Colin Loftin

Regional subculture and homicide: an examination of the Gastil-Hackney thesis.

Two recent comparative studies of regional variation in U.S. homicide rates by Sheldon Hackney and Raymond Gastil have been widely interpreted as verifying the hypothesis that the high levels of interpersonal violence, which characterize the Southern region, result largely from cultural factors that exist independent of the effects of situational variables. A close examination of Gastils and Hackneys studies reveals that their estimates of the independent effects of regional culture are based on poor measurement and arebiased in the direction of the culture of violence hypothesis. A new regression analysis of similar data using slightly modified procedures demonstrates the weaknesses of their procedures by showing that it is possible to obtain estimates of cultural and situational effects vastly different from those obtained by Gastil and Hackney.

Effects of Restrictive Licensing of Handguns on Homicide and Suicide in the District of Columbia

BACKGROUND Whether restricting access to handguns will reduce firearm-related homicides and suicides is currently a matter of intense debate. In 1976 the District of Columbia adopted a law that banned the purchase, sale, transfer, or possession of handguns by civilians. We evaluated the effect of implementing this law on the frequency of homicides and suicides. METHODS Homicides and suicides committed from 1968 through 1987 were classified according to place of occurrence (within the District of Columbia or in adjacent metropolitan areas where the law did not apply), cause (homicide or suicide), mechanism of death (firearms or other means), and time of occurrence (before or after the implementation of the law). The number of suicides and homicides was calculated for each month during the study period, and differences between the mean monthly totals before and after the law went into effect were estimated. RESULTS In Washington, D.C., the adoption of the gun-licensing law coincided with an abrupt decline in homicides by firearms (a reduction of 3.3 per month, or 25 percent) and suicides by firearms (reduction, 0.6 per month, or 23 percent). No similar reductions were observed in the number of homicides or suicides committed by other means, nor were there similar reductions in the adjacent metropolitan areas in Maryland and Virginia. There were also no increases in homicides or suicides by other methods, as would be expected if equally lethal means were substituted for handguns. CONCLUSIONS Restrictive licensing of handguns was associated with a prompt decline in homicides and suicides by firearms in the District of Columbia. No such decline was observed for homicides or suicides in which guns were not used, and no decline was seen in adjacent metropolitan areas where restrictive licensing did not apply. Our data suggest that restrictions on access to guns in the District of Columbia prevented an average of 47 deaths each year after the law was implemented.

Social Inequality and Crime Control

Almost all quantitative research on the determinants of variation in the allocation of public resources to crime control is done within the framework of either rational public choice theory or conflict theory. The rational public choice theory is rooted in notions of economic efficiency, social consensus, and relatively conflict-free democratic political processes. The conflict theory views conflicting interests, exploitation, and differences in power as the essence of politics. Rarely, however, has research tested the relative ability of the two theories to explain political outcomes. This study is an attempt to carry out such a test by considering the sources of variation in the size of urban police departments over time.

A comparison of supplementary homicide reports and national vital statistics system homicide estimates for U.S. counties

This article examines agreement between homicide estimates from the Supplementary Homicide Report (SHR) and the National Vital Statistics System (NVSS) at the county level for 1980 to 1988. NVSS and SHR estimates exactly agree in 22% of the counties (68% if agreement is defined as no more than a difference of four homicides), but in some cases, they differ substantially. Although the NVSS generally exceeds the SHR, the pattern is not uniform: 28% of the counties report more SHR homicides than NVSS homicides. Differences between estimates from the two systems are related to population size. Large-population counties often have substantial differences in homicide counts, but the deflating effect of dividing by population yields small rate differences. In contrast, dividing by population magnifies differences in small-population counties and produces discrepancies in the rates that are not present in the counts. The NVSS and SHR differ somewhat in their definition of cases, and other disagreements result from ambiguities in or failures to follow data collection procedures within each system.

Collective Security and the Demand for Legal Handguns.

One controversial element in the debate on firearms policy is whether crime and civil disorders contribute significantly to the private demand for firearms. In this paper we present a model to explain legal handgum demand that emphasizes the level of collective security as a key factor. Time-series data on legal gun demand in Detroit from 1951 to 1977 are consistent with a model in which individuals respond to three determinants of collective security: high violent crime rates, civil disordes, and police strength. The analysis suggests that low confidence in collective security contributes to both the need for and the resistance to gun control policies.

A spatial autocorrelation model of the effects of population density on fertility.

The relationship between fertility and population density that has been demonstrated with aggregate data is discussed and reexamined here. We argue that models which exclude considerations of spatial processes are incomplete and that therefore parameter estimates for such models are misleading. We then develop a model which incorporates spatial autocorrelation and reanalyze data from one well-known study of the effects of density on fertility using our model. The results show that with one exception the effects of density on fertility are trivial a finding that is quite different from previous research. We conclude that spatial mechanisms are an important consideration in the modeling of social processes that involve geographically defined units. (excerpt)

One With A Gun Gets You Two: Mandatory Sentencing and Firearms Violence in Detroit

Mandatory sentences for crimes committed with a gun are a popular policy because they promise a reduction in gun violence at a relatively low cost. In this article we present some results of a study of the implementa tion of such a law in Detroit, Michigan. Two major questions are discussed: (1) what effect did the Michigan gun law have on the certainty and severity of sentences; and (2) did the gun law reduce the number of serious violent crimes in Detroit? We find that, although the law required a two-year mandatory sentence for felonies committed with a gun and the prosecutor followed a strict policy of not reducing the gun law charge, there was little change in the certainty or severity of sentences that could be attributed to the effects of the gun law. Only in the case of assault was there a significant change in the expected sentence. Also serious violent crimes—murder, robbery, and assault—follow patterns over time that lead us to conclude that the gun law did not significantly alter the number or type of serious crimes in Detroit.

Underreporting of Justifiable Homicides Committed by Police Officers in the United States, 1976-1998

OBJECTIVES This study assessed the consistency of estimates of the number of justifiable homicides committed by US police officers and identified sources of underreporting. METHODS The number of justifiable homicides committed by police officers between 1976 and 1998 was estimated from supplementary homicide report (SHR) and National Vital Statistics System (NVSS) mortality data. RESULTS Nationally, the SHR estimate was 29% larger than the NVSS estimate. However, in most states this pattern was reversed, with more deaths reported in the NVSS. CONCLUSIONS Both systems underreport, but for different reasons. The NVSS misclassifies cases as homicides, rather than justifiable homicides committed by police officers, because certifiers fail to mention police involvement. The SHR misses cases because some jurisdictions fail to file reports or omit justifiable homicides committed by police officers.

The validity of robbery-murder classifications in Baltimore.

The most widely used source of nationwide data on homicide in the United States is the Supplementary Homicide Report (SHR) data compiled by the FBI as part of its Uniform Crime Reporting System. This paper describes a study of the reliability of robbery-murder classifications by the SHR for Baltimore, Maryland during 1983. The research is exploratory, but indicates a high level of inconsistency in the data. Of the 42 cases that were classified as robbery murders by the SHR or the replication study, only 20 were classified that way in both studies. The high level of unreliability can be attributed to three major problems: (1) the SHR codes are mutually exclusive, but many homicides could be placed in several of the categories; (2) there are few systematic rules for classifying ambiguously motivated homicides; and (3) the SHR placed too many cases in the “unknown” category. It appears that the reliability of the coding could be increased substantially with relatively minor changes in the procedures currently used.

The analysis of case-control studies in criminology

Case-control designs, in which subjects are selected conditionally on the value of a dependent variable, are well suited for research in criminology. For many research issues, case-control studies should be preferred over cohort studies because they can provide estimates of the same parameters, with equal standard errors, for less cost. This paper describes the case-control design and some of the problems that arise in interpreting them. An approach to analysis and interpretation that has developed in epidemiology is presented.