К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 

Scope can be subdivided into the types of events that are captured and

the populations covered. The scope of the NCVS, for example, is restricted

to nonfatal incidents and to the characteristics of crime victims rather than

offenders. Vital statistics and hospital-based information on firearm violence

is also limited to the victims. The UCR, by contrast, captures information

on both crime victims and offenders, but they are limited to offenses

that are known to and recorded by law enforcement agencies. The NCVS

includes data on both crimes reported to the police and those that victims

do not report. Household-based surveys such as the NCVS and the GSS are

limited to the population of persons with stable residences, thereby omitting

transients and other persons at high risk for firearm violence. Such

persons are included in the ADAM program, which collects information on

persons who come into contact with the criminal justice system.

Geographic coverage is another dimension of scope. The GSS, for example,

is representative of the United States and the nine census regions,

but it is too sparse geographically to support conclusions at finer levels of

geographical aggregation. This lack of individual-level data from small

geographical areas is a significant shortcoming in the firearms data. Presumably,

we would like to be able to make statements about, for example,

the probability that an individual commits suicide conditional on owning a

gun (or having one available) and other covariates. This cannot be done if

the smallest geographical unit that the data resolve is a multistate region.

Similar statements can be made about other forms of gun violence.

Perhaps no better illustration of the patchwork character of information

on firearms violence in the United States exists than the multiple and

nonoverlapping or partially overlapping coverage of the data sets. That

should come as little surprise, inasmuch as many of the data sets were

expressly intended to provide information about crime, violence, or injury

that was not available from other sources. The major impetus for the

development of the NCVS, for example, was to gather information on

crime incidents that do not come to the attention of law enforcement agencies.

The collection of information on violence from hospitals and emergency

departments is intended to reveal types of violence, such as partner

abuse, thought to be underreported in crime data sources.

The patchwork of existing data sources, in other words, has been created

with the best of intentions and has shed light on aspects of violence,

including the role of firearms, that otherwise would have remained hidden

from view, such as the burden on hospital emergency departments of firearm

injuries (Zawitz and Strom, 2000). However, insufficient attention has

been devoted to linkages across data in population coverage and the types

of firearm violence covered. Can data from the UCR, the NCVS, and emergency

departments be effectively linked to draw inferences about the firearms

violence in the population? As with data standardization, continuing

assessments of remaining gaps in the scope of firearms data should be part

of an ongoing program of methodological research on firearm violence.