III. Exploring the Social Class–Crime Link
Whereas the origin of social classes is relatively clear, the effect of social class divisions on crime is less so. Annual reports of the characteristics of people arrested in the United States provide insight concerning gender, age, race, and ethnicity but tell us little about social class characteristics such as income, occupation, or residence. Consequently, the best information we have regarding the social class characteristics of the individuals who inhabit U.S. prisons and jails derives from surveys of prison inmates and, interestingly, the government provides very little money to fund research into these characteristics. The last detailed survey of prisoners serving felony sentences in state penitentiaries, who make up the majority of those incarcerated in the United States in any given year, was conducted in 1993. Apparently, the federal government has little interest in regularly gathering information about the class and other social characteristics of prisoners.
An examination of the U.S. correctional population leaves little doubt that most of the people serving time for criminal offenses come from the lower end of society’s socioeconomic continuum. Government statistics show that criminal offenders in prison tend to be less well educated, more likely to be unemployed, and to earn far lower incomes than the general population. A 2002 survey conducted on inmates incarcerated in local jails revealed a similar pattern: Only about half of jail inmates were employed full-time at the time of their arrest, even though the national unemployment rate was below 5%, and over half of jail inmates earned less than $15,000 a year (Bureau of Justice Statistics, U.S. Department of Justice, 2004).
Although these statistics may be somewhat skewed by the fact that better-off offenders who are charged with street crimes are more likely to avoid imprisonment, there is little reason to believe the degree of error is substantial. All one has to do is observe any urban police station or city court to know that very few middle- or upper class citizens are arrested and prosecuted for common street crimes. Clearly, the criminal justice net hauls in the poorest of the poor. What this tells us about the link between social class and criminal behavior, however, remains controversial. Some scholars argue that the disproportionate representation of poor people in prison is indicative of their overinvolvement in crime, whereas others suggest this disparity is the result of a criminal justice system that unfairly targets the poor.
The contradictory perceptions about the relationship between social class and criminality are, in part, the product of disparate research findings. There is no shortage of research studies that have examined this relationship; however, there is little consensus because of inconsistent findings and inclusive results. For example, some studies have concluded that crime is more likely among people in higher social classes, whereas others have found criminality more prevalent among the lower classes. Some of these inconsistencies are traced to the different research methods used to study this relationship. These include different data collection methods; different measures of social class, crime, and criminality; different samples; and different methods of data analyses.
An examination of the past research reveals that the earliest of these studies (those conducted before the 1950s) tended to find more criminality among the lower classes than the upper classes. These findings in turn provided the foundation for numerous theories of crime and delinquency that attempted to explain why poverty was criminogenic, focusing on factors such as individual and cultural deficiencies, lack of opportunity, and differential (and harsher) treatment of individuals in poorer communities by the criminal justice system. Many of these theories, however, were only tenuously rooted in empirical research.
Although the social class–crime link was widely accepted, there were criminologists of the time who took issue with the methods that produced the correlation between social class and criminality. Most commonly, they argued that measuring crime through the use of official data (i.e., arrest data, prison statistics) presented a biased picture of crime. This measure of crime simply did not take into account the reality that many crimes go unnoticed or unreported, or for some other reason simply do not become known to those who wish to count them. This unknown and uncounted crime is referred to as the dark figure of crime. The problem, as they saw it, was that there was no way to determine whether accurately measuring the dark figure of crime would or would not show crime to be more broadly distributed. Some criminologists also argued that official measures of crime may actually better measure police practices than actual levels of crime; that is, in reality they may simply reflect, at least in part, the discretionary practices of police officers concerning whom to arrest and whom not to arrest, or a judge’s propensity for sending particular offenders to prison while reserving alternative, community-based sanctions for others.
The development of self-report data in the 1950s intensified the ongoing debate. Researchers administered surveys to individuals randomly selected from the population and asked them to report their criminal behaviors. Although many of the earliest of these studies did not support the belief that lower social classes were more criminal, there was also enough research that found contradictory results to ensure that the issue would not be resolved. Furthermore, there were as many sociologists and criminologists who attacked the validity of self-report data as there were those who took issue with the validity of official measures of crime. Their argument was that there is no way to determine whether people in self-report studies are telling the truth about their criminal behavior. Doubters suggested that self-report surveys were better measures of a participant’s willingness to tell the truth about his or her criminality. They also speculated that people from the lower classes were underreporting their deviant and criminal behavior while those in the upper classes were overreporting, thereby artificially reducing the magnitude of the correlation between lower-class status and criminality.
Tittle, Villemez, and Smith (1978) reviewed 35 research studies that had examined the social class–crime link and concluded that there was an extremely small relationship, with the members of the lower classes exhibiting slightly more criminality. They also noted that this relationship had become smaller over the past four decades.
This by no means settled the debate; instead, research became the impetus for even more extensive and complicated empirical efforts. Much of these later efforts attempted to discover the conditions under which social class influences criminality. One set of studies attempted to determine whether the manner in which social class and crime were measured affects the likelihood of discovering a link between social class and crime. In terms of social class, several studies suggested that this relationship may exist only among people in the lowest economic strata, the group sometimes referred to as the underclass. These studies then measured social class by dividing populations into dichotomous categories such as welfare recipients and nonrecipients or, for school-age children, those who receive free lunches and those who do not. Other studies used a composite measure of social class, which often included education, occupation, and income. Still other studies used Marxian classifications of social class, conceptualizing social class in terms of an individual’s (or his or her parents’) relationship to the means of production— specifically, whether they owned some means of production or sold their labor for a wage. Still others expanded this fairly simple classification to include other variables, such as whether one has control over the means of production and/or control over the labor of others. The emphasis on control helped to distinguish between wage workers who have managerial positions and those who do not, an increasingly prevalent distinction in modern society.
Crime was also measured in a number of different ways in an effort to determine whether it conditions the social class–criminality relationship. For example, a number of studies have examined whether the negative relationship between social class and delinquency existed only for the most serious criminal offenses or the most frequent offenders. Also, the source of crime data was thought to have an effect on whether a relationship between social class and crime was uncovered. Some criminologists held that crime would be shown to be more prevalent among the lower class if official police data or court records are used to determine criminality. As previously mentioned, they argued that people from lower classes are more likely to underreport their criminal behavior on self-report surveys.
A number of studies also sought to determine whether demographic and environmental variables had important conditioning effects on the class–crime relationship. For example, some studies examined whether the effect of social class on criminality was greater among blacks than it was for whites or among males than among females. Given the contradictory results of these research efforts, it would be difficult to suggest that the social class– criminality relationship was specific to a certain race or gender. Still another set of studies has examined whether this relationship was more likely in areas that were characterized as being more heterogeneous, more urban, or in higher status areas, and again produced mixed results.
Tittle and Meier (1990) reviewed the research literature that examined the relationship between socioeconomic status and delinquency and that attempted to specify whether any of the aforementioned conditions mattered. They concluded that there was little evidence that the link between social class and criminality existed under any of the conditions examined.
More recent and sophisticated studies have generally arrived at similar conclusions, although some studies did help clarify the relationship. For example, Wright and his colleagues (B. R. E. Wright, Caspi, Moffitt, Miech, & Silva, 1999) found that people in lower social classes experienced lower educational and occupational goals and more financial strain, aggression, and alienation, which in turn increase delinquency. Delinquency in the higher social classes, on the other hand, was the result of high socioeconomic status causing increased risk taking and social power, and diminished the commitment to conventional values, all of which then predispose these youth to delinquency. Dunaway, Cullen, Burton, and Evans (2000) examined the relationship between social class (measured in a variety of ways) and criminality (based on self-report surveys) and found that, among adults, the correlation was weak for less serious offenses. They did, however, find a class effect for violent offenses and among non-whites. This study was distinctive in that it measured adult criminality, a surprisingly underresearched population. In the end, the best conclusion that can be drawn about the relationship between social class and the commission of street crimes is that it tends to be weak and present only under certain specified conditions, and criminology researchers must continue to attempt to specify other circumstances that may influence this relationship.
What many of these studies do have in common is that most approached the definition of crime as being nonproblematic instead of acknowledging crime as a multifaceted concept that includes crimes of the disadvantaged as well as crimes of the powerful. Unfortunately, the latter were, and still are, less apt to be considered. This is important, because if studies included offenses that powerful individuals are more likely to commit (e.g., insider trading), and that those in lower classes are in no position to commit, then there would be little question as to whether criminality would appear more evenly distributed across social classes than has traditionally been thought. Moreover, it is only by including a wider variety of offenses that we can consider the social class–crime link as having been more completely and fairly tested.
So, although there has been little advancement toward settling the social class–street crime questions, the introduction of self-report studies has generally confirmed that criminality is more broadly and equally distributed across social classes than previously suspected. In fact, to date, these studies have consistently shown that nearly 90% of Americans have committed at least one crime for which they could have been sentenced to jail or prison. These findings may confirm that the use of official statistics means that we may not actually be measuring the level of crime or propensity for criminality but instead are measuring the decision-making practices of the criminal justice system (i.e., when to file an official police report, whom to arrest, whom to charge, and whom to send to prison).