V. The Lean Times: Social Disorganization in the 1970s and 1980s
After the replications that followed Shaw and McKay’s (1942) research, social disorganization as a theory began to decline. This was primarily a result of attacks on the use of official data in crime studies and growing criticism of theoretical problems with the theory. A few studies, however, continued to follow the principles of social disorganization. The general direction of these studies followed that of Shaw and McKay, but few followed their design closely enough to be considered replications. For example, these studies examined population status through scale measurement and analysis of change in population characteristics rather than single-variable correlations. In analyzing the association between economic status and delinquency, research in this era focused on the economic status of individuals rather than the housing conditions studied by Shaw and McKay. These studies typically measured economic characteristics through educational levels and the occupational status of residents. Because of the contradictory findings of earlier research and the growing contention that foreign birth had little to do with delinquency, these studies began to look to additional measures of population status in an effort to better measure its relationship with delinquency.
Quinney (1964) obtained data from Lexington, Kentucky, in 1960 and analyzed them with social area analysis. Quinney’s research included three dimensions: (1) economic status, (2) family status, and (3) ethnic status. Quinney’s family status was the variable most closely associated with Shaw and McKay’s (1942) physical status. Quinney used census data concerning women in the workforce, fertility rates, and single-structure housing. The results of his analysis showed that family status was negatively correlated with juvenile delinquency. These findings were significant even when interaction effects of economic variables were included. Quinney used two variables to examine economic status: (1) number of school grades completed and (2) number of blue-collar workers. The results showed that juvenile delinquency was negatively correlated with economic status. Quinney used the census variable race to examine ethnic status. The racial makeup of a census tract was found to be the most highly correlated with delinquency. Quinney then conducted a second analysis to determine the degree of delinquency exhibited by each race. This analysis revealed that white delinquency rates were lowest in areas with less than 2% blacks and increased steadily as the proportion of blacks increased, peaking in the 15% to 40% black grouping. Black delinquency, however was highest in areas with less than 2% black or more than 50% black but was lowest when the racial mix was predominantly, but not completely, white. In a third analysis, census tracts were divided into areas of high and low economic status and high and low family status. In this analysis, delinquency rates varied in relation to economic status; however, the presence of high family status always lowered the rate of delinquency.
In a partial replication of Quinney’s (1964) study, and to address the criticism of using official data in social disorganization research, Johnstone (1978) used self-reported data to test social disorganization theory. In this study, Johnstone administered self-reported delinquency questionnaires to 1,124 youth aged 14 through 18 living in Chicago. Johnstone also used a modified Shevsky–Bell social area analysis using “area status measures” and “family status measures.” The results of a factor analysis revealed that area-status-measures had a positive but nonsignificant relationship with fighting and weapon-related crimes and a negative and nonsignificant relationship with all other delinquency measures. In regard to family status measures, lower-class status was significantly associated with fighting and weapons offenses, burglary–larceny–robbery offenses, Uniform Crime Report Index offenses, and arrests.
An enduring criticism of Shaw and McKay’s (1942) research was the assumption of a stable delinquency pattern in the community rather than one experiencing change. Bursik and Webb (1982) attempted to test this hypothesis by examining data from Chicago. They used Shaw and McKay’s own data and updated it to the time of their study to facilitate an examination from 1940 to 1970 in 10-year increments. Data were drawn from all male referrals to the Chicago juvenile court in the years of 1940, 1950, 1960, and 1970 and from census data for the corresponding years. A regression analysis revealed that delinquency was not associated with the indicators of change between 1940 and 1950. For the two following periods, however, this trend was reversed. Bursik and Webb also found that communities exhibiting the most rapid change were characterized by the highest increases in delinquency. The analysis showed that communities with the highest rates of population change had an average of 12 more offenses per 1,000 youth than areas with either moderate or slow change. They concluded on the basis of these findings that it was the nature of the change, not the people involved in the change, that was affecting delinquency. In explaining how their findings differed from Shaw and McKay’s, Bursik and Webb concluded that the earlier study was not wrong but that it was conducted “within a specific historical context and grounded . . . in a model of ecological process that [has] changed dramatically since the publication of the 1942 monograph” (p. 36).
Four years later, Schuerman and Kobrin (1986) conducted a study similar to Bursik and Webb’s (1982) with a 20-year historical analysis of Los Angeles County. This was accomplished by gathering data from the juvenile court for 1950, 1960, and 1970 and correlating them with measures of land use, population composition, socioeconomic status, and subculture.
Schuerman and Kobrin (1986) proposed that neighborhoods travel through three stages: (1) emerging areas, with very low delinquency rates; (2) transitional areas, with moderate levels of delinquency; and (3) enduring areas, which maintain high levels of delinquency for many years. They also proposed that deterioration preceded a rise in delinquency in early stages of transition (supporting Shaw and McKay, 1942) but that as the city moved to the enduring stage, rises in the delinquency rate preceded deterioration.
In analyzing the relationship between land use (physical status) and delinquency, Schuerman and Kobrin (1986) found that the number of homes owned and land use type was inversely related with delinquency. They also found high mobility levels in persons living in high-delinquency areas. A cross-lagged regression analysis revealed that physical deterioration was most highly associated with increases in delinquency in emerging areas. As the area continued to deteriorate and delinquency rose, however, the most significant factors shifted to economic characteristics. Schuerman and Kobrin argued that the speed of change rather than the change itself that resulted in a neighborhood moving from low to high crime rates.
In analyzing the influence of socioeconomic factors on delinquency, Schuerman and Kobrin (1986) examined the occupation, unemployment, education, and housing characteristics of census tracts. This analysis revealed expected results of a low number of professional and skilled workers and a low percentage of people with advanced education in high-delinquency areas. The trend among housing characteristics in Schuerman and Kobrin’s study also supported the findings of Shaw and McKay (1942) and the replications. There was a general trend from owner- to renter-occupied housing and from single to multiple housing units as one moved from low-delinquency to high-delinquency areas. This supported Shaw and McKay’s proposal that delinquency was positively correlated with the percentage of people renting and negatively correlated with the percentage of homes owned. There were also significant increases in the degree of overcrowding in high-delinquency areas, which supported the findings of Lander (1954), Bordua (1959), and Chilton (1964). Unlike physical status characteristics, economic variables were not a significant factor of delinquency in emerging areas of Schuerman and Kobrin’s study. Socioeconomic status preceded increases in delinquency only in transitional and enduring stages.
Schuerman and Kobrin (1986) examined four population characteristics: (1) white and (2) non-white population and (3) white and (4) non-white female participation in the labor force. In high-delinquency areas, the percentage of blacks in the population rose slightly from 1950 through 1970, while the percentage of whites decreased dramatically in the same areas. Similar trends occurred in the female labor force participation. From 1950 through 1970, the black female participation in the labor force dropped slightly in high-delinquency areas, but the white female labor force dropped substantially. These findings were even more substantial in the cross-lagged analysis. Schuerman and Kobrin concluded from this analysis that rapid change in population characteristics, along with high rates of deterioration and population turnover, were preceding and greatly influencing the rate of increase in delinquency.
Sampson and Groves (1989) tested social disorganization theory using data from a survey of 10,905 residents in 238 localities in Great Britain. Their rationale was that previous research had relied on census data that were not valid measures of community structure or crime. Sampson and Groves also argued that survey data were superior to Shaw and McKay’s (1942) reliance on official crime. They also proposed that “low economic status, ethnic heterogeneity, residential mobility, and family disruption lead to community social disorganization, which in turn, increased crime and delinquency rates” (p. 775). On the basis of their analysis, Sampson and Groves concluded that social disorganization theory was supported, stating that “between-community variations in social disorganization transmit much of the effect of community structural characteristics on rates of both criminal victimization and criminal offending” (p. 774). Furthermore, they argued for expanded support for social disorganization theory in that “Shaw and McKay’s model explains crime and delinquency rates in a culture other than the United States” (p. 776).
An ironic major drawback of social disorganization research has been the relative lack of theory to guide or explain the research (Bursik, 1988). Much of the research in this area has paid tribute to social disorganization in the literature review and then simply conducted analyses with little theoretical explanation for the findings. Two authors (Sampson, 1986, and Stark, 1987) attempted to advance the theory itself and to provide a better link between neighborhood-oriented research and the theoretical foundation.
Responding to criticisms that ecological research lacked an intervening factor between the variables and criminality, Sampson (1986) proposed that a breakdown in informal social controls is this link. With this premise in mind, Sampson set out to show the link among ecological characteristics, social disorganization, loss of informal social control, and delinquency. The first link he attempted to make concerned the structural density of a neighborhood. In an earlier work, Sampson (1985) proposed that increases in density reduced the ability of a neighborhood to maintain surveillance and guardianship of youth and strangers. As the number of persons in a given living area increased, it was more difficult to know who lived in the area. When this occurred, residents were less able to recognize their neighbors or be concerned with their activities, resulting in an increased opportunity for delinquency. Sampson also proposed a link with residential mobility whereby he argued that neighborhoods with a high population turnover had a greater number of new faces, making it difficult to distinguish between new residents and strangers. Sampson proposed that economic status was related to delinquency through the attachment or social bond a person had to the neighborhood and the neighborhood’s willingness to maintain informal social control. He also proposed that people who owned their own homes had a greater attachment and commitment to the neighborhood and took steps to maintain neighborhood networks and social control. He examined two-parent versus one-parent families and their relative ability to maintain informal social control. Sampson proposed that two-parent families provided increased supervision and that because of this they were aware of and intervened in predecessors of involvement in more serious delinquent activities.
Stark (1987) furthered Sampson’s (1985, 1986) effort to add a theoretical framework to social disorganization research by formalizing some of the more important aspects of Shaw and McKay’s (1942) findings in developing a set of 30 propositions. The primary focus of Stark’s propositional framework was on Shaw and McKay’s physical status variables. The factors Stark used to analyze population status were transience of population, mixed-use neighborhoods, and overcrowding. Stark (1987) proposed that “transience weakens voluntary organizations, thereby directly reducing both informal and formal sources of social control” (p. 900). Stark also sought to provide a basis for understanding how proximity to industry and mixed-use areas influenced delinquency. Stark argued that in areas where residents lived close to commercial or industrial businesses there was more opportunity to commit delinquent acts (e.g., theft) because targets were readily available and close by. In purely residential areas, however, juveniles who wanted to commit such thefts might have to travel a great distance to get to a place where such acts could be committed. Stark proposed that economic status was linked to delinquency in two ways (physical status and population status). First, he proposed that homes in poor areas were typically more crowded; therefore, there was more anonymity and less supervision of children. Stark also linked economic status to delinquency through physical status in his proposition that “poor, dense neighborhoods tend to be mixed-use neighborhoods” (p. 902). In relating population status to delinquency, Stark proposed that physically unattractive areas reduced people’s commitment to their neighborhood. This proposition also supported Shaw and McKay’s conclusion that physically deteriorated areas in close proximity to industry and with a highly transient population cannot maintain commitment to the area by the residents and cannot maintain social control of delinquency.