued that applying today’s global perspective to comparative criminology is misleading because comparative studies of crime have been in existence for two centuries. Yet, comparative research has received relatively little attention from scholars in our field. Marsh noted in 1967 that a tiny number of articles in social science journals at that time dealt with crime in two or more cultures, and Beirne (1997, xiii) pointed out that this lack of interest persisted into the 1980s. It is clear, however, that engagement in comparative criminology has increased sig- nificantly in the last decade (see Adler, Mueller, and Laufer 1994; Ebbe 1996;
MEASUREMENT AND ANALYSIS OF CRIME AND JUSTICE
VOLUME 4
Moore and Fields 1996; Reichel 1994). One major reason is simply that a global perspective on all popular subjects now dominates the world.
The nation-state and the universality of the criminal
justice system
It has become increasingly clear that much of what we call criminal justice depends to a large extent on the operations and structure of modern nation- states. It is the modern nation-state that has given us the tripartite criminal justice system (i.e., police, courts, and corrections) as it is commonly studied today (Newman 1999). Though it may rest within diverse legal traditions and cultures, this basic tripartite structure is similar in all modern nation-states (for authoritative discussions on legal traditions, see David and Brierley 1985 and Wigmore 1936), and investigations that compare nations do so on the assump- tion that these nations have similar structures of criminal justice. Thus, more attention has been given recently to comparing whole nations in addition to comparing cultures. This is partly the result of globalization, but much is due to the more easily definable boundaries and structures of nation-states when com- pared with cultures. (The latter, of course, often transcend national boundaries.)
The single most important difference between cultures and nation-states is that nation-states are political entities while cultures are ways of life. Nation-states may be composed of many cultures, as the recent war in the former Yugoslavia attests, and a single culture (depending on how broadly the term is defined) may span several countries (e.g., Roman Catholicism, Islam, Judaism, American popular culture). Because the nation-state is the operational basis for the crime control activities of criminal justice agencies, and because it is more easily defined than the elusive concept of culture, most recent comparative work in criminology has examined similarities and differ- ences with respect to the nation-state (or relevant political subdivisions such as the state or province).
Definitional diversity of crime
One endemic problem confronting comparative crim- inology is the enormous diversity in the way different cultures and nation-states define crime, justice, and other relevant concepts. As we shall see, this has not stopped many researchers from conducting a wide variety of studies comparing crime and delinquency
The greatest advantage— although often considered an impediment—for research in compar- ative criminology is the great diversity that exists cross- nationally with regard to social, economic, and political indicators.
MEASUREMENT AND ANALYSIS OF CRIME AND JUSTICE
VOLUME 4
research, however, and attention will be paid to these issues in cross-national terms in the pages to come.
The goals of comparative research
There are several goals of comparative research in criminology. Some are obvi- ous applications of the traditional canons of the scientific method, and some are unique to the study of crime in an international setting. Although compara- tive criminology attends mainly to understanding criminal and deviant behavior as it is manifested globally, these studies will inevitably yield useful insights about the control of antisocial activity. Thus, the study of criminology will nat- urally intersect with the field of criminal justice if criminological observations are taken to their logical policy conclusions. With respect to the scientific import of comparative work in criminology, a few important goals are noted here.
Extending theories beyond cultural and national boundaries
Comparative research provides an opportunity for criminological theories, which are typically generated within the context of particular nation-states, to be given a wider hearing (Mueller and Adler 1996). Do the theories developed to explain crime rates in the United States, Finland, Japan, or South Africa serve with equal force to account for criminal violations in other nations around the world? Do theories that try to account for the police use of force in the United States, Russia, Australia, or Brazil help to understand police behavior in other countries? These are questions of replication, and they stand at the heart of the scientific enterprise. Beside permitting criminologists to assess the generalizability of important theoretical propositions, comparative research also assists in the elaboration and specification of theory. If a specific theoretical model does not account for variations in crime in other nation-states, perhaps some refinement can be identified on the basis of this investigation, thereby improving the explanatory power of the theory. Finally, as the world becomes a smaller place through the expansion of globalization and crime and criminal justice become increasingly transnational as a result, comparative investigations ensure that theories of crime and criminal justice will remain relevant to the exigencies of history.
Assessing the performance of national criminal justice
systems
Another important goal for comparative work in criminology is the assessment of national criminal justice systems. For example, an article by Maguire, Howard, and Newman (1998) developed an index by which the performance
THEORY, METHOD , AND DATA IN COMPARATIVE CRIMINOLOGY
CRIMINAL JUSTICE 2000
of national criminal justice systems can be validly compared across nations. The idea of a criminal justice system is of relatively recent vintage, most forcefully espoused in the 1967 report of the President’s Commission on Law Enforcement and Administration of Justice, The Challenge of Crime in a Free Society. Despite this conceptualization of criminal justice, which aims to make the practices of crime control more efficient and less fragmented, the promise of the system idea has hardly been realized more than 30 years later, as Inciardi’s (1984) charac- terization of the U.S. criminal justice system as a nonsystem suggests.
An index of national criminal justice systems’ per- formances, however, can serve to overcome the fragmented operations of the administration of justice that have been so widely attacked in the academic lit- erature and popular media alike. In short, if the various institutions of criminal justice (i.e., police, courts, and corrections) are to work as a system charged with the control of criminal behavior, there must be some way to assess their performance as an operational unit. The performance measures of national criminal justice systems establish this type of benchmark. Moreover, compar- isons of the performance of national criminal justice systems inevitably raise questions of policy and, in the long run, of justice. These are questions for which comparative criminologists should be able to field reasonable answers.
Evaluating national criminal justice policy
Comparative criminology and criminal justice also promise to yield insights into the efficacy of various policy initiatives. For instance, are high levels of gun violence inevitable in the United States because it harbors a gun culture? Perhaps there are other countries that have a high level of gun ownership but a low rate of gun crime (see Killias 1993). Would the legalization of drugs lead to an epidemic of drug use, as is often argued? Perhaps other countries have had a different experience. This is not to say that experiences with crime and its control in one nation should be copied wholesale to another. But when we see different and successful ways of dealing with crime in other countries, we at least know that it is possible and that the current state of affairs in a particular country is not preordained or inevitable. The work of the International Center for Crime Prevention has done much to highlight success stories in crime pre- vention throughout the world (Waller and Welsh 1999).
If the various institu- tions of criminal justice (i.e., police, courts, and correc- tions) are to work as a system charged with the control of criminal behavior, there must be some way to assess their performance as an operational unit.
THEORY, METHOD , AND DATA IN COMPARATIVE CRIMINOLOGY
CRIMINAL JUSTICE 2000
Theoretical Explanations for Criminal Behavior in a Cross-National Context
1
Three general theoretical frameworks are commonly employed in explaining the variation of crime rates among nations. The first—grand theories—are metanarratives that entail a high level of abstraction and usually assume that one major theoretical construct, such as a nation’s level of modernization or its placement in the world’s political economy, has the greatest impact on its level of crime.^2 Structural theories, on the other hand, traditionally have been employed with smaller units of analysis, such as cities or states within a partic- ular nation, and attempt to explain the spatial variation in rates of offending via subcultures (i.e., social learning), status-induced strain, or social control. These theoretical explanations are increasingly being tested at the cross-national level. Finally, a nation’s demographic characteristics, such as its age and sex struc- ture, may also be employed in an attempt to explain levels of violence and property crime. Each section following contains a brief description of theoreti- cal concepts, common measurements for each of these concepts, and a selected list of studies that analyze crime from the particular approach. In all cases, these studies represent theories that were originally developed within one cultural and/or national tradition but that have been recast and applied to cross-national research designs. Thus, many of the theoretical approaches will initially appear familiar to criminologists. However, the employment of these theories in cross- national research promises either to extend the explanatory power of these the- ories or to demonstrate their limitations.
Grand theories
Modernization theory
Durkheim’s notion of anomie is the basis for contemporary theories that pinpoint the effects of modernization as the main causes of crime. Durkheim ([1893] 1964) argued that, as nations develop, they are characterized by an increasingly intricate web of social and economic relations. These complex divisions are suspected of undermining mechanical solidarity and its control over the collective conscience. Thus, rapid social change engenders the break- down of traditional values, resulting in, among other things, a higher crime rate. Eventually, however, organic solidarity and more formal mechanisms of social control should halt rising crime rates, although they are expected to remain at higher levels than before development.
Within cross-national criminology, Clinard and Abbott (1973) and Shelley (1981) have provided notable contributions to Durkheim’s idea of modernization and
MEASUREMENT AND ANALYSIS OF CRIME AND JUSTICE
VOLUME 4
crime. In short, the contention is that each nation experiences similar phases of development. The stimulus for modernization is technological advancement, and this catalyst leads to political, economic, and demographic changes within a society (Strasser and Randall 1981). Industrialization and urbanization are key elements of this social transformation. Their effects—which include the tension between groups that accompanies increased social differentiation and the socioeconomic inequality many presume to follow modernization—are viewed as the main contributors to rising crime rates (Heiland and Shelley 1992). The outcomes of this process are expected to provide stronger explanations of crime than any distinct national or cultural characteristics have; thus, all nations are expected to experience similar trends in crime rates as they develop.^3 Finally, changes in both types and rates of crime are expected as a result of develop- ment. In general, the overall rate of crime is expected to increase but eventually level off as modernization progresses, while another expected result is the pre- dominance of property and economic crimes over crimes against people.
Recent examples of international research on modernization and crime include:
n (^) Bouley and Vaughn’s (1995) study of violent crime in Colombia, in which regression analysis revealed support for grand theories with respect to the crimes of theft and robbery but not for the more violent crimes of assault and homicide.
n (^) Mahabir’s (1988) work on urban gangs in the Caribbean.
n (^) Industrialization and crime in the Russian region of Tuva (Balakina 1994), which official data show has the highest homicide rate in the country (Pridemore 1999).
n (^) Huang’s (1995) multivariate analysis of 29 countries employing United Nations data, which showed support for the modernization hypothesis.
n (^) Ortega and colleagues’ (1992) analysis of 51 nations using International Criminal Police Organization (Interpol) crime data, which revealed support for the Durkheimian-modernization thesis.
n (^) Johnson’s (1990) work, which examined crime and industrial development in Germany and concluded that modernization theory did not hold in this case.
n (^) Matsuoka and Kelly’s (1988) study of the negative impact of resort develop- ment and tourism on Native Hawaiians.
n (^) Neuman and Berger’s (1988) evaluation of modernization, Marxian world system, and ecological opportunity theories, which lent only weak support to the modernization theory.
MEASUREMENT AND ANALYSIS OF CRIME AND JUSTICE
VOLUME 4
1914, revealing that the climbing literacy rate was associated with a decrease in the rate of crimes of passion (e.g., homicide and certain other types of violence) and an increase in the suicide rate. Gillis argued that this information represents strong support for the hypothesis, which predicted that, as the civi- lization process occurs, interpersonal violence will decrease and violence direct against the self will increase.
Opportunity theories
In recognizing the complex social-structural changes that accompany societal evolution, opportunity theo- ries are similar to both modernization and civilization theories. However, while the modernization hypothe- sis focuses on how changes compromise traditional values and the civilization thesis expects the develop- ment of internal self-restraint (as opposed to the external social control of the state) to inhibit harmful behavior, opportunity theories suggest that modern economies and social organization provide increased opportunities to engage in criminal behavior (Cohen and Felson 1979). For example, expanding economies create an increase in expendable income in the average household, which people can then spend on a growing variety of con- sumer goods, which in turn are increasingly available for theft. At the same time, technological gains produce smaller and more portable electronic devices that are easily stolen. Similarly, work (e.g., both spouses working instead of only one) and leisure activities (e.g., a larger amount of expendable income to spend on entertainment) may mean less time spent at home for many, which results in less guardianship over household items that may be stolen. At the same time, increased residential mobility and cultural heterogeneity may lead to weaker community ties than in the past, resulting in communities composed of people who are less willing to guard the personal safety and private property of neighbors they barely know.
It should be made clear that opportunity theorists do not expect the conditions of modern society to create forces that cause criminal behavior; rather, they assume that we are all motivated offenders who will act criminally in a situation given the presence of a suitable victim and the absence of a capable guardian (Cohen and Felson 1979; Felson and Cohen 1980). Thus, opportunity theory posits an increase in property crimes and a decrease in violent crimes over
Opportunity theo- rists do not expect the conditions of modern society to create forces that cause criminal behavior; rather, they assume that we are all motivated offenders who will act criminally in a situation given the presence of a suitable victim and the absence of a capable guardian.
THEORY, METHOD , AND DATA IN COMPARATIVE CRIMINOLOGY
CRIMINAL JUSTICE 2000
time. Studies that directly test opportunity theories are not widespread at the cross-national level (for example, see Bennett 1991a, 1991b; Gartner 1990; Kick and LaFree 1985; Ortega et al. 1992). However, though it may be difficult to measure the amount of time citizens of different countries spend away from home or the average weight of their televisions, studies that examine the effect of modernization and crime may also be interpreted in terms of this thesis. For example, while Bennett (1991b); Groves, McCleary, and Newman (1985); Neapolitan (1994, 1996); and several others show no significant increase in the homicide rate with economic development, others, such as Stack (1984) and Hartnagel (1982), have shown that increases in property crime accompany economic development and/or urbanism, which can result in an increase in the number of victims and, due to mobility and heterogeneity, a decrease in guardianship.
World system theory
World system theory borrows from the Marxist perspective to explain the impact of an ever-expanding capitalism on nations that vary in their level of development (see Chirot and Hall 1982). This theory recognizes that the legiti- macy of a market economy is spreading around the globe and that its expansion is uneven, meaning that (1) nations are no longer autonomous political and economic entities but are instead actors in an international political-economic system, and (2) weaker countries are politically and economically exploited by stronger ones (Smith 1984). This uneven expansion results in nations that are placed into one of three categories:
n (^) Core nations are industrialized, and market relations in them are highly advanced.
n (^) Periphery nations are characterized by a history of colonialism and the pos- session of natural and human resources that are underdeveloped but available for exploitation by the industrialized sectors of core nations. These nations are viewed as economically dependent on the core nations.
n (^) Semi-periphery nations are underdeveloped, perhaps only partially industrialized, and at the mercy of both the core and the periphery (Evans and Timberlake 1980; Walton 1982).
As capitalism expands, world system theory maintains, it disrupts indigenous cultures and traditional means of subsistence, producing exploitation from the outside and new inequalities within. Political and legal formations are disrupt- ed, and social dislocations become widespread. The rural population begins to migrate to cities in search of employment, creating class conflict and
THEORY, METHOD , AND DATA IN COMPARATIVE CRIMINOLOGY
CRIMINAL JUSTICE 2000
it may be argued that a relationship between economic distress and crime within a country reveals support for a Marxist approach, but this does not necessarily speak to a larger world system and whether this system in fact has created the poor economic conditions within a nation. A cross-sectional design is not an entirely inappropriate methodology, and its predominance has perhaps been necessary due to the limited availability of data on nations over time. However, given the increasing availability, reliability, and validity of cross-national data and the growing sophistication of scientific techniques employed by criminolo- gists, we must take care to design studies that best answer the proposed theoret- ical hypotheses and recognize design limitations when drawing our conclusions.
Structural theories There are three general approaches to the study of the etiology of crime: social learning, strain, and social control. The structural analogs of these theoretical strategies are culture, strain (usually represented by absolute or relative eco- nomic deprivation), and social disorganization. These three theories, together with attempts to measure them and identify their impact on crime at the cross- national level, are outlined briefly.
Culture
The attempt to explain crime via cultural variation is a longstanding enterprise. In the first half of the 19th century, Guerry blamed the high rates of violence in the south of France on regional differences in culture that resulted from migra- tion and settlement patterns (Corzine, Huff-Corzine, and Whitt 1998). In Italy, many argued that southern Italians, such as the Neapolitans and Sicilians, pos- sessed cultural traits responsible for the high levels of crime in the regions they inhabited. In the United States, Redfield (1880) began a research tradition that continues today, when his systematic research revealed heightened rates of homicide in the American South.
Despite the fact that southerners everywhere seem to be viewed as a rather impetuous lot, cultural theories of crime are actually grounded in social learn- ing processes and cultural norms. Researchers do not believe, for example, that southerners are born violent; instead, they submit that residents of the South learn violent traits from those in close proximity to them. 4
Modern cultural theories, then, build on Sutherland’s (1947) notion of differen- tial association, in which norms conducive to violence are transmitted to indi- viduals and across generations via processes of social learning. Having acquired these cultural values, the individual is provided with a “tool kit” for living (Swidler 1986), as culture provides one with the means to interpret interpersonal
MEASUREMENT AND ANALYSIS OF CRIME AND JUSTICE
VOLUME 4
interactions while at the same time providing an accepted repertoire of responses appropriate for each situation (Wolfgang and Ferracuti 1967). Individuals in one culture, for example, may be offended by an action that someone in another culture dismisses as unimportant. Similarly, Luckenbill and Doyle (1989) argue that individuals in some cultures are more like- ly to view a negative interaction, no matter how slight, as an injustice that demands revenge. Thus, cultural norms may promote or condone violence in certain situations, meaning that the attitudes and val- ues of individuals within cultures that have higher rates of crime or violence should be distinguishable from those with lower rates.
Measuring and analyzing the effects of culture are extremely difficult tasks for the researcher. Some suggest that culture dictates the form of social institu- tions within a society (Lynch 1995a), while others argue that institutional/structural conditions have a hand in creating (sub)cultur- al values and that these values mediate the effects of social structure on behav- ior (Curtis 1975). These two views are not mutually exclusive and both are probably correct. However, attempts to measure culture at this level of analysis have been weak at best—usually taking national, regional, or ethnic group membership as a measure of culture—thereby making efforts to separate the discrete effects of culture and structure nearly impossible.
Research in the United States has focused on sub culturally violent groups (see Wolfgang and Ferracuti 1967; Messner 1983a, 1983b; Nisbett and Cohen 1996), but those interested in cross-national comparisons must focus on the nation as the unit of analysis. This makes it difficult to examine cultural influ- ences on crime because a cross-national sample requires an assumption that countries are culturally homogeneous, which is rarely the case. Therefore, although the literature (especially anthropological) abounds with case studies that examine the relationship between sociocultural attributes and levels and types of violence, there have been few empirical attempts to measure the effects of culture on crime with a cross-national sample.^5 This is not a futile task, however. It is possible to devise a standard survey instrument that might measure beliefs concerning situational acceptance of violent behavior and other similar attitudes among members of different nations.^6 To say that cultural vari- ation does not affect levels and types of violence among countries makes little sense; most agree that it does. Cross-national researchers, however, have yet to
Thus, cultural norms may promote or condone violence in certain situations, meaning that the attitudes and values of individuals within cultures that have higher rates of crime or violence should be distinguishable from those with lower rates.
MEASUREMENT AND ANALYSIS OF CRIME AND JUSTICE
VOLUME 4
or frustration. The Gini coefficient (often referred to as the index of income concentration), ranges from 0 to 1, with 0 indicating perfect equality and 1 indicating perfect inequality within a population. 7 A few cross-national studies that employed the Gini coefficient as a measure of inequality include Avison and Loring’s (1986) study of population diversity and homicide, Gartner’s (1990) work on homicide victims, Messner’s (1989) analysis of economic dis- crimination and homicide rates, and Messner and Rosenfeld’s (1997) exploration of the relationship between institutional anomie and homicide rates. Alternative measures of inequality are defined in terms of the percentage of overall income received by a specific proportion of the population or as a ratio of the earnings of one segment of the population to that of another. For example, Fiala and LaFree (1988) used both the percentage of total income received by the bottom 20 percent of the population and the ratio of the percentage of income received by the top 20 percent to that received by the bottom 40 percent of each nation.
Social disorganization
While cultural models are founded in social learning principles and strain theo- ries argue that economic and other forms of distress may propel people or groups toward criminal behavior, the theory of social disorganization posits that crime occurs as a result of a breakdown in social bonds. Structural forces act to disrupt social ties and group solidarity, thus interfering with community mechanisms (both formal and informal) of control. Detached from their social bonds and in the absence of the community’s ability to control the behavior of its members, people are free to become involved in criminal behavior (Bursik 1988; Sampson, Raudenbush, and Earls 1997). Thus, social disorganization is an extension of the modernization thesis outlined previously (e.g., one situation in which normative controls may be broken is when shifts in urbanization and the political economy occur during development). This view of community dis- organization is similar to the notion of social control discussed by Park and Burgess (1924) in their work on human ecology.
The community or neighborhood is usually considered the most appropriate level of analysis for testing social disorganization models. However, structural forces outside the community, such as political-economic shifts that redirect the distribution of jobs and services (Bursik 1988), obviously have an impact on social cohesion within a neighborhood. Thus, this theory is often tested at higher levels of analysis, including cross-nationally. The commonly accepted elements of these macrolevel models of social disorganization are poverty, population density, ethnic heterogeneity, residential mobility, and family dis- ruption. Each is suspected of disrupting social integration and cohesion within communities, thereby weakening controls and allowing increases in crime rates (Sampson, Raudenbush, and Earls 1997).
THEORY, METHOD , AND DATA IN COMPARATIVE CRIMINOLOGY
CRIMINAL JUSTICE 2000
The structural elements of social disorganization are described theoretically in a manner that makes them relatively easy to operationalize in measurement models for cross-national analysis. For example, poverty can be measured as the proportion of the population living below the poverty line; population density can be captured as the percentage of the population that resides in urban areas (see Krahn, Hartnagel, and Gartrell 1986; Messner 1989; Ortega et al. 1992) or even the density of the whole country (see Avison and Loring 1986; Neapolitan 1994). Several definitions of heterogeneity are available, including ethnic and linguistic differentiation (commonly consisting of the percentage of the population that is not of the same ethnic background—or that speaks an alternative first language—as the majority; see Gartner 1990; Hansmann and Quigley 1982; Messner 1989). Residential mobility can be operationalized in terms of either urban (Braithwaite and Braithwaite 1980; Fiala and LaFree 1988) or total population (see LaFree and Kick 1986; Schichor 1990) growth or decline. Family disruption is most often measured as a nation’s divorce rate (see Gartner, Baker, and Pampel 1990; Rosenfeld and Messner 1991).
Within the past 15 years, elements of social disorganization have become wide- ly tested in cross-national empirical studies. Even if not directly discussed in terms of this theory, the concepts involved are commonly employed as control variables. No doubt there are problems—in terms of theory, measurement, and the congruence of the two—with each of the measures discussed in the previ- ous paragraphs, which authors of these studies sometimes address. However, the theoretical model for social disorganization is constructed in such a way as to make the production of a measurement model relatively simple, especially because data on these concepts are often readily available cross-nationally. Finally, tests of this model result in similar findings at several levels of aggregation (e.g., community, city, nation), suggesting that it might be a viable model that deserves further scientific attention at the cross-national level.
The demographic correlates of crime Although they do not provide causal mechanisms, the demographic attributes of a nation’s population are often used as control variables and, more recently, as elements of opportunity theories. In cross-national research, these demo- graphic correlates are commonly accepted to be sex and age. 8 In most studies, the sex and age categories are combined, the suggestion being that young males have the highest offending and victimization rates, and thus, as their proportion of the population increases, so will crime rates. Empirically, the cross-national findings examining this relationship have been inconsistent at best. Using various measures (age categories include younger than 15 years, 15–19, 15–24,
